Enzyme Stabilization by Covalent

Enzyme Stabilization by Covalent Binding in Nanoporous Sol-Gel Glass for Nonaqueous Biocatalysis Ping Wang,1 Sheng Dai,2 S. D. Waezsada,2 Alice Y. Tsao,2 Brian H. Davison2 1 Department of Chemical Engineering, University of Akron, Akron, Ohio, 44325 2 Chemical Technology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6226; telephone: 865-576-8522; fax: 865-574-6442; e-mail: davisonbh@ornl.gov Received 9 August 2000; accepted 18 February 2001 Abstract: A unique nanoporous sol-gel glass possessing a highly ordered porous structure (with a pore size of 153 Å in diameter) was examined for use as a support material for enzyme immobilization. A model enzyme, -chymotrypsin, was efficiently bound onto the glass via a bifunctional ligand, trimethoxysilylpropanal, with an active enzyme loading of 0.54 wt%. The glass-bound chymotrypsin exhibited greatly enhanced stability both in aqueous solution and organic solvents. The half-life of the glass-bound -chymotrypsin was >1000-fold higher than that of the native enzyme, as measured either in aqueous buffer or anhydrous methanol. The enhanced stability in methanol, which excludes the possibility of enzyme autolysis, particularly reflected that the covalent binding provides effective protection against enzyme inactivation caused by structural denaturation. In addition, the activity of the immobilized -chymotrypsin was also much higher than that of the native enzyme in various organic solvents. From these results, it appears that the glass–enzyme complex developed in the present work can be used as a high-performance biocatalyst for various chemical processing applications, particularly in organic media. Published by John Wiley & Sons Biotechnol Bioeng 74: 249–255, 2001. Keywords: nonaqueous biocatalysis; sol-gel silica glass; -chymotrypsin; organic solvents; enzyme immobilization INTRODUCTION Nontraditional biocatalysis, which applies to enzymes in nonaqueous media for biotransformations of xenobiotic chemicals, has been investigated extensively over the past two decades. Progress in this area will ultimately reshape the configurations of bioprocessing for large-scale applications, as outlined by Davison et al. (1997). Often the low Correspondence to: B. Davison Contract grant sponsor: Office of Industrial Technologies, U.S. Department of Energy/UT–Battelle, LLC The submitted manuscript has been authored by a contractor of the U.S. Government under contract DE-AC05-00OR22725. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes. activity and stability of enzymes placed in nonaqueous environments are the limiting factors for industrial applications. The observed activities of native enzymes in organic media are usually two to six orders of magnitude lower than their aqueous activities (Dordick, 1992). Factors including structural denaturation and diffusional limitation (due to the insolubility of native enzymes) may contribute to the limited nonaqueous activities of enzymes (Klibanov, 1997). Various activation and stabilization methodologies, including covalently modifying enzymes with chemical groups (Pina et al., 1989; Takahashi et al., 1985; Vazquez-Duhalt et al., 1992; Wang et al., 1999), complexation with surfactants (Paradkar et al., 1994; Powers et al., 1993) or polymers (Secundo et al., 1999), freeze-drying with inorganic salts (Ru et al., 2000), and incorporation of enzymes into polymeric supports (Wang et al., 1997; Yang et al., 1995a, 1995b), have been examined for biocatalysis in organic solvents. In particular, the covalent binding of enzymes to solid supports can effectively extend the lifetime of the biocatalysts by protecting the native three-dimensional (3D) structure of enzyme molecules. Traditional enzyme immobilization technologies have been developed, based mostly on the consideration of the reuse of biocatalysts (Malcata et al., 1990; Tischer and Wedekind, 1999), whereas the detrimental mass transfer effect usually leads to very low apparent enzyme activity, and thus considerably limits the effectiveness of the biocatalysts. In contrast, attachment of enzyme to solid supports may result in enhanced enzyme activity in organic solvents as compared with that of native enzyme in the same reaction media. It has been demonstrated that the incorporation of enzymes into synthetic polymers, especially those via multiple covalent bonds, can significantly improve their activities in a nonaqueous environment (Miyanaga et al., 1999; Wang et al., 1997; Yang et al., 1995a, 1995b). Specifically, plastic enzymes showed activities that were comparable or even higher than those of enzymes solubilized via ion pairing with surfactant in organic solvents (Wang et al., 1997). The previous efforts for the multiple bonding of enzymes have been conducted mostly with organic synthetic poly- © 2001 John Wiley & Sons, Inc. This article is a US Government Work and, as such, is in the public domain in the United States of America. mers. Inorganic materials, on the other hand, have attracted much less attention. This is presumably because of the ease of fabrication of desirable structures and the availability of reactive functional groups of organic polymers. In situ polymerization methods in particular can achieve considerably high enzyme loading (up to 50 wt%) (Wang et al., 1997). The in situ incorporation of enzymes into inorganic materials is rather difficult when compared with synthetic polymers. Nevertheless, sol-gel silica materials have been investigated by different groups as an alternative to organic carriers (Greaves and Rotello, 1997; Li et al., 1998). The requirement for the cosolvent to overcome the low aqueous solubility of alkylsilicate precursors and the formation of alcohols during the gelation reaction process, however, can adversely affect enzyme activity. Although different precursors [such as poly(glycerylsilicate)] have been examined in an effort to generate milder sol-gel conditions (Gill and Ballesteros, 1998), in situ sol-gel incorporation has been applied mostly in the development of enzyme-coated electrodes for analytical applications, where enzyme is entrapped in a sol-gel thin film attached to electrodes (Li et al, 1998; Ogura et al., 1998; Subramanian et al., 1999). In most of these applications, effective enzyme loading is not as critical as in other bioprocessing applications. Recent breakthroughs (Firouzi et al., 1995; Kresge et al., 1992; Tian et al., 1997) in catalyst synthesis have resulted in a novel methodology for preparing nanoporous inorganic materials with extremely high surface areas and ordered mesostructure. Nanoporous silicon, aluminum, and transition metal oxides have been prepared. The essence of this new methodology is the use of molecular assemblies of surfactants or related substances as structure directors during the formation of oxides. The surfactants used in synthesis can be cationic, anionic, and neutral, depending on the charge of the inorganic precursors. Unlike microporous materials, which interact only with enzyme molecules on outer surfaces, nanoporous materials have larger pore openings that allow access to the inner pore surfaces. In a recently published study, nanoporous silica materials (pore diameter range from 27 to 92 Å) were used to physically absorb horseradish peroxidase, and improved enzyme activity in organic solvents and enhanced thermostability in aqueous solutions were demonstrated (Takahashi et al., 2000). The use of highly porous glass may allow the enzyme to be hosted inside the microchannel via multiple-point attachment, which is expected to improve the enzyme’s structural stability (Martinek et al., 1977). In addition, the mass transfer of organic chemicals in these materials is far more efficient than in conventional microporous catalysts, such as zeolites, because of their unique nanoporous pore diameters (20 to 250 Å). These considerations prompted us to propose to synthesize a new class of enzyme catalysts supported in ordered mesoporous hosts. In the present work, a model enzyme, -chymotrypsin, is covalently bound into nanoporous silica via a bifunctional agent, and the catalytic behaviors of the immobilized en- zyme are examined in terms of enzyme stability and activity at elevated temperatures and in organic solvents. MATERIALS AND METHODS Materials Chemicals, including -chymotrypsin, N-succinyl-Ala-AlaPro-Phe p-nitroanilide (SAAPPN), N-acetyl-Lphenylalanine ethylester (APEE), n-propanol, 4-methylumbelliferyl p-trimethylammonium cinnamate chloride (MUTMAC), and 4-methylumbelliferone, were purchased from Sigma Chemical Co. (St. Louis, MO). Highperformance liquid chromatography (HPLC)-grade solvents, acetonitrile, hexane, isooctane, water, and acetone, were purchased from VWR Scientific Products (Suwanee, GA). Fresh organic solvents were dried over 4-Å molecular sieves and used without any further treatment. Vinyltrimethoxysilane, tetraethylorthosilicate, RhH(CO)(PPh3), and PPh3 were purchased from Aldrich Chemical Co. Enzymatic Activity Hydrolysis activity was measured using SAAPPN in 0.1 M Bis–Tris (pH 7.8) buffer. Pure p-nitroaniline was used as the product standard. In a typical measurement, 1 mL of substrate solution containing 0.4 mM SAAPPN was mixed with 10 L of native enzyme solution (∼1 mg/mL) in a 1.7-mL quartz UV-spec cuvette, and absorbance at 410 nm was immediately monitored for 1 min at room temperature. In the case of immobilized enzyme, 30 mg of dry sol-gel was added to 5 mL of substrate solution with stirring, and a well-mixed 1-mL aliquot was periodically taken and filtered through a 0.22- m PTFE syringe filter for absorbance measurement at 410 nm. The sampling solution was taken under stirring to ensure uniform catalyst concentration, and the sampling solution was not added back to the reactor once removed. Up to five samples were analyzed for product concentration at different reaction timepoints to elucidate the initial reaction rate for each reaction with immobilized enzyme. Transesterification reactions were conducted in organic solvents containing 0.5 to 20 mM APEE and 0.5 M n-propanol. Catalyst particles were removed from the reaction solutions by centrifugation, and the concentration of the product and substrate (APEE) was measured by HPLC. The extinction coefficient of the product was assumed to be the same as APEE in the calculation of the extent of reaction of the product. Although methanol is a potential reactant for the transesterification reaction, no methylester product peak was detected. The activity reported for the immobilized enzyme in methanol was calculated based on the same analyses as in other organic solvents. Each reaction for activity measurement was repeated at least three times, and the error was within 10%. Enzyme Stability Hydrolytic enzyme activity was measured (in the same way as mentioned earlier for enzyme activity tests) after a de- 250 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 74, NO. 3, AUGUST 5, 2001 termined incubation time to assess enzyme stability. In a typical procedure, a measured amount of enzyme sample was added to organic solvents or aqueous buffer and placed in a temperature-controlled water bath. After a predetermined incubation time, the vial was removed from the water bath and cooled to room temperature. SAAPPN solution was added directly to the aqueous enzyme solutions and the reaction rates were measured at room temperature for aqueous stability tests (the dilution effect was counted in the calculation of the concentrations of enzyme and substrate), whereas the solvent was purged by blowing nitrogen before the addition of aqueous substrate solution for stability tests in organic solvents. For all tests, native enzyme was used directly from the bottle, and immobilized enzyme was used after drying from deionized water. Active Site Titration The concentration of active sites of both native and immobilized -chymotrypsin was measured according to a method reported earlier (Gabel, 1974). In brief, 100 L of 2 mg/mL native enzyme or 5 mg of enzyme-containing sol-gel was mixed with 2 mL of 0.025 mg/mL MUTMAC solution (0.1 M Na-borate buffer, pH 7.5), and the fluorescence of the reaction solution was measured (excitation 360 nm, emission 450 nm). In the case of immobilized enzyme, the suspended particles were removed by filtration. Solutions of 4-methylumbelliferone were used as the standard. Tests with the out-of-bottle enzyme powder showed that 57.2 wt% was active enzyme by active site titration. The total amount of protein in aqueous solutions was measured by a protein assay (Bradford Coomassie Assay, Bio-Rad) with bovine serum albumin (BSA) as the standard. Analysis Hydrolysis reactions were monitored using a spectrophotometer (Model UV-2101PC, Shimadzu). Fluorescence was performed on a luminescence spectrometer (Model LS50B, Perkin-Elmer Analytical Instruments, Norwalk, CT). HPLC analysis for APEE transesterification was performed using a Shimadzu system equipped with a C18 column (4.6 × 250 mm; Waters Corp., Milford, MA). The mobile phase was a mixture of water and MeCN at a flow rate of 1.0 mL/min. A solvent gradient, in which the MeCN content was increased from 40% to 100% within 15 min, was applied for the transesterification reaction samples. The absorbance at 254 nm of the eluant was monitored through an on-line UV detector. Preparation of Ordered Nanoporous Silica Support The typical procedure (Zhao et al., 1998) involved the mixing of 4 g of a nonionic triblock copolymer (L123, BASF), 20 g of water, and 80 g of 2 M HCl. To this solution, 8 g of tetraethylorthosilicate (TEOS) was added at room tempera- ture. The mixture was then stirred at room temperature for 24 h. The solid product was recovered by filtration. The removal of the polymer template was done by calcination at 500°C for 5 h. The pore size of the calcined and ordered nanoporous silica samples was measured by small-angle X-ray scattering and the Barret–Joyner–Halenda method. Samples having an average pore size of 153 Å were used in the current study. Ligand Synthesis Trimethoxysilylpropanal was synthesized via a method similar to that reported by Takeuchi and Sato (1990). A solution of 20.0 g (135 mmol) vinyltrimethoxysilane, 92 mg (0.01 mmol) RhH(CO)(PPh3), and 0.21 g (0.8 mmol) PPh3 in 100 mL toluene was placed in a 300-mL autoclave. The vessel was loaded with 600 psi CO and stirred for 10 min, followed by addition of H2 to 1200 psi total pressure (ratio of CO:H2 was 1:1). It was heated at 80°C for 4 h under stirring (70 rpm). The products (90%) were isolated by distillation under vacuum (0.5 torr, 45° to 50°C). The products consisted of a mixture of two isomers (normal and iso) whose structures are depicted in Scheme I. The isomer ratio of normal:iso was 95:5. Enzyme Immobilization In a typical immobilization procedure, 0.5 g of porous silica glass was added to 5 mL of ethanol at 40°C. The mixture was stirred for 1 h to wet the glass thoroughly, and 0.2 g of the trimethoxysilylpropanal ligand and 0.5 mL of water were added to the mixture. One hour later the mixture was centrifuged and the supernatant solution was removed. The reagent-coated gel was then washed extensively with ethanol, then water, to remove residual reagent. The washed sol-gel was mixed with 3.5 mL of 5 mg/mL chymotrypsin solution (in water) and stirred at 4°C for 48 h. The mixture was then centrifuged and the protein content and enzyme activity (hydrolysis) of the supernatant solution were measured and compared with the data from the initial enzyme solution. As a result, the supernatant protein concentration was reduced by 53% (by UV absorbance). The enzymebound glass was washed several times (at least five) with 25 mL of deionized water until no absorbance at 280 nm was Scheme I. Chemical structure of biofunctional ligand for the covalent binding of enzyme to sol-gel glass. WANG ET AL.: ENZYME ACTIVATION AND STABILIZATION FOR NONAQUEOUS BIOCATALYSIS 251 observed in the washing solution, and then dried by purging N2 gas. The weight of the final dry solid was around 0.6 g. RESULTS AND DISCUSSION Attachment of enzymes to the surface of glass beads or wool, either via physical adsorption or covalent binding, has been practiced widely as one of the traditional enzyme immobilization technologies. Although glass provides desirable mechanical strength and thermochemical stability, it is difficult to achieve high enzyme loading as required for efficient bioprocessing applications. A straightforward solution to this problem is the use of porous glass that can provide much more surface area, and thus increase the enzyme loading. A nanoporous silica glass with a unique pore configuration (Fig. 1) was synthesized and applied in this work. Such a highly ordered configuration of pores would allow maximum porosity of the materials. The samples used in this study had a pore size of 153 Å and a surface area of 560 m2/g. In addition to the consideration on enzyme loading, we also believe that such a nanoporous structure may provide another advantage in achieving multipoint attachment, which has been proven effective in stabilizing the enzyme structure (Martinek et al., 1977; Mozhaev et al., 1990; Wang et al., 1997; Yang et al., 1995a, 1995b). The large pore size of the glass will allow the enzyme molecules be held inside the nanochannels (the radius of the -chymotrypsin is ∼22 Å). The concave surface (Fig. 2A) can provide a larger contact area for multipoint attachment with the enzyme than flat or convex surfaces (Fig. 2B), especially when the radius of curvature is comparable to the size of the protein molecules. Figure 2. Configurations of the covalent binding of enzyme molecules onto different surfaces: (A) Nanoporous sol-gel glass, which provides the potential for multiple-point attachment; and (B) a flat surface, where a single bond is most likely to be achieved. Figure 1. Scanning electron microscopic image of nanoporous silica glass. The configuration of pores is highly ordered. The pore size (diameter) used in the present work is 153 Å. The amount of available enzyme immobilized on the glass was measured by active site titration, which detects the amount of active enzyme as opposed to the gross protein content via protein assay. Prior to the titration, the sol-gel samples were washed extensively with water to remove any physically adsorbed enzyme, followed by purging N2 gas to dry. Results showed that the immobilization procedure led to an active enzyme loading of 0.54 ± 0.05% (wt/wt). This observation is comparable to results achieved using synthetic polymers (Yang et al., 1995a, 1995b). Mass balance analysis based on measurements of active enzyme concentration (by active site titration) in the solution before and after contact with glass particles showed that 42% of active enzyme was retained in the aqueous solution, and 36% recovered (corresponding to 67% of the missed active enzyme) in the final and purified sol-gel product. The rest of the active enzyme may be lost during the washing of the sol-gel product or inactivated through the processing. Because the focus of this work was on the stability and activity of the glass-bound enzyme, no effort was made to optimize the immobilization procedure. The apparent activities of immobilized enzymes in aqueous solutions are usually much lower than those of their parent free enzymes. This is due mostly to the effect of diffusional limitation, because reactions in aqueous solutions are usually quite rapid. Our tests have shown that the sol-gel chymotrypsin has an apparent aqueous activity of ∼5% of the free enzyme, as measured by the hydrolysis reaction and calculated based on the amount of active enzyme. From an economic perspective for bioprocessing, the reduced apparent activity in aqueous solution can be offset by the extended effective lifetime of the immobilized enzyme. For applications in nonaqueous media, however, diffusional limitation is usually much less prominent, especially for samples with enzyme loading of <1% (wt/wt) (Wang et al., 1997). On the basis of our tests, glass-bound chymotrypsin showed much higher activity than the native enzyme suspended in the same organic solvent. Table I shows the data of the normalized reaction rates in various solvents. Values of the kinetic parameters (Kcat and KM) were not obtained as the data showed much scattering and deviated from Michaelis–Menten kinetics. This could be a 252 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 74, NO. 3, AUGUST 5, 2001 Table 1. V/[E] (M/h M−1) (native enzyme) 0.007 0.25 0.015 0.03 ND 0.03 0.11 V/[E] (M/h M−1) (sol-gel enzyme) 0.79 1.4 0.98 1.5 1.37 1.95 2.68 Solvent Hexane Hexane Acetonitrile Acetonitrile Methanol Isooctane Isooctane [APEE] (mM) 0.5 5 0.5 10 0.5 0.5 5 Enhancement 110 5.6 65 50 — 65 24 result of random factors, such the variation of the size of enzyme clusters formed in the suspension. Nevertheless, the tabulated data sufficiently demonstrate the activation effect of the immobilization, and the activity of glass– chymotrypsin was up to two orders of magnitude higher than that of native enzyme. The relatively higher activity of the immobilized enzyme may have two explanations. First, immobilized enzyme is spread on the surface of the porous glass and thus affords much better contact with the reaction solution than the native enzyme simply suspended in the organic solution. Second, the expected covalent binding may provide the enzyme with protection against structural denaturation due to the unfavorable solvent–protein interactions, and thus result in an apparent activation effect. Although much more work is apparently needed to identify the significance of these two putative mechanisms, in what follows we focus on the experimental proof of a stabilization effect of covalent binding for -chymotrypsin. The thermostability of -chymotrypsin was examined at 40°C and 50°C in aqueous solution through a hydrolysis reaction. It appeared that the glass-bound enzyme was remarkably more stable than the native enzyme. For example, although the native enzyme lost >90% of its original activity within 10 min of heating at 40°C, the immobilized enzyme retained >90% of its original activity after a period of 2 h (Fig. 3). The estimated half-life of immobilized -chymotrypsin is >2300 min at 40°C, about three orders of magnitude higher than that of native enzyme (2.5 min as measured under the same condition). Because the autolysis of proteases in aqueous solutions may significantly inactivate the enzyme, the restricted interactions among the glass-bound enzyme molecules, as opposed to the effect of structural stabilization, could play an important role in retaining the enzyme’s aqueous activity. The effect of autolysis can be eliminated in pure organic solvents, and one recent study has shown that structural denaturation was the predominant cause of enzyme inactivation in water-miscible solvents (Miyanaga et al., 1999). To determine whether covalent binding provides any protection against protein structural denaturation, the assessment of stability of native and immobilized -chymotrypsin in methanol was performed. Interestingly, immobilized -chymotrypsin again showed much higher stability than the native enzyme (Fig. 4). The native enzyme was inactivated almost immediately upon contact with methanol (halflife estimated to be <1 min), whereas >35% of the original activity of the immobilized enzyme was retained after a 2.5-h incubation in the same solvent (half-life about 100 Figure 3. Thermostability of native and immobilized -chymotrypsin. Enzymes were incubated in 0.1 M Bis–Tris (pH 7.8) buffer, and their hydrolysis activities were measured after being cooled to room temperature with 0.4 mM SAAPPN. ( ) Glass-bound enzyme at 40°C; ( ) glass-bound enzyme at 50°C; ( ) native enzyme at 40°C; ( ) native enzyme at 50°C. Figure 4. Solvent stability of -chymotrypsin in anhydrous methanol. Enzymes were placed in pure methanol and incubated at room temperature for a certain period of time, the solvent was then removed by blowing N2, and hydrolysis activity of the enzymes was measured in aqueous buffer. ( ) Native enzyme; ( ) glass enzyme. WANG ET AL.: ENZYME ACTIVATION AND STABILIZATION FOR NONAQUEOUS BIOCATALYSIS 253 min). These observations prove that multipoint covalent binding significantly improved the enzyme’s stability against structural denaturation in organic solvents, and the same mechanism may have also contributed to the largest part of the stabilization effect observed in aqueous solutions. A similar stabilization effect has been reported for enzymes incorporated into synthetic organic polymers (Martinek et al., 1977; Novick and Dordick, 1998; Wang et al., 1997). However, stability data have also been reported for aqueous solutions (Martinek et al., 1977; Novick and Dordick, 1998) or for organic solvents (Wang et al., 1997). Similarly, only aqueous thermostability was reported for enzyme adsorbed on silica glass (Takahashi et al., 2000). Few studies have demonstrated a significant enzyme stabilization effect in both aqueous and organic media, as observed in this work. In studies of subtilisin covalently bound to polyacrylates (Yang et al., 1995a, 1995b), the half-life of the enzyme was improved by over 500-fold in water, whereas it was only about twofold better in hexane. Different inactivation mechanisms can be postulated for aqueous or organic reaction media considering the different microenvironmental interactions involved, even though much further study is needed. Nevertheless, the data presented herein reflect the promising potential of using nanoporous materials to stabilize enzymes. CONCLUSIONS The unique configuration of nanoporous silica glass seems to provide some advantages, such as high enzyme loading and multipoint covalent attachment, for enzyme immobilization. The glass-bound -chymotrypsin exhibited much improved activity and stability, especially in organic solvents, as compared with its native parent enzyme. Because glass is chemically inert and thermally stable, it is expected that enzymes immobilized by the present technology can be used as ideal biocatalysts for long-term chemical processing applications involving elevated temperatures and nonaqueous reaction media. A. Y. Tsao thanks the ORNL-ERULF summer internship program for partial financial support during this work. Funding from USDOE Office of Industrial Technologies. References Davison BH, Barton JW, Petersen GR. 1997. Nomenclature and methodology for classification of nontraditional biocatalysis. Biotechnol Progr 13:512–518. Dordick JS. 1992. Designing enzymes for use in organic solvents. Biotechnol Progr 8:259–267. Firouzi A, Kumar D, Bull LM, Besier T, Sieger P, Huo Q, Walker SA, Zasadzinski JA, Glinka C, Nicol J, Margoleses D, Stucky GD, Chmelka BF. 1995. Cooperative organization of inorganic–surfactant and biomimetic assemblies. Science 267:1138–1143. Gabel D. 1974. Active site titration of immobilized chymotrypsin with a fluorogenic reagent. FEBS Lett 49:280–281. Gill I, Ballesteros A. 1998. Encapsulation of biologicals within silicate, siloxane, and hybrid sol-gel polymers: An efficient and generic approach. J Am Chem Soc 120:8587–8598. Greaves MD, Rotello VM. 1997. Model systems for flavoenzyme activity. Specific hydrogen bond recognition of flavin in a silicate sol-gel. J Am Chem Soc 119:10569–10572. Klibanov AM. 1997. Why are enzymes less active in organic solvent than in water? Trends Biotechnol 15:97–101. Kresge CT, Leonowicz ME, Roth WJ, Vartuli JC, Beck JS. 1992. Ordered mesoporous molecular-sieves synthesized by a liquid–crystal template mechanism. Nature 359:710–712. Li J, Tan SN, Oh JT. 1998. Silica sol-gel immobilized amperometric enzyme electrode for peroxide determination in the organic phase. J Electroanal Chem 448:69–77. Malcata FX, Reyes HR, Garcia HS, Hill CG. 1990. Immobilized lipase reactors for modification of fats and oils–a review. J Am Oil Chem Soc 67:890–910. Martinek K, Klibanov AM, Goldmacher VS, Berezin IV. 1977. The principles of enzyme stabilization: I. Increase in thermostability of enzymes covalently bound to a complementary surface of a polymer support in a multipoint fashion. Biochim Biophys 485:1–12. Miyanaga M, Ohmori M, Imamura K, Sakiyama T, Nakanishi K. 1999. Stability of immobilized thermolysin in organic solvents. J Biosci Bioeng 87:463–472. Mozhaev VV, Sergeeva MV, Belova AB, Khmelnitsky YL. 1990. Multipoint attachment to a support protects enzyme from inactivation by organic solvents: -chymotrypsin in aqueous solutions of alcohols and diols. Biotechnol Bioeng 35:653–659. Novick SJ, Dordick JS. 1998. Preparation of active and stable biocatalytic hydrogels for use in selective transformations. Chem Mater 10: 955–958. Ogura K, Nakaoka K, Nakayama M, Kobayashi M, Fujii A. 1998. Themogravimetry/mass spectrometry of urease-immobilized sol-gel silica and the application of such a urease-modified electrode to the potentiometric determination of urea. Anal Chim Acta 384:219–225. Paradkar VM, Dordick JS. 1994. Aqueous-like activity of -chymotrypsin dissolved in nearly anhydrous organic solvents. J Am Chem Soc 116: 5009–5010. Pina C, Clark DS, Blanch HW 1989. The activity of PEG-modified chymotrypsin in aqueous and organic media. Biotechnol Biotechniq 3: 333–338. Powers ME, Matsuura J, Brassell J, Manning, MC, Shefter E. 1993. Enhanced solubility of proteins and peptides in nonpolar solvents through hydrophobic ion pairing. Biopolymers 33:927–932. Ru MT, Hirokane SY, Lo AS, Dordick JS, Reimer JA, Clark DS. 2000. On the salt-induced activation of lyophilized enzymes in organic solvents: Effect of salt kosmotropicity on enzyme activity. J Am Chem Soc 22:1565–1571. Secundo F, Carrea G, Vecchio G, Zambianchi F. 1999. Spectroscopic investigation of lipase from Pseudomonas cepacia solubilized in 1,4dioxane by non-covalent complexation with methoxypoly(ethylene glycol). Biotechnol Bioeng 64:624–629. Subramanian A, Kennel SJ, Oden PI, Jacobson KB, Woodward J, Doktycz MJ. 1999. Comparison of techniques for enzyme immobilization on silicon supports. Enz Microb Technol 24:26–34. Takahashi H, Li B, Sasaki T, Miyazaki C, Kajino T, Inagaki S. 2000. Catalytic activity in organic solvents and stability of immobilized enzymes depend on the pore size and surface characteristics of mesoporous silica. Chem Mater 12:3301–3305. Takahashi K, Ajima A, Yoshimoto T, Okada M, Matsushima A, Tamaura Y, Inada J. 1985. Chemical reactions by polyethylene glycol modified enzymes in chlorinated hydrocarbons. J Org Chem 40:3414–3415. Takeuchi R, Sato N. 1990. Hydroformylation of alkenes having organosilicon substituents. J Organometal Chem 393:1–10. Tian ZR, Tong W, Wang JY, Duan NG, Vikrishnan VV, Suib SL. 1997. Manganese oxide mesoporous structures: Mixed-valent semiconducting catalysts. Science 276:926–930. Tischer W, Wedekind FG. 1999. Immobilized enzymes: Methods and applications. Topics Curr Chem 2000:95–126. 254 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 74, NO. 3, AUGUST 5, 2001 Vazquez-Duhalt R, Fedorak PM, Westlake DWS. 1992. Role of enzyme hydrophobicity in biocatalysis in organic solvents. Enz Microb Technol 14:837–841. Wang P, Sergeeva MV, Lim L, Dordick JS. 1997. Biocatalytic plastics as active and stable materials for biotransformations. Nat Biotechnol 15: 789–793. Wang P, Woodward CA, Kaufman EN. 1999. Poly(ethylene glycol)modified ligninase enhances pentacholorophenol biodegradation in water–solvent mixtures. Biotechnol Bioeng 3:290–297. Yang Z, Mesiano AJ, Venkatasubramanian S, Gross SH, Harris JM, Russell AJ. 1995a. Activity and stability of enzymes incorporated into acrylic polymers. J Am Chem Soc 117:4843–4850. Yang Z, Williams D, Russell AJ. 1995b. Synthesis of protein-containing polymers in organic solvents. Biotechnol Bioeng 45:10–17. Zhao DY, Huo QS, Feng JL, Chmelka BF, Stucky GD. 1998. Nonionic triblock and star diblock copolymer and oligomeric surfactant syntheses of highly ordered, hydrothermally stable, mesoporous silica structures. J Am Chem Soc 120:6024–6036. WANG ET AL.: ENZYME ACTIVATION AND STABILIZATION FOR NONAQUEOUS BIOCATALYSIS 255