A high throughput screening strategy for the assessment of nitrile-hydrolyzing activity towards the production of enantiopure β-hydroxy acids

Article abstract of DOI:10.1016/j.molcatb.2013.08.001

Nitrile hydrolysing enzymes have found wide use in the pharmaceutical industry for the production of fine chemicals. This work presents a strategy that facilitates the rapid identification of bacterial isolates demonstrating nitrile hydrolysing activity. The strategy incorporates toxicity, starvation and induction studies along with subsequent colorimetric screening for activity, further focusing the assessment towards the substrates of interest. This high-throughput strategy uses a 96 well plate system, and has enabled the rapid biocatalytic screening of 256 novel bacterial isolates towards β-hydroxynitriles. Results demonstrate the strategy's potential to rapidly assess a variety of β-hydroxynitriles including aliphatic, aromatic and dinitriles. A whole cell catalyst Rhodococcus erythropolis SET1 was identified and found to catalyse the hydrolysis of 3-hydroxybutyronitrile with remarkably high enantioselectivity under mild conditions, to afford (S)-3-hydroxybutyric acid in 42% yield and >99.9% ee. The biocatalytic capability of this strain including the variation of parameters such as temperature and time were further investigated and all results indicate the presence of a highly enantioselective if not enantiospecific nitrilase enzyme within the microbial whole cell.

Full text of DOI:10.1016/j.molcatb.2013.08.001

Journal of Molecular Catalysis B: Enzymatic 97 (2013) 150–155
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
A high throughput screening strategy for the assessment of
nitrile-hydrolyzing activity towards the production of enantiopure
ˇ-hydroxy acids
Tracey M. Coady, Lee V. Coffey, Catherine O‘Reilly, Erica B. Owens, Claire M. Lennon∗
Waterford Institute of Technology, Pharmaceutical & Molecular Biotechnology Research Centre, Department of Chemical and Life Sciences,
Cork Road Campus, Waterford, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history:
Nitrile hydrolysing enzymes have found wide use in the pharmaceutical industry for the production of
fine chemicals. This work presents a strategy that facilitates the rapid identification of bacterial isolates
demonstrating nitrile hydrolysing activity. The strategy incorporates toxicity, starvation and induc-
tion studies along with subsequent colorimetric screening for activity, further focusing the assessment
towards the substrates of interest. This high-throughput strategy uses a 96 well plate system, and has
enabled the rapid biocatalytic screening of 256 novel bacterial isolates towards ˇ-hydroxynitriles. Results
demonstrate the strategy’s potential to rapidly assess a variety of ˇ-hydroxynitriles including aliphatic,
aromatic and dinitriles. A whole cell catalyst Rhodococcus erythropolis SET1 was identified and found to
catalyse the hydrolysis of 3-hydroxybutyronitrile with remarkably high enantioselectivity under mild
conditions, to afford (S)-3-hydroxybutyric acid in 42% yield and >99.9% ee. The biocatalytic capability of
this strain including the variation of parameters such as temperature and time were further investigated
and all results indicate the presence of a highly enantioselective if not enantiospecific nitrilase enzyme
within the microbial whole cell.
Received 19 April 2013
Received in revised form 2 August 2013
Accepted 2 August 2013
Keywords:
Biotransformation
Bacterial isolates
Nitrile hydrolysing activity
High throughput screening
Enantioselectivity
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
acids via hydrolysis reactions however can be prone to elimination
reactions under classical acid/base conditions [7]. Alternatively,
The awareness of the importance of chirality in conjunction
with biological activity has led to an increasing demand for effi-
cient methods for the industrial synthesis of enantiomerically pure
compounds [1]. Many chiral drugs now employ chiral synthons in
their synthesis typically utilising the chiral pool, kinetic resolution
techniques, or asymmetric synthesis [2] to afford single enantiomer
compounds. Drawbacks associated with these approaches can be
the use of expensive substrates and/or chiral metal complex cata-
lysts/reagents which also in the latter case may require specialised
conditions, can be toxic and not easily recovered/reused [3]. The
combination of chemical procedures with biocatalytic methods can
offer an excellent alternative strategy for the production of such
compounds [4,5].
The distinct features of enzymatic transformations of nitriles
in particular are the formation of enantiopure carboxylic acids
and the generation of enantiopure amides, which are valuable
compounds in synthetic chemistry [6]. ˇ-Hydroxy nitriles such as
3-hydroxybutyronitrile, 3-hydroxyglutaronitrile and 3-hydroxy-3-
phenylpropionitriles can act as sources of ˇ-hydroxy carboxylic
nitrile biocatalysis can selectively facilitate this hydrolysis without
affecting other acid- or alkali-labile functional groups present [8,9].
This potential has resulted in significant work to identify bacte-
ria and fungi capable of hydrolysing such nitriles [10,11], whose
products can be widely used as chiral precursors for pharmaceu-
tical compounds. For example, 3-hydroxy-3-phenylpropionic acid
and its derivatives, have been used as precursors to chiral drugs
such as nisoxetine [12], fluoxetine [13] and tomoxetine [12]. Addi-
tionally, of particular commercial interest, is the nitrilase catalysed
hydrolysis of 3-hydroxyglutaronitrile, the ethyl-ester of which is an
intermediate to the cholesterol lowering drug (atorvastatin) Lipi-
tor [14,10]. However the biotransformations of substrates having
such a chiral centre remote from the cyano functional group have
been reported to proceed with, in many cases, disappointingly low
enantioselectivity and chemical yield [6,15–18].
A major issue in the development of a specific biotransformation
is to find the appropriate biocatalyst. If there are no commercially
available enzyme preparations the desired activities must be found
either by screening for enzymatic activity in strains from culture
collections or using microbial selection methods [19,20].
Selective enrichment is a method of choice for the isolation
of nitrile hydrolysing micro-organisms, where nitrile substrates
are used as a nitrogen source [19]. It is subsequently important
∗
Corresponding author. Tel.: +353 51834048.
E-mail address: clennon@wit.ie (C.M. Lennon).
1381-1177/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2013.08.001

T.M. Coady et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 150–155
151
to efficiently screen libraries of isolated strains to identify those
possessing the desired reactivity and selectivity. Developing
conventional methods such as high performance liquid chromatog-
raphy (HPLC) or gas chromatography can be tedious and time
consuming for assaying a large number of isolates.
agar-agar ultrapure (15% w/v) prepared in M9-media), containing
ˇ-hydroxy nitrile (10 mM) and incubated at 25 ◦C for 72 h.
2.4. Nitrogen starvation and induction of isolates
Various high-throughput screening methods have been devel-
oped for nitrile hydrolysing enzymes, which have been reviewed
by Martinkova [20] and Gong [21] recently. For example He et al.
reported a simple, rapid and high throughput screening method
for assaying nitrile hydrolysing enzymes based on ferric hydroxy-
mate spectrophotometry [22]. Lin et al. found that a combination
of ferrous and ferric ions could be used to distinguish ˛-amino
nitriles, ˛-amino amides and ˛-amino acids in aqueous solu-
tion [23]. Zhang et al. recently developed a solid screen selective
culture medium plate with bromcresol purple to screen isolates
capable of converting iminodiacetonitrile to iminodiacetic acid
[24]. However, colorimetric determination of ammonia as a com-
mon product of nitrilases and NHases/amidase activity provides a
promising method for rapid screening [25]. For example the Berthe-
lot method which quantifies ammonia formation colorimetrically
at 640 nm after reaction with a phenol hypochlorite reagent [26].
Alternatively addition of Nesslers reagent to a solution containing
ammonia produces an orange brown colour after development for
a certain amount of time [27].
2.4.1. Nitrogen starvation
Successful isolates from Section 2.3 were inoculated to 250 ␮l
M9 minimal media broth [31] with modifications [32], contain-
ing 10 mM ˇ-hydroxynitrile as the sole source of N. Each well
was then inoculated from M9-agar containing ˇ-hydroxynitrile
(10 mM) from Section 2.3 before incubation at 25 ◦C with orbital
shaking at 250 rpm for 24 h.
2.4.2. Induction of nitrile metabolising enzyme activity
96-well Megablocks® containing 500 ␮L of M9 minimal media
broth [31] with modifications [32] and ˇ-hydroxy nitrile (10 mM)
were inoculated with nitrogen starved isolates (20 ␮L) from Section
2.4.1 and incubated with shaking for 72 h at 25 ◦C and 250 rpm.
500 ␮L of 50% glycerol solution was then added to each culture
before storage at −80 ◦C. These cultures of induced isolates served
as inoculation and activity screening stocks for subsequent analy-
ses.
As mentioned previously highly effective and enantioselective
biotransformations of ˇ-hydroxynitriles would provide a conve-
nient approach to producing optically active ˇ-hydroxy carboxylic
acids [28]. The objective of this study was to screen and iden-
tify bacteria with activity towards the ˇ-hydroxynitriles from a
novel bacterial isolate bank (previously collected from soil and
seaweed) in our laboratory [29,30]. From the soil enrichment cul-
tures a total of 187 isolates were purified, and 67 isolates were
obtained from the seaweed enrichments [30]. We now report a high
throughput screening strategy for the rapid identification of bacte-
rial isolates possessing nitrile hydrolysing activity towards a variety
ˇ-hydroxynitriles including 3-hydroxybutyronitrile, 3-hydroxy-3-
phenylpropionitrile and 3-hydroxyglutaronitrile, which represent
an aliphatic, aromatic nitrile and dinitrile respectively. We also
demonstrate the application of this strategy to a successful enan-
tioselectivity screening study, resulting in the identification of an
isolate demonstrating >99.9% ee for 3-hydroxybutyronitrile specif-
ically.
2.5. An investigation of temperature on the growth of various
isolates
96-well Megablocks® containing 500 ␮L of M9 broth and ˇ-
hydroxy nitrile (10 mM) were inoculated with nitrile induced
isolates (20 ␮L) from Section 2.4.2 and incubated with shaking for
72 h at 15 ◦C, 20 ◦C, 25 ◦C and 30 ◦C at 250 rpm. The OD600nm mea-
sured and recorded for future studies.
2.6. Standard curve for Nesslers colorimetric assay
Ammonium chloride (100 mM, 5.35 g) was dissolved in
deionised water (1000 mL), this solution was used as a stock solu-
tion to prepare standards with the following concentration 1 mM,
2 mM, 3 mM, 4 mM and 5 mM. The absorbance was then measured
at 425 nm. At NH3 concentrations up to 10 mmol/L, the A425 nm was
directly proportional to NH3 concentration R2 = 0.9992.
2. Materials and methods
2.1. Materials
2.7. Nesslers microscale ammonia assay
Racemic
3-hydroxybutyronitrile
and
3-hydroxy-3-
Induced isolates from Section 2.4.2 were initially screened for
activity towards ˇ-hydroxynitriles using the Nesslers colorimet-
ric assay [27] modified and outlined in [33], in 96-well microtitre
plates (Sarstedt Ltd.). Fresh cultures of each isolate were grown
in M9-minimal media containing 10 mM nitrile as in Section 2.4.2
before washing 3 times with 500 ␮L of phosphate buffer). Each
150 ␮l reaction contained 10 mM nitrile and cells (OD600 nm = 0.1)
in potassium phosphate buffer (100 mM, pH 7.0), Microtitre plates
were sealed using adhesive film (Sarstedt) and incubated at 25 ◦C
at 250 rpm for 24 h. The reaction was then quenched by adding
37.5 ␮l of 250 mM HCl. Plates were centrifuged at 500 g for 10 min
to pellet the cells/debris. 20 ␮l of the supernatant was trans-
ferred to a microtitre plate, 181 ␮L of assay mastermix was added
(155 ␮L deionised water, 1 ␮L 10 N NaOH, 25 ␮L Nesslers reagent
(Merck). The reaction was allowed to stand for 10 min and then
the absorbance read at 425 nm. Cell blanks contained cells @
OD600nm = 0.1 in phosphate buffer. Nitrile blanks contained 150 ␮L
of 10 mM nitrile in phosphate buffer.
phenylpropionitrile were purchased from Sigma–Aldrich. All
other chemicals were of analytical grade and obtained from
various commercial sources.
2.2. Bacterial isolates
The 254 bacterial isolates used in this study are those described
in Coffey et al. [30].
2.3. Nitrile toxicity studies
96 well Megablock® plates (Sarstedt Ltd.) containing 250 ␮l LB
broth (Merck) and 10 mM ˇ-hydroxy nitrile were inoculated from
glycerol stocks of the isolate library and incubated at 25 ◦C with
shaking at 250 rpm for 72 h. 5 ␮L of these cultures then transferred
in a 96 well format to M9 [31] agar with modifications [32], (Merck

152
T.M. Coady et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 150–155
2.8. General procedure for enantioselectivity screening towards
3-hydroxybutyronitrile
Racemic nitrile (5.1 mg, 5.9 ␮L, 10 mM) was added in one portion
to a solution of potassium phosphate buffer (0.1 M, pH = 7.2, 6 mL)
containing induced cells (OD600 nm = 1), and activated at 25◦ C for
30 min with orbital shaking (250 rpm). The reaction was quenched
after 24 h by removal of the biomass by centrifugation at 3000 × g.
The resulting aqueous solution was acidified by the addition of
1 M HCl (200 ␮L). The aqueous portion was then extracted with
ethyl acetate, the extracts were dried over MgSO4 and the solvent
removed under vacuum. Silver oxide (1 equiv, 0.06 mmol, 13.6 mg),
benzylbromide (4 equiv, 0.24 mmol, 28 ␮L) and dichloromethane
(2 mL) were added and the mixture stirred in the dark for 24 h. The
reaction mixture was diluted with acetone and filtered through a
0.45 ␮m filter and solvent was removed under vacuum. 1 mL of
mobile phase (90% hexane: 10% IPA) was added before the solution
was injected on the Chiral HPLC system. All experiments were per-
formed in triplicate. % enantiomeric excess is calculated from the
ratio of the enantiomer by the difference in peak area divided by
the sum of the peak areas for the major and minor enantiomers.
Fig. 1. Nesslers colorimetric activity assay towards 3-hydroxybutyronitrile, an
example of library screening for nitrile-hydrolyzing enzyme activity in a 96-well
plate format. The biotransformation was carried out at 25 ◦C with 10 mM nitrile.
The reaction was then quenched by the addition of HCl. The biomass was removed
by centrifugation. 20 ␮l of the supernatant was transferred to a microtitre plate,
181 ␮L of assay mastermix was added. The reaction was allowed to stand for 10 min
and then the absorbance read at 425 nm.
(Promega UK), 15 pmoles each primer and 30 ng DNA. The PCR con-
ditions used for amplification were; 1 cycle of 95 ◦C for 5 min, 30
cycles of 95 ◦C for 1 min, 56 ◦C for 1 min, 72 ◦C for 1 min, followed
by 1 cycle of 72 ◦C for 5 min. PCR products were purified using the
DNA Clean and Concentrator – 5 kit (Zymo Research, CA, USA), as
per the manufacturer’s instructions. DNA sequencing of PCR prod-
ucts was performed using the BigDye 3.1 kit (Applied Biosystems)
as per the manufacturer’s instructions and analysed using an ABI
Prism 310 Genetic Analyser (Applied Biosystems, CA, USA).
Nucleotide sequences were analysed using the BLASTn or
BLASTx software [37] (http://www.ncbi.nlm.nih.gov/BLAST/) from
the GenBank (NCBI) database The sequence of isolate SET1 was
deposited in GenBank with accession number KF156942.
2.9. General procedure for large scale biotransformation towards
3-hydroxybutyronitrile
The procedure for large scale biotransformation of racemic 3-
hydroxybutyronitrile was similar to the general procedure, with
the exception of the use of potassium phosphate buffer (0.1 M, pH
7.0, 100 mL) containing 3-hydroxybutyronitrile (85.1 mg, 10 mM).
The resulting aqueous solution was basified to pH 12 with aque-
ous NaOH (2 M) and extracted with ethyl acetate (3 × 100 mL). The
aqueous solution was acidified using aqueous HCl (2 M) to pH 2
and extracted with ethyl acetate (3 × 100 mL), dried over MgSO4
and the solvent removed under vacuum. The crude product was
subjected to silica gel column chromatography eluted with a mix-
ture of hexane and ethyl acetate (1:1) to give 3-hydroxybutyric
acid in 42% yield (44 mg, 4.23 mmol) as clear oil. The configura-
tion of the corresponding acid was determined by comparing the
direction of specific rotation with that of an authentic sample. Enan-
tiomeric excess values were obtained from HPLC analysis using
a column of chiral stationary phase and correlated with litera-
ture. (R)-enantiomer elutes at 11.94 min, (S)-enantiomer elutes at
12.34 min [34]. 1H NMR (400 MHz, CDCl3) ı = 4.19–4.27 (1H, m),
ı = 2.45–2.58 (2H, m), ı = 1.23 (3H, d, J = 6.3 Hz). 13C NMR (400 MHz,
CDCl3) ı = 117, 76, 64, 42, 22 [˛]d25 = +4.0 (c = 2.5, MeOH), and com-
pared with that in the literature [˛]d25 = +4.1 (c = 2.7, MeOH) [35].
This experiment was performed in triplicate.
3. Results and discussion
3.1. High throughput activity screening
The bacterial isolates were first subjected to toxicity studies
with the nitriles of interest. By attempting to grow the differ-
ent isolates in rich medium in the presence of ˇ-hydroxynitriles,
the isolates that were sensitive to the nitrile were identified and
excluded from further screening. 145 of the initial 256 isolates
could grow in the presence of 3-hydroxybutyronitrile, with 77
demonstrating growth after a 48 h period and the remaining 68
after 144 h. A total of 65 of the isolates grew in the presence of
3-hydroxy-3-phenylpropionitrile and 107 in the presence of 3-
hydroxyglutaronitrile. The toxicity study demonstrated that some
isolates show no growth in the presence of the nitriles, where
as 16 isolates showed growth against two or more nitriles such
as isolate SS1-12. The tolerant isolates then underwent a starva-
tion process in order to ensure that the ˇ-hydroxynitrile would
be utilised as the nitrogen source during the induction stage,
which involved attempting to grow the isolates in minimal media
using the ˇ-hydroxynitrile as the sole nitrogen source. As bacte-
rial isolates utilising ˇ-hydroxynitriles as a sole nitrogen source
result in the release of ammonia, enzyme activity was monitored
using the technique of Nesslerisation. A high-throughput technique
was developed using the Nesslers microscale colorimetric assay
to screen for activity in the hydrolysis of ˇ-hydroxynitriles to ˇ-
hydroxycarboxylic acids using the Nesslers colorimetric assay [27]
modified and outlined in [33], in 96 well microtitre plates (Sarstedt
Ltd.) as shown in Fig. 1. Nitrile hydrolyzing activity was calculated
by measuring the absorbance of the sample at 425 nm, and relat-
ing the absorbance to the standard curve in order to determine the
concentration of NH3 in the sample.
2.10. Chiral HPLC separations
Chiralcel AD-H and OJ-H columns (all from Daicel Chemical
Industries) were used for chiral analysis. Chiralcel AD-H was used
for the resolution of ˇ-hydroxyacids. Analytical conditions applied:
90% hexane, 10% IPA and 0.1% TFA, with a flow rate of 0.8 mL/min
and a detection wavelength of 215 nm. Chiralcel OJ-H was used
for the resolution of ˇ-hydroxyamides and nitriles using the same
mobile phase conditions with the exception of TFA. The biotransfor-
mation products of 3-hydroxybutyronitrile were first derivatised to
their corresponding ˇ-benzyloxyethers before analysis.
2.11. Conventional PCR screening and DNA Sequencing
Bacterial genomic DNA was purified using the DNeasy Tissue Kit
(Qiagen, Germany) as per the manufacturer’s instructions. The 16S
ribosomal DNA was amplified using primers 63f and 1387r [36].
The 15 ␮l reaction mix contained 7.5 ␮l GoTaq® Green Master Mix

T.M. Coady et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 150–155
153
Table 1
Activity screening towards ˇ-hydroxynitriles.
Isolate
R
Temperature for optimum growth (◦C)
Optimum activity temperature (◦C)
Activitya (mmol/L)
3-Hydroxyglutaronitrile
F41
F30
CH2CN
15
15
20
15
15
20
15
15
25
15
25
25
25
25
25
20
25
25
25
25
6.29
CH2CN
CH2CN
CH2CN
CH2CN
CH2CN
CH2CN
CH2CN
CH2CN
CH2CN
2.150
1.790
1.726
1.468
1.433
1.144
1.140
0.910
0.772
NN30
F32
SS1-4
SS1-24
SET-1
F37
SS1-18
SS1-12
3-Hydroxybutyronitrile
NN32
F71
NN30
SS1-31
LC3c
SS1-12
F69
SET-1
Sw2-33
F36
CH3
15
15
15
20
30
15
20
15
30
20
25
25
20
30
20
25
20
25
20
25
9.91
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
4.389
3.120
2.42
1.521
1.430
1.206
1.021
1.00
0.515
3-Hydroxy-3-phenylpropionitrile
SS1-14
F62
SS1-12
SS1-24
Ph
Ph
Ph
Ph
15
15
15
15
25
25
25
25
8.07
7.64
1.61
0.21
a
Activity determined using Nessler’s colorimetric activity (24 h) assay expressed in terms of concentration (mmol/L) of NH3 produced @ reported optimum temperature
(column 4).
Reaction temperature is a key parameter in biotransformations
of isolates demonstrated their highest activity at 25 ◦C however in
contrast optimum growth was observed at lower temperatures in
many cases. Table 1 indicates the temperature for optimum growth
and also the temperature which resulted in the optimum activity
of the isolate.
which can significantly influence the activity, enantioselectiv-
ity and stability of a biocatalyst [38]. The isolate library was
first screened for the hydrolysis of ˇ-hydroxynitriles to ˇ-
hydroxycarboxylic acids at 25 ◦C and subsequently at various
temperatures (Table 1). Based on the activity assay results the
isolates were then divided into three subcategories based on
the concentration of NH3 produced (0.08–0.6 mmol/L = low activ-
ity, 0.6–5.8 mmol/L = good activity and >5.8 mmol/L = high activity).
Preliminary results revealed that 47 isolates demonstrated
activity towards 3-hydroxybutyronitrile, 34 isolates towards 3-
hydroxyglutaronitrile while only 4 of the initial 256 isolates
catalysed the hydrolysis of 3-hydroxy-3-phenylpropionitrile. This
indicated sensitivity towards the phenyl ring at the b-position
on the enzymatic reaction, and this is demonstrated in the
ability of isolate to SS1-24 to show good activity towards
3-hydroxyglutaronitrile (1.43 mmol/L) where the same isolate
demonstrates low activity (0.21 mmol/L) towards 3-hydroxy-3-
phenylpropionitrile. This enabled the 10 isolates demonstrating
the highest activity towards the target substrates to be selected
for subsequent screening and optimisation.
As the bacterial strains have been isolated from various locations
such as soil and sea weed, these isolates may show different growth
patterns at various temperatures. It was decided also to monitor the
growth of the 24 chosen isolates in the presence of the key nitriles
at various temperatures ranging from 15 to 30 ◦C, and in each case
an optimum temperature for induction was also determined for all
future studies.
No significant influence of reaction temperatures on the activity
of the chosen isolates was observed when the isolates were assayed
at various temperatures ranging from 15 ◦C to 30 ◦C. The majority
This high throughput screening strategy allowed for the
rapid identification of isolates demonstrating activity towards
ˇ-hydroxynitriles. The toxicity testing, nitrogen starvation and
activity screening serve as a preliminary study where active iso-
lates can be identified. The ultimate aim of this work was to
identify novel enantioselective strains. In previous studies strains
were isolated by enrichment with nitriles such as acetonitrile, ben-
zonitrile, adiponitrile and acrylonitrile [29]. Our work attempted
to induce strains directly with the target substrates to deter-
mine if enantioselective enzyme systems could be produced.
Following the preliminary studies reported above the 10 isolates
demonstrating the highest activity towards 3-hydroxybutyronitrile
were chosen for further enantioselectivity screening towards 3-
hydroxybutyronitrile (Table 1) to identify a link between our
activity screening and enantioselectivity, results are presented in
Table 2. This work serves as a demonstration of the application
of this high-throughput screening strategy for the identification of
both active and enantioselective nitrile hydrolysing isolates.
3.2. Screening for enantioselective biotransformations of
3-hydroxybutyronitrile
Having reduced the bacterial isolates from 256 to 24 having good
to high activity towards ˇ-hydroxynitriles, attention was focused
on determining the enantioselectivity of the isolates. Initial enan-
tioselectivity screening involved the biotransformation of racemic

154
T.M. Coady et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 150–155
Table 2
Enantioselective hydrolysis of 3-hydroxybutyronitrile catalysed by novel bacterial isolates.
Entry
Isolate
Optimum temperatureb (◦C)
Nitrile eec (%)
Acid eec (%)
Activityd
Refs.
1a
2a
3a
4a
5a
6a
7a
8a
9a
10a
F36
F69
F71
SS1-12
15
15
15
25
15
15
15
15
15
25
13.9
1.4
12.5
26.2
ND
1.3
5.2
2.3
22.3
16.2
31.6
27.6
2.0
14.0
99.9
25.5
13.1
ND
Low
Good
Good
Good
High
High
High
Low
SET1
Paenibacillus sp. NN32
Burkholderia sp. LC3c
SW2 33, Serratia sp.
Bacillus sp. NN30
SS1-31
[1]
[2]
58.5
18.7
Good
Good
[1]
a
Biotransformation was carried out by incubating 3-hydroxybutyronitrile (10 mmol/L) in a suspension of the named isolate (OD = 1) in phosphate buffer (pH 7.0) for 24 h.
Temperature for optimumenantioselectivity.
Determined by HPLC analysis using a chiral column (see Section 2.8).
b
c
d
Activity determined using Nessler’s colorimetric activity assay (at optimum temperature).
3-hydroxybutyronitrile. The reaction was carried out conveniently
isolated from the reaction mixture using column chromatography
in 41% yield and >99.9% ee. The unreacted (R)-nitrile could not be
recovered from the reaction mixture.
using whole cells of the isolates at 25◦ C in aqueous phosphate
buffer (pH 7.0). The hydrolysis products were transformed to the
corresponding benzyloxy derivatives in situ for ease of detection
using HPLC. As the specific nitrile metabolising enzyme content
of the cells was unknown the isolates were assayed for activity to
produce both amide and the corresponding carboxylic acid. It was
assumed that the cells possessed a nitrilase system as during the
hydrolysis of nitriles with resting cells, no amide could be detected
and, furthermore the amide was not used as a substrate. In all cases
the nitrile was partially consumed during the incubation period,
which is in agreement with the results obtained using the high
throughput screening strategy.
Having identified this very promising isolate SET1 we decided
to further focus our efforts on this isolate and explore the
high enantioselectivity isolate SET1 demonstrated towards 3-
hydroxybutyronitrile. An enantioselectivity vs. time study was
carried out where the biotransformation was monitored after 3,
6, 9, 24 and 36 h; results indicate that isolate SET1 is highly
enantioselective and possibly enantiospecific with >99.9% ee of
(S)-acid at each sample point. It is believed the hydrolysis of
the (S)-nitrile proceeds rapidly, as only the remaining (R)-nitrile
is detected after 3 h. This was further demonstrated by the
formation of only (S)-3-hydroxybutyric acid after a 10 day incu-
bation period. While the reaction afforded the corresponding
(S)-3-hydroxybutyric acid with excellent enantioselectivity, nei-
ther amide nor the starting nitrile could be recovered after this
time. In addition the reduction in intensity of the remaining (R)-
nitrile is also evident after various timepoints, with complete
consumption after 24 h, indicating the potential metabolism of
the unreacted (R)-nitrile along with the desired biotransforma-
tion.
The exponential decay of the unreacted nitrile may be due to
other enzyme systems present within the microbial whole cells.
A possibility may be hydrolysis of the nitrile to the corresponding
acid followed by isomerisation. An investigation into the deracemi-
sation of the racemic 3-hydroxybutyric acid was carried out by
incubation with the isolate over a 24 h period; however unchanged
racemic acid was recovered. Other options include the possibility
of an isomerase converting the remaining (R)-nitrile to (S)-nitrile
which may then be converted to the corresponding (S)-acid, how-
ever the isolated yield never exceeds 50%. Other enzymes such as
an aldoxime or alcohol dehydrogenase may also account for the
loss of mass balance.
This chiral screening study identified a single isolate SET1
which demonstrated exceptional enantioselectivity towards 3-
hydroxybutyronitrile (>99.9% ee), (entry 5, Table 2). The effect of
temperature on the enantioselectivity of the biotransformation was
examined at 15 ◦C and 25 ◦C. Several isolates demonstrated higher
enantioselectivity at lower temperatures, for example an increase
in enantioselectivity was observed in the case of F36 from 21.3%
to 31.6% ee, when the temperature was varied from 25 ◦C to 15 ◦C
(entry 1, Table 2). In the case of isolate SET1 at both temperatures
the corresponding acid was produced in >99.9% ee. In this case only
a slight decrease in activity from 1.02 mmol/L to 0.98 mmol/L was
observed when the temperature was reduced from 25 ◦C to 15 ◦C
respectively.
Isolate SET1 (entry 5, Table 2), fell within the ‘high’ activity
range of the screening scale however it did not have the high-
est activity as determined using the Nesslers colorimetric assay.
In comparison, isolate NN32 (entry 6, Table 2) showed the highest
activity (Table 1) but demonstrated much lower enantiocontrol.
Isolate SET1 also only generated the corresponding (S)-carboxylic
acid, and no amide was detected; thus it is believed that this isolate
possesses a highly enantioselective nitrilase. It is also surprising to
note that in the case of SET-1 the corresponding unreacted (R)-
nitrile could not be detected in the reaction mixture after the 24 h
reaction time.
It has been reported that the enzyme catalysed kinetic
resolution of ˇ-substituted nitriles can demonstrate low enan-
tioselectivity in some cases [39,15,16,17,40,18]. For example
in the biotransformation of 3-hydroxybutyronitrile using whole
To determine yield and configuration of the acid prod-
uct formed using isolate SET1in the hydrolysis of racemic
3-hydroxybutyronitrile, a large scale biotransformation was car-
ried out over 24 h. The ee value of 3-hydroxybutyric acid was
measured by chiral HPLC analysis, and the absolute configura-
tion was determined as (S) by comparison of optical rotation with
an authentic sample [35]. The product 3-hydroxybutyric acid was
cells of Rhodococcus erythropolis AJ270 containing
a nitrile
hydratase/amidase system, low enantiocontrol was observed with
the formation of (S)-3-hydroxybutyric acid (17.8% ee) [34]. The
enantioselectivity of the reaction was dramatically improved using
a benzyl protection/docking strategy which resulted in the for-
mation of the acid with 86.2% enantiomeric excess. In the case

T.M. Coady et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 150–155
155
of SET1 >99.9% ee may be achieved using 3-hydroxybutyronitrile
Appendix A. Supplementary data
with the free unprotected hydroxy group. The high enantios-
electivity demonstrated by isolate SET1 may be as a result of
the position of the chiral recognition site in the enzyme which
is possibly located remote to the catalytic centre [35], this
has also been observed by previous authors [6,34]. While the
biocatalytic hydrolysis of ˇ-hydroxynitriles to corresponding ˇ-
hydroxy acids and amides using microorganisms possessing nitrile
hydratase/amidase have been well reported [41,18,39], in con-
trast microorganisms with nitrilase activity often demonstrate
low or extremely low activity towards the hydrolysis of ˇ-
hydroxynitriles [7,9,18,39]. To the best of our knowledge this is the
first isolate potentially demonstrating enantiospecificity towards
3-hydroxybutyronitrile.
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.molcatb.
2013.08.001.
References
[1] Q. Ren, K. Ruth, L. Thöny-Meyer, M. Zinn, Appl. Microbiol. Biotechnol. 87 (2010)
41–52.
[2] R.A. Sheldon, Chimia 50 (1996) 418–419.
[3] T.U. Onuegbu, E.M.O. OAS, J. Basic Phys. Res. 2 (2011) 86–92.
[4] V. Gotor, Org. Process Res. Dev. 6 (2002) 420–426.
[5] W.A. Loughlin, Bioresour. Technol. 74 (2000) 49–62.
[6] M. Da-You, W. De-Xian, P. Jie, H. Zhi-Tang, W. Mei-Xiang, J. Org. Chem. 73 (2008)
4087–4091.
[7] E.C. Hann, A.E. Sigmund, S.K. Fager, F.B. Cooling, J.E. Gavagan, A. Ben-Bassat,
S. Chauhan, M.S. Payne, S.M. Hennessey, R. DiCosimo, Adv. Synth. Catal. 345
(2003) 775–782.
[8] S. Kamila, D. Zhu, E.R. Biehl, L. Hua, Org. Lett. 8 (2006) 4429–4431.
[9] D. Brady, A. Beeton, J. Zeevaart, C. Kgaje, F. Rantwijk, R.A. Sheldon, Appl. Micro-
biol. Biotechnol. 64 (2004) 76–85.
[10] G. DeSantis, Z. Zhu, W.A. Greenberg, K. Wong, J. Chaplin, S.R. Hanson, B. Farwell,
L.W. Nicholson, C.L. Rand, D.P. Weiner, D.E. Robertson, M.J. Burk, J. Am. Chem.
Soc. 124 (2002) 9024–9025.
[11] C. O‘Reilly, P.D. Turner, J. Appl. Microbiol. 95 (2003) 1161–1174.
[12] A. Kamal, G.B.R. Khanna, R. Ramu, Tetrahedron Asymmetry 13 (2002)
2039–2051.
[13] R.J. Hammond, B.W. Poston, I. Ghiviriga, B.D. Feske, Tetrahedron Lett. 48 (2007)
1217–1219.
[14] S. Bergeron, D.A. Chaplin, J.H. Edwards, B.S.W. Ellis, C.L. Hill, K. Holt-Tiffin, J.R.
Knight, T. Mahoney, A.P. Osborne, G. Ruecroft, Org. Process Res. Dev. 10 (2006)
661–665.
[15] M.L. Gradley, C.J. Knowles, Biotechnol. Lett. 16 (1994) 41–46.
[16] N. Klempier, A. de Raadt, K. Faber, H. Griengl, Tetrahedron Lett. 32 (1991)
341–344.
DNA sequencing of a 432 bp region of the 16s rDNA of isolate
SET1 indicates that it is a strain of Rhodococcus erythropolis. It is
100% identical over the sequenced region to Rhodococcus sp. TMS1-
19 (GenBank acession JX949804) a strain isolated from a glacier in
China.
4. Conclusion
In summary, this work describes the development of a high-
throughput screening strategy for the identification of isolates
with nitrile hydrolysing activity towards ˇ-hydroxynitriles. This
high-throughput method was used to identify a range of iso-
lates demonstrating activity towards several ˇ-hydroxynitriles.
In a subsequent enantioselectiveity screening study of isolates
demonstrating the desired activity towards 3-hydroxybutronitrile,
an isolate demonstrating a highly enantioselective if not enan-
tiospecific activity towards 3-hydroxybutyronitrile was identified.
[17] M. Yokoyama, N. Imai, T. Sugai, H. Ohta, J. Mol. Catal. B Enzym. 1 (1996) 135–141.
[18] Z.-L. Wu, Z.-Y. Li, J. Mol. Catal. B Enzym. 22 (2003) 105–112.
[19] N. Layh, B. Hirrlinger, A. Stolz, H.J. Knackmuss, Appl. Microbiol. Biotechnol. 47
(1997) 668–674.
Hence
a biosynthetic pathway for the production of (S)-3-
[20] L. Martínková, V. Vejvoda, V. Krˇen, J. Biotechnol. 133 (2008) 318–326.
[21] J.-S. Gong, Z.-M. Lu, H. Li, J.-S. Shi, Z.-M. Zhou, Z.-H. Xu, Microb. Cell Factories
11 (2012) 142.
[22] Y.-C. He, C.-L. Ma, J.-H. Xu, L. Zhou, Appl. Microbiol. Biotechnol. 89 (2011)
817–823.
hydroxybutyric acid from racemic 3-hydroxybutyronitrile was
established using this novel Rhodococcus erythropolis strain
SET1. This cost effective screening strategy developed in this
study can be used to screen a large number of microorgan-
isms for their ability to transform aliphatic, aromatic nitriles
[23] Z.-J. Lin, R.-C. Zheng, L.-H. Lei, Y.-G. Zheng, Y.-C. Shen, J. Microbiol. Methods 85
(2011) 214–220.
and dinitriles within
a short time period, in particular the
[24] J.-F. Zhang, Z.-Q. Liu, Y.-G. Zheng, Y.-C. Shen, Eng. Life Sci. 12 (2012) 69–78.
[25] K. Shubhangi, R. Sonali, U.C. Banerjee, Int. J. Pharm. Sci. Technol. 6 (2011) 36–43.
[26] U. Schreiner, B. Hecher, S. Obrowsky, K. Waich, N. Klempier, G. Steinkellner, K.
Gruber, J.D. Rozzell, A. Glieder, M. Winkler, Enzyme Microb. Technol. 47 (2010)
140–146.
[27] D. Snell, J. Colby, Enzyme Microb. Technol. 24 (1999) 160–163.
[28] A.J.J. Straathof, S. Panke, A. Schmid, Curr. Opin. Biotechnol. 13 (2002) 548–556.
[29] L. Coffey, A. Clarke, P. Duggan, K. Tambling, S. Horgan, D. Dowling, C.O. Reilly,
Arch. Microbiol. 191 (2009) 761–771.
[30] L. Coffey, E. Owens, K. Tambling, D. O’Neill, L. O’Connor, C. O’Reilly, Antonie Van
Leeuwenhoek 98 (2010) 455–463.
[31] J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning: A Laboratory Manual,
2nd Ed., Cold Spring Harbor University Press, 1989.
[32] R. O‘Mahony, Waterford Institute of Technology. The molecular analysis of
the structure and regulation of the nitrile hydratase/amidase operons of three
novel Rhodococcal species, 2004.
[33] O. Cahill, Chemical and Life Science, Waterford Institute of Technology, Water-
ford, 2004.
[34] D.-Y. Ma, Q.-Y. Zheng, D.-X. Wang, M.-X. Wang, Org. Lett. 8 (2006) 3231–3234.
[35] D.-Y. Ma, D.-X. Wang, J. Pan, Z.-T. Huang, M.-X. Wang, J. Org. Chem. 73 (2008)
4087–4091.
addition of a toxicity screen towards the various nitriles, fur-
ther focuses the strategy. Further screening techniques can
be applied to candidates to determine their enantioselectiv-
ity; of note in this particular study isolates with higher
enantioselectivity generally did not display the highest activ-
ity, for example SET1 demonstrated moderate activity towards
3-hydroxybutyronitrile, however it had the highest enantiose-
lectivity. This highlights the importance of carefully selecting
the activity specifications of isolates chosen for enantioselec-
tivity screening. Further studies of R. erythropolis SET1 towards
the understanding of its remarkable enantioselectivity and the
substrate scope towards various ˇ-hydroxynitriles, including
the effect of various substituents on the enantioselectivity of
the biotransformations are actively being investigated in our
laboratory.
[36] J.R. Marchesi, T. Sato, A.J. Weightman, T.A. Martin, J.C. Fry, S.J. Hiom, D. Dymock,
W.G. Wade, Appl. Environ. Microbiol. 64 (1998) 795–799.
[37] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, J. Mol. Biol. 215 (1990)
403–410.
Acknowledgements
The authors would like to thank the Wales Ireland Net-
work for Scientific Skills for financial support under the Ireland
Wales 2007–2013 INTERREG 4A programme sanctioned to
the department. We would also like to thank Kilkenny VEC
and Waterford Institute of Technology for additional funding
received.
[38] P.S. Robert, Trends Biotechnol. 14 (1996) 13–16.
[39] M.-X. Wang, Y. Wu, Org. Biomol. Chem. 1 (2003) 535–540.
[40] S.K. Taylor, N.H. Chmiel, L.J. Simons, J.R. Vyvyan, J. Org. Chem. 61 (1996)
9084–9085.
[41] L. Martıꢀnková, N. Klempier, J. Bardakji, A. Kandelbauer, M. Ovesná, T.
Podarˇilová, M. Kuzma, I. Prˇepechalová, H. Griengl, V.R. Kren, J. Mol. Catal. B
Enzym. 14 (2001) 95–99.