Cg

GI

a

ARRAA

KELGSEA

1

wbs[dempcatmascpiaaep

0d

Colloids and Surfaces B: Biointerfaces 78 (2010) 351–356

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces

journa l homepage: www.e lsev ier .com/ locate /co lsur fb

omparative study of properties of immobilized lipase ontolutaraldehyde-activated amino-silica gel via different methods

uang Yang, Jianping Wu, Gang Xu, Lirong Yang ∗

nstitute of Bioengineering, College of Material Science and Chemical Engineering, Zhejiang University, Hangzhou 310027, China

r t i c l e i n f o

rticle history:eceived 4 December 2008eceived in revised form 15 March 2010ccepted 23 March 2010vailable online 27 March 2010

a b s t r a c t

The enzyme-aggregate coating method was performed to immobilize Arthrobacter sp. lipase in order toachieve better catalytic properties comparable to the conventional covalent attachment and covalentattachment plus cross-linking. The glutaraldehyde-activated amino-silica gel which was synthesized bysol–gel technique was used as the support, and the catalytic characteristics of the lipase preparationswere tested in the asymmetric acylation of 4-hydroxy-3-methyl-2-(2-propenyl)-2-cyclopenten-1-one

eywords:nzyme-aggregate coatingipaselutaraldehyde-activated amino-silica geltability

(HMPC) in organic solvents. The results showed that the immobilized lipase by enzyme-aggregate coatingpossessed both higher activity and stability than those by other methods, e.g. it obtained an activity of82.6 U/g and remained 42% and 93% of the original activity after incubation in vinyl acetate at 60 ◦C for16 h and 9 times recycles, respectively, while the covalently attached lipase got an activity of 67.4 U/gand left 33% and 73% of the original under the same conditions, and the enzyme prepared by covalent

kingall t

nantioselectivitysymmetric acylation

attachment plus cross-lin(E ≥ 400) was achieved by

. Introduction

Lipases (triacylglycerol ester hydrolase, E.C. 3.1.1.3) have beenidely applied as versatile biocatalysts in biotransformation

ecause of their excellent substrate specificity, region- and enantio-electivity [1–3]. They can catalyze esterification [4], hydrolysis5], aminolysis [6] and transesterification [7] reactions to pro-uce many useful chemical products. However, like many othernzymes, lipases are usually limited in industrial applicationsainly because of their short lifetimes which might cause high

roduction cost. So improvement of the stability of enzyme is cru-ial for its further practical applications. There have been manypproaches to improve the enzyme stability: enzyme immobiliza-ion [8], enzyme modification [9], protein engineering [10] and

edium engineering [11]. In most cases, especially for industrialpplications, enzymes are preferably used in their immobilizedtates owing to many advantages, such as simple recovery ofatalyst and products, ready reutilization of the catalyst andossibility of continuous operation. There are many kinds of

mmobilization techniques [12–15], among which, the covalent

ttachment between enzyme molecules and host materials is oftenccepted as an efficient method to achieve high stability of thenzyme by increasing the rigidity of its structure and reducingrotein unfolding. For example, �-chymotrypsin was efficiently

∗ Corresponding author. Tel.: +86 571 87952363; fax: +86 571 87952363.E-mail address: lryang@zju.edu.cn (L. Yang).

927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.colsurfb.2010.03.022

exhibited the lowest activity yield. Moreover, excellent enantioselectivityhe three prepared lipases in our paper (E = 85 for the free enzyme).

© 2010 Elsevier B.V. All rights reserved.

attached onto the nanoporous sol–gel glass via a bifunctional lig-and, trimethoxysilylpropanal, with a greatly enhanced half-lifeof more than 1000-fold higher than that of the native enzyme,both in aqueous solution and organic solvents [16]. Grazú usedthiol-functionalized epoxy supports to site-specific immobilizepenicillin G acylase and lipase from Rhizomucor miehei by cova-lent attachment and the results showed that the thermal stabilitiesof the immobilized enzymes, measured as half-lives, were 12–15folds of that of the free one, with good preservation of activ-ity (>65%) [17]. In recent years, the chemically inert inorganicoxide sol–gel materials have been focused as a new kind of hostcarriers for enzyme immobilization due to their special meritssuch as negligible swelling effects, controllable surface parameters,and high thermal and chemical stabilities. Most critically, surfacemodification with different organic functional groups can be eas-ily achieved by co-condensation of trialkoxyorganosilanes withtetraethylorthosilicate [18,19]. A kind of non-porous silica spheresprepared by co-condensation of (3-aminopropyl)trimethoxysilaneand tetraethylorthosilicate was synthesized to covalently immo-bilize glucose oxidase and the prepared enzyme obtained higherstability [20].

However, the maximal loading capacity of protein by cova-lent attachment is usually limited as a result of the monolayer

coverage of enzyme molecules. In recent years, a novel immo-bilization method enzyme-aggregate coating [21] has been putforward. It involves two steps: the ‘seed’ enzyme molecules are firstcovalently attached onto the support and then additional enzymemolecules and their aggregates in the solution are cross-linked by

352 G. Yang et al. / Colloids and Surfaces B: Biointerfaces 78 (2010) 351–356

PC w

gpaphac

cawlrtpaeaasanwwtdasocba

patcleota(

2

2

o((cRw(a

Protein was determined according to Bradford’s method [28]

Scheme 1. Kinetic resolution of HM

lutaraldehyde onto the ‘seed’ molecules. This resulting enzymereparation can be thought of as bunches of cross-linked enzymeggregates (CLEAs) [22] which are covalently attached to the sup-ort via the linking of each seed enzyme molecule. By this method,igher enzyme loading can be achieved and hence increased over-ll activity and even higher stability of the immobilized biocatalystompared to those only by covalent attachment.

Racemic compound of 4-hydroxy-3-methyl-2-(2-propenyl)-2-yclopenten-1-one (HMPC) [23] has been widely used as thelcoholic moiety of allethrin, a member of synthetic pyrethroidshich are a group of esters with splendid insecticidal activities and

ow toxicity to mammals. However, tested with structure–activityelationship, allethrin of S-HMPC moiety presents higher activityhan that of R-form by four folds. Therefore, preparation of opticallyure S-HMPC is essential for the improvement of the insecticidalctivity and the decrease of environmental contaminations. Dandat al. [24] prepared the S-HMPC product with an e.e. value of 78%nd yield of 80% by combination with the enzymatic hydrolysisnd stereochemical inversion. Wu et al. [25] reported a kind ofurfactant-modified lipase (Pseudomonas sp.) for the asymmetriccylation of HMPC and realized a high e.e. value of the S-formearly 100%. However, no study about the stabilities of the enzymeas reported therein. In our previous work, optically pure S-HMPCith an e.e. value above 99% was obtained by asymmetric acyla-

ion of HMPC via Arthrobacter sp. lipase. Additionally, the low-costiatomite granules suffering surface modification were adopteds the support to immobilize the lipase by different methods,uch as interfacial adsorption and enzyme-aggregate coating basedn adsorption. The results showed that the enzyme-aggregateoated lipase exhibited the highest activity and operational sta-ility among the preparations, with 85% of initial activity remainedfter 10 recycles [26].

In this paper, the glutaraldehyde-activated amino-silica gel wasrepared as the support to immobilize Arthrobacter sp. lipase withn aim to further improve its catalytic characteristics, especiallyhe stability. Various immobilization methods were tried includingovalent attachment, the enzyme-aggregate coating and cova-ent attachment plus cross-linking. The catalytic characteristics,.g. activity, enantioselectivity, thermal and operational stabilityf the immobilized lipases were assessed in the kinetic resolu-ion of HMPC by asymmetric acylation. Vinyl acetate was useds both the acyl donor and organic solvent in our experimentScheme 1).

. Experimental

.1. Materials

Lipase from Arthrobacter sp. was a gift from Institute of Microbi-logy, Chinese Academy of China (IMCAC). Bovine serum albuminBSA), glutaraldehyde (GA) solution of 25%, tetraethoxysiliconeTEOS) and �-(aminopropyl)triethoxy silane (APTES) were pur-

hased from Sigma. The SEPABEADS® EC-EP resin came fromesindion S.R.L., (Mitsubishi Chemical Corporation). rac-HMPCas gifted from Changzhou Kangmei Chemical Industry Co., Ltd

purity > 98% by gas chromatograph). All other chemicals were ofnalytical grade and commercial available.

ith lipase by asymmetric acylation.

2.2. Synthesis of amino-silica gel

Amino-silica gel was synthesized by sol–gel technique accord-ing to Ref. [27]. First, a mixture of 1.0 ml of APTES and 0.5 ml ofTEOS was added into a 50-ml centrifuge vial. Then 0.4 ml of deion-ized water and 30 �l of 1.0 M HCl solution were slowly addedunder strong agitation at room temperature. After that the mixturewas stirred continuously for 6 h and then aged at room temper-ature overnight. Finally the wet gel was dried at 50 ◦C for 3 daysand mechanically broken into particles with a diameter of about100 mesh. The gel particles were first washed with ethanol, deion-ized water for several times and then dried under vacuum.

2.3. Glutaraldehyde activation of amino-silica gel

The amino-silica gel particles were suspended in 10% glutaralde-hyde solution at a ratio of 35 ml of solution per gram gel particles.The suspension was stirred at room temperature for 24 h followedby vacuum filtration. Then the solids were washed extensively withdeionized water to remove glutaraldehyde and dried under vac-uum. By this way GA-activated silica gel was obtained.

2.4. Immobilization of lipase

Immobilization of lipase onto GA-activated silica gel by enzyme-aggregate coating was prepared according to Ref. [21] with somemodifications. 20 mg of GA-activated silica gel were weighed into a50-ml centrifuge vial. Then 4 ml of enzyme solution (pH 8.0, 0.03 Mphosphate buffer) containing some amount of lipase powders wasadded into the vial and the mixture was vigorously stirred at 0 ◦C for24 h. Thereafter 1 ml of a certain concentration of GA solution wasintroduced into the mixture (e.g. the final concentration of GA solu-tion was 0.2% when 1% of GA was added) and stirred continuouslyfor 10 h at 0 ◦C. After that the solids were collected by centrifu-gation, washed with at least 5× 5 ml of pH 8.0, 0.03 M phosphatebuffer for 30 min.

The sample by covalent attachment plus cross-linking wasprepared with the same mixture in enzyme-aggregate coatingmethod, and the mixture was vigorously stirred at 0 ◦C for 24 h.After that, the solids were filtered and treated with 0.4 ml of0.2% GA solution for 2 h. Finally, the solids were filtered out andtreated in the same procedure as the enzyme-aggregate coatingmethod.

The sample by covalent attachment was prepared as that byenzyme-aggregate coating but without GA treatment.

All the immobilized lipases were dried over silica gel under vac-uum for 24 h and then stored in closed vessels over silica gel at4 ◦C.

2.5. Protein assay

using BSA as a standard. The amount of bound protein was deter-mined indirectly by comparing the difference between the amountof protein introduced into the supports and the amount of proteinboth in the filtrate and in the washing solutions after immobiliza-tion.

G. Yang et al. / Colloids and Surfaces B: Biointerfaces 78 (2010) 351–356 353

GA-ac

2

amaI1tiwiaiiosccwoe

2

maswatc

2

saaru

2

Gws(a

3

3

w

from 85 to above 400 and e.e. value of S-HMPC above 99%. The cova-lent attachment plus cross-linking method resulted in lower totalactivity yield than that by covalent attachment alone, which was inagreement to other reports [30]. However, the enzyme-aggregate

Scheme 2. Preparation of

.6. Assay of lipase activity and enantioselectivity

The asymmetric acylation of HMPC was used to determine thectivity and enantioselectivity of the prepared lipases. All the com-ercial organic solvents and rac-HMPC were dehydrated with

nhydrous CaCl2 at room temperature for 2–3 days before used.n a general experiment, a mixture of 3 ml vinyl acetate containing.8 mmol HMPC was placed into a 5-ml screw-capped vial. Thenhe reaction was started by adding 25 mg of free lipase or 30 mg ofmmobilized lipase and carried out at 30 ◦C and 220 rpm. Aliquots

ere withdrawn from the reaction mixture at different reactingntervals and analyzed by gas chromatography (carrier gas: N2 at

flow rate 50 ml/min; column: 120 ◦C, FID detector: 230 ◦C andnjector: 230 ◦C) equipped with a chiral column (WCOT fused sil-ca column with CP Chirasil Dex CB, 25 m × 0.32 mm). One unitf enzyme activity was defined as the amount of enzyme neces-ary to produce 1 �mol of HMPC acetate per min under the assayonditions. Enantioselectivity was expressed as E value and cal-ulated as follows: E = ln [1 − C(1 + e.e.p)]/ln[1 − C(1 − e.e.p)] [29],here C = e.e.s/(e.e.s + e.e.p); C, e.e.s and e.e.p denoted conversion

f substrate, enantiomeric excess of substrate and enantiomericxcess of product, respectively.

.7. Determination of thermal stability

The thermal stability of the free or immobilized lipase was deter-ined as follows: lipases were incubated in 3 ml of pure vinyl

cetate at 60 ◦C for 6 and 16 h, respectively. After incubation, theolvent was removed carefully, and the fresh reaction solutionhich was incubated at 30 ◦C for 10 min was added. The residual

ctivity was determined under the same reaction conditions men-ioned above. The activity of the enzyme without incubation wasonsidered as the control (100%).

.8. Determination of operational stability

A batch-wise fashion was adopted to investigate the operationaltability of the immobilized lipase. The activity was determinedccording to the above method. Each reaction was terminated atpproximately 50% conversation. Then the immobilized lipase wasecovered, washed by 3× 5 ml of vinyl acetate and subsequentlysed in the next reaction.

.9. Characterization

Specimens of the amino-silica gel, GA-activated silica gel andA-activated silica gel after enzyme immobilization were analyzedith Fourier transform infrared spectroscopy (FTIR). The infrared

pectra were obtained using a NICOLET 5700 infrared spectrometerThermo Nicolet, USA) to analyze pellets formed with mixed KBrnd solid samples (1% wt.).

. Results and discussion

.1. Characterization of support and immobilized lipase

The procedure of the preparation of GA-activated silica gelas illustrated in Scheme 2. The infrared spectra of amino-

tivated silica gel support.

silica gel, GA-activated silica gel and GA-activated silica gel afterenzyme immobilization were shown in Fig. 1. As expected, theC–H stretching vibration frequency at near 2930 cm−1 was seenin all spectra, displaying contributions from the organosilane, glu-taraldehyde and enzyme. Furthermore, the corresponding simplebending vibrations were seen between 1500 and 1300 cm−1. Thepresence of free aldehyde groups (at 1713 cm−1) was identified inthe spectrum of GA-activated silica gel but not in other spectra,indicating a successful activation of amino groups on amino-silicagel by glutaraldehyde and consumption of the groups duringlipase immobilization. A peak at 1647 cm−1, seen in the spectraof GA-activated silica gel and that after enzyme immobilization,presumably represented the imine (N C) Schiff-base produced bythe amine-glutaraldehyde reaction. The strong bands appear inthe range of 1250–1000 cm−1 related to the Si–O–Si asymmet-ric stretching vibration and the Si–OH bending bands appear at925 cm−1.

3.2. Immobilization of lipase by enzyme-aggregate coating

In this paper, three different methods were tried to immo-bilize Arthrobacter sp. lipase in order to improve its catalyticproperties especially enantioselectivity and stability. They werecovalent attachment, enzyme-aggregate coating and covalentattachment plus cross-linking. Some literature reported that thepost-treatment by cross-linking after covalent attachment couldfurther improve the stability of the enzyme comparable to thatby covalent attachment alone [30]. However, the higher stabil-ity was achieved at the cost of much lower activity yield. Hence,in order to obtain both higher stability and activity, the enzyme-aggregate coating method was tried. The characteristics of the threelipase preparations were exhibited in Table 1. It was found thatthe enantioselectivity of all the three enzymes after immobilizationimproved much compared to that of the free one with an E value

Fig. 1. FTIR analysis of (a) amino-silica gel; (b) GA-activated silica gel and (c) GA-activated silica gel after enzyme immobilization.

354 G. Yang et al. / Colloids and Surfaces B: Biointerfaces 78 (2010) 351–356

Fpt

ca6imcsnthnla

3

caotawtptbwatttm

TI

m

ig. 2. Thermal stability and enantioselectivity of the free and immobilized lipasesrepared by different immobilization methods. The immobilization conditions ofhe prepared lipases were the same as those in Table 1.

oated lipase exhibited higher bound protein than the covalentlyttached one, together with an increase of activity by 15.9% (from7.4 to 78.1 U/g). This meant that additional enzyme molecules

n solution were indeed linked onto the support via the ‘seed’olecules. The specific activity of the lipase by enzyme-aggregate

oating was lower than that by covalent attachment. The main rea-ons for the phenomena might comprise two aspects: firstly, theonselective intermolecular cross-linkage by GA might happen inhe functional domain of the enzyme molecular, which resulted inarmful enzyme conformation changes; secondly, the cross-linkedetwork offered by GA molecules might bring about diffusional

imitations and prevent the reactant molecules from reaching thective sites in the enzymes.

.3. Thermal and operational stability of immobilized lipases

Besides activity and enantioselectivity, the stability was also arucial standard for the immobilized enzyme, especially in practicalpplications. We investigated the thermal and operational stabilityf the prepared lipases in Table 1 and compared them with those ofhe free form. Shown in Fig. 2 were the thermal stabilities of the freend immobilized lipases. Obviously, improved thermal resistanceas achieved by the enzymes after immobilization. Among them,

he lipases by enzyme-aggregate coating and covalent attachmentlus cross-linking remained higher activity (42% residual activity)han the covalently attached one (33% residual activity) after incu-ation for 16 h. Moreover, the enantioselectivity of the free enzymeas found decreased after the thermal treatment for 16 h, with

n E value from 85 to 56. However, almost no decreases of enan-ioselectivity were observed for the immobilized lipases by all thehree methods with the E value of above 400. This illustrated thathe bonding of enzyme to the support was favorable to keep the

olecule structure stable.

able 1mmobilization of lipase by different methods.

Entry Lipase derivativesa Bound protein (mg/gimmobilized lipase)

Activity

1 EAC-lipase 8.7 ± 0.5 78.1 ± 32 CL-lipase 7.1 ± 0.2 52.3 ± 13 CA-lipase 7.1 ± 0.4 67.4 ± 24 Free lipase – 187 ± 9

a The concentration of GA solution and lipase loading were 0.2% and 0.5 mg/mg support,ethod; CL-lipase meant the lipase preparation by covalent attachment plus cross-linkinb Total activity yield (%) = (the total activity of the immobilized enzymes/the total activ

Fig. 3. Operational stability of the free and immobilized lipases prepared by dif-ferent immobilization methods (every cycle was about 24 h). The immobilizationconditions of the prepared lipases were the same as those in Table 1.

In Fig. 3, significant improvement of operational stability couldbe observed by the immobilized enzymes compared to the freeone. Moreover, the lipase immobilized by enzyme-aggregate coat-ing and covalent attachment plus cross-linking retained 93% of theoriginal activity after nine times recycles (every cycle was about24 h), which was higher than that by covalent attachment (73%residual activity).

Hence, the enzyme-aggregate coating method was thoughtbetter than other immobilization methods comprehensively con-sidering the total activity yield and thermal and operationalstabilities.

3.4. Effect of immobilization condition on enzyme-aggregatecoating preparation

In order to further investigate the enzyme-aggregate coatingmethod, the influences of lipase loading and the concentrationof GA solution were studied. It was seen in Fig. 4 that both thebound protein and activity increased clearly with the rising con-centration of GA solution at below 0.2%, then the bound proteinascended slowly and activity descended with the further increase.With respect to the specific activity, the higher GA concentrationwas, the lower values were obtained. This illustrated that higherGA concentration was beneficial to increase the protein loading.However, the activity was a compromise between several fac-tors. When the GA concentration was lower, the activity increasedmainly resulted from the clear increase of the bound protein. How-ever, at higher GA concentration the nonselective cross-linkage by

GA might result in harmful conformation changes of the enzymeand also prevent the reactants from reaching the active sites in theenzymes due to diffusional limitations offered by the cross-linkednetwork.

(U/g) Specific activity(×104 U/g protein)

Total activity yieldb (%) E

.6 0.90 83.5 ≥400

.9 0.74 55.95 ≥400

.7 0.95 72.1 ≥4001.0 – 85

respectively. EAC-lipase meant the lipase preparation by enzyme-aggregate coatingg and CA-lipase meant the lipase preparation by covalent attachment alone.ity of the free enzymes before immobilization) × 100%.

G. Yang et al. / Colloids and Surfaces B:

Fia

osHmoov

cbwtsstpttt0atctTcs7

Fls

ig. 4. Effect of concentration of GA solution on activity and bound protein of themmobilized lipase. (�) Represents bound protein; (�) represents specific activitynd (�) represents activity. The lipase loading was 1.0 mg/mg support.

As shown in Fig. 5, the bound protein and activity increasedbviously when the lipase loading ascended from 0.5 to 1.5 mg/mgupport and then kept almost the same level with a further increase.igher lipase loading led to decreased specific activity, and thisight mainly result from the steric hindrance effect. No influences

f lipase loading and the concentration of GA solution were noticedn the enantioselectivity of the immobilized lipases and a high Ealue of above 400 was obtained.

We also investigated the influences of lipase loading and theoncentration of GA solution on the thermal and operational sta-ility of the lipase prepared by enzyme-aggregate coating (the dataere not shown here). The results showed that higher concentra-

ion of GA solution produced a bit higher thermal and operationaltability of the lipase preparation. This might attribute to thetronger cross-linking effect by GA which made the molecule struc-ure of enzyme harder to be unfolded. In addition, the enzymereparation with higher lipase loading seemed a little unstableo thermal incubation and its operational stability was also rela-ive poor comparable to that with lower enzyme loading. Amonghem, the preparation of 1.0 mg/mg support of lipase loading and.5% of the GA concentration was found almost no loss of activityfter nine recycles (every cycle was 24 h). This result was betterhan some reported ones, e.g. Chaubey et al. [31] reported theovalent immobilization of Arthrobacter sp. lipase on aminopropyl-riethoxy silane-activated silica gel and CNBr-activated sepharose.

he results showed that 87% residual activity was remained for 10ycles, but every cycle was only 30 min. Bhushan et al. [32] synthe-ized two kinds of alkylated glycidyl epoxy polymers GMA-EGDM5-20(I) and GMA-EGDM 75-30(I), and the covalently immobilized

ig. 5. Effect of lipase loading on activity and bound protein of the immobilizedipase. (�) represents bound protein; (�) represents specific activity and (�) repre-ents activity. The concentration of GA solution was 0.2%.

[[[[[[

[

[

[[

[

[

[[

Biointerfaces 78 (2010) 351–356 355

Arthrobacter sp. lipase on these supports kept above 95% residualactivity after 10 cycles and each cycle was 30 min. Additionally,we also tried to immobilize the lipase onto the commercial EC-EPepoxy resin (the procedure were the same with GA-activated silicagel by covalent attachment), and the results exhibited that the ther-mal and operational stability of the enzyme preparation was not asgood as those of the enzyme-aggregate coated one on GA-activatedsilica gel.

4. Conclusion

Lipase from Arthrobacter sp. was immobilized on GA-activatedsilica gel which was prepared in this work for asymmetric acylationof HMPC in organic solvents. Three different immobilization meth-ods were performed in order to obtain better catalytic properties ofenzyme, including covalent attachment, covalent attachment pluscross-linking and enzyme-aggregate coating method. The resultsshowed that excellent enantioselectivity (E ≥ 400, with e.e. = 99%of S-HMPC) was achieved by all the three immobilization methodsin our work (E = 85 for the free enzyme). The lipase immobilizedby enzyme-aggregate coating obtained both higher activity andstability (e.g. thermal and operational stability) than those bycovalent attachment alone and covalent attachment plus cross-linking. Among them, the enzyme-aggregate coated preparation of1.0 mg/mg support of lipase loading and 0.5% of the GA concentra-tion was found almost no loss of activity after nine recycles (everycycle was 24 h). Hence, we thought that the enzyme-aggregatecoating method was prospective to obtain much more stableenzyme preparations even with higher activity than the conven-tional immobilization method.

Acknowledgements

The authors would like to thank the Chinese National NaturalScience Foundation (Nos. 20606030 and 20336010), Key Project ofChinese National Programs for Fundamental Research and Develop-ment (No. 2003CB716008) and Hi-Tech Research and DevelopmentProgram of China (No. 2006AA02Z238) for financial support.

References

[1] K.M. Koeller, C.H. Wong, Nature 409 (2001) 232–240.[2] M.T. Reetz, Curr. Opin. Chem. Biol. 6 (2002) 145–150.[3] V. Gotor-Fernández, R. Brieva, V. Gotor, J. Mol. Catal. B: Enzyme 40 (2006)

111–120.[4] A. Sharma, S. Chattopadhyay, J. Mol. Catal. B: Enzyme 10 (2000) 531–534.[5] T. Miura, S. Umetsu, D. Kanamori, N. Tsuyama, Y. Jyo, Y. Kawashima, N. Koyata,

Y. Murakami, N. Imai, Tetrahedron 64 (2008) 9305–9308.[6] C. Pilissäo, M.G. Nascimento, Tetrahedron: Asymmetry 17 (2006) 428–433.[7] S. Shah, M.N. Gupta, Process Biochem. 42 (2007) 409–414.[8] S. Pahujani, S.S. Kanwar, G. Chauhan, R. Gupta, Bioresour. Technol. 99 (2008)

2566–2570.[9] A. Slavica, I. Dib, B. Nidetzky, Biotechnol. Bioeng. 96 (2007) 9–17.10] R.J. Sola, W. Al-Azzam, K. Griebenow, Biotechnol. Bioeng. 94 (2006) 1072–1079.11] S.C. Mohapatra, J.T. Hsu, Biotechnol. Bioeng. 64 (1999) 213–220.12] P. Ye, J. Jiang, Z.K. Xu, Colloids Surf. B: Biointerfaces 60 (2007) 62–67.13] Y.X. Bai, Y.F. Li, Y. Yang, L.X. Yi, J. Biotechnol. 125 (2006) 574–582.14] O. Orcaire, P. Buisson, A.C. Pierre, J. Mol. Catal. B: Enzyme 42 (2006) 106–113.15] H.W. Yu, H. Chen, X. Wang, Y.Y. Yang, C.B. Ching, J. Mol. Catal. B: Enzyme 43

(2006) 124–127.16] P. Wang, S. Dai, S.D. Waezsada, A.Y. Tsao, B.H. Davison, Biotechnol. Bioeng. 74

(2001) 249–255.17] V. Grazú, O. Abian, C. Mateo, F. Batista-Viera, R. Fernandez-Lafuente, J.M. Guisan,

Biotechnol. Bioeng. 90 (2005) 597–605.18] M.T. Reetz, A. Zonta, J. Simpelkamp, Biotechnol. Bioeng. 49 (1996) 527–534.19] C.M.F. Soares, O.A. Santos, J.E. Olivo, H.F. Castro, F.F. Moraes, G.M. Zanin, J. Mol.

Catal. B: Enzyme 29 (2004) 69–79.

20] C. Oh, J.H. Lee, Y.G. Lee, Y.H. Lee, J.W. Kim, H.H. Kang, S.G. Oh, Colloid Surf. B:

Biointerfaces 53 (2006) 225–232.21] B.C. Kim, S. Nair, J. Kim, J.H. Kwak, J.W. Grate, S.H. Kim, M.B. Gu, Nanotechnology

16 (2005) S382–S388.22] A.L. Cao, F. Rantwijk, R.A. Sheldon, Org. Lett. 2 (2000) 1361–1364.23] S. Kenji, Y. Hiroshi, M. Masayoshi, European Patent, EP 0039089A2, 1981.11.04.

3 ces B:

[

[[[

[

[

56 G. Yang et al. / Colloids and Surfa

24] H. Danda, T. Nagatomi, A. Maehara, T. Umemura, Tetrahedron 47 (1991)

8701–8716.

25] H.Y. Wu, J.H. Xu, S.F. Tsang, Enzyme Microb. Technol. 34 (2004) 523–528.26] G. Yang, J.P. Wu, G. Xu, L.R. Yang, J. Mol. Catal. B: Enzyme 57 (2009) 96–103.27] A.E. David, N.S. Wang, V.C. Yang, A.J. Yang, J. Biotechnol. 125 (2006) 395–

407.28] M.M. Bradford, Anal. Biochem. 72 (1976) 248–255.

[[

[

Biointerfaces 78 (2010) 351–356

29] C.S. Chen, Y. Fujimoto, G. Girdaukas, C.J. Sih, J. Am. Chem. Soc. 104 (1982)

7294–7299.

30] B.K. Vaidya, R.S. Singhal, Colloids Surf. B: Biointerfaces 61 (2008) 101–105.31] A. Chaubey, R. Parshad, S. Koul, S.C. Taneja, G.N. Qazi, Appl. Microbiol. Biotech-

nol. 73 (2006) 598–606.32] I. Bhushan, R. Parshad, G.N. Qazi, G. Ingavle, C.R. Rajan, S. Ponrathnam, V.K.

Gupta, Process Biochem. 43 (2008) 321–330.