Kinetic studies of lipase-catalyzed esterification in water-in-oil microemulsions and the catalytic behavior of immobilized lipase in MBGs

Kinetic studies of lipase-catalyzed esterification in water-in-oil microemulsions and the catalytic behavior of immobilized lipase in MBGs

Colloids and Surfaces A: Physicochemical and Engineering Aspects 194 (2001) 41 – 47 Kinetic studies of lipase-cataly...

135KB Sizes 0 Downloads 20 Views

Colloids and Surfaces A: Physicochemical and Engineering Aspects 194 (2001) 41 – 47

Kinetic studies of lipase-catalyzed esterification in water-in-oil microemulsions and the catalytic behavior of immobilized lipase in MBGs Guo-Wei Zhou, Gan-Zuo Li *, Jian Xu, Qiang Sheng Key Laboratory for Colloid and Interface Chemistry of State Education Ministry, Shandong Uni6ersity, Jinan 250100, People’s Republic of China Received 14 November 2000; accepted 11 April 2001

Abstract The esterification kinetics of octanoic acid with 1-octanol, catalyzed by Candida lypolytical (CL) lipase, was studied in water-in-oil microemulsions formed by water/bis-(2-ethylhexyl)sulfosuccinate sodium (AOT)/isooctane. Kinetic studies showed that the reaction follows a Ping-Pong Bi Bi mechanism with inhibition by excess of 1-octanol. The values of all apparent kinetic parameters were determined to be 6max = 4.7× 10 − 3 mmol l − 1 min − 1 mg − 1, Km acid =49.3 mmol l − 1, and Km alcohol =47.6 mmol l − 1, respectively. CL lipase has also been immobilized in gelatin-containing AOT microemulsion-based organogels (MBGs) with retention of catalytic activity. These lipasecontaining MBGs were proved to be a novel solid-phase catalyst for use in apolar organic solvents. The behavior of this novel, predominantly hydrophobic matrix as an esterification catalyst was also examined. © 2001 Elsevier Science B.V. All rights reserved. Keywords: W/O microemulsions; Enzyme catalysis; Immobilized enzyme; Microemulsion-based gels; Kinetics

1. Introduction It is well established that many enzymes can be entrapped in water-in-oil (W/O) microemulsions or reverse micelles, retaining their catalytic activity [1]. Among the enzymes studied to date, lipases are most attractive due to their numerous biotechnological applications in the preparation

* Corresponding author. Tel.: + 86-531-8564750; fax: + 86531-8565167. E-mail address: [email protected] (G.-Z. Li).

of fine chemicals, food and pharmaceutical industry. One of the most intensively studied has been the technique of solubilizing enzymes in W/O microemulsions. A major attraction of this procedure is that the lipase is dispersed at the molecular level, rather than as a solid aggregate, in a thermodynamically stable liquid solution that is capable of solubilizing polar, apolar and interfacially active substrates. The main advantages of this system are the possibility to provide an adequate environment to the enzyme, and therefore protect it against denaturation by the organic solvent. Additionally, mi-

0927-7757/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 7 - 7 7 5 7 ( 0 1 ) 0 0 7 4 5 - 2


G.-W. Zhou et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 194 (2001) 41–47

Fig. 1. Scheme for the interaction of the enzyme and water-insoluble compound in W/O microemulsion: 1, enzyme; 2, surfactant; 3, water-insoluble compound.

croemulsions can provide an extremely large interfacial area between the oil-continuous phase and the dispersed aqueous phase (approximately 100 m2 ml − 1), which is especially important for the reaction catalyzed by lipases [2,3]. The mechanism of the interaction is schematically represented in Fig. 1 [1]: the water-insoluble compound is solubilized by the surface layer of the microemulsions and therefore can come into contact with the entrapped enzyme. Interestingly, many W/O microemulsions can be ‘gelled’ by the addition of gelatin aqueous, yielding a matrix suitable for enzyme immobilization. The preparation of gelatin-containing microemulsion-based organogels (MBGs) was first described in 1986 [4], and their physical–structural characterization has since been the subject of a number of studies [5,6]. It is proposed that the MBGs comprise an extensive, rigid, interconnected network of gelatin/water rods stabilized by a monolayer of surfactant, in coexistence with a population of ‘conventional’ W/O microemulsion droplets. The schematic model is shown in Fig. 2 [7]. The MBGs are stable in contact with apolar solvents, and fully retained the surfactant, gelatin, and water components in most conditions. They could be used as a solid-phase biocatalyst in the presence of organic solvents or liquid substrates. The MBGs offer considerable advantages over W/O microemulsions, such as higher enzyme stabilities, product isolation and facilitated reuse. In the present work, we have investigated the kinetic behavior for the esterification of octanoic

Fig. 2. A proposed model for gelatin-containing MBGs: 1, W/O microemulsion-type droplets; 2, gelatin/water channels.

acid with 1-octanol, catalyzed by Candida lipolytical (CL) lipase, in bis-(2-ethylhexyl) sulfosuccinate sodium (AOT) W/O microemulsions, and determined the apparent kinetic parameters. In addition, the catalytic behavior of the CL lipase immobilized in AOT microemulsion-based gels, for the esterification of octanoic acid with several primary alcohol and 1-octanol with fatty acid, has also been studied.

2. Experimental section

2.1. Materials CL lipase (EC was supplied by Wuxi enzyme factory, its specific activity of 88 U mg − 1 solid was obtained by using titration of free fatty acid release from olive oil; bis-(2-ethylhexyl) sulfosuccinate sodium (AOT) and gelatin (Bloom300) were purchased from Sigma Chemical Company, and isooctane(AR grade) was obtained from Beijing Yili Fine Chemical Company. Threetimes distilled water was used throughout this study; All other chemicals were of AR grade and were stored over a type 4A molecular sieve.

2.2. Preparation of W/O microemulsions and CL lipase-containing MBGs W/O microemulsions were prepared by adding

G.-W. Zhou et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 194 (2001) 41–47

appropriate phosphate buffer (pH 7.17) into 0.1 and 0.38 mol l − 1 AOT/isooctane solution, and the final water content W0([H2O]/[AOT]) was 10 and 17, respectively. Two clear solution microemulsions (A and B) were obtained after a gentle shaking for a few seconds. Lipase-containing MBGs were prepared according to the literature [8]: 0.47 g gelatin was allowed to swell for 1 h in 0.6 g water and then the solution was stirred in a 55°C water bath for 10 min in order to achieve dissolution of the polymer, and 1.71 g pre-incubated microemulsions (B) that contained 100 mg CL lipase were then added. The system was vigorously stirred for 2 min at 55°C, and then allowed to cool in air to room temperature.

2.3. Enzymatic reactions in W/O microemulsions and in MBGs Ten milliliters of microemulsions (A) containing adequate concentration of octanoic acid and 1-octanol were added to a 50 ml capped conical flask in a thermostatic and shaky water bath at 37°C, 150 r.p.m. One milliliter of the reaction mixture was withdrawn at selected time intervals and then added to 2 ml V(ethanol):V(acetone) = 1:1 solutions. The depletion of octanoic acid was assayed by titration with 0.05 mol l − 1 NaOH aqueous according to the literature [9,10]. Curves of depletion in octanoic acid concentration as a function of time were linear in 30 min, thus allowing a reliable determination of the initial slopes by linear-regression calculations. The initial rate of reaction equals the depletion of octanoic acid concentration (mol l − 1 min − 1 mg − 1). A sample without lipase was used as a blank. The enzymatic reaction in MBGs was as follows: the MBGs were sectioned into pieces in order to increase the contact area between the gels and the surrounding reaction medium. The volume of MBGs was 3.33 cm3 and the volume of isooctane containing 0.1 mol l − 1 fatty acids with primary alcohols was 10 ml. All the reactions were then performed in a 50 ml capped conical flask in a thermostatic and shaky water bath at 30°C, 150 r.p.m. Substrate concentrations given in the text refer to the concentrations in the isooc-


tane phase prior to contact with a substrate-free MBGs. When the reaction was completed, MBGs were immersed in 5 ml isooctane twice for 1 h each time. The adsorbed substrates and products were then extracted from MBGs. The measurements are the same as for microemulsions. The conversion percent of fatty acid was defined as the consumption of fatty acid divided by the initial amount of fatty acid.

3. Results and discussion

3.1. Kinetic theoretical model analysis Information regarding the kinetic mechanism for enzymatic two substrate reactions may often be obtained by measuring the initial rates of reaction under conditions in which the concentration of one substrate is held constant and the concentration of the second substrate is varied and vice versa [11]. In the case of lipase-catalyzed esterifications, the mechanism may be either sequential or nonsequential (Ping-Pong). The general rate equations for these two mechanisms are shown as: 60 =

6max[A0][B0] [A0][B0]+ KmB[A0]+ KmA[B0]+ KmBK 0mAB

for sequential 60 =


6max[A0][B0] [A0][B0]+ KmB[A0]+ KmA[B0]

for Ping-Pong


where 60 is the initial rate of reaction, 6max is the maximum rate, [A0] and [B0] are the initial substrate concentrations, KmA and KmB are the Michaelis constants for octanoic acid and 1-octanol, respectively, and K 0mAB is the apparent dissociation constant for ternary complexes (EAB). These kinetic parameters represent apparent ones, since the reaction does not take place in a homogeneous medium. When [B0] is held constant, Eqs. (1) and (2) may be written as Eqs. (3) and (4): 60 =

6%[A0] K%m + [A0]


G.-W. Zhou et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 194 (2001) 41–47


60 =

6%[A0] K¦m +[A0]


where 6%=

6max[B0] , KmB +[B0]

K¦m =

K%m =

KmBK 0mAB +KmA[B0] , KmB +[B0]

KmA[B0] KmB + [B0]

6%, K%m and K¦m are different from 6max and KmA, and they vary with [B0]. When [B0] KmB, 6%, K%m and K¦m approach 6max and KmA, then the two substrate reactions may be transformed into a quasi-single substrate reaction. The double reciprocal form of Eqs. (1) and (2) may be written as Eqs. (5) and (6):


1 K 1 K K0 K = 1+ mA + mB + mAB mB 60 [A0] [B0] [A0][B0] 6max


1 1 K K = 1+ mA + mB 60 [A0] [B0] 6max


By constructing a series of double reciprocal 1 1 plots, i.e. 6 − versus [A0] − 1 or 6 − versus [B0] − 1, 0 0 simple visual inspection may distinguish between the two mechanisms. Thus, sequential kinetics are characterized by a series of converging straight 1 lines meeting at a focal point to the left of the 6 − 0 axis, while true Ping-Pong kinetics give rise to a

Fig. 3. Effect of 1-octanol concentration on the initial rate of esterification determined at fixed concentration of octanoic acid in W/O microemulsions. [Octanoic acid]: ( ) 0.04 mol l − 1, ( ) 0.08 mol l − 1, () 0.10 mol l − 1, () 0.12 mol l − 1, (") 0.20 mol l − 1. [AOT]= 0.1 mol l − 1, [CL lipase] = 10 mg ml − 1, W0 = 10, 37°C.

family of parallel straight lines. The double recip1 rocal plot of 6 − versus [A0] − 1 is a primary plot, 0 the plot of the ordinate intercept of the primary plot versus [B0] − 1 is a secondary plot. The 6max and KmB are calculated from the intercept and slope of the secondary plot: intercept= 1/wmax, slope= KmB/wmax. The same method can be used to obtain KmA.

3.2. Kinetic studies of CL lipase catalysis in AOT W/O microemulsions The kinetics of the esterification were investigated by studying the effect of the concentration of both octanoic acid and 1-octanol on the initial rate of the reaction. The 1-octanol concentration was varied at different fixed concentrations of octanoic acid and vice versa. The concentration of octanoic acid and 1-octanol were varied over ranges of 0.04–0.20 mol l − 1, respectively. Fig. 3 shows the variation of the initial rate of the esterification as a function of 1-octanol concentration at various octanoic acid concentrations in microemulsions. It is observed that high concentration of 1-octanol inhibits the lipase activity. A hypothesis explaining the inhibitory effect of 1-octanol could be the following: 1-octanol reacts with the free lipase, a dead-end complex is formed, and the lipase cannot further participate in the reaction [12]. The Lineweaver–Burk plots of reciprocal initial rate versus the reciprocal 1-octanol concentration at several octanoic acid concentrations are illustrated in Fig. 4. The families of lines are essentially parallel, indicating that a PingPong Bi Bi mechanism operates in this system. Fig. 5 is a secondary plot from the ordinate intercept of Fig. 4 versus the reciprocal octanoic acid concentration. From the slope and the ordinate intercept of Fig. 5, 6max and Km acid are calculated to be 4.7×10 − 3 mmol l − 1 min − 1 mg − 1 and 49.3 mmol l − 1, respectively. By a similar analysis, Km alcohol is calculated to be 47.6 mmol l − 1. The reaction scheme for the lipase-catalyzed esterification of octanoic acid and 1-octanol in AOT W/O microemulsions is shown in Fig. 6. According to this mechanism, the lipase initially reacts with octanoic acid to form the lipase–oc-

G.-W. Zhou et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 194 (2001) 41–47


lipase. Thus, it is possible that some of the free lipase could interact with 1-octanol to form the dead-end complex. The modified lipase now reacts with 1-octanol to form the modified lipase–octanol complex, which is also isomerized by a unimolecular reaction to a lipase–octyl octoate complex, which then yields the product octyl octoate and the free lipase.

3.3. Operational and storage stabilities of lipase-containing MBGs

Fig. 4. Double reciprocal plot of the initial rate of esterification versus 1-octanol concentration. [Octanoic acid]: ( ) 0.04 mol l − 1, ( ) 0.08 mol l − 1, () 0.10 mol l − 1, () 0.12 mol l − 1, (") 0.20 mol l − 1. [AOT]= 0.1 mol l − 1, [CL lipase] =10 mg ml − 1, W0 = 10, 37°C.

The MBGs do not dissolve in their parent oils, nor is there any appreciable loss of surfactant, water or enzyme in the substrate-containing organic solvent with which they are in contact. Probably, the ester first accumulates in the hydrophilic gel rather than in the hydrophobic solvent. But with the synthesis proceeding, the ester could enter the oil phase, which depends on the distribution effect and diffusion effect of the reactants and products between the oil phase and MBGs phase. That is to say, the polarity of the participants is a determining factor. So, on completion of each synthesis, the external oil phase is removed, and the MBGs purged of residual substrates/products by incubation at room temperature with fresh aliquots of isooctane (2× 2.5 ml) over a period of about 2 h, and then are replaced with fresh isooctane prior to repeating the measurement with a fresh substrate solution. The operational and storage(non-operational) stabilities of an immobilized lipase are important parameters from which one can determine the

Fig. 5. Secondary plot of Fig. 4 intercepts versus reciprocal octanoic acid concentration.

tanoic acid complex, which is subsequently transformed by a unimolecular isomerization reaction into the acyl– lipase intermediate with the concomitant release of water. The lipase may also combine with 1-octanol to form the lipase–octanol dead-end complex. The reason for the formation of a dead-end complex is that the modified lipase, which is the acyl– lipase complex formed after the release of water in the first step of the reaction, may be structurally similar to the free

Fig. 6. Presentation of a Ping-Pong Bi Bi reaction in W/O microemulsions. A and B, The substrates octanoic acid and 1-octanol; W and P, the products, water and octyl octoate; L and ML, the free lipase and modified lipase; L-A and L-B, the lipase – octanoic acid complex and lipase – octanol dead-end complex; ML-W and ML-B, the modified lipase– water complex and modified lipase– octanol complex; L-P, the lipase – octyl octoate complex.

G.-W. Zhou et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 194 (2001) 41–47


Table 1 Effect of run numbers on the conversion of octanoic acid and 1-octanol (0.1 mol l−1 each) Run number











Conversion (%)











Incubation conditions: 3.33 cm3 MBGs and 10 ml isooctane solution containing [octanoic acid] = [1-octanol]=0.1 mol l−1, 30°C, 24 h.

Table 2 Effect of storage duration on the conversion of octanoic acid and 1-octanol (0.1 mol l−1 each) Storage duration (months)







Conversion (%)







Reaction conditions are the same as in Table 1.

Table 3 Effect of carbon numbers of fatty acid on the conversion of different fatty acid with 1-octanol (0.1 mol l−1 each) Fatty acid

Butanoic acid

Pentanoic acid

Hexanoic acid

Heptanoic acid

Octanoic acid

Nonanoic acid

Lauric acid









Reaction conditions are the same as in Table 1.

economic viability of a biosynthetic process. Octanoic acid and octanol were used as the substrate (0.1mol l − 1 each), isooctane as solvent. The MBGs were allowed to proceed for ten runs performed over a 20-day period. The conversion percent at the end run is hold 90% of conversion percent at the initial run. Data from this study are shown in Table 1. Nascimento and co-workers [13,14] have also concluded that the Chromobacterium 6iscosum (CV) lipase immobilized in MBGs retained its activity after several runs, e.g. average yields of 75% could be obtained with the same MBGs through 15 conversions in 30 days. The decreases of the conversion after many runs are due to the turnover of large quantities of substrate that results in the production of substantial quantities of water (as coproduct). The water is almost retained in the microemulsions of the MBG matrix. The accumulation of water in the MBGs is also apparent from visual inspection of the MBG sections, which undergo progressive swelling with success.

The storage stability data for CL lipase immobilized in MBGs are presented in Table 2. The immobilized lipase exhibits good stability with no significant decreasing of conversion for storage periods of 10 months.

3.4. Effect of chain length of fatty acids and alcohols on the con6ersion in MBGs The role of the chain length of fatty acids in esterification of 1-octanol catalyzed by CL lipasecontaining MBGs in isooctane was studied using butanoic, pentanoic, hexanoic, heptanoic, octanoic, nonanoic and lauric acids. The role of the chain length of alcohol in esterification of octanoic acid was studied using 1-butanol, 1-pentanol, 1-hexanol, 1-heptanol and 1-octanol. Tables 3 and 4 present the conversion percent of these esterifications. It can be seen that the shorter chain length (nB 7) of acids shows lower conversion for the esterification with 1-octanol, but the chain length

G.-W. Zhou et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 194 (2001) 41–47


Table 4 Effect of carbon numbers primary alcohol on the conversion of octanoic acid with different primary alcohol (0.1 mol l−1 each) Alcohol






Conversion (%)






Reaction conditions are the same as in Table 1.

(n=4–8) of the primary alcohols has little effect on the conversion for the esterification with oc-tanoic acid. Our result is in a good agreement with the conclusions of Robinson and Backlund and co-workers [13,15,16]. Robinson and co-workers [13] have shown that CV lipase immobilized in AOT MBGs successfully catalyses esterification of medium- and long-chain fatty acids with short-, medium- and long-chain fatty alcohols. Backlund et al. [15,16] have also shown that the reaction yields of racemic 2-octanol with different fatty acids at 298 K in hexane using CV lipase immobilized in AOT MBGs is slower for the shorter acid (heptanoic acid) than for the longer ones (undecanoic acid). But they did not explain the mechanism of the reactions of the different chain length of fatty acids and alcohols in organic solvent using immobilized MBGs. The probable explanation is the following. The conversion of the esterification of fatty acids with alcohols catalyzed by lipase in microemulsions varies with the type and structure of substrates [17]. The dissociative degree of fatty acid enlarges with the shortening of the carbon chain length, so the ionic strength in the microemulsion ‘water pool’ enlarges. That may affect the activity conformation of the enzyme. On the other hand, the carboxylate radical of the fatty acid dissociation in the interface layer of microemulsions cannot react with the alcohol. The hydroxy of the alcohol does not dissociate, the ionic strength of the microemulsion ‘water pool’ and the state of the alcohol do not change, so the carbon chain length of alcohol has no effect on the activity of lipase. Similar results have also been reported for CV lipase immobilized in MBGs catalyzing the esterification of oleic acid with various alcohols [14].

Acknowledgements Support of this work by the State Natural Science Fund of China is gratefully acknowledged (Grant No. 29903006).

References [1] K. Martinek, A.V. Levashov, N.L. Klyachko, Y.L. Khmelnitsky, I.V. Berezin, Eur. J. Biochem. 155 (1986) 453. [2] N. Krieger, M.A. Taipa, E.H.M. Melo, J.L. LimaFilho, M.R. Aires-Barros, J.M.S. Cabral, Appl. Biochem. Biotechnol. 67 (1997) 87. [3] E.P. Melo, M.R. Aires-Barros, J.M.S. Cabral, Appl. Biochem. Biotechnol. 50 (1995) 45. [4] G. Haering, P.L. Luisi, J. Phys. Chem. 90 (1986) 5892. [5] P.L. Luisi, R. Scartazzini, G. Haering, P. Schurtenberger, Collid Polym. Sci. 268 (1990) 356. [6] C. Quellet, H.F. Eicke, W. Sager, J. Phys. Chem. 95 (1991) 5642. [7] P.J. Atkinson, B.H. Robinson, A.M. Howe, R.K. Heenan, J. Chem. Soc. Faraday Trans. 87 (1991) 3389. [8] G.D. Rees, T.R.J. Jenta, M.G. Nascimento, M. Catauro, B.H. Robison, G.R. Stephenson, R.D.G. Olphert, Indian J. Chem. 32B (1993) 30. [9] C.P. Singh, P. Skagerlind, K. Holmberg, D.O. Shah, J. Am. Oil Chem. Soc. 71 (1994) 1405. [10] P. Skagerlind, M. Jansson, K. Hult, J. Chem. Tech. Biotechnol. 54 (1992) 277. [11] K. Nagayama, Y. Yamaoka, T. Doi, M. Imai, M. Shimizu, J. Chem. Eng. Jpn. 29 (1996) 185. [12] D.G. Hayes, E. Gulari, Biotechnol. Bioeng. 38 (1991) 507. [13] G.D. Rees, M.G. Nascimento, T.R.J. Jenta, B.H. Robinson, Biochim. Biophys. Acta 1073 (1991) 493. [14] M.G. Nascimento, M.C. Rezende, R.D. Vecchia, P.C. Jesus, L.M.Z. Aguiar, Tetrahedron Lett. 33 (1992) 5891. [15] S. Backlund, F. Eriksson, L.T. Kanerva, M. Rantala, Collids Surfaces B: Biointerfaces 4 (1995) 121. [16] S. Backlund, F. Eriksson, G. Hedstrom, A. Laine, M. Rantala, Colloid Polym.Sci. 274 (1996) 540. [17] H. Stamatis, A. Xenakis, M. Provelegiou, F.N. Kolisis, Biotechnol. Bioeng. 42 (1993) 103.