Preparation of cross-linked enzyme aggregates of l-aminoacylase via co-aggregation with polyethyleneimine

Article abstract of DOI:10.1016/j.molcatb.2011.10.003

l-Aminoacylase from Aspergillus melleus was co-aggregated with polyethyleneimine and subsequently cross-linked with glutaraldehyde to obtain aminoacylase-polyethyleneimine cross-linked enzyme aggregates (termed as AP-CLEA). Under the optimum conditions, AP-CLEA expressed 74.9% activity recovery and 81.2% aggregation yield. The said method of co-aggregation and cross-linking significantly improved the catalytic stability of l-aminoacylase with respect to temperature and storage. AP-CLEA were employed for enantioselective synthesis of three unnatural amino acids (namely: phenylglycine, homophenylalanine and 2-naphthylalanine) via chiral resolution of their ester-, amide- and N-acetyl derivatives. The enantioselectivity of AP-CLEA was the highest for hydrolysis of amino acid amides; was moderate for hydrolysis of N-acetyl amino acids and was the least for hydrolysis of amino acid esters. Furthermore, AP-CLEA were found to retain more than 92% of the initial activity after five consecutive batches of (RS)-homophenylalanine hydrolysis suggesting an adequate operational stability of the biocatalyst.

Full text of DOI:10.1016/j.molcatb.2011.10.003

Journal of Molecular Catalysis B: Enzymatic 74 (2012) 184–191
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Preparation of cross-linked enzyme aggregates of l-aminoacylase via
co-aggregation with polyethyleneimine
Bhalchandra K. Vaidya∗, Suyog S. Kuwar, Sandeep B. Golegaonkar, Sanjay N. Nene
Biochemical Engineering Department, Chemical Engineering and Process Development Division, National Chemical Laboratory, Dr. Homi Bhabha Road, Pune 411008, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 12 May 2011
Received in revised form 9 September 2011
Accepted 5 October 2011
Available online 12 October 2011
l-Aminoacylase from Aspergillus melleus was co-aggregated with polyethyleneimine and subsequently
cross-linked with glutaraldehyde to obtain aminoacylase–polyethyleneimine cross-linked enzyme
aggregates (termed as AP-CLEA). Under the optimum conditions, AP-CLEA expressed 74.9% activity
recovery and 81.2% aggregation yield. The said method of co-aggregation and cross-linking significantly
improved the catalytic stability of l-aminoacylase with respect to temperature and storage. AP-CLEA
were employed for enantioselective synthesis of three unnatural amino acids (namely: phenylglycine,
homophenylalanine and 2-naphthylalanine) via chiral resolution of their ester-, amide- and N-acetyl
derivatives. The enantioselectivity of AP-CLEA was the highest for hydrolysis of amino acid amides; was
moderate for hydrolysis of N-acetyl amino acids and was the least for hydrolysis of amino acid esters. Fur-
thermore, AP-CLEA were found to retain more than 92% of the initial activity after five consecutive batches
of (RS)-homophenylalanine hydrolysis suggesting an adequate operational stability of the biocatalyst.
© 2011 Elsevier B.V. All rights reserved.
Keywords:
Cross-linked enzyme aggregates
l-Aminoacylase
Polyethyleneimine
Chiral resolution
Unnatural amino acids
1. Introduction
of amines [10,14]. Owing to higher enantioselectivity and broader
substrate specificity, the commercial importance of Aspergillus l-
l-Aminoacylase (N-acyl amino acid amidohydrolase or acylase-
I; EC 3.5.1.14) has long been utilized for industrial production
of enantiopure l-amino acids from N-acyl dl-amino acids [1].
The enzyme is widely distributed in variety of plants, animals
and microorganisms. The most commonly used l-aminoacylases
are those from hog kidney, porcine kidney, Aspergillus oryzae
and Aspergillus melleus. Moreover, arecheal l-aminoacylases espe-
cially those from Pyrococcus furiosus and Thermococcus litoralis are
receiving increasing attention due to their capability to operate at
elevated temperatures [2,3]. l-Aminoacylases from Aspergillus sp.
are well suited for large scale industrial biotransformations as they
are inexpensive, readily available and more stable as compared to
the others [4]. Aspergillus l-aminoacylases have been extensively
utilized for the industrial production of natural amino acids (e.g.
l-alanine, l-methionine, l-valine, etc.) as well as unnatural amino
acids (l-␣-amino butyric acid, l-norvaline, l-norleucine, etc.) [5].
The usefulness of crude l-aminoacylase of Aspergillus sp. in
enantioselective hydrolysis of amino acid esters and amides is
reported [6]. Furthermore, in the presence of organic medium, the
enzyme can catalyze a number of synthetic transformations such
as regioselective alcoholysis of carboxylic acid esters [7–9], acy-
lation of primary and secondary alcohols [10–13], and acylation
aminoacylase is expected to expand from its current state in years
to come.
The present work is aimed at developing a facile method for syn-
thesizing catalytically active and stable cross-linked biocatalyst of
A. melleus l-aminoacylase. Cross-linked enzyme technology in the
past couple of decades, has emerged as an attractive alternative for
enhancing stability of enzymes [15]. The cross-linking of enzymes
by means of bi-functional cross-linking agents results in formation
of stable heterogeneous biocatalyst which offers distinct benefits
over conventionally immobilized enzyme. Immobilized enzymes
are ‘carrier-bound biocatalysts’ and the presence of a large propor-
tion of non-catalytic carrier (about 90–99% of total mass) causes
dilution of their volumetric activity. On the other hand, cross-linked
enzymes are referred as ‘carrier-free biocatalysts’ and express very
high catalytic activity per unit volume thereby maximizing volu-
metric productivity and space-time yields [16].
The carrier-free cross-linked biocatalyst can be obtained by
either of following three strategies namely: (i) direct cross-linking
of free enzyme which gives Cross-Linked Enzymes (CLE); (ii) cross-
linking of crystalline enzyme which yields Cross-Linked Enzyme
Crystals (CLEC) and (iii) cross-linking of physically aggregated
enzyme which yields Cross-Linked Enzyme Aggregates (CLEA) [15].
More often than not, CLE strategy suffers from several limitations
such as low activity retention, poor reproducibility, poor mechan-
ical stability and difficulty in handling the gelatinous CLE. Need of
highly pure enzyme and crystallization protocols (which are often
∗
Corresponding author. Tel.: +91 20 25903057; fax: +91 20 25902612.
E-mail address: bk.vaidya@rediffmail.com (B.K. Vaidya).
1381-1177/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2011.10.003

B.K. Vaidya et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 184–191
185
expensive, laborious and time consuming) are major limitations of
CLEC strategy. Aforementioned limitations of CLE and CLEC can be
eliminated by means of a strategy called as CLEA [17].
the enzyme precipitate and the mixture was kept under stirring
for 1 h. Then the volume was doubled by adding 100 mM sodium
bicarbonate buffer (pH 10) and a total amount of 75 mg of sodium
borohydride powder was added to reduce the Schiff’s bases formed.
After 15 min, an additional 75 mg of sodium borohydride powder
was added and allowed to react for 15 min. The resultant precip-
itate of AP-CLEA was repeatedly washed with sodium phosphate
buffer (100 mM, pH 7) and centrifuged at 12,000 rpm for 15 min.
Finally the CLEA were dried at 50 ◦C for 24 h using vacuum oven to
remove residual moisture.
Synthesis of CLEA involves precipitation or aggregation of an
enzyme (not necessarily in its pure form) followed by chem-
ical cross-linking of the resulting enzyme aggregates [18,19].
Precipitation/aggregation is generally induced by addition of a pre-
cipitant (acid, salt, organic solvent or non-ionic polymer) to an
aqueous solution of the enzyme. These physical aggregates are
supramolecular structures held together by non-covalent bonding
and re-dissolve when the precipitant is removed. Cross-linking of
these aggregates gives CLEA which remain insoluble in the absence
of the precipitant.
2.3. Optimization studies
Only external amino groups (i.e. amino groups mostly of lysine
which are available on the surface) of an enzyme can participate
in the process of cross-linking. Hence, enzymes having suffi-
cient number of external amino groups (typically electropositive
enzymes) undergo effective cross-linking and form stable CLEA.
On the other hand, enzymes having low number of external amino
groups (typically electronegative enzymes such as l-aminoacylase)
undergo inadequate cross-linking and form mechanically fragile
CLEA that often release enzyme into the reaction medium during a
biocatalytic reaction [20].
Interestingly, Wilson et al. have demonstrated that if the co-
aggregation of enzyme is induced with polyethyleneimine (PEI)
then the cross-linking efficiency of the enzyme can be improved
significantly [21]. PEI is water soluble, cationic polymer consist-
ing of large number of terminal amino groups. The co-aggregation
of enzyme with PEI allows the extension of polymer branches
(having terminal amino groups) closer to some of the embedded
amino groups of enzyme favouring cross-linking between them
[20,22]. Thus, besides external amino groups, a few embedded
amino groups (which are otherwise not accessible during conven-
tional cross-linking procedures) can be utilized in the formation
of stable intra- and inter-molecular cross-links. Thus, the PEI
induced co-aggregation technique for synthesizing CLEA is highly
advantageous strategy especially for electronegative enzymes like
l-aminoacylase.
Different process parameters such as enzyme–PEI ratio, glu-
taraldehyde concentration and cross-linking time were optimized
on the basis on two assessment parameters: namely, activity recov-
ery (Eq. (1)) and aggregation yield (Eq. (2)) as described earlier
[22].
ꢀ
ꢁ
ACLEA
Activity recovery =
× 100
(1)
(2)
AFree × VFree
ꢂ
ꢀ
ꢁꢃ
AResidual × VResidual
Aggregation yield = 100 −
× 100
AFree × VFree
where ACLEA is activity expressed by AP-CLEA; AFree is activity of free
enzyme (U/mL); VFree is volume (mL) of free enzyme used for prepa-
ration of AP-CLEA; AResidual is activity (U/mL) of residual enzyme
solution; and VResidual is volume (mL) of residual enzyme solution
remained after formation of CLEA. All experiments were performed
at least in triplicate and the results are presented as their mean
value. Standard deviation of results never exceeded 5%.
2.4. Characterization of AP-CLEA
2.4.1. Physical properties of AP-CLEA
The surface morphology of AP-CLEA was studied by scanning
electron microscopy (SEM). Micrographs were taken on a JEOL
JSM-5200 SEM instrument. Pore size and pore volume of AP-CLEA
were determined by mercury intrusion porosimetry using Auto-
scan 60 Mercury Porosimeter (Quantachrome, USA) in the range of
0–4000 kg/cm2.
The present study explores the feasibility of the PEI induced
co-aggregation technique for synthesizing CLEA of A. melleus
l-aminoacylase. Herein, the enzyme was co-aggregated with
polyethyleneimine and subsequently cross-linked with glutaralde-
hyde to obtain aminoacylase–polyethyleneimine cross-linked
enzyme aggregates (termed as AP-CLEA). Furthermore, AP-CLEA
were systematically characterized with respect to their physical
properties, catalytic stability and enantioselectivity.
2.4.2. Study on release of enzyme subunit from AP-CLEA
The stability of AP-CLEA against release of enzyme subunit(s)
from the aggregates was evaluated according to the method
described earlier by López-Gallego et al. [20]. Both free aminoa-
cylase and AP-CLEA were boiled separately (∼95 ◦C) in 2 volumes
of 2% sodium dodecyl sulfate (SDS). Then, supernatant of AP-CLEA
and supernatant of free aminoacylase were subjected to SDS-
polyacrylamide gel electrophoresis (SDS-PAGE), the gel was stained
with silver stain and analyzed by densitometry.
2. Materials and methods
2.1. Materials
l-Aminoacylase and N-acetyl-l-methionine were purchased
from Fluka Chemicals, USA. PEI was purchased from Sigma–Aldrich,
USA. Unnatural amino acids (namely: phenylglycine, homopheny-
lalanine and 2-naphthylalanine) and derivatives thereof were
purchased from Bachem Chemicals, Switzerland. All other chem-
icals were of analytical grade and purchased from Merck India Ltd.
2.4.3. Thermal stability of AP-CLEA
Free aminoacylase and AP-CLEA were incubated in phosphate
buffer (50 mM, pH 8) at different temperatures ranging between 20
and 90 ◦C and their activities were determined as described else-
where in the manuscript. The residual activities were calculated
as a percentile ratio of the activity of enzyme after incubation to
the activity enzyme at the optimum temperature. A plot of residual
activity against temperature was obtained to compare the thermal
stability of AP-CLEA over that of free aminoacylase.
2.2. Preparation of AP-CLEA
AP-CLEA were prepared by method described by López-Gallego
et al. [20]. To 25 mL solution of PEI (750 kDa, 25 mg/mL) 25 mL
of aminoacylase (25 mg/mL) solution was added under agitation.
The mixture was left under gentle stirring for 10 min. After 10 min,
2.0 mL of glutaraldehyde solution (25%, v/v) was added to cross-link
2.4.4. Thermal deactivation kinetics of aminoacylase before and
after cross-linking
Kinetics of thermal deactivation of free aminoacylase and AP-
CLEA were studied at different temperatures ranging between 30 ◦C

186
B.K. Vaidya et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 184–191
and 70 ◦C in a shaking water bath (Julabo Ltd.). The samples were
withdrawn at regular time interval, rapidly cooled to 37 ◦C and
immediately analyzed for the residual aminoacylase activity. The
thermal deactivation rate constant (kd) was determined from the
Arrhenius equation (Eq. (3)) [23].
ꢀ
ꢁ
Ed
ln kd = ln A −
(3)
RT
where Ed is the deactivation energy and R is the universal gas con-
stant (8.314 × 10−3 kJ kmol−1 K−1). Natural logarithm of percentile
residual activities of different temperatures was plotted against
time wherein the slope of lines indicates the values of deacti-
vation rate constant, kd at the corresponding temperature [24].
Half-life of free aminoacylase and AP-CLEA for each temperature
was calculated from their respective linear trend line equations.
Further, the Arrhenius plot, i.e. plot of natural logarithm of kd (ln kd)
versus reciprocal of temperature in Kelvin scale (1/T) was obtained
wherein slope of line indicates the deactivation energy (Ed). The
difference in the deactivation energies (ꢀEd) of free aminoacylase
and AP-CLEA was calculated to quantify the thermal stability before
and after cross-linking of the enzyme.
Fig. 1. Schematic representation of stirred cell reactor; 1: membrane support, 2:
outlet, 3: membrane, 4: magnetic stirrer, 5: reaction mixture containing AP-CLEA,
6: sample port and 7: pressure gauze.
collected, washed thrice with cold acidified water and dried at 60 ◦C
under vacuum.
2.4.5. Storage stability of AP-CLEA
Free aminoacylase and AP-CLEA were incubated in phosphate
buffer (50 mM, pH 8) at 4 ◦C and their activities were determined
as described elsewhere in the manuscript at regular time intervals
up to 168 h. The storage stability of free enzyme and AP-CLEA was
compared from the plot of log residual activity against storage time.
2.6. Analytical methods
2.6.1. Aminoacylase activity assay
The activity of aminoacylase was measured using the standard
hydrolytic assay of N-acetyl l-methionine as described earlier [13].
One unit (U) is defined as amount of aminoacylase necessary to
liberate 1 ␮mol of l-methionine per hour at 37 ◦C at pH 8. The activ-
ity of AP-CLEA was determined using 0.1 g of CLEA. The activity of
AP-CLEA is expressed in terms of aminoacylase units per gram of
AP-CLEA.
2.5. AP-CLEA catalyzed chiral resolution of unnatural amino acid
derivatives
2.5.1. General procedure for chiral resolution
The biotransformation reaction was conducted in 25 mL stop-
pered flasks. Racemic substrate (0.01 g) was dissolved in 10 mL
phosphate buffer (50 mM phosphate buffer, pH 8). Co+2 ions act as
cofactor for aminoacylase enzyme therefore CoCl2 (1.19 mg) was
added in the reaction mixture. The biotransformation reaction was
initiated by adding 0.05 g of AP-CLEA. The flask was incubated on
oscillatory shaker at 30 ◦C at 120 rpm. The samples were periodi-
cally withdrawn and analyzed by HPLC.
2.6.2. HPLC analysis
All reaction profiles were monitored by HPLC (Thermo Sep-
aration Products, Fremont, CA, USA). The quantitative analysis
of different amino acids and their esters was carried out using
a reverse phase C-18 column (125 mm × 4 mm, prepacked col-
umn supplemented with a 4 mm × 4 mm guard column procured
from LiChrospher®, Merck, Darmstadt, Germany) eluted isocrat-
ically using acetonitrile–water mobile phase (30:70, v/v) at a
flow rate of 0.6–0.8 mL/min. The enantiomeric excess (e.e.) of
the product was determined by using a chiral CHIROBIOTIC® col-
umn (250 mm × 4.6 mm prepacked column supplemented with
20 mm × 4 mm guard column procured from Astec Inc., USA) eluted
isocratically using water–methanol mobile phase (40:60, v/v) at a
flow rate of 0.6–0.8 mL/min. Analytes were detected using an UV
detector at 215 nm.
2.5.2. Chiral resolution of (RS)-homophenylalanine amide in
stirred cell reactor
The biotransformation reaction was carried out in a laboratory
scale Millipore stirred cell reactor (XFUF04701, Millipore Inc., USA).
A schematic diagram of stirred cell reactor assembly is shown in
Fig. 1. A macroporous filtration fabric cloth (47 mm diameter) was
placed on the membrane support. The reactor was charged with
0.1 g of (RS)-homophenylalanine amide and 25 mL of 50 mM of
phosphate buffer (pH 7) containing 2.98 mg of CoCl2. The biotrans-
formation reaction was initiated by addition of 0.1 g of AP-CLEA.
The temperature of the reaction mixture within the reactor was
maintained at 30 ◦C under constant stirring at 200 rpm. The sam-
ples were periodically withdrawn and analyzed by HPLC. After
about 45% conversion, the reaction was stopped by separating
the reaction mixture. The reactor was thoroughly washed twice
with phosphate buffer (50 mM, pH 7.0) to ensure the complete
removal of product and unreacted substrate. The washings and
reaction mixture were collected together and the pH of the resul-
tant solution was adjusted to 9 with aqueous 2 M NaOH solution.
The unreacted substrate, i.e. (R)-homophenylalanine amide was
extracted with dichloromethane (3× 15 mL). The remaining aque-
ous solution was acidified (to pH 6.3) with 2 M HCl solution and
maintained at 0–2 ◦C. After ∼2 h, the precipitate of (S)-HPA was
2.6.3. Determination of enantiomeric ratio
Enantioselectivity of AP-CLEA was calculated in terms of enan-
tiomeric ratio (E) by using Chen’s formula as described earlier [25].
The value of E mentioned in the present manuscript represents an
average value of triplicate observations where the standard error
was not more than 5%.
3. Results and discussion
3.1. Optimization of process parameters
‘Extent of cross-linking’ is the key parameter which decides
an activity of CLEA (calculated in terms of activity recovery) and
an extent of enzyme aggregation (calculated in terms of aggre-
gation yield). Hence the ‘extent of cross-linking’ in CLEA should

B.K. Vaidya et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 184–191
187
Table 1
Table 3
Effect of enzyme:PEI ratio.
Effect of cross-linking time.
Enzyme:PEI ratio
Activity recovery (%)
Aggregation yield (%)
Cross-linking time (h)
Activity recovery (%)
Aggregation yield (%)
1:3
1:2
1:1
2:1
3:1
1:0
23.3
33.1
50.4
44.0
37.5
6.9
68.0
62.6
60.4
42.8
34.6
11.3
1
6
12
24
36
48
45.0
58.2
65.5
74.9
74.5
70.9
51.2
74.2
79.8
81.2
81.6
82.3
within 24 h where maximum activity recovery (74.9%) and aggre-
gation yield (81.2%) were obtained. Beyond 24 h, small decrease
in the activity recovery was observed while the aggregation yield
remained almost unchanged.
be critically controlled. Adequate cross-linking is essential to form
stable enzyme aggregates. Excessive cross-linking however, inacti-
vates enzyme as well as lowers the porosity of CLEA consequently
decreases the activity recovery [15]. To summarize, up to the opti-
mum point, the activity recovery and the aggregation yield increase
with extent of cross-linking. However, additional cross-linking
leads to decrease in the activity recovery. The aggregation yield
may increase or remain constant beyond the optimum. The extent
of cross-linking in the present study was controlled by optimizing
three process variables namely enzyme to PEI ratio, concentration
of glutaraldehyde and time of glutaraldehyde treatment.
3.2. Characterization of AP-CLEA
3.2.1. Physical properties of AP-CLEA
Scanning electron micrographs of AP-CLEA are given in Fig. 2.
The SEM of AP-CLEA shows the cross-linked aggregates of enzyme
and PEI. The SEM also indicates the highly porous structure of the
aggregate. The average particle size of AP-CLEA as estimated from
SEM data was 50 10 ␮m. The values of pore surface area and pore
volume of AP-CLEA, as determined by mercury porosimetry, were
66 0.5 m2/g and 0.5 0.01 cm3/g respectively.
3.1.1. Effect of enzyme:PEI ratio
To evaluate the effect of enzyme–PEI ratio, aminoacylase and PEI
solutions (25 mg/mL each) were mixed in different concentrations
(1:3, 1:2, 1:1, 2:1 and 3:1) and the resultant AP-CLEA were ana-
lyzed for the activity recovery and aggregation yield as described
elsewhere in the manuscript. Table 1 gives the effect of enzyme:PEI
ratio. Among the different enzyme–PEI ratios studied, 1:1 enzyme
to PEI ratio was found to be optimum and gave maximum for activ-
ity recovery (50.4%) and therefore 1:1 enzyme to PEI ratio was
used in all further experiments. It is interesting to note that, in
the absence of PEI poor values of activity recovery and aggregation
yield (i.e. 6.9% and 11.3% respectively) were obtained.
3.1.2. Effect of glutaraldehyde concentration
Different concentrations of glutaraldehyde (100–600 mM) were
used to determine the optimum quantity required for stable
cross-linking of aminoacylase and PEI. The activity recovery and
aggregation yield of AP-CLEA were found to increase initially with
increase in glutaraldehyde concentration from 100 to 500 mM
(Table 2). Further increase in glutaraldehyde concentration above
500 mM resulted in small decrease in the activity recovery with no
significant change in the aggregation yield. The small decrease in
the activity recovery could possibly be due to an excessive cross-
linking of enzyme molecules making them catalytically inactive.
Glutaraldehyde concentration of 500 mM gave maximum activity
recovery (61.3%) and aggregation yield (74.8%) and hence used in
all further experiments.
3.1.3. Effect of cross-linking time
The cross-linking time was studied from 1 h to 48 h for sta-
ble cross-linking of aminoacylase and PEI (Table 3). The increase
in cross-linking time resulted in significant increase in the activ-
ity recovery of AP-CLEA. The cross-linking reached to optimum
Table 2
Effect of glutaraldehyde concentration.
Glutaraldehyde (mM)
Activity recovery (%)
Aggregation yield (%)
100
200
300
400
500
600
50.9
55.8
58.0
60.0
61.3
56.9
60.8
63.3
68.0
70.2
74.8
74.7
Fig. 2. SEM of AP-CLEA: (a) individual CLEA particle observed at lower magnification
(1000×); (b) macroporous nature of CLEA observed at higher magnification (2500×).

188
B.K. Vaidya et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 184–191
100
80
60
40
20
0
30
40
50
60
70
80
90
Temperature, o
C
Free aminoacylase
AP-CLEA
Fig. 3. Thermal stability of AP-CLEA.
3.2.2. Study on release of enzyme subunit from AP-CLEA
Boiling of free enzyme with SDS causes dissociation of the pro-
tein into its polypeptide subunits. A. melleus aminoacylase is dimer
and each subunit is having approximate molecular weight of 37 kDa
[26]. The supernatant collected after boiling of free enzyme with
SDS gave two distinct bands on the gel at around 37 kDa which is
in agreement with the published data (results not shown). Boiling
of AP-CLEA with SDS allows dissociation of non-covalently bound
enzyme subunits (if any) from the aggregates into the supernatant
solution. Interestingly, the supernatant of AP-CLEA did not give any
band indeed indicate the absence of non-covalently bound enzyme
subunits in AP-CLEA (results not shown). Moreover, these results
highlight the potential of said method for stable and effective cross-
linking of multimeric enzymes.
Fig. 4. Thermal deactivation of (a) free aminoacylase and (b) AP-CLEA in tempera-
ture range of 30–70 ◦C.
3.2.3. Thermal stability of AP-CLEA
The activity profiles of free aminoacylase and AP-CLEA at differ-
ent temperatures are represented in Fig. 3. At 50 ◦C, the free enzyme
retained only 66.4% residual activity while AP-CLEA 98.5% of its ini-
tial activity. Likewise, at 70 ◦C the free enzyme retained only 4.7%
residual activity while AP-CLEA was found to retain 76.6% of its ini-
tial activity. Thus, the cross-linking of aminoacylase with PEI was
observed to confer excellent thermal stability to the enzyme.
mainly ascribed to inter- and intramolecular cross-links which
maintain the catalytically active conformation of enzyme and also
prevent the dissociation of multimeric enzyme into the individual
subunits at the elevated temperatures [21,22].
3.2.5. Storage stability of AP-CLEA
The storage stability of AP-CLEA was determined and compared
with free aminoacylase. The plot of log residual activity verses stor-
age time is shown in Fig. 6. The linear trend lines were drawn to
the activity profiles of free aminoacylase and AP-CLEA. The stor-
age half-life of each enzyme preparation was determined from its
3.2.4. Thermal deactivation kinetics of aminoacylase before and
after cross-linking
The temperature dependant loss of enzyme activity of free
aminoacylase and CLEA is shown in Fig. 4(a) and (b) respectively.
From these temperature dependant activity profiles, the Arrhe-
nius plot was obtained. The Arrhenius plot for free aminoacylase
and CLEA is shown in Fig. 5. The slope of Arrhenius plot indicates
the deactivation energy (Ed). The deactivation energy of AP-CLEA
and free aminoacylase were found to be −8281.1 kJ kmol−1 K−1
and −9369.3 kJ kmol−1 K−1. When an enzyme forms cross-linked
aggregates, a large number of stabilizing linkages are formed
between individual enzyme molecules. Energy must be put into
the system in order to disrupt these stabilizing linkages. As a
result, CLEA has higher deactivation energy than free enzyme
[24]. Higher deactivation energy of AP-CLEA than that of free
enzyme (ꢀEd = 1088.2 kJ kmol−1 K−1) indicates the excellent ther-
mal stabilization of the cross-linked enzyme. The half-life (t1/2) and
deactivation rate constants (kd) of free aminoacylase and AP-CLEA
are given in Table 4. On an average, upon cross-linking, there was
4.4 fold increase in the half-life of enzyme.
Co-aggregation followed by cross-linking is known to confer
better thermal stability to the enzyme than conventional immo-
bilization procedures. The increased thermal stability of CLEA is
Fig. 5. Arrhenius plot for free aminoacylase and AP-CLEA.

B.K. Vaidya et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 184–191
189
Table 4
Thermal deactivation coefficient (kd) and half-life (t1/2) of free enzyme and AP-CLEA.
Temperature (◦C)
kd (min−1
)
t1/2 (min)
Fold increase
in t1/2
Free enzyme
AP-CLEA
Free enzyme
AP-CLEA
30
40
50
60
70
−0.00141
−0.00319
−0.00840
−0.01979
−0.05262
−0.00038
−0.00081
−0.00189
−0.00404
−0.01015
496.45
219.43
83.33
35.37
13.30
1842.11
864.19
370.37
173.26
68.96
3.7
3.9
4.4
4.9
5.1
4.4
Average of fold increase in the half-life over the range of 30–70 ◦
C
respective trend line equation. The storage half-life of free aminoa-
cylase at 4 ◦C was found to be ∼1 day while that of AP-CLEA was
found to be ∼40 days. Thus, the present method of preparation of
CLEA conferred extended storage life to the enzyme.
prepared the CLEA of glutaryl acylase using the PEI induced co-
aggregation technique wherein the use of PEI prevented the release
of enzyme molecules from the aggregate. Besides this, the PEI
induced CLEA showed enhanced thermal and storage stability [20].
Wilson et al. employed this co-aggregation technique for syn-
thesizing CLEA of Penicillin-G acylase to enhance the stability of
enzyme in organic medium. Here the CLEA prepared by PEI co-
aggregation technique showed increased stability (∼25 fold higher)
than the CLEA prepared by the conventional method [21]. The co-
aggregation method has also been used to prepare stable CLEA of
Alcaligenes sp. lipase and Candida antarctica B lipase. The catalytic
properties of the lipase CLEA (such as specificity and enantiose-
lectivity) were greatly modulated by the nature of co-aggregating
polymers [22]. Pan et al. have recently reported the usefulness of PEI
induced method of preparing CLEA for Serratia marcescens lipase. In
this, CLEA showed enhanced thermal stability and excellent opera-
tional stability in repetitive use in aqueous-toluene biphasic system
for asymmetric hydrolysis of trans-3-(4ꢀ-methoxyphenyl)glycidic
acid methyl ester [30]. Mateo et al. prepared the co-aggregates of
the nitrilases with high molecular weight PEI. Here, PEI induced co-
aggregation technique increased the oxygen-resistance of nitrilase
than free enzyme and conventionally cross-linked enzyme [31].
In vitro stability of enzymes remains a critical issue in the field
of biocatalysis. The practical usefulness of an enzyme is often
decided by its storage and operational stability. Analogous to pre-
vious reports, the present study manifests that the PEI induced
co-aggregation method of synthesizing CLEA confers great stability
to the enzyme. In AP-CLEA, enzyme molecules are presumably held
together by multipoint enzyme–PEI linkages which would have sta-
bilized the native (and active) conformation of the enzyme against
denaturizing forces. Besides this, the improved stability of AP-CLEA
is complemented by the protein-protective nature of PEI. PEI in par-
ticular, is known to impede the process of oxidation of sulfhydryl
groups of enzyme (which is a prevailing mechanism of enzyme
inactivation during storage) [32].
The cross-linking of aminoacylase has been achieved by dif-
ferent methods [26–29]. Bode et al. employed ethylene glycol
dimethyl ether (as precipitant) and glutaraldehyde (as a cross-
linking agent) to prepare aminoacylase CLEA. Interestingly this
report also states that commercial l-aminoacylase is a mixture
of several hydrolases [26]. Immobilization of thermophilic l-
aminoacylase from T. litoralis was studied by Hickey et al. [27].
Here, two immobilization approaches were used and both involved
cross-linking of l-aminoacylase. In the first approach, enzyme was
attached to monoliths while in the second approach previously
cross-linked enzyme was trapped using frits in the micro-fluidic
channels. The cross-linked immobilized l-aminoacylase was suc-
cessfully used in the form of miniaturized flow reactors wherein
the enzyme exhibited excellent operational stability. Honda et al.
reported a novel strategy wherein poly-l-lysine was used as a
booster for attaining efficient cross-linking of aminoacylase [28].
Recently, Dong et al. have demonstrated that use of bovine serum
albumin facilitates cross-linking of aminoacylase thereby allow-
ing formation of catalytically stable CLEA of the enzyme [29].
To the best of our knowledge, PEI induced co-aggregation tech-
nique has not yet been studied for synthesizing aminoacylase
CLEA.
The PEI induced co-aggregation technique for synthesizing CLEA
is yet to be studied extensively. Till date, this technique of syn-
thesizing CLEA has been studied for limited number of enzymes
viz. glutaryl acylase [20], penicillin-G acylase [21], lipases [22,30]
and Pseudomonas fluorescens nitrilase [31]. López-Gallego et al.
3.3. AP-CLEA catalyzed chiral resolution
Aminoacylases have the ability to hydrolyze various amino acid
derivatives (such as amino acid esters, amino acid amides and
N-acetyl amino acids) in an enantioselective manner to yield enan-
tiopure amino acids. AP-CLEA were employed for enantioselective
synthesis of three unnatural amino acids (namely: phenylglycine,
homophenylalanine and 2-naphthylalanine) via chiral resolution
of their ester-, amide- and N-acetyl derivatives. These unnatural
amino acids are key chiral components of variety of pharmaceutical
agents. For instance, phenylglycine is used in synthesis of ampi-
cillin; (S)-homophenylalanine is a central chiral intermediate for
synthesis of variety of antihypertensive agents (benzpril, enalapril,
lisinopril, etc.); and (S)-2-naphthylalanine is a component of a pep-
tide drug nafareline, which is used in symptomatic treatment of
endometriosis [33].
Fig. 6. Storage stability of AP-CLEA (Linear trend line equations were used to cal-
culate the respective storage half-lives. The horizontal line (—) at log(50) is drawn
merely for a graphical representation of the storage half-lives.).

190
B.K. Vaidya et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 184–191
Table 5
AP-CLEA catalyzed chiral resolutions.
Substrate
Preferred configuration
eep (%)
C (%)
E
(a) Chiral resolution of amino acid esters
(RS)-Phenylglycine ethyl ester
(RS)-Homophenylalanine ethyl ester
(RS)-2-Naphthylalanine ethyl ester
(b) Chiral resolution of amino acid amides
(RS)-Phenylglycine amide
S
S
S
64.7
82.4
67
35.2
27.9
41.7
6.5
14.1
8.1
S
S
S
98.5
98.2
96.4
44.2
41.6
43.9
>200
>200
124.4
(RS)-Homophenylalanine amide
(RS)-2-Naphthylalanine amide
(c) Chiral resolution of N-acetyl amino acids
(RS)-N-acetyl-phenylglycine
(RS)-N-acetyl-homophenylalanine
(RS)-N-acetyl-2-naphthylalanine
S
S
S
92.5
93.5
87.4
43.3
41.8
32.9
54.3
60.0
22.6
Table 6
Summary of five repetitive batches of hydrolysis of rac-HPA-amide.
Batch number
eep (%)
C (%)
E
Yield of
% Yield
(S)-homophenylalanine (mg)
1
2
3
4
5
Total
98.5
98.0
98.6
98.2
98.3
98.3
44.3
45.4
44.1
44.3
44.0
–
>200
>200
>200
>200
>200
–
42.6
43.7
43.0
42.4
42.2
213.9
84.7
86.9
85.5
84.3
83.9
85.0
3.3.1. Chiral resolution of amino acid esters
To summarize, enantiomeric ratios of the AP-CLEA were
The results of AP-CLEA catalyzed enantioselective hydroly-
sis of amino acid esters are given in Table 5(a). AP-CLEA were
observed to preferentially hydrolyze the S-enantiomer of amino
acid esters. The enantiomeric ratios of AP-CLEA towards hydrolysis
of phenylglycine ethyl ester, homophenylalanine ethyl ester and
2-naphthylalanine ethyl ester were 6.5, 14.1 and 8.1 respectively.
excellent (E ≈ 124–200) towards hydrolysis of amides, moderate
(E ≈ 23–60) towards hydrolysis of N-acetyl derivatives and poor
(E < 15) towards the hydrolysis of amino acid esters. Youshko
et al. reported similar observations wherein free A. melleus l-
aminoacylase enzyme gave higher enantiomeric ratios towards
hydrolysis of phenylglycine amide and homophenylalanine amide
(E > 200) than that towards hydrolysis of N-acetyl-phenylglycine
and N-acetyl-homophenylalanine (E = 96 and 70 respectively) [6].
3.3.2. Chiral resolution of amino acid amides
The results of AP-CLEA catalyzed enantioselective hydrolysis
of amino acid amides are given in Table 5(b). AP-CLEA were
observed to preferentially hydrolyze the S-enantiomer of amino
acid amides. The E values of hydrolysis of phenylglycine amide
and homophenylalanine amide were >200 and that of hydrolysis
of 2-naphthylalanine amide was >100 indicating the remarkable
enantioselectivity of CLEA towards these hydrolytic reactions.
3.4. Operational stability of AP-CLEA
AP-CLEA catalyzed chiral resolution of (RS)-homophenylalanine
amide was conducted in repetitive batch mode. The initial activ-
ity of AP-CLEA was 211.7 U g−1 and that after the fifth batch was
196.8 U g−1. Thus, AP-CLEA were found to retain more than 92% of
the initial activity after five consecutive batches which indicates
an adequate operational stability of AP-CLEA. In the first batch, the
maximum of 98.5% enantio-enrichment of the product (i.e. eep) was
obtained at the reaction time of 70 min. In order to maintain the
maximum possible enantio-enrichment of the product, the reac-
tion time of all consecutive batches was carefully monitored. From
five repeated batches, 213.9 mg of (S)-homophenylalanine having
enantiomeric excess of 98.3% was obtained (Table 6).
3.3.3. Chiral resolution of N-acetyl amino acids
The results of AP-CLEA catalyzed enantioselective hydroly-
sis of N-acetyl amino acids are given in Table 5(c). AP-CLEA
were observed to preferentially hydrolyze the S-enantiomer of
N-acetyl amino acids. CLEA exhibited higher enantioselectivity
towards hydrolysis of N-acetyl-phenylglycine and N-acetyl-
homophenylalanine as indicated from the higher E values (54.3
and 60.0 respectively) than that towards hydrolysis of N-acetyl-
2-naphthylalanine (E = 22.6).
4. Conclusions
The process of co-aggregation followed by cross-linking occa-
sionally alters the catalytic properties (e.g. substrate specificity,
enantioselectivity) of an enzyme. For instance, drastic alterations
in the enantioselectivity of lipase were observed when the enzyme
was cross-linked by PEI-induced co-aggregation technique [22].
The extent of change in the catalytic properties of enzyme pos-
sibly depends on the nature of polymer, nature of enzyme and
type of protein–polymer interactions. Considering the possibility
of enantioselectivity modulation, we compared the enantiomeric
ratios of free aminoacylase with that of AP-CLEA towards hydroly-
sis of unnatural amino acid amides. For the given amide substrate,
no significant difference in the enantiomeric ratio of free enzyme
and that of AP-CLEA was observed.
Co-aggregation of enzyme with polyethyleneimine (PEI) facili-
tates formation of stable Cross-Linked Enzyme Aggregates (CLEA)
of l-aminoacylase. The said method of co-aggregation and cross-
linking confers significantly higher temperature and storage
stability to the enzyme without affecting its enantioselectivity.
Aminoacylase-PEI CLEA (termed as AP-CLEA) can catalyze enantios-
elective hydrolysis of amino acid amides, N-acetyl amino acids and
amino acid esters. The enantioselectivity of AP-CLEA was remark-
ably high for hydrolysis of amino acid amides; was moderate for
hydrolysis of N-acetyl amino acids and was poor for hydrolysis of
amino acid esters.

B.K. Vaidya et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 184–191
191
Acknowledgement
[15] J.J. Roy, T.E. Abraham, Chem. Rev. 104 (2004) 3705–3721.
[16] R.A. Sheldon, R. Schoevaart, L.M. van Langen, Biocatal. Biotransform. 23 (2005)
141–147.
[17] R. Schoevaart, M.W. Wolbers, M. Golubovic, M. Ottens, A.P.G. Kieboom, F. van
Rantwijk, L.A.M. van der Wielen, R.A. Sheldon, Biotechnol. Bioeng. 87 (2004)
754–761.
BKV would like to acknowledge the Council of Scientific and
Industrial Research (CSIR), India for the senior research fellowship.
[18] L.Q. Cao, F. van Rantwijk, R.A. Sheldon, Org. Lett. 2 (2000) 1361–1364.
[19] L. Cao, F. van Langen, F. van Rantwijk, R.A. Sheldon, J. Mol. Catal. B: Enzym. 11
(2001) 665–670.
[20] F. López-Gallego, L. Betancor, A. Hidalgo, N. Alonso, R. Fernández-Lafuente, J.M.
Guisán, Biomacromolecules 6 (2005) 1839–1842.
References
[1] A. Bommarius, K. Drauz, C. Wandrey, Ann. N. Y. Acad. Sci. 672 (1992) 126–136.
[2] S.V. Story, A.M. Grunden, M.W.W. Adams, J. Bacteriol. 183 (2001) 4259–4268.
[3] H.S. Toogood, E.J. Hollingsworth, R.C. Brown, I.N. Taylor, S.J.C. Taylor, R.
McCague, J.A. Littlechild, Extremophiles 6 (2002) 111–122.
[21] L. Wilson, A. Illanes, O. Abian, B. Pessela, R. Fernández-Lafuente, J.M. Guisán,
Biomacromolecules 5 (2004) 852–857.
[4] A.S. Bommarius, K. Drauz, U. Groeger, C. Wandrey, in: A.N. Collins, G.N. Shel-
drake, J. Crosby (Eds.), Chirality in Industry, John Wiley & Sons Inc., New York,
1992, pp. 371–397.
[22] L. Wilson, G.F. Fernández-Lorente, R.F. Fernández-Lafuente, A. Illanes, J.M.
Guisán, J.M. Polomo, Enzyme Microb. Technol. 39 (2006) 750–755.
[23] M.L. Shuler, F. Kargi, Bioprocess Engineering: Basic Concepts, 2nd ed., Prentice-
Hall, New Jersey, 2002.
[24] M.D. Gouda, S.A. Singh, A.G.A. Rao, M.S. Thakur, N.G. Karanth, J. Biol. Chem. 278
(2003) 24324–24333.
[25] A.J.J. Straathof, J.A. Jongejan, Enzyme Microb. Technol. 21 (1997) 559–571.
[26] M.L. Bode, F. van Rantwijk, R.A. Sheldon, Biotechnol. Bioeng. 84 (2003) 710–713.
[27] A.M. Hickey, L. Marle, T. McCreedy, P. Watts, G.M. Greenway, J.A. Littlechild,
Biochem. Soc. Trans. 35 (2007) 1621–1623.
[5] A. Liese, K. Seelbach, C. Wandrey, Industrial Biotransformations, Wiley-VCH,
Weinheim, Germany, 2000.
ˇ
[6] M.I. Youshko, L.M. van Langen, R.A. Sheldon, V.K. Svedas, Tetrahedron: Asym-
metry 15 (2004) 1933–1936.
[7] A. Liljeblad, L.T. Kanerva, Tetrahedron: Asymmetry 10 (1999) 4405–4415.
[8] A. Liljeblad, J. Lindborg, L.T. Kanerva, Tetrahedron: Asymmetry 11 (2000)
3957–3966.
[9] A. Liljeblad, R. Aksela, L.T. Kanerva, Tetrahedron: Asymmetry 12 (2001)
2059–2066.
[28] T. Honda, M. Miyazaki, H. Nakamura, H. Maeda, Adv. Synth. Catal. 348 (2006)
2163–2171.
[10] B. Herradón, S. Valverde, Synlett (1995) 599–602.
[11] P. Noheda, G. García, M.C. Pozuelo, B. Herradón, Tetrahedron: Asymmetry 7
(1996) 2801–2804.
[29] T. Dong, L. Zhao, Y. Huang, X. Tan, Bioresour. Technol. 101 (2010) 6569–6571.
[30] J. Pan, X.-D. Kong, C.-X. Li, Q. Ye, J.-H. Xu, T. Imanaka, J. Mol. Catal. B: Enzym. 68
(2011) 256–261.
[12] J. Faraldos, E. Arroyo, B. Herradón, Synlett (1997) 367–370.
[13] M. Bakker, A.S. Spruijt, F. van Rantwijk, R.A. Sheldon, Tetrahedron: Asymmetry
11 (2000) 1801–1808.
[14] M.I. Youshko, F. van Rantwijk, R.A. Sheldon, Tetrahedron: Asymmetry 12 (2001)
3267–3271.
[31] C. Mateo, B. Fernandes, F. van Rantwijk, A. Stolz, R.A. Sheldon, J. Mol. Catal. B:
Enzym. 38 (2006) 154–157.
[32] M.M. Andersson, R. Hatti-Kaul, J. Biotechnol. 72 (1999) 21–31.
[33] J.S. Ma, Chimica Oggi/Chemistry Today June (2003) 65–68.