Purification, characterization and in-silico analysis of nitrilase from Gordonia terrae

Article abstract of DOI:10.2174/0929866521666140909154537

An inducible and aromatic nitrilase from Gordonia terrae was purified with a yield of 19%. The enzyme had turnover number of 63 s^-1 × 10^-3, K_m1.4 mM and V_max95 Umg^-1 protein for benzonitrile. The nitrilase of G. terrae was active at basic pH (7-10), moderate temperature (20-45 °C) and has a half-life of 4 h at 35 °C. MALDI analysis and amino acid sequence deduced from cloned nucleotide fragment showed 97% homology with putative amidohydrolase of Gordonia sputi NBRC 100414 and G. namibiensis. The enzyme showed regioselectivity towards hydroxybenzonitriles, as different position of hydroxyl group i.e. meta-, para- and orthosubstitutions on benzonitrile effect enzyme activity. The in-silico interactions of these substrates with the predicted 3D model of this enzyme also showed differential interaction between hydroxyl group of substrates and the polar amino acids surrounding enzyme's active site. This leads to different proximity and orientation of substrates vis-a-vis their interaction with catalytic residues.

Full text of DOI:10.2174/0929866521666140909154537

Send Orders for Reprints to reprints@benthamscience.net
Protein & Peptide Letters, 2015, 22, 52-62
5
2
Purification, Characterization and In-silico Analysis of Nitrilase from
Gordonia terrae
ꢀ
ꢁ
ꢀ
ꢀ
ꢀ
Vijay Kumar , Amit Seth , Vijaya Kumari , Virender Kumar and Tek C. Bhalla *
ꢀ
Department of Biotechnology, Himachal Pradesh University, Summer Hill Shimla, Himachal Pradesh-171005, India;
ꢁ
Present address: School of Bioengineering and Food Technology, Shoolini University, Solan, Himachal Pradesh-
1
73212, India
Abstract: An inducible and aromatic nitrilase from Gordonia terrae was purified with a yield of 19%. The en-
-1
-3
-1
m
zyme had turnover number of 63 s x 10 , K 1.4 mM and Vmax 95 Umg protein for benzonitrile. The nitrilase
of G. terrae was active at basic pH (7-10), moderate temperature (20-45 ºC) and has a half-life of 4 h at 35 ºC.
MALDI analysis and amino acid sequence deduced from cloned nucleotide fragment showed 97% homology
with putative amidohydrolase of Gordonia sputi NBRC 100414 and G. namibiensis. The enzyme showed regio-
selectivity towards hydroxybenzonitriles, as different position of hydroxyl group i.e. meta-, para- and ortho-
substitutions on benzonitrile effect enzyme activity. The in-silico interactions of these substrates with the predicted 3D
model of this enzyme also showed differential interaction between hydroxyl group of substrates and the polar amino acids
surrounding enzyme’s active site. This leads to different proximity and orientation of substrates vis-a-vis their interaction
with catalytic residues.
Keywords: Gordonia terrae, in-silico analysis, nitrilase, purification.
Author profile: Dr. Tek Chand Bhalla is Professor in Biotechnology at Himachal Pradesh University, Shimla, India. The thrust
area of his research is biocatalysis and biotransformation. In addition, he is working on protein purification, molecular cloning,
in-silico analysis of various enzymes and developing bioprocesses for the synthesis of industrially important compounds.
INTRODUCTION
chemicals [3-6]. The role of nitriles in chemical synthesis is
inevitable; therefore their enzymatic hydrolysis to non toxic
and high value acids is of active interest. Recently nitrilase
has been used as biocatalyst to synthesize a number of im-
portant organic acids i.e. p-hydroxybenzoic acid [8], nico-
tinic acid [9, 10], mandelic acid [11, 12] and have potential
for many other important bio-transformations [5]. Resting
cells of G. terrae showed higher activity for benzonitrile
followed by meta and para substituted hydroxybenzonitrile,
and for ortho substitution it exhibited negligible activity [8].
The molecular mechanism of such regio-specificity of nitri-
lase remains impaired due to lack of crystal structure of the
nitrilase. Recently crystal structure of thermoactive nitrilase
from Pyrococcus abyssi (PDB id 3IVZ) [13], and some en-
zymes of the nitrilase super family have been reported. Nitri-
lase enzymes from some microorganisms have been purified
and characterized [14-20], and few of them have been mod-
eled using homology modeling in order to understand the
enzyme substrate interaction and substrate specificity [18,
21, 22]. In the present study, nitrilase of G. terrae has been
purified and characterized. Gene encoding for this nitrilase
was cloned, sequenced and a protein model was generated
using homology modeling. On the basis of in-silico interac-
tions of enzyme and substrates (molecular docking), a
mechanism for regio-specificity of G. terrae nitrilase to-
wards positional isomers of hydroxybenzonitriles (ortho,
meta and para-hydroxybenzonitrile) has been predicted.
Nitrilases (EC 3.5.5.1) are important hydrolytic enzymes
which convert nitriles into the corresponding carboxylic ac-
ids and ammonia [1-2]. Nitrile compounds occur in biologi-
cal systems as cyanoglycosides, cyanolipids, ricinine, pheny-
lacetonitrile, etc. [3]. Although most nitriles are highly toxic,
mutagenic, and carcinogenic due to their cyano group, yet
they are widely used in the synthesis of industrially impor-
tant amides and acids [4-6]. Nitrilases have emerged as key
enzymes of potential applications for bioremediation of ni-
trile contaminated soil, waste water and biotransformation of
nitriles to corresponding acids [1-6]. In recent years, nitrilase
mediated biocatalysis has attracted substantial attention and
intensive research has been carried out on basic and applied
aspect in the past three decades [5].
An actinobacterium G. terrae having nitrilase activity
was isolated in our laboratory and used for transformation of
3
-cyanopyridine to nicotinic acid and p-hydroxybenzonitrile
to p-hydroxybenzoic acid [7, 8]. This enzyme converts ben-
zonitrile and its hydroxyl substituted nitriles into correspond-
ing carboxylic acids (Fig. 1). Benzonitrile and hydroxyben-
zonitriles are used for the synthesis of solvents, herbicides
and starting materials for other industrially important
*
Address correspondence to this author at the Department of Biotechnology,
Himachal Pradesh University, Summer Hill Shimla, Himachal Pradesh-
71005, India; Tel: +91-177-2831948, Fax: +91-177-283154;
E-mail: bhallatc@rediffmail.com
1
1
875-5305/15 $58.00+.00
© 2015 Bentham Science Publishers

Purification, Characterization and In-silico Analysis
Protein & Peptide Letters, 2015, Vol. 22, No. 1 53
N
O
OH
X1
H
H
X1
Nitrilase
+
NH
+2HO
3
X4
X2
2
X4
X2
X3
X3
Hydroxybenzoic acid
Ammonia
Hydroxybenzonitrile
1 2 4 3
Figure 1. Nitrilase mediated conversion of hydroxybenzonitrile to hydroxybenzoic acid. X =OH for ortho, X or X =OH for meta, X = OH
for para-hydroxybenzonitrile and corresponding acids. For benzonitrile and benzoic acid, all X are Hydrogen.
TM
MATERIALS AND METHODS
Chemicals
beads (0.1 mm) in bead beater (Model no 1107900 Bio-
spec) upto 6 cycles of disruption. Each cycle comprised of 1
min disruption and 1 min pause. The cell lysate from each
cycle was centrifuged at 10000 g for 30 min. The quantita-
tive assay for nitrilase activity was done with supernatant i.e.
cell free extract (CFE) and pellet (i.e. the aggregate of cell
debris and intact cells) of each sample by ammonia assay
method. Amount of protein in CFE was also quantified by
Bradford assay method.
Pre-packed gel filtration column (HiPrepTM Sephacryl
S-100 HR), ion-exchange column (HiPrepTM DEAE-
Sepharose FF), reagent of polyacrylamide gel electrophoresis
(
PAGE), and agarose gel electrophoresis were purchased
from Sigma. Sodium dodecyl sulphate (SDS), native PAGE
protein molecular size markers and plasmid isolation kit
were procured from Bangalore GeNei, (Merck bioscience).
T4 DNA ligase, Taq DNA polymerase, dNTPs, isopropyl-1-
thio-ꢀ-D-galactopyranoside (IPTG) and X-gal were also ob-
tained from Banglore GeNei, (Merck bioscience). Antibiotic
ampicillin and chloramphenicol were purchased from USB,
Cleveland Ohio, USA. Media ingredients were from Hi-
Media, India and all other chemicals and reagents were of
analytical grade.
Ammonium Sulphate Fractionation
The CFE was subjected to various saturations of ammo-
nium sulphate (10–70% saturation) and the precipitates were
collected by centrifugation and suspended in 0.1 M potas-
sium phosphate buffer (pH 8). The ammonium sulphate frac-
tion (ASF) containing maximum nitrilase activity was proc-
essed further.
Bacterial Strain, Host and Plasmid
Gel Exclusion Chromatography
Culture of G. terrae MTCC 8139 was procured from the
Department of Biotechnology, Himachal Pradesh University
Shimla-171005, India and used as a source for nitrilase. It
was maintained and cultured as described previously [8]. E.
coli DH5ꢁ host strain and pGEM T cloning vector were pur-
chased from Invitrogen, USA.
The ASF was filtered through 0.45 μm filter and loaded
on to pre-packed gel filtration column (S-100) equilibrated
with 0.1 M potassium phosphate buffer (pH 8). The gel fil-
tration chromatography was performed using FPLC system
at a flow rate of 1 ml/min of elution buffer (same as equili-
bration buffer). The fractions exhibiting higher nitrilase ac-
tivity were pooled.
Purification of Nitrilase
The nitrilase of G. terrae was purified using fast per-
formance liquid chromatography (FPLC) system (Akta
Prime, Amersham Biosciences, Uppsala, Sweden). All steps
were carried out at 4 ºC, and 100 mM potassium phosphate
Ion Exchange Chromatography
The concentrated protein solution after size exclusion
chromatography was applied to a DEAE anion exchange
column. After loading protein sample, the column was
washed for 20 min with buffer A (0.1 M potassium phos-
phate buffer pH 8) until there was no further elution of pro-
teins. The enzyme was then eluted with buffer B comprising
buffer A with a linear gradient of NaCl from 0 to 0.5 M.
(
pH 8) containing 1mM dithiothreitol (DTT) was used as
buffer for the purification of this enzyme. The protein con-
tent of all fractions at various stages of purification was es-
timated by Bradford method.
Preparation of Cell Free Extract
Gel Electrophoresis
G. terrae cells grown aerobically at 30 °C for 24 h in
nutrient broth and 2% (v/v) of this precultured medium trans-
ferred to production medium containing (g/L): 5, peptone; 3,
yeast extract; 3, malt extract; 1% (v/v) glycerol, pH 8 and
induced with 0.3% (v/v) isobutyronitrile. After cultivation
for 42 h and 150 rpm at 30 °C, culture was harvested by cen-
trifugation at 10000 g for 10 min. The cell pallet was washed
twice and suspended in 40 ml of 0.1 M potassium phosphate
buffer having pH 8. The cells were disrupted using glass
Gel electrophoresis was performed in 10% SDS-PAGE
and 8% Native PAGE gels with a Tris–Glycine buffer sys-
tem and protein bands in the gels were visualized by silver
staining.
Assay of Nitrilase Activity
The nitrilase assay system contained 0.1 M potassium
phosphate buffer (pH 8), 20 μl protein and 10 mM benzoni-

5
4
Protein & Peptide Letters, 2015, Vol. 22, No. 1
Kumar et al.
trile as substrate in 1 mL reaction. The reaction was carried
out at 35 °C for 1 h and stopped by the addition of an equal
amount of 0.1 M HCl. The amount of ammonia produced in
the reaction was determined by ammonia assay method ac-
cording to Fawcett and Scott (1960). Enzyme activity was
defined as the amount of enzyme required to release 1 ꢀmol
of ammonia per min under assay conditions.
Daltonics) and spectra were interpreted using Brukar
BioTools Version 2.2 (Bruker Daltonik). Homologous pro-
teins were searched in the NCBI database using the BLAST
program [23] (http://www.ncbi.nlm.nih.gov/BLAST/).
Molecular Cloning and Sequencing
Primers were designed based on published literature [24]
and protein sequence obtained from MALDI TOF analysis
with following sequences: forward 5’-ACGCATATGGTCG
AATACACAA3’ and reverse 5’-AAGCTTCGAGTCAG
ATGGAGG3’. The reaction mixtures for PCR contained
1ꢂPCR buffer, each deoxynucleoside triphosphate (dNTP) at
a concentration of 200 μM, each primer at a concentration of
1 μM, DNA templates at a concentration of 1 ng/μl, and 1 U
of Taq DNA polymerase in a final volume of 50 μl. DNA
amplification was performed with a Labnet thermocycler
(Labnet international Inc. USA) using the following pro-
gram: a 5-min hot start at 95 °C, followed by 30 cycles con-
sisting of denaturation (30 s at 94 °C), annealing (60 s at 54
°C), and extending (90 s at 72 °C); and a final extending at
Effect of pH and Temperature on Purified Nitrilase
Effect of pH and temperature on purified nitrilase was
determined by incubating the enzyme with benzonitrile in
different buffer system, pH (6–10), molarity and at various
o
temperatures (20–60 C). The pH stability of the purified ni-
trilase was determined by incubating the enzyme in phos-
phate buffer of different pH (6-8) while the temperature sta-
bility was assessed by incubating the nitrilase at different
temperatures (30, 40 and 50 ºC) in phosphate buffer contain-
ing 1 mM DTT.
Effect of Different Compounds on Purified Nitrilase
7
2 °C for 7 min. Amplified DNA fragment of G. terrae was
+
2+
2+
2+
Influence of various metal ions i.e. Ag Pb , Hg , Mg ,
ligated with T overhang of pGEMT cloning vector using T4
DNA ligase according to manufacturer protocol and cloned
into E. coli DH5ꢁ. Transformed colonies were screened on
LB agar plate having ampicillin with blue white screening
and further verified by colony PCR. For colony PCR a loop-
ful of transformed E. coli cells were boiled in 50 ꢀl Tris-
EDTA buffer and centrifuged at 1000 g for 5 min. Super-
natant was used as DNA template for colony PCR reaction
and all other conditions were same as mentioned above.
Plasmid DNA from transformed culture was isolated and
purified using plasmid isolation kit as per manufacturer’s
protocol. Nucleotide sequencing of nitrilase gene was carried
out by Xceleris Genomic, Ahmadabad, India.
2
+
2+
2+
Ca , Co and Zn , and DTT on the catalytic activity of
nitrilase was determined by pre-incubating the enzyme with
different compounds in 0.1 M phosphate buffer (pH 8) for 30
min.
Substrate Specificity
Specific activity of purified nitrilase of G. terrae towards
some aromatic nitriles i.e. benzonitrile, hydroxybenzonitrile,
aminobenzonitrile, tolunitrile and aliphatic nitriles i.e. aceto-
nitrile, butyronitrile, isobutyronitrile were measured by am-
monia assay method.
Kinetic Parameters of Enzyme
Homology Modeling and Docking
In the kinetic study, initial velocity of the nitrilase cata-
lyzed reaction with various concentrations (0.25–10 mM) of
substrate (benzonitrile, ortho-, meta- and para- hydroxyben-
zonitrile) was measured under the optimized reaction condi-
tions. The kinetic parameters were determined by fitting the
data to the Michalis-Menten equation and Lineweaver Burk
plot.
Nucleotide sequence of this nitrilase was translated into
protein using ExPASy translation tool (http://web.expasy.
org/translate/) and a protein BLAST search was performed
against PDB (Protein Databank) to retrieve the correspond-
ing template for the enzyme [23]. The model was built by
MODELLER9v7 program and it uses an automated approach
to comparative protein structure modeling [25]. Model vali-
dation was done by analyzing its Ramachandran plot and
verified by PROCHECK, in SAVES server [26]. Three di-
mensional structures of different substituted benzonitriles
were prepared using PRODRG server [27] and docked in the
model of nitrilase using the Autodock 4.1 program [28]. The
active site was defined as a collection of amino acid residues
enclosed within a sphere of 5 Å (1 Å=0.1 nm) radius cen-
tered on the catalytic triad of E48, K131 and C165. A ran-
dom start was used to generate the substrate position within
the docking box. The substrate orientation giving the lowest
interaction energy was chosen for additional rounds of dock-
ing.
MALDI TOF Analysis of Purified Protein
Matrix assisted laser desorption ionization-time of flight
(
MALDI-TOF) analysis of the purified nitrilase of G. terrae
was performed at Central Instrumentation Facility, School of
Life Sciences, Jawaharlal Nehru University, New Delhi. Pro-
tein band was excised from silver stained gel, destained with
1
00 mM sodium thiosulphate and 30 mM potassium ferri-
cyanide. The destained band was washed with ammonium
bicarbonate: acetonitrile (1:1 ratio) and incubated in 100%
acetonitrile for 1 h at room temperature for dehydration. The
band was digested with trypsin and incubated at 37 ºC for 16
h. The digested peptides were dried and resuspended in 0.1%
trifluoroacetic acid. This suspension (0.5 mL) was mixed
with same amount of ꢁ-cyano-4-hydroxycinnamic acid (10
mg/mL in 50% acetonitrile, 0.1% trifluoroacetic acid) and
spotted onto a MALDI-target plate for analysis. The peptides
were analyzed by Bruker Autoflex MALDI-TOF (Bruker
RESULTS
Purification of Nitrilase from G. terrae
The nitrilase enzyme was purified from cell free extract
(
CFE) of G. terrae by employing a combination of ammo-

Purification, Characterization and In-silico Analysis
Protein & Peptide Letters, 2015, Vol. 22, No. 1 55
nium sulphate fractionation, gel exclusion and ion exchange
chromatographies. CFE from fifth cycle in bead beater have
highest specific activity and was subjected to ammonium
sulphate precipitation from 10-70% saturation. The desired
protein got precipitated at 25-35% saturation of this salt. The
protein from this saturation was suspended in potassium
phosphate buffer, loaded onto gel permeation column and
eluted under high pressure in FPLC system. Aliquots having
higher nitrilase activity were pooled, loaded in DEAE ion
exchange column and eluted with linear gradient of NaCl
from 0.01-0.5 M concentration. The fractions of desired pro-
tein having nitrilase activity were eluted around 0.20–0.25 M
concentration of NaCl in anion exchange chromatography.
The enzyme was purified to 28 fold with a yield of approxi-
mately 19% with specific activity of 95 Umg proteins for
benzonitrile as substrate in the present study. A summary of
the nitrilase purification scheme is shown in Table 1. The
enzyme preparation after DEAE anion exchange showed
only one band on SDS-PAGE gel electrophoresis having
molecular weight 40 kDa (Fig. 2). Native-PAGE analysis
also showed single band around 40 kDa, therefore, func-
tional form of the enzyme appeared to be monomer.
Characterization of Purified Nitrilase of G. terrae
Effect of pH: The optimum pH stability of purified nitri-
lase of G. terrae was 8; however, it was active within pH 7
to 9.5. The resting cells of G. terrae has exhibited the opti-
mum nitrilase activity at wide pH range from 6 to 10 show-
ing that this enzyme is more stable within the cells.
Effect of temperature: The optimum temperature of G.
terrae nitrilase was observed at 40 °C. Below and above 40
°C, nitrilase activity of G. terrae decreased drastically. Initial
rate of reaction at different temperatures were plotted in Ar-
rhenius plot [ln (k) vs 1/T] yielded a linear profile from
which activation energy of 56.04 kJ/mol was calculated (Fig.
3a). Based on the first-order enzyme deactivation assump-
-
1
tion (d[E]/dt = -k
d
x [E]), the G. terrae deactivation rate co-
-5 -5
efficient k at 25, 35, 45, 55 °C were 1.8x10 , 4.7x10 ,
d
-
4
-4 -1
1.8x10 , 7.3x10 s and its corresponding half-lives t 1/2 (t
/0.693) were 10, 4, 1, 0.3 h, respectively (Fig. 3b).
1/2 = k
d
The nitrilase within the resting cells of G. terrae showed a
longer half life at 35°C and 45 °C, which further indicated
that this enzyme was more stable within the cells.
Effect of different compounds on purified nitrilase: The
effect of various metal ions on the nitrilase activity of G.
+
terrae was studied. Among metal ions tested, Ag caused
2
+
2+
complete inhibition, while Pb and Hg inhibited the nitri-
lase to the extent of 95% and 89% respectively. Some other
2
+
2+
2+
2+
metal ions i.e. Mg , Ca , Co and Zn decreased 10-25%
of nitrilase activity. Dithiothreitol (DTT) enhanced the sta-
bility of enzyme and also resulted in 10% increase in its ac-
tivity.
Substrate specificity: Purified nitrilase of G. terrae
showed specificity towards aromatic nitriles, it showed high-
est activity 95 U/mg protein for benzonitrile and considera-
bly good activity for meta substituted benzonitrile followed
by para-substituted benzonitrile. However, substitution at
meta and para position in aromatic ring relative to cyano
group resulted in lowering nitrilase activity as compared to
benzonitrile. However purified enzyme did not showed any
activity towards ortho- substitutive benzonitrile i.e. ortho-
hydroxybenzonitrile and ortho-aminobenzonitrile (Table 2).
This enzyme had very low activity towards aliphatic nitriles
i.e. acetonitrile, isobutyronitrile and butyronitrile (Table 2).
Figure 2. Purification of nitrilase from G. terrae: Sodium dodecyl
sulphate polyacrylamide gel electrophoresis of purified nitrilase of
wild G. terrae. Protein molecular weight marker (Lane F), ap-
proximately 15μg protein of cell free extract (lane A), 15 μg protein
of the ammonium sulphate fraction (lane B), 15 μg of dialysed pro-
tein, 8 μg protein of Sephacryl S-300 HR column chromatography
fraction (lane D), 6 μg protein of DEAE column chromatography
fraction (lane E) were applied to the gel.
Kinetic parameters: The ability of the purified nitrilase of
G. terrae to catalyze the hydrolysis of benzonitrile and ortho,
meta and para-hydroxybenzonitrile were studied (Table 3).
This enzyme exhibited high specific activity (95 U/mg) and
m
low K (1.4 mM) towards benzonitrile and for its hydroxyl
Table 1. Purification table for nitrilase of G. terrae
Purification steps
Protein
Total Protein
(mg)
Specific activ-
ity (U/mg)
Total activity
Yield (%)
Purification fold
(mg/ml)
Cell free extract
1.67
2.86
165.9
22.92
3.804
1.176
3.38
561
448
186
112
100
79
1
(
NH
4
)SO
Sephacryl-100
DEAE anion exchange
4
fraction (25-35%)
19.54
49.07
95.24
5.78
14.51
28.18
0.19
33
0.059
19

5
6
Protein & Peptide Letters, 2015, Vol. 22, No. 1
a)
Kumar et al.
(
9
8
7
6
5
4
3
2
1
0
̊
55 C
̊
5 C
̊
5 C
̊
5 C
4
3
2
0
1
2
3
4
5
6
Time (h)
(b)
3
2
1
0
.5
y = -6884.5x + 23.599
R² = 0.9752
2
.5
1
.5
0
0.003
0.0031
0.0032
0.0033
0.0034
1
/Temprature (1/K)
Figure 3. (a) Thermostability of G. terrae nitrilase: biotransformation were performed in 0.1 M pH 7.5 potassium phosphate buffer (b) Ef-
fect of temperature on enzyme activity-Arrhenius plot for the determination of activation energy of nitrilase (Initial rate of reaction was plot-
ted against temperature and the activation energy (Ea)= slopeꢁR, was calculated; where R=8.314 J/ K/mol).
substitution its activity gets decreased while K
order of meta and para position of OH on benzonitrile. For
meta-hydroxybenzonitrile its Vmax and K was 36 U/mg and
.5 mM, in case of para-hydroxybenzonitrile kinetic con-
stant were 20U/mg and 2.5 mM respectively, while for ortho
substitution it showed no activity. This indicated that the
presence of any other group like OH on benzene ring of ben-
zonitrile results in hindrance of proper interaction between
enzyme and substrate with maximum obstruction caused to
substrate with ortho substitution.
m
increases in
ampicillin resistance, blue white colony selection and colony
PCR. Recombinant plasmids were isolated, purified and se-
quenced. Nucleotide sequence obtained was 1101 nucleotide
long, GC rich (62.9%) and code for 366 amino acids. Nu-
cleotide sequence of this nitrilase was deposited in NCBI
with accession number KF177346.
m
1
Analysis of MALDI Fragment and Amino Acid Sequence
Peptide sequences obtained from MALDI analysis
showed 100% similarity with nitrilase protein deduced from
nucleotide sequence. Amino acid sequence deduced from the
gene showed 99% homology with R. rhodochrous tga1 and
Nocardia globerula NHB2 but they significantly differ in
their substrate affinity. Nitrilase from these organisms
showed different affinity towards hydroxybenzonitrile, N-
substituted aromatic nitriles and acrylonitrile (Table 4).
BLAST results also showed 97% homology with hypotheti-
Molecular Cloning and Sequencing
PCR reaction of genomic DNA results in amplification of
1
.2 kb DNA fragment which was analysed in 1% agarose
gel. This fragment was purified, ligated into pGEM-T by T4
DNA ligase and recombinant plasmid was transformed into
E. coli DH5ꢀ. Screening of recombinants was carried out by

Purification, Characterization and In-silico Analysis
Protein & Peptide Letters, 2015, Vol. 22, No. 1 57
Table 2. Substrate specificity of nitrilase towards different aromatic and aliphatic nitriles
Substrate
Specific activity (U/mg protein)
Relative activity
Benzonitrile
m-Hydroxybenzonitrile
p-Hydroxybenzonitrile
o-Hydroxybenzonitrile
o-Aminobenzonitrile
m-Aminobenzonitrile
p-Aminobenzonitrile
m-Tolunitrile
95
36
20
0
100
38
21
0
0
0
40
19
22
18
12
3
42
20
23
19
13
3
p-Tolunitrile
4
-Hydroxy-3-methoxybenzonitrile
Acetonitrile
Isobutyronitrile
2
2
Butyronitrile
5
5
Phenylacetonitrile
0
0
Table 3. Kinetic parameter of purified nitrilase of G. terrae
Substrate
Resting cells
Purified Enzyme
K
m
V
max
K
m
V
max
K
cat
K
cat/K
-1
m
-
1
-1
-1
-1
-3
-1
-3
(
mM)
(U mg dcw)
(mM)
(ꢀmole min mg )
( s x 10 )
(s mM x 10 )
Benzonitrile
5
8
1.2
0.37
0.29
0
1.4
1.5
2.5
0
95
36
20
0
63
24
13
0
45
16
5
m-Hydroxybenzonitrile
p-Hydroxybenzonitrile
o-Hydroxybenzonitrile
11
0
0
Table 4. Substrate specificity of nitrilase from G. terrae, N. globerula and R. rhodochrous tga1 towards different nitriles
BZN
2-HBN
3-HBN
4-HBN
2-CP
3-CP
4-CP
ARN
References
G. terrae
1.41
1.7
-
nd
nd
-
0.34
0.14
-
0.21
0.09
-
nd
0.016
-
1.17
1.62
2.2
1.64
1.32
-
1.22
1.1
This study
N. globerula NHB2
R. rhodochrous tga1
Sharma et al., 2011
Luo et al., 2010
5.5
BZN = benzonitrile, HBN = hydroxybenzonitrile, CP = cyanopyridine, ARN = acrylonitrile, nd = not detected
cal amidohydrolase from Gordonia sputi NBRC 100414
WP005202355.1) and putative nitrilase of Gordonia na-
mibiensis NBRC 108229 (WP006868295.1).
PDB blast revealed that this template has query coverage of
75% and an identity of 25% with the query. The generated
structure was verified by PROCHECK, in SAVES server
(
[
26] and by the DOPE (discrete optimized potential energy)
Homology Modeling and Docking
score of MODELLER output. Ramachandran plot of gener-
ated model showed 81% residue in favourable region. The
refined model was docked with substrates: benzonitrile, or-
tho, meta and para-hydroxybenzonitrile, 100 replicas of each
substrate were generated and randomly distributed around
A homology model of this nitrilase was predicted from
the coordinates (Protein Data Bank) of thermoactive nitrilase
of P. abyssi using MODELLER9v7 program [25] (Fig. 4a).

5
8
Protein & Peptide Letters, 2015, Vol. 22, No. 1
a)
Kumar et al.
(
(b)
Figure 4. (a) Predicted three dimensional structure of nitrilase of G. terrae (b) Catalytic triad (Glu 48, Lys 131, Cys 165).
the centre of active site i.e. Glu 48, Lys 131 and Cys 164
DISCUSSION
(
EKC) (Fig. 4b). Docking of ortho, meta and para-
hydroxybenzonitrile as substrates with modeled nitrilase
revealed that hydroxyl group of substituted benzonitrile in-
teracts with polar amino acids i.e. Tyr 54, Tyr 56, Thr135,
Glu138 at vicinity of active pocket (EKC). These amino ac-
ids hindered the proximity and orientation of substrate inter-
action with catalytic triad. OH group at ortho position in the
ring face hindrance from Lys 131, and also seems to face a
strong pull from polar residue i.e. Thr 135 and Glu 138 (Fig.
Nitrilase of G. terrae has been purified to 28 fold and
characterized. SDS and native PAGE analysis have revealed
that this protein is monomer of 40 kD size. Most of the puri-
fied nitrilases have molecular weight around 40 kD and are
multimeric [29], however nitrilase from R. rhodochrous
PA34 [15], Arthrobacter sp. J1 [30], Pseudomonas sp. S1
[
31], are monomeric. This enzyme has optimal pH at 8
whereas the optimum pH of previously reported nitrilases
varied from 6 to 9 and most of them have pH optimum
around pH 7 [5, 28]. The pH below 6 and above 10 drasti-
cally lowered the activity of purified nitrilase and it may be
due to structural change in the enzyme since a change in pH
affect the ionization state of side chain of amino acid that are
involved in the maintenance of proper folding of protein
needed for its catalytic activity. This nitrilase showed maxi-
mum activity at 35 °C indicates its mesophilic nature. Gen-
erally the mesophilic nitrilases have exhibited maximum
activity near ambient temperature between 30 and 40 °C
5
). This interaction increases the distance between Cys164
and CN group of ortho-hydroxybenzonitrile. Purified en-
zyme did not show any activity towards this hydroxyben-
zonitrile. However, with meta and para-hydroxybenzonitrile,
there was no hindrance between Lys 131 and OH group of
substrate, but still there was a pull from the adjacent amino
acid residues which lowered the activity as compared with
benzonitrile. Distance between the active site Cys164 and
CN group of substrate were 3.05, 3.11, 4.49 Å using meta,
para and ortho-hydroxybenzonitrile respectively whereas for
benzonitrile this distance was 2.93 Å (Fig. 5).
[
29]. However, thermophilic nitrilase of P. abyssi exhibited

Purification, Characterization and In-silico Analysis
Protein & Peptide Letters, 2015, Vol. 22, No. 1 59
(
a)
(b)
(c)

6
0
Protein & Peptide Letters, 2015, Vol. 22, No. 1
Kumar et al.
(d)
Figure 5. Docking results showing (a) OH group of o-hydroxybenzonitrile interact with Thr 135, Glu 138 and it face a strong stearic hin-
drance from Lys 131, whereas (b) m-hydroxybenzonitrile and (c) p-hydroxybenzonitrile did not encountered the stearic hindrance, they only
feel an attraction from surrounding amino acids i.e. Thr 135, Glu 138, Tyr 54 and Tyr56, therefore have lower activity as compared to (d)
benzonitrile. Distance between the SH group of cysteine and CN group of respective nitrile were (a) 4.49 Å (b) 3.05 Å (c) 3.11 Å and (d)
2
.93 Å.
maximum activity at 60 °C [13]. The low activation energy
of G. terrae nitrilase (56.04 kJ/mol) implied the faster rate of
hydrolysis of benzonitrile to benzoic acid. P. putida nitrilase
has also shown such low activation energy (49.07 kJ/mol)
and have quick conversion of mandelonitrile to mandelic
bent whereas R. rhodochrous PA34, Pseudomonas sp. S1,
Acidovorax facilis 72W had preference towards aliphatic
nitriles [29]. Gene coding for this nitrilase has been cloned,
sequenced and was found to be 1101 nucleotide in length
with high GC content which is a characteristic feature of
actinobacteria [35]. Amino acid sequence of G. terrae
showed 99% homology with R. rhodochrous tga1 and No-
cardia globerulla NHB2 with difference of two amino acids
from NHB2 and one amino acid from R. rhodochrous tga1.
BLAST results also showed 97% homology with hypotheti-
cal amidohydrolase from Gordonia sputi NBRC 100414
+
acid [17]. Metal ions Ag caused complete inhibition, while
2
+
2+
Pb and Hg also inhibited the activity of purified enzyme,
this meant that cysteine is present in the active site of the
enzyme and these results supported the earlier findings. Ni-
trilase from P. putida [17] rapidly lost activity when stored
in phosphate buffer having 1mM DTT. However, nitrilase of
(
WP005202355.1) and putative nitrilase of Gordonia na-
G. terrae was more stable in phosphate buffer (pH 8) having
mibiensis NBRC 108229 (WP006868295.1), but none of
these have been characterized and purified. Most of the ear-
lier reported nitrilases were of the similar size and have the
conserved catalytic (ECK) triad [29]. Peptide fragments ob-
tained from MALDI TOF analysis confirm that purified pro-
tein was nitrilase and have 100% identity with translated
sequence. Previously we have reported that this enzyme
showed higher activity towards meta-substituted hydroxy-
benzonitrile followed by para-substitution, and for ortho
substituted hydroxybenzonitrile it showed very less activity
1
mM DTT as compared to buffer without DTT at 4 °C. Re-
ducing agent i.e. DTT, ꢀ-mercaptoethanol, etc. maintain re-
ducing environment and break the disulphide bonds between
two cysteine residues. Nitrilase enzyme of G. terrae was
found to be monomer so inter-chain disulphide bonds were
not necessary for the enzymes stability but reducing envi-
ronment restricts the Cys residue for participating in the di-
sulphide bond formation thereby maintaining its activity and
stability in DTT. The purified enzyme exhibited a specific
activity of 95 U/mg for benzonitrile which is comparably
higher to the previously reported nitrilases from other
sources: R. rhodochrous PA34, 3.52 U/mg [15], R. rhodo-
chrous J1, 15.9 U/mg [32], mutant of R. rhodochrous ATCC
[
8]. This trend was similar with purified enzyme, but for
ortho-substituted benzonitrile it showed no activity. This
enzyme showed higher activity for benzonitrile followed by
meta and para-hydroxybenzonitrile. In order to find out the
possible explanation for this regioselectivity of this nitrilase,
a three dimensional structure of this enzyme has been pre-
dicted using thermoactive nitrilase of P. abyssi (3IVZ) [13]
as template. The accuracy of the model is limited as it is
based on homology modelling and not a crystal structure. To
confirm our molecular modelling and structural analysis, the
crystal structure of nitrilase of G. terrae needs to be deter-
mined. Earlier Kim et al. (2009) have used hypothetical pro-
3
3278, 24.4 U/mg [21] and comparable with Rhodococcus
sp., 84 U/mg [33], Aspergillus niger, 91.6 U/mg [34]. K of
m
this enzyme was 1.4 mM, comparable to that of Rhodococ-
cus spp. i.e. R. rhodochrous ATCC 33278, 1.068 mM [21]
and Rhodococcus sp., 3.8 mM [33]. This enzyme showed
clear cut preference for aromatic nitriles i.e. benzonitrile
over the aliphatic, thus can be categorized as an aromatic
nitrilase. Nitrilase from Nocardia sp., A. niger, R. rhodo-
chrous ATCC 33278, Arthrobacter sp. J1 showed aromatic

Purification, Characterization and In-silico Analysis
Protein & Peptide Letters, 2015, Vol. 22, No. 1 61
tein of Pyrococcus horikoshii (1J31) [18] and Yeom et al.
REFERENCES
(
(
2008) employed N-carbamyl-D-amino acid amidohydrolase
1ERZ) and (1FO6) as template [21]. Nitrilase from P.
[
1]
Vesela, A.B.; Franc, M.; Pelantova, H.; Kubac, D.; Vejvoda, V.;
Sulc, M.; Bhalla, T.C.; Mackova, M.; Loveckaa, P.; Janu, P.;
Demnerova, K.; Martinkova, L. Hydrolysis of benzonitrile
herbicides by soil actinobacteria and metabolite toxicity.
Biodegradation, 2010, 2, 761-770.
Banerjee, A.; Sharma, R.; Banerjee, U.C. Nitrile degrading en-
zymes: current status and future prospects. Appl. Microbiol. Bio-
technol., 2002, 60, 33-44.
abyssi was preferred template due to higher query coverage
and homology with nitrilase of G. terrae. Docking studies
with modeled nitrilase revealed that the observed regio-
specificity of enzyme towards substituted benzonitriles is
due to the interaction of OH group of substrate with some
polar amino acids i.e. Thr135, Glu138, Tyr 54, Tyr 56 in the
vicinity of the active site hindering the proximity and orien-
tation of substrate with catalytic triad. Due to the ortho posi-
tion of hydroxyl group, Lys 131 cause steric hindrances in
addition to the pull caused by above mentioned residues
which results in no activity toward ortho-hydroxyben-
zonitrile. Nitrilase of R. rhodochrous ATCC 33278 also have
similar kind of trend toward ortho, meta and para substituted
benzonitriles [21, 22]. Docking and site directed mutagenesis
study on this enzyme has earlier revealed that a positively
charged amino acid is necessary at position 129 for enzyme
activity towards meta- substituted benzonitriles [21]. Nitri-
lase enzyme of P. fluorescens Pf-5 has been reported to have
higher activity for dinitriles as compared to mononitrile as
studied with in-silico substrate docking into model protein
[
2]
[3]
O’Reilly, C.; Turner, P.D. The nitrilase family of CN hydrolysing
enzymes-a comparative study. J. Appl. Microbiol., 2003, 95, 1161-
1
174.
[
4]
Martinkova, L.; Kren, V. Biotransformations with nitrilases. Curr.
Opinion Chem. Biol., 2010, 14, 130-137.
Gong, J.S.; Lu, Z.M.; Li, H.; Shi, J.S.; Zhou, Z.M.; Xu, Z.H.
Nitrilases in nitrile biocatalysis: recent progress and forthcoming
research. Microb. Cell Fact., 2012, 11, 142-148.
Chen, J.; Zheng, R.C.; Zheng, Y.G.; Shen, Y.C.ꢀ Microbial
transformation of nitriles to high-value acids or amides. Adv. Bio-
chem. Engin. Biotechnol., 2009, 113, 33-77.
Prasad, S.; Misra, A.E.; Jangir, V.P.; Awasthi, A., Raj, J.; Bhalla,
T.C. A propionitrile-induced nitrilase of Rhodococcus sp. NDB
[5]
[6]
[7]
1
165 and its application in nicotinic acid synthesis. World J. Mi-
crob. Biotechnol., 2007, 23, 345-353.
[
8]
9]
Kumar, V.; Bhalla, T.C. Transformation of p-hydroxybenzonitrile
to p- hydroxybenzoic acid using nitrilase activity of Gordonia
terrae. Biocat. Biotransform., 2013, 31, 42-48.
Sharma, N.N.; Sharma, M.; Bhalla, T.C. An improved nitrilase-
mediated bioprocess for synthesis of nicotinic acid from 3-
cyanopyridine with hyperinduced Nocardia globerula NHB-2. J.
Ind. Microbiol. Biotechnol., 2011, 38, 235-1243.
Pai, O.; Banoth, L.; Ghosh, S.; Chisti, Y.; Banerjee, U.C. Biotrans-
formation of 3-cyanopyridine to nicotinic acid by free and immobi-
lized cells of recombinant Escherichia coli. Process Biochem.,
[
[
18]. The predicted 3D model of this enzyme showed prefer-
ence for benzonitrile followed by meta- and para-
hydroxybenzonitrile, which was further strengthened by the
data obtained from wet lab experiments. In this nitrilase
some polar amino acid residues i.e. Thr135, Glu138, Tyr 54,
Tyr56 were found to be mainly responsible for its substrate
affinity towards benzonitriles and various hydroxybenzoni-
trile. Site directed mutagenesis at these positions may en-
hance the activity and affinity towards hydroxybenzonitriles.
[10]
2
014, 49, 655-659.
[
11]
12]
Bhatia, S.K.; Mehta, P.K.; Bhatia, R.K. Bhatia, T.C. Optimization
of arylacetonitrilase production from Alcaligenes sp. MTCC 10675
and its application in mandelic acid synthesis. Appl. Microbiol.
Biotechnol., 2013, 98, 83-94.
Liu, Z.Q.; Zhang, X.H.; Xue, Y.P.; Xu, M.; Zheng, Y.G. Improve-
ment of Alcaligenes faecalis nitrilase by gene site saturation
mutagenesis and its application in stereospecific biosynthesis of
[
CONCLUSIONS
Nitrilase of G. terrae exists as a monomer and is an aro-
matic nitrilase. Despite the high similarity with earlier re-
ported nitrilase, this enzyme showed very high affinity to-
wards hydroxybenzonitriles. This enzyme showed higher
activity for benzonitrile followed by meta and para-
hydroxybenzonitrile. Docking studies with modelled nitrilase
revealed that the observed regio-selectivity of enzyme to-
wards substituted benzonitriles is due to the interaction of
OH group of substrate with some polar amino acids around
catalytic triad and stearic hindrance caused by position spe-
cific substitution.
(
R)-(ꢀ)-mandelic acid. J. Agric. Food Chem., 2014, 62, 4685-4694.
[13]
Raczynska, J.E.; Vorgias, C.E.; Antranikian, G.; Rypniewski, W.
Crystallographic analysis of a thermoactive nitrilase. J. Struct.
Biol., 2011, 173, 294-302.
[
14]
Kobayashi, M.; Komeda, H.; Yanaka, N.; Nagasawa, T.; Yamada,
H. Nitrilase from Rhodococcus rhodochrous J1, sequencing and
over-expression of the gene and identification of an essential
cysteine residue. J. Biol. Chem., 1992, 267, 20746-20751.
Bhalla, T.C.; Miura, A.; Wakamoto, A.; Ohba, Y.; Furuhasi, K.;
Asymmteric hydrolysis of ꢁ-aminonitriles to optically active
aminoacids by a nitrilase of Rhodococcus rhodochrous PA-34.
Appl. Microbiol. Biotechnol., 1992, 37, 184-190.
Chauhan, S.; Wu, S.; Blumerman, W.S.; Fallon, R.D.; Gavagan,
J.E.; DiCosimo, R.; Payne, M.S. Purification, cloning, sequencing
and over-expression in Escherichia coli of a regioselective aliphatic
nitrilase from Acidovorax facilis 72W. Appl. Microbiol. Biotech-
nol., 2003, 61, 118-122.
[15]
[
16]
17]
CONFLICT OF INTEREST
The authors confirm that this article content has no con-
flicts of interest.
[
Banerjee, A.; Kaul, P.; Banerjee, U.C. Purification and
characterization of an enantioselective arylacetonitrilase from
Pseudomonas putida. Arch. Microbiol., 2006, 184, 407-418.
Kim, J.S.; Tiwari, M.K.; Ramu, T.; Oh, D.K.; Kim, I.W., Lee, J.K.
ACKNOWLEDGEMENTS
[18]
Identification and characterization of
Pseudomonas fluorescens Pf-5. Appl. Microbiol. Biotechnol., 2009,
2, 273-283.
a novel nitrilase from
We acknowledge University Grants Commission (UGC)
and Council of Scientific and Industrial Research (CSIR)
New Delhi, India, for financial support in the form of SRF to
Vijay Kumar and Amit Seth respectively and UGC for finan-
cial support in the form of JRF to Vijaya Kumari and Viren-
der Kumar. We are also thankful to computational facility at
Sub Distributed Information Centre, Himachal Pradesh Uni-
versity, Shimla-171005, India.
8
[
19]
20]
Petrickova, A.; Vesela, A.B.; Kaplan, O.; Kubac, D.; Uhnakova,
B.; Malandra, A.; Felsberg, J.; Rinagelova A.; Weyrauch P.; Kren,
V.; Bezouska, K.; Martinkova, L. Purification and characterization
of heterologously expressed nitrilases from filamentous fungi.
Appl. Microbiol. Biotechnol., 2012, 93, 1553-1561.
Bhatia, S.K.; Mehta, P.K; Bhatia, R.K.; Bhalla, T.C.ꢀPurification
and characterization of arylacetonitrileꢀspecific nitrilase of
Alcaligenes sp. MTCC 10675. Biotechnol. Appl. Biochem., 2014,
doi: 10.1002/bab.1192 [Epub ahead of print].
[

6
2
Protein & Peptide Letters, 2015, Vol. 22, No. 1
Kumar et al.
[
21]
Yeom, S.J.; Kim, H.J.; Lee, J.K.; Kim, D.E.; Oh, D.K. An amino
acid at position 142 in nitrilase from Rhodococcus rhodochrous
ATCC 33278 determines the substrate specificity for aliphatic and
aromatic nitriles. Biochem. J., 2008, 415, 401-407.
Yeom, S.J.; Lee, J.K.; Oh, D.K. A positively charged amino acid at
position 129 in nitrilase from Rhodococcus rhodochrous ATCC
[29]
[30]
Thuku, R.N.; Brady, D.; Benedik, M.J.; Sewell, B.T. Microbial
nitrilases: versatile, spiral forming, industrial enzymes. J. Appl. Mi-
crobiol., 2009, 106, 703-727.
Bandyopadhyay, A.K.; Nagasawa, T.; Asano, Y.; Fujishiro, K.;
Tani, Y.; Yamada, H. Purification and characterization of
benzonitrilases from Arthrobacter sp. strain J1. Appl. Env. Micro-
biol., 1986, 51, 302-306.
[22]
3
3278 is an essential residue for the activity with meta-substituted
benzonitriles. FEBS Lett., 2010, 584, 106-110.
[31]
[32]
[33]
Dhillon, J.K.; Shivaraman, N. Biodegradation of cyanide
compounds by a Pseudomonas species S1. Can. J. Microbiol.,
1999, 45, 201-208.
Kobayashi, M.; Nagasawa, T.; Yamada, H. Nitrilase of
Rhodococcus rhodochrous J1: Purification and characterization.
Eur. J. Biochem., 1989, 182, 349-356.
Stevenson, D.E., Feng, R.; Dumas, F.; Groleau, D.; Mihoc, A.;
Storer, A.C. Mechanistic and structural studies on Rhodococcus
ATCC 39484 nitrilase. Biotechnol. Appl. Biochem., 1992, 15, 283-
302.
[
23]
24]
Altschul, S.F.; Gish, W.; Miller, W.; Myers, E.W.; Lipman, D.J.
Basic local alignment search tool. J. Mol. Biol., 1990, 215, 403-
4
10.
[
Luo, H.; Fan, L.; Chang, Y.; Ma, J.; Yu, H.; Shen, H.Z. Gene
cloning, overexpression, and characterization of the nitrilase from
Rhodococcus rhodochrous tg1-A6 in E. coli. Appl. Biochem. Bio-
technol., 2010, 60, 393-400.
Eswar, N.; Eramian, D.; Webb, B.; Shen, M.Y.; Sali, A. Protein
structure modelling with MODELLER. Methods Mol. Biol., 2008,
[25]
[26]
[27]
[28]
4
26, 145-159.
[34]
[35]
Kaplan, O.; Vejvoda, V.; Plihal, O.; Pompach, P.; Kavan, D.;
Bojarova, P.; Bezouska, K.; Mackova, M.; Cantarella, M.; Jirku,
V.; Kren, V.; Martinkova, L. Purification and characterization of a
nitrilase from Aspergillus niger K10. Appl. Microbiol. Biotech-
nol., 2006, 73, 567-575.
Ventura, M.; Canchaya, C.; Tauch, A.; Chandra, G.; Fitzgerald,
G.F.; Chater, K.F.; Sinderen, V. Genomics of Actinobacteria:
tracing the evolutionary history of an ancient phylum. Microbiol.
Mol. Biol. Rev., 2007, 71, 495-548.
Laskowski, R.A.; MacArthur, M.W.; Moss, D.S.; Thornton, J.M. A
program to check the stereo-chemical quality of protein structures.
J. Appl. Cryst., 1993, 26, 283-291.
Schuttelkopf, A.W.; VanAalten, D.M.F. PRODRG: a tool for high-
throughput crystallography of protein-ligand complexes. Acta
Crystallogr., 2004, 60, 1355-1363.
Morris, G.M.; Goodsell, D.S.; Halliday, R.S.; Huey, R.; Hart,
W.E.; Belew, R.K.; Olson, A.J. Automated docking using a
lamarckian genetic algorithm and an empirical binding free energy
function. J. Comp. Chem., 1998, 19, 1639-1662.
Received: June 14, 2014
Revised: July 18, 2014
Accepted: August 26, 2014