Hydrogel Network Entrapping Cholesterol Oxidase and

Since the 1970s, optical sensors have been developing very quickly.51,52 The ... for some enzymatic biosensing processes.56,57 From a practical point ...
0 downloads 0 Views 103KB Size
Download PDF
Anal. Chem. 2003, 75, 4019-4027

Hydrogel Network Entrapping Cholesterol Oxidase and Octadecylsilica for Optical Biosensing in Hydrophobic Organic or Aqueous Micelle Solvents Xiao Jun Wu and Martin M. F. Choi*

Department of Chemistry, Hong Kong Baptist University, Kowloon Tong, Hong Kong SAR, P.R. China

Two optical cholesterol biosensors have been fabricated by immobilizing cholesterol oxidase (ChOx) and octadecylsilica (ODS) particles in hydrogel network matrixes of copolymer of poly(vinyl alcohol) (PVA)/hydroxyethyl carboxymethyl cellulose (HECMC), and sol-gel, respectively. In conjunction with an optical oxygen transducer, the immobilized ChOx in the sol-gel/ODS matrix was assembled as an optical cholesterol biosensor to continuously detect free cholesterol in aqueous micelle solution, while the immobilized ChOx in the PVA/HECMC/ODS matrix was constructed as an organic-phase optical cholesterol biosensor for the continuous analysis of free cholesterol in hydrophobic organic solvent. The compositions and properties of the immobilization matrixes, the effects of solvents and the analytical features were studied in detail. Both biosensors showed stable and reliable responses toward free cholesterol. For the aqueous micelle cholesterol biosensor, the analytical working range was from 0.05 to 8.0 mM cholesterol, the response time was 7-12 min, the operation life was more than 35 assays, and the shelf life was ∼4 months. For the organicphase cholesterol biosensor, the analytical working range was from 0.07 to 18.0 mM cholesterol, the response time was 4-8 min, the operation life was more than 120 assays, and the shelf life was longer than 5 months. The organic-phase cholesterol biosensor has been successfully applied to determine the free cholesterol content in commercial butter samples. Enzyme immobilization is one of the most important subjects for any enzyme-based biosensor research. Considerable efforts have been invested in this topic for a number of years.1-4 The biosensing process of the immobilized enzyme is regarded as a heterogeneous phase reaction; thus, the main consideration for enzyme immobilization is to achieve stable and high enzymatic activity with low mass-transfer resistance.5 If the enzyme is * Corresponding author. Fax: +852 3411 7348. E-mail: [email protected]. (1) Straathof, A. J. J., Adlercreutz, P., Eds. Applied Biocatalysis, 2nd ed.; Harwood Academic Publishers: Amsterdam, 2000. (2) Bickerstaff, G. F., Ed. Immobilization of Enzymes and Cells; Humana Press: Totowa, NJ, 1997. (3) Scouten, W. H. A Survey of Enzyme Coupling Techniques. In Methods in Enzymology: Immobilized Enzymes and Cells; Mosbach, K., Ed.; Academic Press: Orlando, Florida, 1987; Vol. 135, pp 30-65. (4) Tischer, W.; Wedekind, F. BiocatalysissFrom Discovery to Application; Fessner, W. D., Ed.; Springer-Verlag: Heidelberg, 1999; pp 95-126. 10.1021/ac020736+ CCC: $25.00 Published on Web 07/01/2003

© 2003 American Chemical Society

entrapped, the immobilization-supporting matrix has to provide sufficient permeability for both solvent and substrate molecules.6 In general, the massive-entrapment technique, with which high enzyme loading is possible, is preferable, since the immobilized enzyme can be well-protected by a large amount of supporting matrixes so that the biosensing processes suffer less interference from solvent, sample, and other external factors. A typical hydrogel network of poly(vinyl alcohol) (PVA) is an ideal matrix for the entrapment of cell7,8 and enzyme.9 An analogous hydrogel network of PVA modified by comixing it with small portions of hydroxyethyl carboxymethyl cellulose (HECMC) has also been successfully applied for enzyme entrapment.10 The sol-gel technique has been demonstrated to be an effective method to tailor-make the silicate cavity of porous hydrogel network so that the enzyme is securely microencapsulated by the silicate polymerization process.11-15 Hydrogel is normally considered as biocompatible with enzymes.16 It is a type of waterswollen/filled and cross-linked polymer formed by the gelling process and features a highly hydrophilic structure of threedimensional networks.16-18 The water, including both the bonding and mobility states in the hydrogel network, is highly thermodynamically exchangeable in all the microphases or regions by virtue of the swelling process.17,19 With this advantage, the enzyme, which is massively entrapped by the hydrogel network with a (5) Gorton, L.; Marko-Varga, G.; Dominguez, E.; Emneus, J. Immobilized Enzyme Reactors: Development, Practical and Theoretical Considerations. In Analytical Applications of Immobilized Enzyme Reactors; Lam, S., Malikin, G., Eds.; Blackie Academic & Professional: Glasgow, 1994. (6) Brink, L. E. S.; Tramper, J. Biotechnol. Bioeng. 1985, 27, 1258-1269. (7) Wu, K.-Y. A.; Wisecarver, K. D. Biotechnol. Bioeng. 1992, 39, 447-449. (8) Chen, K.-C.; Houng, J.-Y., Cell Immobilization with Phosphorylated Polyvinyl Alcohol (PVA) Gel. In Immobilization of Enzymes and Cells; Bickerstaff, G. F., Ed.; Humana Press: NJ, 1997; Chapter 24, pp 207-216. (9) Zhang, J.; Li, B.; Xu, G.; Cheng, G.; Dong, S. Analyst 1999, 124, 699-703. (10) Dong, S.; Guo, Y. Anal. Chem. 1994, 66, 3895-3899. (11) Jin, W.; Brennan, J. D. Anal. Chim. Acta 2002, 461, 1-36. (12) Walcarius, A. Chem. Mater. 2001, 13, 3351-3372. (13) Yamanaka, S. A.; Nishida, F.; Ellerby, L. M.; Nishida, C. R.; Dunn, B.; Valentine, J. S.; Zink, J. I. Chem. Mater. 1992, 3, 495-497. (14) Dave, B. C.; Dunn, B.; Valentine, J. S.; Zink, J. I. Anal. Chem. 1994, 66, 1120A-1127A. (15) Lev, O.; Tsionsky, M.; Rabinovich, L.; Glezer, V.; Sampath, S.; Pankratov, I.; Gun, J. Anal. Chem. 1995, 67, 22A-30A. (16) Lee, K. Y.; Mooney, D. J. Chem. Rev. 2001, 101, 1869-1879. (17) Mathur, A. M.; Scranton, A. B. Biomaterials 1996, 17, 547-557. (18) Ofstead R. F.; Poser, C. I. Semicrystalline Poly(vinyl alcohol) Hydrogels. In Polymers in Aqueous Media; Glass, J. E., Ed.; American Chemical Society: Washington, DC, 1989; Chapter 4, pp 61-72.8. (19) McConville, P.; Pope, J. M. Polymer 2001, 42, 3559-3568.

Analytical Chemistry, Vol. 75, No. 16, August 15, 2003 4019

reasonably high water content, can principally maintain its hydration state well. In addition, if the hydrogel contains a large amount of water, the external hydrophobic organic solvent is unable to distort the native conformation of the entrapped enzyme. In general, the rate of mass transfer for the hydrophobic substrate, however, would sharply decrease in a massive hydrogel network environment. Thus, the traditional hydrogel techniques of enzyme entrapment are mainly applied for hydrophilic solvents and substrates, but not for hydrophobic organic analytes. The hydrophilic-hydrophobic property of the enzyme-immobilized supporting material is heavily related to the mass transfer and enzyme activity of the biosensing system.6,11,20 In fact, it is more significant and important for biocatalytic reaction systems working particularly in organic solvent conditions.21,22 Common ways of addressing these contradictory requirements are to optimize the hydrophilicity structure of the supporting matrix in a unique composition.11,12,15,23 A number of satisfactory biosensing systems can be achieved by immobilizing the enzyme on the surface of such an optimal hydrophilic supporting matrix or by entrapping the enzyme within or between its thin films in multilayer structures.9,11,12,24 In general, the enzyme activity of these immobilization systems, however, would be significantly altered by the water balance between the supporting matrix and the solvent if the biocatalytic reactions were performed in waterunsaturated or water-free organic solvents.21 Consequently, it is hard for the biosensing systems to maintain their stable response signals if the retained water content within the supporting matrix is insufficient and has to be considerably balanced to the outside solvent. Moreover, for hydrophobic substrates, the biosensing systems also have difficulties in taking advantage of the enzyme entrapment in the massive supporting matrix unless the micelle technique is employed.6 In this article, we propose a novel strategy of structural optimization of the hydrophilic-hydrophobic property of the enzyme immobilization system. Our immobilization architecture is constructed with the hydrophilic hydrogel network of PVA/ HECMC or sol-gel accompanied by octadecylsilica (ODS) particles to massively entrap the enzyme for detecting hydrophobic analyte. Octadecylsilica is a typical reversed-phase silica gel and shows good mechanical properties, high hydrophobicity, and chemical inertness. The proposed enzyme immobilization multiple structure aims to reduce the mass transfer resistance of the substrate solutions within the hydrophilic immobilization hydrogel matrixes and at the same time to maintain stable and sensitive biosensing processes. Herein, we present two types of immobilized cholesterol oxidase (ChOx) that were separately entrapped in massive PVA/HECMC/ODS and sol-gel/ODS hydrogel matrixes, respectively. Currently there is a flourishing interest in monitoring and determining cholesterol because of its association with human health24 and some food products.25 In general, two main types of cholesterol biosensors, namely, amperometric26-42 and optical,43-50 (20) Diaz-Garcia, M. E.; Valencia-Gonzalez, M. J. Talanta 1995, 42, 1763-1773. (21) Zaks, A.; Klibanov, A. M. J. Biol. Chem. 1988, 263, 8017-8021. (22) Laane, C.; Boeren, S.; Vos, K.; Veeger, C. Biotechnol. Bioeng. 1987, 30, 81-87. (23) Zhang, S.; Wright, G.; Yang, Y. Biosens. Bioelectron. 2000, 15, 273-282. (24) Chaubey, A.; Malhotra, B. D. Biosens. Bioelectron. 2002, 17, 441-456. (25) Byrne, K. P. Understanding and Managing Cholesterol: A Guide for Wellness Professionals; Human Kinetics Publishers: Champaign, 1991.

4020

Analytical Chemistry, Vol. 75, No. 16, August 15, 2003

have been reported in the literature. All of them must function in solvents containing some water, such as aqueous micelle solvents, reverse micelle solvents, or water-saturated hydrophobic solvents, no matter how the ChOx is immobilized. Even though the solubility of cholesterol in aqueous micelle solution is not high and the resulting cholesterol micelle solution is not stable after few days, the aqueous micelle technique is still widely used since the whole biosensing operation is relatively simple. Unfortunately, these massive enzyme immobilization hydrogel matrixes fabricated from the common sol-gel methods33,37 are slow to be penetrated by cholesterol in aqueous micelle solution as a result of its large mass transfer resistance. For these reasons, for different applications, development of a novel sol-gel cholesterol biosensor that can operate in aqueous micelle solution with good and stable sensitivity, and at a reasonably fast response time would be significant. On the other hand, it is also a big challenge to develop new enzyme immobilization techniques for biosensors such that hydrophobic organic analytes such as cholesterol can be determined without adopting any micelle technique. Since the 1970s, optical sensors have been developing very quickly.51,52 The oxygen-quenching property of luminescent dyes and ruthenium(II) complexes have received increasing attention over the past two decades.53-55 The oxygen-sensitive silica gel (26) Hall, G. F.; Turner, A. P. F. Anal. Lett. 1991, 24, 1375-1388. (27) Tatsuma, T.; Watanabe, T. Anal. Chim. Acta 1991, 242, 85-89. (28) Oyama, N.; Ikeda, S.; Suzuki, M.; Obsaka, T. Electroanalysis 1991, 3, 665671. (29) Trettnak, W.; Lionti, I.; Mascini, M. Electroanalysis 1993, 5, 753-763. (30) Dong, S.; Deng, Q.; Cheng, G. Anal. Chim. Acta 1993, 279, 235-240. (31) Gilmartin, M. A. T.; Hart, J. P. Analyst 1994, 119, 2331-2336. (32) Boguslavsky, L.; Kalash, H.; Xu, Z.; Beckles, D.; Geng, L.; Skotheim, T.; Laurinavicius, V.; Lee, H. S. Anal. Chim. Acta 1995, 311, 15-21. (33) Yao, T.; Takashima, K. Biosens. Bioelectron. 1998, 13, 67-73. (34) Vidal, J. C.; Garcia, E.; Castillo, J. R. Anal. Chim. Acta 1999, 385, 213222. (35) Nakaminami, T.; Ito, S.-I.; Kuwabata, S.; Yoneyama, H. Anal. Chem. 1999, 71, 1068-1076. (36) Wang, H.; Mu, S. Sens. Actuators, B 1999, 56, 22-30. (37) Kumar, A.; Malhotra, R.; Malhotra, B. D.; Grover, S. K. Anal. Chim. Acta 2000, 414, 43-50. (38) Vidal, J. C.; Garcia-Ruiz, E.; Castillo, J. R. J. Pharm. Biomed. Anal. 2000, 24, 51-63. (39) Ram, M. K.; Bertoncello, P.; Ding, H.; Paddeu, S.; Nicolini, C. Biosens. Bioelectron. 2001, 16, 849-856. (40) Brahim, S.; Narinesingh, D.; Guiseppi-Elie, A. Anal. Chim. Acta 2001, 448, 27-36. (41) Pena, N.; Ruiz, G.; Reviejo, A. J.; Pingarron, J. M. Anal. Chem., 2001, 73, 1190-1195. (42) Vidal, J.-C.; Espuelas, J.; Garcia-Ruiz, E.; Castillo, J.-R. Anal. Lett. 2002, 35, 837-853. (43) Trettnak, W.; Wolfbeis, O. S. Anal. Biochem. 1990, 184, 124-127. (44) Braco, L.; Daros, J. A.; de la Guardia, M. Anal. Chem. 1992, 64, 129-133. (45) Krug, A.; Suleiman, A. A.; Guilbault, G. G.; Kellner, R. Enzyme Microb. Technol. 1992, 14, 313-316. (46) Krug, A.; Suleiman, A. A.; Guilbault, G. G. Anal. Chim. Acta 1992, 256, 263-268. (47) Valencia-Gomez, M. J.; Diaz-Garcia, M. E. Anal. Chem. 1994, 66, 27262731. (48) Marazuela, M. D.; Cuesta, B.; Moreno-Bondi, M. C.; Quejido, A. Biosens. Bioelectron. 1997, 12, 233-240. (49) Pineiro-Avila, G.; Salvador, A.; de la Guardia, M. Analyst 1998, 123, 9991003. (50) Marquette, C. A.; Ravaud, S.; Blum, L. J. Anal. Lett. 2000, 33, 1779-1796. (51) Wolfbeis, O. S. Anal. Chem. 2002, 74, 2663-2678. (52) Kuswandi, B.; Andres, R.; Narayanaswamy, R. Analyst 2001, 126, 14691491. (53) Demas, J. N.; DeGraff, B. A. Anal. Chem. 1991, 63, 829A-837A. (54) Kneas, K. A.; Demas, J. N.; Nguyen, B.; Lockhart, A.; Xu, W.; DeGraff, B. A. Anal. Chem. 2002, 74, 1111-1118.

particles adsorbed with tris(4,7-diphenyl-1,10-phenanthroline)ruthenium(II) didodecyl sulfate can be employed as an effective optical oxygen transducer for some enzymatic biosensing processes.56,57 From a practical point of view, it is still useful to incorporate an immobilized enzyme with this optical transducer so as to produce an optical biosensor which is stable and sensitive, possesses fast response, and also performs well in both aqueous and organic solvents.20 In this study, this optical oxygen transducer was separately coupled with the immobilized ChOx in the PVA/ HECMC/ODS hydrogel matrix to construct an organic-phase cholesterol biosensor or in the sol-gel/ODS matrix to form an aqueous micelle cholesterol biosensor. The response mechanism of both types of optical cholesterol biosensors was based on the enzymatic oxidation by oxygen molecules in the carriers. During the reaction, a cholesterol molecule was enzymatically oxidized by an oxygen molecule, releasing stoichiometrically equivalent hydrogen peroxide and cholest-4-en-3-one.58,59 ChOx

Cholesterol + O2 98 cholest-4-en-3-one + H2O2 (1) The depletion of the oxygen level of the reaction systems was followed by the changes in fluorescence intensity that were detected by the optical oxygen transducer. These two types of biosensing systems were characterized by stable and reasonably rapid responses toward cholesterol, and they showed rather long lifetimes, even though their responses showed a little bit of solvent dependence. It was the first time that the organic-phase cholesterol biosensor was demonstrated to perform well in either water-free or water-saturated hydrophobic organic solvents without pH disturbance or relying on any micelle technique. The organicphase cholesterol biosensor was satisfactorily employed to determine the free cholesterol content in some commercial butter samples. Compared with other reported biosensors, both proposed flow-through cholesterol biosensing systems are anticipated to be suitable for on-line monitoring of free cholesterol in the food industry. EXPERIMENTAL SECTION Reagents. Cholesterol oxidase (E.C.1.1.3.6. from Pseudomonas species) with a specific activity of 4.4 u/mg of solid, cholesterol, and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (sodium salt) (HEPES) were obtained from Sigma (St. Louis, MO). Poly(vinyl alcohol) 56-98, carboxymethyl cellulose sodium salt (medium viscosity), Borax, sodium dihydrogen phosphate, and disodium hydrogen phosphate were purchased from Fluka (Buchs, Switzerland). Octadecylsilica (32-63 µm, 100 Å) and Triton X-100 were obtained from Acros Organics (Geel, Belgium). Tetraethylorthosilica (TEOS), propan-2-ol, 4,7-diphenyl-1,10-phenanthroline, and ruthenium(III) chloride hydrate were from Aldrich Chemical (Milwaukee, WI). Teflon tape was a product of Shuang Xing Teflon Factory (Zhuhai, China). Hydroxyethyl carboxymethyl cellulose (55) Klimant, I.; Wolfbeis, O. S. Anal. Chem. 1995, 67, 3160-3166. (56) Wu, X.; Choi, M. M. F.; Xiao, D. Analyst 2000, 125, 157-162. (57) Lau, R. C. W.; Choi, M. M. F.; Lu, J. Talanta 1999, 48, 321-331. (58) Vrielink, A.; Li, J.; Brick, P.; Blow, D. Structure and Mechanism of Cholesterol Oxidase. In Molecular Recognition: Chemical and Biochemical Problems II; Roberts, S. M., Ed.; The Royal Society of Chemistry: Cambridge, 1992; pp 83-93. (59) Wallimann, P.; Marti, T.; Furer, A.; Diederich, F. Chem. Rev. 1997, 97, 1567-1608.

sodium salt was synthesized from carboxymethyl cellulose sodium salt according to a modified method.60 Tris(4,7-diphenyl-1,10phenanthroline)ruthenium(II) didodecyl sulfate dye ion pair was synthesized and oxygen-sensitive silica gel particles were prepared as described in the literature.56 Water-free toluene, n-hexane, and chloroform were prepared from HPLC-grade solvents (Aldrich, Milwaukee, WI) dried with 4-Å molecular sieve pellets. All reagents were of analytical-reagent grade or above and were used without further purification unless otherwise indicated. All aqueous solutions were prepared with deionized water (DI). Oxygen (21% v/v) balanced with nitrogen cylinder gas was purchased from Chun Wang Industrial Gases (Shenzhen, China), and 1.0% v/v oxygen balanced with nitrogen cylinder gas was from Hong Kong Oxygen & Acetylene Co., Ltd. (Hong Kong SAR, China). Enzyme Immobilization in PVA/HECMC/ODS Hydrogel Network Matrix. A PVA/HECMC stock solution was prepared by mixing 10 g of 12% (w/v) PVA aqueous solution and 5 g of 4% (w/v) HECMC aqueous solution, and its pH value was adjusted to 7.1 by a 0.025 M HEPES buffer (pH 7.1). An 8-mg portion of ChOx was dissolved in 0.10 mL of HEPES buffer (pH 7.1) containing 1.0% (w/v) Triton X-100. The enzyme solution was gently stirred with 32 mg ODS particles, which were previously washed by acetone, chloroform, and toluene, respectively, and dried at 105 °C for 3 h. After standing for 40 min, the resulting emulsoid was kept under vacuum for 20 min. Then it was mildly mixed with 0.35 mL of the PVA/HECMC stock solution and, last, spread onto a clean glass plate to form a membrane under ambient conditions within 3 h. A 0.25-mL aliquot of 1.2% (w/v) Borax solution with its pH value adjusted to 7.5 by the HEPES buffer was dropped onto the membrane to generate the hydrogel network. After 15 min, the swollen copolymer hydrogel was dried under vacuum for 75 min, then rinsed with 0.5 mL of the HEPES buffer three times to remove the residual Borax, and last, dried again under ambient conditions for ∼6 h. The water contents in the PVA/HECMC/ODS copolymer matrixes were determined by an NMR method.19 Enzyme Immobilization in Sol-Gel/ODS Matrix. A silica sol stock solution was prepared according to the literature procedure.56 A 10.4-g portion of TEOS, 1.8 g of water, 4.6 g of ethanol, and 30 µL of 0.1 M HCl were mixed and refluxed with stirring at room temperature for ∼8 h until a clear solution was obtained. A 3-mL aliquot of the silica sol stock solution was placed in a 5-mL vial and stirred under vacuum for 20 min to evaporate most of the ethanol. Then the pH of the sol solution was adjusted to ∼4.5 by adding 20 mM pH 7.4 phosphate buffer; solution A was then obtained. In a separate 5-mL vial, 12 mg of ChOx was first dissolved in 0.30 mL of 0.010 M HEPES buffer solution (pH 7.5) containing 0.15 g of Triton X-100, then mixed with solution A, and last, 60 mg of ODS was added to the mixture with gentle stirring. A vacuum was applied to the stirred mixture until a gel was formed. After 2 h, the gel was rinsed twice with 2 mL of HEPES buffer, then it was again maintained under vacuum for 15 min and, finally, kept in 1 mL of the buffer solution overnight at 4 °C. After the buffer solution was removed, the gel was allowed to dry for 6 days. Finally, the xerogel/ODS matrix was collected and was ground to powder form. (60) Li, H.; Huang, C. Y.; Yang, Z. L. Polymer Materials Science and Engineering 1998, 14, 34-37.

Analytical Chemistry, Vol. 75, No. 16, August 15, 2003

4021

Assembly of Biosensor. The typical flow-through cell was machined from stainless steel and had a chamber volume of 0.3 mL (20 mm i.d. × 1.0 mm). A small portion of oxygen-sensitive particles (∼10 mg) was evenly spread over the window of the flowthrough cell, then covered with a piece of Teflon membrane. Last, the matrix of the immobilized ChOx was closely packed into the flow-through cell to form a cholesterol biosensor. When the biosensor was not in use, it was stored at 4 °C. A schematic diagram of the flow-through cell is provided as Supporting Information. Apparatus. The cholesterol biosensor was situated in a PerkinElmer LS-50B spectrofluorometer (Buckinghamshire, U.K.). Fluorescence intensity of the biosensor was controlled and measured by an FL WinLab software (Perkin-Elmer, Buckinghamshire, U.K.). The emission intensity at 602 nm was collected at an excitation wavelength of 440 nm under batch conditions at 20 ( 2 °C and at a pressure of 101.3 kPa. All fluorescence measurements were made with 3-nm bandwidths for both the emission and excitation monochromators. A MasterFlex C/L model 7712062 peristaltic pump (Cole-Parmer Instrument Co., Chicago, IL), or a homemade liquid flow controller was used to deliver the cholesterol solutions through the cholesterol biosensors at flow rates of 0-1.2 mL/min. Unless otherwise stated, all solvents and solutions were presaturated by 21% v/v oxygen gas for 20 min before measurement. Proton nuclear magnetic resonance (1H NMR) spectra were recorded on an INOVA 400-MHz NMR spectrometer (Varian, Palo Alto, CA). Determination of Free Cholesterol in Butter Samples. The standard addition method was employed to determine the content of free cholesterol in the samples. For each sample, five samples of 5.00 g each of butter were individually dispersed with 15 mL toluene in five separating funnels. Cholesterol standard toluene solution (10.0 mM) volumes of 0.0, 1.0, 2.0, 3.0, and 4.0 mL were added into the individual separating funnels. Into each separating funnel, 2 mL of water was again introduced. After mixing and standing for 20 min, the insoluble residue at the lower water layer was discharged. The upper clear toluene solution was washed using 2 mL of DI water three times, transferred into a 25-mL volumetric flask, and topped with pure toluene to the mark. The solutions were then subjected to determination by the PVA/ HECMC/ODS optical cholesterol biosensor. RESULTS AND DISCUSSIONS Optical Response of Cholesterol Biosensors. Cholesterol and oxygen had to diffuse into the enzyme immobilization phase of the hydrogel network matrixes and to the enzyme, while cholest-4-en-3-one and other products had to diffuse out. The enzymatic reaction induced the depletion of the oxygen level surrounding the enzyme so that an oxygen gradient was created between and within two segments, and finally, induced the response of the oxygen transducer. The optical oxygen sensing relied on the fluorescence quenching of tris(4,7-diphenyl-1,10phenanthroline)ruthenium(II) didodecyl sulfate molecules by oxygen molecules.53,55 Figures 1 and 2 separately display the response curves of the two types of biosensors on exposure to successive step changes of various concentrations of cholesterol in the aqueous micelle and toluene solutions, respectively. In both biosensing systems, an increase in the fluorescence intensity was observed in the presence of cholesterol as oxygen was consumed in the enzymatic reaction. The higher the cholesterol concentra4022

Analytical Chemistry, Vol. 75, No. 16, August 15, 2003

Figure 1. Response curves of the sol-gel/ODS cholesterol biosensor subjected to various concentrations of cholesterol. The testing solutions were composed of various concentrations of cholesterol in aqueous micelle solutions of phosphate buffer (50 mM, pH 7.4), propan-2-ol, and Triton X-100 (43:5:2; v/v): (1) 0.0, (2) 1.0, (3) 3.0, and (4) 6.0 mM free cholesterol.

Figure 2. Response curves of the PVA/HECMC/ODS cholesterol biosensor subjected to various concentrations of cholesterol in toluene: (1) 0, (2) 0.50, (3) 1.00, (4) 2.50, (5) 5.00, (6) 7.50, (7) 10.0, and (8) 15.0 mM free cholesterol.

tion, the larger the fluorescence intensity was recorded. The overall process correlates to the concentration of cholesterol in the carriers, which can be quantified by intensity quenching measurements and described as the following equation.56

RS ) (Itest)/(Ibaseline)

(2)

RS is defined as the relative signal change of the biosensor, which is a function of the cholesterol concentration; the terms of Itest and Ibaseline represent the detected fluorescence signals of the biosensor when it is exposed to the cholesterol solution and the solvent, respectively. Typical calibration curves of the biosensors are shown in Figure 3. The sol-gel/ODS cholesterol biosensor has an analytical working range from 0.05 to 8.0 mM cholesterol,

Figure 3. Calibration curves at various concentrations of free cholesterol: (1) the sol-gel/ODS cholesterol biosensor in aqueous buffered micelle solution and (2) the PVA/HECMC/ODS cholesterol biosensor in toluene.

and the signal ratio of the maximum to baseline is over 3.6-fold, with the detection limit ∼50 µM (S/N ) 3) when it operates in aqueous micelle solution. For the PVA/HECMC/ODS cholesterol biosensor in the toluene solvent system, the analytical working range is from 0.07 to 18.0 mM cholesterol, and the signal ratio of the maximum to baseline is nearly 6-fold, with the detection limit ∼70 µM (S/N ) 3). Because the solubility of cholesterol in the aqueous micelle solution is much lower than that in toluene, the working concentration range for the sol-gel/ODS cholesterol biosensor would be narrower than that of the PVA/HECMC/ODS cholesterol biosensor. On the other hand, since the dissolved oxygen concentration is higher in toluene than that in the aqueous micelle solvent, the percentage of oxygen consumption is smaller for the organic-phase cholesterol biosensor. This is even more so when the enzymatic oxidization reactions take place in test solutions in which the cholesterol concentrations are near both ends of the detection limits. Consequently, the sol-gel/ODS cholesterol biosensor could acquire a better detection limit when the same oxygen transducer was used. In brief, both biosensors can provide efficacious responses to free cholesterol in aqueous micelle solution or toluene solvent, respectively, although the relationships of the RS and the cholesterol concentration are not linear across the whole analytical working ranges. Composition of Immobilization Hydrogel Matrix. Although both hydrogel network structures have been commendably proved to possess some excellent properties for enzyme immobilization, our preliminary studies suggested that neither the cholesterol aqueous micelle solution nor the cholesterol hydrophobic organic solution could efficiently diffuse through the normal sol-gel monolithic film or the copolymer membrane of the PVA/HECMC hydrogel. In addition, the PVA/HECMC hydrogel network would lose its permeability in any aqueous solution within a few minutes because of its extremely high water uptake during the swelling process.16,18 When ChOx was solely entrapped in the sol-gel or PVA/HECMC hydrogel networks, the immobilized enzyme showed very low and unstable apparent biocatalytic activity to cholesterol. However, once the hydrogels had been doped with ODS particles, the resulting sol-gel/ODS matrix was then well-penetrated by

Figure 4. Response curves of the PVA/HECMC/ODS cholesterol biosensor toward 12.5 mM cholesterol in (1) toluene, (2) n-hexane, and (3) chloroform solutions. All solutions were presaturated by 1% oxygen gas for 20 min before measurements.

Figure 5. Effect of temperature on the response of the sol-gel/ ODS cholesterol biosensor switching from 0.0 to 1.0 and back again to 0.0 mM aqueous micelle cholesterol solution at various temperatures: (1) 18, (2) 25, (3) 31, (4) 37, and (5) 42 °C.

the cholesterol micelle solutions, while the resulting PVA/ HECMC/ODS copolymer hydrogel matrix was penetrated by the cholesterol organic solution. When PVA/HECMC copolymer solution was in situ gelled by the Borax solution7 together with the ODS particles, on which ChOx was previously deposited, the resulting swollen hydrogel matrixes showed remarkable and stable biocatalytic activity toward cholesterol in toluene, n-hexane, or chloroform, as displayed in Figures 2 and 4. Similarly, when ChOx was immobilized in sol-gel matrix by the silicate polymerization process in the presence of ODS particles, the resulting immobilized-enzyme matrix showed quite high biocatalytic activities toward cholesterol in the aqueous micelle solution, as displayed in Figures 1 and 5. Moreover, both types of cholesterol biosensors could have some small responses when the amount of sol solution used to encapsulate ChOx was as large as 6 mL and 120 mg ODS particles or when the PVA/HECMC copolymer solution used was up to 0.65 mL and 60 mg of ODS particles. On the other hand, if silica gel particles were used instead of ODS Analytical Chemistry, Vol. 75, No. 16, August 15, 2003

4023

particles, the responses of both types of biosensors were rather weak and unstable, which was very similar to that without the silica gel particles. Their responses would disappear in a very short time (i.e.,