Kinetic study of enantioselective hydrolysis of (R, S)-ketoprofen ethyl ester using immobilized T. laibacchii lipase

Yuan-Yuan Zhang,Jun-Hong Liu

DOI: https://doi.org/10.1016/j.bej.2011.01.005

IF: 3.9

2011-01-01

Biochemical Engineering Journal

Abstract:Immobilized Trichosporon laibacchii lipase was used to catalyzed enantioselective hydrolysis of ( S )-ketoprofen from ( R , S )-ketoprofen ethyl ester in an aqueous system containing Tween-80 as the emulsifier. An inherent kinetic model based on irreversible ping-pong bi–bi mechanism with competitive inhibition of both enantiomer products was developed. The diffusion limitations and enzyme deactivation were proved to be negligible. The solution of ordinary differential equations “ode45” was successfully combined with Matlab optimization function “fmincon” to estimate kinetic parameters. The kinetic experiments were carried out in a 100 mL stirred tank reactor at various temperatures (30, 35, 40 and 45 °C) at 200 rpm. The simulated values of the model fitted the experimental values well and the kinetic constant has a good fitness with Arrhenius equation. Thus, the kinetic model proposed here is feasible and can be used for mechanism interpretation of the hydrolytic resolution of racemic ketoprofen ester and the reactor scale-up design. High enantiomeric ratio ( E of 87) indicates that this biotransformation is a promising example for the industrial production of ( S )-ketoprofen. Keywords Immobilized enzymes Lipase Trichosporon laibacchii Kinetics Ketoprofen ethyl ester Enantioselective hydrolysis 1 Introduction Ketoprofen (2-(3-benzoylphenyl)-propionic acid) is one of the most useful nonstreroidal anti-inflammatory drugs (NSAIDs) [1] . However, only ( S )-ketoprofen has the therapeutic activity of reducing inflammation and relieving pains [2] . Kinetic resolution via enzymes is one of the most popular methods among many resolution techniques to prepare optically pure isomers [3] . In recent years, lipases (EC 3.1.1.3) have been successfully applied for the resolution of ( R , S )-ketoprofen through mainly direct enantioselective esterification in organic media or enantioselective hydrolysis of its chemically synthesized racemic ester [4–6] . The lipase from Trichosporon laibacchii is an intracellular lipolytic enzyme that showed ( S )-preference in the resolution of racemic ketoprofen [7] . However, the high cost of lipase production still remains some obstacles to its large-scale industrial applications as catalysts [8] . This problem can be overcome by employing lipase in immobilized form, where the enzyme having high stability can be reutilized easily. Although numerous studies dealt with biocatalytic kinetic resolutions, only few of them actually addressed the determination of reaction kinetics due to the fact that the reactions were heterogeneous and the added complications with determination of different enantiomers at the same time. An accurate kinetic characterization of the catalyst is necessary in order to optimize the design and operation of enzymatic reactors. Many studies [9–11] reported that the ping-pong bi–bi model could be used to describe the catalytic action of lipases irrespective of the type of biotransformation. Thorough investigation of the kinetic mechanism involved and the determination of the corresponding parameters is a key step in process development. It can be used to evaluate the potential utility of the reaction system and to support reaction engineering investigation for process development. In this study, immobilized T. laibacchii lipase was used to catalyze the enantioselective hydrolysis of ketoprofen ester to optically active ( S )-ketoprofen. The reaction was carried out in a 100 mL stirred tank reactor using an aqueous system containing Tween-80 as the emulsifier. An irreversible ping-pong bi–bi model with product inhibition of ( S )- and ( R )-enantiomer using King–Altman method was proposed to describe the reaction mechanism of the hydrolysis. 2 Materials and methods 2.1 Chemicals Racemic ketoprofen was purchased from Xinan Synthetic Pharmaceutical Factory (Chongqing, China). ( S )-Ketoprofen was purchased from Sigma. The ethyl ester of racemic ketoprofen was prepared using the method described by Kim et al. [12] . Tween-80 and PEG4000 were obtained from Shanghai Chemical Reagent Co. Ltd. (Shanghai, China). Diatomites (particle diameter 0.2–2 mm, pore diameter 25–30 nm, specific area 350–450 m 2 g −1 ) were purchased from Chengzhou Huali Diatomite Co. Ltd. (Jiangsu, China). All other chemicals were also obtained commercially and of analytical grade. 2.2 The preparation of immobilized lipase Free lipase was produced from a mutant of T. laibacchii CBS5791 by fermentation at 23 °C, pH 6.5, with an agitation of 200 rpm, aeration of 0.5 vvm for 13 h in a 15 L fermentor (DS-Y, Zhenjiang Dasen Fermentation Instruments Co. Ltd., China). The compositions of seed medium and fermentation medium are the same as described in our previous work [13] . After the fermentation period, the cells were harvested by centrifuging at 5000 rpm for 10 min and then resuspended in phosphate buffer (pH 6.5) to a final concentration of 150 g L −1 . Cell suspensions were disintegrated by high pressure homogenizer (JG-IA type, Scientz Co. Ltd., China) at 70 MPa for 3 cycles at 4 °C. Then the cell homogenate obtained (crude lipase) was subjected to purification by aqueous two-phase system (ATPS) formed by PEG4000/K 2 HPO 4 system of 12/13 (%, w/w) at pH 7.0, 2% NaCl. The immobilized lipase was prepared by an in situ immobilization through adsorption-crosslinking, where immobilization carriers were directly added to the top PEG-rich phase of ATPS. The immobilization conditions were as follows [14] : the additive amounts of the diatomites carriers 1.2 mg/U lipase, adsorption at 30 °C for 3 h, the final concentration of glutaraldehyde 0.15% (w/w), crosslinking at 15 °C for 2 h. 2.3 Kinetic experiments Time-course measurements of the hydrolysis reaction catalyzed by immobilized T. laibacchii lipase were performed in a temperature controlled 100 mL reactor with a turbine stirrer at various temperatures (30, 35, 40 and 45 °C). To 2.7 g immobilized lipase (1334 U/g) with particle diameter of 0.5 mm was added 50 mL phosphate buffer (pH 7.5) containing 1.5% (w/v) Tween-80 and 6% (w/v) racemic ketoprofen ethyl ester which was emulsified for 40 s using a homogenizer before the addition of lipase. The reaction time for each batch was 40 h. Samples were removed and injected onto the HPLC system at different time intervals for analysis. A numerical integral approach was used to fit the kinetic model to the experiment data: E (enantiomeric ratio)-, ee p (enantiomeric excess of the product)-, X S (convertion ratio of S-enantiomer)-, X R (convertion ratio of R -enantiomer)-, X (total convertion ratio)-time courses were compared to empirical data from the batch reactor. 2.4 External and internal diffusion experiments Using immobilized lipase with particle diameter of 0.5 mm, the effect of external diffusion limitation was studied by carrying out hydrolysis experiments at different stirring speeds from 50 to 300 rpm for 10 h at 45 °C. Other experimental conditions were the same as described in kinetic experiments. The effect of internal diffusion limitation was also studied by carrying out hydrolysis experiments using immobilized lipase with different particle diameter range from 0.2 to 1.7 mm at 200 rpm for 10 h at 45 °C. 2.5 HPLC analysis method Ketoprofen was analyzed by HPLC using a Chiracel OJ-H column (4.6 mm × 250 mm). Samples (5 μL) were eluted by a mixture of n-hexane:2-propanol:acetic acid (90:10:0.5, v/v/v) at a rate of 1.0 mL/min, and detected at a wavelength of 254 nm. The retention times of the unresolved ester, ( R )-ketoprofen and ( S )-ketoprofen, were 4.4, 13.6 and 17.8 min, respectively. Enantioselectivity was expressed as E value and calculated as follows: E = ln   [ 1 − X ( 1 + e e p ) ] ln   [ 1 − X   ( 1 − e e p ) ] where X represents the conversion ratio and ee p represents the enantiomeric excess of the product. The ee p value was calculated as [( S − R )/( S + R )] × 100%, where S and R were the chromatographic peak areas of the S - and R -enantiomers, respectively. 2.6 Estimation of kinetic parameters in the reaction rate equations The kinetic parameters in the reaction rate equations were estimated by combining ordinary differential equations solver with the optimization method. The proposed kinetic model was solved using the Matlab programming environment (Matlab 6.5, the Math Works, Natick, MA). The Matlab optimization function “fmincon” was used to fit the model and estimate kinetic parameters, Eqs. (3) and (4) , based on experimental data from kinetic experiments. The fourth-order Runge-Kutta method with variable time step size defined by the Matlab function “ode45” was used as the ordinary differential equations (the kinetic rate equations, Eqs. (5) and (6) ) solver. The Matlab optimization function “fminsearch” was employed to estimate parameters related to the temperature, activation energy, in the reaction rate equations, i.e., Eq. (7) . 3 Theory Model selection and discrimination is an important step in a kinetic investigation. The kinetic resolution of racemic ketoprofen ethyl ester catalyzed by immobilized T. laibacchii lipase may be written as two parallel occurring reactions ( Scheme 1 ). As T. laibacchii lipase shows ( S )-preference in the resolution of ( R , S )-profen, the hydrolysis rate of ( S )-ketoprofen ethyl ester is much higher than the ( R )-enantiomer, which results in the production of ( S )-ketoprofen as the preponderant enantiomer product. In our study, the hydrolytic resolution reaction of the ( R , S )-ketoprofen ethyl ester is carried out in the kinetic-controlled range by eliminating the internal and external diffusion limitation in the experiments. Therefore, the kinetic model will be established based on inherent kinetics theory. A reversible ping-pong bi–bi mechanism could be employed to describe the kinetic behavior of lipase-catalyzed hydrolysis of ketoprofen ester in aqueous solution. Referring to Scheme 2 , the first substrate ( S )- or ( R )-ketoprofen ester ( S S / S R ) binds to the enzyme ( E ) forming a lipase–ketoprofen ester complex ( ES S / ES R ), after which the corresponding product ethanol ( Q ) is released to form an acyl–enzyme complex ( F S / F R ). After that, the second substrate (water) binds to the acyl–enzyme complexes to form another binary complex ( F S S W / F R S W ). Finally the second product ( S )- or ( R )-ketoprofen ( P S / P R ) is released and enzyme is regenerated. Enantiomer product ( S )- or ( R )-ketoprofen causes product competing inhibition to form an enantiomer product–lipase complex ( EP S / EP R ). In the development of the model equations, the corresponding assumptions were made: (1) the enzyme is uniformly distributed throughout the particles, that is, enzyme adsorption during enzyme immobilization occurs inside the whole particle. This assumptions is reasonable, because the size of the T. laibacchii lipase used is about 3 nm × 4 nm × 5 nm according to Candida antarctica lipase B [15] , thus, the surface of the enzyme (12–20 nm 2 ) is much smaller than the pores in the diatomite support (1200–2700 nm 2 ). (2) Particles are spherical and uniform in size [16] . (3) The hydrolytic resolution reaction is irreversible. The hydrolysis of ketoprofen ethyl ester by immobilized lipase in aqueous systems where ethanol is a good leaving group and the reactant water is available at saturating conditions makes this assumption reasonable. (4) External mass transfer resistance between bulk liquid and particles is negligible. Experimental observations ( Fig. 1 a ) confirm this assumption, as the agitation is strong enough to ensure that external diffusion limitation is negligible when stirring speed is higher than 200 rpm. (5) Internal mass transfer resistance is also negligible. This assumption can be demonstrated by the experimental observations ( Fig. 1 b) where internal mass transfer resistance is negligible when particle diameter of immobilized lipase is about 0.5 mm. (6) Enzyme deactivation is negligible. This assumption is reasonable, because the reaction was carried out at relatively lower temperature (30–45 °C) for 40 h. Moreover, it was found that the immobilized T. laibacchii lipase has a good thermal stability and operation stability, and no noticeable deactivation occurs during the 40 h runs, for the range of the temperature used here (data not shown). Based on the above assumptions and experimental observations, the reversible ping-pong bi–bi mechanism of the immobilized lipase reactions could be simplified to an irreversible Michaelis–Menten mechanism. The King–Altman plot in Scheme 3 shows the interconversion patterns of enzyme, substrate, and product species for such a system. Both enantiomers can be considered as alternative substrates for the same active site. Therefore, it is easy to drive the rate equations for both enantiomers as follows: (1) V S = k 2 S [ S S ] [ E 0 ] / K M S 1 + [ S S ] / K M S + [ S R ] / K M R + [ P S ] / K P S + [ P R ] / K P R (2) V R = k 2 R [ S R ] [ E 0 ] / K M R 1 + [ S S ] / K M S + [ S R ] / K M R + [ P S ] / K P S + [ P R ] / K P R Notations [ E 0 ], [ P S ], [ P R ], [ S S ] and [ S R ] denote the concentrations of enzyme, ( S )-ketoprofen, ( R )-ketoprofen, ( S )- and ( R )-ketoprofen ester, respectively. Moreover, k 2 R , K MR , k 2 S , K MS , K PS and K PR are the kinetic constants in Michaelis–Menten kinetics and inhibition constants, respectively. Let: K 1 = k 2 S ; K 2 = 1 K M S ; K 3 = 1 K M R ; K 4 = 1 K P S ; K 5 = 1 K P R ; K 6 = k 2 R thus, Eqs. (1) and (2) change to Eqs. (3) and (4) , respectively, as following: (3) V S = K 1 [ E 0 ] K 2 [ S S ] 1 + K 2 [ S S ] + K 3 [ S R ] + K 4 [ P S ] + K 5 [ P R ] (4) V R = K 6 [ E 0 ] K 3 [ S R ] 1 + K 2 [ S S ] + K 3 [ S R ] + K 4 [ P S ] + K 5 [ P R ] Based on Eqs. (3) and (4) , the reaction can be described by the following differential equations with conversion ( X ): (5) d X S / d t = V S × M ester / ( M acid × S Z 0 )       = 1.1101 / 30 × V S       = 0.037   V S (6) d X R / d t = V R × M ester / ( M acid × R Z 0 )       = 0.037   V R where M acid , M ester represent the molecular weight of ketoprofen and ketoprofen ethyl ester, respectively; X S , X R represent the mass conversion of ( S )- and ( R )-ketoprofen, respectively; V S , V R represent the reaction rate (g/h L) of ( S )- and ( R )-ketoprofen, respectively; S Z 0 , R Z 0 represent the initial concentration (30 g/L) of ( S )- and ( R )-ketoprofen ethyl ester, respectively. 4 Results and discussion 4.1 Effect of internal and external diffusion Due to the use of an immobilized enzyme, the possible influences of internal and external diffusion limitations on the kinetics were investigated. The dependence of the conversion of ketoprofen ester versus stirring speed is shown in Fig. 1 a. The conversion increased rapidly with speeds from 50 to 150 rpm and was almost unchanged when the rotation speed was higher than 150 rpm, which indicated that the effect of external diffusion was not significant when the stirring speed was greater than 150 rpm. Therefore, the external diffusion limitation was negligible when the stirring speed was higher than 150 rpm. Thus, the kinetic measurements in this work were investigated at stirring speed of 200 rpm. With highly stirring speed, to enhance the efficiency of immobilized enzyme the external diffusion limitation was often negligible in the reaction catalyzed by the enzyme. Chen and Tsai [17] developed an enantioselective esterification process for the synthesis of 2- N -morpholinoethyl ( S )-ibuprofen ester prodrug from racemic ibuprofen by using Candida rugosa lipase immobilized on Accurel MP1000 in cyclohexane. The effect of external mass transfer on the reaction rate may be neglected in their work. Gonc¸alves et al. [18] concluded that external diffusion was not significant when the stirring speed is higher than 400 rpm in their work in which they studied influence of mass transfer limitations on the enzymatic synthesis of β-lactam antibiotics catalyzed by penicillin G acylase immobilized on glioxil-agarose. Based on Biot number calculations, Calcium-alginate gel beads were not subject to external mass transfer resistances in the diffusion experiments of chlorferon and diethyl thiophosphate in Ca-alginate gel beads [19] . Effect of the external mass transfer was neglected when the rotating speed was greater than 200 rpm [20] . Using a commercial immobilized lipase, Lipozyme TL IM, to catalyze the hydrolysis of palm olein, the external diffusion was not significant when the stirring speed was higher than 200 rpm [21] . Fig. 1 b shows the dependence of the conversion of ketoprofen ester versus diameter of the immobilized lipase. The conversion decreased rapidly with the increase of the diameters from 0.5 to 1.7 mm, however, the conversion was almost unchanged when the diameters increased from 0.2 to 0.5 mm, which indicated that there was no significant effect of the internal diffusion on the kinetics, and internal mass transfer limitation could be neglected. Therefore, the kinetic experiments were investigated at stirring speed of 200 rpm and diameter of 0.5 mm in our work. Thus, in our kinetic experiments, the substrate and product concentrations in bulk liquid were directly used for the determination of inherent kinetics, because the reaction was considered to be controlled by the inherent kinetics. Intrinsic kinetics was not considered to be employed because immobilization may affect the conformation of the lipase. To reduce internal mass transfer limitation of immobilized enzyme, small particle size and low load of enzyme are often employed. Chen and Tsai [17] conducted that the effect of internal diffusion on the enzymatic reaction was neglected in an enantioselective synthesis of ( S )-ibuprofen ester prodrug by immobilized C. rugosa lipase with particle size <1 mm, pore size 0.05–0.5 μm. In the synthesis of antibiotics catalyzed by immobilized penicillin G acylase, no intra-particle diffusion limitation was affecting the system when enzyme load was kept lower [18] . Using calcium-alginate gel beads with radii of 1.65–1.70 mm, effectiveness factors indicated that internal mass transfer resistance was negligible [19] , therefore, the degradation rates of chlorferon and DETP inside Ca-alginate gel beads were reaction-limited. Using Novozym 435 particles with diameter of 0.6 mm, the Thiele modulus, Φ was calculated to be 0.017, which indicated that internal mass transfer limitation could be neglected [20] . In general, when the rate of reaction is much lower than the rate of diffusion inside the particle, the inherent kinetic parameters may be determined by using measurements with immobilized enzymes. Most frequently, this method has been applied in combination with irreversible Michaelis–Menten models. 4.2 Kinetic experiments and modeling In this kinetic investigation, the reaction rate equations (Eqs. (5) and (6) ) are solved numerically. The kinetic parameters are estimated directly using a solution of ordinary differential equations and minimizing the relative error between experimentally determined and calculated time course of the mass conversion of ( S )- and ( R )-ketoprofen. One of the advantages of using time course instead of initial rate measurements is that there is no need for transformation of the data (e.g., when using Lineweaver–Burk or Eadie–Hofstee plots) and, thus, there is no distortion in error distribution [22] . Reactions catalyzed by lipase are often characterized by a ping-pong bi–bi mechanism, as discussed in Section 3 , the kinetic resolution of ketoprofen ester catalyzed by this lipase may be considered as an irreversible reaction. Based on this assumption, an irreversible ping-pong bi–bi model (Eqs. (5) and (6) ) was selected for the current reaction system. The parameters were estimated by focusing on measurements of time course profiles at different temperatures, 30, 35, 40 and 45 °C, respectively. The resulting model fit is shown in Fig. 2 and the corresponding kinetic parameters for Eqs. (3) and (4) are listed in Table 1 . A good model description is achieved for the conversion of the ( S )- and ( R )-enantiomers at various temperatures (30–40 °C). Overall, the model of both enantiomers conversion describes the experimental data well; only at 45 °C do deviations occur, i.e., simulations are some higher than experiments when reaction time is more than 30 h. The fact that experimental results of both enantiomers are well described by the model during the full time-course over the temperature range studied (30–45 °C), proves the model assumptions mentioned in Section 3 to be reasonable, as their influence on this reaction system is not significant enough to cause lack of model prediction. In general, each parameter in a kinetic model is temperature dependent. While the temperature dependency of rate constants, such as K i may be modeled by Arrhenius equations, Eq. (7) , assuming the active enzyme concentration is constant with temperature, affinity constants may be described by the van’t Hoff equation [23] : (7) K i = A e − E a i / R T ,   ( i = 1,2....6 ) where E ai is activation energy (kJ/mol); A , pre-exponential factor (g/h U); R , universal gas constant (kJ/mol K); and T , temperature (K). The already estimated kinetic parameters at various temperatures (30, 35, 40 and 45 °C) are used for the determination of the relevant temperature parameters including activation energy, E a , and pre-exponential factor, A . Also, no significant increase in relative error was observed when the effect of temperature on Michaelis–Menten constants was neglected. The resulting model fit to experimental data is shown in Fig. 3 in which the line of K 4 is almost superposition with that of K 5 , and the corresponding parameters is shown in Table 2 . The active energy reported in Table 2 is of the same magnitude as those found previously for other reactions catalyzed by lipase [11,24,25] . As shown in Fig. 3 , E a (the slope of the line) was constant over the temperature range studied, as implies that the reaction mechanism may not be changed during the range of temperature. K 4 and K 5 are inhibition constants of ( S )- and ( R )-enantiomer, respectively, shown in Table 1 . K 4 is almost equal to K 5 , and the active energy of K 4 also equates approximately that of K 5 , as implies that the work mechanism of ( S )-enantiomer may be the same as that of ( R )-enantiomer. More competitive inhibition to the hydrolysis reaction, however, may be resulted by ( S )-enantiomer, because the concentration of ( S )-enantiomer is much higher than that of ( R )-enantiomer. The most frequently used parameters to quantify the quality of a kinetic resolution reaction are the enantioselectivity (or enantiomeric ratio), also named E value and enantiomeric excess, ee p , respectively. Figs. 4 and 5 give X , E and ee p profile in the hydrolysis of ( S , R )-ketoprofen ethyl ester. Shown from Fig. 4 a, X increased rapidly with temperatures from 30 to 45 °C, as indicates that the higher temperature may lead a higher conversion. The highest conversion of ( S )-ketoprofen was achieved at 45 °C. About 46.3% conversion was obtained for reaction time of 33 h at the reaction temperature of 45 °C, and after 33 h, X increased little ( X of 47.2% for 40 h). Generally, with increasing reaction temperature, the conversion of ( S )-ketoprofen increases. As shown in Fig. 4 a, only 13.1% conversion was achieved at 30 °C for reaction time of 40 h. However, about 2.6 times increment in X was observed when the reaction temperature increased to 45 °C. The immobilized enzyme was also effective at 45 °C before denaturation occurs. According to Fig. 4 b, E increased some rapidly at 45 °C until 15 h, and then E changed little until 33 h, after 33 h, E decreased due to the reduction of ee p at 45 °C shown from Fig. 5 . The ee p decreased pronouncedly due to the reduction of the reaction rate of ( S )-ketoprofen ester caused by the decrease of the concentration of the ester, in fact, X s is 45% ( Fig. 2 a) and most of ( S )-ketoprofen ester, about 90%, has been converted to ( S )-ketoprofen. In other words, the reason for the decrease of ee p value at 45 °C shown in Fig. 5 is that the conversion at 40 h is the highest, so that the ee p value is the lowest. In general, ee p value decreases with the increase of conversion. At 45 °C for reaction time of 33 h, E and ee p were 87.4 and 94.4%, respectively, which indicates the lipase has high enantioselectivity, and the kinetic resolution is well-suited for industrial purposes. The optimum operation conditions may be 45 °C of temperature and 33–35 h of reaction time. 5 Conclusions An inherent kinetic model of immobilized lipase-catalyzed enantioselective hydrolysis of ketoprofen ester to ( S )-ketoprofen based on irreversible ping-pong bi–bi mechanism was developed, taking into account the competitive product inhibition by both ( S )-and ( R )-enantiomer. The work mechanism of ( S )-enantiomer may be the same as that of ( R )-enantiomer, as the values of inhibition constants and active energy of ( S )-enantiomer are almost equal to those of ( R )-enantiomer. With 200 rpm of stirring speed and 0.5 mm of the diameters of immobilized lipase, internal and external mass transfer limitations could be neglected based on experiments. The solution of ordinary differential equations was successfully combined with optimization to estimate kinetic parameters in the reaction rate equations. The simulated values of the model fitted the experimental values well, and such kinetic models can be used as a basis for process engineering studies and accelerate the development of biocatalyst for industrial applications. With optimum operation conditions, 45 °C of temperature and 33–35 h of reaction time, enantiomeric ratio E and enantiomeric excess of product ee p were 87.4 and 94.4%, respectively, which indicates the lipase has high enantioselectivity and the kinetic resolution is well-suited for industrial purposes. Acknowledgement The authors sincerely thank the Key Subject of Chemical Engineering of Qingdao University of Science and Technology for financial support of the project. References [1] A.L. Ong A.H. Kamaruddin S. Bhatia Current technologies for the production of ( S )-ketoprofen: process perspective Process Biochem. 40 2005 3526 3535 [2] W.W. Xi J.H. Xu Preparation of enantiopure ( S )-ketoprofen by immobilized Candida rugosa lipase in packed bed reactor Process Biochem. 40 2005 2161 2166 [3] G. Duan C.B. Ching E. Lim C.H. Ang Kinetic study of enantioselective esterification of ketoprofen with n-propanol catalysed by a lipase in an organic medium Biotechnol. Lett. 19 1997 1051 1055 [4] J.C. Wu H.R. Low Y.J. Leng Y. Chow R.J. Li M.M.R. Talukder J.C. Choi Ketoprofen resolution by enzymatic esterification and hydrolysis of the ester product Biotechnol. Bioprocess Eng. 11 2006 211 214 [5] E. Henke S. Schuster H. Yang U.T. Bornscheuer Lipase-catalyzed resolution of ibuprofen Monatsh. Chem. 131 2000 633 638 [6] P.O. Carvalho F.J. Contesini R. Bizaco S.A. Calafatti F.A. Macedo Optimization of enantioselective resolution of racemic ibuprofen by native lipase from Aspergillus niger J. Ind. Microbiol. Biotechnol. 33 2006 713 718 [7] J.H. Liu Y.Y. Zhang L.H. Qiu F.K. Yang L. Ye Y.M. Xia Kinetic resolution of ketoprofen ester catalyzed by lipase from a mutant of CBS 5791 J. Ind. Microbiol. Biotechnol. 31 2004 459 499 [8] J.H. Liu Y.Y. Zhang Y.M. Xia F. Su Optimization of immobilization conditions of Candida antarctica lipase based on response surface methodology Chem. Biochem. Eng. Q. 24 2010 203 209 [9] M.V. Flores P.J. Halling Full model for reversible kinetics of lipase-catalyzed sugar-ester synthesis in 2-methyl-2-butanol Biotechnol. Bioeng. 78 2002 794 800 [10] M. Rizzi P. Stylos A. Riek M. Reuss A kinetic study of immobilized lipase catalysing the synthesis of isoamyl acetate by transesterification in n-hexane Enzyme Microb. Technol. 14 1992 709 714 [11] M. Martinelle K. Hult Kinetics of acyl transfer reactions in organic media catalyzed by Candida antarctica lipase B Biochim. Biophys. Acta 1251 1995 191 197 [12] S.H. Kim T.K. Kim G.S. Shin K.W. Lee H.D. Shin Y.H. Lee Enantioselective hydrolysis of insoluble ( R , S )-ketoprofen ethyl ester in dispersed aqueous reaction system induced by chiral cyclodextrin Biotechnol. Lett. 26 2004 965 969 [13] Y.Y. Zhang J.H. Liu L.H. Qiu Y.J. Shi Z.L. Su Y.M. Xia F.M. Li Optimisation of cultivation conditions of a mutant of Trichosporon laibachii CBS 5791 for enantioselective resolution of ketoprofen ester Ann. Microbiol. 55 2005 101 106 [14] Y.Y. Zhang J.H. Liu Purification and in situ immobilization of lipase from of a mutant of Trichosporon laibacchii using aqueous two-phase systems J. Chromatogr. B 878 2010 909 912 [15] J. Uppenberg M.T. Hansen S. Patkar T.A. Jones The sequence, crystal structure determination and refinement of two crystal forms of lipase B from Candida antarctica Structure 2 1994 293 308 [16] P. Eigtved, Immobilization of lipase by adsorption on a macroporous resin, US Patent US 5,156,963, 1992. [17] J.C. Chen S.W. Tsai Enantioselective synthesis of ( S )-ibuprofen ester prodrug in cyclohexane by Candida rugosa lipase immobilized on accurel MP1000 Biotechnol. Prog. 16 2000 986 992 [18] L.R.B. Gonc¸alves A.L.O. Ferreira R. Fernandez-Lafuente J.M. Guisan R.C. Giordano Influence of mass transfer limitations on the enzymatic synthesis of β-lactam antibiotics catalyzed by penicillin G acylase immobilized on glioxil-agarose Bioprocess Biosyst. Eng. 31 2008 411 418 [19] J. Ha C.R. Engler S.J. Lee Determination of diffusion coefficients and diffusion characteristics for chlorferon and diethyl thiophosphate in Ca-alginate gel beads Biotechnol. Bioeng. 100 2008 698 706 [20] S.Z. Wang J.P. Wu G. Xu L.R. Yang Kinetic modelling of lipase-catalyzed remote resolution of citalopram intermediate in solvent-free system Biochem. Eng. J. 45 2009 113 119 [21] Y.H. Chew L.S. Chua K.K. Cheng M.R. Sarmidi R.A. Aziz C.T. Lee Kinetic study on the hydrolysis of palm olein using immobilized lipase Biochem. Eng. J. 39 2008 516 520 [22] J.L.L. Rakels B. Romein A.J.J. Straathof J.J. Heijnen Kinetic analysis of enzymatic chiral resolution by progress curve evaluation Biotechnol. Bioeng. 43 1993 411 422 [23] I.H. Segel Enzyme Kinetics Behavior and Analysis of Rapid Equilibrium and Steady-state Enzyme Systems 1993 John Wiley & Sons New York p. 127 [24] W.R. Berendsen G. Gendrot S. Resnick M. Reuss Kinetic modeling of lipase catalyzed hydrolysis of ( R / S )-1-methoxy-2-propyl-acetate as a model reaction for production of chiral secondary alcohols J. Biotechnol. 121 2006 213 226 [25] W.R. Berendsen A. Lapin M. Reuss Investigations of reaction kinetics for immobilized enzymes identification of parameters in the presence of diffusion limitation Biotechnol. Prog. 22 2006 1305 1312