Biotransformation of iminodiacetonitrile to iminodiacetic acid by Alcaligenes faecalis cells immobilized in ACA-membrane liquid-core capsules

Article abstract of DOI:10.2478/s11696-013-0423-8

Biotransformation of iminodiacetonitrile (IDAN) to iminodiacetic acid (IDA) was investigated with a newly isolated Alcaligenes faecalis ZJUTBX11 strain showing nitrilase activity in the immobilized form. To reduce the mass transfer resistance and to increase the toleration ability of the microorganisms to the toxic substrate as well as to enhance their ability to be reused, encapsulation of the whole cells in alginate-chitosan-alginate (ACA) membrane liquid-core capsules was attempted in the present study. The optimal pH and temperature for nitrilase activity of encapsulated A. faecalis ZJUTBX11 cells were 7.5 C and 35 C, respectively, which is consistent with free cells. Based on the Michaelis-Menten model, kinetic parameters of the conversion reaction with IDAN as the substrate were: K _m = (17.6 ± 0.3) mmol L^-1 and V _max = (97.6 ± 1.2) μmol min^-1 g ^-1 of dry cell mass for encapsulated cells and (16.8 ± 0.4) mmol L^-1 and (108.0 ± 2.7) μmol min^-1 g ^-1 of dry cell mass for free cells, respectively. After being recycled ten times, the whole cells encapsulated in ACA capsules still retained 90 % of the initial nitrilase activity while only 35 % were retained by free cells. Lab scale production of IDA using encapsulated cells in a bubble column reactor and a packed bed reactor were performed respectively.

Full text of DOI:10.2478/s11696-013-0423-8

Chemical Papers
DOI: 10.2478/s11696-013-0423-8
ORIGINAL PAPER
Biotransformation of iminodiacetonitrile to iminodiacetic acid
by cells immobilized
in ACA-membrane liquid-core capsules
a,bJin-Feng Zhang, a,bZhi-Qiang Liu, a,bXin-Hong Zhang, a,bYu-Guo Zheng*
aInstitute of Bioengineering, bEngineering Research Center of Bioconversion and Biopurification of the Ministry of Education,
Zhejiang University of Technology, Hangzhou, Zhejiang 310014, China
Received 15 February 2013; Revised 3 April 2013; Accepted 5 April 2013
Biotransformation of iminodiacetonitrile (IDAN) to iminodiacetic acid (IDA) was investigated
with a newly isolated Alcaligenes faecalis ZJUTBX11 strain showing nitrilase activity in the im-
mobilized form. To reduce the mass transfer resistance and to increase the toleration ability of the
microorganisms to the toxic substrate as well as to enhance their ability to be reused, encapsulation
of the whole cells in alginate–chitosan–alginate (ACA) membrane liquid-core capsules was attempted
in the present study. The optimal pH and temperature for nitrilase activity of encapsulated A. fae-
calis ZJUTBX11 cells were 7.5◦C and 35◦C, respectively, which is consistent with free cells. Based
on the Michaelis–Menten model, kinetic parameters of the conversion reaction with IDAN as the
substrate were: Km = (17.6 0.3) mmol L−1 and Vmax = (97.6 1.2) µmol min−1 g−1 of dry cell
mass for encapsulated cells and (16.8
0.4) mmol L−1 and (108.0
2.7) µmol min−1 g−1 of dry
cell mass for free cells, respectively. After being recycled ten times, the whole cells encapsulated in
ACA capsules still retained 90 % of the initial nitrilase activity while only 35 % were retained by
free cells. Lab scale production of IDA using encapsulated cells in a bubble column reactor and a
packed bed reactor were performed respectively.
c
ꢀ 2013 Institute of Chemistry, Slovak Academy of Sciences
Keywords: biotransformation, iminodiacetic acid, microencapsulation, Alcaligenes faecalis, nitri-
lase, bubble column reactor
Introduction
Xue et al., 2011; Zhang et al., 2011; Jin et al.,
2013).
Nitrilases have recently attracted increasing at-
tention and interest in the biochemistry fields ow-
ing to their ability of conducting nitrile hydrolysis
at ambient conditions with high chemo-, regio- and
enantioselectivity, which also meet the demand of
green chemistry and environmental protection (Baner-
jee et al., 2002; O’Reilly & Turner, 2003; Huang
& Xu, 2006; Martínková & Křen, 2010). A lot
of microorganisms possessing high nitrile-converting
enzyme activity were obtained from soil environ-
ment by employing the high-throughput screening
methods (Sosedov et al., 2009; He et al., 2011;
Iminodiacetic acid (IDA) is an important com-
pound widely utilized as a key intermediate in fine
chemical industry, especially in the production of
glyphosate which has been a dominant herbicide all
over the world for the past decade (Woodburn, 2000).
Biotransformation of IDA from iminodiacetonitrile
(IDAN) by nitrilase led to a promising technique with
the potential to substitute the conventional process
(He et al., 2011; Liu et al., 2011a, 2011b, 2012; Zhang
et al., 2012, 2013). Regarding the future industrial
applications, it is necessary to achieve a biocatalyst
with high enzyme activity and to improve the re-
*Corresponding author, e-mail: zhengyg@zjut.edu.cn

ii
J. F. Zhang et al./Chemical Papers
peated use capacity as well as the operational stability
(Kabaivanova et al., 2005; Banerjee et al., 2006; Kaul
et al., 2006). Great efforts have been made to achieve
these goals: formation of gene recombination microor-
ganisms by using nitrilase separated and purified from
different kinds of sources (Xue et al., 2010; Liu et al.,
2011a, 2011b) and design of a bioreactor that can be
operated in continuous regime even at high dilution
rates (Shukla et al., 2001; Nigam et al., 2009).
A. faecalis ZJUTBX11 strain from culture collec-
tions screened by our laboratory was used in this work
and stored at 4◦C on a BPD (beef and peptone de-
posit) agar slant. The agar slant medium consisted of
beef extract (5 g), peptone (10 g), NaCl (5 g), and
agar (15 g) per liter (pH 7.0). The strain was grown
aerobically at 30◦C for 48–72 h in 50 mL of a sterile
medium consisting of glucose (10 g), peptone (12 g),
yeast powder (5 g), ammonium acetate (7.5 g), NaCl
(5 g), MgSO4 (0.2 g), KH2PO4 (2 g), and K2HPO4
(3 g) per liter, and the pH was adjusted to 7.5 with
1 M NaOH before the autoclaving, and then supple-
mented with butanenitrile (butyronitrile) at the fi-
nal concentration of 0.1 vol. % after the sterilization
for the induction of nitrilase. The microbial cultures
were carried out at 30◦C and 150 min−1 on a shak-
ing incubator. Cells were harvested by centrifugation
at 10000 × g for 20 min under 4◦C and then thor-
oughly washed three times with distilled water to re-
move residuals from the culture medium.
Encapsulation technology has been one of the most
used to immobilize microbial cells since 1993; it has
some obvious advantages over conventional gel beads
entrapment including lower mass transfer resistance
and larger growing space for viable microbial cells
within the capsules (Park & Chang, 2000; Dembczyn-
ski & Jankowski, 2000; Kuan et al., 2010). Microen-
capsulation of whole cells consists in the enclosing of
the active material within a polymeric matrix sur-
rounded by a semi-permeable membrane. A lot of ma-
terials such as alginate, chitosan, polyvinyl alcohol
and other polymers are used to form microcapsules.
An alginate–polylysine–alginate (APA) microcapsule
is one of the most widely studied capsules for its ex-
cellent biocompatibility and characteristics, but the
high cost of polylysine may limit its use in cell en-
capsulation (Chang et al., 1994; Gugerli et al., 2002).
Subsequently, chitosan has been studied to substitute
polylysine for its good properties and lower cost of
the ACA microcapsule preparation; here, alginate and
chitosan, natural anionic or cationic polysaccharides
widely used for immobilization, were combined to form
an alginate–chitosan membrane of the capsules (Lin et
al., 2008).
In the present study, Alcaligenes faecalis
ZJUTBX11 strain, isolated previously from soil in
our laboratory and demonstrated to possess nitrilase
enzyme activity, was used in the biotransformation
of IDA from IDAN. Several immobilization methods
based on the alginate matrix were attempted and var-
ious properties of the immobilized cells such as opti-
mal pH and temperature, pH and temperature stabil-
ity, Michaelis-Menten kinetic parameters with IDAN
as the substrate, and operational stability of the en-
capsulated cell enzyme were also investigated.
Methods of whole cells immobilization
For entrapment of cells, 2 g (wet mass) of cells
were suspended in 10 mL of a sodium chloride solu-
tion (8.5 g L−1) and then mixed well with 10 mL of
a sodium alginate solution (40 g L−1) under thorough
stirring at room temperature. The mixture was ex-
truded drop wise by means of a hypodermic needle
with the diameter of 0.5 mm into a calcium chloride
aqueous solution (20 g L−1) to yield calcium alginate
gel beads with a 2 mm diameter. The resulting gel
beads were hardened in the same aqueous solution for
2 h under appropriate agitation, and stored in a cal-
cium chloride solution (50 g L−1) before use.
Multilayer hydrogel microcapsules were prepared
by a ionic gelation method at room temperature.
2 g (wet mass) of microorganism cells suspended in
10 mL of a normal saline solution was mixed well
with same volume of alginate solutions (40 g L−1
)
prepared by dissolving sodium alginate in an aqueous
solution under heating. The cell-alginate suspension
was then extruded through a 0.5 mm needle into a
20 g L−1 calcium chloride solution to form calcium
alginate gel beads. After a 30 min gelation, micro gel
beads containing microbial cells were obtained; they
were washed with sterile deionized water. The beads
were incubated in a 50 g L−1 chitosan solution dis-
solved in a 1 vol. % acetic acid solution at the vol-
ume ratio of 1 : 10 (beads : solution) to form poly-
electrolyte membranes for a 30 min incubation. Af-
ter being rinsed three times with sterilized physiolog-
ical saline, a 20 g L−1 alginate solution was added to
counteract the excessive positive charge on the mem-
brane; the membranes were rinsed three times. Then,
microcapsules were incubated in a chitosan and al-
ginate solutions again to form dense membranes. As
a result, ACA microcapsules containing cells with a
Experimental
High viscosity sodium alginate (SA) and chi-
tosan (CTS, degree of deacetylation: 95 %) were pur-
chased from Sinopharm Chemical Reagent (Shang-
hai, China). Iminodiacetonitrile (IDAN, > 95 mass %)
was obtained from Zhejiang Wynca Chemical Industry
Group (Hangzhou, China). Iminodiacetic acid (IDA,
> 98 mass %) was purchased from J&K Chemical
(Shanghai, China). Solvents used for HPLC were pur-
chased from Merck & Co. (Shanghai, China). All other
reagents and chemicals purchased from commercial
source were of analytical grade.

J. F. Zhang et al./Chemical Papers
iii
solid core and the diameter of 2 mm were prepared
and stored in physiological saline at 4◦C prior to use.
Furthermore, the ACA microcapsules containing mi-
croorganism cells with a liquid core were also prepared
using the treated microcapsules mentioned above in a
55 mM sodium citrate solution for 20 min under con-
stant stirring to give capsules with a 1.5 mm diameter.
For the preparation of ACA capsules with liquid
cores by a one-step method, 10 mL of the microorgan-
ism suspension containing 2 g (wet mass) of cells and
3 % of calcium chloride were mixed well with 10 mL of
a chitosan solution (50 g L−1) prepared by dissolving
a certain amount of chitosan in a 10 g L−1 acetic acid
solution under the conditions of heating. The mixture
was then extruded drop by drop into 300 mL of a
50 g L−1 sodium alginate solution using a 0.5 mm
hypodermic needle under constant stirring at room
temperature. After 30 min of the membrane forma-
tion process, capsules with the diameter of approxi-
mately 2 mm were washed three times with physio-
logical saline and stored at 4◦C prior to use.
The surface and the layer structure of the cap-
sules membrane were observed by a scanning electric
microscope (SEM). Both capsule samples, with and
without microbial whole cells, were dehydrated on a
lyophilizer and mounted on specimen stubs with sil-
ver paint, gold-coated and examined by SEM at the
acceleration of 20 kV.
Kinetic parameters including the maximum reac-
tion rate (Vmax) and the apparent Michaelis–Menten
constant (Km) were determined based on different
substrate concentrations, from 20 mM to 240 mM, and
the yield of the product in the biotransformation of
IDAN to IDA by free and encapsulated cells at 35◦C
and pH 7.5. Apparent Km sand Vmax values were cal-
7.5, different concentrations of IDAN from 50 mM to
200 mM and a certain amount of the ACA membrane
capsules containing 10 g of wet cells (corresponding to
1.49 g of dry cells). The reactor was incubated at 35◦C
in a controlled temperature chamber. Air was supplied
from the bottom of the reactor with an air pump at the
rates of 0.6 L min−1, 0.9 L min−1, and 1.2 L min−1
.
Every batch, the conversion mixture was withdrawn
at intervals from the reactor and the concentration of
IDA was analyzed using HPLC. All experiments were
carried out in triplicates.
Continuous biotransformation of IDAN to IDA
was also carried out in a packed-bed reactor made
of glass with the ACA-membrane cells-encapsulated
capsules as the catalyst. Parameters of the reactor
were: height of 35 cm and diameter of 4.5 cm, the re-
actor was packed with a certain amount of capsules
containing 10 g of wet cells (approximately 1.49 g
of DCM). Temperature of the reaction system was
controlled at 35◦C by water circulation in the outer
jacket of the glass column. The substrate solution
(various amounts of IDAN dissolved in the 50 mM
Tris-HCl buffer of pH 7.5 to yield a certain con-
centration of the influent) was fed into the bioreac-
tor at the bottom of the column using a peristaltic
pump to regulate the flow rate. The biotransforma-
tion was conducted at different residence times and
the effluent samples were collected at defined inter-
vals from the top of the column and subjected to
HPLC. All experiments were carried out in tripli-
cates.
Analytical methods
The nitrile-hydrolyzing enzyme activities were as-
sayed using IDAN as the substrate according to Liu
et al. (2011b). One unit of the enzyme activity was
defined as the amount of enzyme which produced
1 µmol of IDA per min at 35◦C and pH 7.5. All pre-
sented data were the mean values of triplicate assays
in which the standard deviations were always lower
than 5 %.
The amount of IDA was also determined by
high performance liquid chromatography (HPLC, LC-
10AS, Shimadazu, Japan) using a 4.6 mm × 250 mm
× 5 µm Hypersil SAX column (Elite, Dalian, China).
Here, 1 mL of the reaction mixture was centrifuged
by 10000 × g for 10 min under 4◦C and then microfil-
trated by a 0.45 µm membrane (Millipore, USA). The
resultant 20 µL of the supernatant were subjected to
HPLC to determine the amount of IDA formed. A UV
detector set at the wavelength of 210 nm and a mo-
bile phase of 20 mM (NH4)2HPO4 buffer (pH 4.0) at
the flow rate of 1.0 mL min−1 were used for the detec-
tion. The mobile phase was filtered through a 0.45 µm
filter membrane. The analysis time was 10 min per
sample.
culated by function fitting. Catalytic efficiency (Kcat
)
was expressed as the ratio Vmax/Km. All experiments
were carried out in triplicates.
To investigate the performance of encapsulated
cells in the long-term use, repeated batch conversion
of IDAN to IDA was carried out using the resus-
pended whole cells and encapsulated cells of A. faecalis
ZJUTBX11. The reaction mixture (10 mL), contain-
ing a Tris-HCl buffer (50 mM, pH 7.5), 0.2 g of wet
mass of resting cells (equivalent of 0.03 g dry cell mass,
DCW) or the equivalent cells-immobilizing capsules
and 100 mM IDAN, was immobilized on an orbital
shaker at 35◦C and 150 min−1. After each cycle, the
cells or the capsules were recovered by filtration and
then washed with distilled water prior to their trans-
fer into another fresh reaction mixture for the next
conversion run under the same conditions.
The batch biotransformation of IDA from IDAN
was conducted in a 500 mL bubble column reactor
made of glass (height: 35 cm, inner diameter: 5 cm)
with the working volume of 400 mL. The conversion
system consisted of: 100 mM Tris-HCl buffer of pH

iv
J. F. Zhang et al./Chemical Papers
Results and discussion
tain compounds, such as nitriles or azepan-2-one (ε-
caprolactam), except for Bacillus subtilis ZJB-063
(Chen et al., 2008) and Rhodococcus R 312 (CBS
717.73) (Osprian et al., 1999). A comparative study
of the biotransformation of IDAN to IDA was also
carried out using non-induced and butyronitrile-intro-
duced A. faecalis ZJUTBX11 cells. Non-induced cells
yielded 8 % of IDA form IDAN after 3 h of the con-
version reaction, whereas 35 % of IDA for induced
cells were achieved, which revealed that the A. fae-
calis ZJUTBX11 nitrilase is primarily inducible.
The effect of IDAN concentration ranging from
50 mM to 250 mM on the conversion to IDA was in-
vestigated at 35◦C in a 50 mM Tris-HCl buffer (pH
7.5) (data not shown). Increasing the IDAN concen-
tration from 50 mM to 250 mM led to an increased
IDA concentration and an extended reaction time. In
the batch reaction, 150 mM of IDAN were completely
converted in 8 h of the incubation with free cells as
the biocatalyst. Furthermore, no substrate and prod-
uct inhibition was observed even though the IDAN
concentration was increased up to 200 mM . In view
of the results, this strain can be considered as the most
efficient microorganism for the production of IDA in
future studies.
Biotransformation by free A. faecalis
ZJUTBX11 cells
A. faecalis ZJUTBX11, possessing nitrile-convert-
ing activities, was isolated from soil samples in our
laboratory using IDAN as the sole nitrogen source of
the biotransformation of IDAN to IDA. IDAN is a
dicyano compound mainly used in the production of
glyphosate and it is a poor substrate for the nitrile
hydrolysis to the nitrilase reported previously. It has
been reported that A. faecalis JM3 arylacetonitrilase
is completely inactive against IDAN (Nagasawa et al.,
1990). However, Kobayashi et al. (1990) firstly demon-
strated the nitrilase from Rhodococcus rhodochrous
K22 to be active on aliphatic dinitriles, such as pen-
tanedinitrile (glutaronitrile), 2,2ꢀ-iminodiacetonitrile,
3,3ꢀ-oxydipropionitrile, and 3,3ꢀ-iminodipropionitrile,
with nitrilase activity of 1.924 µmol min−1 mg−1 of
the protein to iminodiacetonitrile. Nitrilase from Al-
caligenes sp. ECU0401 also showed low activity over
IDAN, only 0.49 µmol min−1 mg−1 protein (Zhang
et al., 2011). Except these enzymes, no more infor-
mation about the hydrolysis of IDAN was reported.
In recent years, our research group focused our at-
tention on the biotransformation of IDAN to IDA for
the improvement of green production of glyphosate. It
has been indicated that A. faecalis ZJUTBX11 exhib-
ited excellent specific nitrilase activity and thus can
hydrolyze IDAN effectively employing nitrilase. Spe-
cific enzyme activity of the A. faecalis ZJUTBX11 is
109.4 µmol min−1 g−1 of dry cells considering the
increase of the nitrilase-producing capacity and op-
timization of the fermentation conditions.
Selection of immobilization method based on
alginate matrix
As it is well known, the cost of complex purifica-
tion and immobilization of enzymes is too high for
it to be used in industrial scale production; however,
immobilization of whole cells is simpler and provides
higher enzyme activity recovery than the purified en-
zymes used (Léonard et al., 2011). In this regard, for
a large-scale biotransformation process, immobilized
whole cells of A. faecalis ZJUTBX11 were used as the
biocatalyst in the production of IDA from IDAN.
Microbial cells immobilized in a hydrogel matrix
can be protected from outer harsh environmental con-
ditions such as pH, temperature, organic solvent, and
toxicity of substrate or product. Various methods
In the previous study, the fact that the A. fae-
calis ZJUTBX11 strain produces nitrilase rather than
nitrile hydratase and amidase systems was confirmed
for no amide by-product detected throughout the hy-
drolysis reaction, also no activity was observed when
some amide compounds were used as the substrates.
As it is known, most nitrilases are inducible by cer-
Table 1. Comparison of different methods of whole cells immobilization (n = 3)
Cells loading rate
Enzyme activity recovery
%
Total intact beads
rate after 10 runs
Method
Carrier
Entrapment
Ca-alginate beads
> 99
65
79
55
67
88
1.5
1
45
67
65
87
3
1
2
4
ACA capsules with liquid core
(two-step method)
2
Microencapsulation
ACA capsules with solid core
(two-step method)
> 99
> 95
3
ACA capsules with liquid core
(one-step method)
2.5

J. F. Zhang et al./Chemical Papers
v
products can easily cross through the semi-permeable
membrane (Bučko et al., 2005). The capsule mem-
brane also protects the cells from contact with the
toxic organic solvent and provides a mild, comfortable
environment for the cells. The encapsulation method
enables high cell loading, low mass transfer resistance,
and better ability of repeated use in industrial appli-
cations than any other immobilization method used
until now (Park & Chang, 2000).
In view of the results mentioned above, whole cells
of A. faecalis ZJUTBX11 were encapsulated into the
ACA capsules with liquid core using the one-step
method for the production of IDA from IDAN in the
present study.
based on alginate carrier were utilized for the immo-
bilization of whole cells of A. faecalis ZJUTBX11. Re-
sults of the immobilization methods used are listed in
Table 1. Conventional entrapment of microbial cells
in Ca-alginate beads is the most widely used immobi-
lization technique for whole cells as well as for purified
enzymes. However, the problems of Ca-alginate beads
entrapment are mainly related to the membrane weak-
ening with time. The presence of chelating reagents in-
cluding phosphate, lactate and citrate and the growth
of microorganisms inside the beads can result in a
burst of the Ca-alginate bead (Kabaivanova et al.,
2005).
In the present study, only 45 % of total intact
Ca-alginate beads remained after ten conversion runs,
whereas 67 %, 65 %, and 87 % of ACA capsules with
liquid and solid core prepared via the two-step method
and ACA capsules prepared via the one-step method
remained respectively, though the cells loading rate
of the entrapment in Ca-alginate beads was 100 %.
Meanwhile a large loss of whole cells was observed
in the liquefaction procedure of the microencapsula-
tion via the two-step method, which led to a 65 %
of the cells loading rate compared to approximately
100 % of the other methods. The achieved 67 % of
enzyme activity recovery of ACA capsules with solid
core via the two-step method is similar to that of the
Ca-alginate beads entrapment but much lower than
that of the ACA capsules with liquid core, which is
primarily caused by the diffusion limitations from the
surface into the interior of the beads. Among the four
cell immobilization methods, ACA capsules with liq-
uid core prepared via the one-step method showed the
highest enzyme activity recovery and total intact cell
rate, which means the probability of long-term use,
although the diameter of the capsules (2 mm) pre-
pared using this method was slightly larger than that
of the other ones. Furthermore, encapsulation of vi-
able whole cells in one step can significantly simplify
the preparation procedure and subsequently improve
the recovery of enzyme activity.
Bioencapsulation has been widely used as one
of the most important immobilization techniques in
a number of applications, including biotechnologi-
cal (Park & Chang, 2000), environmental (Stormo &
Crawford, 1992), etc. In particular, microbial whole
cells and enzymes were encapsulated as industrial bio-
catalysts for the production of fine chemicals and
pharmaceuticals (Liese et al., 2006; Schmid et al.,
2001; Straathof et al., 2002). Liquid core capsules have
many advantages over other methods in microbial cell
immobilization, such as adsorption, entrapment, etc.
(Fo¨rster et al., 1996). Capsules made of high viscos-
ity sodium alginate and chitosan have shown a high
potential in biotransformation. Their liquid cores pro-
vide a larger space for the encapsulated cells and re-
duces the mass transfer resistance. The cells can be
confined inside the capsule and the substrates and the
Characterization of ACA membrane liquid-
core capsules
Morphologies of both capsules with cells and empty
samples were investigated. The capsules with the di-
ameter of 1.5–2.0 mm were smooth, spherical, and
covered with a thin layer of sodium alginate which
diminished during the conversion process due to the
shear force or was redissolved in the reaction system.
The size of the liquid-core capsules is controlled by the
gravity force, drag force, and surface tension (Park &
Chang, 2000), while in the present experiment, the size
of a liquid drop falling from a needle was due to the
gravity and the drag force larger than 1 mm. It is very
important for the capsules to have good mechanical
stability to support their proliferation and long-term
use in bioreactors (Kurillova et al., 2000). Whole cells
of A. faecalis ZJUTBX11 within the capsules were uni-
formly distributed in the interior of the capsules con-
taining chitosan because of its incomplete combination
with sodium alginate. However, cells encapsulated in
the capsules with liquid core tended to form large ag-
gregates in standing state, especially in a packed bed
reactor. The surface and the layer structure of the
ACA membrane of the capsules were also observed by
a scanning electric microscope (SEM), Fig. 1. The thin
gelatin film formed around the droplet spontaneously
when it trickled into the alginate solution because of
the electrostatic interaction of the oppositely charged
polymers. The layer thickness of the ACA membrane
of the capsules was approximately 15 µm which is
negligible compared to their diameter of 2 mm. En-
capsulated whole cells can be formed for every type
of enzyme using genetically engineered microbial cells
(Park et al., 1999).
Temperature and pH profiles
The effect of basic enzymatic properties such as pH
and temperature on the nitrile-converting enzyme ac-
tivity of free and encapsulated cells was investigated.
In all cases, the highest enzyme activity of the free and
encapsulated cells was taken as 100 % and the relative

vi
J. F. Zhang et al./Chemical Papers
Fig. 1. SEM photographs of ACA-membrane capsules: empty capsules (A, C, E) and cell-encapsulated capsules (B, D, F).
activity was expressed in percentage of the maximum
activity.
ones in the pH range of 5–10, and 65 % of the initial
enzyme activity were recovered for encapsulated cells
when stored at pH 6.0 for 3 h compared to 44 % for
free cells; and 57 % for encapsulated cells compared
to 45 % for free cells when incubated at pH 10.0.
The effect of temperature on the enzyme activity
of free and encapsulated cells was studied in the range
of 20–50◦C in the Tris-HCl buffer system (50 mM, pH
7.5) and it is presented in Fig. 2C. The maximum ac-
tivity was observed at 35◦C for both free and encapsu-
lated cells. These results show that the encapsulation
procedure has no obvious effect on the physical and
chemical properties of the enzyme. Generally, thermal
stability of the biocatalyst has to be improved by its
immobilization before its utilization in the industrial
scale. In the current study, the temperature stability
of the enzyme in free and encapsulated cells was evalu-
The effect of pH on the enzyme activity of free
and encapsulated cells was assessed using IDAN as the
substrate in the pH range of 4.0–10.0 at 35◦C, as well
as an acetate buffer (pH 4–6), Tris-HCl buffer (pH 7–
9), and glycine–NaOH buffer (pH 8.5–10). As can be
seen from Fig. 2A, the optimal pH value of encapsu-
lated cells in the Tris-HCl buffer system was 7.5, which
is consistent with that of the free cells. This result is
in agreement with the fact that the optimal pH of ni-
trilase from A. faecalis is always between pH 7–8, e.g.
pH 7.5 for nitrilase from A. faecalis JM3 (Nagasawa
et al., 1990), or pH 7.0 for nitrilase from A. faecalis
ECU0401 (He et al., 2010). Fig. 2B depicts the pH sta-
bility of the free and encapsulated cells. As it can be
seen, encapsulated cells are more stable than the free

J. F. Zhang et al./Chemical Papers
vii
Fig. 2. Effect of optimal pH (A) and pH stability (B), optimal temperature (C) and thermal stability (D) on the enzyme activity.
Highest enzyme activities of free and encapsulated cells obtained were assigned to be 100 %. Free cells ( ), encapsulated
cells ( ), free cells at 45◦C ( ), at 55◦C ( ), encapsulated cells at 45◦C (ꢀ), at 55◦C (ꢁ).
◦
•
ated by measuring the residual enzyme activity at dif-
ferent times of the incubation at 45◦C and 55◦C. The
activity of both types of cells at 0 h of the incubation
was taken to be 100 %. The results in Fig. 2D show
an increase in the thermal stability both at 45◦C and
55◦C after the encapsulation which is probably due
to the better protection provided by the membrane
of the capsules. Thermal stability of the encapsulated
cells was enhanced by immobilization which is a great
advantage in large scale applications.
Kinetic parameters of free and encapsulated
cells nitrilase
Fig. 3. Initial reaction rate at different concentrations of IDAN
in its bioconversion to IDA catalyzed by free cells and
the encapsulated counterparts ( ) with fitting curves
(—).
◦
The biotransformation reaction of IDAN to IDA
followed the Michaelis–Menten kinetics and the cor-
responding kinetic parameters of the free and immo-
bilized cells were determined at pH 7.5 and 35◦C us-
ing the Lineweaver–Burk plot to calculate the Vmax
and the apparent Km values (Fig. 3). Both free and
encapsulated cells showed a typical hyperbolic re-
sponse with the increasing substrate concentration
from 20 mM to 240 mM. The Vmax and apparent Km
values were found to be (108.0 2.7) µmol min−1 g−1
and (97.6
1.2) µmol min−1 g−1 of DCM and
(17.6 0.3) mmol L−1 for encapsulated cells, respec-
tively (Table 2). The Km value obtained for encap-
sulated cells was slightly higher than that obtained
for their free counterparts showing a decrease in the
affinity of the enzyme of encapsulated cells to the
substrate compared to that of free cells. Generally,
of DCM and (16.8
0.4) mmol L−1 for free cells,

viii
J. F. Zhang et al./Chemical Papers
Table 2. Kinetic parameters of free and encapsulate cells at pH 7.5 and 35◦C (n = 3)
Parameter
Free cells
Encapsulated cells
Apparent Km/(mmol L−1
)
16.8
108.0
0.4
2.7
17.6
97.6
0.017
0.3
1.2
Vmax/(µmol min−1 g−1
)
Catalytic efficiency, Vmax/Km/min−1
0.018
the differences observed in the enzyme affinity can
be due to the conformational changes of the enzyme
caused by its immobilization or the lower accessibility
of active sites of the immobilized enzyme. The Vmax
value for encapsulated cells was lower than that for
free cells which probably attributed to the restricted
diffusion of the substrate into the capsules. Bučko et
al. (2005) prepared polyelectrolyte complex capsules
immobilizing a whole-cell epoxide-hydrolyzing enzyme
with sodium alginate and cellulose sulfate as polyan-
ions and poly(methylene-co-guanidine) as a polyca-
tion. Encapsulation of the biocatalyst led to a two-fold
increase in the CES hydrolase activity and a shorter
time required for total biotransformation.
Fig. 4. Operational stability of free (A) and encapsulated (B)
cells. The enzyme activity of both types of cells at 0 h
was considered to be 100 %.
Operational stability of encapsulated cells
A significant advantage of the immobilized enzyme
is its excellent recycling stability compared to its free
form. Therefore, enzyme activity of the ACA mem-
brane encapsulated biocatalyst is very stable during
the sequencing batch or continuous reaction processes.
To investigate the long-term operational stability, sev-
eral repetitive applications of encapsulated cells in the
conversion of IDAN to IDA in a batch reaction pro-
cess were performed under optimal conditions. Initial
activity of the encapsulated cells in the first run of
the reaction was considered as 100 %, the relative
enzyme activities of encapsulated cells in each run
were calculated by the ratio of nitrilase activity to the
total enzyme activity in the first run. After ten re-
peated batches, the encapsulated cells still retained
80 % of their original activity, whereas only 40 %
were observed for free cells (Fig. 4). The results re-
vealed that cells encapsulated in the ACA membrane
capsules have better operational stability compared to
their suspended counterparts, which indicates that the
encapsulation of whole cells is a promising alternative
technology in the long-term production field. Further-
more, only a small amount of cell loss was observed
in the conversion reaction system within six cycles,
indicating excellent mechanical strength of the ACA
capsules; another controlled experiment with alginate
gel beads used in the conversion system was carried
out, and a large amount of cells observed in the reac-
tion mixture due to their leakage from the hydrogel
revealed weak mechanical strength of the membrane.
Kabaivanova et al. (2005) demonstrated the immo-
bilization of whole cells of moderately thermophilic
Bacillus sp. UG-5B harboring nitrilase on polysulfone
membranes using chemical binding, the retention of
100 % activity was obtained after eight reaction cy-
cles.
Batch production of IDA in a bubble column
reactor (BCR)
In order to reuse the biocatalyst improving thus
the productivity and reducing the production costs,
consecutive batch reactions with capsule immobilized
whole cells of A. faecalis ZJUTBX11 were performed
in a bubble column bioreactor containing 100 mM
IDAN. Conversion of the substrate and productivity
of the reactor were determined at various incubation
times under different conditions possibly affecting the
bioconversion process in the BCR system.
The reason why a BCR reactor was used in the
present study is the mix function provided by the up-
flow air bubble, since the shear force in a BCR is rather
lower than that caused by the impeller in a stirred
tank reactor (STR), which significantly reduced the
breakage of the cell immobilized capsules in the con-
version process (Xu et al., 2005). Factors significantly
influencing the production of IDA in a BCR are mainly
pH, temperature, air flow rate, and concentration of
the substrate. In the present experiments, pH and
temperature of the reaction system were fixed at 7.5
and 35◦
C by the Tris-HCl buffer and a thermostatic
water cycle was set up to maintain optimal conditions
of nitrilase-mediated hydrolysis of IDAN catalyzed by
encapsulated A. faecalis ZJUTBX11 cells. The air flow
rates of 0.6 L min−1, 0.9 L min−1, and 1.2 L min−1

J. F. Zhang et al./Chemical Papers
ix
Fig. 5. Time course of the batch production of IDA with BCR at different air flow rates: 0.6 L min−1 ( ), 0.9 L min−1
min−1 (ꢀ) and different concentrations of IDAN: 50 mM (A), 1.00 mM (B), 150 mM (C), 200 mM (D).
(
), 1.2 L
◦
were applied to evaluate the effect of the bubble rate
on the production of IDA because flow rates lower
than 0.6 L min−1 cannot provide efficient lifting force
to achieve good mixing results and flow rates higher
than 1.2 L min−1 result in a large amount of air bub-
bles possibly causing overflow of the reaction mixture
from the reactor; this could be resolved by an addition
of antifoaming agents but at higher production costs.
All batch operations were terminated after 8 h of the
conversion and at predetermined interferes; 1 mL of
the reaction mixture was withdrawn to determine the
concentration of IDA by HPLC.
The effect of the air flow rate on the production
of IDA at different concentrations of the substrate are
shown in Fig. 5, where four different concentrations
of the substrate resulted in an increase of the product
content with the increase of the air flow rate in the
BCR system. The faster the air flow rates the better
the mix of the substrate and the biocatalyst in the sys-
tem which subsequently decreased mass transfer resis-
tance and resulted in an increase of the reaction rate.
The results show the conversion of the reaction and
the productivity as a function of the substrate concen-
tration and air flow rate (Figs. 6A and B). As it can
be seen from the figure, the conversion ratio decreased
with the increase of the substrate concentration from
Fig. 6. Conversion (A) and productivity (B) of IDAN to IDA
by encapsulated cells in a bubble column reactor at
different concentrations of the substrate under air flow
rate of: 0.6 L min−1 ( ), 0.9 L min−1
( ), and 1.2 L
•
min−1(ꢁ), respectively.
50 mM to 200 mM whereas the difference in the con-
version ratio under various air flow rates diminished
as the substrate concentration increased. As the sub-

x
J. F. Zhang et al./Chemical Papers
Table 3. Results of the batch production of IDA in a bubble column reactor at different concentrations of IDAN (n = 5)
Concentration of IDAN
mM
Conversion
%
Initial specific activity
Biocatalyst productivity (DCW)
g g−1
µmol min−1 g−1
50
100
150
200
89.4
67.4
56.5
42.9
2.5
3.2
1.4
2.3
47.5
65.1
85.4
88.2
3.6
1.4
2.1
2.8
1.4
2.1
2.7
2.7
0.05
0.02
0.03
0.01
strate content is too high and will inhibit the biocat-
alyst, the conversion ratio cannot be improved by en-
hancing the stirring rate. On the other hand, produc-
tivity of the conversion system increased significantly,
from 4.5 mmol L−1 h−1 to 10.6 mmol L−1 −1, as the
h
substrate concentration increased up to 150 mM, and
then it did not increase though the continuous increase
of the substrate concentration and the air flow rate. As
it can be seen from Table 3, the initial specific activ-
ity reached the highest value of 88.2 µmol min−1 g−1
of DCM when applying the substrate concentration of
150 mM and the air flow rate of 1.2 L min−1, achieving
thus the highest biocatalyst productivity of 2.7 g g−1
of DCM.
During the conversion process, no breakage of the
capsules and no cell accumulation in the reactor was
observed, which shows that the mechanical strength
of the capsule membrane was strong enough to bear
the shear force caused by air bubble flowing up and
indicates that the BCR system is a good alternative
for the biotransformation applying the encapsulation
technique.
Fig. 7. Conversion (A) and productivity (B) of IDAN to IDA
by encapsulated cells in a packed bed reactor at differ-
ent residence times at the substrate concentration of:
50mM ( ), 100 mM ( ), 150 mM (ꢀ), and 200 mM
◦
(ꢂ), respectively.
Continuous biotransformation of IDAN in a
packed bed reactor (PBR)
ever, in a PBR, only approximately 1 h is required
to achieve a steady state and can maintain in the best
conditions for a longer time than that in a BCR. Obvi-
ously higher conversion ratio can be obtained by pro-
longing the residence time whereas the productivity
of the reactor decreases significantly (Fig. 7B). The
maximum productivity of 45.6 mmol L−1 h−1 was
achieved at the residence time of 0.6 h and the sub-
strate concentration of 200 mM ; the conversion ratio
was rather low compared to the results obtained in a
BCR.
Based on the above results, continuous biotrans-
formation of IDAN to IDA in a PBR using encap-
sulated A. faecalis ZJUTBX11 cells was performed
(Fig. 8). The results showed that the concentration
of the product changes a little even after ten hours
of the process, which indicates perfect performance of
the cell immobilized capsules in a PCR; low concentra-
tion of cells leaking from the capsules indicate that the
ACA membrane liquid core capsules are a promising
immobilization technique for the biocatalysis indus-
try.
To enhance the IDA productivity, continuous bio-
transformation of IDAN to IDA in a 400 mL PBR
reactor was performed at 35◦C with continuous feed-
ing of IDAN from the bottom of the reactor at the
concentrations of 50 mM, 100 mM, 150 mM, and
200 mM, pH 7.5 and different flow rates. The sub-
strate residence time was set to 0.4 h, 0.6 h, 0.8 h,
1.3 h, and 1.9 h, respectively, by changing the flow
rate controlled by a peristaltic pump. In each case
when achieving a steady state of the reactor, when
the pH and concentration of the product remained
constant, relevant results were determined calculating
the conversion ratio and productivity of the reaction
in the PBR. As it can be seen in Fig. 7A, the conver-
sion ratio increased rapidly as the residence time was
prolonged from 0.4 h to 1.9 h and it was obviously
reduced with the increase of the substrate concentra-
tion. When applying the residence time of 1.9 h and
the substrate concentration of 50 mM, the conversion
ratio reached the maximum value of 45.2 % which is
lower than that obtained in the BCR system; how-

J. F. Zhang et al./Chemical Papers
xi
Chen, J., Zheng, Y. G., & Shen, Y. C. (2008). Biotransformation
of p-methoxyphenylacetonitrile into p-methoxyphenylacetic
acid by resting cells of Bacillus subtilis. Biotechnology and
Applied Biochemistry, 50, 147–153. DOI: 10.1042/ba20070
106.
Dembczynski, R., & Jankowski, T. (2000). Characterisation of
small molecules diffusion in hydrogel-membrane liquid-core
capsules. Biochemical Engineering Journal, 6, 41–44. DOI:
10.1016/s1369-703x(00)00070-x.
F¨orster, M., Mansfeld, J., Dautzenberg, H., & Schellenberger, A.
(1996). Immobilization in polyelectrolyte complex capsules:
Encapsulation of a gluconate-oxidizing Serratia marcescens
strain. Enzyme and Microbial Technology, 19, 572–577. DOI:
10.1016/0141-0229(95)00193-x.
Fig. 8. Continuous production of IDA from IDAN by encapsu-
lated cells in a packed bed reactor at the residence time
of 0.6 h and the substrate concentration of 150 mM.
Gugerli, R., Cantana, E., Heinzen, C., von Stockar, U., &
Marison, I. W. (2002). Quantitative study of the pro-
duction and properties of alginate/poly-L-lysine microcap-
sules. Journal of Microencapsulation, 19, 571–590. DOI:
10.1080/02652040210140490.
He, Y. C., Zhang, Z. J., Xu, J. H., & Liu, Y. Y. (2010). Biocat-
alytic synthesis of (R)-(–)-mandelic acid from racemic man-
delonitrile by cetyltrimethylammonium bromide-permeabi-
lized cells of Alcaligenes faecalis ECU0401. Journal of In-
dustrial Microbiology & Biotechnology, 37, 741–750. DOI:
10.1007/s10295-010-0720-y.
He, Y. C., Liu, Y. Y., Ma, C. L., & Xu, J. H. (2011). Modified
ferric hydroxamate spectrophotometry for assaying glycolic
acid from the hydrolysis of glycolonitrile by Rhodococcus sp
CCZU10−1. Biotechnology and Bioprocess Engineering, 16,
901–907. DOI: 10.1007/s12257-011-0156-x.
Conclusions
Biotransformation of IDAN to IDA by the immobi-
lization of whole cell nitrilase of A. faecalis ZJUTBX11
in an ACA capsule with liquid core was attempted. In
addition, the excellent mechanical performance of the
capsules enabled the utilization of the encapsulated
cells in a larger scale biotransformation under long-
term continuous operational conditions. No breakage
of capsules and no cell accumulation were observed
in either of the reactors indicating lower shear force
in the reactors and higher mechanical strength of the
capsules. However, the low enzyme activity is a draw-
back of the presented method. Increasing the activity
of the enzyme in its genetically engineered strain is
attempted in our lab.
Huang, H. R.,
& Xu, J. H. (2006). Preparation of (S)-
mandelic acid from racemate using growing cells of Pseu-
domonas putida ECU1009 with (R)-mandelate degradation
activity. Biochemical Engineering Journal, 30, 11–15. DOI:
10.1016/j.bej.2006.01.010.
Jin, L. Q., Liu, Z. Q., Xu, J. M., & Zheng, Y. G. (2013).
Biosynthesis of nicotinic acid from 3-cyanopyridine by a
newly isolated Fusarium proliferatum ZJB-09150. World
Journal of Microbiology & Biotechnology, 29, 431–440. DOI:
10.1007/s11274-012-1195-y.
Kabaivanova, L., Dobreva, E., Dimitrov, P., & Emanuilova, E.
(2005). Immobilization of cells with nitrilase activity from
a thermophilic bacterial strain. Journal of Industrial Micro-
biology and Biotechnology, 32, 7–11. DOI: 10.1007/s10295-
004-0189-7.
Acknowledgements. The authors gratefully acknowledge the
financial support of the National High Technology Research
and Development Program of China (863 Program) (No.
2011AA02A210075), Natural Science Foundation of the Zhe-
jiang Province (No. R3110155, Z4090612 and Y4110409) and
the Qianjiang Talent Project of the Zhejiang Province.
Kaul, P., Banerjee, A., & Banerjee, U. C. (2006). Stereoselec-
tive nitrile hydrolysis by immobilized whole-cell biocatalyst.
Biomacromolecules, 7, 1536–1541. DOI: 10.1021/bm0507913.
Kobayashi, M., Yanaka, N., Nagasawa, T., & Yamada, H.
(1990). Purification and characterization of a novel nitrilase
of Rhodococcus rhodochrous K22 that acts on aliphatic ni-
triles. Journal of Bacteriology, 172, 4807–4815.
References
Banerjee, A., Sharma, R., & Banerjee, U. C. (2002). The nitrile-
degrading enzymes: current status and future prospects.
Applied Microbiology and Biotechnology, 60, 33–44. DOI:
10.1007/s00253-002-1062-0.
Kuan, I. C., Wu, J. C., Lee, S. L., Tsai, C. W., Chuang, C.
A., & Yu, C. Y. (2010). Stabilization of D-amino acid ox-
idase from Rhodosporidium toruloides by encapsulation in
polyallylamine-mediated biomimetic silica. Biochemical En-
gineering Journal, 49, 408–413. DOI: 10.1016/j.bej.2010.02.
003.
Kurillova, L., Gemeiner, P., Vikartovska, A., Mikova, H., Rosen-
berg, M., & Ilavsky, M. (2000). Calcium pectate gel beads
for cell entrapment. 6. Morphology of stabilized and hard-
ened calcium pectate gel beads with cells for immobilized
biotechnology. Journal of Microencapsulation, 17, 279–296.
DOI: 10.1080/026520400288265.
Banerjee, A., Kaul, P., & Banerjee, U. C. (2006). Enhancing the
catalytic potential of nitrilase from Pseudomonas putida for
stereoselective nitrile hydrolysis. Applied Microbiology and
Biotechnology, 72, 77–87. DOI: 10.1007/s00253-005-0255-8.
Bučko, M., Vikartovská, A., Lacík, I., Kolláriková, G., Gemei-
ner, P., Pätoprstý, V., & Brygin, M. (2005). Immobilization
of a whole-cell epoxide-hydrolyzing biocatalyst in sodium
alginate-cellulose sulfate-poly(methylene-co-guanidine) cap-
sules using a controlled encapsulation process. Enzyme and
Microbial Technology, 36, 118–126. DOI: 10.1016/j.enzmictec.
2004.07.006.
Chang, P. L., Hortelano, G., Awrey, D. E., & Tse, M. (1994).
Growth of recombinant fibroblasts in alginate microcap-
sules. Biotechnology and Bioengineering, 43, 925–933. DOI:
10.1002/bit.260431005.
Léonard, A., Dandoy, P., Danloy, E., Leroux, G., Meunier, C.
F., Rooke, J. C., & Su, B. L. (2011). Whole-cell based hybrid
materials for green energy production, environmental reme-

xii
J. F. Zhang et al./Chemical Papers
diation and smart cell-therapy. Chemical Society Reviews,
40, 860–885. DOI: 10.1039/c0cs00024h.
Schmid, A., Dordick, J. S., Hauer, B., Kiener, A., Wubbolts,
M., & Witholt, B. (2001). Industrial biocatalysis today and
tomorrow. Nature, 409, 258–268. DOI: 10.1038/35051736.
Liese, A., Seelbach, K., Wandrey, C., Rao, N. N., Lu¨tz, S., Seel-
bach, K., & Liese, A. (2006). Basics of bioreaction engineer-
ing. In A. Liese, K. Seelbach, and C. Wandrey (Eds.), Indus-
trial biotransformations (pp. 57–91). Weinheim, Germany:
Wiley. DOI: 10.1002/9783527608188.ch5.
Shukla, V. B., Veera, U. P., Kulkarni, P. R.,
& Pandit,
A. B. (2001). Scale-up of biotransformation process in
stirred tank reactor using dual impeller bioreactor. Biochem-
ical Engineering Journal, 8, 19–29. DOI: 10.1016/s1369-
703x(00)00130-3.
Lin, J. Z., Yu, W. T., Liu, X. D., Xie, H. G., Wang, W., & Ma, X.
J. (2008). In Vitro and in Vivo characterization of alginate–
chitosan–alginate artificial microcapsules for therapeutic oral
delivery of live bacterial cells. Journal of Bioscience and Bio-
engineering, 105, 660–665. DOI: 10.1263/jbb.105.660.
Liu, Z. Q., Dong, L. Z., Cheng, F., Xue, Y. P., Wang, Y. S., Ding,
J. N., Zheng, Y. G., & Shen, Y. C. (2011a). Gene cloning,
expression, and characterization of a nitrilase from Alcali-
genes faecalis ZJUTB10. Journal of Agricultural and Food
Chemistry, 59, 11560–11570. DOI: 10.1021/jf202746a.
Liu, Z. Q., Li, F. F., Cheng, F., Zhang, T., You, Z. Y., Xu, J.
M., Xue, Y. P., Zheng, Y. G., & Shen, Y. C. (2011b). A novel
synthesis of iminodiacetic acid: Biocatalysis by whole Al-
caligenes faecalis ZJB-09133 cells from iminodiacetonitrile.
Biotechnology Progress, 27, 698–705. DOI: 10.1002/btpr.603.
Liu, Z. Q., Zhou, M., Zhang, X. H., Xu, J. M., Xue, Y. P., &
Zheng, Y. G. (2012). Biosynthesis of iminodiacetic acid from
iminodiacetonitrile by immobilized recombinant Escherichia
coli harboring nitrilase. Journal of Molecular Microbiology
and Biotechnology, 22, 35–47. DOI: 10.1159/000337055.
Martínková, L., & Křen, V. (2010). Biotransformations with
nitrilases. Current Opinion in Chemical Biology, 14, 130–
137. DOI: 10.1016/j.cbpa.2009.11.018.
Nagasawa, T., Mauger, J., & Yamada, H. (1990). A novel nitri-
lase, arylacetonitrilase, of Alcaligenes Faecalis JM3. Purifi-
cation and characterization. European Journal of Biochem-
istry, 194, 765–772. DOI: 10.1111/j.1432-1033.1990.tb19
467.x.
Nigam, V. K., Khandelwal, A. K., Gothwal, R. K., Mohan, M.
K., Choudhury, B., Vidyarthi, A. S., & Ghosh, P. (2009).
Nitrilase-catalysed conversion of acrylonitrile by free and im-
mobilized cells of Streptomyces sp. Journal of Biosciences,
34, 21–26. DOI: 10.1007/s12038-009-0005-7.
O’Reilly, C., & Turner, P. D. (2003). The nitrilase family of
CN hydrolysing enzymes – a comparative study. Journal of
Applied Microbiology, 95, 1161–1174. DOI: 10.1046/j.1365-
2672.2003.02123.x.
Osprian, I., Jarret, C., Strauss, U., Kroutil, W., Orru, R. V. A.,
Felfer, U., Willetts, A. J., & Faber, K. (1999). Large-scale
preparation of a nitrile-hydrolysing biocatalyst: Rhodococ-
cus R 312 (CBS 717.73). Journal of Molecular Catalysis B:
Enzymatic, 6, 555–560. DOI: 10.1016/s1381-1177(99)00009-
0.
Park, J. K., Jin, Y. B., & Chang, H. N. (1999). Reusable
biosorbents in capsules from Zoogloea ramigera cells for cad-
mium removal. Biotechnology and Bioengineering, 63, 116–
121. DOI: 10.1002/(sici)1097-0290(19990405)63:1<116::aid-
bit12>3.0.co;2-k.
Park, J. K., & Chang, H. N. (2000). Microencapsulation of
microbial cells. Biotechnology Advances, 18, 303–319. DOI:
10.1016/s0734-9750(00)00040-9.
Sosedov, O., Matzer, K., Bu¨rger, S., Kiziak, C., Baum, S.,
Altenbuchner, J., Chmura, A., van Rantwijk, F., & Stolz,
A. (2009). Construction of recombinant Escherichia coli
catalysts which simultaneously express an (S)-oxynitrilase
and different nitrilase variants for the synthesis of (S)-
mandelic acid and (S)-mandelic amide from benzaldehyde
and cyanide. Advanced Synthesis & Catalysis, 351, 1531–
1538. DOI: 10.1002/adsc.200900087.
Stormo, K. E., & Crawford, R. L. (1992). Preparation of encap-
sulated microbial cells for environmental applications. Ap-
plied and Environmental Microbiology, 58, 727–730.
Straathof, A. J., Panke, S., & Schmid, A. (2002). The pro-
duction of fine chemicals by biotransformations. Current
Opinion in Biotechnology, 13, 548–556. DOI: 10.1016/s0958-
1669(02)00360-9.
Woodburn, A. T. (2000). Glyphosate: production, pricing and
use worldwide. Pest Management Science, 56, 309–312. DOI:
10.1002/(sici)1526-4998(200004)56:4<309::aid-ps143>3.3.co;
2-3.
Xu, S. K., Qu, Y. H., Chaouki, J., & Guy, C. (2005). Character-
ization of homogeneity of bubble flows in bubble columns us-
ing RPT and fibre optics. International Journal of Chemical
Reactor Engineering, 3, A54. DOI: 10.2202/1542-6580.1264.
Xue, Y. P., Liu, Z. Q., Xu, M., Wang, Y. J., Zheng, Y. G., &
Shen, Y. C. (2010). Enhanced biotransformation of (R,S)-
mandelonitrile to (R)-(–)-mandelic acid with in situ produc-
tion removal by addition of resin. Biochemical Engineering
Journal, 53, 143–149. DOI: 10.1016/j.bej.2010.10.009.
Xue, Y. P., Liu, Z. Q., Xu, M., Wang, Y. J., & Zheng,
Y. G. (2011). Efficient separation of (R)-(–)-mandelic acid
biosynthesized from (R,S)-mandelonitrile by nitrilase using
ion-exchange process. Journal of Chemical Technology and
Biotechnology, 86, 391–397. DOI: 10.1002/jctb.2528.
Zhang, Z. J., Xu, J. H., He, Y. C., Ouyang, L. M., & Liu, Y. Y.
(2011). Cloning and biochemical properties of a highly ther-
mostable and enantioselective nitrilase from Alcaligenes sp
ECU0401 and its potential for (R)-(–)-mandelic acid produc-
tion. Bioprocess and Biosystems Engineering, 34, 315–322.
DOI: 10.1007/s00449-010-0473-z.
Zhang, J. F., Liu, Z. Q., Zheng, Y. G., & Shen, Y. C. (2012).
Screening and characterization of microorganisms capable of
converting iminodiacetonitrile to iminodiacetic acid. Engi-
neering in Life Sciences, 12, 69–78. DOI: 10.1002/elsc.2011
00090.
Zhang, J. F., Liu, Z. Q., & Zheng, Y. G. (2013). Improve-
ment of nitrilase production from a newly isolated Al-
caligenes faecalis mutant for biotransformation of imin-
odiacetonitrile to iminodiacetic acid. Journal of the Tai-
wan Institute of Chemical Engineers, 44, 169–176. DOI:
10.1016/j.jtice.2012.11.010.