Engineering actively magnetic crosslinked inclusion bodies of Candida antarctica lipase B: An efficient and stable biocatalyst for enzyme-catalyzed reactions

Article abstract of DOI:10.1016/j.mcat.2021.111467

In our previous study, the soluble expression of recombinant Candida antarctica lipase B (rCalB) in E. coli is achieved through fusion with polyamino acid tails. The present study showed that even in the form of inclusion bodies (rCalB-IBs), rCalB-IBs wer

Full text of DOI:10.1016/j.mcat.2021.111467

Molecular Catalysis 504 (2021) 111467
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.journals.elsevier.com/molecular-catalysis
Engineering actively magnetic crosslinked inclusion bodies of Candida
antarctica lipase B: An efficient and stable biocatalyst for
enzyme-catalyzed reactions
Yu Han, Xinying Zhang, Liangyu Zheng*
Key Laboratory for Molecular Enzymology and Engineering of Ministry of Education, College of Life Sciences, Jilin University, Changchun 130012, People’s Republic of
China
A R T I C L E I N F O
A B S T R A C T
Keywords:
In our previous study, the soluble expression of recombinant Candida antarctica lipase B (rCalB) in E. coli is
achieved through fusion with polyamino acid tails. The present study showed that even in the form of inclusion
bodies (rCalB-IBs), rCalB-IBs were still active towards enzyme-catalyzed reactions. rCalB-IBs could be directly
Candida antarctica lipase B
Active inclusion bodies
Magnetic enzyme aggregates
Enantioselective hydrolysis reaction
Carbon–carbon bond-forming reaction
3 4
crosslinked by the oxidized dextran in the presence of Fe O nanoparticles to obtain magnetic aggregates (rCalB-
mclIBs), and an activity recovery of 113.9 % ± 2.2 % could be achieved under optimized conditions. The ob-
tained rCalB-mclIBs exhibited excellent thermal and storage stabilities, and a good tolerance in changing pH
environmental. rCalB-mclIBs could be effectively recycled and could retain 85.4 % ± 2.5 % of its initial activity,
and an e.e.
p
of 84.4 % ± 2.2 % in (R, S)-NEMPA-ME hydrolysis was observed after the 10th cycle. After the
application of rCalB-mclIBs into a promiscuous carbon–carbon bond-forming reaction, the MBH and the Aldol
products could be obtained simultaneously. This study provides a simple and inexpensive strategy to highlight
the application potential of rCalB-IBs as a biocatalyst.
1
. Introduction
The Candida antarctica lipase B (CalB) is one of the most widely used
IBs) can also display catalytic abilities, which are similar to those of
soluble expressed rCalBs. rCalB-IBs can be effectively applied in enzyme-
catalyzed reactions, and similar phenomena are also observed by other
groups [13]. Beta-galactosidase IBs in E. coli are first reported to be
enzymatically active with a specific activity of one-third of its soluble
forms [14]. The aggregated recombinant IBs of human dihydrofolate
reductase and fluorescent protein are also reported to be used in reaction
mixtures for efficient catalysis [15]. The lipase A of Bacillus subtilis and
the hydroxynitrile lyase of Arabidopsis thaliana are also successfully
produced as catalytically active IBs and used in aqueous and micro-
aqueous organic solvent-based reaction systems [16]. These studies
break previous ideas that IBs are products of misfolding in the protein
expression and exist in an inactivated form. However, at present, the
precipitated IBs can be an ordered dynamic structure and are aggregated
as “native-like structures” with catalytic activities [14,17,18]. Precipi-
tated IBs can be used as an efficient catalyst for enzyme-catalyzed re-
actions. Unlike soluble proteins, IBs can be directly separated from a
fermentation broth as a single protein, thereby avoiding the tedious
enzyme purification. Furthermore, IBs are in the form of precipitate and
may be used as a similarly immobilized enzyme. If the precipitated IBs
biocatalysts in industrial processes because of its high degree of activity
and selectivity [1,2]. This biocatalyst tolerates varied experimental
conditions, and numerous publications have shown its efficiency in
catalyzing many organic reactions to produce enantiopure compounds
that can be used as chiral building blocks for commercial scale [3–5]. In
terms of its wide industrial application, the achievement of a large
amount of soluble expression of CalB is particularly important. Although
E. coli is an ideal expression host for the expression of exogenous protein,
the soluble CalB cannot be easily achieved in E. coli [6–8]. The low
protein solubility of recombinant enzymes in E. coli is a major factor
hindering their application [9,10]. Recent research (also including our
work) has shown that recombinant CalBs (rCalBs) with polycationic
amino acids, such as n-histidines (nHis), n-arginines (nArg), or n-lysines
(
nLys) at N and C terminals, can enhance their soluble expression in
E. coli [11,12].
In the process of our efforts to achieve the soluble expression of
CalBs, we have found that the expressed rCalB inclusion bodies (rCalB-
*
Corresponding author.
E-mail address: lyzheng@jlu.edu.cn (L. Zheng).
https://doi.org/10.1016/j.mcat.2021.111467
Received 14 August 2020; Received in revised form 5 February 2021; Accepted 9 February 2021
Available online 24 February 2021
2
468-8231/© 2021 Elsevier B.V. All rights reserved.

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
are rationally modified, the stabilities of modified IBs maybe further
improved.
3 4
magnetic properties of the magnetic Fe O nanoparticles, the resulting
rCalB-mclIBs can be easily controlled and separated from the reaction
mixture through the application of magnetic field, thereby eliminating
the need for centrifugation and filtration.
In the present study, the catalytic abilities of rCalB-IBs are first
verified. The enzyme-catalyzed enantioselective hydrolysis of racemic
(
R, S)-N-(2-ethyl-6-methylphenyl)alanine methyl ester ((R, S)-NEMPA-
Subsequently, the advantages of rCalB-mclIBs as a biocatalyst are
investigated and compared with those of rCalB-clIBs and rCalB-IBs. The
characteristics of rCalB-mclIBs, rCalB-clIBs, and rCalB-IBs are deter-
mined using Scanning electron microscopy (SEM), Fourier transform
infrared (FTIR) spectroscopy and Circular dichroism (CD) spectroscopy.
The thermal, pH, and storage stabilities and reusability of rCalB-mclIBs,
rCalB-clIBs, and rCalB-IBs are studied and compared. The catalytic
abilities of rCalB-mclIBs in the enantioselective hydrolysis of (R, S)-
NEMPA-ME, are also evaluated and compared with those of rCalB-clIBs,
rCalB-IBs, free rCalB, and commercial Novozym 435. The promiscuous
carbon–carbon bond-forming reaction, such as the Morita–Baylis–Hill-
man (MBH) reaction between 4-nitrobenzaldehyde and 2-cyclohexen-1-
one, is also selected to evaluate the catalytic abilities of rCalB-mclIBs
and determine the diversity of rCalB-mclIB-catalyzed reactions. The
MBH reaction is selected here because it is an important carbon–carbon
ME) is selected as a model reaction to obtain (S)-N-(2-ethyl-6-methyl-
phenyl)alanine ((S)-NEMPA). The corresponding product (S)-NEMPA is
a key chiral building block for the synthesis of most widely used her-
bicides, such as (S)-metolachlor [19]. The commercial Novozym 435
(
CalB immobilized on a macroporous acrylic resin) is found active to-
wards the enantioselective hydrolysis, and nearly 50 % of (R, S)-NEM-
◦
PA-ME can be hydrolyzed after 2 h at 37 C but displays poor
enantioselective ratio (E value, 10.6) towards the (S)-NEMPA. The
introduction of some additives, such as organic compounds (e.g., diethyl
ether), macrocyclic dioxotetraamine (e.g., n-C
6
H13-MDTA), ionic liquid
(
e.g., [ETOMG]BF ), amide compounds (e.g., formamide), and amino
4
acids (e.g., His or Lys), in the reaction system can enhance the enan-
tioselectivity of Novozym 435 to some extent [20]. On the basis of the
aforementioned additive strategy, the rational enhancement of enzy-
me‑catalyzed enantioselective reaction is subsequently established
through the construction of recombinant enzymes. In this strategy, the
recombinant CalBs (rCalBs) with polycationic amino acids, such as nHis
or nLys at its N and/or C terminals, are used to catalyze the enantiose-
lective hydrolysis of (R, S)-NEMPA-ME, and the E value of the reaction
can be improved from 12.1 to 20.3 [21]. The establishment of the above
enzyme-catalyzed enantioselective hydrolysis reaction and the rational
regulation aiming at the reaction originate from our laboratory. Herein,
we hope to further validate the applicability of rCalB-IBs. Thus, the
familiar model reaction is still being used.
bond-forming reaction between the -carbon of a conjugated carbonyl
α
compound and a carbon electrophile and has become one of the most
useful and popular reactions with enormous synthetic utilities [36,37].
The resulting MBH adducts have many applications in the synthesis of
medicinally relevant compounds and certain complex natural products
[38]. The lipase Novozym 435 is generally considered as a particularly
efficient promiscuous enzyme that can be used to catalyze the formation
of carbon–carbon bonds [39,40]. Hence, Reetz and coauthors have
selected the Novozym 435 as a biocatalyst in the MBH reaction between
4-nitrobenzaldehyde and 2-cyclohexen-1-one. However, the Novozym
435 shows no activity toward the MBH reaction, and the reason for this
inactivity remains unclear [41]. Out of great interest in the reaction, the
rCalB-mclIBs prepared in this study are attempted to be used in the same
MBH reaction and observe if the different forms of CalB, such as
rCalB-mclIBs, have catalytic effects that are different from that of the
Novozym 435. This study aims to confirm the applicability of rCalB-IBs
and construct rCalB-mclIBs with excellent operational stability and
reusability for use as a valuable biocatalyst for enzyme-catalyzed
reactions.
Regarding the possibility that rCalB-IBs may display weak stability in
its application, the crosslinked enzyme aggregate (CLEA) technology is
considered. The preparation of CLEAs includes the precipitation of en-
zymes by using agents, such as inorganic salts or organic solvents
without undergoing denaturation, to form aggregates and subsequently
crosslink the aggregates by using bifunctional agents, such as suitable
dialdehydes via the reaction of amino groups of Lys residues on the
external surface of the enzyme [22–24]. Here, rCalB-IBs fused with
poly-Lys are expressed in a precipitated form, whether it can replace the
precipitation step in the CLEA preparation, and can be directly modified
using the one-step dialdehyde crosslinking process to obtain crosslinked
rCalB-IBs (rCalB-clIBs) by using poly-Lys. If feasible, the obtained
rCalB-clIBs can display the advantages of CLEAs, such as the elimination
of additional carriers, achievement of highly concentrated enzyme ac-
tivity and stability, and recycling several times without appreciable loss
of activity [25,26]. The technological advantages established here are
evident because the wild CalB is hard to be crosslinked by dialdehydes
due to its low surface-reactive amino groups [27], and the CalB-CLEAs
can only be achieved in the crosslinking process by adding feeder pro-
teins that are rich in the Lys residue, such as bovine serum albumin [28].
Herein, the enzyme rCalB-IBs are fused with the poly-Lys tail, and rich
Lys residues may positively affect the crosslinking process, thereby
achieving sufficient crosslinking between rCalB-IBs and dialdehyde.
2. Materials and methods
2.1. Materials
A recombinant plasmid PA1K-pPIC9K containing the original lipase
CalB gene was available in our laboratory. The rCalBs modified by
polyamino acid tails were constructed, and expressed in our laboratory
[11]. p-Nitrophenyl caprylate (pNPC) was purchased from J&K Scien-
3 4
tific Ltd. (Beijing, China). Magnetic Fe O nanoparticles (diameter of 10
nm) were purchased from Aladdin (Shanghai, China). (R,
S)-N-(2-ethyl-6-methylphenyl)alanine methyl ester ((R, S)-NEMPA-ME),
and the product standard (S)-N-(2-ethyl-6-methylphenyl)alanine
((S)-NEMPA) were synthesized, and purified in our laboratory [19].
Dextran (MW 70,000), sodium borohydride, and sodium metaperiodate
was purchased from Energy Chemical (Shanghai, China). 4-Nitrobenzal-
dehyde was purchased from Shanghai Darui Fine chemical Ltd., China.
2-Cyclohexen-1-one was purchased from Chengdu Best Reagent Ltd.,
China. Chromatographic grade acetonitrile, n-hexane, and isopropanol
were bought from J & K Scientific Ltd., China. Other chemicals used in
this study were all commercially available reagents of analytical grade.
3 4
Magnetic Fe O nanoparticles are simultaneously introduced in the
crosslinking process of rCalB-IBs to further simplify the separation and
the recovery process of rCalB-clIBs and avoid the internal mass-transfer
limitations of the increased size (clumping) of CLEA clusters due to the
separation of CLEAs from the reaction mixture by centrifugation or
filtration [29–31]. Magnetic Fe
properties, such as low toxicity, regular shape, large specific surface
area, low cost, and easy synthesis [32]. Magnetic Fe can hardly
3 4
O nanoparticles have many excellent
3 4
O
produce adverse effects on the enzyme activity, which makes it widely
2.2. Preparation of rCalB-IBs
used in the enzyme immobilization process [33–35]. When rCalB-IBs
with large number of amino groups and Fe
3
O
4
nanoparticles are treated
The single colony containing rCalB-pET21a were inoculated into
with dialdehydes, the insoluble magnetic rCalB-clIBs (rCalB-mclIBs) can
be easily formed, and Fe nanoparticles can be coated together with
rCalB-IBs during the crosslinking process. Furthermore, given the
Luria broth (10 g/L of tryptone, 5 g/L of yeast extract, and 10 g/L of
◦
3
O
4
NaCl) containing ampicillin (50
μg/mL). Cultures were grown at 37 C to
an optical density of 0.6–0.8 at 600 nm. Isopropyl β-D-
2

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
thiogalactopyranoside (IPTG, final concentration of 0.1 mM) was added,
the added amount of rCalB-IBs, and rCalB-clIBs was changed to 12.5 mg/
◦
and the cells were incubated at 16 C for 16 h to induce the expression of
mL.
◦
rCalBs. Then cells were harvested by centrifuging at 6793×g at 4 C for 5
min. The collected cells were re-suspended in sodium phosphate buffer
(
100 mM, pH 9.0), and treated with ultrasonic cell disruptor (JY92-IIN,
2.7. Enantioselective hydrolysis of (R, S)-NEMPA-ME
SCIENTZ, Ningbo, China). rCalB-IBs were separated from the soluble
◦
fraction by centrifugation at 14, 800×g at 4 C for 10 min, and washed
(R, S)-NEMPA-ME (14.0 mmol/L) was added in 1.0 mL of sodium
phosphate buffer (100 mM, pH 7.0) containing 15 U of rCalB-mclIBs,
two times with sodium phosphate buffer (100 mM, pH 9.0), and then
◦
◦
stored at 4 C for further use. The soluble and precipitated rCalBs were
rCalB-clIBs, or rCalB-IBs and the reaction was carried out at 25 C under
analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis
nitrogen protection for 5 h. After completion of the reaction, the enzyme
was removed, and the residue was extracted with ethyl acetate (2.0 mL)
for three times. The organic layer was separated, dried over Na SO , and
(
SDS-PAGE). The gene sequences of rCalBs were analyzed at Comate
Bioscience Co., Ltd. (Jilin, China), and listed in the supplementary ma-
terials. The amounts of 242.7 ± 2.5, 352.3 ± 1.5, and 105.2 ± 2.9 mg
could be obtained for 6His–CalB-IBs, 6His–CalB–5Lys-IBs, and
2
4
concentrated under reduced pressure. The enantioselective hydrolysis
was monitored by high performance liquid chromatography (HPLC,
Agress1100, Elite, Dalian, China) using a Daicel Chiralpak AD-H column
capable of separating the internal standard of benzoic acid. The mobile
phase was a mixture of n-hexane/isopropanol/trifluoroacetic acid (97/
6
His–CalB–10Lys-IBs from 0.4 g cells, respectively.
2
.3. Oxidation of dextran
3
/0.05, v/v/v) at a flow rate of 1.0 mL/min. UV detection at 254 nm was
Dextran (1.7 g) was dissolved in 50 mL of water, and 3.9 g of sodium
employed for quantification. The retention times of benzoic acid, (R)-,
and (S)-NEMPA were 9.1 min, 6.7 min, and 7.3 min, respectively. The
enantiomeric ratio (E value) was calculated from the conversion (c) and
metaperiodate was added. The resulting solution was stirred at room
temperature for 90 min. Subsequently, the solution was dialyzed five
times using a MW cut-off of 12 kDa against 5 L water each time at room
temperature during >2 h and under stirring [42].
the enantiomeric excess (e.e.
E = ln[1ꢀ c(1 + e.e. )]/ln[1ꢀ c(1 ꢀ e.e.
), while c and c are the concentrations of (S)- and (R)-NEMPA,
respectively.
p
) of (S)-NEMPA by using the equation [44]:
p
p
)], where e.e. ꢀ c )/(c
p
= (c
S
R
S
+
c
R
S
R
2
.4. Preparation of rCalB-clIBs
Wet rCalB-IBs (100 mg/mL) were mixed in a sodium phosphate
buffer (100 mM, pH 7.0, 1.0 mL), and oxidized dextran stock solution
2.8. Comparing the stabilities of rCalB-mclIBs, rCalB-clIBs, and rCalB-IBs
(
(
0.4 mmol/L) was subsequently added to a final concentration of 10 %
◦
v/v). The mixture was kept at 10 C for 3 h with shaking. The prepared
The samples of rCalB-mclIBs (25.0 mg/mL), rCalB-clIBs (12.5 mg/
mL), and rCalB-IBs (12.5 mg/mL) were used to compare their thermal,
pH, and storage stabilities, respectively. To compare the thermal sta-
bilities, the enzymes were incubated in Trisꢀ HCl buffer (50 mM, pH 8.0,
◦
rCalB-clIBs were collected by centrifugation at 14,800×g at 4 C for 10
min, and washed two times with sodium phosphate buffer (100 mM, pH
7
.0, 1.0 mL). The resulting rCalB-clIBs were then re-suspended in 4.0 mL
◦
of sodium borohydride solution (1.0 mg/mL) to be blocked. The final
blocking of rCalB-clIBs with sodium borohydride permitted to eliminate
the chemical reactivity of the support avoiding undesired enzyme-
support [43]. This mixture was allowed to be reacted for 45 min at 4
4.0 mL) at different temperatures ranging from 20 to 70 C for 0.5 h. The
pH stabilities were evaluated by measuring their residual enzymatic
activities after incubation in 50 mM of Trisꢀ HCl solution (4.0 mL) at
different pH ranging from 5.0 to 10.0 for 0.5 h. To check the storage
stabilities, the samples were stored in 50 mM of Trisꢀ HCl buffer (pH 8.0,
◦
C. After that, rCalB-clIBs were washed two times with sodium phos-
◦
phate buffer (100 mM, pH 7.0, 1.0 mL). The obtained rCalB-clIBs were
4.0 mL) at 4 C for a certain period of time. The residual enzymatic
◦
◦
stored at 4 C for further use.
activities were measured by hydrolysis of pNPC (0.1 mmol/L) at 40 C.
The maximum specific activity values of rCalB-mclIBs, rCalB-clIBs, and
rCalB-IBs under initial conditions are regarded as 100 %, and the ratio of
the specific activity obtained under other conditions to the maximum
specific activity is referred to as the relative activity.
2
.5. Preparation of rCalB-mclIBs
Samples of wet rCalB-IBs (50 mg/mL) in a sodium phosphate buffer
(
100 mM, pH 7.0, 2.0 mL), were mixed with magnetic Fe
3 4
O nano-
particles (50 mg/mL) under stirring. Oxidized dextran stock solution
(
(
0.4 mmol/L) was subsequently added to a final concentration of 10 %
2.9. Structural characterization
◦
v/v). The mixture was kept at 10 C for 3 h with shaking. The prepared
rCalB-mclIBs were collected by an external magnetic field for 10 min,
FTIR spectroscopy was conducted for rCalB-mclIBs, rCalB-clIBs,
and washed two times with sodium phosphate buffer (100 mM, pH 7.0,
rCalB-IBs, free rCalB, and Fe
3
O
4
nanoparticles on an IFS 66v/S spec-
◦
ꢀ 1
1
.0 mL). The obtained rCalB-mclIBs were stored at 4 C for further use.
trophotometer (Bruker, Germany) with a resolution of 4 cm using the
standard KBr disk method. The second derivative FTIR spectrum anal-
ysis of the rCalB-IBs and free rCalB in the Amide I band was performed
after a 25 point smoothing by the Savitzky-Golay method.
2
.6. Lipase activity determination
The assay was performed by measuring the increase in the absor-
CD spectroscopy detection was acquired from 200 to 300 nm with a
0.125 nm step and 0.2 s acquisiton time on a MOS-450 Circular Di-
chroism spectrometer (Bio-Logic, France). The sample preparation of
free rCalB, rCalB-IBs and rCalB-clIBs were the same procedure as FTIR
detection. Final spectra were recorded as an average of 3 scans and the
background was subtracted.
bance of p-nitrophenol (pNP) at 420 nm produced by the hydrolysis of
pNPC. The reaction began with 25.0 mg/mL of rCalB-mclIBs added in
4
.0 mL of Trisꢀ HCl buffer (50 mM, pH 8.0) containing pNPC (0.1 mmol/
◦
L) at 40 C. The samples were removed at different time intervals, and
the absorbance values of pNP produced from the reactions was moni-
◦
tored at 40 C by Shimadzu-2700 UV–vis spectrophotometer equipped
3 4
The morphology of rCalB-mclIBs, rCalB-clIBs, rCalB-IBs, and Fe O
with a temperature controller. The absorbance values of pNP were
nanoparticles were characterized with a scanning electron microscopy
(JEOL JSM6700 F, Japan) in a field-emission FEI Inspect F operated at 3
kV. Samples were suspended in ethanol for ultrasonic dispersion, dried,
and fixed onto silicon wafers. The samples were immediately sputter-
coated with platinum before observation.
limited to a range from 0.2 to 0.8. One unit of enzyme activity (U) was
defined as the amount of rCalB-mclIBs required to liberate 1 mol of pNP
μ
from the substrate pNPC per minute. The activities of rCalB-IBs, and
rCalB-clIBs were also measured by using the same procedure, and only
3

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
2
.10. Reusability of rCalB-mclIBs, rCalB-clIBs, and rCalB-IBs
3. Results and discussion
The reaction systems consisted of rCalB-mclIBs (19.6 U/mL), rCalB-
3.1. Catalytic performance of rCalB-IBs
clIBs (15.7 U/mL) or rCalB-IBs (17.2 U/mL), and (R, S)-NEMPA-ME
(
14.0 mmol/L) in a sodium phosphate buffer (100 mM, pH 7.0, 1.0 mL).
In our previous study, the recombinant rCalBs with polycationic
amino acids, such as nHis (n = 5–15), or nLys (n = 5–15) at its N and/or
C terminals, were found to increase its soluble expression in E. coli. The
detailed description is shown in our previous paper [11]. The SDS–PAGE
of the broken bacterial liquid confirmed that some constructed rCalBs
could be expressed as soluble proteins. However, some rCalBs were
expressed in the form of inclusion bodies (rCalB-IBs), and the precipi-
tated amount was changed by varying the length of polyamino acids. For
6His–CalB–nLys, the 6His–CalB–10Lys was expressed in a semisoluble
and semi-inclusion form, and others, such as 6His–CalB and
6His–CalB–5Lys, were expressed as an inclusion body (Fig. 1). The
precipitated amounts of rCalB-IBs from different recombinant bacteria
were detected in the same batch, and the same recombinant bacterial
amount (wet weight = 0.4 g) was broken to obtain rCalB-IBs in each
assay. The results in Table 1 indicated that the precipitated amounts of
rCalB-IBs were different from each other and that the precipitated
amounts of 242.7 ± 2.5, 352.3 ± 1.5, and 105.2 ± 2.9 mg could be
obtained for 6His–CalB-IBs, 6His–CalB–5Lys-IBs, and 6His–CalB
–10Lys-IBs, respectively.
During the enzyme activity determination, the 6His–CalB-IBs,
6His–CalB–5Lys-IBs, and 6His–CalB–10Lys-IBs, which were expressed as
inclusion bodies, could also display their catalytic abilities towards the
substrate. The 6His–CalB-IBs, 6His–CalB–5Lys-IBs, and 6His–CalB–
10Lys-IBs had similar catalytic performances and specific activities of
0.16 ± 0.03, 0.17 ± 0.06, and 0.18 ± 0.07 U/mg, respectively (Table 1).
Results suggested that the precipitated rCalB-IBs were different from the
regularly expressed inclusion body protein without activity, which could
aggregate as an active form.
◦
Each cycle lasted for 2 h at 25 C. rCalB-mclIBs were isolated from the
reaction system by an external magnetic field. rCalB-clIBs, and rCalB-IBs
were isolated from the reaction system by centrifugation at 14,800×g at
◦
4
C. After being washed three times with sodium phosphate buffer (100
mM, pH 7.0, 2.0 mL), the isolated enzymes were re-suspended in a fresh
reaction mixture consisted of (R, S)-NEMPA-ME (14.0 mmol/L) and
sodium phosphate buffer (100 mM, pH 7.0, 1.0 mL), and repeatedly used
for the next reaction cycle under the same reaction conditions. The re-
sidual enzyme activities of each cycle were calculated by regarding the
enzyme activity in the first cycle as 100 %.
2
.11. Promiscuous carbon–carbon bond-forming reactions at an
analytical scale
The reaction mixture (0.2 mL) consisted of 4-nitrobenzaldehyde
(
0.007 mmol, 35.0 mmol/L) and 2-cyclohexen-1-one (0.14 mmol, 700
mmol/L) in a sodium phosphate buffer (100 mM, pH 8.0) was ultra-
sonicated for 1 min to form a solution, and then the same enzyme ac-
tivity (10 U) of rCalB-mclIBs, rCalB-clIBs, rCalB-IBs, free rCalB, or 1.0
mg of Novozym 435 were added in the mixture. The resulting mixture
◦
was stirred at 40 C under nitrogen protection. The products were
monitored on silica gel plates (Qingdao Haiyang Chemical Co., Ltd.,
China) using UV light for spot detection. When the substrate aldehyde
was entirely converted, the mixture was extracted with ethyl acetate
(
2.0 mL) for three times. The organic layer was separated, dried over
Na SO , and concentrated under reduced pressure. The yields of the
2
4
MBH and Aldol reactions were analyzed by high-performance liquid
chromatography (HPLC, Agress1100, Elite, Dalian, China) using a
Thermo Scientific C-18 column by external standard method. The mo-
bile phase was a mixture of acetonitrile/water (30/70, v/v). The flow
rate was 1.0 mL/min, and the detection was achieved under UV light at
After rCalB-IBs were found to have catalytic activities, these IBs
should be applied in enzyme-catalyzed reactions to validate their actual
applicability. Here, the enzyme-catalyzed enantioselective hydrolysis of
(R, S)-NEMPA-ME to obtain (S)-NEMPA was selected as the model re-
action. In each assay, the same enzyme activity (15 U) was used to
compare the catalytic performance of rCalB-IBs. As expected, rCalB-IBs
could effectively catalyze the hydrolysis reaction, and all rCalB-IBs
showed strong specificity toward the (S)-NEMPA-ME. The catalytic
conversions of 6His–CalB-IBs, 6His–CalB–5Lys-IBs, and 6His–CalB–
2
54 nm. The retention times of the MBH product 2-(hydroxy(4-nitro-
phenyl)methyl)cyclohex-2-enone and the Aldol product 6-(hydroxy(4-
nitrophenyl)methyl)cyclohex-2-enone were 19.5 min, and 23.5 min,
respectively, which confirmed based on their standards (purity >99.5
%
).
1
0Lys-IBs were 17.9 % ± 1.9 %, 24.1 % ± 1.5 %, and 18.1 % ± 1.1 %,
respectively, with e.e.
values of 87.5 % ± 1.3 %, 86.9 % ± 1.2 %, and
86.4 % ± 0.9 %, respectively, and E values of 18.8 ± 2.4, 18.7 ± 1.5, and
2
.12. Promiscuous carbon–carbon bond-forming reactions at a
p
preparative scale
1
6.6 ± 1.4, respectively (Table 1), which were similar to those of the
The reaction mixture (20.0 mL) consisted of 4-nitrobenzaldehyde
previously reported free rCalBs [11]. These results demonstrated that
rCalB-IBs could maintain their catalytic advantages by fusing nHis/nLys
tails.
(
0.7 mmol, 0.1 g, 35.0 mmol/L) and 2-cyclohexen-1-one (14.0 mmol,
.2 mL, 700 mmol/L) in sodium phosphate buffer (100 mM, pH 8.0, 18.8
1
mL) was ultrasonicated for 1 min to form a solution, and then the rCalB-
mclIBs (1000 U) was added in the mixture. The resulting mixture was
◦
stirred at 40 C under nitrogen protection. After the aldehyde was
entirely converted, the enzymes were removed, and the residue was
extracted with ethyl acetate (10.0 mL) for three times. The organic layer
2 4
was separated, dried over Na SO , and concentrated under reduced
pressure. The MBH and Aldol products were purified through column
chromatography on silica gel (Qingdao Haiyang Chemical Co., Ltd.,
China), and eluted with ethyl acetate: petroleum ether (1:3, v/v),
affording the MBH product (light yellow solid, 28.7 mg) in 17.8 % yield
and Aldol product (light yellow solid, 22.5 mg) in 13.9 % yield. The
structure authenticity of the MBH and Aldol products were confirmed by
HPLC analysis based on their standards (purity >99.5 %).
Fig. 1. SDS-PAGE analysis of recombinant rCalB expressed in E. coli. M: mo-
lecular weight marker; Lane 1–2, 3–4 and 5–6 supernatant and precipitation for
6
His–CalB, 6His–CalB–5Lys, and 6His–CalB–10Lys, respectively.
4

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
Table 1
well modified by the oxidized dextran, and the modified rCalB-clIBs
could retain their excellent catalytic activities (Table 2). The presence of
the nLys tail fused on rCalB-IBs could positively affect the crosslinking of
rCalB-IBs by the oxidized dextran.
a
The precipitated amounts and catalysis parameters of rCalB-IBs .
b
rCalB-IBs
IBs
Specific
Conv.
e.e.
p
E
c
d
d
e
(
mg)
activity (U/
mg)
(%)
(%)
value
The concentration of the oxidized dextran was an important
parameter to determine the particle size of rCalB-clIBs. Higher oxidized
dextran concentrations tended to form enzyme aggregates with
increased particles sizes, thereby affecting mass transfer and generating
diffusion limitations. By contrast, if the oxidized dextran concentration
tested was low, the formed enzyme aggregates were unstable and tended
to dissolve. Also, the low amount of oxidized dextran might not be
conducive to the stability improvement of rCalB-IBs, and the enzyme
molecule might still be too flexible. A large amount of oxidized dextran
could result in the loss of the minimum flexibility needed for the enzyme
activity. Thus, the concentration of the oxidized dextran was optimized
subsequently [46]. Table 2 shows that rCalB-IBs could be successfully
modified by the oxidized dextran and that the activity recovery of
rCalB-clIBs increased with increased concentration of the oxidized
dextran. At the final oxidized dextran concentration of 10 % (v/v) and
crosslinking time of 3 h, a high activity recovery of 91.0 % ± 2.2 % could
be reached for 6His–CalB–5Lys-clIBs, which indicated that almost all
crosslinked rCalB-clIBs remained active and that the effects of diffusion
limitations were not evident.
6
6
6
His–CalB-IBs
242.7 ±
0.16 ± 0.03
0.17 ± 0.06
0.18 ± 0.07
17.9 ±
1.9
87.5 ±
1.3
18.8 ±
2.4
2
.5
His–CalB–5Lys-
IBs
352.3 ±
1.5
24.1 ±
1.5
86.9 ±
1.2
18.7 ±
1.5
His–CalB–10Lys-
IBs
105.2 ±
2.9
18.1 ±
1.1
86.4 ±
0.9
16.6 ±
1.4
a
All measurements were performed three times in parallel. Experiment data:
Mean ± standard deviation (n = 3). The conversions and e.e.
p
were determined
by HPLC analysis.
b
The same recombinant bacteria amount (wet weight, 0.4 g) was broken to
obtain rCalB-IBs.
c
Activity determination: pNPC (0.1 mmol/L); Reaction media: Trisꢀ HCl
◦
buffer (50 mM, pH 8.0, 4.0 mL); Reaction temperature: 40 C.
d
Reaction conditions: (R, S)-NEMPA-ME: 14.0 mmol/L; rCalB-IBs 15 U; Reac-
tion media: sodium phosphate buffer (100 mM, pH 7.0, 1.0 mL); Reaction
◦
temperature: 25 C; Reaction time: 5 h.
e
E value was calculated from the equation: E = ln[1 ꢀ c(1 + e.e.
)].
p
)]/ln[1 ꢀ c (1
ꢀ
e.e.
p
3
.2. Further modification of rCalB-IBs
Following this methodology, crosslinked rCalB-clIBs were treated
with sodium borohydride to reduce the Schiff’s base moieties and check
whether they could form irreversible linkages and eliminate the chem-
ical reactivity of aggregates, thereby avoiding undesired reaction and
being a useful reaction endpoint [47]. Results showed that the activity
recovery of 91.8 % ± 1.2 % for rCalB-clIBs treated with sodium boro-
hydride was similar to that without treatment (91.0 % ± 2.2 %), which
indicated no leakage of enzymatic activity from rCalB-clIBs and feasible
crosslinking by the oxidized dextran.
Considering that the rCalB-IBs could be separated directly from a
fermentation broth and used as an effective solid catalyst for the enan-
tioselective hydrolysis of (R, S)-NEMPA-ME, it was necessary to check
whether it could be as a similarly immobilized enzyme to be recycled
usage. Unfortunately, the operation stabilities and the reusability of
rCalB-IBs were not especially ideal (data not shown). Based on this sit-
uation, the crosslinking strategy was introduced to further modify rCalB-
IBs. Here, the glutaraldehyde and the oxidized dextran were selected as
crosslinkers to modify rCalB-IBs (6His–CalB–5Lys-IBs) and achieve
rCalB-clIBs (6His–CalB–5Lys-clIBs). Results revealed that glutaralde-
hyde, the most commonly used crosslinking agent, could remarkably
decrease the catalytic activities of rCalB-clIBs, which might be because
of the easy entry to the active center of the enzyme and reaction with the
amino acids in the active site, thereby losing enzyme activity [23,45]. By
contrast, the oxidized dextran, which contained a large amount of
aldehyde groups and a high molecular weight, could decrease the
drawbacks of using glutaraldehyde as a crosslinker. rCalB-IBs could be
3 4
Magnetic Fe O nanoparticles were subsequently attempted to be
added in the crosslinking process to simplify the recovery process of
rCalB-clIBs, and the resulting magnetic rCalB-mclIBs could be achieved.
Similarly, the additive amount of Fe
at the final oxidized dextran concentration of 10 % (v/v), and the effects
of Fe nanoparticles on the catalytic activities of rCalB-mclIBs were
3 4
O nanoparticles was first checked
3 4
O
also observed simultaneously. Table 3 shows that the activity recovery
of 6His–CalB–5Lys-mclIBs was slightly improved through the addition of
Fe
without Fe
could be achieved in the presence of 50 mg/mL Fe
His–CalB–5Lys-mclIBs could be easily recovered from the reaction
medium by using a magnet because of the magnetic character of Fe
The amount of Fe nanoparticles (< 50 mg/mL) was not enough to
recover the rCalB-mclIBs by using an external magnetic field. By
contrast, Fe nanoparticles could recover rCalB-mclIBs but could
3
O
4
nanoparticles compared with that of 6His–CalB–5Lys-clIBs
3
O
4
(Table 2). A high activity recovery of 93.4 % ± 2.3 %
3 4
O
nanoparticles, and
6
Table 2
3 4
O .
Effects of the crosslinking parameters on the modification of 6His–CalB–5Lys-
3 4
O
a
IBs .
d
Parameters
Activity recovery (%)
3 4
O
b
Oxidized dextran final concentration (%, v/v)
2
5
1
1
2
82.2 ± 1.5
86.4 ± 2.6
89.7 ± 2.1
84.0 ± 2.7
74.3 ± 2.4
remarkably decrease the activity recovery of rCalB-mclIBs (Table 3). On
the basis of the optimized final oxidized dextran concentration of 10 %
0
5
0
(
3 4
v/v) and Fe O amount of 50 mg/mL, other crosslinking parameters,
including pH of sodium phosphate buffer, temperature, time, stirring
speed, and Triton X-100 surfactant amount, were also optimized.
c
Crosslinking time (h)
0.5
1
79.8 ± 1.7
85.5 ± 1.5
91.0 ± 2.2
88.3 ± 1.9
3 4
The crosslinking of rCalB-IBs and Fe O nanoparticles by using the
oxidized dextran involved the reaction between highly reactive surface
amino groups of poly-Lys residues and aldehyde groups of the oxidized
dextran. Therefore, the charge of the poly-Lys tail fused on rCalB-IBs
might regulate the crosslinking process, and the reactivity of rCalB-IBs
might be different at different pH values of sodium phosphate buffer
3
5
a
All measurements were performed three times in parallel. Experiment data:
Mean ± standard deviation (n = 3).
b
Conditions: 6His–CalB–5Lys-IBs (100 mg/mL); Sodium phosphate buffer
[
48]. For this purpose, the crosslinking time courses at different pH
(
100 mM, pH 7.0, 1.0 mL); Oxidized dextran (0.4 mmol/L, final concentration of
values ranging from 5.0 to 9.0 were considered. As expected, the
crosslinking process of rCalB-IBs at pH 7.0 and 9.0 proceeded rapidly,
and activity recoveries of 97.1 % ± 1.9 % and 97.1 % ± 1.6 %,
respectively, could be obtained for rCalB-mclIBs in 3 h. The relative
enzyme activities of the supernatant were only 3.8 % ± 0.3 % and 4.7 %
± 0.6 % at pH 7.0 and 9.0, respectively, during the 3 h crosslinking
2
–20 %, v/v); Crosslinking time: (2 h).
c
Conditions: 6His–CalB–5Lys-IBs (100 mg/mL); Sodium phosphate buffer
(
100 mM, pH 7.0, 1.0 mL); Oxidized dextran (0.4 mmol/L, final concentration of
1
0 % v/v); Crosslinking time: (0.5–5 h).
d
Activity determination: pNPC (0.1 mmol/L); Reaction media: Trisꢀ HCl
◦
buffer (50 mM, pH 7.0, 4.0 mL); Reaction temperature: 40 C.
5

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
Table 3
the control rCalB-IBs decreased to 7.9 % ± 0.3 % and 8.7 % ± 1.2 % at
pH 7.0 and 9.0, respectively, in 3 h (Fig. 2B and C), which indicated that
rCalB-IBs were relatively stable. All results suggested that the cross-
linking strategy described here was suitable for the preparation of
rCalB-mclIBs.
Effects of the magnetic crosslinking parameters on the modification of
a
6
His–CalB–5Lys-IBs .
f
Parameters
Activity recovery (%)
b
Fe
3
O
4
nanoparticles (mg/mL)
25
91.4 ± 1.9
93.4 ± 2.3
90.3 ± 1.5
88.2 ± 2.2
83.6 ± 2.5
79.9 ± 2.8
76.1 ± 1.9
The crosslinking rate might depend on the temperature for an
enzyme with a typical number of accessible poly-Lys amino groups. The
crosslinking requires long reaction time at low temperatures and short
reaction time at high temperatures. Table 3 shows that the activity re-
covery of rCalB-mclIBs did not change evidently at the temperature
5
7
1
1
1
2
0
5
00
25
50
00
◦
◦
ranging from 4 C to 15 C. Although most of the enzyme aggregate
◦
c
◦
Crosslinking temperature ( C)
4
1
1
2
92.6 ± 2.2
95.4 ± 1.4
90.8 ± 1.8
86.4 ± 2.0
formation was carried out at low temperature (4 C) due to the heat-
◦
0
5
0
labile nature of the enzymes, the crosslinking of rCalB-IBs at 10
C
could achieve an activity recovery of 95.4 % ± 1.4 %. This shift in the
optimum temperature might be due to the covalent bond formation
between rCalB-IBs caused by the oxidized dextran, which might
decrease the conformational flexibility of the enzyme [49]. The forma-
d
Crosslinking time (h)
1
2
3
4
5
90.9 ± 2.3
93.2 ± 1.4
97.1 ± 1.9
94.6 ± 2.1
87.3 ± 2.5
◦
tion of rCalB-mclIBs at 20 C resulted in minimally active rCalB-mclIBs,
which was probably because of the thermal degradation of enzymes at
high temperature.
e
Considering that the crosslinking reaction of the oxidized dextran
with Lys residues progressed with time depending on the accessibility of
the Lys amino groups, the crosslinking time should be optimized. From
the data in Table 3, the rCalB-mclIBs showed an increase in activity from
Stirring speed (rpm)
100
90.9 ± 2.2
101.4 ± 2.3
95.1 ± 2.0
2
2
00
80
a
All measurements were performed three times in parallel. Experiment data:
9
0.9 % ± 2.3%–97.1% ± 1.9 % upon increasing the crosslinking time
Mean ± standard deviation (n = 3).
b
from 1 h to 3 h. The optimum crosslinking time for the preparation of
rCalB-mclIBs was 3 h. The short crosslinking time could lead to poor
activity recovery of enzyme aggregates, whereas the prolonged cross-
linking time might restrict the enzyme flexibility and abolish the enzyme
activity due to intensive crosslinking. The optimum crosslinking time
could involve a compromise between the efficient crosslinking and the
enzyme stability. Also, the stirring speed of 200 rpm could help to
further improve the activity recovery of rCalB-mclIBs to 101.4 % ± 2.3
Conditions: 6His–CalB–5Lys-IBs (50 mg/mL); Oxidized dextran (0.4 mmol/L,
final concentration of 10 % v/v); Sodium phosphate buffer (100 mM, pH 7.0, 2.0
mL); Fe
3
O
4
nanoparticles (25–200 mg/mL); Crosslinking temperature (4 ◦C);
Crosslinking time (3 h).
c
Conditions: 6His–CalB–5Lys-IBs (50 mg/mL); Oxidized dextran (0.4 mmol/L,
final concentration of 10 % v/v); Sodium phosphate buffer (100 mM, pH 7.0, 2.0
mL); Fe
3
O
4
nanoparticles (50 mg/mL); Crosslinking temperature (4–20 ◦C);
Crosslinking time (3 h).
d
Conditions: 6His–CalB–5Lys-IBs (50 mg/mL); Oxidized dextran (0.4 mmol/L,
%
by minimizing the mass transfer problems (Table 3).
rCalB-IBs and Fe nanoparticles were crosslinked in the presence
final concentration of 10 % v/v); Sodium phosphate buffer (100 mM, pH 7.0, 2.0
◦
3 4
O
mL); Fe
3
O
4
nanoparticles (50 mg/mL); Crosslinking temperature (10 C);
of the Triton X-100 surfactant to check if the surfactant could further
improve the activity recovery of rCalB-mclIBs by modifying the enzyme
conformation. The Triton X-100 was selected here in accordance with a
previous report [50,51]. Results (Fig. 3) showed that the Triton X-100
could exhibit some positive effects on the activity of rCalB-mclIBs and
that the activity recovery could be further improved to 113.9 % ± 2.2 %
Crosslinking time (1–5 h).
e
Conditions: 6His–CalB–5Lys-IBs (50 mg/mL); Oxidized dextran (0.4 mmol/L,
final concentration of 10 % v/v); Sodium phosphate buffer (100 mM, pH 7.0, 2.0
◦
mL); Fe
3
O
4
nanoparticles (50 mg/mL); Crosslinking temperature (10 C);
Crosslinking time (3 h); Stirring speed (100–280 rpm).
f
Activity determination: pNPC (0.1 mmol/L); Reaction media: Trisꢀ HCl
◦
buffer (50 mM, pH 8.0, 4.0 mL); Reaction temperature: 40 C.
at low surfactant concentration (0.25 mol/L). rCalB had a small lid,
μ
which did not fully isolate the enzyme from the reaction medium but
could still be affected by the Triton X-100 surfactant. This finding
indicated that the surfactant could help fix its open form to expose the
active site and activated the enzyme [50,52]. Also, the Triton X-100
promoted the dissociation of large aggregates [51,52], thereby
increasing the activity of rCalB-mclIBs.
process (Fig. 2B and C), suggesting that rCalB-IBs were nearly
completely crosslinked by the oxidized dextran at pH 7.0 and 9.0 and
easily formed rCalB-mclIBs aggregates with much space. This finding
prevented the occurrence of diffusional restrictions, resulting in high
activity recovery and stability of rCalB-mclIBs. At low pH value of 5.0,
rCalB-mclIBs showed a remarkable decrease in the enzyme activity of
Lastly, the excellent activity recovery of 113.9 % ± 2.2 % for rCalB-
mclIBs could be achieved at rCalB-IBs (50 mg/mL), oxidized dextran
1
9.8 % ± 1.2 % after 3 h (Fig. 2A), which might be due to the formation
of rCalB-mclIBs with compact and minimal space [38]. The activity of
(
3 4
0.4 mmol/L, final concentration of 10 % [v/v]), Fe O nanoparticles
Fig. 2. The crosslinking time courses of 6His–CalB–5Lys-mclIBs in 100 mM of sodium phosphate buffer at pH 5.0 (A); pH 7.0 (B), and pH 9.0 (C).
6

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
maintained their activities. However, rCalB-IBs and rCalB-clIBs only
retained 49.8 % ± 1.3 % and 60.9 % ± 1.7 %, respectively, of their initial
activities at pH 9.0.
◦
Under the same storage conditions at 4 C, the pNPC hydrolytic ac-
tivities of rCalB-mclIBs decreased at a slower rate than those of rCalB-
clIBs and rCalB-IBs (Fig. 4C). rCalB-mclIBs maintained 86.4 % ± 2.2 % of
its initial activity after 10 days of storage, whereas rCalB-clIBs only
retained 67.2 % ± 1.2 % of its initial activity. rCalB-IBs lost 61.5 % ± 2.1
%
of its initial activity after the same storage period. This result indi-
cated that the stabilities of rCalB-IBs improved considerably after the
magnetic crosslinking modification. rCalB-mclIBs could obtain minimal
possible distortion effects from the buffer solution on its active sites.
These results strongly suggested that the thermal, pH, and storage
stabilities of rCalB-IBs improved after crosslinking with the oxidized
3 4
dextran to obtain rCalB-clIBs. The magnetic Fe O nanoparticles added
Fig. 3. Effects of Triton X-100 on the activities of 6His–CalB–5Lys-mclIBs.
in the crosslinking process could further enhance the stabilities of rCalB-
mclIBs, which might be ascribed to the suitable microenvironment
(
50 mg/mL), sodium phosphate buffer (100 mM, pH 7.0), crosslinking
created by Fe O4 nanoparticles for the enzyme molecules. In addition,
3
◦
temperature (10 C), crosslinking time (3 h), stirring speed (200 rpm),
the steric constrained the aggregates imposed on the enzyme structure,
and Triton X-100 (0.25
μ
mol/L). Its loading capacity could reach 0.98 ±
.01 g wet rCalB-IBs per g wet rCalB-mclIBs. The activity recovery of
13.9 % ± 2.2 % for magnetic rCalB-mclIBs was remarkably higher than
making its denaturation difficult [53–55].
0
1
that of nonmagnetic rCalB-clIBs (91.0 % ± 2.2 %), which indicated that
the magnetic Fe nanoparticles added during the crosslinking process
3.4. Structural characteristics
3 4
O
could be enveloped in the formed aggregates and produce a favorable
The FTIR spectroscopy was first used to compare the structural
characteristics of rCalB-IBs (6His–CalB–5Lys-IBs) with those [11] of free
rCalB (6His–CalB–10Lys). Fig. 5A and B demonstrate that the charac-
teristic lipase peaks, including the amide I and II bands in the
influence on the activities of rCalB-mclIBs.
ꢀ 1
3
.3. Comparison of the stabilities of rCalB-mclIBs, rCalB-clIBs, and
1630–1690 and the 1500–1580 cm
regions, and the secondary
rCalB-IBs
structure of 6His–CalB–5Lys-IBs were similar to those of free
6
His–CalB–10Lys. Thus, rCalB-IBs could still display catalytic abilities
The stabilities of enzyme in enzyme-catalyzed reactions were the key
even in the form of precipitate, which was ascribed to the maintenance
of the actively dynamic structure.
factors to determine if it was valuable for its usage. Thus, the thermal,
pH, and storage stabilities of rCalB-mclIBs (6His–CalB–5Lys-mclIBs)
were examined and compared with those of rCalB-clIBs
Because the position, intensity and width of the characteristic pro-
tein peaks generated by the internal vibration mode of protein can be
detected by FTIR spectroscopy, FTIR spectra is often used to confirm the
secondary structure of protein, and determine the correct combination
of the protein and carrier during enzyme immobilization process
[56–58]. In view of this, the FTIR spectroscopy was used here to detect
(
6His–CalB–5Lys-clIBs) and rCalB-IBs (6His–CalB–5Lys-IBs).
The rCalB-mclIBs, rCalB-clIBs, and rCalB-IBs were incubated at
◦
◦
temperature ranging from 20 C to 70 C for 0.5 h, and their activities
were measured periodically. From the data in Fig. 4A, the catalytic ac-
tivities of all enzymes had different extents of reduction, but the inac-
tivation of rCalB-IBs and rCalB-clIBs was evident compared with that of
rCalB-mclIBs with increased incubation temperature. At low tempera-
the successful modification of rCalB-IBs by using Fe
3 4
O nanoparticles
and oxidized dextran. For comparison, rCalB-clIBs, rCalB-IBs, and Fe
3 4
O
nanoparticles were also analyzed and used as the control. Characteristic
◦
ture (such as 20 C), rCalB-IBs and rCalB-clIBs lost 38.9 % ± 1.4 % and
lipase peaks were observed in the 1630–1690 (v [amide-type I]; 1653.9,
ꢀ 1
3
1.2 % ± 2.0 %, respectively, of their initial activities, whereas rCalB-
1651.9, and 1649.0 cm ) and the 1500–1580 (v [amide-type II];
◦
ꢀ 1
ꢀ 1
mclIBs maintained 100 % of its initial activity. At 70 C, rCalB-mclIBs
still retained a residual activity of 60.8 % ± 2.3 %, but the activities of
rCalB-IBs and rCalB-clIBs declined greatly, and only 27.2 % ± 1.8 % and
1538.1, 1537.1, and 1534.3 cm ) cm
regions for 6His–CalB–
5Lys-IBs, 6His–CalB–5Lys-clIBs, and 6His–CalB–5Lys-mclIBs, respec-
tively (Fig. 6A). In view of the FTIR spectra for 6His–CalB–5Lys-mclIBs,
ꢀ 1
–
–
the band at 1649.0 cm was associated with the C O stretching vi-
3
5.9 % ± 2.2 % of their initial activities, respectively, were observed.
The pH stabilities of rCalB-mclIBs, rCalB-clIBs, and rCalB-IBs were
ꢀ 1
bration in the amide-type I, and that at 1534.3 cm was related to the
H bending and CN stretching vibrations in the amide-type II. The
determined at varying pH (5.0–10.0) for 0.5 h. Results are shown in
Fig. 4B. Notably, rCalB-mclIBs presented a good tolerance at changing
environmental pH. At high pH (9.0–10.0), rCalB-mclIBs almost
N
–
–
amide-type I and the amide-type II peaks could be attributed to the
protein backbone of CalB, and the amide-type I band could be attributed
Fig. 4. The temperature stabilities (A), pH stabilities (B), and storage stabilities (C) of 6His–CalB–5Lys-mclIBs, 6His–CalB–5Lys-clIBs, and 6His–CalB–5Lys-IBs.
7

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
Fig. 5. FTIR spectra of characteristic lipase peaks (A), and secondary structure (B) for 6His–CalB–5Lys-IBs, and free 6His–CalB–10Lys.
Fig. 6. (A): FTIR spectra of 6His–CalB–5Lys-IBs, 6His–CalB–5Lys-clIBs, 6His–CalB–5Lys-mclIBs, and Fe
His–CalB–5Lys-IBs, and 6His–CalB–5Lys-clIBs.
3
O
4
;
(B): CD spectra of free 6His–CalB–10Lys,
6
to the β-sheets and the
conformation of CalB. A similar phenomenon was also reported by other
groups [59,60]. The FTIR image of the control Fe nanoparticles
Fig. 6A) showed the characteristic absorption of 588.3 cm , which was
attributed to the Fe–O stretching vibration [61,62]. The characteristic
peaks of lipase CalB and Fe nanoparticles were observed in the FTIR
spectra of 6His–CalB–5Lys-mclIBs (Fig. 6A), which indicated that rCal-
B-IBs and Fe nanoparticles could be successfully crosslinked together
α
-helices of CalB, representing the native
corresponding rCalB-IBs appeared poorly well ordered, forming sheets
of micrometer-sized flakes (Fig. 7A). After the modification of rCalB-IBs
by the oxidized dextran, the morphology changed to round or barrel-like
3 4
O
ꢀ 1
(
3 4
shapes (Fig. 7B). After further addition of magnetic Fe O nanoparticles
in the oxidized dextran crosslinking process, almost spherical-shaped
3
O
4
and regularly arranged structures were finally obtained in rCalB-
3 4
mclIBs (Fig. 7C) and similar to that of the Fe O (control, Fig. 7D). The
3
O
4
regular spherical-shaped morphology and the excellent activity of rCalB-
mclIBs indicated that the particle size and the morphology of rCalB-
mclIBs had been well controlled by the crosslinking optimization. rCalB-
mclIBs might be classified as type 1 aggregates because CLEAs appeared
to have fine-grained structures with high activities [28]. Consequently,
rCalB-mclIBs with type 1 morphology aggregates permitted excellent
mass transfer and showed excellent activities. Fig. 7E shows that the
prepared rCalB-mclIBs could be easily recovered by the external mag-
netic field. The above results showed that the magnetic complex be-
by the oxidized dextran. The FTIR spectroscopy results combined with
the high activity recovery (113.9 % ± 2.2 %) of rCalB-mclIBs indicated
that the native active conformation (including tertiary and quarternay
structure) of the enzyme on 6His–CalB–5Lys-mclIBs could be completely
reserved to achieve their catalytic functions.
To further check the accuracy of FTIR detection above, CD spec-
troscopy was subsequently selected to observe the structure changes of
rCalBs, rCalB-IBs, and rCalB-clIBs, and compared with each other. It
should be pointed out that rCalB-clIBs was detected here as a sample,
3 4
tween rCalB-IBs and Fe O was successfully formed. Similar results were
because the magnetic Fe
3
O
4
nanoparticles in rCalB-mclIBs could disturb
also reported by other groups [66–68].
the CD detection. According to the CD spectra in Fig. 6B, the negative
peaks at 206.5 and 232 nm for free rCalB (6His-CalB-10Lys), 209 and
3.5. Recovery and reusability of rCalB-mclIBs, rCalB-clIBs, and rCalB-
IBs
2
30 nm for rCalB-IBs (6His-CalB-5Lys-IBs), and 210 and 229 nm for
rCalB-clIBs (6His-CalB-5Lys-clIBs), could be observed, respectively. The
observed negative peaks could be attributed to the -helices of CalB
63–65]. And especially, the CD curve and the negative peak position of
α
To further exhibit the usage superiority of prepared magnetic rCalB-
mclIBs, the reusability of 6His–CalB–5Lys-mclIBs in the enantioselective
hydrolysis of (R, S)-NEMPA-ME was studied. The modified
[
rCalB-clIBs was similar to those of rCalB-IBs and free rCalB, which
indicated that no obvious structure changes occurred when rCalB-IBs
was crosslinked by the oxidized dextran, and the native structure of
rCalB could be retained.
6His–CalB–5Lys-clIBs without Fe
3
O
4
and the 6His–CalB–5Lys-IBs
without any modification were also selected as control samples. Before
studying the reusability of the enzymes, the catalytic performances of
6His–CalB–5Lys-mclIBs, 6His–CalB–5Lys-clIBs, and 6His–CalB–5Lys-IBs
were compared with that of soluble 6His–CalB–10Lys to confirm simi-
larities, and the same enzyme amount of 15 U was selected for all
The SEM revealed that the morphologies of rCalB-IBs varied with the
change in the modification process and could help in understanding the
3 4
interactions of rCalB-IBs, Fe O , and oxidized dextran. The structures of
8

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
Fig. 7. SEM images of 6His–CalB–5Lys-IBs (A), 6His–CalB–5Lys-clIBs (B), 6His–CalB–5Lys-mclIBs (C), and Fe
3
O
4
(D). Recovery of 6His–CalB–5Lys-mclIBs by
external magnetic field (E), and the control sample (F).
enzyme-catalyzed reactions. Results (Table 4) indicated that
comparison between rCalB-mclIBs and Novozym 435 here. Although the
product yield by per mg of CalB for rCalB-mclIBs (0.3 % ± 0.005 %) was
greatly lower than that of the Novozym 435 (592 % ± 37 %) because of
the high specific activity of Novozym 435, the enantioselectivity of
6
His–CalB–5Lys-mclIBs, 6His–CalB–5Lys-clIBs, and 6His–CalB–5Lys-IBs
could display catalytic abilities towards the hydrolysis of (R, S)-NEMPA-
ME similar to that of soluble 6His–CalB–10Lys, and the (S)-NEMPA
could be obtained. The e.e.
p
values of 6His–CalB–5Lys-mclIBs,
rCalB-mclIB-catalyzed hydrolysis (e.e.
16.6 ± 2.8, Table 4) was higher than that of Novozym 435-catalyzed
hydrolysis (e.e.
p
of 86.5 % ± 2.7 % and E value of
6
His–CalB–5Lys-clIBs, and 6His–CalB–5Lys-IBs could be maintained
over 86.5 % ± 2.7 %, 87.0 % ± 2.3 %, and 86.9 % ± 1.2 %, respectively.
The catalytic conversion of rCalB-mclIBs (15.1 % ± 2.4 %) was similar to
that of soluble 6His–CalB–10Lys (17.9 % ± 2.1 %) and slightly lower
than those of rCalB-clIBs (21.3 % ± 1.1 %) and rCalB-IBs (24.1 % ± 1.5
p
of 70.3 % ± 1.1 % and E value of 7.7 ± 0.5, Table 4).
rCalB-mclIBs had the advantages in stereoselectivity and might have a
potential to be used in enzyme-catalyzed enantioselective reactions. The
decreased catalytic activities for rCalB-mclIBs were reasonable because
%
). The E values of 16.6 ± 2.8, 18.5 ± 2.6, 18.7 ± 1.5, and 15.4 ± 2.1
3 4
the addition of Fe O and oxidized dextran in the crosslinking process of
could be achieved for 6His–CalB–5Lys-mclIB-, 6His–CalB–5Lys-clIB-,
His–CalB–5Lys-IB-, and soluble 6His–CalB–10Lys-catalyzed reactions,
rCalB-IBs might produce the mass transfer resistance and reduce the
chance of interactions between enzyme and substrates.
6
respectively, and the trend was similar to those of reaction conversions.
All results indicated that rCalB-mclIBs could be an effective biocatalyst
for use in the enantioselective hydrolysis of (R, S)-NEMPA-ME.
The catalytic abilities of rCalB-mclIBs were subsequently compared
with that of Novozym 435 to evaluate whether rCalB-mclIBs have an
application potential similar to that of Novozym 435. Because Novozym
After ensuring that 6His–CalB–5Lys-mclIBs could also display cata-
lytic activities and enantioselectivity towards the substrate (R, S)-
NEMPA-ME, their reusability was subsequently investigated through the
◦
repeated batch hydrolysis reaction at 25 C and compared with those of
6His–CalB–5Lys-clIBs and 6His–CalB–5Lys-IBs. After 2 h for each run,
the enzyme was reused with fresh substrates. The relative activities of
rCalB-mclIBs, rCalB-clIBs, and rCalB-IBs after each recycle are illustrated
in Fig. 8A. Results showed that rCalB-IBs without modification lost their
4
35 is from commercial sources, and added in milligrams in this reac-
tion, the product yield by per mg of CalB protein is calculated for
Table 4
a
Enantioselective hydrolysis of (R, S)-NEMPA-ME catalyzed by free rCalB, rCalB-IBs, rCalB-clIBs, rCalB-mclIBs, and Novozym 435, respectively .
d
e
Enzymes
Conv. (%)
e.e.
p
(%)
E value
Yield (%)
b
Free 6His-CalB-10Lys
17.9 ± 2.1
24.1 ± 1.5
21.3 ± 1.1
15.1 ± 2.4
29.7 ± 0.7
85.3 ± 1.0
86.9 ± 1.2
87.0 ± 2.3
86.5 ± 2.7
70.3 ± 1.1
15.4 ± 2.1
18.7 ± 1.5
18.5 ± 2.6
16.6 ± 2.8
7.7 ± 0.5
–
–
–
b
6
6
6
His-CalB-5Lys-IBs
His-CalB-5Lys-clIBs
b
b
His-CalB-5Lys-mclIBs
0.3 ± 0.05
c
Novozym 435
592 ± 37
a
All measurements were performed three times in parallel. Experiment data: Mean ± standard deviation (n = 3). The conversions and e.e.
p
were determined by
HPLC analysis.
b
Reaction conditions: (R, S)-NEMPA-ME: 14.0 mmol/L; Enzymes: 15 U; Reaction media: sodium phosphate buffer (100 mM, pH 7.0, 1.0 mL); Reaction temperature:
◦
2
5
c
C; Reaction time: 5 h.
Reaction conditions: (R, S)-NEMPA-ME: 34.0 mmol/L; Novozym 435: 1.0 mg; Reaction media: sodium phosphate buffer (100 mM, pH 7.0, 0.5 mL); Reaction
◦
temperature: 25 C; Reaction time:0.5 h.
d
E value was calculated from the equation: E = ln[1 ꢀ c(1 + e.e.
p
)]/ln[1 ꢀ c (1 ꢀ e.e.
p
)].
e
The product yield by per mg of CalB protein.
9

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
Fig. 8. Reusability of 6His–CalB–5Lys-IBs, 6His–CalB–5Lys-clIBs, 6His–CalB–5Lys-mclIBs. The relative activity (A), and e.e.
p
(B).
full activity only in the first five cycles. The reusability of crosslinked
rCalB-clIBs by the oxidized dextran could be improved to some extent
and could maintain nearly 69.4 % ± 1.9 % of its initial activity after six
times of recycling. The reusability of magnetic rCalB-mclIBs could be
was difficult to achieve when using the Novozym 435 [41]. Results are
shown in Table 5. The MBH reaction yields of the product (2-(hydroxy
(4-nitrophenyl)methyl)cyclohex-2-enone) catalyzed by 6His–CalB–
5Lys- mclIBs, 6His–CalB–5Lys-clIBs, and 6His–CalB–5Lys-IBs were 19.7
% ± 0.4 %, 23.6 % ± 2.0 %, and 23.4 % ± 1.7 %, respectively, which
were similar to that catalyzed by soluble 6His–CalB–10Lys (21.0 % ± 1.9
%). The control soluble 6His–CalB–10Lys could achieve the MBH reac-
tion. This finding indicated that the modified polyamino acid tails might
be the key factor for achieving the MBH reaction because the difference
between 6His–CalB–10Lys and CalB on the Novozym 435 was the
introduction of polyamino acid tails only. Given that rCalB-IBs in a
precipitated form with “native-like structures” could be used as an
efficient catalyst to catalyze the MBH reaction, the modified rCalB-m-
clIBs and rCalB-clIBs could also display catalytic activities towards the
reaction when rCalB-IBs were further subjected to the crosslinking
further enhanced by introducing Fe
activity of rCalB-mclIBs retained 85.4 % ± 2.5 % of its initial activity
after 10 cycles. Moreover, the e.e.
3 4
O nanoparticles, and the catalytic
p
of (S)-NEMPA (84.4 % ± 2.2 %) could
be maintained in all cycles (Fig. 8B). The small decrease in activities
might be partly caused by the leaching of the enzyme into the reaction
medium, which is a common disadvantage of the immobilization tech-
nique [4,69]. A second possibility is due to the long-term mechanical
stirring under continuous reactions, which causes conformational
changes that distort the active site [70]. A high activity retention after
1
0 reusability cycles revealed that the developed magnetic crosslinking
method could serve as an efficient and practical strategy for boosting the
catalytic performance of rCalB-mclIBs. Fe nanoparticles were
3
O
4
3 4
modification with or without Fe O . Although all soluble rCalB, rCal-
embedded into rCalB-clIBs, which might produce a role like the carrier
skeleton and improve the operational stability of rCalB-mclIBs. Also,
B-IBs, rCalB-clIBs, and rCalB-mclIBs could realize the MBH reaction with
low MBH yields and the enantioselectivity of rCalB was not found, re-
sults were still important because the MBH reaction previously consid-
ered impossible by Novozym 435 was easily realized in the simple
reaction system established here. The use of pure sodium phosphate
buffer as a reaction medium and 18 h as reaction time was greatly su-
perior to other catalytic systems with organic or aqueous–organic me-
dium and long reaction time ranging from 30 h to 120 h [71,72]. The
catalytic mechanism which rCalBs can achieve the MBH reaction, and
the optimization of reaction parameters to obtain the high MBH yields
and enantioselectivity, are continuing to be done in our laboratory and
will be reported in due course.
3 4
Fe O nanoparticles may enhance the rigidity of rCalB-mclIBs to a
certain extent, which makes rCalB-mclIBs maintain the catalytic activity
for a long time during the continuous stirring reaction [30].
3
.6. Achievement of the promiscuous carbon–carbon bond-forming
reactions
Besides the enantioselective hydrolysis reaction, rCalB-IBs, rCalB-
clIBs, and rCalB-mclIBs could display catalytic abilities towards car-
bon–carbon bond-forming reactions, such as the MBH reaction between
During the process of studying the above enzyme-catalyzed MBH
4
-nitrobenzaldehyde and 2-cyclohexen-1-one, and this phenomenon
Table 5
a
Promiscuous carbon–carbon bond-forming reactions catalyzed by free rCalB, rCalB-IBs, rCalB-clIBs, rCalB-mclIBs, and Novozym 435, respectively .
Enzymes
MBH yields (%)
Aldol yields (%)
Total yields (%)
b
Free 6His–CalB–10Lys
21.0 ± 1.9
23.4 ± 1.7
23.6 ± 2.0
19.7 ± 0.4
1.0 ± 0.3
17.9 ± 0.6
21.1 ± 0.9
19.0 ± 0.3
15.0 ± 1.1
0.8 ± 0.4
38.9 ± 2.4
45.2 ± 0.8
43.0 ± 1.6
34.7 ± 0.4
1.8 ± 0.1
b
6
6
6
His–CalB–5Lys-IBs
His–CalB–5Lys-clIBs
b
b
His–CalB–5Lys-mclIBs
c
Novozym 435
a
All measurements were performed three times in parallel. Experiment data: Mean ± standard deviation (n = 3). The yields were determined by HPLC analysis.
b
Reaction conditions: 4-nitrobenzaldehyde: 35.0 mmol/L; 2-cyclohexen-1-one: 700 mmol/L; Enzymes: 10 U; Reaction media: sodium phosphate buffer (100 mM, pH
◦
8
.0, 0.2 mL); Reaction temperature: 40 C; Reaction time: 18 h.
c
Reaction conditions: 4-nitrobenzaldehyde: 35.0 mmol/L; 2-cyclohexen-1-one: 700 mmol/L; Novozym 435: 1.0 mg; Reaction media: sodium phosphate buffer (100
◦
mM, pH 8.0, 0.2 mL); Reaction temperature: 40 C; Reaction time: 18 h.
1
0

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
reactions, another carbon–carbon bond-forming reaction could occur
simultaneously between 4-nitrobenzaldehyde and 2-cyclohexen-1-one.
The reaction was confirmed to belong to the Aldol reaction by check-
ing the product 6-(hydroxy(4-nitrophenyl)methyl)cyclohex-2-enone.
The Aldol reaction yields of soluble rCalB-, rCalB-IB-, rCalB-clIB-, and
rCalB-mclIB-catalyzed reactions were similar, and the yields of 17.9 % ±
References
[
1] E.M. Anderson, K.M. Larsson, O. Kirk, One biocatalyst–many applications: the use
of Candida antarctica B-Lipase in organic synthesis, Biocatal. Biotransfor. 16 (1998)
1
81–204, https://doi.org/10.3109/10242429809003198.
[2] V. Gotor-Fernandez, E. Busto, V. Gotor, Candida antarctica lipase B: an ideal
biocatalyst for the preparation of nitrogenated organic compounds, Adv. Synth.
Catal. 348 (2006) 797–812, https://doi.org/10.1002/adsc.200606057.
0
.6 %, 21.1 % ± 0.9 %, 19.0 % ± 0.3 %, and 15.0 % ± 1.1 %, respec-
[
3] B. Wang, C. Zhang, Q. He, H. Qin, G. Liang, W. Liu, Efficient resolution of (R, S)-1-
(1-naphthyl)ethylamne by Candida antarctica lipase B in ionic liquids, Mol. Catal.
tively, could be reached (Table 5). The Aldol side reactions that simul-
taneously occurred should be a direct factor leading to the low MBH
conversion because of the substrates shared by both reactions. However,
from another point, results were also interesting. The MBH and the Aldol
products could be simultaneously obtained in a one-pot reaction, and
total yields of 38.9 % ± 2.4 %, 45.2 % ± 0.8 %, 43.0 % ± 1.6 %, and 34.7
4
48 (2018) 116–121, https://doi.org/10.1016/j.mcat.2018.01.026.
[
4] M.P. Pinheiro, N.S. Rios, T. de S. Fonseca, F. de A. Bezerra, E. Rodríguez-Castell o´ n,
R. Fernandez-Lafuente, M. Carlos de Mattos, J.C.S. dos Santos, L.R.B. Gonçalves,
Kinetic resolution of drug intermediates catalyzed by lipase B from Candida
antarctica immobilized on immobead-350, Biotechnol. Prog. 34 (2018) 878–889,
https://doi.org/10.1002/btpr.2630.
[
[
5] E. Busto, V. Gotor-Fern a´ ndez, V. Gotor, Hydrolases in the stereoselective synthesis
of N-heterocyclic amines and amino acid derivatives, Chem. Rev. 111 (2011)
3998–4035, https://doi.org/10.1021/cr100287w.
%
± 0.4 % could be reached for soluble rCalB-, rCalB-IB-, rCalB-clIB-,
and rCalB-mclIB-catalyzed reactions, respectively.
6] S.Y. Hong, E.J. Jung, J.C. Joo, Y.J. Yoo, Soluble expression of Candida antarctica
lipase B in Escherichia coli by fusion with Skp chaperone, Biotechnol. Bioprocess
Eng. 17 (2012) 687–692, https://doi.org/10.1007/s12257-012-0093-3.
As a further extension of this synthetic approach, rCalB-mclIB-cata-
lyzed MBH and Aldol reactions at a preparative scale were established.
Similar to that at the analytical scale, at the preparative scale, the iso-
lated yields of 17.8 % and 13.9 % could be obtained for the MBH and the
Aldol reactions, respectively. The corresponding MBH product of 2-
[7] M.W. Larsen, U.T. Bornscheuer, K. Hult, Expression of Candida antarctica lipase B in
Pichia pastoris and various Escherichia coli systems, Protein Expr. Purif. 62 (2008)
9
0–97, https://doi.org/10.1016/j.pep.2008.07.012.
[
8] H.S. Seo, S.E. Kim, K.Y. Han, J.S. Park, Y.H. Kim, S.J. Sim, J. Lee, Functional fusion
mutant of Candida antarctica lipase B (CalB) expressed in Escherichia coli, Biochim.
Biophys. Acta 1794 (2009) 519–525, https://doi.org/10.1016/j.
(
hydroxy(4-nitrophenyl)methyl)cyclohex-2-enone (28.7 mg) and Aldol
product of 6-(hydroxy(4-nitrophenyl)methyl)cyclohex-2-enone (22.5
mg) could be finally reached. The rCalB-mclIBs prepared in this study
could be used as an effective biocatalyst for different enzyme-catalyzed
reactions.
bbapap.2008.12.007.
[
9] Y. Asano, M. Dadashipour, M. Yamazaki, N. Doi, H. Komeda, Functional expression
of a plant hydroxynitrile lyase in Escherichia coli by directed evolution: creation and
characterization of highly in vivo soluble mutants, Protein Eng. Des. Sel. 24 (2011)
6
07–616, https://doi.org/10.1093/protein/gzr030.
[
[
10] L. Bivona, Z. Zou, N. Stutzman, P.D. Sun, Influence of the second amino acid on
recombinant protein expression, Protein Expr. Purif. 74 (2010) 248–256, https://
doi.org/10.1016/j.pep.2010.06.005.
4
. Conclusion
11] X. Zhou, Y. Han, Z. Lv, X. Tian, H. Li, P. Xie, L. Zheng, Simultaneously achieve
soluble expression and biomimetic immobilization of Candida antarctica lipase B by
introducing polyamine tags, J. Biotechnol. 249 (2017) 1–9, https://doi.org/
In conclusion, rCalB-IBs were determined to display catalytic activ-
ities similar to that of soluble rCalBs and could be successfully used in
the enzyme-catalyzed enantioselective hydrolysis reaction to prepare
the (S)-NEMPA. After rCalB-IBs were successfully modified by the
1
0.1016/j.jbiotec.2017.03.015.
[
[
12] S.K. Kim, Y.C. Park, H.H. Lee, S.T. Jeon, W.K. Min, J.H. Seo, Simple amino acid tags
improve both expression and secretion of Candida antarctica lipase B in
recombinant Escherichia coli, Biotechnol. Bioeng. 112 (2015) 346–355, https://doi.
org/10.1002/bit.25361.
13] V.D. J a¨ ger, R. Lamm, K. Küsters, G. Olçücü, M. Oldiges, K.E. Jaeger, J. Büchs,
U. Krauss, Catalytically-active inclusion bodies for biotechnology—general
concepts, optimization, and application, Appl. Microbiol. Biotechnol. 104 (2020)
3 4
oxidized dextran and magnetic Fe O nanoparticles, the obtained rCalB-
mclIBs could further exhibit excellent thermal, pH, and storage stabil-
ities and reusability. The modification strategy adopted in this study had
evident advantages, such as mildness, low cost, and simple preparation.
Also, the prepared rCalB-mclIBs could be used as an effective biocatalyst
for the enantioselective hydrolysis reaction to prepare (S)-NEMPA and
promiscuous carbon–carbon bond-forming reactions to obtain MBH and
Aldol products simultaneously.
¨
7
313–7329, https://doi.org/10.1007/s00253-020-10760-3.
[
[
14] J. Mukherjee, M.N. Gupta, Paradigm shifts in our views on inclusion bodies, Curr.
Biochem. Eng. 3 (2016) 47–55, https://doi.org/10.2174/
2212711902666150302212051.
15] E. García-fruit o´ s, N. Gonz a´ lez-montalb ´a n, M. Morell, A. Vera, R.M. Ferraz, A. Arís,
S. Ventura, A. Villaverde, Aggregation as bacterial inclusion bodies does not imply
inactivation of enzymes and fluorescent proteins, Microb. Cell Fact. 4 (2005) 27,
https://doi.org/10.1186/1475-2859-4-27.
CRediT authorship contribution statement
[
16] M. Diener, B. Kopka, M. Pohl, K.E. Jaeger, U. Krauss, Fusion of a coiled-coil domain
facilitates the high-level production of catalytically active enzyme inclusion bodies,
Chem. Cat. Chem. 8 (2016) 142–152, https://doi.org/10.1002/cctc.201501001.
Yu Han: Investigation, Data curation, Methodology, Writing - orig-
inal draft. Xinying Zhang: Investigation, Writing - original draft.
Liangyu Zheng: Supervision, Writing - review & editing, Funding
acquisition.
[17] R. Kloss, T. Karmainski, V.D. J ¨a ger, D. Hahn, A. Grünberger, M. Baumgart,
U. Krauss, Karl-Erich Jaeger, W. Wiechert, M. Pohl, Tailor-made catalytically active
inclusion bodies for different applications in biocatalysis, Catal. Sci. Technol. 8
(
2018) 5816–5826, https://doi.org/10.1039/C8CY01891J.
[
[
[
18] D. Humer, O. Spadiut, Wanted: more monitoring and control during inclusion body
processing, World J. Microbiol. Biotechnol. 34 (2018) 158, https://doi.org/
1
0.1007/s11274-018-2541-5.
Declaration of Competing Interest
19] L. Zheng, S. Zhang, Y. Feng, S. Cao, J. Ma, L. Zhao, G. Gao, Enantioselective lipase-
catalyzed kinetic resolution of N-(2-ethyl-6-methylphenyl)alanine, J. Mol. Catal., B
Enzym. 31 (2004) 117–122, https://doi.org/10.1016/j.molcatb.2004.06.005.
20] L. Zheng, S. Zhang, X. Yu, L. Zhao, G. Gao, X. Yang, H. Duan, S. Cao, Enhancement
of enantioselectivity in lipase-catalyzed resolution of N-(2-ethyl-6-methylphenyl)
alanine by additives, J. Mol. Catal., B Enzym. 38 (2006) 17–23, https://doi.org/
The authors declare no conflict of interest.
Acknowledgements
1
0.1016/j.molcatb.2005.10.003.
[
21] Y. Han, X. Zhou, L. Zheng, Rational enhancement of enzyme-catalyzed
enantioselective reaction by construction of recombinant enzymes based on
additive strategy, Bioprocess Biosyst. Eng. 42 (2019) 1739–1746, https://doi.org/
The authors are grateful for the financial support provided by the
National Natural Science Foundation of China (Nos. 21672081 and
1
0.1007/s00449-019-02170-1.
2
1372098), Jilin Provincial Science and Technology Sustentation Pro-
[
[
22] A. Basso, S. Serban, Industrial applications of immobilized enzymes—a review,
gram (Nos. 20200404104YY and 20190201165JC), Jilin Province
Development and Reform Commission (Nos. 2016C047-1).
Mol. Catal. 479 (2019), 110607, https://doi.org/10.3390/catal8050174.
@
23] R.A. Sheldon, Cross-linked enzyme aggregates (CLEA s): stable and recyclable
biocatalysts, Biochem. Soc. Trans. 35 (2007) 1583–1587, https://doi.org/10.1042/
bst0351583.
Appendix A. Supplementary data
[24] H. Yamaguchi, Y. Kiyota, M. Miyazaki, Techniques for preparation of cross-linked
enzyme aggregates and their applications in bioconversions, Catalysts 8 (2018)
1
74, https://doi.org/10.3390/catal8050174.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2021.111467.
1
1

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
[
[
25] D.G. Filho, A.G. Silva, C.Z. Guidini, Lipases: sources, immobilization methods, and
industrial applications, Appl. Microbiol. Biotechnol. 103 (2019) 7399–7423,
https://doi.org/10.1007/s00253-019-10027-6.
26] R.A. Sheldon, Characteristic features and biotechnological applications of cross-
linked enzyme aggregates (CLEAs), Appl. Microbiol. Biotechnol. 92 (2011)
enzymes via multipoint covalent attachment. Application to chymotrypsin, RSC
Adv. 5 (2015) 20639–20649, https://doi.org/10.1039/c4ra16926c.
[49] A. Sanchez, J. Cruz, N. Rueda, J.C.S. dos Santos, R. Torres, C. Ortiz, R. Villalonga,
R. Fernandez-Lafuente, Inactivation of immobilized trypsin under dissimilar
conditions produces trypsin molecules with different structures, RSC Adv. 6 (2016)
27329–27334, https://doi.org/10.1039/C6RA03627A.
4
67–477, https://doi.org/10.1007/s00253-011-3554-2.
[
[
27] R.A. Sheldon, Cross-linked enzyme aggregates as industrial biocatalysts, Org.
Process Res. Dev. 15 (2011) 213–223, https://doi.org/10.1021/op100289f.
28] S. Talekar, A. Joshi, G. Joshi, P. Kamat, R. Haripurkar, S. Kambale, Parameters in
preparation and characterization of crosslinked enzyme aggregates (CLEAs), RSC
Adv. 3 (2013) 12485–12511, https://doi.org/10.1039/C3RA40818C.
[50] H. Yang, W. Zhang, Surfactant imprinting hyperactivated immobilized lipase as
efficient biocatalyst for biodiesel production from waste cooking oil, Catalysts 9
(2019) 914, https://doi.org/10.3390/catal9110914.
[51] U.M.F. de Oliveira, L.J.B. Lima de Matos, M.C.M. de Souza, B.B. Pinheiro, J.C.
S. dos Santos, L.R.B. Gonçalves, Effect of the presence of surfactants and
immobilization conditions on catalysts’ properties of rhizomucor miehei lipase onto
chitosan, Appl. Biochem. Biotechnol. 184 (2018) 1263–1285, https://doi.org/
10.1007/s12010-017-2622-1.
[52] M. Galvisa, O. Barbosaa, M. Ruiza, J. Cruza, C. Ortizb, R. Torresa, R. Fernandez-
Lafuente, Chemical amination of lipase B from Candida antarctica is an efficient
solution for the preparation of crosslinked enzyme aggregates, Process Biochem. 47
(2012) 2373–2378, https://doi.org/10.1016/j.procbio.2012.09.018.
[53] J.M. Palomo, G. Mu n˜ oz, G. Fern ´a ndez-Lorente, C. Mateo, R. Fern a´ ndez-Lafuente, J.
M. Guis ´a n, Interfacial adsorption of lipases on very hydrophobic support
(octadecyl–sepabeads): immobilization, hyperactivation and stabilization of the
open form of lipases, J. Mol. Catal., B Enzym. 19–20 (2002) 279–286, https://doi.
org/10.1016/S1381-1177(02)00178-9.
[
[
29] J.C.S. dos Santos, O. Barbosa, C. Ortiz, A. Berenguer-Murcia, R.C. Rodrigues,
R. Fernandez-Lafuente, Importance of the support properties for immobilization or
purification of enzymes, ChemCatChem 7 (2015) 2413–2432, https://doi.org/
1
0.1002/cctc.201500310.
30] W.S. Galv a˜ o, B.B. Pinheiro, L.R.B. Golçalves, M.Cde Mattos, T.S. Fonseca, T. Regis,
D. Zampieri, J.C.S. dos Santos, L.S. Costa, M.A. Correa, F. Bohn, P.B.A. Fechine,
Novel nanohybrid biocatalyst: application in the kinetic resolution of secondary
alcohols, J. Mater. Sci. 53 (2018) 14121–14137, https://doi.org/10.1007/s10853-
0
18-2641-5.
[
31] S. Talekar, V. Ghodake, T. Ghotage, P. Rathod, P. Deshmukh, S. Nadar, M. Mulla,
M. Ladole, Novel magnetic cross-linked enzyme aggregates (magnetic CLEAs) of
alpha amylase, Bioresour. Technol. 123 (2012) 542–547, https://doi.org/10.1016/
j.biortech.2012.07.044.
32] M. Bilal, Y. Zhao, T. Rasheed, H.M.N. Iqbal, Magnetic nanoparticles as versatile
carriers for enzymes immobilization: a review, Int. J. Biol. Macromol. 120 (2018)
[54] G.H. Peters, O.H. Olsen, A. Svendsen, R.C. Wade, Theoretical investigation of the
dynamics of the active site lid in Rhizomucor miehei lipase, Biophys. J. 71 (1996)
119–129, https://doi.org/10.1016/S0006-3495(96)79207-X.
[
[
2
530–2544, https://doi.org/10.1016/j.ijbiomac.2018.09.025.
[55] J. Cui, L. Cui, S. Jia, Z. Su, S. Zhang, Hybrid cross-linked lipase aggregates with
magnetic nanoparticles: a robust and recyclable biocatalysis for the epoxidation of
oleic acid, J. Agric. Food Chem. 64 (2016) 7179–7187, https://doi.org/10.1021/
acs.jafc.6b01939.
33] E. Su, M. Yang, C. Ning, X. Ma, S. Deng, Magnetic combined cross-linked enzyme
aggregates (Combi-CLEAs) for cofactor regeneration in the synthesis of chiral
alcohol, J. Biotechnol. 271 (2018) 1–7, https://doi.org/10.1016/j.
jbiotec.2018.02.007.
[56] A. Barth, Infrared spectroscopy of proteins, Biochim. Biophys. Acta Bioenerg. 1767
(2007) 1073–1101, https://doi.org/10.1016/j.bbabio.2007.06.004.
[57] P.I. Haris, Probing protein–protein interaction in biomembranes using Fourier
transform infrared spectroscopy, Biochim. Biophys. Acta 1828 (2013) 2265–2271,
https://doi.org/10.1016/j.bbamem.2013.04.008.
[58] K.S. Moreira, A.L.B. de Oliveira, L.S.M. Júnior, R.R.C. Monteiro, T.N. da Rocha, F.
L. Menezes, L.M.U.D. Fechine, J.C. Denardin, S. Michea, R.M. Freire, P.B.
A. Fechine, M.C.M. Souza, J.C.S. dos Santos, Lipase from Rhizomucor miehei
immobilized on magnetic nanoparticles: performance in fatty acid ethyl ester
(FAEE) optimized production by the Taguchi method, Front. Bioeng. Biotechnol. 8
(2020) 693, https://doi.org/10.3389/fbioe.2020.00693.
[59] J.C. Cruza, P.H. Pfromma, J.M. Tomichb, M.E. Rezac, Conformational changes and
catalytic competency of hydrolases adsorbing on fumed silica nanoparticles: II.
Secondary structure, Colloids Surf. B 81 (2010) 1–10, https://doi.org/10.1016/j.
colsurfb.2010.06.005.
[60] T. Nematian, A. Shakeri, Z. Salehi, A.A. Saboury, Lipase immobilized on
functionalized superparamagnetic few-layer graphene oxide as an efficient
nanobiocatalyst for biodiesel production from Chlorella vulgaris bio-oil, Biotechnol.
Biofuels 13 (2020) 57, https://doi.org/10.1186/s13068-020-01688-x.
[61] J. Cui, B. Sun, T. Lin, Y. Feng, S. Jia, Enzyme shielding by mesoporous organosilica
[
[
34] A. Badoei-dalfard, Z. Karami, S. Malekabadi, Construction of CLEAs-lipase on
magnetic graphene oxide nanocomposite: an efficient nanobiocatalyst for biodiesel
production, Bioresour. Technol. 278 (2019) 473–476, https://doi.org/10.1016/j.
biortech.2019.01.050.
35] J. Zhou, R. Zhang, T. Yang, Q. Liu, J. Zheng, F. Wang, F. Liu, M. Xu, X. Zhang,
Z. Rao, Relieving allosteric inhibition by designing active inclusion bodies and
coating of the inclusion bodies with Fe
butyric acid production, ACS Catal. 8 (2018) 8889–8901, https://doi.org/
0.1021/acscatal.8b03181.
3 4
O nanomaterials for sustainable 2-Oxo-
1
[
[
[
[
36] D. Basavaiah, G. Veeraraghavaiah, The Baylis–hillman reaction: a novel concept
for creativity in chemistry, Chem. Soc. Rev. 41 (2012) 68–78, https://doi.org/
1
0.1039/c1cs15174f.
37] Y. Wei, M. Shi, Recent advances in organocatalytic asymmetric morita-baylis-
hillman/aza- morita-baylis-hillman reactions, Chem. Rev. 113 (2013) 6659–6690,
https://doi.org/10.1021/cr300192h.
38] D. Basavaiah, A.J. Rao, T. Satyanarayana, Recent advances in the Baylis-Hillman
reaction and applications, Chem. Rev. 103 (2003) 811–891, https://doi.org/
1
0.1021/cr010043d.
39] F. Hasan, A.A. Shah, A. Hameed, Industrial applications of microbial lipases,
Enzyme Microb. Technol. 39 (2006) 235–251, https://doi.org/10.1016/j.
enzmictec.2005.10.016.
3 4
shell on Fe O @silica yolk-shell nanospheres, Int. J. Biol. Macromol. 117 (2018)
673–682, https://doi.org/10.1016/j.ijbiomac.2018.05.227.
[
[
40] M.T. Reetz, Lipases as practical biocatalysts, Biocatal. Biotransfor. 6 (2002)
3 4
[62] Y. Shao, T. Jing, J. Tian, Y. Zheng, Graphene oxide-based Fe O nanoparticles as a
1
45–150, https://doi.org/10.1016/S1367-5931(02)00297-1.
novel scaffold for the immobilization of porcine pancreatic lipase, RSC Adv. 5
(2015) 103943–103955, https://doi.org/10.1039/C5RA19276E.
[63] N.J. Greenfield, Using circular dichroism spectra to estimate protein secondary
structure, Nat. Protoc. 1 (2006) 2876–2890, https://doi.org/10.1038/
nprot.2006.202.
[64] B. Nian, C. Cao, Y. Liu, How Candida antarctica lipase B can be activated in natural
deep eutectic solvents: experimental and molecular dynamics studies, J. Chem.
Technol. Biotechnol. 95 (2020) 86–93, https://doi.org/10.1002/jctb.6209.MM.
[65] Q. Jia, R. Wen, Q. Meng, Novel highly soluble chimeric recombinant spidroins with
high yield, Int. J. Mol. Sci. 21 (2020) 6905, https://doi.org/10.3390/
ijms21186905.
41] M.T. Reetz, R. Mondi `e re, J.D. Carballeira, Enzyme promiscuity: first protein-
catalyzed Morita-Baylis-Hillman reaction, Tetrahedron Lett. 48 (2007) 1679–1681,
https://doi.org/10.1016/j.tetlet.2007.01.063.
[
[
[
[
[
42] H. Li, W. Xiao, P. Xie, L. Zheng, Co-immobilization of enoatereductase with a
cofactor-recycling partner enzyme, Enzyme Microb. Technol. 109 (2018) 66–73,
https://doi.org/10.1016/j.enzmictec.2017.09.013.
43] C. Mateo, J.M. Palomo, L.M. van Langen, F. van Rantwijk, R.A. Sheldon, A New,
mild cross-linking methodology to prepare cross-linked enzyme aggregates,
Biotechnol. Bioeng. 86 (2004) 273–276, https://doi.org/10.1002/bit.20033.
44] A.J.J. Straathof, J.A. Jongejan, The enantiomeric ratio: origin, determination and
prediction, Enzyme Microb. Technol. 21 (1997) 559–571, https://doi.org/
[66] J.H. Liou, Z.H. Wang, I.H. Chen, S.S.S. Wang, S.C. How, J.S. Jan, Catalase
immobilized in polypeptide/silica nanocomposites via emulsion and
biomineralization with improved activities, Int. J. Biol. Macromol. 159 (2020)
931–940, https://doi.org/10.1016/j.ijbiomac.2020.05.138.
1
0.1016/S0141-0229(97)00066-5.
45] L. Tavernini, C. Ottone, A. Illanes, L. Wilson, Entrapment of enzyme aggregates in
chitosan beads for aroma release in white wines, Int. J. Biol. Macromol. 154 (2020)
1
082–1090, https://doi.org/10.1016/j.ijbiomac.2020.03.031.
[67] R.M. Bezerra, R.R.C. Monteiro, D.M.A. Neto, F.F.M. da Silva, R.C.M. de Paula, T.L.
G. de Lemos, P.B.A. Fechine, M.A. Correa, F. Bohn, L.R.B. Gonçalves, J.C.S. dos
Santos, A new heterofunctional support for enzyme immobilization: PEI
46] M. Amaral-Fonseca, W. Kopp, R.L.C. Giordano, R. Fern a´ ndez-Lafuente, P.
W. Tardioli, Preparation of magnetic cross-linked amyloglucosidase aggregates:
solving some activity problems, Catalysts 8 (2018) 496, https://doi.org/10.3390/
catal8110496.
3 4
functionalized Fe O MNPs activated with divinylsulfone. Application in the
immobilization of lipase from Thermomyces lanuginosus, Enzyme Microb.Technol.
138 (2020), 109560, https://doi.org/10.1016/j.enzmictec.2020.109560.
[68] A. Arca-Ramos, V.V. Kumar, G. Eibes, M.T. Moreira, H. Cabana, Recyclable cross-
linked laccase aggregates coupled to magnetic silica microbeads for elimination of
pharmaceuticals from municipal wastewater, Environ. Sci. Pollut. Res. 23 (2016)
8929–8939, https://doi.org/10.1007/s11356-016-6139-x.
[
[
47] J.C.S. dos Santos, N. Rueda, O. Barbosa, Mdel C. Mill ´a n-Linares, J. Pedroche, M. del
Mar Yuste, L.R.B. Gonçalves, R. Fernandez-Lafuente, Bovine trypsin
immobilization on agarose activated with divinylsulfone: improved activity and
stability via multipoint covalent attachment, J. Mol. Catal., B Enzym. 117 (2015)
3
8–44, https://doi.org/10.1016/j.molcatb.2015.04.008.
48] J.C.S. dos Santos, N. Rueda, O. Barbosa, J.F. Fern ´a ndez-S ´a nchez, A.L. Medina-
Castillo, T. Ram o´ n-M ´a rquez, M.C. Arias-Martos, M.C. Mill a´ n-Linares, J. Pedroche,
M. del Mar Yust, L.R.B. Gonçalves, R. Fernandez-Lafuente, Characterization of
supports activated with divinyl sulfone as a tool to immobilize and stabilize
[69] G.V. Lima, M.R. da Silva, T. de Sousa Fonseca, L.B. de Lima, M. da C. F. de Oliveira,
T.L.G. de Lemos, D. Zampieri, J.C.S. dos Santos, N.S. Rios, L.R.B. Gonçalves,
F. Molinari, M.C. de Mattos, Chemoenzymatic synthesis of (S)-Pindolol using
1
2

Y. Han et al.
Molecular Catalysis 504 (2021) 111467
lipases, Appl. Catal. A Gen. 546 (2017) 7–14, https://doi.org/10.1016/j.
[71] G.N. Ma, J.J. Jiang, M. Shi, Y. Wei, Recent extensions of the Morita-Baylis-Hillman
reaction, Chem. Commun. 37 (2009) 5496–5514, https://doi.org/10.1039/
b909405a.
[72] Y. Nakayama, T. Gotanda, K. Ito, Asymmetric Morita-Baylis-Hillman reactions of 2-
cyclohexen-1-one catalyzed by chiral biaryl-based bis(thiourea) organocatalysts,
Tetrahedron Lett. 52 (2011) 6234–6237, https://doi.org/10.1016/j.
tetlet.2011.09.064.
apcata.2017.08.003.
[
70] P. Paitaid, A. H-Kittikun, Magnetic cross-linked enzyme aggregates of aspergillus
oryzae ST11 lipase using polyacrylonitrile coated magnetic nanoparticles for
biodiesel production, Appl. Biochem. Biotechnol. 190 (2020) 1319–1332, https://
doi.org/10.1007/s12010-019-03196-7.
1
3