Soluble and functional expression of a recombinant enantioselective amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its biochemical characterization

Article abstract of DOI:10.1016/j.procbio.2015.05.005

A gene encoding an enantioselective amidase (KamH) was cloned from Klebsiella oxytoca KCTC 1686 and inserted into the EcoRI and HindIII sites of the vector pET-30a(+). When KamH with a peptide containing a His-tag and an enterokinase cleavage site was overexpressed in Escherichia coli, approximately half was found in the soluble fraction, but it lacked activity. After cleavage of the peptide by enterokinase, the enzyme activity was partly restored, reaching 420.2 ± 33.62 U/g dry cell weight (DCW). Another recombinant plasmid was constructed by inserting the KamH gene into the NdeI and EcoRI sites of pET-30a(+) to express KamH in its native form. The overexpressed amidase was found primarily in the soluble fraction and its maximum activity was 3613.4 ± 201.68 U/g DCW. This indicated that the peptide influenced not only soluble expression but also activity of KamH, perhaps by blocking the substrate-binding tunnel of KamH. Similar results were obtained with heterologously expressed amidases from Rhodococcus erythropolis MP50 and Agrobacterium tumefaciens d3. All of these amidases have an N-terminal α-helical domain. Therefore, amidases of this type may be functionally expressed in their native form. KamH hydrolyzed a range of aliphatic and aromatic amides and exhibited strict S-selectivity towards racemic amides.

Full text of DOI:10.1016/j.procbio.2015.05.005

G Model
PRBI-10418; No. of Pages8
ARTICLE IN PRESS
Process Biochemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio
Soluble and functional expression of a recombinant enantioselective
amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its
biochemical characterization
∗
∗
Fa-Mou Guo, Jian-Ping Wu , Li-Rong Yang , Gang Xu
Institute of Biological Engineering, College of Chemical and Biological Engineering, Zhejiang University, 310027 Hangzhou, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A gene encoding an enantioselective amidase (KamH) was cloned from Klebsiella oxytoca KCTC 1686 and
inserted into the EcoRI and HindIII sites of the vector pET-30a(+). When KamH with a peptide containing
a His-tag and an enterokinase cleavage site was overexpressed in Escherichia coli, approximately half was
found in the soluble fraction, but it lacked activity. After cleavage of the peptide by enterokinase, the
enzyme activity was partly restored, reaching 420.2 ± 33.62 U/g dry cell weight (DCW). Another recom-
binant plasmid was constructed by inserting the KamH gene into the NdeI and EcoRI sites of pET-30a(+) to
express KamH in its native form. The overexpressed amidase was found primarily in the soluble fraction
and its maximum activity was 3613.4 ± 201.68 U/g DCW. This indicated that the peptide influenced not
only soluble expression but also activity of KamH, perhaps by blocking the substrate-binding tunnel of
KamH. Similar results were obtained with heterologously expressed amidases from Rhodococcus erythro-
polis MP50 and Agrobacterium tumefaciens d3. All of these amidases have an N-terminal ␣-helical domain.
Therefore, amidases of this type may be functionally expressed in their native form. KamH hydrolyzed a
range of aliphatic and aromatic amides and exhibited strict S-selectivity towards racemic amides.
Received 15 February 2015
Received in revised form 9 May 2015
Accepted 11 May 2015
Available online xxx
Keywords:
Amidase
Klebsiella oxytoca KCTC 1686
Biotransformation
Stereoselectivity
Soluble expression
©
2015 Elsevier Ltd. All rights reserved.
1
. Introduction
fine chemical and pharmaceutical industries. Amidases have also
been applied in the bioremediation of toxic chemicals [17,18].
Recombinant expression can increase the expression level of
an exogenous gene, and Escherichia coli is the preferred recom-
binant expression host because of its high biomass-to-cost ratio.
A number of amidase genes have been cloned and heterologously
expressed. However, in most cases the expression efficiency has
been low because of the formation of inactive or insoluble intracel-
lular enzyme [2,15,19–23]. There have been numerous attempts to
improve expression efficiency. First, different heterologous expres-
sion strategies have been investigated, such as using Bacillus subtilis
as a heterologous expression host [23], expressing the amidase
gene together with upstream and downstream flanking sequences
[24], and using pET22b as the expression vector to add a six-
histidine (6 × His)-tag to the C-terminal end of the protein [25].
Second, some studies have focused on the optimization of expres-
sion conditions, for example by using auto-induction media [25,26],
adding ethanol to the culture medium [27], lowering the growth
Amidases (EC 3.5.1.4), which catalyze the hydrolysis of amides
to acid and ammonia, have been discovered in a variety of
organisms, including bacteria, archaea, and fungi. A number
of amidase-harboring microbes have been studied in detail,
such as Pseudomonas chlororaphis [1], Delftia tsuruhatensis [2],
Agrobacterium tumefaciens [3], Klebsiella oxytoca [4], Brevibac-
terium epidermidis [5], Rhodococcus erythropolis [6,7], Sulfolobus
tokodaii [8], and Pyrococcus yayanosii CH1 [9]. Some amidases
have very strict chemo-, regio-, and enantioselectivities, allow-
ing them to be used to produce optically pure amides, carboxylic
acids, and related derivatives. They have been used to produce
(
2
(
S)-2,2-dimethylcyclopropanecarboxamide [2,10–12], piperazine-
-carboxylic acid [13], phenylalanine [14], S-naproxen [15], and
S)-3,3,3-trifluoro-2-hydroxy-2-methylpropionamide [16]. These
chemicals and their derivatives have important applications in the
◦
temperature to 29 C, and cultivating cells without IPTG induction
[
28]. Although these strategies improved expression efficiency to a
∗ Corresponding authors at: Zhejiang University, Institute of Biological Engi-
neering, College of Chemical and Biological Engineering, Hangzhou 310027, China.
Tel.: +86 0571 87952363; fax: +86 0571 87952009.
certain extent, they did not eliminate the problems of low expres-
sion level and formation of inclusion bodies. Suzuki and Ohta have
reported the denaturation and refolding of an insoluble aggregated
E-mail addresses: wjp@zju.edu.cn (J.-P. Wu), lryang@zju.edu.cn (L.-R. Yang).
http://dx.doi.org/10.1016/j.procbio.2015.05.005
1
359-5113/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Guo F-M, et al. Soluble and functional expression of a recombinant enantioselective
amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its biochemical characterization. Process Biochem (2015),
http://dx.doi.org/10.1016/j.procbio.2015.05.005

G Model
PRBI-10418; No. of Pages8
ARTICLE IN PRESS
F.-M. Guo et al. / Process Biochemistry xxx (2015) xxx–xxx
2
6
× His-tagged fusion protein, but the steps were cumbersome and
2.3. Plasmid construction
the yield was unsatisfactory [8].
In this study, we cloned an amidase gene from K. oxytoca KCTC
Ami1 was digested with EcoRI and HindIII, and Ami2 was
digested with NdeI and EcoRI. Subsequently, the digested DNA
fragments were inserted into the same restriction sites of the pET-
30a(+) expression vector under the control of the T7 promoter.
Ligations were performed according to the manufacturer’s proto-
col (Novagen). The constructed recombinant plasmids were named
pET-Ami1 and pET-Ami2.
1
686 [29] and studied how to achieve efficient expression in E. coli.
First, the amidase (KamH) was expressed as a fusion protein with a
kDa peptide containing a 6 × His-tag and an enterokinase cleav-
6
age site at its N-terminus; approximately half of the recombinant
protein was expressed in soluble form, but it had no amidase activ-
ity. Previous studies have indicated that extra amino acid residues
at the terminus of overexpressed proteins can have a profound
effect on the folding and solubility of these proteins [30–34]. There-
fore, we used enterokinase to remove the extra amino acid residues
from the fusion protein and found that the activity was partly
restored. When KamH was expressed in its native form, almost
no inclusion bodies were formed, and the recombinant protein
had maximum activity. Similar results were achieved with het-
erologous expression of other amidases containing an N-terminus
2.4. Expression of KamH in E. coli
Transformation of E. coli BL21(DE3) with pET-Ami1 and pET-
Ami2 was carried out essentially as recommended by Novagen.
Transformed E. coli BL21(DE3) cells were cultivated in LB medium
supplemented with 50 ␮g/ml kanamycin. Cells were incubated
◦
at 37 C on a gyratory shaker until the OD6 reached 1.0. Then,
00
␣
-helical domain. Therefore, amidases of this type may be func-
isopropyl-␤-d-thiogalactopyranoside (IPTG) was added to a final
tionally expressed in their native form, and this strategy may be
an alternative method for the efficient expression of amidases.
Finally, characterization of KamH indicated that this enzyme has
great application potential in the fine chemicals and pharmaceuti-
cal intermediates industries.
concentration of 0.5 mM and the cultures were incubated for 12 h
◦
at 18 C. The cells were harvested by centrifugation (12,000 × g for
10 min), and cell suspensions in 50 mM Tris–HCl buffer (pH 8.0)
were disrupted by sonication (Sonicator 400, Misonix, USA). The
soluble and insoluble fractions were separated by centrifugation
◦
(
12,000 × g for 20 min) at 4 C and analyzed by sodium dodecyl
sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) on 12%
polyacrylamide gels with a 4% stacking gel. Gels were stained with
Coomassie Brilliant Blue.
2
. Materials and methods
2.1. Bacterial strains, plasmids, enzymes, and chemical reagents
2.5. Enterokinase cleavage of the extra peptide
The microbial strain K. oxytoca KCTC 1686 was procured from
◦
KCTC (Daejeon, Korea) and was routinely grown at 30 C in lysogeny
broth (LB) medium (1.0% [w/v] bactotryptone, 0.5% [w/v] yeast
extract, 1.0% [w/v] NaCl, pH 7.2). E. coli strains DH5␣ and BL21(DE3)
The recombinant protein produced by E. coli BL21(DE3)/pET-
Ami1 was purified by a combination of Ni-NTA affinity chromatog-
raphy and Superose S10-300 gel filtration chromatography. The
purified protein was dissolved in enterokinase cleavage buffer
(
Novagen, Madison, USA) were used as hosts for plasmid construc-
tion and recombinant amidase production, respectively. Plasmid
pET-30a(+) (Novagen, Madison, USA) was used to construct recom-
binant vectors. The E. coli strains harboring recombinant vector
(50 mM Tris–HCl, pH 8.0, 2 mM CaCl , 0.1% Tween 80, 50 mM NaCl)
2
to yield a protein solution with an apparent protein concentration
of 6.0 mg/ml. A 2-ml aliquot of this enzyme solution was incubated
◦
◦
were routinely cultured at 37 C in LB liquid medium or on LB agar
with 3 U of enterokinase (Stratagene, Shanghai, China) at 35 C
solid medium supplemented with kanamycin (50 ␮g/ml) as appro-
priate. Restriction enzymes (EcoRI, HindIII, NdeI) and T4 DNA ligase
were purchased from Takara, Ltd. (Dalian, China). FastPfu DNA
polymerase and protein markers were purchased from TransGen
Biotech, Ltd. (Beijing, China). All other reagents were commercial
products of analytical grade or higher and were purchased from
standard suppliers.
for 20 h. Then, 100 ␮l of the enterokinase-digested protein was
incubated with 60 ␮l of STI-agarose (Stratagene, Shanghai, China)
and 1 ml of His-tag affinity resin at 4 C for 1 h. The His-tag and
◦
enterokinase were removed by low-speed centrifugation (1000 × g
for 10 min) and the supernatant was used to test enzyme activity.
2.6. Functional expression of other amidases
The microbial strains R. erythropolis MP50 (DSM 9675) and A.
tumefaciens d3 (DSM 9674) were procured from DSMZ (Braun-
schweig, Germany) and were routinely grown in LB medium.
Extraction of genomic DNA was performed as described for KamH.
Amidase nucleotide sequence data from R. erythropolis MP50
2
.2. Genomic DNA extraction and amidase gene subcloning
K. oxytoca KCTC 1686 was cultured in LB liquid medium
◦
at 30 C and cells were harvested from the culture medium
in the late log phase. Genomic DNA (GenBank accession No.
CP003218.1) was extracted using
Kit (Axygen, Hangzhou, China) according to the manufacturer’s
instructions.
Genomic DNA of K. oxytoca KCTC 1686 was used as template to
amplify the amidase gene by polymerase chain reaction (PCR). Two
sets of primers were designed to amplify the gene: primers A1 (for-
ward) and A2 (reverse), with EcoRI and HindIII sites, respectively,
and primers A3 (forward) and A4 (reverse), with NdeI and EcoRI
sites, respectively. The primer sequences and restriction sites are
listed in Table 1. The PCR program was carried out as follows: 1
(MamH) and A. tumefaciens d3 (DamH) appear in the GenBank
a Genomic DNA Isolation
nucleotide sequence database under the number AY026386 and
AF315580. The primer sequences (A5–A12) used to amplify the
amidase genes from these two strains and the restriction sites are
listed in Table 1. The PCR, recombinant plasmid construction, trans-
formation, and expression of recombinant amidases were carried
out as described for KamH.
2.7. Amidase activity assays and determination of protein
concentration
◦
◦
◦
cycle at 94 C for 5 min; 30 cycles of 94 C for 20 s, 57 C for 20 s,
Amidase activity was determined in reaction mixtures (1.0 ml)
consisting of Tris–HCl buffer (50 mM, pH 8.0), 10 mM benzamide,
and an appropriate amount of enzyme. After a 5-min preincuba-
◦
◦
and 72 C for 90 s; and a final extension at 72 C for 10 min. The PCR
products were analyzed by 0.8% agarose gel electrophoresis. The
products were named Ami1 and Ami2, respectively.
◦
tion at 35 C with shaking, the reaction was started by addition
Please cite this article in press as: Guo F-M, et al. Soluble and functional expression of a recombinant enantioselective
amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its biochemical characterization. Process Biochem (2015),
http://dx.doi.org/10.1016/j.procbio.2015.05.005

G Model
PRBI-10418; No. of Pages8
ARTICLE IN PRESS
F.-M. Guo et al. / Process Biochemistry xxx (2015) xxx–xxx
3
Table 1
Primers and restriction sites used in this work.
Amidase gene
Seq. no.
Sequence
Restriction enzyme
ꢀ
ꢀ
ꢀ
Klebsiella oxytoca 1686
A1
A2
A3
A4
A5
A6
A7
A8
5 -CCGGAATTCATGGCTATTCAACGTCCCACTG-3
EcoRI
HindIII
NdeI
EcoRI
EcoRI
HindIII
NdeI
HindIII
EcoRI
HindIII
NdeI
ꢀ
ꢀ
5 -TTCCCAAGCTTGTTAAAACGTCCGCCAGTCACCG-3
ꢀ ꢀ
5 -GGAATTCCATATGGCTATTCAACGTCCCACT G-3
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
5 -CCGGAATTCTTAAAACGTCCGCCAGTCACCG-3
Rhodococcus
erythropolis MP50
5 -CCGGAATTCATGCGACCCAATCGCCCATTCG-3
5 -TTCCCAAGCTTGTTACCGCAGCACCGGTGCGCTC-3
ꢀ
5 -GGAATTCCATATGCGACCCAATCGCCCATTC G-3
ꢀ
5 -TTCCCAAGCTTGTTACCGCAGCACCGGTGCGCTC-3
ꢀ
Agrobacterium
tumefaciens d3
A9
5 -CCGGAATTCATGAGTCTGGGACCAGAGCTTG-3
ꢀ
A10
A11
A12
5 -TTCCCAAGCTTGTTACATTTTCCGCCAGTCGCCG-3
ꢀ
5 -GGAATTCCATATGAGTCTGGGACCAGAGCTTG-3
ꢀ
ꢀ
5 -TTCCCAAGCTTGTTACATTTTCCGCCAGTCGCCG-3
HindIII
of substrate. The reaction was carried out for 10 min under the
same conditions and was stopped by addition of an equal vol-
ume of acetonitrile. The same reaction system without enzyme
was used as a reaction blank. The protein was removed by
centrifugation at 12,000 × g for 10 min. The acid product was
determined by its absorbance at 220 nm using reversed-phase
high-performance liquid chromatography (HPLC) with a C18 col-
umn (5 ␮m, 4.6 × 250 mm). The mobile phase contained 30% (v/v)
acetonitrile plus 0.1% (v/v) acetic acid, and the flow rate was
mandelamide, 2,2-dimethylcyclopropanecarboxamide, and 2-(4-
chlorophenyl)-3-methylbutyramide. Enantiomeric excess values
were obtained by HPLC analysis using an AY-RH column. The
enantiomeric excess and enantiomeric ratio of the product were
determined according to Chen et al. [37].
3. Results and discussion
3.1. Cloning of the amidase gene and construction of plasmids
0
.5 ml/min. The enantioselectivity of KamH was analyzed by HPLC
®
on a CHIRALPAK AY-RH column (5 ␮m, 4.6 × 150 mm). One unit
of amidase activity is defined as the amount that converts 1 ␮mol
of amide substrate to acid product in 1 min. Protein concentra-
tions were determined using the Bradford method [35], and bovine
serum albumin (Promega, USA) was used as the standard. The Km
and Vmax for benzamide were determined with benzamide con-
centrations of 0.25, 0.50, 1.0, 2.0, 4.0, 8.0, 16, and 32 mM, using
a direct linear plot. Each measurement was repeated at least three
times. Statistical analyses were performed using the statistical soft-
ware SPSS 19.0. A difference was considered significant at a level
of p < 0.05.
The primer sets A1/A2 and A3/A4, harboring different restriction
enzyme sites, were designed based on the putative amidase gene
sequences of K. oxytoca KCTC 1686 available in nucleotide sequence
databases. Two 1515-bp fragments, Ami1 and Ami2, were obtained
by PCR amplification and cloned into pET-30a(+) to generate pET-
Ami1 and pET-Ami2. The constructed plasmids were sequenced by
Sangon Biotech Co. Ltd.
3.2. Heterologous expression of KamH and activity assays
The recombinant plasmid pET-Ami1 was used to transform
E. coli BL21(DE3) cells, generating the recombinant strain E. coli
BL21(DE3)/pET-Ami1. This strain was induced by 0.5 mM IPTG,
and the recombinant amidase was expressed as a fusion protein
with a 6 kDa peptide containing a 6 × His-tag and an enteroki-
nase cleavage site at the N-terminus. Approximately half of the
recombinant protein was found in the soluble fraction (Fig. 1a),
but no enzymatic activity was detected. The extra amino acid
residues of the recombinant amidase (soluble fraction) from E. coli
BL21(DE3)/pET-Ami1 were removed by enterokinase cleavage,
resulting in partial restoration of amidase activity. The activity of
the enterokinase-digested protein reached 420.2 ± 33.62 U/g dry
cell weight DCW (Table 2). To determine whether the fusion peptide
influenced the expression and activity of the recombinant ami-
dase, we transformed the second recombinant plasmid, pET-Ami2,
into E. coli BL21(DE3), obtaining E. coli BL21(DE3)/pET-Ami2. This
strain was induced as described above, resulting in expression of
KamH in its native form. SDS–PAGE analysis showed that most
of the recombinant amidase was present in the soluble fraction
(Fig. 1b). The enzymatic activity with benzamide as substrate was
2
.8. Determination of pH and temperature optima and thermal
stability
The recombinant amidase produced by E. coli BL21(DE3)/pET-
Ami2, purified by a combination of ammonium sulfate fractionation
and Superose S10-300 gel filtration, was used to study the opti-
mum pH, optimal reaction temperature, and thermal stability.
The optimum pH of recombinant KamH was tested by incubat-
ing the enzyme with 10 mM benzamide under standard conditions
for 10 min in the following buffers: sodium citrate (pH 4.0–6.5),
potassium phosphate (pH 6.5–7.5), Tris–HCl (pH 7.5–8.8), and
glycine–NaOH (pH 8.8–10.0). The optimal reaction temperature
was investigated using 10 mM benzamide as substrate under
standard conditions at various temperatures (20–55 C). The ther-
mal stability was determined by incubating the enzyme at 30, 35,
and 40 C for different lengths of time and measuring the activity
under standard conditions.
◦
◦
2.9. Substrate specificity and stereoselectivity
3
613.4 ± 201.68 U/g DCW (Table 2). Statistical analysis indicated
The substrate specificity of recombinant KamH was tested
that the fusion peptide had a significant effect (p < 0.05) on the
enzymatic activity of KamH. When hexanamide, naphthalene-2-
carboxamide, or 2,2-dimethylcyclopropanecarboxamide was used
as substrate, the native amidase had much higher enzymatic
activity than the His-tagged fusion protein. With benzamide as
substrate, the Vmax and Km of the purified native enzyme were
22.6 ± 3.54 ␮mol/(min mg protein) and 1.1 ± 0.09 mM, respec-
tively. These results are similar to those obtained for the amidase
from R. erythropolis AJ270 [25].
using various aliphatic and aromatic amides. Cell-free extract of
E. coli BL21(DE3)/pET-Ami2 was mixed with various concentra-
tions of substrate under standard reaction conditions; 10% (v/v)
methanol was added as cosolvent to solubilize the substrate. The
enzyme activity toward the various amides was determined by
monitoring the formation of either carboxylic acid or ammonia
[
36]. Several racemic amides were used to study the stereoselec-
tivity of the recombinant enzyme, including 2-phenylacetamide,
Please cite this article in press as: Guo F-M, et al. Soluble and functional expression of a recombinant enantioselective
amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its biochemical characterization. Process Biochem (2015),
http://dx.doi.org/10.1016/j.procbio.2015.05.005

G Model
PRBI-10418; No. of Pages8
ARTICLE IN PRESS
F.-M. Guo et al. / Process Biochemistry xxx (2015) xxx–xxx
4
Fig. 1. SDS–PAGE analysis of KamH expression. (a) SDS–PAGE analysis of recombinant amidase from E. coli BL21(DE3)/pET-Ami1. M, protein marker (80, 60, 40, 30, 20 kDa);
W, total protein of uninduced BL21(DE3) cells harboring the amidase gene in pET-30a(+); T1, S1, and I1, total protein, soluble fraction, and insoluble fraction of induced E. coli
BL21(DE3)/pET-Ami1 cells, respectively. (b) SDS–PAGE analysis of recombinant amidase from E. coli BL21(DE3)/pET-Ami2. T2, S2, and I2, total protein, soluble fraction, and
insoluble fraction of induced E. coli BL21(DE3)/pET-Ami2 cells, respectively. Arrows indicate the position of the amidase.
Table 2
Enzymatic activity of various forms of the recombinant amidases.
Various forms
KamH (U/g DCW)
MamH (U/g DCW)
DamH (U/g DCW)
Fusion form protein
Enterokinase-digested protein
Native form protein
Vmax (␮mol/(min mg protein))
Km (mM)
0
0
0
420.2 ± 33.62
3613.4 ± 201.68
22.6 ± 3.54
1.1 ± 0.09
322.6 ± 32.26
245.9 ± 32.79
2741.9 ± 169.35
2295.1 ± 155.74
–
–
–
–
Note: Enzymatic activity was measured with benzamide as substrate.
High-level expression of amidases in E. coli has often been dis-
appointing owing to the formation of inclusion bodies or inactive
protein. As described above, similar problems were encountered
with the expression of KamH in E. coli BL21(DE3) cells. We also
used an Origami strain to express KamH and adjusted the IPTG
concentration. When KamH was expressed as a fusion protein in
E. coli Origami(DE3), induced by a different concentration of IPTG,
a large number of inclusion bodies were formed and no enzymatic
activity was detected. Moreover, inclusion body formation and a
lack of enzyme activity were also observed when RSFDuet-1, pET-
(Table 3). These results are similar to the findings of other scholars
[19,25].
Previous studies have indicated that extra amino acid residues
at the terminus of overexpressed proteins can improve the yield
of the recombinant proteins, help protect them from intracellu-
lar proteolysis, and enhance their solubility [31,38–40]. However,
these extra residues also have the potential to interfere with the
biological activity of the target protein, and different proteins
react differently to the presence of the fusion partner [30,33].
The complete deduced amino acid sequence of KamH showed
47.7% sequence identity to an amidase from Rhodococcus sp. N-771
(RamH) [41]. Studies of the crystal structure of RamH showed that
this enzyme has three domains (an N-terminal ␣-helical domain,
a small domain, and a large domain) and that a Ser-cisSer-Lys cat-
alytic triad is located in the large domain. The N-terminal ␣-helical
domain, together with the small domain, forms a narrow substrate-
binding tunnel. We performed homology modeling and built a
protein structural model. Then, based on this model, we studied
the influence of a peptide at the N- or C-terminus on the solu-
bility and functional expression of KamH. Results indicated that
a peptide at the N- or C-terminus may lead to misfolding during
2
1a(+), pET-22b(+), and pET-28a(+) were used as expression vectors
to express KamH as a fusion protein in E. coli. Conversely, most of
the recombinant protein was present in the soluble fraction and
had maximum amidase activity when RSFDuet-1, pET-21a(+), pET-
2
2b(+), and pET-28a(+) were used as expression vectors to express
KamH in its native form. To investigate the influence of other fusion
tags on the functional expression of KamH, we added an Arg-tag
or S-tag at the N-terminus of KamH. When the recombinant pro-
teins were expressed in E. coli, a large number of inclusion bodies
were formed and no enzymatic activity was detected. This indi-
cated that the Arg-tag and S-tag also affected the solubility and
enzymatic activity of KamH. When KamH was expressed with-
out any extra terminal residues, the recombinant amidase was
found primarily in the soluble fraction and had maximum ami-
dase activity. These results indicated that extra amino acid residues
at the N-terminus affected not only the soluble expression of
KamH but also its activity. We also added a His-tag, Arg-tag, or
S-tag at the C-terminus of KamH. When the recombinant pro-
teins were expressed in E. coli, a large number of inclusion bodies
were formed and the recombinant proteins had very low activity
Table 3
Influence of different fusion tags at different positions on the functional expression
of KamH.
Location
His-tag (U/g DCW)
Arg-tag (U/g DCW)
S-tag (U/g DCW)
N-terminus
C-terminus
0
0
0
672.3 ± 42.02
756.3 ± 33.61
84.0 ± 16.81
Note: Enzymatic activity was measured with benzamide as substrate.
Please cite this article in press as: Guo F-M, et al. Soluble and functional expression of a recombinant enantioselective
amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its biochemical characterization. Process Biochem (2015),
http://dx.doi.org/10.1016/j.procbio.2015.05.005

G Model
PRBI-10418; No. of Pages8
ARTICLE IN PRESS
F.-M. Guo et al. / Process Biochemistry xxx (2015) xxx–xxx
5
Fig. 3. SDS–PAGE analysis of MamH expression. SDS–PAGE analysis of recombi-
nant amidase from E. coli BL21(DE3)/pET-Ami5 and E. coli BL21(DE3)/pET-Ami6. M,
protein marker (80, 60, 40, 30, 20 kDa); W, total protein of uninduced BL21(DE3)
cells harboring the amidase gene in pET-30a(+); T1, S1, and I1, total protein, soluble
fraction, and insoluble fraction of induced E. coli BL21(DE3)/pET-Ami5 cells, respec-
tively; T2, S2, and I2, total protein, soluble fraction, and insoluble fraction of induced
E. coli BL21(DE3)/pET-Ami6 cells, respectively. Arrows indicate the position of the
amidase.
Fig. 2. SDS–PAGE analysis of DamH expression. SDS–PAGE analysis of recombi-
nant amidase from E. coli BL21(DE3)/pET-Ami3 and E. coli BL21(DE3)/pET-Ami4.
M, protein marker (80, 60, 40, 30, 20 kDa); W, total protein of uninduced BL21(DE3)
cells harboring the amidase gene in pET-30a(+); T1, S1, and I1, total protein, soluble
fraction, and insoluble fraction of induced E. coli BL21(DE3)/pET-Ami3 cells, respec-
tively; T2, S2, and I2, total protein, soluble fraction, and insoluble fraction of induced
E. coli BL21(DE3)/pET-Ami4 cells, respectively. Arrows indicate the position of the
amidase.
SDS–PAGE analysis of the recombinant amidase was performed
expression of this amidase and prevent substrate entry into the
narrow substrate-binding tunnel.
(Fig. 3). We found that almost all of the recombinant amidase was
expressed in the form of inclusion bodies, and no enzymatic activ-
ity was detected. When E. coli BL21(DE3)/pET-Ami6 was induced
by IPTG, the SDS–PAGE analysis showed that the insoluble frac-
tion constituted only a small part of the total expressed protein
The heterologous expression of KamH in its native form was
considerably better than that of some other amidases. For example,
when an amidase from R. erythropolis was expressed in E. coli with a
signal peptide on the N-terminus, a portion of the protein was found
in inclusion bodies and the mass of expressed protein did not differ
from that in the native strain [23]. In another study, an amidase
from Paracoccus sp. M-1 was successfully expressed only in auto-
(Fig. 3). Maximum amidase activity of this recombinant enzyme
was 2741.9 ± 169.35 U/g DCW (Table 2). Therefore, expression
of DamH and MamH in their native form is a powerful strat-
egy.
We also performed homology modeling for the amidases from A.
tumefaciens d3 and R. erythropolis MP50, and two protein structural
◦
induction medium under low-temperature incubation at 25 C [26].
When the amidase from the hyperthermophilic archaeon Sulfolobus
solfataricus was expressed in E. coli with a 6 × His-tag at the N-
or C-terminus, the major fraction of the recombinant protein was
insoluble [19].
3.3. Functional expression of other amidases
The amidase gene of A. tumefaciens d3 was amplified from
genomic DNA using the relevant primers (Table 1). Thereafter,
two plasmids, pET-Ami3 (with extra amino acid residues at the
N-terminus) and pET-Ami4 (without extra amino acid residues),
were constructed and transformed into E. coli BL21(DE3). The
recombinant strains, E. coli BL21(DE3)/pET-Ami3 and E. coli
BL21(DE3)/pET-Ami4, were subsequently cultivated and induced.
When the amidase was expressed in E. coli BL21(DE3)/pET-Ami3,
a large proportion of the protein was soluble (Fig. 2). However,
as observed with heterologous expression of KamH, no enzymatic
activity was detected. When the amidase was expressed in E. coli
BL21(DE3)/pET-Ami4 under the same cultivation and induction
conditions, most of the recombinant protein was present in the
soluble fraction (Fig. 2), and its activity was 2295.1 ± 155.74 U/g
DCW (Table 2). The amidase gene of R. erythropolis MP50 was
also amplified, using the same method. For heterologous expres-
sion of this amidase gene, pET-Ami5 (with extra amino acid
residues at the N-terminus) and pET-Ami6 (without extra amino
acid residues) were constructed and transformed into E. coli
BL21(DE3). E. coli BL21(DE3)/pET-Ami5 was induced by IPTG, and
◦
Fig. 4. Effect of pH on KamH activity. Reactions were performed at 35 C for 10 min
in buffers of varying pH with 10 mM benzamide as substrate. The amidase activity
at pH 8.0 was set as 100%.
Please cite this article in press as: Guo F-M, et al. Soluble and functional expression of a recombinant enantioselective
amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its biochemical characterization. Process Biochem (2015),
http://dx.doi.org/10.1016/j.procbio.2015.05.005

G Model
PRBI-10418; No. of Pages8
ARTICLE IN PRESS
6
F.-M. Guo et al. / Process Biochemistry xxx (2015) xxx–xxx
Table 4
Substrate spectrum of KamH.
Compound
Acetamide
Structural formula
Relative activity (%)
e.e. value (%)a
14 ± 1.8
–
Propionamide
Butanamide
65 ± 7.2
84 ± 6.8
–
–
Hexanamide
Acrylamide
205 ± 7.6
6 ± 4.6
–
–
Isobutyramide
82 ± 6.2
–
2
,2-Dimethylcyclopropanecarboxamide
13 ± 1.5
96
Benzamide
100 ± 2.8
37 ± 5.4
–
–
Phenylacetamide
2
-Phenylacetamide
76 ± 6.7
28 ± 3.2
97
96
Mandelamide
2
-(4-Chlorophenyl)-3-methylbutyramide
1.8 ± 0.2
201 ± 8.4
99
Naphthalene-2-carboxamide
–
a
Only racemic amides have an enantiomeric excess (e.e.) value.
models were built. These models indicated that DamH and MamH
also have three domains (an N-terminal ␣-helical domain, a small
domain, and a large domain). A peptide at the N- or C-terminus
may lead to misfolding during expression of DamH and MamH and
prevent substrate entry into the narrow substrate-binding tunnel.
When DamH and MamH were expressed as fusion proteins in E.
coli, a large proportion of insoluble protein was discovered and no
amidase activity was found in the soluble fraction. Therefore, ami-
dases of this type may need to be expressed in their native form to
achieve full function.
3.4. Effects of pH and temperature on KamH
The optimum pH was determined by measuring the enzyme
activity at different pH values, using a variety of buffers. The opti-
mum pH was 8.0, and the enzyme showed relatively high activity
over a broad pH range from 6.5 to 9.0. The enzyme exhibited little
activity below pH 6.0 or above pH 9.5 (Fig. 4). The optimal reac-
tion temperature was determined by assaying the enzyme activity
◦
at different temperatures from 20 to 55 C (Fig. 5). The activity
of KamH was affected by temperature and the enzyme showed
Please cite this article in press as: Guo F-M, et al. Soluble and functional expression of a recombinant enantioselective
amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its biochemical characterization. Process Biochem (2015),
http://dx.doi.org/10.1016/j.procbio.2015.05.005

G Model
PRBI-10418; No. of Pages8
ARTICLE IN PRESS
F.-M. Guo et al. / Process Biochemistry xxx (2015) xxx–xxx
7
GDI-211 prefers aromatic amides [45]. In addition to a wide
substrate spectrum, recombinant KamH also displayed strict (S)-
enantioselectivity to racemic 2-phenylacetamide, mandelamide,
2
3
,2-dimethylcyclopropanecarboxamide, and 2-(4-chlorophenyl)-
-methylbutyramide. KamH hydrolyzed these racemic amides with
high (S)-enantioselectivity (>95%) at 50% conversion. Optically pure
mandelamide and 2,2-dimethylcyclopropanecarboxamide have
important applications in the pharmaceutical industry.
4. Conclusion
In this study, a putative amidase gene was cloned from K. oxy-
toca KCTC 1686 and functionally expressed in E. coli BL21(DE3)
without an affinity tag. When KamH was expressed as a fusion pro-
tein, approximately half of the recombinant protein was expressed
in soluble form, but it had no amidase activity. Similar results
were achieved with heterologously expressed amidases contain-
ing an N-terminal ␣-helical domain. On this basis, it was proposed
that amidases containing an N-terminal ␣-helical domain may be
functionally expressed in their native form and that this strategy
may be an alternative method for the efficient expression of ami-
dases. In addition to a wide substrate spectrum, recombinant KamH
also displayed strict (S)-enantioselectivity to racemic amides. This
study lays the groundwork for further research into the application
of KamH in the fine chemicals and pharmaceutical intermediates
industries.
Fig. 5. Determination of the optimal reaction temperature for KamH. The amidase
◦
activity was measured in 50 mM Tris–HCl buffer, pH 8.0, at 20–55 C for 10 min. The
◦
amidase activity at 40 C was set as 100%.
Conflict of interest statement
There is no conflict of interest in this study.
Acknowledgments
This work was supported by the National Basic Research Pro-
gram of China (973 Program, No. 2011CB710800) and the Hi-Tech
Research and Development Program of China (863 Program, No.
2
010AA101502). These two programs are both supported by the
Ministry of Science and Technology of the People’s Republic of
China.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.procbio.2015.05.005
◦
Fig. 6. Thermal stability of KamH. The enzyme was incubated at 30, 35, or 40
C
for different lengths of time. The residual activity was measured under standard
conditions. The activity of amidase incubated at 30, 35, or 40 C for 0 h was set as
◦
1
00%.
References
[
1] Ciskanik LM, Wilczek JM, Fallon RD. Purification and characterization of an
enantioselective amidase from Pseudomonas chlororaphis B23. Appl Environ
Microbiol 1995;61:998–1003.
◦
maximum activity at 40 C. Thermal stability is an important con-
sideration for all enzymes, so the thermal stability of KamH at 30,
5, and 40 C was determined (Fig. 6). About 50% of the amidase
activity remained after incubation at 35 C for 4 h, and the enzyme
was quite stable at 30 C.
◦
[2] 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;46:182–7.
[3] Trott S, Bauer R, Knackmuss HJ, Stolz A. Genetic and biochemical characteriza-
tion of an enantioselective amidase from Agrobacterium tumefaciens strain d3.
Microbiology 2001;147:1815–24.
3
◦
◦
3.5. Substrate specificity of KamH
[4] Shaw NM, Naughton AB. The substrate specificity of the heat-stable stereospe-
cific amidase from Klebsiella oxytoca. Tetrahedron 2004;60:747–52.
[
5] Jin SJ, Zheng RC, Zheng YG, Shen YC. R-enantioselective hydrolysis of 2,2-
dimethylcyclopropanecarboxamide by amidase from a newly isolated strain
Brevibacterium epidermidis ZJB-07021. J Appl Microbiol 2008;105:1150–7.
6] Hirrlinger B, Stolz A, Knackmuss HJ. Purification and properties of an amidase
from Rhodococcus erythropolis MP50 which enantioselectively hydrolyzes 2-
arylpropionamides. J Bacteriol 1996;178:3501–7.
Various aliphatic and aromatic amides were used to study
the substrate spectrum of recombinant KamH. The recombinant
amidase exhibited excellent activity toward most of the sub-
strates (Table 4). These results are similar to those reported
for amidases from Comamonas acidovorans KPO-2771-4 [42], R.
erythropolis AJ270 [25], and D. tsuruhatensis CCTCCM 205114
10]. In contrast, amidases from P. yayanosii CH1 [9], Heli-
cobacter pylori [43], and Rhodococcus rhodochrous M8 prefer
aliphatic amides [44], while an amidase from Pseudomonas sp.
[
[7] Reisinger C, Osprian I, Glieder A, Schoemaker HE, Griengl H, Schwab H. Enzy-
matic hydrolysis of cyanohydrins with recombinant nitrile hydratase and
amidase from Rhodococcus erythropolis. Biotechnol Lett 2004;26:1675–80.
[
[
8] Suzuki Y, Ohta H. Identification of a thermostable and enantioselective ami-
dase from the thermoacidophilic archaeon Sulfolobus tokodaii strain 7. Protein
Express Purif 2006;45:368–73.
Please cite this article in press as: Guo F-M, et al. Soluble and functional expression of a recombinant enantioselective
amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its biochemical characterization. Process Biochem (2015),
http://dx.doi.org/10.1016/j.procbio.2015.05.005

G Model
PRBI-10418; No. of Pages8
ARTICLE IN PRESS
F.-M. Guo et al. / Process Biochemistry xxx (2015) xxx–xxx
8
[
9] Fu L, Li X, Xiao X, Xu J. Purification and characterization of a thermostable
aliphatic amidase from the hyperthermophilic archaeon Pyrococcus yayanosii
CH1. Extremophiles 2014;18:429–40.
[26] Shen W, Chen H, Jia K, Ni J, Yan X, Li S. Cloning and characterization of a novel
amidase from Paracoccus sp M-1, showing aryl acylamidase and acyl transferase
activities. Appl Microbiol Biotechnol 2012;94:1007–18.
[
[
[
10] Wang YS, Cheng F, Zheng RC, Wang YJ, Zheng YG. Characterization of an enan-
tioselective amidase with potential application to asymmetric hydrolysis of
[27] Borges P, Pacheco R, Karmali A. Pseudomonas aeruginosa amidase: aggregation
in recombinant Escherichia coli. Biotechnol J 2011;6:888–97.
[28] Bigey F, Chebrou H, Fournand D, Arnaud A. Transcriptional analysis of the
nitrile-degrading operon from Rhodococcus sp ACV2 and high level produc-
tion of recombinant amidase with an Escherichia coli-T7 expression system. J
Appl Microbiol 1999;86:752–60.
[29] Shin SH, Kim S, Kim JY, Lee S, Um Y, Oh MK, et al. Complete genome sequence of
Klebsiella oxytoca KCTC 1686, used in production of 2, 3-butanediol. J Bacteriol
2012;194:2371–2.
[30] Bucher MH, Evdokimov AG, Waugh DS. Differential effects of short affinity tags
on the crystallization of Pyrococcus furiosus maltodextrin-binding protein. Acta
Crystallogr, Sect D: Biol Crystallogr 2002;58:392–7.
(R, S)-2, 2-dimethylcyclopropane carboxamide. World J Microbiol Biotechnol
2
011;27:2885–92.
11] Wang YS, Zheng RC, Xu JM, Liu ZQ, Cheng F, Feng ZH, et al. Enantioselective
hydrolysis of (R)-2, 2-dimethylcyclopropane carboxamide by immobilized cells
of an R-amidase-producing bacterium, Delftia tsuruhatensis CCTCC M 205114,
on an alginate capsule carrier. J Ind Microbiol Biotecnol 2010;37:503–10.
12] Yeom SJ, Kim HJ, Oh DK. Enantioselective production of 2,2-
dimethylcyclopropane carboxylic acid from 2,2-dimethylcyclopropane
carbonitrile using the nitrile hydratase and amidase of Rhodococcus erythropolis
ATCC 25544. Enzyme Microb Technol 2007;41:842–8.
[
[
[
[
13] Komeda H, Harada H, Washika S, Sakamoto T, Ueda M, Asano Y. A novel R-
stereoselective amidase from Pseudomonas sp MCI3434 acting on piperazine-
[31] Esposito D, Chatterjee DK. Enhancement of soluble protein expression through
the use of fusion tags. Curr Opin Biotechnol 2006;17:353–8.
[32] Sati SP, Singh SK, Kumar N, Sharma A. Extra terminal residues have a
2
-tert-butylcarboxamide. Eur J Biochem 2004;271:1580–90.
14] Yasukawa K, Asano Y. Enzymatic synthesis of chiral phenylalanine derivatives
by a dynamic kinetic resolution of corresponding amide and nitrile substrates
with a multi-enzyme system. Adv Synth Catal 2012;354:3327–32.
15] Trott S, Burger S, Calaminus C, Stolz A. Cloning and heterologous expression of
an enantioselective amidase from Rhodococcus erythropolis strain MP50. Appl
Environ Microbiol 2002;68:3279–86.
profound effect on the folding and solubility of a Plasmodium falciparum sex-
ual stage-specific protein over-expressed in Escherichia coli. Eur J Biochem
2002;269:5259–63.
[33] Terpe K. Overview of tag protein fusions: from molecular and biochemical
fundamentals to commercial systems. Appl Microbiol Biotechnol 2003;60:
523–33.
[34] Liu BH, Zhang J, Fang Z, Du GC, Chen J, Liao XR. Functional analysis of the
C-terminal propeptide of keratinase from Bacillus licheniformis BBE11-1 and
its effect on the production of keratinase in Bacillus subtilis. Process Biochem
2014;49:1538–42.
16] Shaw NM, Naughton A, Robins K, Tinschert A, Schmid E, Hischier
ML, et al. 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 acids and (S)-3,3,3-trifluoro-2-
[35] Bradford MM. A rapid and sensitive method for the quantitation of micro-
gram quantities of protein utilizing the principle of protein-dye binding. Anal
Biochem 1976;72:248–54.
[36] Kaplan A. The determination of urea, ammonia, and urease. Methods Biochem
Anal 1969;17:311–24.
[37] Chen CS, Fujimoto Y, Girdaukas G, Sih CJ. Quantitative analyses of biochemical
kinetic resolutions of enantiomers. J Am Chem Soc 1982;104:7294–9.
[38] Chatterjee DK, Esposito D. Enhanced soluble protein expression using two new
fusion tags. Protein Expression Purif 2006;46:122–9.
hydroxy-2-methylpropionamide. Org Process Res Dev 2002;6:497–504.
17] Kohyama E, Yoshimura A, Aoshima D, Yoshida T, Kawamoto H, Nagasawa T.
Convenient treatment of acetonitrile-containing wastes using the tandem com-
bination of nitrile hydratase and amidase-producing microorganisms. Appl
Microbiol Biotechnol 2006;72:600–6.
18] Vesela AB, Pelantova H, Sulc M, Mackova M, Lovecka P, Thimova M, et al. Bio-
transformation of benzonitrile herbicides via the nitrile hydratase-amidase
pathway in rhodococci. J Ind Microbiol Biot 2012;39:1811–9.
[
[
[
[
[
19] d’Abusco AS, Ammendola S, Scandurra R, Politi L. Molecular and biochem-
ical characterization of the recombinant amidase from hyperthermophilic
archaeon Sulfolobus solfataricus. Extremophiles 2001;5:183–92.
20] Doran JP, Duggan P, Masterson M, Turner PD, O’Reilly C. Expression and purifi-
cation of a recombinant enantioselective amidase. Protein Expression Purif
[39] Corchero J, Viaplana E, Benito A, Villaverde A. The position of the heterologous
domain can influence the solubility and proteolysis of ␤-galactosidase fusion
proteins in E. coli. J Biotechnol 1996;48:191–200.
[40] Marblestone JG, Edavettal SC, Lim Y, Lim P, Zuo X, Butt TR. Comparison of SUMO
fusion technology with traditional gene fusion systems: enhanced expression
and solubility with SUMO. Protein Sci 2006;15:182–9.
[41] Ohtaki A, Murata K, Sato Y, Noguchi K, Miyatake H, Dohmae N, et al. Structure
and characterization of amidase from Rhodococcus sp N-771: insight into the
molecular mechanism of substrate recognition. Biochim Biophys Acta-Proteins
Proteomics 2010;1804:184–92.
[42] Hayashi T, Yamamoto K, Matsuo A, Otsubo K, Muramatsu S, Matsuda A, et al.
Characterization and cloning of an enantioselective amidase from Comamonas
acidovorans KPO-2771-4. J Ferment Bioeng 1997;83:139–45.
[43] Skouloubris S, Labigne A, DeReuse H. Identification and characterization
of an aliphatic amidase in Helicobacter pylori. Mol Microbiol 1997;25:
989–98.
2
005;40:190–6.
21] Kim SH, Oriel P. Cloning and expression of the nitrile hydratase and amidase
genes from Bacillus sp BR449 into Escherichia coli. Enzyme Microb Technol
2
000;27:492–501.
[
[
22] Wu SJ, Fallon RD, Payne MS. Cloning and nucleotide sequence of amidase gene
from Pseudomonas putida. DNA Cell Biol 1998;17:915–20.
23] Yue Y, Lian J, Tian P, Tan T. Cloning of amidase gene from Rhodococcus ery-
thropolis and expression by distinct promoters in Bacillus subtilis. J Mol Catal
B-Enzym 2009;56:89–95.
[
[
24] Cheong TK, Oriel PJ. Cloning of a wide-spectrum amidase from Bacil-
lus stearothermophilus BR388 in Escherichia coli and marked enhancement
of amidase expression using directed evolution. Enzyme Microb Technol
[44] Ryabchenko LE, Podchernyaev DA, Kotlova EK, Yanenko AS. Cloning the
amidase gene from Rhodococcus rhodochrous M8 and its expression in
Escherichia coli. Russ J Genet 2006;42:886–92.
[45] Kagayama T, Ohe T. Purification and properties of an aromatic amidase from
Pseudomonas sp. GDI 211. Agric Biol Chem 1990;54:2565–71.
2
000;26:152–8.
25] Xue Z, Chao Y, Wang D, Wang M, Qian S. Overexpression of a recombi-
nant amidase in a complex auto-inducing culture: purification, biochemical
characterization, and regio- and stereoselectivity.
011;38:1931–8.
J Ind Microbiol Biot
2
Please cite this article in press as: Guo F-M, et al. Soluble and functional expression of a recombinant enantioselective
amidase from Klebsiella oxytoca KCTC 1686 in Escherichia coli and its biochemical characterization. Process Biochem (2015),
http://dx.doi.org/10.1016/j.procbio.2015.05.005