Covalent Immobilization of Glucose Oxidase on Well-Defined Poly

The Si-g-P(GMA) chips were transferred to 3 mL of the 0.1 M PBS (pH 7.4) containing ... The telescope with a magnification power of 23× was equipped ...
0 downloads 0 Views 154KB Size
Biomacromolecules 2005, 6, 1012-1020

1012

Covalent Immobilization of Glucose Oxidase on Well-Defined Poly(glycidyl methacrylate)-Si(111) Hybrids from Surface-Initiated Atom-Transfer Radical Polymerization F. J. Xu, Q. J. Cai, Y. L. Li, E. T. Kang,* and K. G. Neoh Department of Chemical and Biomolecular Engineering, National University of Singapore, Kent Ridge, Singapore 119260 Received October 27, 2004; Revised Manuscript Received December 6, 2004

A simple one-step procedure was employed for the covalent immobilization of an atom-transfer radical polymerization (ATRP) initiator, via the robust Si-C bond, on the hydrogen-terminated Si(111) surface (Si-H surface). Well-defined poly(glycidyl methacrylate) [P(GMA)] brushes, tethered directly on the (111)oriented single-crystal silicon surface, were prepared via surface-initiated ATRP. Kinetics study on the surfaceinitiated ATRP of glycidyl methacrylate revealed that the chain growth from the silicon surface was consistent with a “controlled” process. A relatively high concentration of glucose oxidase (GOD; above 0.2 mg/cm2) could be coupled directly to the well-defined P(GMA) brushes via the ring-opening reaction of the epoxide groups with the amine moieties of the enzyme. The resultant GOD-functionalized P(GMA) brushes, with the accompanying hydroxyl groups from the ring-opening reaction of the epoxide groups, serves as an effective spacer to provide the GOD with a higher degree of conformational freedom and a more hydrophilic environment. An equivalent enzyme activity above 1.6 units/cm2 [µmoles of β-D-(+)-glucose oxidized to D-gluconolactone per minute per square centimeter] and a corresponding relative activity of about 60% could be readily achieved. The immobilized GOD also exhibited an improved stability during storage over that of the free enzyme. The GOD-functionalized silicon substrates are potentially useful to the development of silicon-based glucose biosensors. Introduction Glucose biosensors have attracted a great deal of interest because of the increasing incidence of diabetes.1,2 The enzyme, glucose oxidase (GOD), is well-known as a biological sensing material for the quantitative determination of β-Dglucose in solution because of its substrate specificity.3,4 For the GOD catalyzed glucose oxidation process, the general reaction mechanism is as follows:4-6 glucose + GO(FAD) f gluconolactone + GO(FADH2), where GO(FAD) and GO(FADH2) represent the oxidized and reduced forms of GOD, respectively. The subsequent reaction for the regeneration of GOD is given by GO(FADH2) + O2 f GO(FAD) + H2O2. A number of techniques for enzyme immobilization on various matrixes, such as covalent linkage, encapsulation, layer-by-layer deposition, and cross-linking, have been developed for the fabrication of biosensing devices.4-10 Some matrixes, such as polymer films, polymer gels, and conducting polymers, have been used for GOD immobilization in the development of glucose biosensors.3-5,11,12 Silicon-based biosensors and biochips are of increasing importance in biomedical engineering and biotechnology.13-18 The main techniques reported for immobilization of enzymes on silicon substrates include silanization, metal linking, gel entrapment, silicon-carbon surface attachment, and poly(L-lysine) attachment.15,16 Stable polymer brushes covalently * To whom all correspondence should be addressed. Tel: +65-68742189. Fax:+65-6779-1936. E-mail: [email protected].

tethered on silicon surfaces can provide excellent mechanical and chemical protection to the substrate, alter the electrochemical characteristics of the interface, and provide new pathways to the functionalization of silicon surfaces for molecular recognition and sensing.19-23 Atom-transfer radical polymerization (ATRP) is a recently developed “living” or “controlled” radical polymerization method.24-28 It is possible to prepare well-defined polymer brushes on various types of substrates via surface-initiated ATRP.28-31 In the preparation of polymer brushes, ATRP initiators were immobilized on the substrate surfaces, usually in multistep processes. For the silicon substrates, the ATRP initiators can be coupled to the substrate surfaces via organosilanes28,32-35 or via the robust Si-C bonds.36-38 In the present study, an alternative one-step approach to the covalent attachment (Si-C bonding) of an ATRP initiator, 4-vinylbenzyl chloride (VBC), on the hydrogenterminated Si(111) surface (Si-H surface) via radicalinduced hydrosilylation was adopted. Well-defined poly(glycidyl methacrylate) [P(GMA)] brushes covalently tethered on silicon surfaces [Si-g-P(GMA) hybrids] were prepared, via surface-initiated ATRP, on the VBC-coupled silicon (SiVBC) surface (Figure 1). P(GMA) is a known surface linker and spacer for biomolecules.39-45 In this work, the epoxide groups of the grafted P(GMA) brushes were used for the direct coupling of GOD, with the concomitant formation of hydroxyl groups (Figure 1). With the simultaneous presence of the surface spacer and the neighboring hydroxyl groups,

10.1021/bm0493178 CCC: $30.25 © 2005 American Chemical Society Published on Web 01/08/2005

Covalent Immobilization of GOD on P(GMA)-Si(111)

Biomacromolecules, Vol. 6, No. 2, 2005 1013

Figure 1. Schematic diagram illustrating the processes of radical-initiated hydrosilylation of VBC with the Si-H surface to give the Si-VBC surface, surface-initiated ATRP of GMA from the Si-VBC surface at room temperature, and GOD immobilization on the Si-g-P(GMA) surface.

GOD can assume a higher degree of conformational freedom in a more hydrophilic environment. The surface microstructure is advantageous to retaining the activity of the immobilized enzymes.5 The compositions of the modified silicon surfaces were investigated by X-ray photoelectron spectroscopy (XPS). The amount and activity of the immobilized GOD as a function of the P(GMA) thickness were studied in detail. The GOD-functionalized Si-g-P(GMA) hybrids may find useful applications in silicon-based glucose biosensors. Experimental Section 2.1. Materials. (111)-oriented single crystal silicon, or the Si(111) wafer, with a thickness of about 1.5 mm and a diameter of 150 mm, was purchased from Unisil Co. of Santa Clara, CA. The as-received wafers were polished on one side and doped lightly as n-type. The silicon wafers were sliced into square chips of 2 cm × 3 cm in size. To remove the organic residues on the surface, the silicon substrate was washed with the “piranha” solution, consisting of 98 wt % concentrated sulfuric acid (70 vol %) and hydrogen peroxide (30 vol %).21,22 Caution: piranha solution reacts violently with organic materials and should be handled carefully! After rinsing with copious amounts of doubly distilled water, the silicon chips were blown dried with purified argon. The pristine (oxide-covered) silicon chips were immersed in 10 vol % Hydrofluoric acid (HF; Caution: strong acid, handle very carefully!) in individual Teflon vials for 2 min to remove the oxide film and leave behind a uniform Hterminated Si(111) surface (Si-H surface). Details on the preparation of the Si-H surface have been described earlier.21,22 HF (37 wt %), benzoyl peroxide (BPO), VBC, glycidyl methacrylate (GMA), 2,2′-byridine (Bpy), copper(I) chloride, copper(II) chloride, dimethylsulfoxide (DMSO), and dimethylformamide (DMF) were obtained from Aldrich Chemical Co. of Milwaukee, WI. GMA was passed through a silica gel column to remove the inhibitor and stored under an argon atmosphere at -10 °C. GOD (type II, 15 500 units g-1 from Aspergillus niger) was purchased from Sigma Chemical Co., St. Louis, MO. Dulbecco’s phosphate buffered saline (PBS, pH 7.4) solution, used for the enzyme immobilization work, was freshly prepared. Bio-Rad dye reagent for protein assay

(catalog no. 500-0006) was obtained from Bio-Rad, Inc., Hercules, CA. 2.2. Immobilization of the ATRP Initiator on the Si-H Surface. The initiator was immobilized via radical-induced hydrosilylation46 of VBC with the Si-H surface to give rise to a covalently bonded (Si-C bonded) monolayer (the SiVBC surface). The process is shown schematically in Figure 1. For the radical-induced hydrosilylation reaction, the Si-H surface was treated with 1 mL of VBC in 6 mL of DMSO for 60 min at 85 °C, in the presence of BPO (10 mg) as the radical initiator and under a dry argon atmosphere, to give rise to the Si-VBC surface. After the hydrosilylation reaction, the Si-VBC substrate was immersed in about 200 mL of acetone (a good solvent for VBC and VBC polymer) for about 24 h. The solvent was stirred continuously and changed every 8 h. The substrate was subsequently rinsed with copious amounts of fresh acetone to ensure the complete removal of the adhered and physically adsorbed VBC or VBC oligomers. 2.3. Surface-Initiated ATRP of GMA. For the preparation of GMA polymer [P(GMA)] brushes on the Si-VBC surface, the reaction was carried out using a [GMA]/[CuCl]/ [CuCl2]/[Bpy] feed ratio of 100:1:0.2:2 in 4 mL of a mixed solvent [DMF/water (1:1, v/v)] at room temperature in a Pyrex tube containing the Si-VBC substrate. The reaction was allowed to proceed for a predetermined period of time to give rise to the Si-g-P(GMA) surface. Surface-initiated ATRP of other monomers on the surface-functionalized silicon substrates had been described earlier.21,22 After the reaction, the Si-g-P(GMA) hybrid was washed thoroughly by extraction with acetone. The hybrid was subsequently immersed in a large volume of acetone for about 48 h to ensure the complete removal of the adhered and physically adsorbed polymer, if any. The solvent was stirred continuously and changed every 8 h. 2.4. Immobilization of GOD on the Si-g-P(GMA) Surface. The Si-g-P(GMA) hybrid chips of 1 cm × 1 cm in area were used in all experiments with GOD. The Si-gP(GMA) chips were transferred to 3 mL of the 0.1 M PBS (pH 7.4) containing GOD at a concentration of 4 mg mL-1. The immobilization reaction was allowed to proceed at room temperature for a predetermined period of time under continuous stirring. Coupling reactions involving epoxide groups and -NH2 groups of biomolecules had been widely

1014

Biomacromolecules, Vol. 6, No. 2, 2005

reported.39-45 The nucleophilic -NH2 groups of GOD were, thus, expected to react readily and irreversibly with the reactive epoxide groups of the Si-g-P(GMA) surface to give rise to the Si-g-P(GMA)-GOD surface (Figure 1). The reversibly bound GOD on the Si-g-P(GMA)-GOD surface was desorbed in copious amounts of PBS for 24 h at room temperature. The PBS solution was gently stirred and changed every 8 h. The method of GOD immobilization used in the present work, thus, differed from that reported for the immobilization of enzymes on the COOH moiety-functionalized substrates. These substrates usually required preactivation by a coupling agent, such as the water-soluble carbodiimides.4,5,52 2.5. Determination of the Immobilized GOD Concentration. The amount of enzyme immobilized on the Si-gP(GMA)-GOD surface was determined by the modified dye-interaction methods,48,49 using the Bio-Rad protein dye reagent. For the preparation of the dye solution, the BioRad stock dye solution was diluted five times with doubly distilled water. A GOD solution (100 µL) of known concentration was added to 5 mL of the dye solution. The GOD-dye solution was gently stirred for 3 h and centrifuged at 5000 rev/min for 15 min. The GOD-dye complex was precipitated, and the free dye remained in the upper layer. The absorbance of supernatant at 465 nm was used for the standard calibration. For the quantitative determination of immobilized GOD on the silicon surface, the dye solution (5 mL) was added to a test tube containing the Si-gP(GMA)-GOD chip of 1 cm × 1 cm in size. The dye solution was gently stirred for 3 h. The chip was removed, and the absorbance of the dye solution was measured at 465 nm. The amount of GOD immobilized on the chip was calculated with reference to the standard calibration curve. 2.6. Assay of GOD Activity. For the investigation of activity of the immobilized GOD, 5 mL of β-D-(+)-glucose solution (1.8 wt %) was used as the assay medium. The activity was deduced from the consumption rate of the β-D(+)-glucose.7,9 The enzymatic reaction was initiated by the introduction of a Si-g-P(GMA)-GOD chip of 1 cm × 1 cm in size into the assay medium at room temperature with constant agitation. The concentration of the β-D-(+)-glucose solution was measured on a YSI model 2700 SELECT biochemistry analyzer (YSI Inc., Yellow Springs, OH). The activity of the immobilized GOD in units was defined as the number of micromoles of β-D-(+)-glucose oxidized to D-gluconolactone per minute. The relative activity (RA) was defined as the ratio of the observed surface activity over the activity obtained from an equivalent amount of the free enzyme. 2.7. Surface Characterization. The chemical composition of the modified silicon surfaces was determined by XPS. The XPS measurements were performed on a Kratos AXIS HSi spectrometer using a monochromatized Al KR X-ray source (1486.6 eV photons) and procedures similar to those described earlier.21,22 The static water contact angles of the pristine and functionalized Si-H surfaces were measured at 25 °C and 60% relative humidity, using the sessile drop method with a 3-µL water droplet, in a telescopic goniometer [Rame-Hart model 100-00-(230), manufactured by the Rame-

Xu et al.

Hart, Inc., Mountain Lakes, NJ]. The telescope with a magnification power of 23× was equipped with a protractor of 1° graduation. For each angle reported, at least three measurements from different surface locations were averaged. The angle reported was accurate to (3°. The thickness of the polymer brushes grafted on the silicon substrate was determined by ellipsometry.21,22 The measurements were carried out on a variable angle spectroscopic ellipsometer (model VASE, J. A. Woollam, Inc., Lincoln, NE) at incident angles of 70 and 75° and in the wavelength range 2001000 nm. For each sample, the thickness measurements were made on at least four different surface locations. Data were recorded and processed using the WVASE32 software package.21,22 3. Results and Discussion 3.1. Immobilization of the ATRP Initiator on the Si-H Surface. The use of Si-H surfaces, instead of Si/SiO2 surfaces, allowed the preparation of polymer-Si hybrids with robust Si-C linkages. The stability and chemical robustness of the Si-C bonds had been reported earlier.36-38 The ATRP initiators were immobilized via radical-induced hydrosilylation46 of VBC with the Si-H surface. Thus, using BPO as the radical initiator, a stable VBC initiator monolayer was formed on the Si-H surface via the Si-C bonds (the SiVBC surface). The mechanism of radical-initiated hydrosilylation is similar to that of the UV-induced hydrosilylation described earlier.21,22,46,50 The initiator, BPO, undergoes decomposition to form two alkyl radicals and carbon dioxide. The alkyl radical abstracts an H atom from the adjacent Si-H group to give rise to a silicon radical. The latter reacts readily with an alkene and forms a surface-tethered alkyl radical on the β carbon. The radical subsequently abstracts an H atom from the adjacent Si-H bond. The abstraction creates a new reactive silicon radical to allow the above reaction to propagate as a chain reaction on the Si-H surface. The presence of radical-initiated hydrosilylation of VBC on the Si-H surface can be ascertained by comparing the XPS wide scan and Cl(2p) core-level spectra of the Si-H surface (Figure 2a,b) with those of the resultant Si-VBC surface (Figure 2c,d). The latter surface had been subjected to the vigorous washing/extraction procedures after preparation (see Experimental Section). In addition to the significantly enhanced C(1s) signal intensity, a new Cl(2p) spectrum at the binding energy (BE) of about 200 eV, characteristic of the covalently bonded Cl,51 has appeared on the Si-VBC surface. The static water contact angle increased from about 72° for the Si-H surface to about 84° for the Si-VBC surface (Table 1). The increase in surface hydrophobicity is consistent with an extensive coverage of the Si-H surface by VBC. Experimental and theoretical studies have suggested that a maximum of about 60% of the hydrogen sites on the Si-H surface are substituted by an alkyl group at complete surface coverage.38 A control experiment, in which the Si-H substrate was soaked in the similar VBC solution for the same period of time, but in the absence of BPO, was also carried out. No elemental signals associated with the VBC molecules were detected on the

Covalent Immobilization of GOD on P(GMA)-Si(111)

Biomacromolecules, Vol. 6, No. 2, 2005 1015

Figure 2. Wide scan and Cl(2p) core-level spectra of (a, b) the Si-H surface and (c, d) the Si-VBC surface. Table 1. Static Water Contact Angle, Amount of Immobilized GOD, and EA of the GOD-Functionalized Surfaces sample

static contact anglea ((3°)

immobilized GOD (mg/cm2)

EAb (units/cm2)

Si-g-P(GMA)c Si-g-P(GMA)-GODd Si-g-P(GMA)-GODe

69 54 53

0.00 0.17 0.23

0.00 1.50 1.79

Si-g-P(GMA)-GODf Si-g-P(GMA)-GODg

53 52

1.67 1.60

a Static water contact angles for the pristine (oxide-covered) Si(111) surface, the Si-H surface, and the Si-VBC surface are about 20, 72, and 84°, respectively. b The activity of the immobilized GOD in units was defined as micromoles of β-D-(+)-glucose oxidized to D-gluconolactone per minute. c Reaction conditions: [GMA]/[CuCl]/[CuCl2]/[Bpy] ) 100:1: 0.2:2 in DMF/water (1:1, v/v) at room temperature for 5 h; P(GMA) thickness ) 51 nm. d Obtained at the GOD immobilization time of 0.5 h. The starting surface corresponds to the Si-g-P(GMA) surface. e Obtained at the GOD immobilization time of 5 h. The starting surface corresponds to the Si-g-P(GMA) surface. f Obtained after storage in air at 4 °C for 14 days. The starting surface corresponds to the Si-g-P(GMA)-GOD surface. g Obtained after storage in PBS solution at 4 °C for 14 days. The starting surface corresponds to the Si-g-P(GMA)-GOD surface.

Si-H surface after the substrate was rinsed briefly with acetone and subjected to surface analysis by XPS. Thus, benzyl chloride has been successfully immobilized on the Si-H surface, only in the presence of BPO, to cater for the subsequent surface-initiated ATRP of GMA from the Si-

VBC surface. On the basis of the VBC monolayer thickness of about 0.3 nm (as determined by ellipsometry), VBC density of 1.08 g/cm3, and VBC molecular weight of 153 g/mol, the initiator density for the Si-VBC surface was estimated to be about 1.4 initiators/nm2. 3.2. Si-VBC Surface-Initiated ATRP of GMA. P(GMA) is a potential surface linker for biomolecules and has promising applications in advanced biotechnologies, such as DNA separation and proteins (antibodies and enzymes) immobilization.39-45 In this work, the method of addition of Cu(II) complex (CuCl2) was chosen to control the concentration of the deactivating Cu(II) complex21,22 during the surface-initiated ATRP process on the Si-VBC surface. The ratio of [GMA (monomer)]/[CuCl (catalyst)]/[CuCl2 (deactivator)]/[Bpy (ligand)] was controlled at 100:1:0.2:2. The presence of grafted GMA polymer [P(GMA)] on the Si-VBC surface was confirmed by XPS analysis and ellipsometry measurement after the surface had been subjected to vigorous washing and extraction. Figure 3a shows the C(1s) core-level spectrum of the silicon surface with grafted P(GMA) brushes [Si-g-P(GMA) surface] at an ATRP time of 5 h. The C(1s) core-level spectrum can be curvefitted into three peak components with BEs at about 284.6, 286.2, and 288.4 eV, attributable to the CsH, CsO, and

1016

Biomacromolecules, Vol. 6, No. 2, 2005

Xu et al.

Figure 3. C(1s) and N(1s) core-level spectra of (a, b) the Si-g-P(GMA) surface obtained at the ATRP time of 5 h, (c, d) the Si-g-P(GMA)-GOD surface obtained at the GOD immobilization time of 0.5 h, and (e, f) the Si-g-P(GMA)-GOD surface obtained at the GOD immobilization time of 5 h. These surfaces correspond to those described in Table 1.

OdCsO species, respectively, in an approximate ratio of 3:3:1 and consistent with the chemical structure of P(GMA).51 The thickness and water contact angle of the corresponding P(GMA) layer are about 51 nm and 68°, respectively (Table 1). The degree of polymerization (DP) of the Si-g-P(GMA) surface is estimated to be about 154, based on the surface initiator density of 1.4 initiators/nm2 and P(GMA) density of 1.0 g/cm3. An approximately linear increase in P(GMA) thickness and DP of the grafted P(GMA) chains on the Si-VBC surface with polymerization time is observed, as shown by lines a and b, respectively, of Figure 4. The results suggest that the chain growth from the SiVBC surface was consistent with a “controlled” or “living” process.21,22 The persistence of “living” P(GMA) chain ends from surface-initiated ATRP on the Si-H surfaces has also been ascertained by the subsequent growth of a second polymer block, using the P(GMA) brushes as the macroinitiators.52 Control experiments on the pristine (oxide-covered) Si(111) and Si-H surfaces revealed that no increase in organic layer thickness was detected when the two substrates

Figure 4. Dependence of (a) thickness and (b) DP of the grafted P(GMA) chains of the Si-g-P(GMA) surface on the surface-initiated ATRP time.

were subjected to the “surface-initiated” ATRP of GMA under similar reaction conditions. The above results indicate

Covalent Immobilization of GOD on P(GMA)-Si(111)

that GMA has been successfully graft-polymerized on the Si-VBC surfaces via surface-initiated ATRP. 3.3. Immobilization of GOD on the Si-g-P(GMA) Surface. The epoxy groups of the Si-g-P(GMA) surface can react readily and irreversibly with nucleophilic groups, such as -NH2, -SH, and -COOH. Thus, the Si-g-P(GMA) surface with a high density of epoxide groups is well-suited for the immobilization of proteins, enzymes and other biomolecules. Nucleophilic reactions invoving -NH2 moieties of biomolecules and epoxide groups have been widely reported.39-45 In the present work, the reaction of GOD with the Si-g-P(GMA) surface gave rise to the GOD-functionalized silicon surface, or the Si-g-P(GMA)-GOD surface (Figure 1). After the vigorous extraction of the reversibly bound GOD on the Si-g-P(GMA)-GOD surfaces, the C(1s) and N(1s) core-level spectra of the Si-g-P(GMA)-GOD surfaces are compared to those of the starting Si-g-P(GMA) surface in Figure 3. The C(1s) and N(1s) spectral line shapes of the Si-g-P(GMA)-GOD surfaces are significantly different from the corresponding spectral line shapes of the original Si-g-P(GMA) surface. The C(1s) core-level spectra of the Si-g-P(GMA)-GOD surfaces obtained at the GOD immobilization time of 0.5 h (Figure 3c) and 5 h (Figure 3e) can be curve-fitted into five peak components with BEs at about 284.6, 285.5, 286.2, 287.8, and 288.4 eV, attributable to the CsH, CsN, CsO, OdCNH, and OdCsO species, respectively.51 The CsN peak component is associated with the linkages in GOD, as well as the linkages between P(GMA) and GOD. The OdCNH peak component is associated with the peptide bonds in GOD itself. Earlier studies have suggested that most of the amine moieties in the enzyme can become involved in the covalent attachment.53,54 The Si-g-P(GMA) surface becomes more hydrophilic after GOD immobilization, and the contact angle decreases to about 53° (Table 1). Increasing the GOD loading on the surface from 0.17 mg/cm2 to 0.23 mg/cm2 did not result in a further decrease in the water contact angle of the surface. The phenomenon can probably be attributed to the complete coverage of the Si-g-P(GMA) surface by GOD at the surface loading of 0.17 mg/cm2. The further increase in the surface concentration of GOD probably has involved immobilization of the enzyme in the subsurface region, as is also suggested by the sluggish increase in the surface concentration of GOD as a function of the P(GMA) layer thickness at GOD loading above 0.17 mg/cm2 (see Figure 6a). The above results, the appearances of a strong N(1s) signal at the BE of about 399.2 eV (Figure 3d,f), characteristic of the amine species,51 and the increase in the surface [N]/[C] ratio with the immobilization time are consistent with the fact that GOD has been covalently immobilized on the Si-g-P(GMA) surface. The surface concentration of immobilized GOD can be expressed as the weight of immobilized GOD per area of the Si-g-P(GMA)-GOD surface, as determined using the protein-dye interaction method.48,49 The amount of immobilized GOD (Figure 5) increases with the GOD immobilization time. When the immobilization time reaches about 4-5 h, the surface concentration of immobilized GOD levels off. Therefore, a 5-h immobilization time was adopted

Biomacromolecules, Vol. 6, No. 2, 2005 1017

Figure 5. Dependence of the amount of covalently immobilized GOD of the Si-g-P(GMA)-GOD surface on the immobilization time.

in the present work. The overall concentrations of immobilized GOD on the Si-g-P(GMA)-GOD surfaces, obtained at the GOD immobilization times of 0.5 and 5 h, are about 0.17 and 0.23 mg/cm2 (Table 1), respectively. As the maximum amount of GOD immobilized on a 1 cm × 1 cm silicon chip, from a GOD buffer solution containing 12 mg of GOD (3 mL × 4 mg/mL, see Experimental Section), was about 0.23 mg/cm2, approximately 98% of GOD remained in the immobilization solution. Thus, the amount of GOD immobilized was not limited by external diffusion. With the increase in thickness of the grafted P(GMA) layer on the Si-g-P(GMA) surface, the amount of covalently immobilized GOD is expected to increase. Figure 6a shows the increase in GOD concentration of the Si-g-P(GMA)GOD surface, obtained at the GOD immobilization time of 5 h, as a function of the P(GMA) layer thickness. Thus, the amount of immobilized GOD increases rapidly and then levels off gradually with the P(GMA) layer thickness. As the thickness of the P(GMA) layer increases, steric hindrance from the spatially distributed P(GMA) graft chains on the silicon surface also increases and begins to limit the accessibility of GOD only to the epoxide groups in the top surface layer of the P(GMA) brushes. Nevertheless, because the Si-g-P(GMA) surface contains a high density of epoxide groups, a large amount of GOD (above 0.2 mg/cm2) can be readily immobilized, compared to the maximum amount (below 0.16 mg/cm2) of GOD immobilized on polymer films via amide linkage,4,55-57 adsorption,58 and reversible immobilization.59 3.4. Assay of Immobilized GOD Activity. The enzyme activity (EA) and RA of the covalently immobilized GOD on the Si-g-P(GMA)-GOD surface, obtained at a GOD immobilization time of 5 h, as a function of the P(GMA) thickness are shown in Figure 6b. As the P(GMA) thickness increases, the observed EA increases and then gradually levels off at moderate to large thicknesses of the grafted P(GMA) layer. The initial increase in the observed EA must be associated with the increase in the amount of surface immobilized enzyme. The slow increase in EA at moderate to large thicknesses of the grafted P(GMA) layer suggests an increase in diffusion limitation of the glucose substrates to the enzymatic sites. A fraction of the GOD molecules is

1018

Biomacromolecules, Vol. 6, No. 2, 2005

Figure 6. Dependence of (a) the amount and (b) the EA and RA of the covalently immobilized GOD of the Si-g-P(GMA)-GOD surface on the thickness of the grafted P(GMA) layer (GOD immobilization time ) 5 h).

probably embedded in the grafted P(GMA) layer and becomes less accessible to the glucose substrates. In the determination of enzyme activities and stability, an excess amount of glucose substrate was always employed to ensure the absence of diffusion limitation of the enzyme substrate. For instance, at the maximum EA of 1.79 units/cm2 (Table 1), about 88% of β-D-(+)-glucose remained in the assay solution. The RA of the immobilized enzyme decreases gradually with the increase in P(GMA) thickness. The immobilized enzyme retains about 55-65% of the activity of an equivalent amount of the free enzyme (Figure 6b). The decrease in activity is a phenomenon commonly observed in covalently immobilized enzymes55 and is usually interpreted in terms of modification in the tertiary structure of the covalently bonded enzyme or the diffusion limitation of the glucose substrates to the active enzymatic sites arising from the increase in steric hindrance.60 In fact, in the absence of diffusion limitation of the glucose substrates, persistently high RAs (>80%) have been reported for the lower surface concentrations of GOD (0.036 mg/cm2 and 0.023 mg/cm2, respectively) immobilized on polyamide membranes via the glutaraldehyde method56,57 and on ultrafiltration membranes of the acrylonitrile and N-vinylimidazole copolymers.61 In the present work, an EA above 1.6 units/cm2 (1 unit ) 1 µmol of β-D-(+)-glucose oxidized to D-gluconolactone per min) can be readily achieved, compared to the commonly reported activities (below 0.7 units/cm2) of GOD immobilized on polymer films and mmbranes.4,5,9,56,62 In addition, a

Xu et al.

relatively high RA (above 55%) is obtained for GOD immobilized on the well-defined P(GMA) brushes, compared to the maximum RA (below 35%) of GOD immobilized by other methods, such as via amide linkages4,5,63 or entrapping methods.64 The effect of the molecular spacer on the activity of the immobilized enzymes has been described.55,65 A spacer inserted between the substrate surface and the immobilized enzyme molecule helps to retain the activity of the immobilized enzyme. The spacers can reduce the deformation of the immobilized enzymes considerably. In addition, a hydrophilic micro-environment is advantageous to the retention of the activity of the immobilized enzymes.5 In the present work, the epoxide groups of the grafted P(GMA) chains reacted with the nucleophilic -NH2 group of GOD to covalently immobilize the GOD, with the accompanied formation of the hydroxyl groups (Figure 1). The resultant GOD-functionalized P(GMA) chain with the hydroxyl groups acts as a spacer to provide GOD with a higher degree of conformational freedom and a more hydrophilic environment. On the other hand, the present method also avoids the negative effects of pH of the environment on the EA. For example, for enzymes immobilized on acrylic acid (AAc) polymer-modified surfaces,4,5,63 the EA was adversely affected by the decrease in pH of the film surface associated with the surface-grafted AAc polymer. pH is known to be a major factor contributing to the dissociation of key functional groups within the active sites of enzymes.66 Finally, the use of coupling agents can also affect the properties of immobilized enzymes.9,67 Direct coupling of P(GMA) with GOD, thus avoiding the use of a water-soluble carbodiimide intermediate to activate the surface functional groups prior to the coupling reaction,4,5,47,55,63 has the added advantage. 3.5. Stability of the Immobilized GOD. Storage stability is an important advantage of immobilized enzymes over the native (free) enzymes, because native enzymes can lose their activities fairly quickly.68 The storage stability of the immobilized GOD on the Si-g-P(GMA) surfaces, obtained from an immobilization time of 5 h, was examined after the Si-g-P(GMA)-GOD substrates were stored, respectively, in air and in PBS solution at 4 °C for 14 days. According to the XPS analysis results (Figure 7a-d and Table 1), the relative intensities of the nitrogen signal ([N]/[C] ratios) on the two aged surfaces remain almost unchanged compared to that of the freshly prepared Si-g-P(GMA)-GOD surface in Figure 3f. Under both storage conditions, the immobilized GOD molecules still retained more than 90% of their original EA. Free GOD molecules from the commercial preparation, on the other hand, retained only about 80% of their original activity over the same period of time in air at 4 °C and in PBS solution at 4 °C. The result readily indicates that the immobilized GOD exhibits an improved stability over that of the free enzyme. Of the immobilization methods, covalent fixation of enzyme molecules on a surface often gives rise to the highest stabilization effect on enzyme activities because the active conformation of the immobilized enzyme is stabilized by multipoint bond formation between the substrate and the enzyme molecules.65,69 The stability of immobilized GOD during storage also depends strongly on the coupling method. For example, coupling via a 1,3-

Covalent Immobilization of GOD on P(GMA)-Si(111)

Biomacromolecules, Vol. 6, No. 2, 2005 1019

Figure 7. C(1s) and N(1s) core-level spectra of the Si-g-P(GMA)-GOD surface after storage (a, b) in air at 4 °C for 14 days and (c, d) in PBS solution at 4 °C for 14 days. The C(1s) and N(1s) spectra of the original Si-g-P(GMA)-GOD surface correspond to those shown in Figure 3 (e, f).

bifunctional aromatic reagent can result in deformation of the active enzyme conformation and a loss in the multipoint binding ability of the enzyme. As a consequence, the storage stability of the immobilized enzyme is reduced.9,65 Such a problem can be avoided to a large extent through the direct coupling of GOD with the epoxide functional groups of P(GMA), as in the case of the present work. 4. Conclusions A Si-g-P(GMA) hybrid with well-defined and covalently tethered P(GMA) brushes was prepared via surface-initiated ATRP of GMA on the hydrogen-terminated Si(111) surface, pre-immobilized with an ATRP initiator monolayer via a simple one-step hydrosilylation process. Kinetics study revealed an approximately linear increase in thickness and DP of the P(GMA) brushes with the polymerization time. The epoxide groups of the grafted P(GMA) chains can be used for the direct covalent immobilization of GOD through the ring-opening reaction with the amine groups of GOD. The P(GMA) chain with the hydroxyl groups from the ringopening coupling reaction with GOD serves as an effective spacer to provide the immobilized GOD with a higher degree of conformational freedom and a more hydrophilic environment. An equivalent EA above 1.6 units/cm2 and a RA of about 55-65% can be readily achieved for the immobilized GOD. The covalent immobilization process also helps to improve the stability of GOD. With the inherent advantage

of the electronic properties of silicon substrates and surfaces, as well as the retention of a high level of the EA, the Si-gP(GMA)-GOD hybrid is potentially useful for the fabrication of silicon-based glucose biosensors. References and Notes (1) Chinnayelka, S.; McShane, M. J. Biomacromolecules 2004, 5, 1657. (2) Malhotra, B. D.; Chaubey, A. Sens. Actuators, B 2003, 91, 117. (3) Suzuki, H.; Kumagai, A.; Ogawa, K.; Kokufuta, E. Biomacromolecules 2004, 5, 486. (4) Liu, X.; Neoh, K. G.; Cen, L.; Kang, E. T. Biosens. Bioelectron. 2004, 19, 823. (5) Cen, L.; Neoh, K. G.; Kang, E. T. Biosens. Bioelectron. 2003, 18, 363. (6) Deshpande, M. V.; Amalnerkar, D. P. Prog. Polym. Sci. 1993, 18, 623. (7) Ogawa, K.; Kambe, T. N.; Nakahara, T.; Kokufuta, E. Biomacromolecules 2002, 3, 625. (8) Shi, L.; Lu, Y.; Sun, J.; Zhang, J.; Sun, C.; Liu, J.; Shen, J. Biomacromolecules 2003, 4, 1161. (9) Tiller, J. C.; Rieseler, R.; Berlin, P.; Klemm, D. Biomacromolecules 2002, 3, 1021. (10) Gulla, K. C.; Gouda, M. D.; Thakur, M. S.; Karanth, N. G. Biosens. Bioelectron. 2004, 19, 621. (11) Malitesta, C.; Palmisano, F.; Torsi, L.; Zambonin, P. G. Anal. Chem. 1990, 62, 2735. (12) Shinohara, H.; Chiba, T.; Aizawa, M. Sens. Actuators, B 1988, 13, 79. (13) Szili, E.; Thissen, H.; Hayes, J. P.; Voelecker, N. Biosens. Bioelectron. 2004, 19, 1395. (14) Cai, W.; Peck, J. R.; Weide, D. W.; Hamers, R. J. Biosens. Bioelectron. 2004, 19, 1013. (15) Le´tant, S. E.; Hart, B. R.; Kane, S. R.; Hadi, M. Z.; Reynolds, J. G. AdV. Mater. 2004, 16, 689.

1020

Biomacromolecules, Vol. 6, No. 2, 2005

(16) Subramanian, A.; Kennel, S. J.; Oden, P. I.; Jacobson, K. B.; Woodward, J.; Doktycz, M. J. Enzyme Microb. Technol. 1999, 24, 26. (17) Popat, K. C.; Desai, T. A. Biosens. Bioelectron. 2004, 19, 1037. (18) Zhang, L.; Strother, T.; Cai, W.; Cao, X. P.; Simth, L. M.; Hamers, R. J. Langmuir 2002, 18, 788. (19) Royea, W. J.; Juang, A.; Lewis, N. S. Appl. Phys. Lett. 2000, 77, 1988. (20) Wro´blewski, W.; Malinowska, A.; Brzo´zka, E. Z. Talanta 2004, 63, 33 (21) Xu, F. J.; Xu, D.; Kang, E. T.; Neoh, K. G. J. Mater. Chem. 2004, 14, 2674. (22) Xu, F. J.; Zhong, S. P.; Yung, L. Y. L.; Kang, E. T.; Neoh, K. G. Biomacromolecules 2004, 5, 2392. (23) Kato, K.; Uchida, E.; Kang, E. T.; Uyama, Y.; Ikada, Y. Prog. Polym. Sci. 2003, 28, 209. (24) Coessens, V.; Pintauer, T.; Matayjaszewski, K. Prog. Polym. Sci. 2001, 26, 337. (25) Matayjaszewski, K.; Xia, J. H. Chem. ReV. 2001, 101, 2921. (26) Dong, C. M.; Sun, X. L.; Faucher, K. M.; Apkarian, R. P.; Chaikof, E. L. Biomacromolecules 2004, 5, 224. (27) Peeters, J.; Palmans, A. R. A.; Veld, M.; Scheijen, F.; Heise, A.; Meijer, E. W. Biomacromolecules 2004, 5, 1862. (28) Matayjaszewski, K.; Miller, P. J.; Shukla, N.; Immaraporn, B.; Gelman, A.; Luokala, B. B.; Siclovan, T. M.; Kickelbick, G.; Vallant, T.; Hoffmann, H.; Pakula, T. Macromolecules 1999, 32, 8716. (29) Pyun, T.; Kowalewski, T.; Matayjaszewski, K. Macromol. Rapid Commun. 2003, 24, 1043. (30) Joralemon, M. J.; Murthy, K. S.; Remsen, E. E.; Becker, M. L.; Wooley, K. L. Biomacromolecules 2004, 5, 903. (31) Iwata, R.; Suk-In, P.; Hoven, V. P.; Takahara, A.; Akiyoshi, K.; Iwasaki, Y. Biomacromolecules 2004, 5, 2308. (32) Husseman, M.; Malmstro¨m, E. E.; McNamara, M.; Mate, M.; Mecerreyes, D.; Benoit, D. G.; Hedrick, J. L.; Mansky, P.; Huang, E.; Russell, T. P.; Hawker, C. J. Macromolecules 1999, 32, 1424. (33) Ejaz, M.; Yamamoto, S.; Tsujii, Y.; Fukuda, T. Macromolecules 2002, 35, 1412. (34) Edmondson, S.; Huck, W. T. S. J. Mater. Chem. 2004, 14, 730. (35) Mori, H.; Boker, A.; Krausch, G.; Muller, A. H. E. Macromolecules 2001, 34, 6871. (36) Lim, J. E.; Shim, C. B.; Kim, J. M.; Lee, B. Y.; Yue, J. E. Angew. Chem., Int. Ed. 2004, 43, 3839. (37) Sieval, A. B.; Linke, R.; Heij, G.; Meijer, G.; Zuilhof, H.; Sudholter, E. J. R. Langmuir 2001, 17, 7554. (38) Sieval, A. B.; Linke, R.; Zuilhof, H.; Sudholter, E. J. R. AdV. Mater. 2000, 12, 1457. (39) Stears, R. L.; Martinsky, T.; Schena, M. Nat. Med. (N. Y.) 2003, 9, 140. (40) Nishiyama, S.; Goto, A.; Saito, K.; Sugita, K.; Tanmada, M.; Sugo, T.; Funami, T.; Goda, Y.; Fujimoto, S. Anal. Chem. 2002, 74, 4933.

Xu et al. (41) Grano, V.; Diano, N.; Rossi, S.; Portaccio, M.; Attanasio, A.; Cermola, M.; Spiezie, R.; Citton, C.; Mita, D. G. Biotechnol. Prog. 2004, 20, 1393. (42) Arica, M. Y.; Bayramoglu, G.; Bicak, N. Process Biochem. 2004, 39, 2007. (43) Danisman, T.; Tan, S.; Kacar, Y.; Ergene, A. Food Chem. 2004, 85, 461. (44) Carbjal, M. L.; Samolko, E. E.; Grasselli, M. Nucl. Instr. Methods Phys. Res., Sect. B 2003, 208, 416. (45) Eckert, A. W.; Grobe, D.; Rothe, U. Biomaterials 2000, 21, 441. (46) Buriak, J. M. Chem. ReV. 2002, 102, 1272. (47) Kulik, E. A.; Kato, K.; Ivanchenko, M. I.; Ikada, Y. Biomaterials 1993, 14, 63. (48) Kang, I. K.; Kwon, B. K.; Lee, J. H.; Lee, H. B. Biomaterials 1993, 14, 787. (49) Bonde, M.; Pontoppidan, H.; Pepper, D. S. Anal. Biochem. 1993, 200, 195. (50) Linford, M. R.; Fenter, P.; Eisenberger, P. M.; Chidsey, C. E. D. J. Am. Chem. Soc. 1999, 121, 11513. (51) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. In X-ray Photoelectron Spectroscopy; Chastain, J., Ed.; Perkin-Elmer: Eden Prairie, MN, 1992; pp 40, 47, 62, 92. (52) Yu, W. H.; Kang, E. T.; Neoh, K. G. Langmuir 2004, 20, 8294. (53) Wolowacz, S. E.; Hin, B. F. Y. Y.; Lowe, C. R. Anal. Chem. 1992, 64, 1541. (54) Degani, Y.; Heller, A. J. Phys. Chem. 1987, 91, 1285. (55) Ikada, Y. Biomaterials 1994, 15, 725. (56) Vasileva, N.; Godjevargova, T. Mater. Sci. Eng., C 2005, 25, 17. (57) Vasileva, N.; Godjevargova, T. J. Membr. Sci. 2004, 239, 157. (58) Arica, M. Y.; Denizli, A.; Baran, T.; Hasirci, V. Polym. Int. 1998, 46, 345. (59) Arica, M. Y.; Bayramoglu, G. Biochem. Eng. J. 2004, 20, 73. (60) Caliceti, P.; Schiavon, O.; Sartore, L.; Monfardini, C.; Veronese, F. M. J. Bioact. Compat. Polym. 1993, 8, 41. (61) Godjevargova, T.; Konsulov, V.; Dimov, A.; Vasileva, N. J. Membr. Sci. 2000, 172, 279. (62) Yabuki, S.; Shinohara, H.; Aizawa, M. J. Chem. Soc., Chem. Commun. 1989, 945. (63) Godjevargova, T.; Dimov, A.; Ivavova, D. J. Appl. Polym. Sci. 1998, 68, 323. (64) Wang, S. Q.; Yoshimoto, M.; Fukunaga, K.; Nakao, K. Biotechnol. Bioeng. 2003, 83, 444. (65) Wang, C. C.; Hsiue, G. H. J. Appl. Polym. Sci. 1993, 50, 1141. (66) Dumont, J.; Fortier, G. Biotechnol. Bioeng. 1996, 49, 544. (67) Subramanian, A.; Kennel, S. J.; Oden, P. I.; Jacobson, K. B.; Woodward, J. D.; Mitchel, J. Enzyme Microb. Technol. 1999, 24, 26. (68) Buchholz, K.; Klein, J. In Methods Enzymology; Mosbach, K. V., Ed.; Academic Press: New York, 1987; Vol. 135, p 3. (69) Arica, M.; Yakup, H. J. Chem. Technol. Biotechnol. 1993, 58, 287.

BM0493178