Synthesis of novel cyano-cyclitols and their stereoselective biotransformation catalyzed by Rhodococcus erythropolis A4

Article abstract of DOI:10.1016/j.tetasy.2010.04.022

A variety of novel cyano-cyclitols possessing complex stereochemistry have been synthesized. These compounds were subjected to the biocatalyzed hydrolysis of their nitrile groups. The bacterial strain Rhodococcus erythropolis A4, expressing a nitrile hydr

Full text of DOI:10.1016/j.tetasy.2010.04.022

Tetrahedron: Asymmetry 21 (2010) 695–702
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Synthesis of novel cyano-cyclitols and their stereoselective
biotransformation catalyzed by Rhodococcus erythropolis A4
a
a
b
ˇ b
ˇ
Nicola D’Antona a, , Giovanni Nicolosi , Raffaele Morrone , David Kubác , Ondrej Kaplan ,
*
Ludmila Martínková b
a Istituto di Chimica Biomolecolare del CNR—UOS Catania, Via P.Gaifami 18, I-95126 Catania, Italy
Laboratory of Biotransformation, Institute of Microbiology of the Academy of Sciences of the Czech Republic, Vídenská 1083, CZ-142 20 Prague 4, Czech Republic
b
ˇ
a r t i c l e i n f o
a b s t r a c t
Article history:
A variety of novel cyano-cyclitols possessing complex stereochemistry have been synthesized. These
compounds were subjected to the biocatalyzed hydrolysis of their nitrile groups. The bacterial strain Rho-
dococcus erythropolis A4, expressing a nitrile hydratase/amidase bienzymatic system, was able to recog-
nize (1R,2S,3S,4R)/(1S,2R,3R,4S)-1-cyano-2,3,4-trihydroxy-cyclohex-5-ene and trans-3-cyanocyclohexa-
3,5-diene-1,2-diol, and to catalyze their transformations into the corresponding amides and acids. The
kinetic and stereochemical trends of these biotransformations, a rare example of the enantiorecognition
of a rigid bulky aliphatic substrate, are discussed.
Received 18 February 2010
Accepted 13 April 2010
Available online 14 May 2010
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
conditions, high temperatures, formation of undesirable by-prod-
ucts, low yields and the generation of large amounts of waste salts,
The development of molecules with enhanced biological activ-
ity is a continuing primary goal in biomedical research. The need
for compounds interacting with biological receptors characterized
by a high grade of asymmetry leads researchers to synthesize mol-
ecules possessing well defined stereochemical patterns, with the
declared aims of obtaining better substrate-receptor interactions,
minimizing collateral effects due to different isomeric forms, and
reducing the amount of drug administered. An important class of
cyclitols1 characterized by a high stereochemical complexity are
conduritols,2 1,2,3,4-Cyclohexenetetrols possessing four stereo-
genic centers, and consequently existing as four different pairs of
enantiomers, namely conduritol B, C, E, and F, and two meso forms,
conduritol A and D. Only conduritol A and B are present in nature,3
but all possible stereoisomers have been synthesized4,5 and all of
them have exhibited important biological activities,6 making these
cyclitols potentially key molecules in hit-to-lead drug discovery.
The conduritol scaffold can also be found in many natural cells
mediators such as inositols,7 glycerophosphoinositols,8 pseudosu-
gars,9 and antibiotics such as manumycin A10 and tricholomenyn
A,11 or RNA mimics.12 Possible key synthetic equivalents in the
preparation of these valuable compounds are the nitrilic deriva-
tives of conduritols, due to the great versatility of cyano group
transformations.
making this route economically and environmentally inconve-
nient, and unsuitable, especially with unstable substrates such as
polyhydroxylated compounds. A possible alternative is offered by
biocatalysis, since the enzymatic hydrolysis of nitriles can be per-
formed under milder conditions, for example, at approximately
neutral pH.14
In Nature, the biotransformation of nitriles proceeds via two
different pathways (Fig. 1): the direct nitrilase-catalyzed reaction
to yield the corresponding carboxylic acids and ammonia, or a
two-step indirect reaction, in which nitriles are first transformed
to amides by nitrile hydratase, and then the amides are trans-
formed to the corresponding carboxylic acids and ammonia by
amidase.
In contrast with this definite synthetic potential, the use of ni-
trile hydrolyzing enzymes in organic synthesis has remained re-
stricted until now and examples of their application in the
hydrolysis of cyclic aliphatic substrates are negligible in number.15
Moreover, although many of the enzymatic transformations of ni-
triles which have been reported thus far involve straightforward
achiral substrates,16 stereoselective transformations involving
racemic or prochiral substrates are particularly interesting.17
While in the first publications18 to describe the hydrolysis of race-
mic nitriles, the resolution was assumed to only occur in the amide
hydrolysis step under the influence of the amidase, subsequent
work19 showed that, with selected substrates, nitrile hydratases
or nitrilases could also present their intrinsic enantioselectivity.
Due to the unpredictable nature of enantioselective enzymatic ni-
trile hydrolysis, there is still great interest in investigating a variety
of new enzymatic transformations of nitriles.
Unfortunately, even if the conventional chemical hydrolysis of
nitriles is a well known reaction,13 it suffers from several disadvan-
tages, including the requirement for highly acidic or basic reaction
* Corresponding author. Tel.: +39 095 7338342; fax: +39 095 7338310.
E-mail address: nicola.dantona@icb.cnr.it (N. D’Antona).
0957-4166/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2010.04.022

696
N. D’Antona et al. / Tetrahedron: Asymmetry 21 (2010) 695–702
Figure 1. Different pathways of nitrile metabolism in microorganisms.
Herein, we present the synthesis of novel chiral nitrilic condur-
spectroscopic data as well as elemental analysis data for the novel
compounds are given in the Section 4.
itol derivatives, and an investigation of their biotransformation
reactions, aiming to obtain optically active carboxylic acids and
amides for the preparation of valuable bioactive compounds.
2.2. Biotransformations
Compounds ( )-6, ( )-7 and ( )-8 were subjected to a prelimin-
ary general biotransformation protocol (see Section 4), which
could give rise, depending on the catalyst used, to the correspond-
ing amides and/or acids. Bacterial whole-cell preparations of Rho-
dococcus erythropolis A4 grown under conditions to express its
bienzymatic nitrile hydratase/amidase system were used, as well
as those of Rhodococcus sp. NDB 1165 and Nocardia sp. NHB-2
grown under conditions to express their nitrilase activities. A fun-
gal strain of Fusarium solani O1 expressing nitrilase activity was
also assayed. All substrates were dissolved in a Tris-HCl buffer
(pH 8) containing up to 5% of methanol as a cosolvent, to increase
their solubility. The nitrilase-producing strains Rhodococcus sp.,
Nocardia sp. and F. solani O1 did not give any reaction with the
compounds tested. This observation was in accordance with the
characterization of the nitrile-transforming enzymes in all these
organisms as ‘aromatic nitrilases’ with a preference for benzoni-
trile analogues.22 The nitrilases of this type generally exhibit no
or low activities for aliphatic, arylaliphatic or alicyclic nitriles. On
the other hand, whole cells from R. erythropolis A4 recognized sub-
strates ( )-6 and ( )-7, catalyzing their biotransformation to the
corresponding amides and acids. This was in accordance with the
previous finding that nitrile hydratases are largely less sensitive
to steric hindrance than nitrilases and hence they also transform
relatively bulky substrates.23
Since nitrile-hydrolyzing biocatalysts are intracellular enzymes,
with substrates having to cross the cell wall to interact with them,
the observed activity of R. erythropolis A4 could also be a result of
the facilitated uptake of the aforementioned hydrophobic sub-
strates by cells with long aliphatic chains of mycolic acids in their
cell envelopes. In contrast, the more hydrophilic nature of sub-
strate ( )-8 could result in an incapacity to cross the cell wall,
and explain its complete absence of recognition. To verify this
hypothesis, the biotransformation of compound ( )-8 was repeated
in the presence of a cellular extract of R. erythropolis A4, again
yielding no reaction product; this result, which goes against our
previous hypothesis, would indicate that in this case the absence
of reactivity is not due to the polarity/mobility features of the sub-
strate but more generally to its structural properties. Using the
2. Results and discussion
2.1. Synthesis of substrates
The first part of this work was concerned with the development
of a synthetic strategy for the preparation of cyano-deoxyconduri-
tols, novel molecules in which a hydroxyl group of the conduritol
framework has been replaced by a nitrile group. Commercially
available p-benzoquinone 1, an inexpensive and highly versatile
building block, was used as the starting material. Compound 1
was subjected to sequential bromination, reduction and epoxida-
tion reactions (Scheme 1)20, yielding the racemic diepoxide ( )-4.
In the presence of calcium oxide as a heterogeneous catalyst21 and
trimethylsilylcyanide as the nitrilating agent, compound ( )-4 gave a
quantitative yield of non-isolable intermediate (( )-5), that upon
treatment with trifluoroacetic acid produced two main compounds,
which were chromatographically purified and spectroscopically
identified as (1R, 2S, 3S, 4R)/(1S, 2R, 3R, 4S)-1-cyano-2,3,4-trihy-
droxy-cyclohex-5-ene ( )-6 and trans-3-cyanocyclohexa-3,5-diene-
1,2-diol ( )-7 (Scheme 2).
A study of the coupling constants Jvic observed in 1H NMR spec-
tra permitted the assignment of all the relative configurations of
the chiral centres, indicating compound ( )-6 as a derivative of
conduritol B. Analogously, the relative configurations of the stere-
ogenic centers of ( )-7 were determined, and the results were in
agreement with the hypothesis that it is a stable derivative of
( )-6, generated under acidic conditions, by the b-elimination of
a water molecule driven by an increase in conjugation. Compound
( )-7 was further subjected to trans-dihydroxylation via a regiose-
lective epoxidation reaction by treatment with m-chloroperben-
zoic acid (mCPBA), and subsequent stereoselective opening of the
oxirane ring by the action of trifluoroacetic acid (Scheme 3), yield-
ing (1R,2S,3R,4S)/(1S,2R,3S,4R)-5-cyano-1,2,3,4-tetrahydroxycyclo-
hex-5-ene ( )-8 as the sole product.
Again, the analysis of 1H NMR coupling constants Jvic permitted
the assignment of the relative configuration of the stereogenic cen-
tres of compound ( )-8, a direct derivative of conduritol A, in which
the nitrile substituent is directly located on the alkenic bond. The
Scheme 1.

N. D’Antona et al. / Tetrahedron: Asymmetry 21 (2010) 695–702
697
Scheme 2.
Scheme 3.
same cellular extract, compounds ( )-6 and ( )-7 continued to be
recognized, even though the enzymes lost their activity after only
2 h, probably because of a greater instability of the catalytic system
when not operating in its natural cellular environment.
At a reaction time of 3 h and 20% conversion, the concentration
of amide 9 tended to reach a plateau while the concentration of the
acid continued to increase and that of the nitrile to decrease. At 8 h,
the biotransformations were drastically slowed down, and at 24 h
it virtually stopped with final concentrations of 40% both for com-
pounds 6 and 10; this can be mainly explained by the inactivation
of the biocatalyst. The biotransformation displayed a notable level
of enantiorecognition. The first step catalyzed by the nitrile hydra-
tase displayed a low enantioselectivity (E ꢀ 2.5) as demonstrated
by the maximum value of enantiomeric excess (40%, Fig. 3), of ni-
trile recovered after a reaction time of 24 h (approx. 60% conver-
sion, Figure 2).
To study both the kinetic behaviour of the bienzymatic system
and its potential enantioselective features, substrate ( )-6 was
transformed by the whole cells of R. erythropolis A4 and samples
withdrawn at regular intervals were appropriately derivatized and
analyzed using gas chromatography. In the early stages of the reac-
tion, it was possible to observe a fast conversion of nitrile ( )-6 into
the corresponding amide and, in turn, the conversion of the latter
into the acid derivative, as demonstrated by the increase in the con-
centrations of the species 1-carboxamido-2,3,4-trihydroxy-cyclo-
hex-5-ene 9 and 2,3,4-trihydroxy-cyclohex-5-en-1-carboxylic acid
10 in the solution (Fig. 2).
Nevertheless, this observation is noteworthy, since examples of
enantioselective nitrile hydratases are scarce in the scientific liter-
ature and are almost non-existent for aliphatic bulky substrates. In
CN
CONH2
COOH
OH
OH
OH
R.erythropolis A4
R.erythropolis A4
(Amidase)
(Nitrilehy dratase)
OH
OH
OH
OH
( )-6
OH
(+)-9
OH
(+)-10
Only one isomer is drawn for the racemic mixtures.
Figure 2. Kinetic trend of biotransformation reaction of ( )-6 catalyzed by Rhodococcus erythropolis A4.

698
N. D’Antona et al. / Tetrahedron: Asymmetry 21 (2010) 695–702
On the other hand, the kinetic and enantioselective trends of
biotransformation of substrate ( )-7 by whole cells of R. erythrop-
olis A4 were quite different. The first step, catalyzed by the nitrile
hydratase, was very fast; complete conversion to the correspond-
ing amide ( )-13 was achieved within 15 min (Fig. 4); no enantio-
preference being shown by the catalyst. By prolonging the reaction
time, complete transformation of compound ( )-13 into carboxylic
acid ( )-14 catalyzed by the amidase could be observed.
This second step was characterized by a very low value of the
enantiomeric ratio (E = 2.2), both amide 13 and acid 14 being pro-
duced with poor enantiomeric excesses as determined by chiral
HPLC (Figure 5). A comparison with literature data26 relative to
compound 14 enabled the absolute configuration of the preferen-
tially synthesized enantiomer to be identified as (+)-(1S,2S) trans-
3-carboxylic-cyclohexa-3,5-diene-1,2-dihydroxy acid.
3. Conclusions
The chemical synthesis of several novel cyano-cyclitols has
been achieved. Enzymatic hydrolysis of these racemic substrates
was obtained for the first time using, under mild conditions (buffer
solution pH 8, 35 °C), bacterial whole-cell preparations of R. ery-
thropolis A4 expressing the bienzymatic system nitrile hydratase/
amidase. However, both the efficiency and enantioselectivity of
the biotransformations were strongly dependent upon the struc-
tures of the nitrile substrates. The best results were obtained when
(1R,2S,3S,4R)/(1S,2R,3R,4S)-1-cyano-2,3,4-trihydroxy-cyclohex-5-
ene ( )-6 was used as the substrate, with a conversion value of 60%
after 8 h, and enantiomeric excesses values in the range 70–90%
both for the corresponding amide and acid, depending on the reac-
tion time. While this observation shows an important instance of
an (S) enantioselective amidase, even more noteworthy is that it
represents the first case to our knowledge of a nitrile hydratase
that is enantioselective towards bulky aliphatic substrates. The
versatile utility of the resulting optically active novel carboxylic
acids and amide derivatives renders this method very attractive
and practical in organic synthesis. The extension of biotransforma-
tions for the synthesis of various chiral cyclitols and their applica-
tions in organic chemistry to the synthesis of enantiopure bioactive
compounds are currently being investigated in our laboratories.
Figure 3. Stereochemical trend of biotransformation reaction of ( )-6 catalyzed by
Rhodococcus erythropolis A4.
contrast, the second step catalyzed by the amidase exhibited a
moderate to good enantioselectivity, since it was possible to re-
cover both the acid and the amide with an ee of 70–90%, depending
on the reaction time. To better understand the potentialities of the
enzymatic system, the theoretical trends of enantiomeric excesses
in case the catalyst was not inactivated after 20 h were calculated,
by using the simulation software ‘SeKiRe’ developed by Faber’s
group for the study of bienzymatic systems,24 and are shown as
dotted lines in Figure 3. To define the stereopreference displayed
by the bienzymatic system, the absolute configuration of the enan-
tiomerically enriched product 10 was determined after chemical
derivatization (Scheme 4), and the specific rotation of the resulting
compound 12 was measured.
By comparing the sign of the specific rotation determined and
those reported in literature for the enantiomers of 12,25 it was
possible to identify compound 10 as (1S,2R,3R,4S)-2,3,4-trihy-
droxy-cyclohex-5-en-1-carboxylic acid. This data demonstrated
the stereopreference of the amidase towards the substrate with an
(S)-configuration. Considering this information and the enantiose-
lectivity trends shown in Figure 3, it was also possible to identify
amide 9 as (1R,2S,3S,4R)-1-carboxamido-2,3,4-trihydroxy-cyclo-
hex-5-ene, and, as in the previous case, the stereospecificity of the
enzyme (in this case nitrile hydratase) towards a substrate with an
(S)-configuration. The ‘inversion’ in the configuration of the recov-
ered amide was not surprising, since it is a typical effect of a sequen-
tial double resolution with two enzymes having the same
enantiopreference but differing levels of enantioselectivity (the sec-
ond catalytic system being more selective than the first).
4. Experimental
4.1. General
1H and 13C NMR spectra were recorded in the indicated deuter-
ated solvents in a Bruker Avance™ 400 spectrometer at 400.13 and
100.62 MHz, respectively. Chemical shifts (d) are given as parts per
million relative to the residual solvent peak and coupling constants
(J) are in hertz. Column chromatography was performed on Silica
Gel 60 (70–230 mesh) using the specified eluents. GC analyses
were carried out in a ShimadzuÒ Fast-GC-17A equipped with a
FID detector and a SUPELCOÒ Fast-SPB™-5, using n-undecane as
the internal standard. Chiral GC analyses were carried out in a Per-
kin–ElmerÒ 8500 equipped with a FID detector and a MEGAÒ di-
methyl-tert-butylsilyl-b-cyclodextrin (coated on OV 1701)
column. Chiral HPLC analyses were carried out in a DIONEXÒ appa-
COOH
OH
COOCH3
OH
CH2OH
OH
MeOH/H+
Reflux, 8h
NaBH4
THF/MeOH, 30m
OH
OH
OH
OH
OH
OH
11
(+)-10
(+)-12
Scheme 4.

N. D’Antona et al. / Tetrahedron: Asymmetry 21 (2010) 695–702
699
CONH2
OH
COOH
OH
CN
OH
OH
R.erythropolis A4
(Amidase)
R.erythropolis A4
(Nitrile hydratase)
OH
OH
( )-7
( )-13
(+)-14
Only one isomer is drawn for the racemic mixtures.
100
90
80
70
60
50
40
30
20
10
0
100
(b)
(a)
90
80
70
60
50
40
30
20
10
0
% CN (7)
% CONH2 (13)
% COOH (14)
0
5
10
15
20
25
30
35
40
45
50
0
0.5
1
1.5
2
2.5
3
Reaction time (hours)
Reaction time (hours)
Figure 4. Kinetic trend of biotransformation reaction of ( )-7 catalyzed by Rhodococcus erythropolis A4: (a) full reaction time range; (b) expansion in the time range 1–3 h.
100
CN (7)
vacuum at 40 °C. The mixture was purified by silica gel column
CONH2 (13)
chromatography using 100% CH2Cl2 as the eluant, and recovered
as a white solid (54.64 g, 0.20 mol, 87% yield). 1H and 13C NMR
spectra, and melting point range (not reported) are in accordance
with literature data.20
80
60
40
20
0
COOH (14)
4.3. Synthesis of (1S,4S,5R,6R)/(1R,4R,5S,6S)-5,6-dibromocyclohex-
2-ene-1,4-diol, ( )-3
0
20
40
60
80
100
-20
-40
-60
Compound ( )-2 (54.64 g, 0.2 mol) was dissolved in 900 ml of
distilled diethyl ether and cooled in an ice bath. After the addition
of 300 ml of water, NaBH4 (19.0 g, 0.50 mol) was added over
30 min with stirring. After 2.5 h, the clarified organic phase was ex-
tracted with water, dried over Na2SO4 and taken to dryness to yield
compound ( )-3 (47.3 g, 0.17 mol, 87% yield) as a white solid (mp
110–111 °C). MS (ESI+): m/z 295 [M+Na]+. 1H NMR ((CD3)2CO):
5.73 (2H, s, H-2, H-3); 4.85 (2H, d, J = 6.5, OH); 4.6–4.4 (2H, m,
H-1, H-4); 4.4–4.3 (2H, m, H-5, H-6). 13C NMR ((CD3)2CO): 131.1
(2C, C-2, C-3); 73.9 (2C, C-1, C-4); 61.9 (2C, C-5, C-6). Anal. Calcd
for C6H8Br2O2: C, 26.50; H, 2.97. Found: C, 26.61; H, 2.99.
Concentration (%)
Figure 5. Stereochemical trend of biotransformation reaction of ( )-7 catalyzed by
Rhodococcus erythropolis A4.
ratus equipped with a ChiralcelÒ OD or OJ column (Daicel Chemical
Industries) using n-hexane/iso-propanol mixtures as the mobile
phase and detection by UV-vis detector at 220, 250, 266 and
275 nm. Optical densities (OD) were measured on a DAD-UV–visi-
ble AgilentÒ spectrophotometer at 610 nm. Optical rotations were
measured on a DIP 135 JASCOÒ instrument at 25 °C. ESI-MS spectra
were acquired in positive mode in a WatersÒ Micromass ZQ2000,
using a 10 V cone voltage, 3 kV capillary voltage and 150 °C source
temperature. Melting points are uncorrected. The simulation of ki-
netic and enantioselectivity trends were performed using the spe-
cific software ‘Sequential Kinetic Resolution’ (SeKiRe), developed
by Professor K. Faber and his group.24 The enantiomeric ratio (E)
was calculated from the experimental values of conversion and
enantiomeric excess using the equation reported in the litera-
ture.27 The enantiomeric excess of compound 10 was calculated
from the experimental values of concentrations and enantiomeric
excesses of compounds 6 and 9, using the aforementioned ‘SeKiRe’
software. The chemicals, solvents and materials for microbiological
cultures were obtained from Sigma–AldrichÒ, FlukaÒ, Carlo ErbaÒ,
Riedel-de HaenÒ, DifcoÒ and used without further purification.
4.4. Synthesis of (3S,4S,5S,6S)/(3R,4R,5R,6R)-3,4:5,6-diepoxycy
clohex-1-ene, ( )-4
Compound ( )-3 (2.0 g, 7.35 mmol) was dissolved in 30 ml of
anhydrous tetrahydrofuran, under an argon atmosphere at 0 °C,
and KOH (4.7 g, 0.08 mol) was slowly added to the solution. After
2 h, another portion of KOH (1.7 g, 0.03 mol) was added, and the
solution stirred continually for
a further 2 h, centrifuged at
4000 rpm and taken to dryness to yield compound ( )-4 (783 mg,
7.11 mmol, 96.8% yield) as a white solid. 1H and 13C NMR spectra,
and melting point range (not reported) are in accordance with lit-
erature data.20
4.5. Synthesis of (1R,2S,3S,4S)/(1S,2R,3R,4R)-1-cyano-3,4-epoxy-
2-trimethylsilylcyclohex-1-ene, ( )-5
CaO (2 g, 0.036 mol) and Me3SiCN (2 ml, 15.9 mmol) were
added to a solution of ( )-4 (485 mg, 4.40 mmol) in 14 ml of
CH2Cl2/hexane 1:1, and the mixture was stirred for 24 h at room
temperature. The solution was filtered and concentrated under a
nitrogen stream, producing a mixture containing compound ( )-5
that could not be isolated because of its instability.
4.2. Synthesis of (5R,6R)/(5S,6S)-5,6-dibromocyclohex-2-ene-
1,4-dione, ( )-2
Br2 (12.5 ml, 0.24 mol) at 0 °C was added over 2 h to a solution
of p-benzoquinone 1 (25 g, 0.23 mol) in 150 ml of CHCl3. After 1 h,
the reaction was stopped by evaporating the solution under
MS (ESI+): m/z 216 [M+Na]+.

700
N. D’Antona et al. / Tetrahedron: Asymmetry 21 (2010) 695–702
4.6. Synthesis of (1R,2S,3S,4R)/(1S,2R,3R,4S)-1-cyano-2,3,4-
trihydroxy-cyclohex-5-ene ( )-6 and trans-3-cyanocyclohexa-
3,5-diene-1,2-diol ( )-7
was maintained at 4 °C on a modified Czapek-Dox agar31 and
grown in the presence of 2-cyanopyridine (nitrilase inducer) as de-
scribed previously.32
A mixture containing compound ( )-5 was sequentially dis-
4.9. Preparation of whole-cell catalysts
solved in 500
CF3COOH (600
ond amount (500
l
l
l of THF, and 10 ml of H2O. An initial volume of
l) was added dropwise to the solution and a sec-
l) was added 1 h later, with the mixture contin-
Whole cells of bacteria were harvested by centrifugation and
washed with Tris–HCl buffer (50 mM, pH 8). The F. solani mycelium
was harvested by filtration, washed with the same buffer and
lyophilized. Bacterial biomass was immediately used for biotrans-
formations. The lyophilized fungal mycelia were stored at À20 °C
until use, the nitrilase activity being stable for several months un-
der these conditions.
The activities of nitrile-transforming enzymes and amidase
were determined with benzonitrile and benzamide, respectively,
using the previously described assays.33,34 The nitrile hydratase
and amidase activities in whole cells of R. erythropolis A4 were ap-
prox. 1.4 and 0.3 U mgÀ1 of dry cell weight, respectively. The nitri-
lase activities of Nocardia globerula NHB-2, Rhodococcus sp. NDB
1165 and F. solani were approx. 1.9, 1.1 and 0.5 U mgÀ1 of dry cell
weight, respectively.
l
ually stirred. After 1 h, the reaction was stopped by evaporating the
solution under vacuum. The mixture, after being purified by silica
gel column chromatography using MeOH/CH2Cl2 1:9 as the elu-
ents, yielded compound ( )-6 (171.4 mg, 1.10 mmol, 25% yield),
and compound ( )-7 (259.2 mg, 1.89 mmol, 43% yield), both in
the form of oils.
4.6.1. (1R,2S,3S,4R)/(1S,2R,3R,4S)-1-Cyano-2,3,4-trihydroxy-
cyclohex-5-ene ( )-6
MS (ESI+): m/z 178 [M+Na]+. 1H NMR ((CD3)2SO): 5.69 (1H, d,
OH), 5.65 (1H, ddd, J = 10.0, 2.3, 2.7 Hz, H-5), 5.47 (1H, ddd,
J = 10.0, 2.0, 2.1 Hz, H-6), 5.23 (1H, d, OH), 5.14 (1H, d, OH), 3.88
(1H, dd, J = 2.1, 7.9 Hz, H-1), 3.53 (1H, t, J = 9.7 Hz, H-3), 3.42 (1H,
ddd, J = 9.7, 2.9, 2.2 Hz, H-4), 3.17 (1H, dd, J = 9.7, 8.0 Hz, H-2).
13C NMR ((CD3)2SO): 134.4 (1C, C-5), 120.9 (1C, CN), 119.5 (1C,
C-6), 76.6 (1C, C-2), 71.4 (2C, C-1, C-3), 36.6 (1C, C-4). Anal. Calcd
for C7H9NO3: C, 54.19; H, 5.85; N, 9.03. Found: C, 54.00; H, 5.91;
N, 9.05.
4.10. Preparation of cell-free extract of R. erythropolis A4
Whole cells of R. erythropolis A4 suspended in Tris/HCl buffer
(50 mM, pH 8) were disrupted using a Retsch MM-200 oscillation
mill as described previously.33 The cell extracts were immediately
used for biotransformations. The activities of nitrile hydratase and
amidase were approx. 3 and 0.7 U mgÀ1 of protein, respectively.
4.6.2. trans-3-Cyanocyclohexa-3,5-diene-1,2-diol ( )-7
MS (ESI+): m/z 160 [M+Na]+. 1H NMR (CD3OD): 6.77 (1H, dd,
J = 5.60, 1.21 Hz, H-4), 6.23 (1H, ddd, J = 9.8, 3.1, 1.0 Hz, H-6),
6.07 (1H, ddd, J = 9.7, 5.5, 1.6 Hz, H-5), 4.34 (1H, ddd, J = 9.6, 3.1,
1.6 Hz, H-1), 4.30 (1H, dd, J = 9.6, 1.2 Hz, H-2). 13C NMR (CD3OD):
138.2 (1C, C-4), 136.8 (1C, C-6), 122.5 (1C, C-5), 117.9 (1C, C-3),
114.9 (1C, CN), 71.8 (1C, C-1), 71.2 (1C, C-2). Anal. Calcd for
C7H7NO2: C, 61.31; H, 5.14; N, 10.21. Found: C, 61.57; H, 5.19; N,
10.16.
4.11. General procedure for biotransformations
A 10 mM solution of substrate was prepared using a suspension
of biomass in Tris–HCl buffer (50 mM, pH 8), having an optical den-
sity (OD) value of approx. 15. A percentage of MeOH up to 5% was
previously used to solubilise the substrate. The suspension was
shaken (200 rpm) at 35 °C; aliquots were drawn at regular time
intervals and analysed by TLC (MeOH/CH2Cl2 1:1).
4.7. Synthesis of (1R,2S,3R,4S)/(1S,2R,3S,4R)-5-cyano-1,2,3,4-
tetrahydroxycyclohex-5-ene ( )-8
4.12. Biotransformation of ( )-6 catalysed by whole cells from R.
erythropolis A4
At first, m-CPBA (90 mg, 0.52 mmol) was added to a solution of
( )-7 (43.7 mg, 0.32 mmol) in 5 ml of CH2Cl2, and the mixture stir-
red at room temperature for 1 h. After evaporating the solvent un-
Compound ( )-6 (14 mg, 0.09 mmol) and undecane (14
internal standard for gas chromatography analysis) were dissolved
in 500 l of MeOH, and added to a suspension containing whole
ll,
der vacuum, the mixture was redissolved in 50
ll of THF and 10 ml
of H2O, and CF3COOH (200 l) was added. After 3 h of continuous
l
l
stirring, the mixture was filtered, evaporated under vacuum and
purified by silica gel column chromatography using MeOH/CH2Cl2
1:4 as the eluents, yielding compound ( )-8 as an oil (27 mg,
0.16 mmol, 50% yield). MS (ESI+): m/z 194 [M+Na]+. 1H NMR
(CD3OD): 6.56 (1H, d, J = 2.9 Hz, H-6), 4.24 (1H, dd, J = 6.4, 2.7 Hz,
H-1), 4.15 (1H, d, J = 4.6 Hz, H-4), 3.84 (1H, dd, J = 4.6, 2.3 Hz, H-
3), 3.77 (1H, dd, J = 6.4, 2.2 Hz, H-2). 13C NMR (CD3OD): 147.4
(1C, C-6), 118.8 (1C, C-5), 116.8 (1C, CN), 73.9 (1C, C-4), 72.4 (1C,
C-1), 70.7 (1C, C-3), 69.6 (1C, C-2). Anal. Calcd for C7H9NO4: C,
49.12; H, 5.30; N, 8.18. Found: C, 48.90; H, 5.31; N, 8.12.
cells from R. erythropolis A4 (OD = 16; approx. 3.7 mg dry cell
weight mLÀ1) in Tris–HCl buffer (50 mM). The mixture was shaken
(200 rpm) at 35 °C for 2 days; the reaction was stopped by centri-
fugation of the solution, and evaporation under vacuum of the
supernatant at 40 °C. The mixture was purified by silica gel column
chromatography using MeOH/CH2Cl2 1:1 as the eluents, and
yielded compound (À)-9 (3.0 mg, 0.017 mmol, 20% yield,
ee = 91.0%), compound (+)-10 (5.3 mg, 0.030 mmol, 34% yield,
ee = 74.5%), and (À)-6 as an unreacted substrate (4.2 mg,
0.027 mmol, 30% yield, ee = 43%).
The course of the reaction, including the concentration and
enantiomeric excesses values, was monitored by withdrawing ali-
quots at regular time intervals that, after centrifugation, lyophilisa-
tion and derivatization (see below), were analyzed by gas
chromatography.
Derivatization protocol: a 200 ll sample was added to 100 ll of
a sylilating mixture (hexamethyldisilazane/trimethylchlorosilane/
pyridine, 3:1:9), and stirred at 40 °C for 30 min.
4.8. Microorganisms and cultures
The bacterial strains were maintained at 4 °C on meat peptone
agar (in g/L, Bacto beef extract 3, peptone 10, NaCl 5, agar 15). R.
erythropolis A428 was grown for 2 days at 28 °C in shaken 500-mL
Erlenmeyer flasks containing 100 mL of basal salts broth according
to Di Geronimo and Antoine,29 supplemented with 10 g/L of glyc-
erol and 3 g/L of yeast extract. Nocardia globerula NHB-222a and
Rhodococcus sp. NDB 116522b were grown in the presence of isobu-
tyronitrile (nitrilase inducer) as described previously.30 F. solani O1
Fast-GC conditions: injector temperature 250 °C; detector tem-
perature 280 °C; oven 100 °C for 1 min then 10 °C/min up to 280 °C
for 1 min. Chiral-GC conditions: injector temperature 250 °C;

N. D’Antona et al. / Tetrahedron: Asymmetry 21 (2010) 695–702
701
detector temperature 250 °C; oven 100 °C for 1 min then 0.4 °C/
min up to 200 °C for 1 min.
solution, and evaporation under vacuum of the supernatant at
40 °C. The mixture was purified by silica gel column chromatogra-
phy using MeOH/CH2Cl2 1:1 as the eluants, and yielded compound
(À)-13 (7.7 mg, 0.050 mmol, 45% yield, ee = 27.0%), and compound
(+)-14 (8.3 mg, 0.053 mmol, 48.6% yield, ee = 25.0%). The course of
the reaction was monitored by Fast-GC using the same protocols
and device conditions as described above for the biotransformation
of compound ( )-6. The enantiomeric excesses values were ob-
tained by chiral HPLC analysis using a ChiralcelÒ OJ column and
hexane/ethanol 9:1 as the mobile phase for compound 7, and a
ChiralcelÒ OD column and hexane/i-propanol 85:15 as the mobile
phase for compound 13.
4.12.1. (À)-(1R,2S,3S,4R)-1-Carboxamido-2,3,4-trihydroxy-
cyclohex-5-ene, (À)-9
MS (ESI+): m/z 174 [M+Na]+; 1H NMR (CD3OD): 5.66 (1H, dt,
J = 10.0, 2.4 Hz, H-6), 5.55 (1H, dt, J = 10.4, 2.4 Hz, H-5), 4.10–4.05
(1H, m, H-1), 3.82 (1H, t, J = 9.7 Hz, H-3), 3.44 (1H, dd, J = 10.1,
8.0 Hz, H-2), 3.15–3.06 (1H, m, H-4). 13C NMR (CD3OD): 176.0
(1C, -CONH2), 130.8 (1C, C-6), 123.9 (1C, C-5), 76.9 (1C, C-2), 71.8
(1C, C-1), 70.9 (1C, C-3), 50.9 (1C, C-4). [
MeOH, ee = 91.0%). Anal. Calcd for C7H11NO4: C, 48.55; H, 6.40; N,
8.09. Found: C, 48.67; H, 6.46; N, 8.17.
a
]
D = À180.3 (c 0.150,
4.15.1. (À)-(1R,2R)-trans-3-Carboxyamidocyclohexa-3,5-diene-
1,2-diol, (À)-13
4.12.2. (+)-(1S,2R,3R,4S)-2,3,4-trihydroxy-cyclohex-5-en-1-
carboxylic acid, (+)-10
MS (ESI+): m/z 178 [M+Na]+; 1H NMR (CD3OD): 6.91–6.88 (1H,
m, H-4), 6.19 (1H, dd, J = 3.6 Hz, H-6), 6.09 (1H, dd, J = 1.6, 4.0 Hz,
H-5), 4.62–4.58 (1H, m, H-1), 4.32–4.27 (1H, m, H-2). 13C NMR
(CD3OD): 171.9 (1C, CONH2), 134.7 (1C, C-4), 133.6 (1C, C-3),
130.5 (1C, C-6), 124.6 (1C, C-5), 73.8 (1C, C-2), 71.1 (1C, C-1).
MS (ESI-): m/z 173 [M–H]À; 1H NMR (CD3OD): 5.76 (1H, d,
J = 10.1, 2.0 Hz, H-6), 5.55 (1H, d, J = 10.4, 2.3 Hz, H-5), 4.10–4.05
(1H, m, H-1), 3.87 (1H, t, J = 9.6 Hz, H-3), 3.47 (1H, dd, J = 10.1,
7.6 Hz, H-2), 3.00–2.90 (1H, m, H-4). 13C NMR (CD3OD): 178.6
(1C, –COOH), 128.3 (1C, C-5), 126.3 (1C, C-6), 76.9 (1C, C-2), 72.2
[a
]D = À0.8 (c 0.385, MeOH, ee = 27.0%). Anal. Calcd for C7H9NO3:
(1C, C-4), 71.7 (1C, C-3), 52.9 (1C, C-1). [
a]
D = +128.2 (c 0.265,
C, 54.19; H, 5.85; N, 9.03. Found: C, 53.92; H, 5.79; N, 9.06.
MeOH, ee = 74.5%). Anal. Calcd for C7H10O5: C, 48.29; H, 5.79.
Found: C, 48.48; H, 5.74.
4.15.2. (+)-(1S,2S)-trans-3-Carboxylic-cyclohexa-3,5-diene-1,2-
dihydroxy acid, (+)-14
4.13. Synthesis of (1S,2R 3R,4S)-2,3,4-trihydroxy-cyclohex-5-en-
1-methylcarboxylate, 11
MS (ESI-): m/z 155 [MÀH]À. 1H NMR (CD3OD): 6.90–6.86 (1H,
m, H-4), 6.39–6.33 (1H, m, H-6), 6.29–6.24 (1H, m, H-5), 4.60–
4.54 (1H, m, H-1), 4.30–4.23 (1H, m, H-2). 13C NMR (CD3OD):
171.3 (1C, COOH), 134.4 (1C, C-4), 130.5 (1C, C-3), 124.5 (1C, C-
At first, HCl (100 ll) was added to a solution of (+)-10 (4.5 mg,
0.025 mmol) in 2 ml of MeOH, and the mixture stirred under
refluxing conditions for 8 h. After the addition of water, the reac-
tion mixture was repeatedly partitioned with AcOEt; the final or-
ganic phase was dried over anhydrous Na2SO4 and taken to
dryness to yield compound (1S,2R,3R,4S)-2,3,4-trihydroxy-cyclo-
hex-5-en-1-methylcarboxylate, 11 (3.7 mg, 0.019 mmol, 76% yield)
in the form of an oil. MS (ESI+): m/z 211 [M+Na]+; 1H NMR
(CD3OD): 5.75–5.70 (1H, m, H-6), 5.60–5.50 (1H, m, H-5), 4.03–
3.97 (1H, m, H-1), 3.75–3.70 (1H, m, H-3), 3.48–3.42 (1H, m, H-
2), 2.99–2.94 (1H, m, H-4). 13C NMR (CD3OD): 173.0 (1C, -CO),
127.3 (1C, C-6), 125.3 (1C, C-5), 76.9 (1C, C-2), 72.7 (1C, C-3),
70.2 (1C, C-4), 53.9 (1C, C-1), 52.2 (1C, -OCH3). Anal. Calcd for
C8H12O5: C, 51.06; H, 6.43. Found: C, 50.87; H, 6.50.
5), 73.9 (1C, C-2), 71.2 (1C, C-1). [
a]D = +1.0 (c 0.415, H2O,
ee = 25.0%); lit. [
a]
D = +3.9 (c 0.1, H2O).26 Anal. calcd for C7H8O4:
C, 53.85; H, 5.16. Found: C, 53.65; H, 5.22.
Acknowledgements
Financial support via project OC09046 (Ministry of Education,
Czech Rep.), institutional research concept AV0Z50200510 (Insti-
tute of Microbiology), and COST/ESF action CM0701 is gratefully
acknowledged. We would like to thank Professor T.K. Bhalla
(Himachal Pradesh University, Shimla, India) for kindly providing
the Nocardia globerula NHB-2 and Rhodococcus sp. NDB 1165
strains.
4.14. Synthesis of (À)-(1R,2R,3R,4S)-1-
References
(hydroxymethyl)cyclohex-5-en-2,3,4-triol (À)-12
1. Hudlicky, T.; Cebulak, M. Cyclitols and their Derivatives; VCH: New York, 1993.
2. Balci, M.; Sutbeyaz, Y.; Secen, H. Tetrahedron 1990, 46, 3715–3742.
3. Kübler, K. Arch. Phann. Ber. Stsch. Pharm. 1908, 246, 620–622.
4. Balci, M. Pure Appl. Chem. 1997, 69, 97–104.
5. Dangschat, G.; Fischer, H. O. L. Naturwissenschaften 1939, 27, 756–757.
6. Chida, N.; Yamada, K.; Ogawa, S. J. Chem. Soc., Chem. Commun. 1991, 588–590.
7. Mitchell, R. H. Nat. Rev. Mol. Cell Biol. 2008, 9, 151–161.
8. Di Paolo, G.; De Camilli, P. Nature 2006, 443, 651–657.
9. Ogawa, S.; Sekura, R.; Maruyama, A.; Yuasa, H.; Hashimoto, H. Eur. J. Org. Chem.
2000, 2089–2093.
10. Alcaraz, L.; Macdonald, G.; Ragot, J. P.; Lewis, N.; Taylor, R. J. K. J. Org. Chem.
1998, 63, 3526–3527.
11. Miller, M. W.; Johnson, C. R. J. Org. Chem. 1997, 62, 1582–1583.
12. Vijgen, S.; Nauwelaerts, K.; Wang, J.; Van Aerschot, A.; Lagoja, I.; Herdewijn, P. J.
Org. Chem. 2005, 70, 4591–4597.
At first, NaBH4 (4.3 mg, 0.114 mmol) was added to a solution of
(1S,2R,3R,4S)-2,3,4-trihydroxy-cyclohex-5-en-1-methylcarboxy-
late, 11 (3.7 mg, 0.019 mmol) in 2 ml of distilled THF/MeOH (1:3),
and the mixture kept agitated for 30 min. After the addition of
water, the reaction mixture was extracted with AcOEt; the final or-
ganic phase was dried over anhydrous Na2SO4 and taken to dryness
to yield compound (À)-12 (3.0 mg, 0.018 mmol, 96% yield) in the
form of an oil. MS (ESI+): m/z 183 [M+Na]+; 1H NMR data (not re-
ported) are in accordance with literature data.25
[
a
]
D = À11.94 (c
0.3, MeOH, ee = 74.5%), lit. [
a
]
D = À13.4 (c 1.1, MeOH).25 Anal. Calcd
for C7H12O4: C, 52.49; H, 7.55. Found: C, 52.62; H, 7.61.
13. Vogel, A. I. Practical Organic Chemistry, 4th ed.; Longman.
14. Klempier, N.; Harte, G.; de Raadt, A.; Griengl, H.; Braunegg, G. Food Technol.
Biotech. 1996, 34, 67–71.
15. (a) Matoishi, K.; Sano, A.; Imai, N.; Yamazaki, T.; Yokoyama, M.; Sugai, T.; Ohta,
H. Tetrahedron: Asymmetry 1998, 9, 1097–1102; (b) Martínková, L.; Klempier,
4.15. Biotransformation of ( )-7 catalysed by whole cells from R.
erythropolis A4
ˇ
N.; Preiml, M.; Ovesná, M.; Kuzma, M.; Mylerová, V.; Kren, V. Can. J. Chem.
2002, 80, 724–727; (c) Winkler, M.; Meischler, D.; Klempier, N. Adv. Synth.
Catal. 2007, 349, 1475–1480.
16. Mylerová, V.; Martínková, L. Curr. Org. Chem. 2003, 7, 1–17.
17. Martínková, L.; Krˇen, V. Curr. Opin. Chem. Biol. 2010, 14 (in press).
18. Faber, K. Biotransformations in Organic Chemistry; Springer: Berlin, Heidelberg,
New York, 1992.
Compound ( )-7 (15 mg, 0.11 mmol) was dissolved in 550 ll of
MeOH, and added to a suspension containing whole cells from R.
erythropolis A4 (OD = 17; approx. 3.9 mg dry cell weight mLÀ1) in
Tris–HCl buffer (50 mM, pH 8). The mixture was shaken
(200 rpm) at 35 °C for 2 h, then stopped by centrifugation of the

702
N. D’Antona et al. / Tetrahedron: Asymmetry 21 (2010) 695–702
ˇ
19. Wang, M. X. Top. Catal. 2005, 35, 117–130.
20. Esser, T.; Farkas, F.; Mangholz, S. M.; Sequin, U. Tetrahedron 1994, 50, 3709–
3720.
Pelantová, H.; van Pelt, S.; Bolte, J.; Kren, V.; Martínková, L. J. Mol. Catal. B:
Enzym. 2008, 50, 107–113.
29. Di Geronimo, M. J.; Antoine, A. D. Appl. Environ. Microb. 1976, 31, 900–906.
ˇ
21. Sugita, K.; Ohta, A.; Onaka, M.; Izumi, Y. Bull. Chem. Jpn. 1991, 64, 1792–1799.
22. (a) Bhalla, T. C.; Kumar, H. Can. J. Microbiol. 2005, 51, 705–708; (b) Prasad, S.;
Misra, A.; Jangir, V. P.; Awasthi, A.; Raj, J.; Bhalla, T. C. World J. Microb. Bio. 2007,
23, 345–353.
23. Deigner, H.; Blencowe, C.; Freyberg, C. E. J. Mol. Catal. B: Enzym. 1996, 1, 61.
24. Kroutil, W.; Kleewein, A.; Faber, K. Tetrahedron: Asymmetry 1997, 8, 3251–
3261.
30. Veselá, A. B.; Franc, M.; Pelantová, H.; Kubác, D.; Vejvoda, V.; Šulc, M.; Bhalla, T.
C.; Macková, M.; Lovecká, P.; Janu˚ , P.; Demnerová, K.; Martínková, L.
Biodegradation, in press.
31. Kaplan, O.; Vejvoda, V.; Charvátová-Pišvejcová, A.; Martínková, L. J. Ind.
Microbiol. Biot. 2006, 33, 891–896.
32. Vejvoda, V.; Kaplan, O.; Bezouška, K.; Pompach, P.; Šulc, M.; Cantarella, M.;
ˇ
Benada, O.; Uhnáková, B.; Rinágelová, A.; Lutz-Wahl, S.; Fischer, L.; Kren, V.;
25. Hansen, F. G.; Bundgaard, E.; Madsen, R. J. Org. Chem. 2005, 70, 10139–10142.
26. Franke, D.; Lorbach, V.; Esser, S.; Dose, C.; Sprenger, G. A.; Halfar, M.; Thömmes,
J.; Müller, R.; Takors, R.; Müller, M. Chem. Eur. J. 2003, 9, 4188–4196.
27. Chen, C. S.; Sih, C. J. Angew. Chem., Int. Ed. 1989, 28, 695–707.
Martínková, L. J. Mol. Catal. B: Enzym. 2008, 50, 99–106.
ˇ
ˇ
33. Vejvoda, V.; Kaplan, O.; Kubác, D.; Kren, V.; Martínková, L. Biocatal.
Biotransform. 2006, 24, 414–418.
ˇ
ˇ
34. Vejvoda, V.; Šveda, O.; Kaplan, O.; Prikrylová, V.; Elišáková, V.; Himl, M.; Kubác,
D.; Pelantová, H.; Kuzma, M.; Martínková, L. Biotechnol. Lett. 2007, 29, 1119–
1124.
ˇ
28. Kubác, D.; Kaplan, O.; Elišáková, V.; Pátek, M.; Vejvoda, V.; Slámová, K.;
Tóthová, A.; Lemaire, M.; Gallienne, E.; Lutz-Wahl, S.; Fischer, L.; Kuzma, M.;