- Email: [email protected]
Lipase-catalyzed synthesis of sorbitol–fatty acid esters at extremely high substrate concentrations
Journal of Biotechnology 123 (2006) 174–184
Lipase-catalyzed synthesis of sorbitol–fatty acid esters at extremely high substrate concentrations Hyun Jung Kim, Sung Hun Youn, Chul Soo Shin ∗ Department of Biotechnology, College of Engineering, Yonsei University, 134 Shinchon-dong, Seodaemun-gu, Seoul 120-749, South Korea Received 29 April 2005; received in revised form 4 November 2005; accepted 9 November 2005
Abstract Lipase-catalyzed synthesis of sorbitol–fatty acid esters was performed in eutectic media with extremely high substrate concentrations. Homogeneous eutectic melts of sorbitol and fatty acids of C6–C16 were prepared using an adjuvant mixture. Enhanced homogeneity of mixtures was confirmed by X-ray diffraction analysis. The substrate concentration was 3.63–6.67 M in the eutectic media, whereas in organic media the concentration was below 0.10 M. Esters were synthesized with an immobilized Candida antarctica lipase, and optimum conditions were analyzed. Compared to reactions in organic media, the initial reaction rate of ester synthesis and the overall productivity were significantly enhanced in eutectic media while the conversion yields were similar. Based on the kinetic analysis, highly viscous eutectic media were shown to influence the initial reaction rate and the apparent activation energy resulting in diffusion limitations. © 2005 Elsevier B.V. All rights reserved. Keywords: Eutectic media; High concentration; Enhanced homogeneity; Diffusion limitation
1. Introduction The enzyme-catalyzed synthesis of biosurfactants has been reported during the past decade (Fiechter, 1992; Sarney and Vulfson, 1995). Among surfactants analyzed, the sugar–fatty acid esters are potentially useful in the pharmaceutical, cosmetic, and food industries as a non-ionic surfactant (Fregapane et al., 1994; Hill ∗ Corresponding author. Tel.: +82 2 2123 2886; fax: +82 2 362 7265. E-mail address: [email protected] (C.S. Shin).
0168-1656/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jbiotec.2005.11.006
and Rhode, 1999; Hills, 2003). Because of hydrophobic and hydrophilic natures of substrates, heterogeneous immiscible mixtures were induced, which hindered enzymatic reactions. However, lipase-catalyzed synthesis of these esters has been successfully carried out in non-aqueous organic media using hydrophilic solvents such as acetone (Arcos et al., 1998a,b,c, 2001), 2-methyl 2-butanol (Flores and Halling, 2002; Salis et al., 2004), and tert-butanol (Cao et al., 1999; Degn et al., 1999; Moreau et al., 2004). Unfortunately, substrate concentrations in organic media were extremely low at several different mM levels, suggesting non-
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
applicability for industrial uses even though high conversion yields of almost 100% can be attained. Among non-conventional media that are used for biocatalysis, organic solvent media and solid–liquid two-phase systems have been widely used in preparing substrate mixtures (Vermu¨e and Tramper, 1995)–other non-conventional media such as supercritical fluids, ionic liquids, and reverse micelles were not dealt here. Organic solvent media formed a homogeneous liquid phase including a small quantity of substrate (Fig. 1A) owing to a low solubility (Arcos et al., 1998c; Plou et al., 2002). Bioreactions in organic solvent media have been widely used for many applications over the past two decades (Zaks and Klibanov, 1985; Dordick, 1988; Klibanov, 2001). The reaction yields in the organic media have been improved by enhancement of enzyme activity via immobilization, genetic or protein engineering, extractive biocatalysis via integration of reaction and separation, and novel enzyme screening (Vermu¨e and Tramper, 1995; Paiva and Malcata, 1997; Tsuzuki et al., 1999; Bornscheuer et al., 2002; Krishna, 2002; Reetz, 2002). However, concentrations are still extremely low, resulting in limited industrial applications. The ‘solid phase media’ or ‘solid-to-solid system’ (Erbeldinger et al., 1998), which is a type of solid–liquid two-phase system, has a feature of liquid phase similar to that of organic solvent media (Fig. 1B). The solid phase media comprise a small volume of liquid phase and mostly solid substrate. However, since the substrate molecules existed in liquid phase can be only engaged in enzymatic reactions, reactants in liquid phase cannot greatly contribute to enhancement of overall productivity. According to a report of Cao et al. (1997), the initial reaction rate and overall productivity of reaction in solid phase media were enhanced approximately 10 times compared to those in organic solvent media. However, the substrate concentration in the liquid phase was only approximately 0.75 M (Cao et al., 1999) or lower (Yan et al., 1999). Reactions do not always proceed well in the solid-to-solid system, whose switch-like behavior was carefully investigated in several references (Ulijn et al., 2001, 2003; Ulijn and Halling, 2004). Eutectic mixture formation is a well-known melting technique that lowers the melting point of a mixture below the melting point of each pure compound in the mixture (Gill and Vulfson, 1994). An adjuvant usu-
175
Fig. 1. Schemes of substrate mixtures in organic solvent media (A), solid phase media (B), and eutectic media (C); S: substrate; P: product; E: enzyme.
ally helps to decrease the melting point and the molten mixture can be maintained at room temperature or below (L´opez-Fandi˜no et al., 1994a). The eutectic mixture, which exists as a state of homogeneous liquid
176
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
solution, consists of mostly substrate molecules and a small quantity (5–30%, gram-solvent per gram-total substrate mixtures) of organic solvent (Fig. 1C). As the molar ratio of a binary mixture is changed, the lowest melting point can be found at a specific mole fraction. This point is called as the ‘eutectic point’ and the corresponding mole fraction is as the ‘eutectic composition’. Often, the term ‘heterogeneous eutectic mixture’ was used (Gill and Vulfson, 1994; Erbeldinger et al., 1998). Here, we used the term ‘eutectic media’ for eutectic substrate mixtures comparative to organic media. Eutectic media exhibit a transparent and stable liquid phase with extremely high substrate concentrations, corresponding to approximately 100 times higher than the organic media. Since small amounts of organic solvents are used in eutectic media, they can be considered to be more biocompatible for enzymatic reactions than with organic phase media. There are a few reports (Gill and Vulfson, 1993; L´opez-Fandi˜no et al., 1994b; Gill et al., 1996) that extremely high concentrations of eutectic substrate mixtures were prepared in biocatalytic reactions. We have also reported results for enzymic dipeptide synthesis using eutectic media (Kim and Shin, 2001; Ahn et al., 2001; Kim et al., 2001; Shin et al., 2003). Based on the computer-aided molecular dynamics simulations, we confirmed that a eutectic mixture can be formed by the enhancement of overall molecular kinetic energy (Kim et al., 2005). However, this eutectic system has not been thoroughly investigated for industrial application to enzymatic reactions. In this study, we determined biosynthesis of sorbitol–fatty esters as a model system mainly performed in organic solvent media, and tried to enhance the substrate concentration levels via eutectic media engineering. The initial reaction rate, conversion yield, and overall productivity in eutectic media were compared with those in organic solvent media. Based on kinetic analysis, the effect of viscosity on the diffusion limitation in eutectic media was investigated.
2. Materials and methods 2.1. Enzyme and chemicals The Candida antarctica lipase (nonspecific triacylglycerol lipase, EC 3.1.1.3) immobilized on acrylic resin was obtained from Sigma Chemical Co. (a prod-
uct of Novozyme Corp.). All fatty acids (caproic, caprylic, capric, lauric, myristic, and palmitic) and dimethyl sulfoxide (DMSO) were also obtained from Sigma Chemical Co. Sorbitol and sodium hydroxide were purchased from Duksan. 2-Methyl-2-methanol (2M2B) was a product of Fluka. Methanol and acetonitrile (HPLC grade) were products of T.J. Baker. Molecular sieves (potassium, sodium alumino-silicate, ˚ were 8–12 mesh beads) with a nominal pore size of 3 A obtained from Sigma Chemical Co. (USA). 2.2. Synthesis of sorbitol–fatty esters in eutectic media For preparation of eutectic media, sorbitol (2 mmol) and fatty acids (2 mmol) containing an adjuvant were placed into screw-top glass vials and incubated for 30 min in a water bath (Jeio Tech, WB-10M) at 60 ◦ C. NaOH (3 M NaOH solution, 50 l), NaOH/DMSO (50/100 l), and NaOH/DMSO/2M2B (50/100/50 l) were used as adjuvants, respectively. In order to obtain a homogeneous solution, mixtures were cooled and reheated three times. Acylation in eutectic media was performed for 24 h with immobilized C. antarctica lipase in a water bath (Jeio Tech, SWB-03) at 100 rpm and 60 ◦ C. To eliminate the water generated in reaction mixtures, 300 mg of the activated molecular sieves ˚ incubated in a dry oven for 48 h were (pore size: 3 A) added. Temperatures were adjusted for Arrhenius plots from 50 to 80 ◦ C for substrate mixtures containing each adjuvant. All experiments were repeated three times, and the averaged values were used. Samples were intermittently withdrawn with a glass spatula and subjected to HPLC analysis. Water contents of reaction mixtures were measured with the moisture test kit (Riedel-de Ha¨en® , 37858 Hydranal® -Moisture Testkit). 2.3. Synthesis of sorbitol–fatty esters in organic media For preparation in organic media, sorbitol (0.5 mmol) and fatty acids (0.5 mmol) were prepared in a 15 ml screw-top glass vial containing 10 ml of an equimolar solvent mixture of 2M2B and acetonitrile (ACN). Preliminary drying of solvents was done with activated molecular sieves. Reactions were performed in a water bath (Jeio Tech, SWB-03) at 100 rpm and 60 ◦ C for 48 h with 300 mg activated molecular sieves.
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
Aliquots of 10 l were intermittently withdrawn for HPLC analysis. 2.4. Analysis of reaction mixtures A reaction mixture was quantitatively analyzed by HPLC. Samples intermittently obtained from reaction mixtures were diluted approximately 1/100 with the eluent, and immobilized enzymes were immediately removed by filtration (Whatman, PTFE syringe filter, 13 mm, 0.45 m). An HPLC system was used to analyze the product concentration. It was equipped with elution pump (Young-in, Model 910), a column (Thermo electron corporation, Hypersil GOLD column, 250 mm × 4.6 mm), an RI detector (Waters 410, Differential Refractometer), and analytical computer software (Yullin Technology, MultiChroTM ). The mobile phase consisting of a mixture of methanol and acetonitrile (55:45) was passed through the column at a flow rate of 0.5 ml min−1 and 30 l of the diluted reaction mixture was analyzed. The conversion yield was calculated from the amount of fatty acid consumed after reaction. Purified products were obtained by extraction with warm acetone (50 ◦ C) and filtration (Millipore, FluoroporeTM membrane filter, 0.45 m) of chilled extracts at −20 ◦ C (Yan et al., 1999). Dried white crystals were isolated with glass-based preparative TLC (Merck, 25 TLC plates 20 cm × 20 cm Silica gel 60 F254 ), and used to prepare a calibration curve. Identification of esters was performed via LC/MS, and 1 H and 13 C NMR. 2.5. Measurement of viscosity A 50 ml glass vial containing an equimolar mixture of sorbitol (30 mmol) and lauric acid (30 mmol) was incubated in a water bath at 60 ◦ C under various adjuvant conditions. After liquid melts were attained, shear stresses relating to the shear rates of each mixture were measured from 50 to 85 ◦ C using a rheometer (Brookfield, DV-III equipped with a No. 31 spindle). Based on plots of shear stresses versus rates, apparent viscosities from the slopes were calculated using the least square method. 2.6. X-ray diffraction analysis Equimolar mixtures of sorbitol (2 mmol) and lauric acid (2 mmol) were prepared, some containing the
177
NaOH adjuvant (50 l) and others containing an adjuvant mixture of NaOH/DMSO/2M2B (50/100/50 l). After the mixtures before and after eutectic melting treatment were re-solidified at room temperature for 48 h, X-ray diffraction (XRD; Rigaku, D/max; Rint 2000) analysis was performed at 30 kV and 25 mA with a scan speed of 4◦ min−1 . Data were plotted on intensities versus a 2θ diffraction angle.
3. Results 3.1. Preparation of eutectic mixtures We used the eutectic formation technique for enhancement of substrate concentrations. Fatty acids with different carbon numbers of C6–C16 were used as an acyl donor, and sorbitol was used as an acyl acceptor (=an alcohol donor). Based on preliminary analyses, NaOH, DMSO, and 2M2B were used as adjuvants in preparing eutectic media. Substrate mixtures of sorbitol and fatty acid were prepared with extremely high concentrations from 3.63 to 6.67 M (Table 1), which varied with the adjuvant contents. We adopted NaOH (3M) as an adjuvant, which allowed adjustment of the hydrogen ion concentration of biocatalysis. However, a heterogeneous emulsion mixture was formed when the NaOH was used alone. When DMSO and 2M2B were added to substrate mixtures, a homogeneous substrate solution was obtained. In contrast to eutectic media, organic media showed a substrate concentration below 0.1 M, which was approximately 1/40–1/50 of the concentrations in eutectic media. The morphology of a sorbitol–lauric acid substrate mixture was investigated via X-ray diffraction (Fig. 2). The substrate mixtures containing two different types of adjuvants (NaOH and NaOH/DMSO/2M2B) were melted. Both eutectic and non-eutectic mixtures were re-solidified then analyzed. Prior to eutectic melting a noisy base line and high sharp peaks appeared irregularly (Fig. 2, NaOH, Before). After eutectic melting, however, the peaks were shortened and the noisy small peaks disappeared (Fig. 2, NaOH, After). Prior to eutectic melting, when the adjuvant mixture NaOH/DMSO/2M2B was added, sharp and high peaks disappeared. The degree of noises on the base line was almost same as a case with the adjuvant NaOH (Fig. 2, NaOH/DMSO/2M2B, Before). On the contrary, after eutectic melting the
178
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
Table 1 Substrate concentrations in organic and eutectic media Substrate mixtures
Concentrations (M) Organic media
Sorbitol + caproic acid (C6) Sorbitol + caprylic acid (C8) Sorbitol + capric acid (C10) Sorbitol + lauric acid (C12) Sorbitol + myristic acid (C14) Sorbitol + palmitic acid (C16)
<0.10 <0.10 <0.10 <0.10 <0.10 <0.10
Eutectic media Adjuvant A
Adjuvant B
Adjuvant C
6.67 5.99 5.40 4.94 4.51 4.20
5.71 5.22 4.76 4.40 4.05 3.79
5.33 4.90 4.49 4.17 3.86 3.63
An equimolar mixture of 2M2B and ACN of 10 ml was used for a solvent in organic media. Adjuvant A: NaOH (3 M, 50 l); Adjuvant B: mixture of NaOH (3 M, 50 l) and DMSO (100 l); Adjuvant C: mixture of NaOH (3 M, 50 l), DMSO (100 l), and 2M2B (50 l).
intermittent sharp peaks disappeared, and a relatively stable base line with small noises were obtained (Fig. 2, NaOH/DMSO/2M2B, After). These results indicate that more homogeneous property was achieved in eutectic media than in mixture before eutectic melting. 3.2. Acylations in eutectic media
Fig. 2. X-ray diffraction patterns of sorbitol–lauric acid mixtures with adjuvants of NaOH and NaOH/DMSO/2M2B. All samples were analyzed after being re-solidified at room temperature for 48 h.
Lipase-catalyzed acylation of sorbitol was performed with the medium chain fatty acids capric, lauric, and myristic with an NaOH adjuvant (Fig. 3). After immobilized C. antarctica lipase was added to each substrate mixture, capric, lauric and myristic acids were rapidly converted with maximum rates of 6.0, 4.1, and 2.7 mol l−1 h−1 , respectively. Mono-esters were mainly produced and di-esters followed with a concentration of approximately 1/3–1/6 of the mono-ester. The maximum conversion yields for mono- and diesters were 56.5% and 22.5% in a sorbitol–capric acid mixture, 51.3% and 13.0% in a sorbitol–lauric acid mixture, and 43.7% and 10.4% in a sorbitol–myristic acid mixture, respectively. In the next, a mixture of sorbitol and lauric acid was used for reaction. The initial reaction rate and conversion yield of mono- and di-esters increased with an increasing amount of immobilized lipase up to 100 mg. Above that level, the initial reaction rate gradually increased whereas the conversion yield was kept at constant (Fig. 4A). Activated molecular sieves were added for trapping of the water generated during esterification. When the molecular sieves were added, the initial reaction rates for mono- and di-esters were 3.3 and 0.9 mol l−1 h−1 , and the maximum conversion yields were approximately 70% and 12%, respectively
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
179
Fig. 4. Initial reaction rates (open marks) and maximum conversion yields (closed marks) of ester synthesis between sorbitol and lauric acid at various contents of immobilized lipase (A) and molecular sieves (lipase contents: 100 mg) (B). Mono-esters: circle; di-esters: triangle.
Fig. 3. Reaction patterns of sorbitol–fatty acid esters. The immobilized lipase and activated molecular sieves were added at 100 and 300 mg, respectively; () sorbitol; (䊉) mono-ester; () di-ester.
(Fig. 4B). As a result, we confirmed that the critical amount of the enzyme and the molecular sieves were 100 and 300 mg for mixtures of sorbitol/fatty acid (4–5 M), respectively. Based on optimized reaction conditions, ester synthesis was performed with sorbitol and lauric acid. The mono-ester was rapidly produced up to 2.9 M while the di-ester was produced up to 0.4 M (Fig. 5). When the product concentration did not change further, it was assumed that the equilibrium was reached. After equilibrium of mono- and di-esters was reached in 12 h, hydrolysis due to water generation occurred. The concentration of esters was slightly decreased while the
lauric acid concentration was slightly increased in the reaction mixture. After the molecular sieves (300 mg) were added at 12 h, the concentrations of lauric acid, mono-ester, and di-ester were significantly changed. The water content was also rapidly decreased. Equilibrium of the reaction mixtures was disrupted by addition of the molecular sieves at 12 h, causing movement toward ester synthesis for approximately 6 h. The equilibrium concentrations of mono- and di-esters were up to 3.6 and 0.6 M, respectively. Because of a limit in the water-removal capacity, a new equilibrium was established at 19 h. 3.3. Kinetic analysis Kinetic parameters were evaluated from Arrhenius plots with an initial reaction rate of lauric acid (Fig. 6A) in the presence of the adjuvants NaOH, NaOH/DMSO or NaOH/DMSO/2M2B. Linear relationships were achieved within a region in which the
180
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
Fig. 5. Optimized synthesis of sorbitol–laurate ester via immobilized C. antarctica lipase (100 mg). The arrow indicates an additional feeding of activated molecular sieves; () sorbitol; (䊉) mono-ester; () di-ester; (×) water content.
regression coefficient (R2 ) was 0.98 or higher. Apparapp ent activation energies (Ea ) were calculated based on adjuvant type. When the NaOH adjuvant was used alone, the initial reaction rates were not affected by temperature differences in the low temperature region of 50–70 ◦ C and the Arrhenius plots showed a gentle slope app (Ea value: 7.0 kJ mol−1 ). On the contrary, a sharp app slope (Ea value: 21.5 kJ mol−1 ) was observed above 70 ◦ C (dashed line). For the reactions using adjuvant mixtures of NaOH/DMSO and NaOH/DMSO/2M2B, both showed similar patterns in the changes of logarithmic rate versus reverse temperature. At 50–60 ◦ C, the app reaction rates were suddenly elevated and high Ea values of 20.8 and 21.8 kJ mol−1 were obtained for the NaOH/DMSO and NaOH/DMSO/2M2B, respectively. When the temperature was increased above 60 ◦ C (dashed line), the slopes of each reaction mixapp ture plot were diminished and reduced Ea values of −1 12.9 and 15.5 kJ mol were observed. In order to elucidate the relationship between reaction kinetics and medium properties the viscosities of each reaction mixture were analyzed (Fig. 6B). When the NaOH was used alone as an adjuvant, a sticky solu-
Fig. 6. Arrhenius plots of sorbitol–laurate ester synthesis under various adjuvant conditions (A), the viscosities on temperature (B), and the initial reaction rates (open marks) and conversion yields (closed marks) on viscosity (C). Circle: NaOH only; triangle: NaOH/DMSO; quadrangle: NaOH/DMSO/2M2B.
tion with a high viscosity was developed at less than 70 ◦ C (dashed line). While the viscosity showed a big decrease from 2.18 to 0.71 Pa s in the low temperature region of 55–70 ◦ C, there was only a small decrease from 0.71 to 0.31 Pa s in the high temperature region of 70–85 ◦ C. In contrast to the viscosity of NaOH-
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
181
Table 2 Conversion yields and overall productivities for synthesis of sorbitol–fatty acid esters Substrate mixtures
Sorbitol + capric acid Sorbitol + lauric acid Sorbitol + myristic acid
Organic media
Eutectic media
Conversion yielda (%)
Overall productivityb (mmol g−1 h−1 )
Conversion yielda (%)
Overall productivityb (mmol g−1 h−1 )
94 89 83
0.098 0.093 0.086
98 97 94
3.27 3.23 3.13
a
Maximum conversion yields. Overall productivities were calculated from overall reaction rates when it reached a maximum ester production in organic (48 h) and eutectic media (6 h), respectively. b
adjuvant mixtures, the viscosities of the mixtures including NaOH/DMSO and NaOH/DMSO/2M2B as adjuvant mixtures were significantly reduced. Viscosity values decreased from 0.24 Pa s until the temperature reached 60 ◦ C, but slightly decreased above 60 ◦ C (dashed line). The relationships between viscosity, initial reaction rate, and conversion yield are depicted in Fig. 6C. The open and closed marks indicate initial reaction rates and maximum conversion yields of lauric acid, respectively, in all adjuvant types. Initial reaction rates (open) were significantly increased with a decrease in viscosity below 0.5 Pa s, but the conversion yield (closed) was almost the same over this viscosity range. In the high viscosity region, the conversion yield was diminished by approximately 20%. 3.4. Acylations in organic media Lipase-catalyzed acylations of sorbitol with medium chain fatty acids were performed in organic media. Maximum conversion yields in organic media were similar or a little lower than in the eutectic media (Table 2). On the other hand, the overall productivity in each medium was significantly different. The overall productivity in the organic media was observed at 0.086–0.098 mmol product per gram of immobilized lipase per hour (mmol g−1 h−1 ) at an equilibrium at 48 h, whereas 3.13–3.27 mmol g−1 h−1 was observed in the eutectic media at an equilibrium at 6 h.
4. Discussion XRD peaks generally indicate the degree of crystalline property and provide structural information
about materials. Based on XRD results in Fig. 2, the homogeneity of mixtures is quite different under adjuvant conditions. In addition, thermal treatment is essential for eutectic formation, and this thermal energy may promote molecular homogeneity. Enhanced homogeneity is considered to directly stimulate the production of mono-esters, as suggested by Ducret et al. (1995). Because of the low solubility of sorbitol in fatty acids, both reactants tend to exist separately, which leads to poor mono-ester production. Enhanced homogeneity of reactants made by addition of various solvents leads to increased mono-ester production. We found that mono-laurate synthesis was increased (compared to Fig. 3, middle) because of the enhanced homogeneity of reaction mixture under optimized adjuvant conditions regardless of mixing process (Fig. 5). Acylation reactions between sugars and fatty acids/fatty acid esters are condensation reactions that produce water or alcohol as a by-product. Esterification produces water that can induce hydrolysis, which generally reduces the ester production yield. Thus, removal of water is important (Yan et al., 2002). We ˚ for water used molecular sieves with a pore size of 3 A elimination. This method is effective, convenient and useful for removal of the water generated in a smallscale experiment, but is not suitable for biocatalysis on a preparative scale. We found linear relationships between logarithmic enzyme concentrations versus logarithmic initial rates (data not shown). This fact suggests that reactions would not occur under substrate-limiting conditions (Bartling et al., 2001). In reactions of eutectic media, the substrate concentration, the initial reaction rate, and the overall productivity were dramatically enhanced with an accompanying similar conversion yield, compared to the organic media. On the contrary, apparent
16.4 43.6 47.5 55.8 15.7 <50.4 7.0–21.8 All reactions were performed with immobilized C. antarctica lipase; n.d.: not described. a Esterification. b Transesterification.
10–50 35–55 30–70 30–70 30–60 30–60 50–80 0.1 0.004 0.004 0.004 0.002 n.d. 4.2–4.9 n-Hexane Hexane/2-butanone Heptane Heptane Toluene Acetone NaOH/DMSO/2M2B Geraniol Oleyl alcohol Tetrahydrofurfuryl alcohol Tetrahydrofurfuryl alcohol n-Butanol Glucose Sorbitol
Reference (kJ mol−1 ) app
Ea
Temperature range (◦ C) Concentration (M) Solvent Acyl acceptor Acyl donor
Table 3 Comparison of the activation energies for various lipase-catalyzed reactions
Acetic Cinnamic acida Butyric acida Ethyl butyrateb Methyl acetoacetateb Mono-laurateb Lauric acid
app
activation energy (Ea ) values were relatively low compared to similar reactions (Table 3). Regardless app of lipase source, Ea values were attained in a broad range of 15–55 kJ mol−1 in both esterifications and app transesterifications. This result indicates that the Ea value is not always directly dependent on the biocatalyst type, and other factors in the medium such as concentration, density, and viscosity can also affect the app Ea value. app In our investigations, we found that the Ea value was closely related to the viscosity in the eutectic media (Fig. 6B). When the NaOH was only used as an adjuvant, the viscosity of the mixtures was extremely high compared to the mixtures containing NaOH/DMSO or NaOH/DMSO/2M2B as adjuvants. With the NaOH adjuvant the viscosity suddenly changed at 70 ◦ C, and the slope in the 50–70 ◦ C region was approximately 3.4 times higher than that in the 70–85 ◦ C region (Fig. 6B). Although the substrate concentration in case with the adjuvant NaOH were extremely high, the app Ea value at 50–70 ◦ C was very low compared to that at 70–80 ◦ C. In eutectic media, substrate molecules may not move efficiently to the active site of enzyme because of highly viscous mixture, inducing diffusioncontrolled reactions. It has been widely known that a app low Ea value can be obtained in a mixture containing low substrate concentrations (Laidler and Bunting, 1980). When NaOH/DMSO and NaOH/DMSO/2M2B app were used as adjuvants, similar Ea values of 20.8 and 21.8 kJ mol−1 were attained even at low temperatures of 50–60 ◦ C, which correspond to that at higher temperatures in an NaOH adjuvant media. Howapp ever, the lower Ea values (12.9 and 15.5 kJ mol−1 ) were obtained at 60–80 ◦ C, which suggest a diffusion control. This phenomenon was similar to a substratedeficient environment in an immobilized enzyme system at higher temperatures corresponding to an internal diffusion limitation (Laidler and Bunting, 1980). Several investigations have suggested that reaction kinetics in highly viscous eutectic media is directly influenced by the viscosity of the reaction mixture. In other words, the initial velocity of reactions in eutectic media is strongly controlled by the high viscosity, which seriously reduces diffusion despite a concentration gradient 30–40 times higher than in organic media. Thus, we can expect viscous media to reduce the reaction kinetics.
Bartling et al. (2001) Lue et al. (2005) Yadav and Devi (2004) Yadav and Devi (2004) Yadav and Lathi (2005) Arcos et al. (2001) This study
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
acida
182
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184 app
Based on the Ea values and viscosities, kinetic app control is expected at Ea values of 20–21 kJ mol−1 and diffusion control at approximately 7–15 kJ mol−1 . app More specifically, Ea values of 12–15 and 7 kJ mol−1 indicate internal and external diffusion limitations, respectively. In contrast to our results, Garc´ıa-Alles and Gotor (1998) showed that only the enzymes located on the surfaces of immobilized particles are active and, thus, the internal mass transfer limitation was not severe for immobilized C. antarctica lipase. However, their transesterification reactions were carried out in a dilute liquid solution (0.08 M 1-butanol and 1.1 M ethyl acetate), so viscosity was not a factor. Table 2 shows results of the contrast between the eutectic and organic media. Despite similar conversion yields, reactions in eutectic media show 33–36 times higher overall productivity than in organic media. This result is due to the differences in initial substrate concentration and the time necessary to achieve the substrate–product equilibrium. The overall productivity result in organic media is similar to the report of Cao et al. (1997). The overall productivity of solid-phase synthesis was approximately 0.2–0.4 mmol g−1 h−1 , whereas a value of 3.2 mmol g−1 h−1 was obtained in the eutectic media. Even though the apparent substrate concentration of a heterogeneous reaction mixture including 20% (w/w) adjuvant was extremely high in the solid phase synthesis, the actual concentration in the liquid phase was only 0.8 M (Cao et al., 1997, 1999). The eutectic media showed a homogeneous liquid mixture with a concentration of 3.6–6.7 M, involving an adjuvant of 7–25% (w/w). 5. Conclusion In addition to technical advances in biocatalysis via engineering of media, biocatalysts, and proteins (Krishna, 2002), achieving a substrate medium with a high concentration and an efficient reaction system has been a goal of research. We have shown that the eutectic formation technique is useful in preparation of a sorbitol–fatty acid mixture with a remarkably high concentration. In eutectic media containing a homogeneously distributed substrate molecules, significant enhancements in initial reaction rate and overall productivity were achieved. However, the eutectic media may induce a viscous nature that seriously hinders
183
the diffusion. Diffusion limitations can be predicted based on a correlation between the viscosity and the app Ea value in eutectic media. A novel biocatalysis methodology using eutectic formation techniques can be further developed with additional aid of media engineering. References Ahn, J.E., Kim, C., Shin, C.S., 2001. Enzymic synthesis of aspartame precursors from eutectic substrate mixtures. Process Biochem. 37, 279–285. Arcos, J.A., Bernab´e, M., Otero, C., 1998a. Quantitative enzymatic production of 6-O-acylglucose esters. Biotechnol. Bioeng. 57, 505–509. Arcos, J.A., Bernab´e, M., Otero, C., 1998b. Quantitative enzymatic production of 1,6-diacyl sorbitol esters. Biotechnol. Bioeng. 60, 53–60. Arcos, J.A., Bernab´e, M., Otero, C., 1998c. Quantitative enzymatic production of 1,6-diacyl fructofuranoses. Enzyme Microb. Technol. 22, 27–35. Arcos, J.A., Hill Jr., C.G., Otero, C., 2001. Kinetics of the lipasecatalyzed synthesis of glucose esters in acetone. Biotechnol. Bioeng. 73, 104–110. Bartling, K., Thompson, J.U.S., Pfromm, P.H., Czermak, P., Rezac, M.E., 2001. Lipase-catalyzed synthesis of geranyl acetate in n-hexane with membrane-mediated water removal. Biotechnol. Bioeng. 75, 676–681. Bornscheuer, U.T., Bessler, C., Srinivas, R., Krishna, S.H., 2002. Optimizing lipases and related enzymes for efficient application. Trends Biotechnol. 20, 433–437. Cao, L., Fischer, A., Bornscheuer, U.T., Schmid, R.D., 1997. Lipasecatalyzed solid phase synthesis of sugar fatty acid esters. Biocatal. Biotransfor. 14, 269–283. Cao, L., Bornscheuer, U.T., Schmid, R.D., 1999. Lipase-catalyzed solid-phase synthesis of sugar esters. Influence of immobilization on productivity and stability of the enzyme. J. Mol. Catal. B: Enzym. 6, 279–285. Degn, P., Pedersen, L.H., Duus, J.Ø., Zimmermann, W., 1999. Lipase-catalysed synthesis of glucose fatty acid esters in tertbutanol. Biotechnol. Lett. 21, 275–280. Dordick, J.S., 1988. Biocatalysis in nonaqueous media. Patents and literature. Appl. Biochem. Biotechnol. 19, 103–112. Ducret, A., Giroux, A., Trani, M., Lortie, R., 1995. Enzymatic preparation of biosurfactants from sugars or sugar alcohols and fatty acids in organic media under reduced pressure. Biotechnol. Bioeng. 48, 214–221. Erbeldinger, M., Ni, X., Halling, P.J., 1998. Enzymatic synthesis with mainly undissolved substrates at very high concentrations. Enzyme Microb. Technol. 23, 141–148. Fiechter, A., 1992. Biosurfactants: moving towards industrial application. Trends Biotechnol. 10, 208–217. Flores, M.V., Halling, P.J., 2002. Full model for reversible kinetics of lipase-catalyzed sugar–ester synthesis in 2-methyl 2-butanol. Biotechnol. Bioeng. 78, 795–801.
184
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
Fregapane, G., Sarney, D.B., Greenbert, S.G., Knight, D.J., Vulfson, E.N., 1994. Enzymatic synthesis of monosaccharide fatty acid esters and their comparison with conventional products. J. Am. Oil Chem. Soc. 71, 87–91. Garc´ıa-Alles, L.F., Gotor, V., 1998. Lipase-catalyzed transesterification in organic media: solvent effects on equilibrium and individual rate constants. Biotechnol. Bioeng. 59, 684–694. Gill, I., Vulfson, E.N., 1993. Enzymatic synthesis of short peptides in heterogeneous mixtures of substrates. J. Am. Chem. Soc. 115, 3348–3349. Gill, I., Vulfson, E., 1994. Enzymic catalysis in heterogeneous eutectic mixtures of substrates. Trends Biotechnol. 12, 118–122. Gill, I., L´opez-Fandi˜no, R., Jorba, X., Vulfson, E.N., 1996. Biologically active peptides and enzymatic approaches to their production. Enzyme Microb. Technol. 18, 162–183. Hill, K., Rhode, O., 1999. Sugar-based surfactants for consumer products and technical applications. Fett-Lipid 101, 25–33. Hills, G., 2003. Industrial use of lipases to produce fatty acid esters. Eur. J. Lipid Sci. Technol. 105, 601–607. Kim, C., Shin, C.S., 2001. Solvent-free enzymatic synthesis of alitame precursor using eutectic substrate mixtures. Enzyme Microb. Technol. 28, 611–616. Kim, C., Lee, I.K., Ahn, J.E., Shin, C.S., 2001. Synthesis of kyotorphin precursor from eutectic mixtures catalysed by ␣chymotrypsin. Biotechnol. Lett. 23, 1423–1427. Kim, H.J., Kim, J.H., Youn, S.H., Shin, C.S., 2005. Eutectic formation analysis of amino acid mixtures using molecular dynamics simulations. Biotechnol. Progr. 21, 1307–1314. Klibanov, A.M., 2001. Improving enzymes by using them in organic solvents. Nature 409, 241–246. Krishna, S.H., 2002. Developments and trends in enzyme catalysis in nonconventional media. Biotechnol. Adv. 20, 239–267. Laidler, K.J., Bunting, P.S., 1980. The kinetics of immobilized enzyme systems. In: Purich, D.L. (Ed.), Methods in Enzymology, vol. 64. Academic Press, New York, pp. 227–248. L´opez-Fandi˜no, R., Gill, I., Vulfson, E.N., 1994a. Enzymatic catalysis in heterogenous mixtures of substrates: the role of the liquid phase and the effects of “Adjuvants”. Biotechnol. Bioeng. 43, 1016–1023. L´opez-Fandi˜no, R., Gill, I., Vulfson, E.N., 1994b. Protease-catalyzed synthesis of oligopeptides in heterogeneous substrates mixtures. Biotechnol. Bioeng. 43, 1024–1030. Lue, B.-M., Karboune, S., Yeboah, F.K., Kermasha, S., 2005. Lipase-catalyzed esterification of cinnamic acid and oleyl alcohol in organic solvent media. J. Chem. Technol. Biotechnol. 80, 462–468. Moreau, B., Lognay, G.C., Blecker, C., Broh´ee, J.-C., Ch´ery, F., Rollin, P., Paquot, M., Marlier, M., 2004. Synthesis of novel dglucuronic acid fatty esters using Candida antarctica lipase in tert-butanol. Biotechnol. Lett. 26, 419–424.
Paiva, A.L., Malcata, F.X., 1997. Integration of reaction and separation with lipases: an overview. J. Mol. Catal. B: Enzym. 3, 99–109. Plou, F.J., Cruces, M.A., Ferrer, M., Fuentes, G., Pastor, E., Bernab´e, M., Christensen, M., Comelles, F., Parra, J.L., Ballesteros, A., 2002. Enzymatic acylation of di- and trisaccharides with fatty acids: choosing the appropriate enzyme, support and solvent. J. Biotechnol. 96, 55–66. Reetz, M.T., 2002. Lipases as practical biocatalysts. Curr. Opin. Chem. Biol. 6, 145–150. Salis, A., Pinna, M.C., Murgia, S., Monduzzi, M., 2004. Novel mannitol based non-ionic surfactants from biocatalysis. J. Mol. Catal. B: Enzym. 27, 139–146. Sarney, D.B., Vulfson, E.N., 1995. Application of enzymes to the synthesis of surfactants. Trends Biotechnol. 13, 164–172. Shin, G.H., Kim, C., Kim, H.J., Shin, C.S., 2003. Stereoselective enzymatic synthesis of an aspartame precursor of N-CBZ-l-Aspl-PheOMe. J. Mol. Catal. B: Enzym. 26, 201–208. Tsuzuki, W., Kitamura, Y., Suzuki, T., Kobayashi, S., 1999. Synthesis of sugar fatty acid esters by modified lipase. Biotechnol. Bioeng. 64, 267–271. Ulijn, R.V., Janssen, A.E.M., Moore, B.D., Halling, P.J., 2001. Predicting when precipitation-driven synthesis is feasible: application to biocatalysis. Chem. Eur. J. 7, 2089–2098. Ulijn, R.V., De Martin, L., Gardossi, L., Halling, P.J., 2003. Biocatalysis in reaction mixtures with undissolved solid substrates and products. Curr. Org. Chem. 7, 1333–1346. Ulijn, R.V., Halling, P.J., 2004. Solid-to-solid biocatalysis: thermodynamic feasibility and energy efficiency. Green Chem. 6, 488–496. Vermu¨e, M.H., Tramper, J., 1995. Biocatalysis in non-conventional media: medium engineering aspects. Pure Appl. Chem. 67, 345–373. Yadav, G.D., Devi, K.M., 2004. Immobilized lipase-catalyzed esterification and transesterification reactions in non-aqueous media for the synthesis of tetrahydrofurfuryl butyrate: comparison and kinetic modeling. Chem. Eng. Sci. 59, 373–383. Yadav, G.D., Lathi, P.S., 2005. Lipase catalyzed transesterification of methyl acetoacetate with n-butanol. J. Mol. Catal. B: Enzym. 32, 107–113. Yan, Y., Bornscheuer, U.T., Cao, L., Schmid, R.D., 1999. Lipasecatalyzed solid-phase synthesis of sugar fatty acid esters: removal of byproducts by azeotropic distillation. Enzyme Microb. Technol. 25, 725–728. Yan, Y., Bornscheuer, U.T., Schmid, R.D., 2002. Efficient water removal in lipase-catalyzed esterifications using a low-boilingpoint azeotrope. Biotechnol. Bioeng. 78, 31–34. Zaks, A., Klibanov, A.M., 1985. Enzyme-catalyzed processes in organic solvents. Proc. Natl. Acad. Sci. U.S.A. 82, 3192– 3196.
Lipase-catalyzed synthesis of sorbitol–fatty acid esters at extremely high substrate concentrations Hyun Jung Kim, Sung Hun Youn, Chul Soo Shin ∗ Department of Biotechnology, College of Engineering, Yonsei University, 134 Shinchon-dong, Seodaemun-gu, Seoul 120-749, South Korea Received 29 April 2005; received in revised form 4 November 2005; accepted 9 November 2005
Abstract Lipase-catalyzed synthesis of sorbitol–fatty acid esters was performed in eutectic media with extremely high substrate concentrations. Homogeneous eutectic melts of sorbitol and fatty acids of C6–C16 were prepared using an adjuvant mixture. Enhanced homogeneity of mixtures was confirmed by X-ray diffraction analysis. The substrate concentration was 3.63–6.67 M in the eutectic media, whereas in organic media the concentration was below 0.10 M. Esters were synthesized with an immobilized Candida antarctica lipase, and optimum conditions were analyzed. Compared to reactions in organic media, the initial reaction rate of ester synthesis and the overall productivity were significantly enhanced in eutectic media while the conversion yields were similar. Based on the kinetic analysis, highly viscous eutectic media were shown to influence the initial reaction rate and the apparent activation energy resulting in diffusion limitations. © 2005 Elsevier B.V. All rights reserved. Keywords: Eutectic media; High concentration; Enhanced homogeneity; Diffusion limitation
1. Introduction The enzyme-catalyzed synthesis of biosurfactants has been reported during the past decade (Fiechter, 1992; Sarney and Vulfson, 1995). Among surfactants analyzed, the sugar–fatty acid esters are potentially useful in the pharmaceutical, cosmetic, and food industries as a non-ionic surfactant (Fregapane et al., 1994; Hill ∗ Corresponding author. Tel.: +82 2 2123 2886; fax: +82 2 362 7265. E-mail address: [email protected] (C.S. Shin).
0168-1656/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jbiotec.2005.11.006
and Rhode, 1999; Hills, 2003). Because of hydrophobic and hydrophilic natures of substrates, heterogeneous immiscible mixtures were induced, which hindered enzymatic reactions. However, lipase-catalyzed synthesis of these esters has been successfully carried out in non-aqueous organic media using hydrophilic solvents such as acetone (Arcos et al., 1998a,b,c, 2001), 2-methyl 2-butanol (Flores and Halling, 2002; Salis et al., 2004), and tert-butanol (Cao et al., 1999; Degn et al., 1999; Moreau et al., 2004). Unfortunately, substrate concentrations in organic media were extremely low at several different mM levels, suggesting non-
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
applicability for industrial uses even though high conversion yields of almost 100% can be attained. Among non-conventional media that are used for biocatalysis, organic solvent media and solid–liquid two-phase systems have been widely used in preparing substrate mixtures (Vermu¨e and Tramper, 1995)–other non-conventional media such as supercritical fluids, ionic liquids, and reverse micelles were not dealt here. Organic solvent media formed a homogeneous liquid phase including a small quantity of substrate (Fig. 1A) owing to a low solubility (Arcos et al., 1998c; Plou et al., 2002). Bioreactions in organic solvent media have been widely used for many applications over the past two decades (Zaks and Klibanov, 1985; Dordick, 1988; Klibanov, 2001). The reaction yields in the organic media have been improved by enhancement of enzyme activity via immobilization, genetic or protein engineering, extractive biocatalysis via integration of reaction and separation, and novel enzyme screening (Vermu¨e and Tramper, 1995; Paiva and Malcata, 1997; Tsuzuki et al., 1999; Bornscheuer et al., 2002; Krishna, 2002; Reetz, 2002). However, concentrations are still extremely low, resulting in limited industrial applications. The ‘solid phase media’ or ‘solid-to-solid system’ (Erbeldinger et al., 1998), which is a type of solid–liquid two-phase system, has a feature of liquid phase similar to that of organic solvent media (Fig. 1B). The solid phase media comprise a small volume of liquid phase and mostly solid substrate. However, since the substrate molecules existed in liquid phase can be only engaged in enzymatic reactions, reactants in liquid phase cannot greatly contribute to enhancement of overall productivity. According to a report of Cao et al. (1997), the initial reaction rate and overall productivity of reaction in solid phase media were enhanced approximately 10 times compared to those in organic solvent media. However, the substrate concentration in the liquid phase was only approximately 0.75 M (Cao et al., 1999) or lower (Yan et al., 1999). Reactions do not always proceed well in the solid-to-solid system, whose switch-like behavior was carefully investigated in several references (Ulijn et al., 2001, 2003; Ulijn and Halling, 2004). Eutectic mixture formation is a well-known melting technique that lowers the melting point of a mixture below the melting point of each pure compound in the mixture (Gill and Vulfson, 1994). An adjuvant usu-
175
Fig. 1. Schemes of substrate mixtures in organic solvent media (A), solid phase media (B), and eutectic media (C); S: substrate; P: product; E: enzyme.
ally helps to decrease the melting point and the molten mixture can be maintained at room temperature or below (L´opez-Fandi˜no et al., 1994a). The eutectic mixture, which exists as a state of homogeneous liquid
176
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
solution, consists of mostly substrate molecules and a small quantity (5–30%, gram-solvent per gram-total substrate mixtures) of organic solvent (Fig. 1C). As the molar ratio of a binary mixture is changed, the lowest melting point can be found at a specific mole fraction. This point is called as the ‘eutectic point’ and the corresponding mole fraction is as the ‘eutectic composition’. Often, the term ‘heterogeneous eutectic mixture’ was used (Gill and Vulfson, 1994; Erbeldinger et al., 1998). Here, we used the term ‘eutectic media’ for eutectic substrate mixtures comparative to organic media. Eutectic media exhibit a transparent and stable liquid phase with extremely high substrate concentrations, corresponding to approximately 100 times higher than the organic media. Since small amounts of organic solvents are used in eutectic media, they can be considered to be more biocompatible for enzymatic reactions than with organic phase media. There are a few reports (Gill and Vulfson, 1993; L´opez-Fandi˜no et al., 1994b; Gill et al., 1996) that extremely high concentrations of eutectic substrate mixtures were prepared in biocatalytic reactions. We have also reported results for enzymic dipeptide synthesis using eutectic media (Kim and Shin, 2001; Ahn et al., 2001; Kim et al., 2001; Shin et al., 2003). Based on the computer-aided molecular dynamics simulations, we confirmed that a eutectic mixture can be formed by the enhancement of overall molecular kinetic energy (Kim et al., 2005). However, this eutectic system has not been thoroughly investigated for industrial application to enzymatic reactions. In this study, we determined biosynthesis of sorbitol–fatty esters as a model system mainly performed in organic solvent media, and tried to enhance the substrate concentration levels via eutectic media engineering. The initial reaction rate, conversion yield, and overall productivity in eutectic media were compared with those in organic solvent media. Based on kinetic analysis, the effect of viscosity on the diffusion limitation in eutectic media was investigated.
2. Materials and methods 2.1. Enzyme and chemicals The Candida antarctica lipase (nonspecific triacylglycerol lipase, EC 3.1.1.3) immobilized on acrylic resin was obtained from Sigma Chemical Co. (a prod-
uct of Novozyme Corp.). All fatty acids (caproic, caprylic, capric, lauric, myristic, and palmitic) and dimethyl sulfoxide (DMSO) were also obtained from Sigma Chemical Co. Sorbitol and sodium hydroxide were purchased from Duksan. 2-Methyl-2-methanol (2M2B) was a product of Fluka. Methanol and acetonitrile (HPLC grade) were products of T.J. Baker. Molecular sieves (potassium, sodium alumino-silicate, ˚ were 8–12 mesh beads) with a nominal pore size of 3 A obtained from Sigma Chemical Co. (USA). 2.2. Synthesis of sorbitol–fatty esters in eutectic media For preparation of eutectic media, sorbitol (2 mmol) and fatty acids (2 mmol) containing an adjuvant were placed into screw-top glass vials and incubated for 30 min in a water bath (Jeio Tech, WB-10M) at 60 ◦ C. NaOH (3 M NaOH solution, 50 l), NaOH/DMSO (50/100 l), and NaOH/DMSO/2M2B (50/100/50 l) were used as adjuvants, respectively. In order to obtain a homogeneous solution, mixtures were cooled and reheated three times. Acylation in eutectic media was performed for 24 h with immobilized C. antarctica lipase in a water bath (Jeio Tech, SWB-03) at 100 rpm and 60 ◦ C. To eliminate the water generated in reaction mixtures, 300 mg of the activated molecular sieves ˚ incubated in a dry oven for 48 h were (pore size: 3 A) added. Temperatures were adjusted for Arrhenius plots from 50 to 80 ◦ C for substrate mixtures containing each adjuvant. All experiments were repeated three times, and the averaged values were used. Samples were intermittently withdrawn with a glass spatula and subjected to HPLC analysis. Water contents of reaction mixtures were measured with the moisture test kit (Riedel-de Ha¨en® , 37858 Hydranal® -Moisture Testkit). 2.3. Synthesis of sorbitol–fatty esters in organic media For preparation in organic media, sorbitol (0.5 mmol) and fatty acids (0.5 mmol) were prepared in a 15 ml screw-top glass vial containing 10 ml of an equimolar solvent mixture of 2M2B and acetonitrile (ACN). Preliminary drying of solvents was done with activated molecular sieves. Reactions were performed in a water bath (Jeio Tech, SWB-03) at 100 rpm and 60 ◦ C for 48 h with 300 mg activated molecular sieves.
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
Aliquots of 10 l were intermittently withdrawn for HPLC analysis. 2.4. Analysis of reaction mixtures A reaction mixture was quantitatively analyzed by HPLC. Samples intermittently obtained from reaction mixtures were diluted approximately 1/100 with the eluent, and immobilized enzymes were immediately removed by filtration (Whatman, PTFE syringe filter, 13 mm, 0.45 m). An HPLC system was used to analyze the product concentration. It was equipped with elution pump (Young-in, Model 910), a column (Thermo electron corporation, Hypersil GOLD column, 250 mm × 4.6 mm), an RI detector (Waters 410, Differential Refractometer), and analytical computer software (Yullin Technology, MultiChroTM ). The mobile phase consisting of a mixture of methanol and acetonitrile (55:45) was passed through the column at a flow rate of 0.5 ml min−1 and 30 l of the diluted reaction mixture was analyzed. The conversion yield was calculated from the amount of fatty acid consumed after reaction. Purified products were obtained by extraction with warm acetone (50 ◦ C) and filtration (Millipore, FluoroporeTM membrane filter, 0.45 m) of chilled extracts at −20 ◦ C (Yan et al., 1999). Dried white crystals were isolated with glass-based preparative TLC (Merck, 25 TLC plates 20 cm × 20 cm Silica gel 60 F254 ), and used to prepare a calibration curve. Identification of esters was performed via LC/MS, and 1 H and 13 C NMR. 2.5. Measurement of viscosity A 50 ml glass vial containing an equimolar mixture of sorbitol (30 mmol) and lauric acid (30 mmol) was incubated in a water bath at 60 ◦ C under various adjuvant conditions. After liquid melts were attained, shear stresses relating to the shear rates of each mixture were measured from 50 to 85 ◦ C using a rheometer (Brookfield, DV-III equipped with a No. 31 spindle). Based on plots of shear stresses versus rates, apparent viscosities from the slopes were calculated using the least square method. 2.6. X-ray diffraction analysis Equimolar mixtures of sorbitol (2 mmol) and lauric acid (2 mmol) were prepared, some containing the
177
NaOH adjuvant (50 l) and others containing an adjuvant mixture of NaOH/DMSO/2M2B (50/100/50 l). After the mixtures before and after eutectic melting treatment were re-solidified at room temperature for 48 h, X-ray diffraction (XRD; Rigaku, D/max; Rint 2000) analysis was performed at 30 kV and 25 mA with a scan speed of 4◦ min−1 . Data were plotted on intensities versus a 2θ diffraction angle.
3. Results 3.1. Preparation of eutectic mixtures We used the eutectic formation technique for enhancement of substrate concentrations. Fatty acids with different carbon numbers of C6–C16 were used as an acyl donor, and sorbitol was used as an acyl acceptor (=an alcohol donor). Based on preliminary analyses, NaOH, DMSO, and 2M2B were used as adjuvants in preparing eutectic media. Substrate mixtures of sorbitol and fatty acid were prepared with extremely high concentrations from 3.63 to 6.67 M (Table 1), which varied with the adjuvant contents. We adopted NaOH (3M) as an adjuvant, which allowed adjustment of the hydrogen ion concentration of biocatalysis. However, a heterogeneous emulsion mixture was formed when the NaOH was used alone. When DMSO and 2M2B were added to substrate mixtures, a homogeneous substrate solution was obtained. In contrast to eutectic media, organic media showed a substrate concentration below 0.1 M, which was approximately 1/40–1/50 of the concentrations in eutectic media. The morphology of a sorbitol–lauric acid substrate mixture was investigated via X-ray diffraction (Fig. 2). The substrate mixtures containing two different types of adjuvants (NaOH and NaOH/DMSO/2M2B) were melted. Both eutectic and non-eutectic mixtures were re-solidified then analyzed. Prior to eutectic melting a noisy base line and high sharp peaks appeared irregularly (Fig. 2, NaOH, Before). After eutectic melting, however, the peaks were shortened and the noisy small peaks disappeared (Fig. 2, NaOH, After). Prior to eutectic melting, when the adjuvant mixture NaOH/DMSO/2M2B was added, sharp and high peaks disappeared. The degree of noises on the base line was almost same as a case with the adjuvant NaOH (Fig. 2, NaOH/DMSO/2M2B, Before). On the contrary, after eutectic melting the
178
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
Table 1 Substrate concentrations in organic and eutectic media Substrate mixtures
Concentrations (M) Organic media
Sorbitol + caproic acid (C6) Sorbitol + caprylic acid (C8) Sorbitol + capric acid (C10) Sorbitol + lauric acid (C12) Sorbitol + myristic acid (C14) Sorbitol + palmitic acid (C16)
<0.10 <0.10 <0.10 <0.10 <0.10 <0.10
Eutectic media Adjuvant A
Adjuvant B
Adjuvant C
6.67 5.99 5.40 4.94 4.51 4.20
5.71 5.22 4.76 4.40 4.05 3.79
5.33 4.90 4.49 4.17 3.86 3.63
An equimolar mixture of 2M2B and ACN of 10 ml was used for a solvent in organic media. Adjuvant A: NaOH (3 M, 50 l); Adjuvant B: mixture of NaOH (3 M, 50 l) and DMSO (100 l); Adjuvant C: mixture of NaOH (3 M, 50 l), DMSO (100 l), and 2M2B (50 l).
intermittent sharp peaks disappeared, and a relatively stable base line with small noises were obtained (Fig. 2, NaOH/DMSO/2M2B, After). These results indicate that more homogeneous property was achieved in eutectic media than in mixture before eutectic melting. 3.2. Acylations in eutectic media
Fig. 2. X-ray diffraction patterns of sorbitol–lauric acid mixtures with adjuvants of NaOH and NaOH/DMSO/2M2B. All samples were analyzed after being re-solidified at room temperature for 48 h.
Lipase-catalyzed acylation of sorbitol was performed with the medium chain fatty acids capric, lauric, and myristic with an NaOH adjuvant (Fig. 3). After immobilized C. antarctica lipase was added to each substrate mixture, capric, lauric and myristic acids were rapidly converted with maximum rates of 6.0, 4.1, and 2.7 mol l−1 h−1 , respectively. Mono-esters were mainly produced and di-esters followed with a concentration of approximately 1/3–1/6 of the mono-ester. The maximum conversion yields for mono- and diesters were 56.5% and 22.5% in a sorbitol–capric acid mixture, 51.3% and 13.0% in a sorbitol–lauric acid mixture, and 43.7% and 10.4% in a sorbitol–myristic acid mixture, respectively. In the next, a mixture of sorbitol and lauric acid was used for reaction. The initial reaction rate and conversion yield of mono- and di-esters increased with an increasing amount of immobilized lipase up to 100 mg. Above that level, the initial reaction rate gradually increased whereas the conversion yield was kept at constant (Fig. 4A). Activated molecular sieves were added for trapping of the water generated during esterification. When the molecular sieves were added, the initial reaction rates for mono- and di-esters were 3.3 and 0.9 mol l−1 h−1 , and the maximum conversion yields were approximately 70% and 12%, respectively
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
179
Fig. 4. Initial reaction rates (open marks) and maximum conversion yields (closed marks) of ester synthesis between sorbitol and lauric acid at various contents of immobilized lipase (A) and molecular sieves (lipase contents: 100 mg) (B). Mono-esters: circle; di-esters: triangle.
Fig. 3. Reaction patterns of sorbitol–fatty acid esters. The immobilized lipase and activated molecular sieves were added at 100 and 300 mg, respectively; () sorbitol; (䊉) mono-ester; () di-ester.
(Fig. 4B). As a result, we confirmed that the critical amount of the enzyme and the molecular sieves were 100 and 300 mg for mixtures of sorbitol/fatty acid (4–5 M), respectively. Based on optimized reaction conditions, ester synthesis was performed with sorbitol and lauric acid. The mono-ester was rapidly produced up to 2.9 M while the di-ester was produced up to 0.4 M (Fig. 5). When the product concentration did not change further, it was assumed that the equilibrium was reached. After equilibrium of mono- and di-esters was reached in 12 h, hydrolysis due to water generation occurred. The concentration of esters was slightly decreased while the
lauric acid concentration was slightly increased in the reaction mixture. After the molecular sieves (300 mg) were added at 12 h, the concentrations of lauric acid, mono-ester, and di-ester were significantly changed. The water content was also rapidly decreased. Equilibrium of the reaction mixtures was disrupted by addition of the molecular sieves at 12 h, causing movement toward ester synthesis for approximately 6 h. The equilibrium concentrations of mono- and di-esters were up to 3.6 and 0.6 M, respectively. Because of a limit in the water-removal capacity, a new equilibrium was established at 19 h. 3.3. Kinetic analysis Kinetic parameters were evaluated from Arrhenius plots with an initial reaction rate of lauric acid (Fig. 6A) in the presence of the adjuvants NaOH, NaOH/DMSO or NaOH/DMSO/2M2B. Linear relationships were achieved within a region in which the
180
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
Fig. 5. Optimized synthesis of sorbitol–laurate ester via immobilized C. antarctica lipase (100 mg). The arrow indicates an additional feeding of activated molecular sieves; () sorbitol; (䊉) mono-ester; () di-ester; (×) water content.
regression coefficient (R2 ) was 0.98 or higher. Apparapp ent activation energies (Ea ) were calculated based on adjuvant type. When the NaOH adjuvant was used alone, the initial reaction rates were not affected by temperature differences in the low temperature region of 50–70 ◦ C and the Arrhenius plots showed a gentle slope app (Ea value: 7.0 kJ mol−1 ). On the contrary, a sharp app slope (Ea value: 21.5 kJ mol−1 ) was observed above 70 ◦ C (dashed line). For the reactions using adjuvant mixtures of NaOH/DMSO and NaOH/DMSO/2M2B, both showed similar patterns in the changes of logarithmic rate versus reverse temperature. At 50–60 ◦ C, the app reaction rates were suddenly elevated and high Ea values of 20.8 and 21.8 kJ mol−1 were obtained for the NaOH/DMSO and NaOH/DMSO/2M2B, respectively. When the temperature was increased above 60 ◦ C (dashed line), the slopes of each reaction mixapp ture plot were diminished and reduced Ea values of −1 12.9 and 15.5 kJ mol were observed. In order to elucidate the relationship between reaction kinetics and medium properties the viscosities of each reaction mixture were analyzed (Fig. 6B). When the NaOH was used alone as an adjuvant, a sticky solu-
Fig. 6. Arrhenius plots of sorbitol–laurate ester synthesis under various adjuvant conditions (A), the viscosities on temperature (B), and the initial reaction rates (open marks) and conversion yields (closed marks) on viscosity (C). Circle: NaOH only; triangle: NaOH/DMSO; quadrangle: NaOH/DMSO/2M2B.
tion with a high viscosity was developed at less than 70 ◦ C (dashed line). While the viscosity showed a big decrease from 2.18 to 0.71 Pa s in the low temperature region of 55–70 ◦ C, there was only a small decrease from 0.71 to 0.31 Pa s in the high temperature region of 70–85 ◦ C. In contrast to the viscosity of NaOH-
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
181
Table 2 Conversion yields and overall productivities for synthesis of sorbitol–fatty acid esters Substrate mixtures
Sorbitol + capric acid Sorbitol + lauric acid Sorbitol + myristic acid
Organic media
Eutectic media
Conversion yielda (%)
Overall productivityb (mmol g−1 h−1 )
Conversion yielda (%)
Overall productivityb (mmol g−1 h−1 )
94 89 83
0.098 0.093 0.086
98 97 94
3.27 3.23 3.13
a
Maximum conversion yields. Overall productivities were calculated from overall reaction rates when it reached a maximum ester production in organic (48 h) and eutectic media (6 h), respectively. b
adjuvant mixtures, the viscosities of the mixtures including NaOH/DMSO and NaOH/DMSO/2M2B as adjuvant mixtures were significantly reduced. Viscosity values decreased from 0.24 Pa s until the temperature reached 60 ◦ C, but slightly decreased above 60 ◦ C (dashed line). The relationships between viscosity, initial reaction rate, and conversion yield are depicted in Fig. 6C. The open and closed marks indicate initial reaction rates and maximum conversion yields of lauric acid, respectively, in all adjuvant types. Initial reaction rates (open) were significantly increased with a decrease in viscosity below 0.5 Pa s, but the conversion yield (closed) was almost the same over this viscosity range. In the high viscosity region, the conversion yield was diminished by approximately 20%. 3.4. Acylations in organic media Lipase-catalyzed acylations of sorbitol with medium chain fatty acids were performed in organic media. Maximum conversion yields in organic media were similar or a little lower than in the eutectic media (Table 2). On the other hand, the overall productivity in each medium was significantly different. The overall productivity in the organic media was observed at 0.086–0.098 mmol product per gram of immobilized lipase per hour (mmol g−1 h−1 ) at an equilibrium at 48 h, whereas 3.13–3.27 mmol g−1 h−1 was observed in the eutectic media at an equilibrium at 6 h.
4. Discussion XRD peaks generally indicate the degree of crystalline property and provide structural information
about materials. Based on XRD results in Fig. 2, the homogeneity of mixtures is quite different under adjuvant conditions. In addition, thermal treatment is essential for eutectic formation, and this thermal energy may promote molecular homogeneity. Enhanced homogeneity is considered to directly stimulate the production of mono-esters, as suggested by Ducret et al. (1995). Because of the low solubility of sorbitol in fatty acids, both reactants tend to exist separately, which leads to poor mono-ester production. Enhanced homogeneity of reactants made by addition of various solvents leads to increased mono-ester production. We found that mono-laurate synthesis was increased (compared to Fig. 3, middle) because of the enhanced homogeneity of reaction mixture under optimized adjuvant conditions regardless of mixing process (Fig. 5). Acylation reactions between sugars and fatty acids/fatty acid esters are condensation reactions that produce water or alcohol as a by-product. Esterification produces water that can induce hydrolysis, which generally reduces the ester production yield. Thus, removal of water is important (Yan et al., 2002). We ˚ for water used molecular sieves with a pore size of 3 A elimination. This method is effective, convenient and useful for removal of the water generated in a smallscale experiment, but is not suitable for biocatalysis on a preparative scale. We found linear relationships between logarithmic enzyme concentrations versus logarithmic initial rates (data not shown). This fact suggests that reactions would not occur under substrate-limiting conditions (Bartling et al., 2001). In reactions of eutectic media, the substrate concentration, the initial reaction rate, and the overall productivity were dramatically enhanced with an accompanying similar conversion yield, compared to the organic media. On the contrary, apparent
16.4 43.6 47.5 55.8 15.7 <50.4 7.0–21.8 All reactions were performed with immobilized C. antarctica lipase; n.d.: not described. a Esterification. b Transesterification.
10–50 35–55 30–70 30–70 30–60 30–60 50–80 0.1 0.004 0.004 0.004 0.002 n.d. 4.2–4.9 n-Hexane Hexane/2-butanone Heptane Heptane Toluene Acetone NaOH/DMSO/2M2B Geraniol Oleyl alcohol Tetrahydrofurfuryl alcohol Tetrahydrofurfuryl alcohol n-Butanol Glucose Sorbitol
Reference (kJ mol−1 ) app
Ea
Temperature range (◦ C) Concentration (M) Solvent Acyl acceptor Acyl donor
Table 3 Comparison of the activation energies for various lipase-catalyzed reactions
Acetic Cinnamic acida Butyric acida Ethyl butyrateb Methyl acetoacetateb Mono-laurateb Lauric acid
app
activation energy (Ea ) values were relatively low compared to similar reactions (Table 3). Regardless app of lipase source, Ea values were attained in a broad range of 15–55 kJ mol−1 in both esterifications and app transesterifications. This result indicates that the Ea value is not always directly dependent on the biocatalyst type, and other factors in the medium such as concentration, density, and viscosity can also affect the app Ea value. app In our investigations, we found that the Ea value was closely related to the viscosity in the eutectic media (Fig. 6B). When the NaOH was only used as an adjuvant, the viscosity of the mixtures was extremely high compared to the mixtures containing NaOH/DMSO or NaOH/DMSO/2M2B as adjuvants. With the NaOH adjuvant the viscosity suddenly changed at 70 ◦ C, and the slope in the 50–70 ◦ C region was approximately 3.4 times higher than that in the 70–85 ◦ C region (Fig. 6B). Although the substrate concentration in case with the adjuvant NaOH were extremely high, the app Ea value at 50–70 ◦ C was very low compared to that at 70–80 ◦ C. In eutectic media, substrate molecules may not move efficiently to the active site of enzyme because of highly viscous mixture, inducing diffusioncontrolled reactions. It has been widely known that a app low Ea value can be obtained in a mixture containing low substrate concentrations (Laidler and Bunting, 1980). When NaOH/DMSO and NaOH/DMSO/2M2B app were used as adjuvants, similar Ea values of 20.8 and 21.8 kJ mol−1 were attained even at low temperatures of 50–60 ◦ C, which correspond to that at higher temperatures in an NaOH adjuvant media. Howapp ever, the lower Ea values (12.9 and 15.5 kJ mol−1 ) were obtained at 60–80 ◦ C, which suggest a diffusion control. This phenomenon was similar to a substratedeficient environment in an immobilized enzyme system at higher temperatures corresponding to an internal diffusion limitation (Laidler and Bunting, 1980). Several investigations have suggested that reaction kinetics in highly viscous eutectic media is directly influenced by the viscosity of the reaction mixture. In other words, the initial velocity of reactions in eutectic media is strongly controlled by the high viscosity, which seriously reduces diffusion despite a concentration gradient 30–40 times higher than in organic media. Thus, we can expect viscous media to reduce the reaction kinetics.
Bartling et al. (2001) Lue et al. (2005) Yadav and Devi (2004) Yadav and Devi (2004) Yadav and Lathi (2005) Arcos et al. (2001) This study
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
acida
182
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184 app
Based on the Ea values and viscosities, kinetic app control is expected at Ea values of 20–21 kJ mol−1 and diffusion control at approximately 7–15 kJ mol−1 . app More specifically, Ea values of 12–15 and 7 kJ mol−1 indicate internal and external diffusion limitations, respectively. In contrast to our results, Garc´ıa-Alles and Gotor (1998) showed that only the enzymes located on the surfaces of immobilized particles are active and, thus, the internal mass transfer limitation was not severe for immobilized C. antarctica lipase. However, their transesterification reactions were carried out in a dilute liquid solution (0.08 M 1-butanol and 1.1 M ethyl acetate), so viscosity was not a factor. Table 2 shows results of the contrast between the eutectic and organic media. Despite similar conversion yields, reactions in eutectic media show 33–36 times higher overall productivity than in organic media. This result is due to the differences in initial substrate concentration and the time necessary to achieve the substrate–product equilibrium. The overall productivity result in organic media is similar to the report of Cao et al. (1997). The overall productivity of solid-phase synthesis was approximately 0.2–0.4 mmol g−1 h−1 , whereas a value of 3.2 mmol g−1 h−1 was obtained in the eutectic media. Even though the apparent substrate concentration of a heterogeneous reaction mixture including 20% (w/w) adjuvant was extremely high in the solid phase synthesis, the actual concentration in the liquid phase was only 0.8 M (Cao et al., 1997, 1999). The eutectic media showed a homogeneous liquid mixture with a concentration of 3.6–6.7 M, involving an adjuvant of 7–25% (w/w). 5. Conclusion In addition to technical advances in biocatalysis via engineering of media, biocatalysts, and proteins (Krishna, 2002), achieving a substrate medium with a high concentration and an efficient reaction system has been a goal of research. We have shown that the eutectic formation technique is useful in preparation of a sorbitol–fatty acid mixture with a remarkably high concentration. In eutectic media containing a homogeneously distributed substrate molecules, significant enhancements in initial reaction rate and overall productivity were achieved. However, the eutectic media may induce a viscous nature that seriously hinders
183
the diffusion. Diffusion limitations can be predicted based on a correlation between the viscosity and the app Ea value in eutectic media. A novel biocatalysis methodology using eutectic formation techniques can be further developed with additional aid of media engineering. References Ahn, J.E., Kim, C., Shin, C.S., 2001. Enzymic synthesis of aspartame precursors from eutectic substrate mixtures. Process Biochem. 37, 279–285. Arcos, J.A., Bernab´e, M., Otero, C., 1998a. Quantitative enzymatic production of 6-O-acylglucose esters. Biotechnol. Bioeng. 57, 505–509. Arcos, J.A., Bernab´e, M., Otero, C., 1998b. Quantitative enzymatic production of 1,6-diacyl sorbitol esters. Biotechnol. Bioeng. 60, 53–60. Arcos, J.A., Bernab´e, M., Otero, C., 1998c. Quantitative enzymatic production of 1,6-diacyl fructofuranoses. Enzyme Microb. Technol. 22, 27–35. Arcos, J.A., Hill Jr., C.G., Otero, C., 2001. Kinetics of the lipasecatalyzed synthesis of glucose esters in acetone. Biotechnol. Bioeng. 73, 104–110. Bartling, K., Thompson, J.U.S., Pfromm, P.H., Czermak, P., Rezac, M.E., 2001. Lipase-catalyzed synthesis of geranyl acetate in n-hexane with membrane-mediated water removal. Biotechnol. Bioeng. 75, 676–681. Bornscheuer, U.T., Bessler, C., Srinivas, R., Krishna, S.H., 2002. Optimizing lipases and related enzymes for efficient application. Trends Biotechnol. 20, 433–437. Cao, L., Fischer, A., Bornscheuer, U.T., Schmid, R.D., 1997. Lipasecatalyzed solid phase synthesis of sugar fatty acid esters. Biocatal. Biotransfor. 14, 269–283. Cao, L., Bornscheuer, U.T., Schmid, R.D., 1999. Lipase-catalyzed solid-phase synthesis of sugar esters. Influence of immobilization on productivity and stability of the enzyme. J. Mol. Catal. B: Enzym. 6, 279–285. Degn, P., Pedersen, L.H., Duus, J.Ø., Zimmermann, W., 1999. Lipase-catalysed synthesis of glucose fatty acid esters in tertbutanol. Biotechnol. Lett. 21, 275–280. Dordick, J.S., 1988. Biocatalysis in nonaqueous media. Patents and literature. Appl. Biochem. Biotechnol. 19, 103–112. Ducret, A., Giroux, A., Trani, M., Lortie, R., 1995. Enzymatic preparation of biosurfactants from sugars or sugar alcohols and fatty acids in organic media under reduced pressure. Biotechnol. Bioeng. 48, 214–221. Erbeldinger, M., Ni, X., Halling, P.J., 1998. Enzymatic synthesis with mainly undissolved substrates at very high concentrations. Enzyme Microb. Technol. 23, 141–148. Fiechter, A., 1992. Biosurfactants: moving towards industrial application. Trends Biotechnol. 10, 208–217. Flores, M.V., Halling, P.J., 2002. Full model for reversible kinetics of lipase-catalyzed sugar–ester synthesis in 2-methyl 2-butanol. Biotechnol. Bioeng. 78, 795–801.
184
H.J. Kim et al. / Journal of Biotechnology 123 (2006) 174–184
Fregapane, G., Sarney, D.B., Greenbert, S.G., Knight, D.J., Vulfson, E.N., 1994. Enzymatic synthesis of monosaccharide fatty acid esters and their comparison with conventional products. J. Am. Oil Chem. Soc. 71, 87–91. Garc´ıa-Alles, L.F., Gotor, V., 1998. Lipase-catalyzed transesterification in organic media: solvent effects on equilibrium and individual rate constants. Biotechnol. Bioeng. 59, 684–694. Gill, I., Vulfson, E.N., 1993. Enzymatic synthesis of short peptides in heterogeneous mixtures of substrates. J. Am. Chem. Soc. 115, 3348–3349. Gill, I., Vulfson, E., 1994. Enzymic catalysis in heterogeneous eutectic mixtures of substrates. Trends Biotechnol. 12, 118–122. Gill, I., L´opez-Fandi˜no, R., Jorba, X., Vulfson, E.N., 1996. Biologically active peptides and enzymatic approaches to their production. Enzyme Microb. Technol. 18, 162–183. Hill, K., Rhode, O., 1999. Sugar-based surfactants for consumer products and technical applications. Fett-Lipid 101, 25–33. Hills, G., 2003. Industrial use of lipases to produce fatty acid esters. Eur. J. Lipid Sci. Technol. 105, 601–607. Kim, C., Shin, C.S., 2001. Solvent-free enzymatic synthesis of alitame precursor using eutectic substrate mixtures. Enzyme Microb. Technol. 28, 611–616. Kim, C., Lee, I.K., Ahn, J.E., Shin, C.S., 2001. Synthesis of kyotorphin precursor from eutectic mixtures catalysed by ␣chymotrypsin. Biotechnol. Lett. 23, 1423–1427. Kim, H.J., Kim, J.H., Youn, S.H., Shin, C.S., 2005. Eutectic formation analysis of amino acid mixtures using molecular dynamics simulations. Biotechnol. Progr. 21, 1307–1314. Klibanov, A.M., 2001. Improving enzymes by using them in organic solvents. Nature 409, 241–246. Krishna, S.H., 2002. Developments and trends in enzyme catalysis in nonconventional media. Biotechnol. Adv. 20, 239–267. Laidler, K.J., Bunting, P.S., 1980. The kinetics of immobilized enzyme systems. In: Purich, D.L. (Ed.), Methods in Enzymology, vol. 64. Academic Press, New York, pp. 227–248. L´opez-Fandi˜no, R., Gill, I., Vulfson, E.N., 1994a. Enzymatic catalysis in heterogenous mixtures of substrates: the role of the liquid phase and the effects of “Adjuvants”. Biotechnol. Bioeng. 43, 1016–1023. L´opez-Fandi˜no, R., Gill, I., Vulfson, E.N., 1994b. Protease-catalyzed synthesis of oligopeptides in heterogeneous substrates mixtures. Biotechnol. Bioeng. 43, 1024–1030. Lue, B.-M., Karboune, S., Yeboah, F.K., Kermasha, S., 2005. Lipase-catalyzed esterification of cinnamic acid and oleyl alcohol in organic solvent media. J. Chem. Technol. Biotechnol. 80, 462–468. Moreau, B., Lognay, G.C., Blecker, C., Broh´ee, J.-C., Ch´ery, F., Rollin, P., Paquot, M., Marlier, M., 2004. Synthesis of novel dglucuronic acid fatty esters using Candida antarctica lipase in tert-butanol. Biotechnol. Lett. 26, 419–424.
Paiva, A.L., Malcata, F.X., 1997. Integration of reaction and separation with lipases: an overview. J. Mol. Catal. B: Enzym. 3, 99–109. Plou, F.J., Cruces, M.A., Ferrer, M., Fuentes, G., Pastor, E., Bernab´e, M., Christensen, M., Comelles, F., Parra, J.L., Ballesteros, A., 2002. Enzymatic acylation of di- and trisaccharides with fatty acids: choosing the appropriate enzyme, support and solvent. J. Biotechnol. 96, 55–66. Reetz, M.T., 2002. Lipases as practical biocatalysts. Curr. Opin. Chem. Biol. 6, 145–150. Salis, A., Pinna, M.C., Murgia, S., Monduzzi, M., 2004. Novel mannitol based non-ionic surfactants from biocatalysis. J. Mol. Catal. B: Enzym. 27, 139–146. Sarney, D.B., Vulfson, E.N., 1995. Application of enzymes to the synthesis of surfactants. Trends Biotechnol. 13, 164–172. Shin, G.H., Kim, C., Kim, H.J., Shin, C.S., 2003. Stereoselective enzymatic synthesis of an aspartame precursor of N-CBZ-l-Aspl-PheOMe. J. Mol. Catal. B: Enzym. 26, 201–208. Tsuzuki, W., Kitamura, Y., Suzuki, T., Kobayashi, S., 1999. Synthesis of sugar fatty acid esters by modified lipase. Biotechnol. Bioeng. 64, 267–271. Ulijn, R.V., Janssen, A.E.M., Moore, B.D., Halling, P.J., 2001. Predicting when precipitation-driven synthesis is feasible: application to biocatalysis. Chem. Eur. J. 7, 2089–2098. Ulijn, R.V., De Martin, L., Gardossi, L., Halling, P.J., 2003. Biocatalysis in reaction mixtures with undissolved solid substrates and products. Curr. Org. Chem. 7, 1333–1346. Ulijn, R.V., Halling, P.J., 2004. Solid-to-solid biocatalysis: thermodynamic feasibility and energy efficiency. Green Chem. 6, 488–496. Vermu¨e, M.H., Tramper, J., 1995. Biocatalysis in non-conventional media: medium engineering aspects. Pure Appl. Chem. 67, 345–373. Yadav, G.D., Devi, K.M., 2004. Immobilized lipase-catalyzed esterification and transesterification reactions in non-aqueous media for the synthesis of tetrahydrofurfuryl butyrate: comparison and kinetic modeling. Chem. Eng. Sci. 59, 373–383. Yadav, G.D., Lathi, P.S., 2005. Lipase catalyzed transesterification of methyl acetoacetate with n-butanol. J. Mol. Catal. B: Enzym. 32, 107–113. Yan, Y., Bornscheuer, U.T., Cao, L., Schmid, R.D., 1999. Lipasecatalyzed solid-phase synthesis of sugar fatty acid esters: removal of byproducts by azeotropic distillation. Enzyme Microb. Technol. 25, 725–728. Yan, Y., Bornscheuer, U.T., Schmid, R.D., 2002. Efficient water removal in lipase-catalyzed esterifications using a low-boilingpoint azeotrope. Biotechnol. Bioeng. 78, 31–34. Zaks, A., Klibanov, A.M., 1985. Enzyme-catalyzed processes in organic solvents. Proc. Natl. Acad. Sci. U.S.A. 82, 3192– 3196.