Enzymatic production of key intermediate of gabapentin by recombinant amidase from Pantoea sp. with high ratio of substrate to biocatalyst

Article abstract of DOI:10.1016/j.procbio.2016.02.012

1-Cyanocyclohexaneacetic acid is the key intermediate of gabapentin. A novel bioprocess catalyzed by amidase was developed for efficient production of 1-cyanocyclohexaneacetic acid from 1-cyanocyclohexaneacetamide, which can be prepared with high efficiency by nitrile hydratase-catalyzed regioselective hydration of 1-cyanocyclohexaneacetonitrile. Kinetic analysis and molecular docking of three recombinant amidase demonstrated that amidase (Pa-Ami) from Pantoea sp. was the most robust biocatalyst for hydrolysis of 1-cyanocyclohexaneacetamide with the k_cat/K_m value of 208.2 ± 16.2 mM^-1 s^-1. Some key parameters of the bioprocess, such as substrate loading, catalyst loading and product inhibition, were investigated. Enzymatic hydrolysis of 80 g/L of 1-cyanocyclohexaneacetamide was completed within 20 min using 1 g/L wet whole cells of recombinant Escherichia coli BL21, leading to high ratio of substrate to catalyst (S/C-ratio, 80) and high space-time yield (5794.7 g_product L^-1 d^-1). These encouraging results indicated the great potential of Pa-Ami in practical production of gabapentin.

Full text of DOI:10.1016/j.procbio.2016.02.012

Accepted Manuscript
Title: Enzymatic production of key intermediate of gabapentin
by recombinant amidase from Pantoea sp. with high ratio of
substrate to biocatalyst
Author: Zhe-Ming Wu Ren-Chao Zheng Xu Ding Jian-Qiang
Jin Yu-Guo Zheng
PII:
S1359-5113(16)30025-3
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.procbio.2016.02.012
PRBI 10626
To appear in:
Process Biochemistry
Received date:
Revised date:
Accepted date:
26-12-2015
6-2-2016
23-2-2016
Please cite this article as: Wu Zhe-Ming, Zheng Ren-Chao, Ding Xu, Jin Jian-
Qiang, Zheng Yu-Guo.Enzymatic production of key intermediate of gabapentin by
recombinant amidase from Pantoea sp.with high ratio of substrate to biocatalyst.Process
Biochemistry http://dx.doi.org/10.1016/j.procbio.2016.02.012
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Enzymatic production of key intermediate of gabapentin by recombinant
amidase from Pantoea sp. with high ratio of substrate to biocatalyst
Authors and Affiliation:
Zhe-Ming Wu , Ren-Chao Zheng , Xu Ding , Jian-Qiang Jin and Yu-Guo Zhenga,b*
aKey Laboratory of Bioorganic Synthesis of Zhejiang Province, College of Biotechnology and
Bioengineering, Zhejiang University of Technology, Hangzhou 310014, P. R. China
bEngineering Research Center of Bioconversion and Biopurification of Ministry of Education, Zhejiang
University of Technology, Hangzhou 310014, P. R. China
a,b
a,b
a,b
a,b
*
Corresponding author:
Tel: +86-571-88320630, Fax: +86-571-88320630, E-mail: zhengyg@zjut.edu.cn
Graphical abstract
CONH2
COOH
NC
NC
Pa-Am i
Excellent amidase activity
High ratio of substrate to catalyst
1
-CCH AM
0 g/L
1-CCH AA
00% conversion
8
High space-tim e yield
1
1

Highlights:
1
2
3
4
5
. A novel bioprocess for preparing 1-cyanocyclohexaneacetic acid was developed.
. Three amidases were investigated to hydrolyze 1-cyanocyclohexaneacetamide.
. Pa-Ami exhibited high activity (297.6 U/mg) and catalytic efficiency.
-1
-1
. Pa-Ami-catalyzed process afforded a space-time yield of 5794.7 gproduct L d .
. Pa-Ami was inhibited by 1-cyanocyclohexaneacetic acid.
Abstract: 1-Cyanocyclohexaneacetic acid is the key intermediate of gabapentin. A
novel bioprocess catalyzed by amidase was developed for efficient production of
1
-cyanocyclohexaneacetic acid from 1-cyanocyclohexaneacetamide, which can be
prepared with high efficiency by nitrile hydratase-catalyzed regioselective hydration
of 1-cyanocyclohexaneacetonitrile. Kinetic analysis and molecular docking of three
recombinant amidase demonstrated that amidase (Pa-Ami) from Pantoea sp. was the
most robust biocatalyst for hydrolysis of 1-cyanocyclohexaneacetamide with the
-1 -1
k
cat/K value of 208.2±16.2 mM s . Some key parameters of the bioprocess, such as
m
substrate loading, catalyst loading and product inhibition, were investigated.
Enzymatic hydrolysis of 80 g/L of 1-cyanocyclohexaneacetamide was completed
within 20 min using 1 g/L wet whole cells of recombinant Escherichia coli BL21,
leading to high ratio of substrate to catalyst (S/C-ratio, 80) and high space-time yield
-1
-1
(
5794.7 gproduct L d ). These encouraging results indicated the great potential of
Pa-Ami in practical production of gabapentin.
Keywords: Amidase, Gabapentin, Whole-cell biocatalysis, High space-time yield,
Hydrolysis
2

1
. Introduction
Gabapentin, initially known as antiepileptic drug (AED) in 1994, is a structural
analog of inhibitory neurotransmitter γ-aminobutyric acid (GABA) [1]. It has been
approved for the treatment of neuropathic pain related to postherpetic neuralgia [2],
postpoliomyelitis neuropathy [3], and reflex sympathetic dystrophy [4]. Due to its
acceptable side effects and strong safety record, gabapentin is currently touted as one
of the most popular drugs [5, 6].
With the increasing demand of gabapentin in recent years, several chemical
routes have been developed from readily available raw materials [7, 8].
1
-Cyanocyclohexaneacetic acid (1-CCHAA) is a key intermediate of gabapentin. One
developed chemical process via formation of 1-CCHAA as intermediate from
-cyanocyclohexylacetonitrile (1-CCHAN) and followed by hydrogenation, was
1
approved as an attractive approach for gabapentin [9, 10]. However, the chemical
process typically required drastic conditions and produced unwanted by-products and
considerable amount of inorganic wastes. Furthermore, ω-cyanocarboxylic acid
1
-CCHAA was produced in low purity and yield due to poor regioselectivity of
chemical hydrolysis of α,ω-dinitrile 1-CCHAN. An alternative approach to the
ω-cyanocarboxylic acid is the selective hydrolysis of dinitrile catalyzed by nitrilases
[11-14]. However, some practical problems such as low substrate concentration and
high biocatalyst loading restricted its industrial applications. Recently, a bioprocess
for regioselective hydration of 1-CCHAN was developed using Rhodococcus
aetherivorans ZJB1208 nitrile hydratase in our laboratory, and the concentration of
3

product (1-cyanocyclohexaneacetamide, 1-CCHAM) accumulated as high as 966.7 g.
-1
L [15]. Since the selective chemical hydrolysis of cyano and amide bond of
1-CCHAM is virtually impossible, amidase-catalyzed hydrolysis is the only facile
way of preparing 1-CCHAA from 1-CCHAM (Fig. 1). Moreover, nitrile
hydratase-amidase coupled biotransformation will provide a particularly appealing
approach [16].
In our effort to search for efficient biocatalysts, three signature amidase (AS)
family amidases (Pa-Ami from Pantoea sp., Cc-Ami from Comamonas composti and
Dt-Ami 6 from Delftia tsuruhatensis ZJB-05174) have been exploited for their
hydrolytic activity towards 1-CCHAM. Pa-Ami was found to exhibit highest catalytic
efficiency, indicating its great potential in large-scale production of 1-CCHAA for
gabapentin synthesis.
2. Materials and methods
2
.1 Chemicals
Compounds 1-CCHAN and 1-CCHAA were provided by Zhejiang Chiral
Medicine Chemicals Co., Ltd. (Hangzhou, China). Compound 1-CCHAM was
prepared using 1-CCHAN as substrate by R. aetherivorans preserved in our
laboratory [15]. Isopropyl-β-D-thiogalactopyranoside (IPTG) and kanamycin were
purchased from Sigma. All other chemicals were of analytical grade and purchased
from commercial sources.
2
.2 Construction of recombinant vectors
The genomic DNA of D. tsuruhatensis ZJB-05174 was extracted and used as
4

template, and two sets of primers (1 μM each) Dt-Ami 6-F
(5’-CATATGAACGACAGCGAACTGCACC-3’)
and
Dt-Ami
6-R
(5’-AAGCTTTCAGGCAGCAGGGTGCTGTCTGT-3’) were used to amplify the
Dt-Ami 6 gene (GenBank accession no. KP943494). The reaction system (100 μL)
contained 4 U Taq DNA polymerase, 40 ng genomic DNA, 100 μM each dNTP, and
1
3
0 μL 10 × Taq DNA buffer. The PCR program was carried out as follows: 94 °C for
min; followed by 30 cycles at 94 °C for 30 s, 58 °C for 30 s, and 72 °C for 2 min;
the reaction was further carried out at 72 °C for 10 min and finally kept at 4 °C. The
DNA fragment was digested with NdeI and HindⅢ and ligated to the pre-digested
pET-28a(+) with the same restriction enzymes. After confirming the absence of
unintended mutations, the plasmid was introduced into E. coli BL21(DE3).
The amino acid sequences of Pa-Ami (GenBank accession no. WP_008109374)
and Cc-Ami (GenBank accession no. WP_027015397) were reverse-translated on the
basis of the preferred codons by gene designer software against E. coli as host[17].
The synthesized genes with 6 × His-tag were inserted into expression vector
pET-28a(+) between NcoI and EcoRI restriction endonuclease sites. The ligated
plasmids pET28a(+)-amidases were transformed into E. coli BL21 (DE3) by heat
shock method [18].
2
.3 Expression and purification of recombinant amidases
E. coli cells harboring the recombinant plasmids were grown at 37 °C in
Luria-Bertani (LB) medium (peptone 10 g/L, yeast extract 5 g/L, NaCl 10 g/L)
containing kanamycin (50 μg/mL) until the optical density at 600 nm reached ~0.6.
5

The expression of recombinant amidases were induced by 0.1 mM IPTG for 12 h at
8 °C. The cells were collected by centrifugation at 12,000 × g for 10 min and washed
2
twice with Tris-HCl buffer (20 mM, pH 8.0). The harvested cells (3.0 g, wet weight)
were then suspended in 30 mL Tris-HCl buffer (20 mM, pH 8.0) and disrupted by
sonication with a Vibra-Cell VC 505 ultrasonic processor (Sonics and Materials Inc.,
Newtown, CT) at 300W for 15 min. Cell debris were removed by centrifugation at
1
2,000 × g for 20 min at 4 °C. The supernatant was used as crude extract and loaded
onto a nickel-nitrilotriacetic acid (Ni-NTA) superflow column pre-equilibrated with
0 mM Tris-HCl buffer (pH 8.0). After eluting the unbound proteins with 20 mM
2
Tris-HCl buffer (pH 8.0) containing 0.5 M NaCl and 20 mM imidazole, the bound
recombinant amidases were eluted with 20 mM Tris-HCl buffer (pH 8.0) containing
0
.3 M NaCl and 0.3 M imidazole. The collected fractions were dialyzed in 20 mM
Tris-HCl buffer (pH 8.0) overnight at 4 °C and analyzed by 12% sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).
2.4 Enzyme assays
The amidase activity was assayed at 45 °C by monitoring the formation of
-CCHAA. The standard reaction mixture consisted of 20 mM Tris-HCl buffer (pH
.0), 20 mM 1-CCHAM and an appropriate amount of the enzyme in a total volume
1
8
of 1 mL. The reactions were stopped by addition of 50 μL HCl (1 M). The mixtures
were centrifuged at 14,000 × g for 5 min and the concentration of 1-CCHAA was then
analyzed by high-performance liquid chromatography (HPLC) using a ODS column
(
Elite, 250 × 4.6 mm) and a UV detector at 215 nm. The mobile phase consisted of
6

7
6% buffer (containing 0.58 g/L NH
4
H
2
PO
4
and 1.83 g/L NaClO , pH=1.8) and 24%
4
acetonitrile and the flow rate was 1 mL/min. One unit of enzyme activity was defined
as the amount of enzyme that catalyzed the formation of 1 μmol of 1-CCHAA per
min. Protein concentrations were determined by the method of Bradford [19] using
BSA as the protein standard.
2
.5 Effects of temperature and pH on Pa-Ami activity
The reactions were performed in standard reaction mixtures (1 mL) containing
0 mM 1-CCHAM and an appropriate amount of purified Pa-Ami. The effect of
2
temperature on Pa-Ami activity was investigated in 20 mM Tris-HCl buffer (pH 8.0)
at different temperatures (30-65 °C). The optimum pH was determined at various pHs:
Na
2
HPO
4
-NaH
2
PO (pH 6.0 to 7.5), Tris-HCl (pH 7.5 to 9.0) and Gly-NaOH (pH 9.0
4
to 10.0) under the standard assay conditions. All experiments were carried out in
triplicate.
2
.6 Kinetic analysis
Kinetic parameters were determined by measuring the enzyme activity at
different substrate concentrations (0.5-20 mM). The reaction was stopped after 15 min
to ensure that less than 10% of the substrate was consumed. All measurements were
performed in 20 mM Tris-HCl buffer (pH 8.0) with an appropriate amount of amidase
at 45 °C. The K
m
and Vmax were calculated from the Lineweaver-Burk
double-reciprocal plot of the obtained data.
Inhibition kinetic parameters (Ki) were determined at 20 and 40 mM 1-CCHAM
in the presence of different concentrations of 1-CCHAA (7.5 to 250 mM). Dixon plots
7

were adapted to determine the inhibition type.
2
.7 Homology Modeling and Molecular Docking
The three-dimensional (3D) structures of amidases were constructed using the
Build Homology Models (MODELER) module in Discovery Studio (DS) version 2.1
Accelrys, Inc.). The crystal structure of amidase from Thermus thermophilus (PDB
DC0) was used as the template for Pa-Ami, Cc-Ami and Dt-Ami 6 according to the
(
2
results of blast. The loop of the generated 3D models were refined sufficiently using
Modeller 9v4. The molecular dockings were performed using Autodock version 4.0.1
and the protein structure diagrams were prepared with PyMOL [20].
2.8 Hydrolysis of 1-CCHAM catalyzed by whole cells of recombinant E. coli
overexpressing Pa-Ami
Hydrolysis of 1-CCHAM catalyzed by the recombinant E. coli cells
overexpressing Pa-Ami was performed in 50 mL flasks containing 10 mL of reaction
mixtures at 35 °C with stirring (180 rpm). The effect of substrate concentration on
enzymatic hydrolysis was determined with the reaction mixture containing 100 mM
Tris-HCl buffer (pH 8.5), 2 g/L (wet cell weight) resting cells and various
concentrations of 1-CCHAM (60-100 g/L). The optimal catalyst loading was
evaluated with 1-3 g/L resting cells at a substrate concentration of 100 g/L. The effect
of product on hydrolysis was investigated in a reaction mixture consisted of 100 mM
Tris-HCl buffer (pH 8.5), 1 g/L (wet cell weight) resting cells, 80 g/L 1-CCHAM and
10-40 g/L 1-CCHAA. Samples (900 µL) were drawn at regular intervals and the
reaction was quenched by the addition of 100 µL 1M HCl. After appropriate dilution
8

and centrifugation (14,000 × g, 10 min), 100 μL of supernatant was then withdrawn
and analyzed by HPLC as described above.
3
. Results and discussion
3.1 Expression and purification of amidases
Amidases can be assigned into amidase signature (AS) family and nitrilase
superfamily based on their amino acid sequence [21]. The amidase signature family is
defined by the AS sequence, which consist of a highly conserved stretch of
approximately 130 serine- and glycine-rich amino acids and a catalytic triad of
Ser-Ser-Lys (Fig. S1). Members from AS family usually display hydrolytic activity
towards a broad spectrum of amides [22, 23].
D. tsuruhatensis ZJB-05174 was an amidase producing strain previously isolated
in our laboratory [24]. Based on the AS sequence and the available genome sequence
data of D. acidovorans SPH-1, which shared high homology with D. tsuruhatensis, an
AS family enzyme Dt-Ami 6 was mined from D. tsuruhatensis ZJB-05174 and
overexpressed in E. coli. As Dt-Ami 6 exhibited high hydrolytic activity towards
1-CCHAM (175.1 U/mg), several homologous AS amidases which possessing a
catalytic triad of Ser-Ser-Lys in AS sequence were selected, synthesized and
overexpressed in E. coli (Fig. S1). Pa-Ami and Cc-Ami, which shared 63% and 89%
identity with Dt-Ami 6, showed hydrolytic activity towards 1-CCHAM. Under the
standard assay conditions, the specific activity of Pa-Ami (297.6 U/mg) was superior
to those of Dt-Ami 6 and Cc-Ami (41.5 U/mg).
The recombinant enzymes were purified to homogeneity by Ni-NTA affinity
9

chromatography. SDS-PAGE analysis of the purified Dt-Ami 6, Pa-Ami and Cc-Ami
showed single band corresponding to a molecular mass of about 50, 55 and 50 kDa,
respectively, which were consistent with the molecular weight (49.9 kDa of Dt-Ami 6,
57.2 kDa of Pa-Ami and 51.1 kDa of Cc-Ami) calculated from the amino acid
sequence (Fig. S2).
3
.2 Kinetic analysis
Kinetic studies of Pa-Ami, Cc-Ami and Dt-Ami 6 were performed by measuring
initial velocities in the presence of varying concentrations of 1-CCHAM. As shown in
Table 1. Pa-Ami showed the lowest apparent K value (1.21±0.11 mM) and highest
cat (250.1±3.5 s ). The catalytic efficiency of Pa-Ami (kcat/K
=208.2±16.2 mM-1 s-1)
m
-1
k
m
was about 2.3-fold higher than that of Dt-Ami 6. These catalytic properties suggested
Pa-Ami as a robust biocatalyst for hydrolysis of 1-CCHAM.
3
.3 Homology Modeling and Molecular Docking
Several models were obtained
http://nihserver.mbi.ucla.edu/SAVES/). The best quality models were shown in the
and
evaluated
by
Procheck
(
Ramachandran plot, where 90.3%, 91.1% and 90.2% of residues were located in the
most favored regions, 7.4%, 7.7% and 7.0% of residues in the additional allowed
regions, 3.1%, 0.8% and 1.5% of residues in the generously allowed regions, and
0
.7%, 0.5% and 0.0% in the disallowed regions for Pa-Ami, Cc-Ami and Dt-Ami 6,
respectively. And the root-mean-square deviations between the models and 2DC0
crystal structure were 0.247, 0.210 and 0.222 for Pa-Ami, Cc-Ami and Dt-Ami 6,
respectively. These data indicated that the constructed models were reliable.
10

The amidase signature family enzymes possess a catalytic Ser-Ser-Lys triad [25],
and these residues are conserved in Pa-Ami (Ser153, Ser177 and Lys80), Cc-Ami
(Ser155, Ser179 and Lys82) and Dt-Ami 6 (Ser155, Ser179 and Lys82) (Fig. 2).
Based on the proposed reaction mechanism of AS amidase [26], Pa-Ami for example,
the side-chain oxygen atom of Ser177 played a role of a nucleophilic agent and
attacked the carbonyl carbon atom of the substrate. Simultaneously, carbonyl group in
substrate was protonated by Ser153, which in turn deprotonated Ser177. The
protonated Lys80 decreased the nucleophilic character of Ser153 and increased the
ability of Ser153 to protonate the carbonyl oxygen of the substrate and the ability of
Ser177 to protonate the leaving group of the substrate. Next, the amino group
captured the proton from Ser153 and the Ser was protonated by the ammonium (NH
+)
3
group at Lys. The amino group of the amide function was eliminated, the proton of
the hydroxyl group in the tetrahedral enzyme–substrate intermediate was abstracted
by Ser153, and the carbonyl group in the substrate molecule was restored. Finally, the
Ser177 abstracted a proton from water molecular to return the original form and
deprotonated water molecular attack the electrophilic carbonyl C to release the
corresponding acid.
When the substrate 1-CCHAM was docked into the active site of Pa-Ami,
γ
Cc-Ami and Dt-Ami 6, as shown in Fig. 2, the distance between the O of Ser179 in
Cc-Ami, Ser177 in Pa-Ami and Ser179 in Dt-Ami 6 and the amide carbon of the
substrate were 3.5 Å, 3.2 Å, and 3.3 Å (Table S1), respectively. Because of the shorter
γ
distance, the nucleophile attack from O of Ser was facilitated and the catalytic
11

efficiency was improved. Furthermore, the efficiency of the proton relay network of
catalytic triad was also an important issue and was closely related to the activity. The
molecular docking data indicated that Pa-Ami have better proton transformation
ability than Dt-Ami 6 and Cc-Ami. Therefore, Pa-Ami was regarded to exhibit the
highest activity due to the shortest nucleophilic attack distance and best proton
transfer ability, which was in accord with the determined enzyme activities.
3
.4 Effects of pH and temperature on Pa-Ami activity
As shown in Fig. 3, Pa-Ami showed high activity (>90%) towards 1-CCHAM at
pH ranging from 7.0 to 9.0 and the optimum pH was found to be 8.5 in Tris-HCl
buffer. However, a remarkable decrease in hydrolytic activity of Pa-Ami was
observed under acidic and basic conditions. Pa-Ami activity increased as the
temperature increased from 30 to 50 °C. The purified Pa-Ami exhibited a maximum
activity at 50 °C and the enzymatic activity sharply decreased above the optimum
temperature (50 °C) (Fig. 4).
3
.5 Optimization of substrate loading
Whole cell biocatalyst has a series of advantages over isolated enzymes, and has
been widely adopted for the commercial production of bulk chemicals and valuable
pharmaceuticals [27-29]. Therefore, whole cells of E. coli overexpressing Pa-Ami
were used to catalyze the hydrolysis of 1-CCHAM at 35 °C and pH 8.5.
High concentration of substrate is beneficial to an industrial bioprocess because
of lower catalyst consumption and easier product recovery. However, the substrate
concentrations of amidase-catalyzed biotransformation are usually low because of
12

their toxicity [30, 31]. In this study, the effect of substrate loading on the hydrolytic
reaction was investigated. As shown in Fig. 5, the whole-cell biocatalysts (2 g/L, wet
cell weight) exhibited strong tolerance against high substrate concentration (up to 100
g/L). The product concentration increased dramatically at the initial stage of the
reaction and the conversion reached 100% after 10, 15 and 20 min with 60, 80 and
100 g/L of substrate. However, product formation no longer increased when the
substrate concentration was raised up to 150 g/L (data not shown). What's more, the
conversion was only 90.1% at the substrate concentration of 150 g/L, even the
biocatalyst loading was elevated to 5 g/L and reaction time was prolonged (Table S2).
This may be ascribed to the enzyme inactivation caused by a high concentration of
product and this type of inhibition has been reported in many enzymatic hydrolysis
[
32, 33].
3
.6 Optimization of biocatalyst loading
The amount of biocatalyst used was another crucial factor for its potential
application. The influence of biocatalyst loading on the conversion of 1-CCHAM (100
g/L) was investigated. As depicted in Fig. 6A, a conversion of 100% was obtained
with 2 and 3 g/L wet whole cells. The time required for complete conversion was
shortened from 20 to 15 min as the amount of biocatalyst increased from 2 to 3 g/L.
When biocatalyst loading of 1 g/L was employed, the conversion reached 80% after
2
0 min and product formation no longer increased. However, 80 g/L substrate could
be completely hydrolyzed with 1 g/L biocatalyst loading (a S/C-ratio of 80) after 20
min (Fig. 6B). On the other hand, space-time yield, which presents quantitative data
13

on the productivity, is the main intensification criterion of the biotransformation itself.
A high space-time yield, which could reduce the capital costs of the process, requires
a high specific activity of the biocatalyst. This process exhibited a high space-time
-1
-1
yield of 5794.7 gproduct L d , which markedly exceeded the average one of industrial
-1
-1
bioprocess (372 gproduct L d ) [34].
3
.7 Effect of product concentration on hydrolytic reaction
Enzymes were shown to be inhibited by their catalyzed product [35]. To
investigate the effect of product concentration, the conversion was determined at
various initial concentrations of 1-CCHAA (10 to 40 g/L). The conversion
significantly decreased from 84.4% to 56.4% when initial concentration of 1-CCHAA
was increased from 10 to 40 g/L (Fig. 7).
To explore the inhibition mechanism of Pa-Ami, the inhibitory type of
1
-CCHAA was investigated by Dixon plots (Fig. 8). The plots of Pa-Ami with
various concentrations of 1-CCHAA resulted in a family of straight lines intersecting
one another above the x-axis, with a Ki value of 50.62±6.7 mM. Therefore, the
inhibition mode of 1-CCHAA toward Pa-Ami was competitive inhibition where
1
-CCHAA reversibly binds to an allosteric site on Pa-Ami.
. Conclusion
4
In summary, we have exploited a novel amidase for efficient hydrolysis of
-CCHAM to afford 1-CCHAA. The recombinant Pa-Ami exhibited high hydrolytic
1
activity and strong tolerance against high substrate loading. The biocatalytic process
with high S/C-ratio (80) provided a more economical route for industrial application,
14

-1
-1
resulting in an outstanding space-time yield of 5794.7 gproduct L d . The developed
amidase-catalyzed process avoids the use of strong acids and organic solvent and thus
reduces remarkably the toxicity to environment as compared to chemical processes.
On the other hand, it significantly improved the catalytic efficiency as compared to
previously reported biocatalysts. These encouraging results not only demonstrated the
feasibility of in silico mining of novel enzymes based on sequence homology, but also
indicated that amidase-catalyzed industrial production of 1-CCHAA is viable,
technologically efficient and environmentally benign.
Acknowledgements
This work was supported by the National Nature Science Foundation of China
(
No. 21202150), National Science Foundation of Zhejiang (Y4080334, LY13B060004)
and National High Technology Research and Development Program of China
No.2012AA022201B).
(
15

References
[
[
[
[
[
[
1] Robinson AA, Malow BA. Gabapentin shows promise in treating pediatric
insomnia. Sleep 2012;35:A394-A394.
2] Segal AZ, Rordorf G. Gabapentin as a novel treatment for postherpetic neuralgia.
Neurology 1996;46:1175-1176.
3] Zapp JJ. Postpoliomyelitis pain treated with gabapentin. Am Fam Physician
1
996;53:2442-2444.
4] Mellick GA, Mellick LB. Reflex sympathetic dystrophy treated with gabapentin.
Arch Phys Med Rehab 1997;78:98-105.
5] Chang CY, Challa CK, Shah J, Eloy JD. Gabapentin in acute postoperative pain
management. BioMed Res Int 2014;2014: 631756.
6] Attal N, Cruccu G, Baron R, Haanpaa M, Hansson P, Jensen TS, Nurmikko T.
EFNS guidelines on the pharmacological treatment of neuropathic pain: 2010
revision. Eur J Neurol 2010;17:1113-E1188.
[
7] Xu XX, Wei ZL, Bai X. Beckmann rearrangement as a key to gabapentin. Chinese
J Org Chem 2006;26:354-356.
[8] Griffiths G, Mettler H, Mills LS, Previdoli F. Novel syntheses of gabapentin via
addition
of
hydrocyanic
acid
to
cyclohexylidenemalonate
or
cyano(cyclohexylidene)acetate. Helv Chim Acta 1991;74:309-314.
[
9] Jennings RA, Johnson DR, Seamans RE, Zeller JR. Process for cyclic amino acid
anticonvulsant compounds. USA. Patent. US 5319135. 94.06.07.
[
10] Jennings RA, Seamans RE, Zeller JR, Johnson DR. Process for cyclic amino acid
16

anticonvulsant compounds. Eur. Patent EP 0414262 B1. 94.11.30.
11] Zhu D, Mukherjee C, Biehl ER, Hua L. Nitrilase-catalyzed selective hydrolysis
of dinitriles and green access to the cyanocarboxylic acids of pharmaceutical
importance. Adv Synth Catal 2007;349:1667-1670.
[
[
[
12] Wong JW, Burns MP. Biocatalytic preparation of 1-cyano-cyclohexaneacetic acid.
USA. Patent. US 20050009154 A1. 05.01.13.
13]Xue YP, Wang YP, Xu Z, Liu ZQ, Shu XR, Jia DX, Zheng YG, Shen YC.
Chemoenzymatic synthesis of gabapentin by combining nitrilase-mediated
hydrolysis with hydrogenation over Raney-nickel. Catal Commun
2
015;66:121-125.
[
14]Xue YP, Zheng YG, Wang YP, Zou SP. Chemo-enzymatic synthesis of gabapentin.
CHN. Patent. CN104911225A 2015.09.16.
[
15] Zheng RC, Yin XJ, Zheng YG. Highly regioselective and efficient production of
1
-cyanocyclohexaneacetamide by Rhodococcus aetherivorans ZJB1208 nitrile
hydratase. J Chem Technol Biotechnol 2015; DOI: 10.1002/jctb.4724.
16] Burns MP, Weaver JK, Wong JW, John WW, Justin KW, Michel PB.
Stereoselective bioconversion of alphatic dinitriles into cyano carboxylic acids.
EUR. Patent. WO2005100580-A1. 2005.10.27
[
[
[
17] Villalobos A, Ness JE, Gustafsson C, Minshull J, & Govindarajan S. Gene
designer: A synthetic biology tool for constructing artificial DNA segments.
BMC Bioinformatics 2006; 7:2–3.
18] Chung CT, Niemela SL, & Miller Roger H. One-step preparation of competent
17

Escherichia coli: Transformation and storage of bacterial cells in the same
solution. Biochem 1989;86:2172-2175.
[
19] Bradford MM. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal Biochem
1
976;72:248-254.
[
[
20] Hallaway D, Feiner S, Hollerer T. Bridging the gaps: Hybrid tracking for
adaptive mobile augmented reality. Appl Artif Intell 2004;18:477-500.
21] Novo C, Farnaud S, Tata R, Clemente A, Brown PR. Support for a
three-dimensional structure predicting a Cys-Glu-Lys catalytic triad for
Pseudomonas aeruginosa amidase comes from site-directed mutagenesis and
mutations altering substrate specificity. Biochem J 2002;365:731-738.
[
[
22] Neumann S, Kula MR. Gene cloning, overexpression and biochemical
characterization of the peptide amidase from Stenotrophomonas maltophilia.
Appl Microbiol Biotechnol 2002;58:772-780.
23] Cai G, Zhu S, Wang X, Jiang W. Cloning, sequence analysis and expression of
the gene encoding a novel wide-spectrum amidase belonging to the amidase
signature superfamily from Achromobacter xylosoxidans. FEMS Microbiol Lett
2
005;249:15-21.
[24] Zheng RC, Zheng YG, Shen YC. Amidase catalyzed kinetic resolution of racemic
mandelamide to (R)-mandelate by Delftia tsuruhatensis ZJB-05174. J Biotechnol
2008;136:S368-S368.
[25] Oinuma K, Takuwa A, Taniyama K, Doi Y, Takaya N. Hydrazidase, a novel
18

amidase signature enzyme that hydrolyzes acylhydrazides. J Bacteriol
015;197:1115-1124.
2
[
[
26] Labahn J, Neumann S, Buldt G, Kula MR, Granzin J. An alternative mechanism
for amidase signature enzymes. J Mol Biol 2002;322:1053-1064.
27] Shaw NM, Naughton A, Robins K, Tinschert A, Schmid E, Hischier ML, Venetz
V, Werlen J, Zimmermann T, Brieden W, de Riedmatten P, Roduit JP,
Zimmermann B, Neumuller R. Selection, purification, characterisation, and
cloning of a novel heat-stable stereo-specific amidase from Klebsiella oxytoca,
and its application in the synthesis of enantiomerically pure (R)- and
(
(
S)-3,3,3-trifluoro-2-hydroxy-2-methylpropionic
S)-3,3,3-trifluoro-2-hydroxy-2-methylpropionamide. Org Process Res Dev
002;6:497-504.
acids
and
2
[
28] Yang ZY, Ni Y, Lu ZY, Liao XR, Zheng YG, Sun ZH. Industrial production of
S-2,2-dimethylcyclopropanecarboxamide with a novel recombinant R-amidase
from Delftia tsuruhatensis. Process Biochem 2011;46:182-187.
[
[
29] Schmid A, Dordick JS, Hauer B, Kiener A, Wubbolts M, Witholt B. Industrial
biocatalysis today and tomorrow. Nature 2001;409:258-268.
30] Nojiri M, Taoka N, Yasohara Y. Characterization of an enantioselective amidase
from Cupriavidus sp. KNK-J915 (FERM BP-10739) useful for enzymatic
resolution of racemic 3-piperidinecarboxamide. J Mol Catal B-Enzym
2
014;109:136-142.
[31] Fuhshuku K, Watanabe S, Nishii T, Ishii A, Asano Y. Efficient preparation of
19

both enantiomers of 3,3,3-trifluoro-2-hydroxy-2-methylpropanoic acid catalyzed
by Shinella sp. R-6 and Arthrobacter sp. S-2. J Mol Catal B-Enzym
2
014;102:115-119.
32] Xue YP, Liu ZQ, Xu M, Wang YJ, Zheng YG. Efficient separation of
R)-(-)-mandelic acid biosynthesized from (R,S)-mandelonitrile by nitrilase using
[
(
ion-exchange process. J Chem Technol Biotechnol 2011;86:391-397.
[
33] Hua DL, Lin S, Li YF, Chen H, Zhang ZB, Du Y, Zhang XH, Xu P. Enhanced
2
-phenylethanol production from L-phenylalanine via in situ product adsorption.
Biocatal Biotransform 2010;28:259-266.
[
[
34] Bommarius AS, Riebel-Bommarius BR, Biocatalysis: fundamentals and
applications, John Wiley & Sons; 2007. p. 1–2.
35] Sonoki T, Morooka M, Sakamoto K, Otsuka Y, Nakamura M, Jellison J, Goodell
B. Enhancement of protocatechuate decarboxylase activity for the effective
production of muconate from lignin-related aromatic compounds. J Biotechnol
2
014;192:71-77.
20

Figure captions:
Fig. 1 Chemoenzymatic route for synthesis of gabapentin
Fig. 2 Docking of 1-CCHAM to the active pocket of the recombinant Cc-Ami (A),
Pa-Ami (B) and Dt-Ami 6 (C).
Fig. 3 Effects of pH on activity of the purified Pa-Ami. The activities were
determined in standard conditions in different buffers with pH varying from 6.0 to
1
1
0.0. The enzyme activity obtained in 50 mM Tris-HCl buffer (pH 8.5) was taken as
00%. The buffers: NaH PO -Na HPO buffer was used for pH 6.0-7.5(■); Tris-HCl
2
4
2
4
buffer for 7.5-9.0(◇); Gly-NaOH for pH 9.0-10.0 (▲).
Fig. 4 Effects of temperature on activity of the purified Pa-Ami. Assays were
performed at various temperatures (30-65 °C) in Tris-HCl buffer (pH 8.5). The
enzyme activity obtained at 50 °C was taken as 100%.
Fig. 5 Time course of 1-CCHAA production by recombinant whole E. coli cells at
various substrate concentrations. The reactions were performed in 50-mL flasks with
1
1
0 mL of reaction mixtures consisted of various substrate concentrations (60-100 g/L),
00 mM Tris-HCl (pH 8.5) and 2 g/L E. coli cells at 35 °C.
Fig. 6 (A) Time course of 1-CCHAA production by whole E. coli cells at various
biocatalyst loadings. The reactions were performed in reaction mixtures consisted of
various biocatalyst loading (1-3 g/L), 100 mM Tris-HCl (pH 8.5) and 100 g/L
1
-CCHAM at 35 °C; (B) Time course of 1-CCHAA production by 1 g/L whole E. coli
cells. The reactions were performed in reaction mixtures consisted of 80 g/L
-CCHAM and 100 mM Tris-HCl (pH 8.5) at 35 °C.
Fig. 7 Effect of concentrations of pre-added 1-CCHAA on conversion. The reactions
1
21

were performed in reaction mixtures consisted of various concentrations of 1-CCHAA
10-40 g/L), 100 mM Tris-HCl (pH 8.5), 80 g/L 1-CCHAM and 1 g/L E. coli cells at
5 °C.
Fig. 8 Determination of enzyme inhibition characteristics by Dixon plots for
-CCHAA. Pa-Ami activity was measured at different 1-CCHAA concentrations
7.5-240 mM) in the presence of 20 and 40 mM 1-CCHAM.
(
3
1
(
22

O
CN
CONH2
Pa-Ami
COOH
HN
NH2 COOH
Rhodococcus
aetherivorans
ZJB1208
NC
NC
NC
1) HCl
2) Na CO
Raney Ni
2
3
1-CCHAN
1-CCHAM
1-CCHAA
Gabapentin
Fig. 1
23

B
A
C
Fig. 2
24

PBS
100
Tris-HCl
Gly-NaOH
8
6
4
2
0
0
0
0
0
5
.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5
pH
Fig. 3
25

100
8
6
4
2
0
0
0
0
0
2
5
30
35
40
45
50
55
60
65
70
o
Temperature ( C)
Fig. 4
26

1
20
00
1
80
60
40
20
0
6
8
1
0g/L
0g/L
00g/L
0
2
4
6
8
10 12 14 16 18 20 22
Reaction ime (min)
Fig. 5
27

A
1
20
00
1
80
60
40
20
0
1
g/L
2g/L
g/L
3
0
2
4
6
8
10 12 14 16 18 20 22
Reaction time (min)
B
100
80
60
40
20
0
0
2
4
6
8
10 12 14 16 18 20 22
Reaction time (min)
Fig. 6
28

100
80
60
40
20
0
10g/L
2
4
0g/L
0g/L
0
2
4
6
8
10 12 14 16 18 20 22
Reaction time (min)
Fig. 7
29