Process Biochemistry 43 (2008) 10611067

Contents lists available at ScienceDirect

Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio

Interfacially activated lipases against hydrophobic supports: Effect of the support nature on the biocatalytic properties
Gloria Fernandez-Lorente b, Zaida Cabrera a, Cesar Godoy a, Roberto Fernandez-Lafuente a, Jose M. Palomo a,*, Jose M. Guisan a,*
a b

lisis, Instituto de Cata lisis (CSIC), Campus UAM Cantoblanco, 28049 Madrid, Spain Departamento de Biocata Departamento de Microbiologa, Instituto de Fermentaciones Industriales, c/Juan de la Cierva 3, 2006 CSIC, Madrid, Spain

A R T I C L E I N F O

A B S T R A C T

Article history: Received 25 March 2008 Received in revised form 16 May 2008 Accepted 19 May 2008 Keywords: Interfacial activation Modulation of lipase properties Lipase purication Hydrophobic supports Lecitase

Different lipases (lipase B from Candida Antarctica, CAL-B, lipase from Thermomyces lanuginose, TLL and lipase from Bacillus thermocatenulatus, BTL) and a phospholipase (Lecitase1 Ultra) were immobilized by interfacial activation on four different hydrophobic supports (hexyl- and butyl-toyopearl and butyl- and octyl-agarose) and their properties were compared. The results suggested that selection of different supports yielded very different results in terms of recovered activity (ranging from a sevenfold hyperactivation to almost fully inactive biocatalysts), stability, specicity and adsorption strength. Even more interestingly, the enantioselectivity of the enzymes in the hydrolysis of ()-2-O-butyryl-2phenylacetic acid was strongly dependent on the support utilized. For example, BTL immobilized on octylagarose was fully enantiospecic towards the hydrolysis of (S)-2-O-butyryl-2-phenylacetic acid (E > 100), whereas when immobilized on hexyl-toyopearl, the enantiomeric value of the immobilized lipase was only E = 8. However, there is not an optimal support; it depends on the lipase and on the studied parameter. In the asymmetric hydrolysis of phenylglutaric acid diethyl diester, BTL immobilized on hexyl-toyopearl was the most enantioselective catalyst with ee > 99% (A factor >100) in the production of S-monoester product, whereas the enzyme immobilized on butyl-toyopearl only exhibited an A factor of 3. Finally, butyl-agarose was chosen as the most specic support on the lipase adsorption compared to other proteins at low ionic strength yielding the best purication of BTL from crude preparations. 2008 Elsevier Ltd. All rights reserved.

1. Introduction Lipases are the most used enzymes in biocatalysis and organic chemistry [15]. They recognize a wide variety of substrates while exhibiting high regioselectivity and enantiospecicity in many instances [6,7]. Lipases present a specic catalytic mechanism of action, existing in two structural forms, the closed one, where a polypeptide chain (lid or at) isolates the active center from the medium, and the open form, where this lid moves and the active center is exposed [810]. This equilibrium is shifted towards the open form in the presence of hydrophobic surfaces (e.g., droplets of oils), where the lipase becomes adsorbed by the large hydrophobic pocket around their active center and the internal face of the lid [11,12]. Moreover, lipases may become adsorbed to other hydrophobic

* Corresponding authors. Tel.: +34 91 585 4809; fax: +34 91 585 4760. E-mail addresses: josempalomo@icp.csic.es (J.M. Palomo), jmguisan@icp.csic.es (J.M. Guisan). 1359-5113/$ see front matter 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.procbio.2008.05.009

surfaces following a similar mechanism: hydrophobic proteins like hydrophobins [13], other lipase molecules [1417] or on the surface of hydrophobic supports [1820]. This particular mechanism of catalysis has permitted to develop specic protocols for the immobilization of lipases. Thus, the immobilization of lipases via interfacial activation on hydrophobic supports at low ionic strength has been reported to be a very simple and efcient method to immobilize and purify lipases [18]. This protocol xes the open form of lipases via interactions between the hydrophobic surroundings of their active centre the internal face of the lid and the area of the lipase around the active center that interacts with it and the hydrophobic surface of the support. These immobilized biocatalysts are very active against hydrophobic substrates having smallmedium size, even showing higher activity than that of the soluble enzyme when acting on fully soluble substrates [21]. Moreover, these immobilized enzyme preparations are very stable under different experimental conditions [18]. The adsorption of the lipase on the support is quite strong, but it is still reversible, permitting to recover and reuse the support after enzyme inactivation [18].

1062

G. Fernandez-Lorente et al. / Process Biochemistry 43 (2008) 10611067 Sweden). BTL cloned in E. coli was produced as previously described [14]. Butyl- and hexyl-toyopearl were from Tosoh Corporation (Tokyo, Japan). R/S mandelic acid, Triton X-100, hexadecyltrimethylammonium bromide (CTAB), and p-nitrophenyl butyrate (pNPB), were from Sigma. 2-O-Butyryl-2-phenylacetic acid (1) was synthesized as previously described [38]. 2.2. Lipase activity determination The standard assay was performed by measuring the increase in absorbance at 348 nm (isoblastic point of p-nitrophenol) produced by the releasing of pnitrophenol in the hydrolysis of 0.4 mM p-nitrophenyl butyrate (pNPB) in 25 mM sodium phosphate at pH 7 and 25 8C, using a thermostatized spectrophotomer with magnetic stirring. To start the reaction, 0.1 mL of lipase solution or suspension was added to 2.5 mL of substrate solution. An international unit of pNPB activity is dened as the amount of enzyme necessary to hydrolyze 1 mmol of pNPB/min (IU) under the conditions described above. 2.3. Lipase immobilization Ten grams of support were added to 100 mL of lipase solution (0.01 mg prot/mL or 1 mg/mL) in 5 mM sodium phosphate pH 7 at 25 8C under very mild stirring. Periodically, samples of supernatant and suspension were withdrawn and the activities were determined as described above. After immobilization, the immobilized enzymes were washed with an excess of distilled water and stored at 4 8C. The preparations with low enzyme loading, where diffusion problems were avoided, were used to determine the activity with pNPB and inactivation activity. The preparations with high enzyme loading were used in the hydrolysis of 2-Obutyryl-2-phenylacetic acid and phenylglutaric acid diethyl diester. 2.4. Lipase desorption One gram of immobilized lipase preparation was re-suspended in 10 mL of 5 mM sodium phosphate pH7 and 25 8C and detergent (Triton X-100, except for TLL where CTAB was used due to the inhibition caused by Triton X-100 [39]) was progressively added [18]. The immobilized enzymes were incubated under gentle stirring for 30 min before measuring the enzyme activity in the supernatant. Afterwards, whenever necessary, new additions of detergent were performed. References with soluble enzymes submitted to identical treatment were used to determine the effect of the detergent on the enzyme activity. 2.5. Lipase inactivation experiments Immobilized enzyme suspensions were incubated at the indicated temperature and at pH 7. Samples were periodically withdrawn and the activity was determined as previously described. 2.6. Synthesis of 3-phenylglutaric acid diethyl diester (2) A solution of ethanol (70 mmol), 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (3.6 mmol) with dimethylamino-pyridine (0.72 mmol) in diethyl ether (10 mL) was added drop-wise over a stirred solution of 3-phenylglutaric acid (1.44 mmol) in diethyl ether (25 mL) The mixture was stirred at 25 8C for 5 h and the reaction was followed by HPLC. After that, the mixture was extracted with 100 mM NaCl solution and after re-extraction of the water phase with diethyl ether. The combined organic solvents were dried with NaSO4 and the solvent was evaporated. The crude extract was washed several times with cold ether (5 2 mL) and dried in vacuum. Yield: 98%. 1 H NMR (400 MHz CDCl3): d = 7.337.20 (m, 5H, Ph); 4.52 (m, 1H, CH), 4.13 (m, 4H, 2 CH2), 2.54 (m, 4H, 2 CH2), 1.29 (t, 6H, 2 CH3). 2.7. Enzymatic hydrolysis of substrates 1 and 2 One gram of immobilized lipase (prepared using 10 mg of protein per g of support) was added to 10 mL of 5 mM (1) in 25 mM sodium acetate at pH 5 and 25 8C or 1 mM (2) in 25 mM sodium phosphate at pH 7 and 25 8C. During the reaction, the pH value was maintained constant by automatic titration using a pH-stat Mettler Toledo DL50 graphic. The degree of hydrolysis was conrmed by reverse-phase HPLC (Spectra Physic SP 100 coupled with an UV detector Spectra Physic SP 8450) on a Kromasil C18 (25 cm 0.4 cm) column supplied by Analisis Vinicos (Spain). At least triplicates of each assay were made. The elution was isocratic with a mobile phase of acetonitrile (35%) and 10 mM ammonium phosphate buffer (65%) at pH 2.95 and a ow rate of 1.5 mL/min. The elution was monitored by recording the absorbance at 225 nm (substrate 1) and 205 nm (substrate 2). The enzymatic activity was measured in mmol of substrate hydrolyzed per hour per mg of immobilized protein. 2.8. Determination of enantiomeric excess At different conversion degrees, the enantiomeric excess (ee) of the acid (in the hydrolysis of 1) or the formed monoester (in the hydrolysis of 2) was analyzed by

Following this immobilization mechanism, the active center of the lipase is in close contact with the support surface, which will generate a hydrophobic environment around the active center. Thus, changes in the support nature (hydrophobicity, internal morphology, etc.) could alter the strength of the lipase adsorption on the support and even the structure of the active center surroundings and, therefore, the nal lipase catalytic features. It has been reported that due to the large changes in the lipase conformation during catalysis, it is quite simple to modulate their catalytic properties during immobilization (varying the orientation, rigidity, microenvironment, etc.), greatly altering the enantiospecicity [2227] or regioselectivity [28,29] of the immobilized enzymes. Here, the results of a study of the support effects on the properties of lipases (selectivity, specicity, activity, etc.) are presented. In this case, the immobilization of lipases follows the same mechanism: interfacial activation against hydrophobic supports [18]. The lipase active center will be in very close contact with the support surface. However, considering the hydrophilic nature of the lipases out of the active center surroundings, scarce additional inuences of the lipase with the support should be expected. The close contact between the support and the active center may produce large differences in the nal structure of the enzyme active center and, therefore, on the nal enzyme catalytic properties. Thus, the nature of the support that produces that interfacial activation could determine the nal properties of the immobilized enzyme. Moreover, the selectivity in the protein adsorption on the different supports and the adsorption strength of the lipase to the different supports has been analyzed. Two supports with very different internal morphology were used. Agarose beads are formed by wide trunks [30]. Compared to the lipase size, an almost planar surface will be available to interact with the enzyme. Agarose is very hydrophilic, but if coated with octyl or butyl chains, it offers a hydrophobic surface suitable for the lipase interfacial activation [18]. Toyopearl is an acrylic support formed by a mild crosslink (supplier information), giving quite thin bers, even smaller than the size of a lipase molecule. The support matrix has a certain hydrophobic character that may be reinforced by coating the bers surface with butyl or hexyl groups. Previous results suggested that the adsorption of lipases to a hydrophobic support may be driven by reasons more complex that the mere hydrophobicity of the support. Usually, the more hydrophobic the support, the higher the amount of lipases that becomes adsorbed on it [18]. However, in some cases the use of moderately hydrophobic supports permits the adsorption of lipases that did not adsorb on supports with high hydrophobicity, for example pancreas porcine lipase did not adsorb on octadecyl Sepabeads or octyl-agarose but it was strongly adsorbed on phenyl agarose (a support with lower hydrophobicity) [31]. Thus, a more systematic study may help to understand on the reasons for the adsorption of lipases on hydrophobic supports. In this study, we have used three different lipases: those from Bacillus thermocatenulatus (BTL) [14], Candida antarctica B (CAL-B) [32,33] and from Thermomyces lanuginose (TLL) [34]. Moreover, we have included in this study a commercial phospholipase, Lecitase1 Ultra, which is used in degumming processes [35] and that has been described to behave as a lipase in many aspects. This enzyme presents a certain interest in ne chemistry [36,37].
2. Materials and methods 2.1. Materials Lecitase Ultra, CAL-B and TLL were obtained from Novozymes (Denmark). Butyl- and octyl-agarose 4BCL was purchased from GE healthcare (Uppsala,
1

G. Fernandez-Lorente et al. / Process Biochemistry 43 (2008) 10611067 chiral reverse phase HPLC. The column was a Chiracel OD-R. In the case of 1, the mobile phase was an isocratic mixture of 5% acetonitrile and 95% NaClO4/HClO4 0.5 M at pH 2.3 and the analyses were performed at a ow of 0.5 mL/min by recording the absorbance at 225 nm. For 2, the mobile phase was acetonitrile (25%) and 10 mM ammonium phosphate buffer (75%) at pH 2.95 and the analyses were performed at a ow of 0.7 mL/min by recording the absorbance at 205 nm. 2.9. Calculation of E value and A factor The enantiomeric ratio (E) was dened as the ratio between the percentage of hydrolyzed R and S isomers (from racemic mixture) at hydrolysis degrees between 15 and 20%. The asymmetric factor (A) of the produced monoester isomers was dened as the ratio between the percentage of produced R and S isomers of monoester. The absolute conguration was assigned in agreement with the results of Ostaszewski and co-workers [40].

1063

3. Results and discussion 3.1. Immobilization of BTL on different hydrophobic supports After 1 h, BTL had been fully immobilized on all hydrophobic supports. In all cases, the effect of the immobilization of the lipase on its activity against pNPB was positive (Table 1). BTL increased the activity by a twofold factor when it was immobilized on butylor octyl-agarose supports, or hexyl-toyopearl, and by a 1.5-fold factor when the support was butyl-toyopearl. The different adsorbed BTL-preparations were incubated in growing concentrations of detergents and the amount necessary to desorb 100% of the lipase was taken as a measure of the adsorption strength (Table 2). Curiously, BTL adsorbed on both butyl- and octyl-agarose supports support having larger surfaces and where protein-support multi-interactions could be more easily obtained could be desorbed from the supports at the lowest concentration of detergent. However, toyopereal adsorbed more strongly the enzyme, although the bers are very thin and it is difcult to establish many enzymesupport interactions (Table 2). It should be considered that the mechanism of adsorption of the lipases in hydrophobic supports is not via a conventional hydrophobic adsorption (where the surface offered by the support to give multiple interactions may be a key to dene the strength of the adsorption) [22]. Now, the only part of the lipase that needs to interact with the support is the large hydrophobic pocket around the active center of the lipase [18]. Thus, agarose offers a plane for the interaction with the lipase [30], but has a reduced geometrical congruence with the more or less internal pocket formed by the lid
Table 1 Effect on the enzyme activity of the immobilization of BTL on different hydrophobic supports Support Octyl-agarose Butyl-agarose Hexyl-toyopearl Butyl-toyopearl
a

Fig. 1. Inactivation of different preparations of Bacillus thermocatenolatus lipase. Inactivations were performed at pH 7 and 50 8C. Squares, octyl-sepharose-BTL; circles, butyl-sepharose-BTL; triangles, butyl- or hexyl-toyopearl-BTL.

Relative activity (%) a 200 200 200 150

Relative activity considers 100 the activity of the soluble enzyme.

and the hydrophobic areas around the active center and the layer of hydrophobic groups in the support. The use of toyopearl supports made it necessary to double the concentration of detergent to release the lipase from the support, suggesting a much stronger interaction between the thin bers that compose this support and the lipase. It may be assumed that the hydrophobic pocket of the lipase may present a higher geometrical congruence with these thin bers than with large planar surfaces, involving more groups of the enzyme in the adsorption that may adapt to these thin bers. In none of the supports utilized, was possible it to detect a relevant effect of using more hydrophobic moieties coating the support. In fact, in both cases, the BTL adsorbed on the supports activated with the less hydrophobic group required the higher amount of detergent to be desorbed. Results suggest that not only the hydrophobicity of the support surface but also the exact morphology of the support surface may dene the adsorption of a lipase on a support [31]. The thermal stability of the different BTL preparations was again quite dissimilar. Octyl-agarose-BTL was the most stable preparation, being the second most stable butyl-agarose-BTL while both toyopearl preparations presented very similar stability (Fig. 1). Next, the hydrolysis of ()-2-O-butyryl-2-phenylacetic acid (1) (Scheme 1) by the different BTL preparations was studied (Table 3). All the BTL preparations presented very similar activity value. Curiously, the octyl-agarose-BTL preparation was the only one having around 60% of the activity (in the case of pNPB this preparation was the most active). Although all the enzymes preparations preferred to hydrolyze (S)-1, the agarose-BTL preparations were the most enantiospecic in this reaction, with a E value over 100 using octyl-agarose-BTL and over 60 using butyl-agarose-BTL. The E values for the other preparations were much lower. The different BTL preparations were also assayed against di-ethyl phenylglutarate (2) (Scheme 1), a prochiral diester (Table 4). The activities against this substrate were two orders of magnitude lower than using substrate 1, and the specic activities were quite similar for all preparations, ranging from 0.0010 U/mg protein (using butyl-agarose-BTL) to

Table 2 Amount of detergent necessary to fully release BTL adsorbed on different hydrophobic supports Support Octyl-agarose Butyl-agarose Hexyl-toyopearl Butyl-toyopearl Experimental conditions are described in Section 2. Triton X-100 (%) 0.2 0.3 0.4 0.6

Scheme 1. Model compounds used to evaluate the selectivity of different lipase preparations. Racemic 2-O-butyryl-2-phenylacetic acid (Rac 1); 3-phenylglutaric acid diethyl diester (2).

1064

G. Fernandez-Lorente et al. / Process Biochemistry 43 (2008) 10611067 Table 5 Effect on the enzyme activity of the immobilization of lipases on different hydrophobic supports Enzyme CAL-B Support Octyl-agarose Hexyl-toyopearl Octyl-agarose Butyl-toyopearl Octyl-agarose Butyl-toyopearl Relative activitya (%) 130 40 750 60 220 50

Table 3 Effect on the enzyme enantioselectivity of the immobilization of BTL on different hydrophobic supports in the hydrolysis of substrate 1 Support Octyl-agarose Butyl-agarose Hexyl-toyopearl Butyl-toyopearl Activitya 0.25 0.4 0.4 0.38 E >100 66 8 18 Isomer S S S S

TLL

Other details are described in Section 2. a Activity in mmol g1 min1 at 15% conversion. cat Table 4 Asymmetric hydrolysis of 2 catalyzed by immobilized BTL on different hydrophobic supports Support Octyl-agarose Butyl-agarose Hexyl-toyopearl Butyl-toyopearl Activitya 0.0015 0.0010 0.0012 0.0014 A factor (S) 44 29 116 3

Lecitase

Activity was determined against pNPB as described in Section 2. a Relative activity considers 100 the activity of the soluble enzyme.

Table 6 Amount of detergent necessary to fully release the lipases adsorbed on different hydrophobic supports Enzyme CAL-B Support Octyl-agarose Butyl-toyopearl Octyl-agarose Hexyl-toyopearl Butyl-toyopearl Octyl-agarose Butyl-toyopearl Detergent Triton X-100 Concentration (%) 0.2 0.2 0.3 0.8 0.4 0.2 0.6

Experiments were performed as described in Section 2. a Activity in mmol g1 min1 . cat

TLL

CTAB

0.0015 U/mg protein (using octyl-agarose-BTL), with the activities displayed for both toyopearl preparations in between. The differences in enantioselectivity values were much higher. For example, the butyl-toyopearl-BTL gave only an asymmetric factor value of 3, while hexyl-toyopearl-BTL permitted a value over 100, producing only the (S)-monoester product. Therefore, the results showed clear differences in the behaviour of BTL depending on the hydrophobic support employed. In all cases the enzyme was immobilized via interfacial activation on a hydrophobic support, but the change in the internal morphology (plane versus thin ber) or the groups coating the support may greatly alter the enzyme properties. However, the effects are not easy to predict, the optimal catalyst depends on the specic parameter studied, in some aspects one preparation was the most adequate (e.g., octyl-agarose-BTL for the resolution of 1), while in other cases the best preparation was other (e.g., hexyl-toyopearlBTL for the asymmetric hydrolysis of 2). Thus, due to the close contact between the enzyme active center and the support, it seems that the active center of the lipase may be easily altered by just changing the support morphology. 3.2. Immobilization of other lipases on different hydrophobic supports In order to check if these results achieved using BTL were just a peculiarity of BTL or a more general phenomenon, similar studies using CAL-B, TLL and Lecitase were performed. In all cases, immobilization yield of the enzymes was over 80% in only 1 h. Next, we show the results where the largest differences were found between the enzymes immobilized on the different supports. 3.2.1. Effect on enzymatic activity CAL-B slightly increased its activity when it was immobilized on octyl-agarose while the immobilization on both toyopearl supports produced a reduction on the enzyme activity by around 50% (Table 5). TLL increased the activity by a 7.5-fold factor when it was immobilized on octyl-agarose while reducing it to 60% when it was immobilized on butyl-toyopearl (Table 5). Lecitase increased the activity by a twofold factor when immobilized on octyl-agarose but the activity was reduced by 50% when the support was butyltoyopearl. Thus, octyl-agarose afforded the highest activity for the three enzymes studied. Immobilization on toyopearl gave the lowest activities, this could be related to a partial blocking of the

Lecitase

Triton X-100

Experimental conditions are described in Section 2.

active center of the lipases by an excessive congruence between the thin bers that form the support and the hydrophobic pocket formed by the surroundings of the active center of the lipase. 3.2.2. Strength of lipase adsorption on different hydrophobic supports The support that requires the highest amount of detergent to desorb TLL was hexyl-toyopearl whereas in the case of Lecitase, was butyl-toyopearl (Table 6). Toyopearl was the support that gave the strongest adsorption. This suggested again that there is a better congruence between the pocket of the lipase and these thin bers than with large and plane surfaces, in opposition to any other immobilization based in multipoint adsorption [22]. The fact that the adsorption strength is different depending on the lipase used, suggests that the adsorption strength is determined not only by the support (e.g., hydrophobicity of the support surface and size of the ber of the support) but also by the lipase features (size of the lid and hydrophobic residues number). All the supports after the lipase desorption may be utilized to immobilize fresh enzyme, permitting their reuse by several adsorptiondesorption cycles without detecting any effect on the immobilization parameters: immobilization ratio and yield was maintained, the amount of detergents to desorb the enzyme was identical, and the catalytic features of the lipase were maintained within the error limits. 3.2.3. Thermal stability of lipases immobilized on different hydrophobic supports All the lipases studied resulted much more stable after immobilization than in their soluble form [18]. The different lipases exhibited very different results in terms of thermal stability when the obtained immobilized preparations using the different supports were compared. For example, TLL immobilized on hexyl-toyopearl was the most stable, being the enzyme immobilized on butyl-toyopearl the less stable (Fig. 2). However, using CAL-B or Lecitase, all the immobilized enzyme preparations presented almost identical stability (see as an example in Fig. 3 the results with CAL-B).

G. Fernandez-Lorente et al. / Process Biochemistry 43 (2008) 10611067

1065

Fig. 2. Inactivation courses of different preparations of TLL. Inactivations were performed at pH 7 and 65 8C. Circles, hexyl-toyopearl-BTL; triangles, octylsepharose-BTL; squares, butyl-toyopearl-BTL. Fig. 4. SDS-PAGE of the proteins adsorbed on different hydrophobic supports after incubating the supports with crude extract of BTL. Lane 1: molecular weight markers; lane 2: BTL cloned in E. coli crude extract; lane 3: proteins adsorbed on octyl-agarose; lane 4: proteins adsorbed on hexyl-toyopearl; lane 5: proteins adsorbed on butyl-toyopearl; lane 6: proteins adsorbed on butyl-agarose.

Fig. 3. Inactivation courses of different preparations of CAL-B. Inactivations were performed at pH 7 and 65 8C. Circles, octyl-sepharose CAL-B; squares, hexyltoyopearl-CAL-B; triangles, butyl-toyopearl-CAL-B.

against this substrate, although the activity against pNPB was signicantly high. Even more interesting was the fact that the enantiomeric ratio (E) of the different preparations also depended on the support used, CAL-B preferred the isomer R, while the other lipases displayed a preference towards the hydrolysis of S-isomer. CAL-B-octyl-agarose preparation showed the highest E value of 60 compared to the enzyme immobilized on butyl-toyopearl (E = 3). However, TLL and Lecitase presented similar enantioselectivity using the different hydrophobic supports (E = 56 by TLL and E = 810 by Lecitase). 3.3. Selectivity of the lipase adsorption

Again, depending on the enzyme, the effect of the hydrophobic support (in this case on the enzyme stability) was completely different. 3.2.4. Enantioselectivity of lipases immobilized on different hydrophobic supports The different lipase immobilized preparations were used in the hydrolytic resolution of 1. Again, the lipase features depended on the support used to immobilize it (Table 7). The octyl-agarose-lipase preparations were the most active catalysts for all enzymes against pNPB. However, in the hydrolysis of 1, the adsorption on hexyl-toyopearl permitted to obtain the most active catalyst for TLL and Lecitase. On the other hand, TLL and Lecitase immobilized on butyl-toyopearl were almost inactive
Table 7 Activity and enantioselectivity on the hydrolysis of 1 of lipases immobilized on different hydrophobic supports Enzyme CAL-B Support Octyl-agarose Butyl-toyopearl Octyl-agarose Hexyl-toyopearl Butyl-toyopearl Octyl-agarose Hexyl-toyopearl Butyl-toyopearl Activitya 17 27 0.22 0.24 0.001 0.85 1.25 0.001 Hydrolyzed isomer R R S S S S S S Eb 60 3 5 6 10 8

CAL-B, TLL and Lecitase were quite pure lipase preparations that became fully puried after immobilization on any of the supports used in this work, e.g., octyl-agarose [18]. However, BTL extract was a quite crude preparation and other proteins apart from the lipase were adsorbed on octyl-agarose at low ionic strength, reducing the selectivity of the adsorption [41]. Here, we have compared the proteins adsorbed on the different supports. In all cases the purication of the lipase was very high just during the adsorption step, but Fig. 4 shows that the amount of contaminant proteins that became adsorbed on the support increased when changing butyl-agarose, by octyl-agarose or hexyl-toyopearl, but decreased when butyl-toyopearl was used. These results suggested that hexyl-toyopearl was slightly less selective for lipase adsorption that octyl-agarose while butylagarose was signicantly the most selective. 4. Conclusions Lipases may become adsorbed in many different hydrophobic supports, most likely by interfacial activation. The immobilization is very efcient, permitting the full and rapid immobilization of the lipase. The adsorption is strong enough to use the enzyme preparation as an immobilized one, although the enzyme may be desorbed by using detergents, therefore, this immobilization strategy retains the reversibility that is one of the advantages of physical adsorptions versus covalent immobilization. However, depending on the nature of the support that produces the lipase activation, all the properties of the immobilized lipase changed, and the optimal preparation was different depending on the parameter or the lipase studied. Considering that all the

TLL

Lecitase

Other details are described in Section 2. a Activity was dened as: UI/mgprotein 103. It was calculated at 1015% conversion. b E was calculated when 20% of the substrate had been consumed.

1066

G. Fernandez-Lorente et al. / Process Biochemistry 43 (2008) 10611067 [8] Brady L, Brzozowski AM, Derewenda ZS, Dodson E, Dodson G, Tolley S, et al. A serine protease triad forms the catalytic centre of a triacylglycerol lipase. Nature 1990;343:76770. [9] Brzozowski AM, Derewenda U, Derewenda ZS, Dodson GG, Lawson DM, Turkenburg JP, et al. A model for interfacial activation in lipases from the structure of a fungal lipase-inhibitor complex. Nature 1991;351:4914. [10] Derewenda U, Brzozowski AM, Lawson DM, Derewenda ZS. Catalysis at the interface: the anatomy of a conformational change in a triglyceride lipase. Biochemistry 1992;31:153241. [11] Van Tilbeurgh H, Egloff M-P, Martinez C, Rugani N, Verger R, Cambillau C. Interfacial activation of the lipase-procolipase complex by mixed micelles revealed by X-ray crystallography. Nature 1993;362:81420. [12] Verger R. Interfacial activation of lipases: facts and artifacts. Trends Biotechnol 1997;15:328. as [13] Palomo JM, Pen MM, Fernandez-Lorente G, Mateo C, Pisabarro AG, FernandezLafuente R, et al. Solid-phase handling of hydrophobins: immobilized hydrophobins as a new tool to study lipases. Biomacromolecules 2003;4:20410. [14] Rua ML, Schmidt-Dannert C, Wahl S, Sprauer A, Schmid RD. Thermoalkalophilic lipase of Bacillus thermocatenulatus: large-scale production, purication and propertiesaggregation behaviour and its effect on activity. J Biotechnol 1997;56:89102. [15] Dunhaupt A, Lang S, Wagner F. Pseudomonas cepacia lipase: studies on aggregation, purication and on the cleavage of olive oil. Biotechnol Lett 1992;14:9538. [16] Jaeger K-E, Adrian F-J, Meyer HE, Hancock REW, Winkler UK. Extracellular lipase from Pseudomonas aeruginosa is an amphiphilic protein. Biochim Biophys Acta Prot Struct Mol Enzymol 1992;1120:31521. [17] Palomo JM, Fuentes M, Fernandez-Lorente G, Mateo C, Guisan JM, FernandezLafuente R. General trend of lipase to self-assemble giving bimolecular aggregates greatly modies the enzyme functionality. Biomacromolecules 2003;4: 16. [18] Fernandez-Lafuente R, Armisen P, Sabuquillo P, Fernandez-Lorente GM, Gui san J. Immobilization of lipases by selective adsorption on hydrophobic supports. Chem Phys Lipids 1998;93:18597. [19] Foresti ML, Ferreira ML. Analysis of the interaction of lipases with polypropylene of different structure and polypropylene-modied glass surface. Colloid Surf A Physicochem Eng Asp 2007;294:14755. [20] Bryjak J, Trochimczuk AW. Immobilization of lipase and penicillin acylase on hydrophobic acrylic carriers. Enzyme Microb Technol 2006;39:5738. [21] Fernandez-Lorente G, Palomo JM, Cabrera Z, Guisan JM, Fernandez-Lafuente R. Specicity enhancement towards hydrophobic substrates by immobilization of lipases by interfacial activation on hydrophobic supports. Enzyme Microb Technol 2007;41:5659. [22] Mateo C, Palomo JM, Fernandez-Lorente G, Guisan JM, Fernandez-Lafuente R. Improvement of enzyme activity, stability and selectivity via immobilization technique. Enzyme Microb Technol 2007;40:145163. [23] Palocci C, Chronopoulou L, Venditti I, Cernia E, Diociaiuti M, Fratoddi I, et al. Lipolytic enzymes with improved activity and selectivity upon adsorption on polymeric nanoparticles. Biomacromolecules 2007;8:304753. [24] Takac S, Bakkal M. Impressive effect of immobilization conditions on the catalytic activity and enantioselectivity of Candida rugosa lipase toward Snaproxen production. Proc Biochem 2007;42:10217. [25] Koszelewski D, Redzej A, Ostaszewski R. The study on efcient hydrolases immobilization for the kinetic resolution of the a-acetoxyamides. J Mol Catal B Enzym 2007;47:517. [26] Chaubey A, Parshad R, Koul S, Taneja SC, Qazi GN. Enantioselectivity modulation through immobilization of Arthrobacter sp. lipase: kinetic resolution of uoxetine intermediate. J Mol Catal B Enzym 2006;42:3944. [27] Sabbani S, Hedenstrom E, Nordin O. The enantioselectivity of Candida rugosa lipase is inuenced by the particle size of the immobilising support material Accurel. J Mol Catal B Enzym 2006;42:19. [28] Palomo JM, Filice M, Fernandez-Lafuente R, Terreni M, Guisan JM. Regioselective hydrolysis of peracetylated b-monosaccharides by immobilized lipases. Key role of the immobilization protocol. Adv Synth Catal 2007;349:196976. [29] Filice M, Fernandez-Lafuente R, Terreni M, Guisan JM, Palomo JM. Screening of lipases for regioselective hydrolysis of peracetylated-b-monosaccharides. J Mol Cat B Enzym 2007;49:127. [30] Medin A. Studies on structures and properties of agarose. Doctoral Dissertation Uppsala, Sweden, Uppsala University 1995. [31] Segura RL, Betancor L, Palomo JM, Hidalgo A, Fernandez-Lorente G, Terreni M, et al. Purication and identication of different lipases contained in PPL commercial extracts: a minor contaminant is the main responsible of most esterasic activity. Enzyme Microb Technol 2006;39:81723. [32] Anderson EM, Larsson KM, Kirk O. One biocatalystmany applications: the use of Candida antarctica B-lipase in organic synthesis. Biocatal Biotransform 1998;16:181204. [33] Gotor-Fernandez V, Busto E, Gotor V. Candida antarctica lipase B: an ideal biocatalyst for the preparation of nitrogenated organic compounds. Adv Synth Catal 2006;348:797812. [34] Misiunas A, Talaikyte Z, Niaura G, Razumas V, Nylander T. Thermomyces lanuginosus lipase in the liquid-crystalline phases of aqueous phytantriol: X-ray diffraction and vibrational spectroscopic studies. Biophys Chem 2008;134:14456. [35] Bojsen K, Svendsen A, Fuglsang CC, Patear S, Borch K, Vind J, et al. Novozymes A/S, Denmark. PCT International application WO2000/32758, 2000.

immobilized enzymes become adsorbed on the support via interfacial activation, the results suggest that the close contact between the support and the active center of the enzyme may fully alter the enzyme properties. Thus, octyl-agarose seemed to be the support that permitted the higher activity using simple and small substrates [21]. This could be due to the internal morphology of the support; agarose is formed by moderately thick trunks of agarose polymers [30] that reduce the geometrical congruence between the enzyme active center and the support, and therefore reducing the blockage of the active center of the enzyme after enzyme adsorption by the surroundings of the active center. Toyopearl is formed by ne bers that permit a higher geometrical congruence between the pocket formed by active center surroundings and the support, promoting a more signicant blockage of the enzyme active center by the support [21]. This could explain also that lipases may become more strongly adsorbed on these supports than in octylagarose, but simultaneously butyl-toyopearl seems to be the most selective for lipase adsorption. Finally, the use of different supports to interfacially activate the lipases permits to change the specicity and enantiospecicity of the lipases, in some cases in a very signicant way. For example, BTL may be fully enantioselective in the hydrolysis of ()-2-Obutyryl-2-phenylacetic acid hydrolyzing the S-isomer when it was adsorbed on octyl-agarose, while the E value greatly decreased when it was adsorbed on the other supports utilized in this study and previously described. In the asymmetric hydrolysis of phenylglutaric acid diethyl diester, BTL immobilized on hexyl-toyopearl was the most enantioselective catalyst with ee > 99% (A factor >100) in the production of S-monoester product, whereas the enzyme immobilized on butyl-toyopearl only exhibited an A factor of 3. Therefore, even though the hydrophobic supports are able to immobilize the open form of the lipases, depending on the support nature, the nal features of the immobilized enzyme may be fully different. That way, the selection of the support may be a critical step in the design of a lipase biocatalyst. Acknowledgments The authors gratefully recognize the support from the Spanish CICYT (project. BIO-2005-8576). We gratefully recognize CSIC by I3P contracts for Dr. Palomo (founded with FEDER founds), AECI for a fellowship for Ms. Cabrera and MEC for a Ph.D. fellowship for Mr. Godoy and for a RyC contract for Dr. Fernandez-Lorente. The author would also like to thank Mrs. Alicia Palomo D.P.S.I.C (ofcial translator and proof-reader) for the English proof-reading of the manuscript. The help and suggestions from Dr. Angel Berenguer (University of Cambridge) are gratefully acknowledged. References
[1] Villeneuve P. Lipases in lyophilisation reactions. Biotechnol Adv 2007;25:515 36. [2] Tanaka M, Demizu Y, Nagano M, Hama M, Yoshida Y, Kurihara M, et al. Lipasecatalyzed kinetic resolution of cyclic trans-1,2-diols bearing a diester moiety: synthetic application to chiral seven-membered-ring a,a-disubstituted aamino acid. J Org Chem 2007;72:77506. [3] Van Buijtenen J, Van As BAC, Verbruggen M, Roumen L, Vekemans JAJM, Pieterse K, et al. Switching from S- to R-selectivity in the Candida antarctica lipase B-catalyzed ring-opening of v-methylated lactones: tuning polymerizations by ring size. J Am Chem Soc 2007;129:73938. [4] Reetz MT. Lipases as practical biocatalysts. Curr Opin Chem Biol 2002;6:145 50. [5] Schmid RD, Verger R. Lipases: interfacial enzymes with attractive applications. Angew Chem Int Ed 1998;37:160933. [6] Yasmin T, Jiang T, Han B. Novozym 435 catalyzed regioselective acylation of ethane-1,2-diol in the presence of ionic liquids. Catal Lett 2007;116:469. [7] Akai S, Kita Y. Recent progress on the lipase-catalyzed asymmetric syntheses. J Synth Org Chem 2007;65:77282.

G. Fernandez-Lorente et al. / Process Biochemistry 43 (2008) 10611067 [36] Fernandez-Lorente G, Palomo JM, Guisan JM, Fernandez-Lafuente R. Effect of the immobilization protocol in the activity, stability, and enantioslectivity of Lecitase1 Ultra. J Mol Catal B Enzym 2007;47:99104. [37] Fernandez-Lorente G, Filice M, Terreni M, Guisan JM, Fernandez-Lafuente R, Palomo JM. Lecitase1 ultra as regioselective biocatalyst in the hydrolysis of fully protected carbohydrates. Strong modulation by using different immobilization protocols. J Mol Catal B Enzym 2008;51:1107. [38] Palomo JM, Fernandez-Lorente G, Guisan JM, Fernandez-Lafuente R. Modulation of immobilized lipase enantioselectivity via chemical amination. Adv Synth Catal 2007;349:111927.

1067

[39] Fernandez-Lorente G, Palomo JM, Cabrera Z, Fernandez-Lafuente R, Guisan JM. Improved catalytic properties of immobilized lipases by the presence of very low concentrations of detergents in the reaction medium. Biotechnol Bioeng 2007;97:24250. [40] Fryszkowska A, Komar M, Koszelewski D, Ostaszewski R. Tetrahedron Asymmetry 2005;16:247585. [41] Palomo JM, Segura RL, Fernandez-Lorente G, Pernas M, Rua ML, Guisan JM, et al. Purication, immobilization, and stabilization of a lipase from Bacillus thermocatenulatus by interfacial adsorption on hydrophobic supports. Biotechnol Prog 2004;20:6305.