Modeling and optimization of lipase-catalyzed hydrolysis for production of (S)-2-phenylbutyric acid enhanced by hydroxyethyl-β-cyclodextrin

Article abstract of DOI:10.1016/j.procbio.2019.05.011

An efficient reactive system was established to produce (S)-2-phenylbutyric acid (2-PBA) through the enzymatic enantioselective hydrolysis of 2-phenylbutyrate ester (2-PBAE) in aqueous medium. Lipase CALA from Canadian antarctica and hexyl 2-phenylbutyrate (2-PBAHE) were identified upon screening as the best enzyme and substrate, respectively. Adding hydroxyethyl-β-cyclodextrin (HE-β-CD) to improve the solubility of the substrate resulted in a 1.5 times increase in substrate conversion while retaining a high enantioselectivity compared with that when HE-β-CD was not added. The effects of lipase concentration, substrate concentration and HE-β-CD concentration, temperature, pH, and reaction time on enantiomeric excess and conversion rate were investigated, and the optimal conditions were identified using response surface methodology (RSM). Under the optimal conditions, namely 50 mg/mL lipase CALA, 30 mmol/L substrate, 60 mmol/L HE-β-CD, pH of 6.5, temperature of 83 °C and reaction time of 18 h, the enantiomeric excess and overall conversion rate were 96.05% and 27.28%, respectively. This work provides an efficient alternative method for improving the conversion of aromatic ester substrates by including β-cyclodextrin in an aqueous hydrolysis reaction system.

Full text of DOI:10.1016/j.procbio.2019.05.011

Accepted Manuscript
Title: Modeling and optimization of lipase-catalyzed
hydrolysis for production of (S)-2-phenylbutyric acid
enhanced by hydroxyethyl-␤-cyclodextrin
Authors: Panliang Zhang, Qing Cheng, Weifeng Xu, Kewen
Tang
PII:
S1359-5113(19)30280-6
DOI:
Reference:
https://doi.org/10.1016/j.procbio.2019.05.011
PRBI 11653
To appear in:
Process Biochemistry
Received date:
Revised date:
Accepted date:
21 February 2019
28 March 2019
9 May 2019
Please cite this article as: Zhang P, Cheng Q, Xu W, Tang K, Modeling and
optimization of lipase-catalyzed hydrolysis for production of (S)-2-phenylbutyric
acid enhanced by hydroxyethyl-␤-cyclodextrin, Process Biochemistry (2019),
https://doi.org/10.1016/j.procbio.2019.05.011
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Modeling and optimization of lipase-catalyzed hydrolysis for
production of (S)-2-phenylbutyric acid enhanced by hydroxyethyl-
β-cyclodextrin
Panliang Zhang1, Qing Cheng1, Weifeng Xu, Kewen Tang*
Department of Chemistry and Chemical Engineering, Hunan Institute of Science and
Technology, Yueyang 414000, Hunan, China
*Corresponding author.
E-mail: tangkewen@sina.com (K.W. Tang)
1 These authors equally contributed to this work.
Graphical abstract
Research highlights
 Chiral resolution of 2-phenylbutyric acid by lipase-catalyzed with eep of 96.05%.
 Enhancement of substrate conversion by β-cyclodextrin derivatives is up to 1.5 times.
 The reaction system may be extended to a large number of aromatic acid enantiomers.

ABSTRACT
An efficient reactive system was established to produce (S)-2-phenylbutyric acid (2-PBA)
through the enzymatic enantioselective hydrolysis of 2-phenylbutyrate ester (2-PBAE) in
aqueous medium. Lipase CALA from Canadian antarctica and hexyl 2-phenylbutyrate (2-
PBAHE) were identified upon screening as the best enzyme and substrate, respectively. Adding
hydroxyethyl-β-cyclodextrin (HE-β-CD) to improve the solubility of the substrate resulted in
a 1.5 times increase in substrate conversion while retaining a high enantioselectivity compared
with that when HE-β-CD was not added. The effects of lipase concentration, substrate
concentration and HE-β-CD concentration, temperature, pH, and reaction time on enantiomeric
excess and conversion rate were investigated, and the optimal conditions were identified using
response surface methodology (RSM). Under the optimal conditions, namely 50 mg/mL lipase
CALA, 30 mmol/L substrate, 60 mmol/L HE-β-CD, pH of 6.5, temperature of 83 °C and
reaction time of 18 hours, the enantiomeric excess and overall conversion rate were 96.05%
and 27.28%, respectively. This work provides an efficient alternative method for improving
the conversion of aromatic ester substrates by including β-cyclodextrin in an aqueous
hydrolysis reaction system.
Keywords: Lipase; Enantioselective hydrolysis; 2-Phenylbutyric acid; Response surface
methodology
1. Introduction
2-Phenylbutyric acid (2-PBA) (Fig. 1) has a wide range of applications, especially in the
pharmaceutical industry. It is an important intermediate for the synthesis of a variety of
aromatic acid drugs, which are commonly used to treat indigestion and some other diseases

and for their anti-thrombotic effect. For example, indobufen (2-(4-(1-oxoisoindolin-2-yl)
phenyl) butanoic acid), synthesized from 2-phenylbutyric acid [1], is a new drug with
reversible, selective multi-target, and anti-thrombotic properties [2,3]. Studies have shown that
the anti-thrombotic activity of (S)-indobufen is more pronounced than that of (R)-indobufen,
and a similar phenomenon has also been observed for most of the other aromatic acids [4].
Therefore, obtaining optically pure (S)-2-PBA is of great importance in the pharmaceutical
industry. Currently, capillary electrophoresis [5], chromatography [6], and liquid-liquid
extraction [7] are applied for the resolution of 2-PBA enantiomers. However, the low stability
and sensitivity of capillary electrophoresis, high separation cost and low capacity of
chromatographic techniques, and low selectivity of liquid-liquid extraction limit their
applications. Hence, a new and efficient method for the resolution of 2-PBA enantiomers is
urgently needed.
Over the past decades, enzymatic kinetic resolution has been widely used for enantiomeric
resolution because of its excellent stereoselectivity, mild reaction conditions, and few side
reactions [8,9], and most importantly, it is environmentally friendly [10]. Lipase, a type of
hydrolases, is one of the most common classes of enzymes [11]. As reported in the literature,
some free lipases or immobilized lipases, including lipase AYS, lipase AK, lipase PS,
Novozym 40086, lipozyme TL IM, Novozym 435, and lipozyme RM IM, have been widely
used for the chiral resolution of racemic esters, carboxylic acids, and alcohols [12-21]. Lipase
CALA is a member of the α/β-hydrolase family. The excellent thermal stability, activity, and
stereoselectivity of lipase CALA make it critically important for industry [22-25].
Enantioselective hydrolysis is a common method for the enzymatic kinetic resolution of

enantiomers [26,27]. However, the substrate conversion is generally low because of the poor
solubility of the racemic substrate in the aqueous phase. Many methods have been utilized to
address this problem, such as, modification of the lipase [28,29], immobilization of the lipase
[30,31], and addition of a surfactant [32] or crown ether [33].
Cyclodextrin (CD) derivatives have attracted a large amount of attention in the fields of
catalysis and separation [34,35], and they have a three-dimensional ring structure that forms a
conical hollow cylinder with a hydrophobic interior and hydrophilic exterior. This distinct
structure of CD derivatives can effectively increase the solubility of some hydrophobic drugs
in water by forming inclusion complexes [14,36], which may be useful for enhancing the
conversion of an ester substrate for enzyme-catalyzed hydrolysis. The results from several
studies on enzymatic kinetic resolution in aqueous systems have supported this hypothesis
[14,37,38].
The efficiency of lipase-catalyzed resolution depends on many process factors, and
process optimization requires many experiments. Response surface methodology (RSM) is a
powerful tool that has been used to evaluate the interactive effects among multiple process
factors and describe the overall process [39]. In the present work, the lipase-catalyzed
hydrolysis of 2-phenylbutyrate ester (2-PBAE) to prepare (S)-2-PBA with lipase CALA in
aqueous medium was studied. β-CD derivatives were used as a green additive to improve the
conversion of insoluble substrates in the aqueous hydrolysis reaction system. RSM was used
to simulate and optimize the reaction process.
2. Experimental
2.1. Lipase and reagents

Commercially available free lipases, including lipase AYS from Candida rugosa (white
powder, activity 30000 U/g), lipase AK from Pseudomonas fluorescens (light yellow powder,
activity 20000 U/g), and lipase PS from Burkholderia cepacia (white powder, activity 30000
U/g), were purchased from Chengna Chemical Co., Ltd. (Nanjing, China). Lipase CALA from
Candida antarctica (brown liquid, activity 6000 LU/g) was purchased from Gao Ruisen
Technology Co., Ltd. (Beijing, China). Four immobilized lipases, including Novozym 435
from Candida antarctica immobilized on macroporous resin (activity 10000 U/g), lipozyme
RM IM from Rhizomucor miehei immobilized on macroporous anion exchange resin (activity
250 IUN/g), Novozym 40086 from Aspergillus oryzae immobilized on acrylic resin beads
(activity 275 IUN/g), and lipase TL IM from Thermomyces lanuginosus immobilized on silica
(activity 250 IUN/g), were purchased from Novozymes Biopharma DK A/S (Denmark). All of
the lipases were used without further treatment.
(R,S)-2-PBA containing equal proportions (50%) of the two enantiomers (purity > 98%)
was obtained from Huacheng Chemical Industry Co., Ltd. (Shanghai, China). Tris
(hydroxymethyl) methyl aminomethane was obtained from Shanpu Chemical Co., Ltd.
(Shanghai, China). β-CD derivatives were purchased from Qianhui Fine Chemical & Co. Inc.
(Shandong, China). Nine alcohols were supplied by Sinopharm Chemical Reagent Co., Ltd.
(Shanghai, China). The solvents for chromatography were of high performance liquid
chromatography (HPLC) grade and other reagents were of analytical grade, which were
purchased from different suppliers.
2.2. Synthesis of 2-PBAE
(R,S)-2-PBAE was obtained by esterification reactions between (R,S)-2-PBAand different

alcohols, where (R,S)-2-PBA and an alcohol at a molar ratio of 1:1 were dissolved in 40 mL
toluene and reacted at 110 °C with stirring for 12 hours in the presence of 0.05 g p-
toluenesulfonic acid as a catalyst. After completion of the reaction, the resultant mixture was
washed with saturated NaHCO3 solution (3 × 40 mL) to remove the excess substrate and then
washed with deionized water until neutral. The organic phase was separated, dried over
anhydrous MgSO4 for 12 hours, filtered and concentrated under reduced pressure to give the
desired yellow liquid product. The yield and purity of 2-PBAE were higher than 90% and 95%,
respectively. Because the esterification is a non-selective reaction, the proportion of each
enantiomer before and after the reaction did not change.
2.3. Lipase-catalyzed hydrolysis of 2-PBAE
The lipase-catalyzed hydrolysis reaction was carried out as follows: a certain amount of
substrate was first placed in a 25-mLreaction tube, and 1 mL0.1 mol/LTris/HCl buffer solution
containing HE-β-CD was decanted into the tube. Athermostatic apparatus (IKA® RCT B S025)
was employed to maintain a constant temperature for the reaction. After reaching the required
temperature, lipase was added to start the reaction, and the reaction system was stirred with a
magnetic stirrer at 600 rpm. The reaction was terminated with 1 mL of acetonitrile, and then,
the reaction mixture was filtered. The reaction mechanism is shown in Fig. 2.
2.4. HPLC analysis
The concentrations of 2-PBA enantiomers were measured with a HPLC instrument
(Waters 1525 binary pump system) equipped with UV/visible detection set at 225 nm (Waters
2489). The analysis was performed on an Inertsil ODS-3 column (250 mm × 4.6 mm, 5 μm).
The mobile phase was consisted of methanol and 0.5% acetic acid aqueous solution containing

25 mmol/L HP-β-CD (36:64, v/v, pH = 4.0 adjusted with triethylamine). The flow rate was 0.8
mL/min, and the column temperature was set at 30 °C. The sample volume was 10 μL [7]. The
retention times of (S)-2-PBA and (R)-2-PBA were 34.21 and 37.56 min, respectively. The
purity of all synthesized esters was also confirmed by HPLC with a mobile phase consisting of
acetonitrile and water (7:3, v/v).
The enantiomeric excess of the product (eep) and conversion rate (c) are defined as
[(S)  acid][(R)  acid]
eep 
100%
(1)
[(S) acid][(R)  acid]
[(S)  acid][(R)  acid]
[(S) ester]0 [(R) ester]0
c 
100%
(2)
The enantioselectivity, E, is expressed as
ln[1(1 K)c(1 eep )]
E 
(3)
ln[1(1 K)c(1eep )]
where K is the reaction equilibrium constant and the subscript 0 is the initial concentration.
2.5. Experimental design and data analysis
Based on a single factor investigation, RSM was used to optimize the process of the lipase-
catalyzed hydrolysis 2-PBAE. Compared with the common method, the RSM and central
composite design can reflect the effects of process variables and the interactions among them
more clearly, allowing the number of required experimental runs and time for process
optimization to be reduced greatly.
The independent variables examined in this study were temperature, pH of the aqueous
phase, HE-β-CD concentration, enzyme concentration, substrate concentration, and reaction
time. The dependent variables were enantiomeric excess of the product and conversion of the
ester substrate. The code and value range of the independent variables are shown in Table 1.

The experimental data were regression analyzed with the response surface to satisfy the
following quadratic polynomials equations:
k
k
k
ee    X   X 2 
 X X
(4)
(5)


 
p
0
i
i
ii
i
ij
i
j
i1
i1
i1 ji1
k
k
k
c     X   X 2 
 X X


 
0
i
i
ii
i
ij
i
j
i1
i1
i1 ji1
where eep and c are the predicted values of response, α0 and β0 are constants, αi and βi are linear
coefficients, αii and βii are squared coefficients, αij and βij are cross-product coefficients, k is the
number of factors, and Xi and Xj are the independent variables. The experimental data obtained
were analyzed by analysis of variance (ANOVA) and fitted to a second-order polynomial
equation using multiple regression analysis.
3. Results and discussion
3.1. Lipase-catalyzed hydrolysis reaction system
3.1.1. Screening of lipase
The activities of eight lipases were evaluated to screen the optimum lipase for the
hydrolysis resolution of racemic hexyl 2-phenylbutyrate (2-PBAHE). As shown in Table 2,
lipase AK and lipozyme TL IM had no catalytic activity for the hydrolysis of the substrate.
Lipase PS and Novozym 40086 had more activity toward (R)-2-PBAHE. Lipase CALA
exhibited the highest selectivity (eep up to 64.46%) among the lipases; however, the conversion
rate of 5.45% was relatively low. In the case of Novozym 435, the highest conversion rate of
36.6% was obtained, but with nearly no selectivity toward 2-PBAHE. To obtain a high optical
purity of the target product, lipase CALA was selected for further investigation.
3.1.2. Influence of substrate structure

The lipase-catalyzed hydrolysis reaction occurs by forming an intermediate complex
between the active site of the lipase and the substrate, following a strict molecular match. Many
studies have noted that substrate access to the active site of the lipase is one of the main factors
controlling the specificity of catalysis and enantioselectivity [8]. 2-PBAEs with different
carbon chain lengths were investigated under identical conditions. As shown in Table 3, the
carbon chain length of 2-PBAE had a significant effect on eep and c in the lipase-catalyzed
hydrolysis reaction. For substrates with a straight chain, both eep and c increased gradually,
with the increase in carbon chain length ranging from methyl ester to hexyl ester. For branched
chains, eep and c were obviously lower than those for a straight chain with the same number of
carbon atoms. Notably, there were rapid decreases in eep and c for octyl ester. These results
suggest that the substrate specificity may be depend on the binding between the length and
shape of the substrate and the activity center of lipase. Short-chain substrates with lower
specific interactions and long-chain substrates with larger steric hindrance result in decreased
c [40]. Based on the above results, the 2-PBAHE was selected as the optimal substrate for
further experiments.
3.1.3. Screening of CD derivatives
As shown in Table 3, a high eep was achieved with lipase CALA as the catalyst, but the c
of (S)-2-PBAHE was relatively low for industrial-scale production. This may be due to the low
solubility of 2-PBAHE in this aqueous reaction system. Thus, increasing the solubility may be
an effective way to improve the conversion. CD derivatives have a special structure with an
outer hydrophilic area and inner hydrophobic cavity. Many studies have suggested that they
can form inclusion complexes with a variety of hydrophobic aromatic compounds

[14,34,36,38,]. It was found in our previous work that β-CD derivatives can significantly
increase the solubility of hydrophobic enantiomers and recognize a series of enantiomers
[7,41,42]. In this study, five β-CD derivatives were separately added to the aqueous reaction
system (Fig. 3). As shown in Table 4, each β-CD derivative had a different effect on c, possibly
because the solubility was enhanced through the formation of inclusion complexes between 2-
PBAHE and β-CD derivatives. The stability of the inclusion complexes can be influenced by
the chemical structure of β-CD derivatives. It was observed that adding SBE-β-CD resulted in
nearly no improvement in c. This phenomenon can be explained by the dissociation of SBE-β-
CD, which makes it highly polar and then hinders the inclusion of the hydrophobic 2-PBAHE.
The c increased when HP-β-CD, CM-β-CD, HE-β-CD or Me-β-CD was used, and eep remained
nearly unchanged. The reason for these results may be that more stable inclusion complexes
are formed between the above β-CD derivatives and 2-PBAHE because of the better alignment
of their molecular structure. Among them, HE-β-CD was selected as the most suitable additive
because of its low price and easy access.
3.2. Regression analysis and model fitting
The effects of six independent variables on the dependent variables (eep and c) were
investigated (Table S1). These experimental results were fitted with a second-order polynomial
equation, and the values of the regression coefficients were calculated. The multiple quadratic
regression models for eep and c can be expressed by the second-order polynomial equations (6)
and (7):
ee %  92.85+ 2.92A+8.62B-0.96C-3.77D-0.14E-3.32F+8.75BF+0.7CF
+0.084DF-9.36A2 -7.76B2 -1.63C2 -2.94D2 -0.24E2 -0.13F2
p  
(6)

c %  27.70-3.09A-9.49B+6.72C+8.73D-10.80E+9.70F-11.32BF+ 4.80CF
+2.39DF-7.66A2 -7.37B2 -2.74C2 -2.59D2 +1.64E2 -3.46F2
 
(7)
The performance of the models was checked to prevent misleading results. Pareto analysis
of variance (ANOVA) was performed to test the accuracy of the models (Table S2). The smaller
the p-value was, the more significant the corresponding coefficient. A p-value < 0.0500
indicated that the model terms were significant. As shown in Table S2, the p-values for all the
corresponding coefficients were lower than 0.0001, indicating that the quadratic models (6)
and (7) were significant and adequate to simulate the relationship between the responses and
the independent variables. Moreover, the p-value can also be used to check the significance of
each coefficient and indicate the pattern of interactions among variables. In this case, A, C, D,
A2, B2, C2, and D2 were significant for model eep, and the order of significance was A > C >
D > E > F > B. Similarly, A, C, D, E, A2, B2, and F2 were significant for model c, and the order
of significance was A > C > E > D > F > B.
From the analysis of variance results shown in Table S2, the R2 values of eep and c were
0.9850 and 0.9952, respectively, indicating that the models were reliable. The "Pred R-
Squared" values of 0.9681 and 0.9770 were in reasonable agreement with the "Adj R-Squared"
values of 0.9769 and 0.9926, respectively. The coefficients of variation of the models were
0.45% and 4.69%, respectively, indicating that the models had good stability. Therefore, the
eep and c of the lipase-catalyzed hydrolysis resolution 2-PBAHE for the preparation of (S)-2-
PBA can be analyzed and predicted by these models.
3.3. Effects of multiple factors
3.3.1. Effects of temperature and pH value
Fig. 4a and Fig. 4b show the surface response plots for eep and c as functions of pH from

4.0 to 8.0 and temperature from 65 to 90 °C. As shown in Fig. 4, c and eep increased at first
and then decreased as the temperature increased. As the temperature increased, the chance of
collision between the enzyme and the substrate molecules increased, which may favor the
formation of the enzyme-substrate complex, then promoting c. However, an excessive
temperature may result in the decomposition of weak ions and hydrogen bonding, causing the
denaturation of the protein structure [43]. The eep also decreased because the conformation of
the active site of the enzyme was altered at high temperatures, affecting the enantioselectivity
of the enzyme. Moreover, as shown in Fig. 4, an extreme pH will lead to a decrease in eep. The
active center of the lipase is a “catalytic triad” composed of Ser, His and another amino acid
residue. The imidazole moiety of the His directly participates in the reaction, which is the key
step for the catalytic properties of lipase [44]. An extreme pH can disrupt the lipase tertiary
structure and hydrogen bonding, resulting in the loss of enantioselectivity and catalytic activity
[12,13]. In addition, the pH can also alter the existing state of the substrate and lipase, which
affects the binding and catalytic ability of the enzyme towards the substrate. Fig. 4a shows that
the highest eep was obtained at a pH of 6.5. To obtain an eep value greater than 96%, a
temperature of 83 °C and pH of 6.5 were set as the optimal conditions based on the response
surface graphs.
3.3.2. Effects of the concentrations of HE-β-CD and the substrate
As shown in Table 4, the addition of HE-β-CD improved the conversion. In order to
further optimize the reaction system, the effects of the concentrations of HE-β-CD and the
substrate and their mutual interaction on c and eep were studied. As shown in Fig. 5, under a
fixed substrate concentration, eep first increased and then decreased as the HE-β-CD

concentration increased. A possible reason for this result is that HE-β-CD can form inclusion
complexes with 2-PBAHE and has a stronger inclusion ability towards (S)-2-PBAHE than (R)-
2-PBAHE. However, an excessive HE-β-CD concentration (＞60 mmol/L) may lead to more
(R)-2-PBAHE being included, resulting in a drop in eep. Because the 2-PBAHE molecular is
soluble in the reaction system when HE-β-CD was added, c increased rapidly with the increase
in HE-β-CD concentration, and then, the trend became more subtle. There was a decrease in c
as the substrate concentration increased with a fixed HE-β-CD concentration, possibly because
the active sites of the enzyme are limited or a part of the substrate could not be catalyzed at the
fixed time, which lead to a decrease in c. It was concluded that the appropriate concentration
of HE-β-CD was 60 mmol/L to obtain a high eep (>95%) and that the concentration of the
substrate should not be too high.
3.3.3. Effects of the concentrations of lipase CALA and the substrate
The effects of the concentrations of lipase and the substrate on eep and c were further
studied. As shown in Fig. 6, the concentration of the substrate had few effects on the eep but
a large effect on conversion, the same phenomenon observed in Fig. 5. The c increased
continuously as the lipase concentration increased, while the eep showed a small change from
20 to 50 mg/mL and then decreased dramatically with a further increase to 90 mg/mL. This
phenomenon was consistent with the literature [16]. This can be explained by the fact that the
available active sites for substrate binding increase, and the reaction rates for the two
enantiomers of the substrate both increase at a high lipase concentration. (S)-2-PBAHE
reaches its reaction equilibrium faster, and the apparent rate decreases gradually. However,
(R)-2-PBAHE is far from reaction equilibrium, and the reaction rate is enhanced by increasing

the concentration of lipase, which results in a decrease in eep. From the perspective of
economic efficiency, a higher substrate concentration is preferred while obtaining a suitable
conversion rate. To obtain a higher eep and satisfactory c, the proposed concentrations of the
lipase and substrate were 50 mg/mL and 30 mmol/mL, respectively.
3.3.4. Effects of the concentration of HE-β-CD and reaction time
As shown in Fig. 7, eep and c were clearly influenced by the concentration of HE-β-CD
and reaction time. The eep first increased to a maximum value and then decreased with the
further increase in HE-β-CD concentration, showing a similar trend as that observed in Fig. 5.
At any HE-β-CD concentration, eep decreased with time (Fig. 7a). As shown in Fig. 7b, the c
increased slightly under a low HE-β-CD concentration, even though the reaction time
continuously increased. However, under a relatively high concentration of HE-β-CD, c
increased rapidly at first and then increased slowly after 18 hours. This phenomenon can be
explained by the fact that a larger number of complexes between HE-β-CD and the substrate
are formed in the aqueous phase at high HE-β-CD concentrations, which enhances the reaction
as explained above. Therefore, the concentration of HE-β-CD was set at 60 mmol/L to achieve
a maximum eep and relatively high c. Prolonging the reaction time can increase c, but eep will
decrease after 18 hours of reaction time. Therefore, a reaction time of 18 hours was selected as
the suitable reaction time.
3.3.5. Effects of the concentration of lipase CALA and reaction time
In a lipase-catalysis reaction, the reaction rate can be increased by increasing the
concentrate of lipase, but the cost is also increased. Therefore, the concentration of lipase and
reaction time should be optimized. As shown in Fig. 8a, eep decreased rapidly as the lipase

concentration and reaction time increased. In addition, as show in Fig. 8b that increasing the
lipase concentration shortened the reaction time to achieve a high conversion. Combining the
results in Figs. 8a and 8b, 50 mg/mL lipase and a reaction time of 18 hours were proposed for
the hydrolysis of 2-PBAHE to prepare (S)-2-PBA, which conform with the results shown in
Fig. 6 and Fig. 7.
3.4. Validation of the model
In order to further test the validity of the model predictions, additional independent
experiments were carried out at the suggested optimum conditions: a 1 mL Tris/HCl buffer
solution system containing 60 mmol/L HE-β-CD, 50 mg/mL lipase CALA, and 30 mmol/L
substrate at 83 °C and pH of 6.5 and reacting for 18 hours. Three groups of parallel experiments
were performed. Fig. 9 shows the chromatograms of the products of the lipase CALA-catalyzed
hydrolysis of (R,S)-2-PBAHE. The values of eep and c were 96.05% and 27.28%, respectively.
The predicted values of the model were 95.31% and 28.96%, respectively. The relative errors
of eep and c between the experimental value and the predicted value were less than 1.00%,
indicating that the model is reliable to predict the preparation of (S)-2-PBA by the lipase
CALA-catalyzed hydrolysis of 2-PBAHE.
4. Conclusions
The lipase-catalyzed hydrolysis of 2-PBAE to prepare (S)-2-PBA was studied
systematically. The effects of various factors on enantiomeric excess and conversion rate were
investigated, such as the type of β-CD derivatives, substrate concentration, lipase concentration,
HE-β-CD concentration, temperature, pH, and reaction time. Lipase CALA was selected as the
most effective catalyst. The addition of HE-β-CD increased the conversion by 1.5 times. RSM

was used to obtain the optimal process parameters and understand the relationship between the
resolution efficiency and process parameters. The reaction system had an optimal temperature
and pH. The concentration of the substrate had an effect on c but not on eep. Under the optimal
conditions: 50 mg/mL lipase CALA, 30 mmol/L substrate, 60 mmol/L HE-β-CD, pH of 6.5,
temperature of 83 °C and reaction time of 18 hours, the experimental values of eep and c were
96.05% and 27.28%, respectively, which are in good agreement with the predicted values. This
work provides a simple and feasible method for obtaining optically pure (S)-2-PBA and
demonstrates that β-CD derivatives are useful for improving the conversion of an aromatic
ester substrate for the hydrolysis process in an aqueous reaction system. The results provide
theoretical guidance for process design and operation. However, before applying the resolution
system to industrial-scale equipment, for example, a stirred tank reactor, additional factors,
including scaling effects, should be considered.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (No.
21676077), the Open Fund Project of Key Laboratory in Hunan University (No. B060303),
and Aid program for Science and Technology Innovative Research Team in Higher Educational
Institutions of Hunan Province.
References
[1] L.M. Fuccella, G. Corvi, E. Moro, Pharmacokinetic, bioavailability and pharmacodynamic
study of indobufen (K 3920), an inhibitor of platelet aggregation, after a single dose in
man, Eur. J. Clin. Pharmacol. 15 (1979) 323-327. https://doi.org/10.1007/BF00558435.
[2] S. Sahar-Helft, T. Chackartchi, D. Polak, Dental treatment in the era of new anti-thrombotic
agents, Int. Dent. J. 68 (2018) 131-137. https://doi.org/10.1111/idj.12322.

[3] A.G. Rebuzzi, A. Natale, C. Bianchi, Effects of indobufen on platelet thromboxane B2
production in patients with myocardial infarction, Eur. J. Clin. Pharmacol. 39 (1990) 99-
100. https://doi.org/10.1007/BF02657071.
[4] A.J. Hutt, J. Caldwell, The importance of stereochemistry in the clinical pharmacokinetics
of the 2-arylpropionic acid non-steroidal anti-inflammatory drugs, Clin. Pharmacokinet.
9 (1984) 371-373. https://doi.org/10.2165/00003088-198409040-00007.
[5] R.J. Robins, J.G. Woolley, M. Ansarin, Phenyllactic acid but not tropic acid is an
intermediate in the biosynthesis of tropane alkaloids in datura and brugmansia
transformed root cultures, Planta 194 (1994) 86-94. https://doi.org/10.2307/23383030.
[6] T. Jira, A. Bunke, A. Karbaum, Use of chiral and achiral ion-pairing reagents in
combination with cyclodextrins in capillary electrophoresis, J. Chromatogr. A 798 (1998)
281-288. https://doi.org/10.1016/s0021-9673(97)01168-0.
[7] W.F. Xu, G.L. Dai, K.W. Tang, P.L. Zhang, B.Q. Xiong, L. Yu, Continuous chiral
separation of 2-phenylbutyric acid by liquid-liquid extraction in centrifugal contactor
separators,
Sep.
Purif.
Technol.
179
(2017)
53-60.
https://doi.org/10.1016/j.seppur.2017.01.063.
[8] D. Romano, F. Bonomi, M.C. Mattos, Esterases as stereoselective biocatalysts, Biotechnol.
Adv. 33 (2015) 547-565. https://doi.org/10.1016/j.biotechadv.2015.01.006.
[9] D. Gérard, M. Guéroult, L. Casas-Godoy, Efficient resolution of profen ethyl ester
racemates by engineered Yarrowia lipolytica Lip2p lipase, Tetrahedron: Asymmetry 28
(2017) 433-441. https://doi.org/10.1016/j.tetasy.2017.01.014.
[10] V. Gotor-Fernández, R. Brieva, V. Gotor, Lipases: useful biocatalysts for the preparation

of pharmaceuticals, J. Mol. Catal. B-Enzym. 40 (2006) 111-120.
https://doi.org/10.1016/j.molcatb.2006.02.010.
[11] A. Ghanem, V. Schurig, Lipase-catalyzed access to enantiomerically pure (R)- and (S)-
trans-4-phenyl-3-butene-2-ol,
Tetrahedron:
Asymmetry 14
(2003)
57-62.
https://doi.org/10.1016/S0957-4166(02)00745-0.
[12] H.P. Dong, Y.J. Wang, Y.G. Zheng, Enantioselective hydrolysis of diethyl 3-
hydroxyglutarate to ethyl (S)-3-hydroxyglutarate by immobilized Candida antarctica
lipase
B,
J.
Mol.
Catal.
B-Enzym.
66
(2010)
90-94.
https://doi.org/10.1016/j.molcatb.2010.03.009.
[13] S.H. Cho, P.Y. Wang, S.W. Tsai, Lipase-catalyzed hydrolytic resolution of (R, S)-3-
hydroxy-3-phenylpropionates in biphasic media, J. Taiwan Inst. Chem. E. 42 (2011) 408-
412. https://doi.org/10.1016/j.jtice.2010.09.005.
[14] S.H. Kim, T.K. Kim, G.S. Shin, Enantioselective hydrolysis of insoluble (R,S)-ketoprofen
ethyl ester in dispersed aqueous reaction system induced by chiral cyclodextrin,
Biotechnol.
https://doi.org/10.1023/B:BILE.0000030040.13828.d7.
[15] R.E. Deasy, M. Brossat, T.S. Moody, Lipase catalysed kinetic resolutions of 3-
arylalkanoic acids, Tetrahedron: Asymmetry 22 (2011) 47-61.
https://doi.org/10.1016/j.tetasy.2010.12.019.
[16] G.D. Yadav, I.V. Borkar, Kinetic modeling of immobilized lipase catalysis in synthesis
of n-butyl levulinate, Ind. Eng. Chem. Res. 47 (2008) 3358-3363.
https://doi.org/10.1021/ie800193f.
Lett.
26
(2004)
965-969.

[17] S.L. Wei, A.H. Kamaruddin, S. Bhatia, Enzyme kinetics of kinetic resolution of racemic
ibuprofen ester using enzymatic membrane reactor, Chem. Eng. Sci. 60 (2005) 4957-4970.
https://doi.org/10.1016/j.ces.2005.03.016.
[18] K. Won, J.K. Hong, K.J. Kim, Lipase-catalyzed enantioselective esterification of racemic
ibuprofen coupled with pervaporation, Process Biochem. 41 (2006) 264-269.
https://doi.org/10.1016/j.procbio.2005.07.006.
[19] J. Xiong, J.P. Wu, G. Xu, L.R. Yang, Kinetic study of lipase catalyzed asymmetric
transesterification of mandelonitrile in solvent-free system, Chem. Eng. J. 138 (2008)
258-263. https://doi.org/10.1016/j.cej.2007.05.034.
[20] J.B. Sontakke, G.D. Yadav, Kinetic modeling and statistical optimization of lipase
catalyzed enantioselective resolution of (R,S)-2-pentanol, Ind. Eng. Chem. Res. 50
(2011)12975-12983. https://doi.org/10.1021/ie2012032.
[21] G.Y. Liu, P.L. Zhang, W.F. Xu, L.J. Wang, K.W.Tang, Lipase-catalyzed hydrolysis of
(+,-)-2-(4-methylphenyl) propionic methyl ester enhanced by hydroxypropyl-β-
cyclodextrin,
J.
Chem.
Technol.
Biotechnol.
94
(2019)
147-158.
https://doi.org/10.1002/jctb.5756.
[22] D.J. Ericsson, A. Kasrayan, P. Johansson, M, A.G. Sandstr, X-ray structure of Candida
antarctica lipase a shows a novel lid structure and a likely mode of interfacial activation.
J. Mol. Biol. 376 (2008) 109-119. https://doi.org/10.1016/j.jmb.2007.10.079.
[23] S. Kim, Y.K. Choi, J. Hong, J. Park, M.J. Kim, Candida antarctica lipase A and
pseudomonas stutzeri lipase as a pair of stereocomplementary enzymes for the resolution
of 1,2-diarylethanols and 1,2-diarylethanamines, Tetrahedron Lett. 54 (2013) 1185-1188.

https://doi.org/10.1016/j.tetlet.2012.11.147.
[24] D. Kahveci, X. Xu, Enhancement of activity and selectivity of Candida rugosalipase and
Candida antarcticalipase A by bioimprinting and/or immobilization for application in the
selective ethanolysis of fish oil, Biotechnol. Lett. 33 (2011) 2065-2071.
https://doi.org/10.1007/s10529-011-0671-z.
[25] Y. Wikmark, M.S. Humble, J.E. Bäckvall, Combinatorial library based engineering of
Candida antarctica lipase A for enantioselective transacylation of sec-alcohols in organic
solvent,
Angew.
Chem.
Int.
Edit.
54
(2015)
4284-4288.
https://doi.org/10.1002/anie.201410675.
[26] R.N. Patel, Industrial applications of biocatalytic hydrolysis (esters, amides, epoxides,
nitriles) and biocatalytic dynamic kinetic resolution, Comprehensive Chirality 9 (2012)
288-317. https://doi.org/10.1016/B978-0-08-095167-6.00913-7
[27] A.H. Kamaruddin, M.H. Uzir, H.Y. Aboulenein, Chemoenzymatic and microbial dynamic
kinetic resolutions, Chirality 21 (2010) 449-467. https://doi.org/10.1002/chir.20619.
[28] M. Cancino, P. Bauchart, G. Sandoval, J.M. Nicaud, I. André, V. Dossat, A variant of
Yarrowia lipolytica lipase with improved activity and enantioselectivity for resolution of
2-bromo-arylacetic acid esters, Tetrahedron: Asymmetry 19 (2008) 1608-1612.
https://doi.org/10.1016/j.tetasy.2008.06.009.
[29] A. Uyanik, N. Sen, M. Yilmaz, Enhancing effect of calix[4] arene amide derivatives on
lipase performance in enantioselective hydrolysis of racemic arylpropionic acid methyl
esters,
Polycycl.
Aromat.
Compd.
36
(2015)
613-627.
https://doi.org/10.1080/10406638.2015.1037005.

[30] G. Fernández-Lorente, M. Terreni, C. Mateo, Modulation of lipase properties in macro-
aqueous systems by controlled enzyme immobilization: enantioselective hydrolysis of a
chiral ester by immobilized Pseudomonas lipase, Enzyme Microb. Technol. 28 (2001)
389-396. https://doi.org/10.1016/s0141-0229(00)00324-0.
[31] K. Kawakami, M. Ueno, Y. Oda, Application of a Burkholderia cepacia lipase-
immobilized silica monolith micro-bioreactor to continuous-flow kinetic resolution for
transesterification of (R,S)-1-phenylethanol, Process Biochem. 47 (2012) 147-150.
https://doi.org/10.1016/j.procbio.2011.09.017.
[32] D. Brady, L. Steenkamp, E. Skein, Optimisation of the enantioselective biocatalytic
hydrolysis of naproxen ethyl ester using ChiroCLEC-CR, Enzyme Microb. Technol. 34
(2004) 283-291. https://doi.org/10.1016/j.enzmictec.2003.11.002.
[33] Y. Mine, K. Fukunaga, K. Itoh, Enhanced enzyme activity and enantioselectivity of lipases
in organic solvents by crown ethers and cyclodextrins, J. Biosci. Bioeng. 95 (2003) 441-
447. https://doi.org/10.1016/s1389-1723(03)80042-7.
[34] V. Reynaldo, C.A. Roberto, A. Fragoso, Supramolecular chemistry of cyclodextrins in
enzyme
technology,
Chem.
Rev.
107
(2007)
3088-3116.
https://doi.org/10.1002/chin.200743251.
[35] J. Szejtli, Utilization of cyclodextrins in industrial products and processes, J. Mater. Chem.
7 (1997) 575-587. https://doi.org/10.1039/a605235e.
[36] D. Duchene, Cyclodextrins and their inclusion complexes, Akadémiai Kiadó, 1982.
[37] G.S. Shin, K.W. Lee, T.K. Kim, Lipase-catalyzed production of optically active (S)-
flurbiprofen in aqueous phase reaction system containing chiral succinyl β-cyclodextrin,

J. Mol. Catal. B-Enzym. 33 (2005) 93-98. https://doi.org/10.1016/j.molcatb.2005.02.005
[38] H.D. Shin, J.H. Kim, T.K. Kim, Esterification of hydrophobic substrates by lipase in the
cyclodextrin induced emulsion reaction system, Enzyme Microb. Technol. 30 (2002) 835-
842. https://doi.org/10.1016/S0141-0229(02)00025-X.
[39] R.H. Myers, D.C. Montgomery, C.M. Anderson, Response surface methodology: process
and product optimization using designed experiments, John Wiley & Sons, 2016.
[40] A. Torres-Gavilán, E. Castillo, A. López-Munguía, The amidase activity of candida
antarctica, lipase B is dependent on specific structural features of the substrates, J. Mol.
Catal. B-Enzym. 41 (2006) 136-140. https://doi.org/10.1016/j.molcatb.2006.06.001.
[41] K.W. Tang, L.T. Song, Y.B Liu, Separation of flurbiprofen enantiomers by biphasic
recognition chiral extraction, Chem. Eng. J. 158 (2010) 411-417.
https://doi.org/10.1016/j.cej.2010.01.009.
[42] K.W. Tang, P.L. Zhang, C.Y. Pan, H.J. Li, Equilibrium studies on enantioselective
extraction of oxybutynin enantiomers by hydrophilic β-cyclodextrin derivatives, AIChE
J. 57 (2011) 3027-3036. https://doi.org/10.1002/aic.12513.
[43] D.H. Yu, Z. Wang, P. Chen, L. Jin, Y.M. Cheng, J.G. Zhou, S.G. Cao, Microwave-assisted
resolution of (R,S)-2-octanol by enzymatic transesterification, J. Mol. Catal. B-Enzym.
48 (2007) 51-57. https://doi.org/10.1016/j.molcatb.2007.06.009.
[44] P.Y. Wang, S.W. Tsai, T.L. Chen, Improvements of enzyme activity and enantioselectivity
via combined substrate engineering and covalent immobilization, Biotechnol. Bioeng.
101 (2008) 460-469. https://doi.org/10.1002/bit.21916.

Figure legends
Fig. 1. The chemical structure of (S)-2-PBA.
Fig. 2. Reaction diagram of hydrolysis resolution of (R,S)-2-PBAE by lipase.
Fig.3. The mechanism of inclusion interaction between β-CD derivatives and 2-PBAHE.
Fig. 4. Effects of temperature and pH value on (a) eep and (b) c. Conditions: 60 mmol/L HE-β-
CD, 30 mmol/L substrate, 50 mg/mL enzyme and reaction time of 18 hours.
Fig. 5. Effects of the concentrations of HE-β-CD and the substrate on (a) eep and (b) c.
Conditions: 50 mg/mL enzyme, temperature of 83 °C, pH of 6.5 and reaction time of 18 hours.
Fig. 6. Effects of the concentrations of enzyme and the substrate on (a) eep and (b) c. Conditions:
60 mmol/L HE-β-CD, temperature of 83 °C, pH of 6.5 and reaction time of 18 hours.
Fig. 7. Effects of the concentrations of HE-β-CD and reaction time on (a) eep and (b) c.
Conditions: 30 mmol/L substrate, 50 mg/mL enzyme, temperature of 83 °C and pH of 6.5.
Fig. 8. Effects of the concentrations of enzyme and reaction time on (a) eep and (b) c.
Conditions: 60 mmol/L HE-β-CD, 30 mmol/L substrate, temperature of 83 °C and pH of 6.5.
Fig. 9. Chromatograms of (a) the sample of racemic (R,S)-2-PBA and (b) the hydrolysis
products of (R,S)-2-PBAHE.

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6