Journal of Biotechnology 119 (2005) 70–75

Enzyme stabilization by glutaraldehyde crosslinking of adsorbedproteins on aminated supports

Fernando Lopez-Gallego, Lorena Betancor, Cesar Mateo, Aurelio Hidalgo,Noelia Alonso-Morales, Gisela Dellamora-Ortiz, Jose M. Guisan∗,

Roberto Fernandez-Lafuente∗

Departamento de Biocat´alisis, Instituto de Cat´alisis, CSIC, Campus Universidad Autonoma, Cantoblanco, 28049 Madrid, Spain

Received 22 November 2004; received in revised form 13 May 2005; accepted 24 May 2005

Abstract

The stabilization achieved by different immobilization protocols have been compared using three different enzymes (glutarylacylase (GAC),d-aminoacid oxidase (DAAO), and glucose oxidase (GOX)): adsorption on aminated supports, treatment of thisadsorbed enzymes with glutaraldehyde, and immobilization on glutaraldehyde pre-activated supports. In all cases, the treatmentof adsorbed enzymes on amino-supports with glutaraldehyde yielded the higher stabilizations: in the case of GOX, a stabilizationover 400-fold was achieved. After this treatment, the enzymes could no longer be desorbed from the supports using high ionicstrength (suggesting the support-protein reaction). Modification of the enzymes immobilized on supports that did not offer thep g that them od resultsi aldehydep©

K

1

h

f

(

1;04;eis-stelaofaryl.,

o-pid

0

ossibility of react with glutaraldehyde showed the same stability that the non modified preparations demonstratinere chemical modification did not have effect on the enzyme stability. This simple strategy seems to permit very go

n terms of immobilization rate and stability, offering some advantages when compared to the immobilization on glutarre-activated supports.2005 Elsevier B.V. All rights reserved.

eywords:Glucose oxidase; Glutaryl acylase;d-Amino acid oxidase; Glutaraldehyde; Enzyme stabilization; Enzyme immobilization

. Introduction

There are several possibilities of using glutaralde-yde for the immobilization of proteins: the use of

∗ Corresponding authors. Tel.: +34 91 585 48 09;ax: +34 91 585 47 60.E-mail addresses:jmguisan@icp.csic

J.M. Guisan), rfl@icp.csic.es (R. Fernandez-Lafuente).

supports previously activated (Zhou and Chen, 200Magnan et al., 2004; Seyhan and Alptekin, 20Burteau et al., 1989; Van Aken et al., 2000; Dos RCosta et al., 2003; Barros Rui et al., 2003; Lamas Eet al., 2001) or the treatment with glutaraldehydepreviously adsorbed proteins on supports with primamino groups (D’Souza and Kubal, 2002; Hwang et a2004). From an immobilization point of view, both prtocols seem to follow a similar mechanism: a first ra

168-1656/$ – see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.jbiotec.2005.05.021

F. Lopez-Gallego et al. / Journal of Biotechnology 119 (2005) 70–75 71

protein adsorption by ionic exchange, followed by thecovalent reaction (Balcao et al., 2001; Mieglo et al.,2003). However, the first strategy reduces the chemicalmodification of the enzyme to only the groups of theprotein that are involved in the immobilization, while inthe second case all the protein surface may be modified(Fernandez-Lafuente et al., 1995).

When the immobilization is carried out on pre-activated supports, the primary amino groups of theenzyme would react with the aldehyde groups thathave been introduced by modification of the aminogroups of the support (usually with two glutaralde-hyde molecules) (Monsan, 1978; Guisan et al., 1997).The usual pH to perform the immobilization on thissupports is around 7–8.5, because of the low stabilityof glutaraldehyde activated supports at high pH val-ues. At these pH values, the reactivity of�-amino ofLys groups, may be expected to be quite low, and theintensity of the multipoint covalent attachment may benot very high. On the other hand, when the enzyme isfirstly adsorbed on the support, and then treated withglutaraldehyde under mild conditions, all the primaryamino groups of the enzyme and support may be acti-vated with one molecule of glutaraldehyde. This maypermit to have an intense crosslinking under a broadreaction conditions (Fernandez-Lafuente et al., 1995).In this form, the glutaraldehyde molecule bound to the�-amino groups of lysines of the enzyme could cova-lently react with the glutaraldehyde molecule bond tothe primary amino groups of the support establishing am

ilizes lase( e(G

2

2

inod ta .A.( nica l 7-A A.

(Murcia, Spain).d-Aminoacid oxidase (DAAO) fromTrigonopsis variabiliswas purchased from Recor-dati S.A. Glucose oxidase (GOX) fromAspergillusniger, horseradish peroxidase,o-phenylendiamine,2,2′-azino-bis(3-ethylbenzathiazoline)-6-sulfonic acid(ABTS), and ethylenediamine were purchased fromSigma (St. Louis, MO, USA). Glucose was from Pan-reac (Barcelona, Spain). All other reagents were ofanalytical grade.

2.2. Methods

All experiments were carried out in triplicate and theresults are presented as its mean value. Experimentalerror never exceeded 5%.

2.3. Preparation of glutaryl 7-ACA acylase

One milliliter of the commercial preparation of GACwas diluted with 4 mL of 25 mM potassium phosphatebuffer pH 7 and dialyzed against 5 L of 5 mM potas-sium phosphate buffer pH 7. Then the dialyzed enzymewas centrifuged at 12,000 rpm, for 30 min at 4◦C, andthe supernatant (containing 16 IU/mL (see below) and11 mg protein/mL) was used as the enzymatic prepa-ration for further experiments. Over 90% of the initialactivity was recovered after this procedure.

2.4. Determination of enzyme activity

2ed

u idr ysw blee laset s-s -i tionw

untot on-d

22 e( c-

ulti-point covalent enzyme-support attachment.Here, both strategies will be compared to stab

everal industrially relevant enzymes: glutaryl acyGAC) (Fechtig et al., 1968), d-aminoacid oxidasDAAO) (Conlon et al., 1995; Fisher et al., 1995), andlucose oxidase (GOX) (Wilson and Turner, 1992).

. Materials and methods

.1. Materials

Crosslinked 4% agarose beads and their amerivatives, MANAE-agarose (Fernandez-Lafuente el., 1993), were kindly donated by Hispanagar SBurgos, Spain). Glutaryl 7-aminocephalosporacid acylase (GAC), cephalosporin C, and glutaryCA were kindly donated by Bioferma Murcia S.

.4.1. Assay of glutaryl 7-ACA acylase activityGlutaryl acylase (GAC) activity was determin

sing a pH-STAT, by titration of the glutaric aceleased in the hydrolysis of glutaryl 7-ACA. Assaere carried out by the addition of 1–2 IU of solunzyme or suspension of immobilized glutaryl acy

o 10 mL of 10 mM glutaryl 7-ACA in 100 mM potaium phosphate buffer pH 7.5 at 25◦C, in a mechancally stirred vessel. The acid released in the reacas titrated with 25 mM NaOH.One unit of GAC activity was defined as the amo

f enzyme that is necessary to produce 1�mol of glu-aric acid per minute in the previously described citions.

.4.2. Assay ofd-amino acid oxidase activity

.4.2.1. Soluble enzyme.d-Amino acid oxidasDAAO) activity in soluble form was analyzed spe

72 F. Lopez-Gallego et al. / Journal of Biotechnology 119 (2005) 70–75

trophotometrically using cephalosporin C as substratemeasuring the increment of the absorbance at 445 nmpromoted by coupling the oxidative deamination ofthe substrate with the reaction between hydrogenperoxide ando-phenylendiamine catalyzed by perox-idase (Aebi, 1981). The reaction mixture consistedof 1.5 mL of 25 mg/mL cephalosporin C solution in50 mM potassium phosphate at pH 7.5, 0.5 mL of1.85 mM o-phenylendiamine in distilled water and0.1 mL of a 1 mg/mL peroxidase solution in 50 mMpotassium phosphate at pH 7.5. The reagents werekept at 4◦C and pre-incubated at 25◦C. The reactionwas initiated by adding a maximum of 0.1 DAAOunits to the reaction mixture.

One DAAO unit was defined as the amount ofenzyme able to oxidize 1�mol of cephalosporin Cper minute under the previously described assay con-ditions.

2.4.2.2. Immobilized enzyme.DAAO activity inimmobilized form was analyzed by HPLC usingcephalosporin C as substrate, measuring the dis-appearance of cephalosporin C along time. Thereaction mixture consisted of 1.5 mL of 25 mg/mLcephalosporin C solution in 50 mM potassium phos-phate at pH 7.5. The reaction was initiated by addinga maximum of 0.3 units of immobilized DAAO to thereaction mixture. Samples were withdrawn at differenttimes and analyzed by HPLC (Kromasil C8 250 mm,mobile phase was 20 mM ammonium acetate at pH5

stc

2a-

s thea n ofAa bya obi-l ing1 .0,0 ono Lo mp

One unit of glucose oxidase activity was definedas the amount of enzyme that causes the oxidation of1�mol of ABTS per minute, under the specified con-ditions.

2.5. Protein determination

The amount of protein in the different samples wasdetermined according to Bradford method (Bradford,1976), using bovine serum albumin as standard.

2.6. Preparation of pre-activated agarose withglutaraldehyde

Glutaraldehyde agarose preactivated support wasprepared suspending 10 mL of MANAE-agarose(Fernandez-Lafuente et al., 1993) in 20 mL of 15%glutaraldehyde in 200 mM phosphate buffer pH 7.0.The suspension was kept under mild stirring for 15 hat 25◦C. After that, the gel was filtered and washedexhaustively with 25 mM sodium phosphate buffer andthen with distilled water. This support must be usedimmediately due to the low stability of the aldehydegroups.

2.7. Enzyme immobilization onto differentsupports

An amount of 10 g of different supports prepareda re-a ndedi 7c r1 ereg es n ande ove.W en-s -m paredh andc hes rose.I re-s e int thei

.2 and acetonitrile (95/5)).One unit of DAAO activity was defined a

he amount of enzyme able to oxidize 1�mol ofephalosporin C per minute.

.4.3. Assay of glucose oxidase activityGlucose oxidase (GOX) activity was me

ured spectrophotometrically by an increase inbsorbance at 414 nm resulting from the oxidatioBTS in a coupled system with peroxidase (Batemannd Evans, 1995). The reaction was carried outddition of soluble enzyme or suspension of imm

ized glucose oxidase to a reaction mixture contain.2 mL of 100 mM sodium phosphate buffer pH 6.5 mL of 1 M glucose, 0.1 mL of a 27 mg/mL solutif ABTS prepared in distilled water, and 0.1 mf 2 mg/mL peroxidase solution in 100 mM sodiuhosphate buffer pH 6.0 at 25◦C.

s previously described (MANAE-agarose or pctivated agarose with glutaraldehyde) were suspe

n 20 mL of 25 mM potassium phosphate at pHontaining 2 UI/mL of DAAO, 7 UI/mL of GOX, o6 UI/mL in the case of GAC. The suspensions wently stirred at 25◦C. Periodically, samples of thuspensions and the supernatants were withdrawnzyme activity was analyzed as described abhen the immobilization was completed, the susp

ions were filtered and stored at 4◦C for further experients. In all cases, a reference suspension was preaving exactly the same enzyme concentrationonditions (pH,T, ionic strength), adding instead of tupport the corresponding amount of inert wet-agan all cases, the activity of this reference was fully perved during immobilization, therefore, a decreashe supernatant activity can be directly correlated tommobilization yield.

F. Lopez-Gallego et al. / Journal of Biotechnology 119 (2005) 70–75 73

2.8. Crosslinking with glutaraldehyde

The adsorbed derivative on MANAE-agarose pre-pared as previously described was incubated with 0.5%glutaraldehyde solution (unless otherwise specified) in25 mM sodium phosphate buffer at pH 7 and 25◦Cfor 1 h, under mild stirring. This treatment permit-ted to fully modify the primary amino groups of theenzyme and the support with just one glutaraldehydemolecule (Monsan, 1978; Fernandez-Lafuente et al.,1995; Guisan et al., 1997).

The suspension was then filtered and washed with25 mM sodium phosphate buffer pH 7 to remove theexcess of glutaraldehyde. Then the filtered derivativeswere incubated for an additional 20 h period at 25◦Cfor achieving an more intense crosslinking between theenzyme and the support.

2.9. Evaluation of the attachment between theenzyme and the support

The different enzyme derivatives were incubatedwith 1 M NaCl at pH 7.0 and 25◦C for 1 h. After that,enzyme activity was analyzed in the suspension andsupernatant as previously described.

3. Results and discussion

3.1. Immobilization of enzymes onto differents

bi-l e-h rtsa thea lenta n byt ncu-b

che sed1 Xe than8 eda nd4 entw ins

Fig. 1. Thermal stability of the different immobilized prepara-tions of glutaryl acylase (GAC). Triangles, derivative adsorbed ontoMANAE-agarose and then crosslinked with 0.5% glutaraldehydesolution; rhombus, derivatives immobilized onto MANAE-agaroseactivated with glutaraldehyde; cross, derivative immobilized ontoMANAE-agarose activated with glutaraldehyde and then modifiedwith 0.5% glutaraldehyde solution; squares, derivatives adsorbedonto MANAE-agarose; circles, soluble enzyme. The inactivationcourses were carried out by incubating 10 IU/mL 10 mM potassiumphosphate buffer at pH 7.0 and 47◦C. Additional specifications aredescribed in Section2.2.

led to a greater or lower reduction on the expressedactivity of the MANAE adsorbed derivatives depend-ing on each enzyme. Only a slightly decrementwas seen for GAC (by 10%), however, a significantdecrement was observed in the case of GOX (by 55%)and DAAO (by 65%).

3.2. Stability of the different enzyme preparations

Figs. 1–3show that ionically adsorbed preparationsfor all enzymes presented a slightly higher stability thanthe soluble enzyme. The immobilization of the enzymeon MANAE-glut permitted to improve the enzyme sta-bility compared to that of the just adsorbed enzymesin the DAAO and GOX cases, where the stability wasincreased six- and 172-folds, respectively. However, inthe case of GAC, the stability of both preparations wasvery similar.

On the other hand, when the MANAE adsorbed pro-teins were further treated with glutaraldehyde, higherstabilization factors were found in all cases with respectto the covalently derivative on MANAE-glut. In thisform, this glutaraldehyde treatment caused an incre-ment in the enzyme stabilities, with respect to the onlyadsorbed derivative and even more than those immo-bilized on glutaraldehyde pre-activated supports. Thestabilities of the MANAE adsorbed proteins treated

upports

Several enzymes were rapidly and fully immoized on MANAE-agarose activated with glutaraldyde (MANAE-glut) and MANAE-agarose. Suppoctivated with glutaraldehyde or the treatment ofdsorbed enzymes with glutaraldehyde give covattachment of the enzyme to the support, as show

he lack of enzyme release from the support, when iating the preparations in 1 M NaCl.

The expressed activity was different for eanzyme and immobilization protocol. GAC expres00% of the activity using both supports, GOxpressed 100% on MANAE-agarose and more0% on MANAE-glut, DAAO reduced the expressctivity from 100% on the MANAE support to arou0% on the covalently immobilized. The treatmith glutaraldehyde of the MANAE adsorbed prote

74 F. Lopez-Gallego et al. / Journal of Biotechnology 119 (2005) 70–75

Fig. 2. Thermal stability of the different immobilized preparations ofd-aminoacid oxidase (DAAO). Triangles, derivative adsorbed ontoMANAE-agarose and then crosslinked with 0.5% glutaraldehydesolution; rhombus, derivatives immobilized onto MANAE-agaroseactivated with glutaraldehyde; cross, derivative immobilized ontoMANAE-agarose activated with glutaraldehyde and then modifiedwith 0.5% glutaraldehyde solution; squares, derivatives adsorbedonto MANAE-agarose; circles, soluble enzyme. The inactivationcourses were carried out by incubating 0.8 IU/mL 10 mM potassiumphosphate buffer at pH 7.0 and 50◦C. Additional specifications aredescribed in Section2.2.

with glutaraldehyde compared to the stability of theimmobilized proteins on pre-activated supports were:a factor of 9 for DAAO, a factor of 2.5 for GOX, and 13for GAC. This means that the final stabilization in somecases was very high. For example, the new derivative

Fig. 3. Thermal stability of the different immobilized preparationsof glucose oxidase (GOX). Triangles, derivative adsorbed ontoMANAE-agarose and then crosslinked with 0.5% glutaraldehydesolution; rhombus, derivatives immobilized onto MANAE-agaroseactivated with glutaraldehyde; cross, derivative immobilized ontoMANAE-agarose activated with glutaraldehyde and then modifiedwith 0.5% glutaraldehyde solution; squares, derivatives adsorbedonto MANAE-agarose; circles, soluble enzyme. The inactivationcourses were carried out by incubating 0.8 IU/mL 10 mM potassiumphosphate buffer at pH 7.0 and 56◦C. Additional specifications aredescribed in Section2.2.

of GOX was more than 400-fold more stable than thesoluble enzyme.

The modification of the enzyme surface does notseem to be responsible for the increment in enzymaticthermostability because, as shown inFigs. 2–4, thistreatment does not have any effect on the stability ofthe enzymes immobilized onto MANAE-glut where thenew glutaraldehyde molecules only can react with theenzyme but not with the support. Thus, the stabilizationachieved seems to be the result of an different enzyme-support reaction in the case of the crosslinking of theadsorbed enzyme.

4. Conclusion

Immobilization via adsorption onto a cationic sup-port, such as MANAE-agarose, followed by treatmentwith glutaraldehyde has been found to be a very simpleand fast procedure which can be employed to prepareremarkably stable covalently immobilized enzymes.Stabilization factors using this technique were muchhigher than using glutaraldehyde pre-activated sup-ports, suggesting that the possibilities of achieving anintense multipoint covalent attachment (that is, a strongprotein-support interactions) were quite high using amild modification of the amino groups in the sup-port and the protein with glutaraldehyde, in agreementwith previous results showing stabilization of proteinsvia intermolecular crosslinking (Fernandez-Lafuentee eent inog pri-mw , ani sup-p veryh ionf veryh ingo .,2

A

oject( an-

t al., 1995). Thus, it seems that the reaction betwwo glutaraldehyde molecules bond to primary amroups is really easier than the reaction betweenary amino groups of the enzyme (having a pKof 10.7)ith the glutaraldehyde activated supports. Hence

ntense crosslinking between the enzyme and theort seems to be achieved, and this promoted aigh stabilization of the enzyme. In fact, stabilizat

actors achieved by using this strategy becomeigh, and fully comparable with that achieved usther immobilization techniques (Lopez-Gallego et al004; Sarath Babu et al., 2004; Mateo et al., 2000).

cknowledgements

This research has been funded by the EC PrMATINOES G5RD-CT-2002-00752), and the Sp

F. Lopez-Gallego et al. / Journal of Biotechnology 119 (2005) 70–75 75

ish CICYT Projects (BIO2001-2259, and PPQ 2002-01231 A). A post-doctoral fellowship for Gisela M.Dellamora-Ortiz (Fundac¸ao CAPES of the BrazilianGovernment) and a pre-doctoral fellowship for Fer-nando Lopez Gallego from Spanish MCYT are grate-fully acknowledged.

References

Aebi, H., 1981. Catalase. In: Bergmeyer, Hu. (Ed.), Methods of Enzy-matic Analysis, 2nd ed. Verlag Chemie Internantional, DeerfielBeach, pp. 673–684.

Balcao, V.M., Mateo, C., Fernandez-Lafuente, R., Malcata, F.X.,Guisan, J.M., 2001. Structural and functional stabilization ofl-asparaginase upon immobilization onto highly activated sup-ports. Biotechnol. Prog. 17, 537–542.

Barros Rui, M., Extremina, C., Gonc¸alves, I.C., Braga, B.O., Balcao,V.M., Malcata, F.X., 2003. Hydrolysis of�-lactalbumin by car-dosin A immobilized on highly activated supports. EnzymeMicrob. Technol. 7, 908–916.

Bateman Jr., R.C., Evans, J.A., 1995. Using the glucose oxi-dase/peroxidase system in enzyme kinetics. J. Chem. Educ. 72,240.

Bradford, M.M., 1976. A rapid and sensitive method for the quantifi-cation of microgram quantities of protein utilizing the principleof protein-dye binding. Anal. Biochem. 76, 248–253.

Burteau, N., Burton, S., Crichton, R.R., 1989. Stabilisation andimmobilisation of penicillin amidase. FEBS Lett. 258, 185–189.

Conlon, H.D., Baqual, J., Baker, K., Shen, Y.Q., Wong, B.L., Nolles,R., Rauson, C.W., 1995. Two-step immobilized enzyme con-version of cephalosporine C to 7-aminocephaloporanic acid.Biotechnol. Bioeng. 476, 510–513.

D .,y in628.

D ith51–

F theActa

F , C.,ti-l

od.

F .M.,ular

crosslinking with bifunctional reagents. Enzyme Microb. Tech-nol. 17, 517–523.

Fisher, L., Horner, R., Wagner, F., 1995. Production ofl-amino acidsby applyingd-aminoacid oxidase. Ann. N.Y. Acad. Sci. 750,415–420.

Guisan, J.M., Penzol, G., Armisen, P., Bastida, A., Blanco, R.M.,Fernandez-Lafuente, R., Garcıa-Junceda, E., 1997. Immobiliza-tion of enzymes acting on macromolecular substrates: reductionof steric problems. In: Bickerstaff, G. (Ed.), Immobilization ofEnzymes and Cells, Series Methods Biotechnol, vol. 1. TheHumana Press Inc., pp. 261–275.

Hwang, S., Lee, K., Park, J., Min, B., Haam, S., Ahn, I., Jung, J.,2004. Stability analysis ofBacillus stearothermophilusL1 lipaseimmobilized on surface-modified silica gels. Biochem. Eng. J.17, 85–90.

Lamas Estela, M., Barros Rui, M., Balcao, V.M., Malcata, F.X.,2001. Hydrolysis of why proteins by proteases extracted fromCynara cardunculusand immobilized onto highly activated sup-ports. Enzyme Microb. Technol. 28, 642–652.

Lopez-Gallego, F., Betancor, L., Hidalgo, H., Mateo, C., Guisan,J.M., Fernandez-Lafuente, R., 2004. Optimization of an indus-trial biocatalyst of glutarayl acylase. Stabilization of the enzymeby multipoint immobilization on new amino-epoxy-sepabeads.J. Biotechnol. 111, 219–227.

Magnan, E., Catarino, I., Paolucci-Jeanjean, D., Preziosi-Belloy, L.,Belleville, M.P., 2004. Immobilization of lipase on a ceramicmembrane: activity and stability. J. Membr. Sci., 161–166.

Mateo, C., Abian, O., Fernandez-Lafuente, R., Guisan, J.M., 2000.Reversible enzyme immobilization via a very strong and nondis-torting ionic adsorption on support-polyethylenimine compos-ites. Biotechnol. Bioeng. 68, 98–105.

Mieglo, I., Moreira, M.T., Palma, C., Guisan, J.M., Fernandez-Lafuente, R., Feijoo, G., Lema, J.M., 2003. Catlaytic propertiesof immobilized and stabilized manganese peroxidases. EnzymeMicrob. Technol. 32, 769–775.

M of a84.

S 004.rmalens.

S tics. 39,

V Co-diata

W yme.

Z e–74.

os Reis-Costa, L.S., Andreimar, M., Franc¸a, S.C., Trevisan, H.CRoberts, T.J.C., 2003. Immobilization of lipases and assacontinuous fixed bed reactor. Protein Peptide Lett. 10, 619–

’Souza, S.F., Kubal, B.S., 2002. A cloth strip bioreactor wimmobilized glucoamylase. J. Biochem. Biophys. Meth. 51, 1159.

echtig, B., Peter, H., Bickel, H., Visher, E., 1968. Concerningpreparation of 7-aminocephalosporanic acid. Helv. Chim.51, 1109–1120.

ernandez-Lafuente, R., Rosell, C.M., Rodriguez, V., SantanaSoler, G., Bastida, A., Guisan, J.M., 1993. Preparation of acvated supports containing low pK amino groups. A new toofor protein immobilization via the carboxyl coupling methEnzyme Microb. Technol. 15, 546–550.

ernandez-Lafuente, R., Rosell, C.M., Rodriguez, V., Guisan, J1995. Strategies for enzyme stabilization by intramolec

onsan, P., 1978. Optimization of glutaraldehyde activationsupport for enzyme immobilisation. J. Mol. Catal. 3, 371–3

arath Babu, V.R., Kumar, M.A., Karanth, N.G., Thakur, M.S., 2Stabilization of immobilized glucose oxidase against theinactivation by silanization for biosensor applications. BiosBioelectron. 19, 1337–1341.

eyhan, T.S., Alptekin, O., 2004. Immobilization and kineof catalase onto magnesium silicate. Process Biochem2149–2155.

an Aken, B.P., Henry, L., Spiros, N., Agathos, N., 2000.immobilization of manganese peroxidase from Phlebia raon porous silica beads. Biotechnol. Lett. 8, 641–646.

ilson, R., Turner, A.P.F., 1992. Glucose oxidase: an ideal enzBionsens. Biolelectron. 7, 165–185.

hou, Q.Z.K., Chen, X.D., 2001. Immobilization of�-galctosidason graphite surface by glutaraldehyde. J. Food Eng. 48, 69