Enhanced Activity of Enzymes Immobilized in Thermoresponsive Core

Nov 12, 2009 - core onto which a poly(N-isopropylacrylamide) (PNiPA) network is ... immobilized into the PNiPA network without loss of colloidal stabi...
0 downloads 0 Views 2MB Size
Download PDF
J. Phys. Chem. B 2009, 113, 16039–16045

16039

Enhanced Activity of Enzymes Immobilized in Thermoresponsive Core-Shell Microgels Nicole Welsch and Alexander Wittemann Physikalische Chemie I, UniVersity of Bayreuth, 95440, Bayreuth, Germany

Matthias Ballauff* Soft Matter and Functional Materials, Helmholtz-Zentrum Berlin fu¨r Materialien und Energie GmbH, Glienicker Straβe 100, 14109 Berlin and Department of Physics, Humboldt UniVersity Berlin, Newtonstr. 15, 12489 Berlin, Germany ReceiVed: August 4, 2009; ReVised Manuscript ReceiVed: October 15, 2009

We present a quantitative study of the catalytic activity of β-D-glucosidase from almonds adsorbed on thermosensitive microgels. The core-shell particles used as a carrier system consist of a solid polystyrene core onto which a poly(N-isopropylacrylamide) (PNiPA) network is grafted. In the swollen state of this microgel, i.e., below the critical solution temperature (LCST) of PNiPA, high amounts of enzyme can be immobilized into the PNiPA network without loss of colloidal stability. The enzymatic activity of β-D-glucosidase in its native form and in the adsorbed state was analyzed in terms of Michaelis-Menten kinetics. Moreover, the dependence of the enzymatic activity on temperature was investigated. We demonstrate that the enzymatic activity of β-D-glucosidase adsorbed on such a core-shell microgel is increased by a factor of more than three compared to its activity in solution. This is in marked contrast to other carrier systems that usually lead to a strong decrease of enzymatic activity. Both the high loading capacity of the carrier observed and the increase of the catalytic activity of immobilized β-D-glucosidase are traced back to the formation of strong interactions between the enzyme and microgel. Studies by Fourier-transform infrared (FT-IR) spectroscopy identify the formation of hydrogen bonds as driving forces for the adsorption. Hydrogen bonding may also be the reason for the enhanced activity. Introduction Nanoparticles such as latex particles, silica-based systems, and hydrogels have generated much research interest in biomedical domains as these systems might be well-suited as enzyme supports,1-6 drug-delivery systems,7-9 and biosensors.10 Due to their versatile properties, temperature-responsive polymers have been intensively explored as supports.11-18 As microgels based on poly(N-isopropylacrylamide) (PNiPA) thermally respond close to the body temperature, PNiPA is the most investigated thermosensitive polymer.2 At room temperature, PNiPA is hydrophilic and swollen in water. Raising the temperature above 32 °C, the lower critical solution temperature (LCST) of this polymer, a volume transition takes place, and most of the water is expelled from the now hydrophobic network. In general, nanosized hydrogels are characterized by a very short response time to external triggers when compared to macrogels.19 Moreover, functionalized temperature-sensitive microgels can be used as drug-delivery systems for a targeted transport and for drug release after endocytosis in the targeted cells.7 However, immobilization of enzymes on hydrogels often leads to a pronounced loss of catalytic activity.11,18 This decrease of the activity is attributed to a partial unfolding of the protein resulting from the adsorption of proteins on solid surfaces and on colloidal particles.20 Denaturation of proteins on PNiPAbased microgels may also be relevant for the cytotoxicity of these particles.7,9,21 Strong interactions between nanoparticles, proteins, and peptides seem also to promote protein fibrillation. * Corresponding author. E-mail address: [email protected].

Thus, Linse et al.22 investigated the fibril formation of the human protein β2-microglobulin in the presence of copolymer particles consisting of N-isopropylacrylamide and N-tert-butylacrylamide. They found that the adsorption of β2-microglobulin on the nanoparticles enhances the probability of nucleation. The opposite effect, however, is observed for the fibrillation of amyloid β protein.23 Hence, there is a subtle interaction between proteins and the PNiPA network that needs to be explored in detail. Evidently, the catalytic activity of an adsorbed enzyme will give direct information about the interaction of the protein with the network and possible conformational changes thus induced. In particular, the dependence of the enzymatic activity on temperature is expected to elucidate the role of hydrophobic interactions above the volume transition. Despite the enormous number of studies employing microgels, no quantitative investigation of the enzyme activity in terms of the Michaelis-Menten kinetics is available so far. Microgels have been recently introduced as “smart” carriers for metallic nanoparticles, the catalytic activity of which can be switched by the volume transition in these systems.24 However, a similar study on the dependence of the enzymatic activity on temperature is still lacking. Here, we present a quantitative study of the activity of β-Dglucosidase from almonds adsorbed on core-shell microgels as a function of temperature. As depicted in Figure 1a, the core of this polymer consists of polystyrene (PS) with colloidal dimensions (diameter ∼85 nm), and the shell consists of a PNiPA network cross-linked by N,N′-methylenebisacrylamide (BIS). To obtain information about the interactions between

10.1021/jp907508w CCC: $40.75  2009 American Chemical Society Published on Web 11/12/2009

16040

J. Phys. Chem. B, Vol. 113, No. 49, 2009

Welsch et al.

Figure 1. (a) Schematic representation of the adsorption of β-D-glucosidase on PS-PNiPA microgel particles consisting of a PS core onto which a cross-linked PNiPA network is attached. The adsorption was conducted at 4 °C in 10 mM MOPS buffer solution (pH ) 7.2) and with a microgel concentration of 1 wt %. The immobilized enzymes do not impede the volume transition of the microgel above the LCST. (b) The adsorbed amount of β-D-glucosidase per gram of microgel τads is plotted versus the concentration of free enzyme csol in solution. The dashed line represents the fit of the experimental data by the Langmuir isotherm according to eq 2. The arrow marks the amount of entrapped enzyme used for kinetic investigation (620 mg of β-D-glucosidase per gram of microgel). The inset displays the data as a linearized Langmuir plot. (c) Model reaction for the enzymatic hydrolysis of oNPG into D-glucose and o-nitrophenol.

adsorbed enzymes and support, Fourier transform infrared (FTIR) spectroscopy was used. Experimental Methods Materials. All chemicals were purchased from Sigma-Aldrich unless otherwise noted. β-D-Glucosidase (from almonds, Product No. G0395, LOT No. 058K4066) was used without further purification and was assigned to an activity of 5.2 units/mg of solid by the supplier. Styrene was destabilized by washing with aqueous NaOH and subsequently with deionized water. After drying with CaCl2, the monomer was distilled under reduced pressure. Acrylic acid was distilled under reduced pressure prior to use. All other chemicals used in this study were of analytical grade. Water for all purposes was purified using reverse osmosis and was deionized preliminarily (Milli-Q). Electrophoresis of β-D-Glucosidase. Discontinuous sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS/PAGE) was performed in a vertical setup supplied from Biorad (Mini Protean 3 Cell) and Serva (Blue Vertical 101). The 0.75 mm thick slab gels were polymerized just before electrophoresis with an acryl amide concentration of 5% and a pH value of 6.8 for the stacking gel and 12% and a pH value of 8.8 for the resolving gel. Electrophoresis was carried out at a constant current of 25 mA. Isoelectric focusing was performed in a flatbed system for horizontal electrophoresis with a cooling plate to provide homogeneous thermostatting (ServaBlue Horizon). The gels used were purchased as prefabricated ampholyte/polyacrylamide gels with dimensions of 12.5 × 12.5 cm and a thickness of 0.15 mm (Serva Servalyt Precotes Wide Range). These gels are used for the pH range 3-10. The gels were focused at 10 mA, and the electrophoresis was carried out at a constant current of 3.0 mA, whereas the voltage was increased in seven steps from 150 to 2500 V. Synthesis of PS-PNiPA Core-Shell Microgels and Characterization. The synthesis of the PS-core particles and the polymerization of the cross-linked PNiPA shell were performed

by emulsion and seed polymerization, respectively, along lines given previously.25 In this study the content of the cross-linker N,N′-methylenebisacrylamide (BIS) is 2.5 mol % with respect to NiPA. DLS measurements were performed using a Peters-ALV goniometer setup. The samples were diluted to a concentration of 10-2 g/L and filtered through a syringe filter (Supor membrane: 1.2 µm pore width, PALL Acrodisc). Measurements were carried out at different temperatures in a range of 10-60 °C controlled by a thermostatted toluene bath. All experiments were done at a scattering angle of 90°. The fluctuations of the scattered light were analyzed with an ALV-5000 correlator (Peters). The measured intensity time correlation functions were analyzed using the methods of cumulants. Adsorption Experiments. The immobilization of β-Dglucosidase was performed in the same manner as described elsewhere.26 The pH value of the polymer solution was adjusted to 7.2 by ultrafiltration of the aqueous solution against Dmorpholinopropane-sulfonic acid (MOPS) buffer. NaN3 (2 mM) was added to the buffer solution to avoid microbial growth. Given amounts of β-D-glucosidase from almonds were dissolved in 10 mM MOPS buffer to which a constant amount of microgel solution was added. The total volume was 10 mL, and the microgel concentration was accounted for 1 wt %. The samples were stirred for 24 h at 4 °C for equilibration. The surplus of nonadsorbed enzyme was removed by extensive ultrafiltration against MOPS buffer at room temperature. The amount of nonadsorbed enzyme was determined spectroscopically from the eluate at 278 nm (Agilent 8453 UV-visible, Agilent Technologies). The amount of immobilized β-D-glucosidase was calculated by subtracting the amount of native enzyme from the total amount of β-D-glucosidase. Enzyme Activity Assay. The activities of native and immobilized β-D-glucosidase were determined by following the hydrolysis of o-nitrophenyl-β-D-glucopyranoside (oNPG) at 405 nm with an Agilent 8453 spectrophotometer (Agilent 8453

Activity of Enzymes in Core-Shell Microgels

J. Phys. Chem. B, Vol. 113, No. 49, 2009 16041

UV-visible, Agilent Technologies). All kinetic studies were performed in MOPS buffer (pH 7.2) and with an optical path length of 1 cm. The kinetic parameters for β-D-glucosidase were determined at different temperatures (12-60 °C) using the concentration range of oNPG of 1.0-20.0 mM. The final concentration of native enzyme was adjusted to 0.01 g/L for experiments conducted at 12-30 °C and to 0.005 g/L for those performed at temperatures ranging from 35 up to 60 °C. In the case of entrapped enzyme, concentrations of 0.01 g/L were used at temperatures ranging from 14 to 20 °C. For reactions carried out at temperatures ranging from 25 to 34 °C and temperatures ranging from 36 to 55 °C, the concentration of β-D-glucosidase was adjusted to 0.005 and 0.0035 g/L, respectively. The substrate solutions were incubated for 15 min at the appropriate temperatures before the reaction was started by addition of 50 µL of ice-cooled enzyme solution. The absorbance was measured at 405 nm for 12 min with an interval of 10 s. The initial rate ν of enzymatic cleavage was determined from the increase of extinction with time. Therefore, experimental data points are modeled by the quadratic equation

Et ) Ei + kt + ct2

(1)

where Et is the background-corrected extinction at time t and Ei is the background-corrected extinction at the start of the calculation time range. The slope k leads directly to the initial rate of cleavage ν using the extinction coefficient of the product o-nitrophenol. Therefore, the molar spectral extinction coefficient was measured as a function of temperature. The extinction coefficient of o-nitrophenol decreases in a polynomial fashion when temperature is raised (data not shown). FT-IR Spectroscopy. The analysis of the secondary structure of native and immobilized β-D-glucosidase was performed using a FT-IR spectroscopy setup developed for protein analytics (Bruker Optik Confocheck) including a spectrometer (Tensor 27), a calcium fluoride flow-through liquid cell with 6.5 µm path length, and a highly sensitive photovoltaic MCT detector with liquid nitrogen cooling. Native β-D-glucosidase was dissolved in 10 mM MOPS buffer to a concentration of 15 mg/ L, whereas the enzyme-loaded microgels were adjusted with 10 mM MOPS buffer to 2 wt % microgel suspension. Measurements of pure microgels were accomplished at a concentration of 3 wt % latex spheres. Before the samples and reference solutions were injected into the cell, they had been filtrated through a 0.8 µm membrane syringe filter (Acrodisc, PALL) to remove dust particles and polymer agglomerates which could block the integrated filter frit and the cell. At least two spectra with 128 scans and 20 kHz were recorded in transmission mode ranging from 400 to 4000 cm-1 and averaged. The temperature of the measurement cell was maintained at 25 °C using a cryostat (Haake DC 30K20). For temperature-dependent measurements of the bare PSPNiPA microgel, a temperature ramp was chosen, at which the temperature was raised from 20 to 40 °C in steps of 5 °C. The secondary structure, i.e., R-helix and β-sheet content, was evaluated by using the Quant 2 analysis in the spectrometer software OPUS which is predicated on a protein database (Bruker OPTIK Bruker Protein Library). This databank comprises IR spectra of 30 reference proteins with elucidated X-ray structure. The absolute experimental error of the cross correlation of the reference data in the spectrophotometer is 4% for R-helix and 3% for β-sheet.

Results and Discussion Adsorption of β-D-Glucosidase from Almonds on PSPNiPA Microgels. We determined the adsorption capacity of PS-PNiPA microgel particles by investigating the efficiency of immobilization at different enzyme concentrations at 4 °C (Figure 1a and b). Hence, β-D-glucosidase is adsorbed onto this microgel below the LCST. At the chosen conditions, up to 620 mg of β-D-glucosidase per gram of microgel could be adsorbed. This indicates that strong interactions between both species exist in the swollen state of PNiPA. The adsorption behavior of β-D-glucosidase can be described by a Langmuir-type model

τads K · csol ) τads,M 1 + K · csol

(2)

where (τads)/(τads,M) denotes the fraction of adsorption sites occupied as the function of the enzyme concentrations in solution csol and τads,M is the maximum binding capacity. The dashed line in Figure 1b displays the fit of the experimental data by eq 2. K was determined to (0.08 ( 0.03) mL/mg and τads,M to (1400 ( 300) mg/g of microgel. This high capacity demonstrates that enzyme molecules penetrate deeply into the network. Additional experiments confirmed that at given adsorption conditions no leakage or desorption of the β-Dglucosidase occurs. The excellent fit of the data by the Langmuir model indicates that only one type of binding site for the enzyme is present in the microgel. Similar findings have been reported recently for the adsorption of β-D-glucosidase from almonds on spherical polyelectrolyte brushes (SPB)27,28 and other proteins and enzymes on microgels based on PNiPA.13,17 Thus, the present microgel presents a model system providing a chemically welldefined surrounding for the embedded protein. Since the isoelectric point of the isozyme of β-D-glucosidase from almonds used is 4.4, the enzyme has an overall negative charge at the conditions chosen for adsorption. Due to the sulfate groups which were introduced into the polymer through KPS as initiator for the synthesis of the particles, the PS-PNiPA core-shell system is weakly negatively charged. This results in an overall repulsion between particles and enzyme. The main driving forces for immobilization therefore must be sought in the formation of hydrogen bonds between the backbone of the enzyme and the amide side chains of the microgel as well as in hydrophobic bonding.29 Desorption of immobilized enzyme upon temperature increase can be excluded since a higher adsorption capacity above the LCST is observed in general.30 Temperature-Dependent Michaelis-Menten Kinetics. In the following, the hydrolytic activity of adsorbed and native β-D-glucosidase is presented as a function of temperature. As depicted in Figure 1c, o-nitrophenyl-β-D-glucopyranoside (oNPG) was chosen as a substrate. The rate V of enzymatic activity was evaluated in terms of Michaelis-Menten kinetics31

ν)

νmax · [S] kcat · [E]tot · [S] ) Km + [S] Km + [S]

(3)

where [S] is the substrate concentration; νmax is the maximum rate attained at infinite concentration of substrate; Km is the Michaelis-Menten constant; [E]tot is the total enzyme concentration; and kcat is the turnover number. All kinetic experiments related to immobilized β-D-glucosidase were done using solu-

16042

J. Phys. Chem. B, Vol. 113, No. 49, 2009

Welsch et al.

Figure 2. (a) Lineweaver-Burk plots for the hydrolysis of oNPG catalyzed by immobilized (filled circle, 620 mg of enzyme per gram of microgel) and native β-D-glucosidase (filled square) at 20 °C (left) and 40 °C (right) in MOPS buffer (pH ) 7.2). (b) Arrhenius plots of native (filled square) and immobilized (filled circle) β-D-glucosidase (620 mg of enzyme per gram of microgel) in MOPS buffer solution (pH ) 7.2). The turnover number kcat for each temperature was determined from the kinetic analysis according to Figure 2a. The enzyme concentration was located between 0.005 and 0.01 g/L (native enzyme) and 0.0035 to 0.01 g/L (immobilized enzyme), respectively, and the substrate concentration varied between 1.0 and 20.0 mM. The dashed lines are linear fits of the experimental data according to eq 4 to determine the activation energies of the rate-limiting steps of the reaction. In addition, the hydrodynamic radius Rh of the carrier is plotted as a function of temperature T (open tilted square). The inflection point of this curve is calculated to 304.7 K.

TABLE 1: Kinetic Parameters of Free and Immobilized β-D-Glucosidase at 20 and 40 °C enzyme

temperature [°C]

Km [mM]

kcat [s-1]a

free free immobilized immobilized

20 40 20 40

6.8 ( 1 6.8 ( 1 9.4 ( 1 11.3 ( 1

20.4 57.8 65.5 204.2

a Related to β-D-glucosidase in its dimeric form with a molecular weight of 135 kD.

tions with particles containing 620 mg of enzyme per gram of core-shell microgel. Figure 2a displays the resulting rates V and the kinetic constant kcat for native and adsorbed β-Dglucosidase at 20 and 40 °C. The evaluation of the data was done by a nonlinear fit according to eq 3. Table 1 summarizes the kinetic parameters which were obtained for activity measurements of native and adsorbed β-D-glucosidase at both temperatures. The kinetic data for the free enzyme at 20 °C are comparable to data obtained in preceding studies.28 However, as the central result of this study we find that the immobilization of β-D-glucosidase onto PS-PNiPA microgels results in a remarkable enhancement of the hydrolytic activity by a factor of 3.2-3.5. A secondary effect accompanying the adsorption process is the increase of Km, which signals a reduced binding

affinity. However, this modulation is much less pronounced than the impact on kcat. To elucidate this effect further, we investigated the catalytic rate for native and adsorbed β-D-glucosidase between 12 and 60 °C. The values of kcat which were evaluated in terms of Michaelis-Menten kinetics were plotted according to the Arrhenius equation

(

kcat ) A · exp -

Ea R·T

)

(4)

where A is the pre-exponential factor; R is the gas constant; T is the absolute temperature; and Ea is the activation energy. Figure 2b displays the Arrhenius plots for native and entrapped β-D-glucosidase. These experiments unambiguously show that kcat of the adsorbed β-D-glucosidase exceeds the one of the free enzyme for the whole temperature range. In the case of the free enzyme, the Arrhenius plot is characterized by a distinct inflection point at 20 °C, with activation energies of 42.9 kJ/ mol above and 104.8 kJ/mol below the breakpoint. Weber et al. reported that this inflection point is an isokinetic temperature at which the rate-limiting step in hydrolysis catalyzed by β-Dglucosidase from almonds is changed.32 In the case of adsorbed

Activity of Enzymes in Core-Shell Microgels

J. Phys. Chem. B, Vol. 113, No. 49, 2009 16043

Figure 3. (a) FT-IR spectra of pure PS-PNIPA microgel (black line) and PS-PNiPA microgel with immobilized β-D-glucosidase (red line, 620 mg of enzyme per gram of microgel) in MOPS buffer solution (pH ) 7.2) at 25 °C. Both spectra are normalized to a particle concentration of 1 wt %. (b) FT-IR spectra of native β-D-glucosidase (black line) and immobilized β-D-glucosidase (red line, 618 mg of enzyme per gram of microgel). The spectrum of immobilized β-D-glucosidase was obtained from the spectrum of enzyme-loaded PS-PNiPA by subtraction of the pure PS-PNiPA spectrum. The FT-IR spectra of both free and immobilized enzyme are shown after normalization to the same enzyme concentration. The difference spectrum (blue line) was obtained by subtraction of the spectrum of the free enzyme from that of immobilized enzyme. (c) FT-IR spectra of bare microgel at different temperatures: (teal line) 20 °C, (blue line) 30 °C, (pink line) 35 °C, and (black line) 40 °C. The change of the hydrogen binding pattern upon heating above the LCST of the polymer has a great impact on the shape and intensities of the amide I and amide II band: the rise in temperature leads to an increased absorption at 1657 cm-1 and a reduced absorption at 1626 cm-1. The higher intensity at 1657 cm-1 is attributable to the formation of more intramolecular hydrogen bonds, whereas the drop of intensity at 1626 cm-1 is caused by the breakage of hydrogen bonding with water. To compare these temperature-induced changes with the changes of the spectrum of the microgel after adsorption of enzyme, the spectrum of PS-PNiPA microgel loaded with β-D-glucosidase (green dashed-dotted line, 620 mg per gram of microgel) is also depicted.

β-D-glucosidase, a change in the slope is also observed. However, between 20 and 34 °C the Arrhenius plot deviates from linearity. Within this region, kcat was shifted to lower values, and the Km value slightly increased at 32 °C (see below). The activation energies were calculated from the slopes in the linear regions and were found to be 115.4 kJ/mol below 20 °C and 28.9 kJ/mol from 34 to 50 °C. Above the LCST, Ea for the reaction catalyzed by the immobilized β-D-glucosidase is decreased about 30% compared to Ea which was calculated for free β-D-glucosidase. As is obvious from Figure 2b, the adsorption of the enzyme leads to a slowing down of the catalytic rate at temperatures where the volume transition of PNiPA occurs. Moreover, above the LCST a smaller activation energy for the bound enzyme is observed when compared to the free enzyme. Both phenomena can be attributed to the volume phase transition of the PS-PNiPA polymer at 32 °C, where the hydrodynamic radius Rh of the carrier shrinks about ca. 40 nm. Therefore, the reduction of the pore sizes of the PNiPA network upon heating must lead to a

slowing of the diffusion of oNPG. This increase of the diffusional barrier is followed by a slight decrease of kcat. At still higher temperatures, the density within the network is constant, and the linear relation between ln kcat and T-1 is recovered again. The lower Ea above the LCST indicates that diffusional limitations affect the catalytic rate in the shrunken state of the microgel. Similar catalytic results were obtained for metal nanoparticles embedded within microgel particles.24 The increased hydrophobicity of the polymer network at rising temperature leads to the amplification of hydrophobic interactions and to more hydrogen bridges between the gel and the enzyme. These additional interactions changing the microenvironment of entrapped β-D-glucosidase must be the reason for the enhanced catalytic activity. This result is supported by the investigation of the immobilization of enzymes which differ in their sensitivity to hydrophobic environments.33 FT-IR Spectroscopy. To explore the interactions of β-Dglucosidase with the network, we used Fourier transform IR spectroscopy (FT-IR) in transmission mode. FT-IR experiments

16044

J. Phys. Chem. B, Vol. 113, No. 49, 2009

of proteins adsorbed to SPB and of bare PNiPA microgels performed by us34 and by Keerl et al.35 corroborate the use of FT-IR spectroscopy as the appropriate method for this purpose. Due to the presence of amide groups in the PNiPA network, the measured spectrum of immobilized β-D-glucosidase must be adjusted for the absorption of the carrier. Therefore, the spectra of the loaded and unloaded carrier particles were measured at 25 °C and normalized to a microgel concentration of 1 wt % (Figure 3a). After subtraction of the spectra of the bare microgel, the spectra of adsorbed β-D-glucosidase could be obtained in good accuracy. Figure 3b shows this spectrum in comparison to the enzyme before the adsorption studies. The secondary structure information of the native enzyme is derived from the amide I and amide II band of the IR spectrum using a PLS algorithm as described by Dousseau and Pe´zolet.36 In the native state, the R-helix and β-sheet content was determined to (13 ( 4) % and (3 ( 3) %, respectively, which shows that β-D-glucosidase mainly consists of unstructured regions.31 Obviously, the immobilization of β-D-glucosidase induced both a shift of the amide I and amide II band (Figure 3a) and a change in the shape of these peaks (Figure 3b and 3c). Unfolded or aggregated proteins strongly absorb between 1610 and 1628 cm-1 of the amide I band.37 As no increase in this range can be observed, an unfolding of adsorbed β-D-glucosidase can definitely be excluded. This is confirmed by the high catalytic activity after immobilization. The modulations of the FT-IR spectrum of adsorbed hydrolase are comparable to temperature-induced shifts in the spectrum of bare PNiPA microgels. This suggests that analogous interactions are responsible for the changed absorption (Figure 3c): Heating the polymer above the LCST leads to the emergence of a shoulder at 1657 cm-1 at the expense of the intensity at 1626 cm-1. In the case of the amide II band, the phase transition causes a slight blue shift. Similar observations were made previously.35 The volume phase transition induces the breakage of intermolecular hydrogen bonds between water molecules and the polymer network and the formation of intramolecular hydrogen bonds between the CdO and N-H groups of the polymeric chains. As shown in Figure 3c, the absorption spectrum of the microgel with adsorbed enzyme at 25 °C shows some similarities to the spectrum of the bare microgel in the shrunken state. This indicates that upon adsorption hydrogen bonds between enzyme and microgel are formed and hydrogen bridges to water are broken. The differences between the spectra of free and immobilized β-D-glucosidase which occur between 1615 and 1660 cm-1 can not be ascribed to the change of the secondary structure because the reduction of the intensity at lower wavenumbers and the increase at higher wavenumbers conflict with a change of the content of β-sheet or R-helix.37 Consequently, the changes in the absorption of adsorbed β-Dglucosidase are attributed to the rearrangement of intermolecular hydrogen bridges between enzyme and carrier rather than to an altered secondary structure of β-D-glucosidase. The high adsorption capacity of the core-shell microgels used for this study unambiguously shows that strong attractive interactions toward β-D-glucosidase are present. FT-IR experiments indicate that the formation of hydrogen bonds plays an important role in the adsorption process. Moreover, the catalytic constants of the hydrolysis of o-NPG are modulated: The increase of Km is occasionally observed for reactions catalyzed by immobilized enzymes.18 While Km is influenced by the adsorption of β-D-glucosidase only slightly, a dramatic effect is observed in the case of kcat. This pronounced enhancement

Welsch et al. of kcat demonstrates that β-D-glucosidase assumes an even higher enzymatic activity than in the unbound state. The enhanced activity may be explained as follows: Hydrogenbonding, especially interactions at the sugar hydroxyl position 2, are significant for the binding of the substrate and the stabilization of the transition states. The catalytic glutamate residues inside the active side are exactly 5.5 Å apart, allowing the substrate and a water molecule to bind between them.38 Hence, small changes in the hydrogen binding pattern next to the active site involve large changes in the electronic dynamics of the active site and thus in the hydrolytic catalysis. FT-IR experiments give clear evidence for a rearrangement of hydrogen bonds upon adsorption and substantiate this argumentation. Moreover, the highly flexible enzyme may change its tertiary and quaternary structure in response to the changed environment. Consequently, a shift of the equilibrium between the monomeric and dimeric form of β-D-glucosidase39 affects the activity dramatically. Conclusion In conclusion, we could demonstrate that PS-PNiPA microgels serve as superior carriers for the adsorption of β-Dglucosidase: The adsorption of β-D-glucosidase from almonds induces changes within the catalytic center leading to a more than 3-fold increase of the catalytic activity. In addition, the catalytic properties of immobilized enzymes can be manipulated by the temperature-dependent swelling behavior of the microgel. Thus, improved carrier particles for enzymes can be obtained by judicious tailoring of the interaction between protein and the particle. Acknowledgment. Financial support by the Deutsche Forschungsgemeinsschaft, Schwerpunkt “Hydrogele”, and by the SFB 481, Bayreuth, is gratefully acknowledged. References and Notes (1) Luckarift, H. R.; Spain, J. C.; Naik, R. R.; Stone, M. O. Nat. Biotechnol. 2006, 22, 211–213. (2) Nayak, S.; Lyon, L. A. Angew. Chem., Int. Ed. 2005, 44, 7686– 7708. (3) Nagel, B.; Warsinke, A.; Katterle, M. Langmuir 2001, 17, 4704–4707. (4) Park, T. G.; Hoffman, A. S. J. Biomed. Mater. Res. 1990, 24, 21–38. (5) Shiroya, T.; Yasui, M.; Fujimoto, K.; Kawaguchi, H. Colloids Surf. B. 1995, 4, 275–285. (6) Ogawa, K.; Wang, B.; Kokufuta, E. Langmuir 2001, 17, 4704–4707. (7) Das, M.; Mardyani, S.; Chan, W. C. W.; Kumacheva, E. AdV. Mater. 2006, 18, 80–83. (8) Hsiue, G.-H.; Hsu, S.-H.; Yang, Ch.-Ch.; Lee, S.-H.; Yang, I. K. Biomaterials 2002, 23, 457–462. (9) Nayak, S.; Lee, H.; Chmielewski, J.; Lyon, L. A. J. Am. Chem. Soc. 2004, 126, 10258–10259. (10) Retama, J. R.; Lope´z, M. S.-P.; Pe´rez, J. P. H.; Cabanillas, G. F.; Lo´pez-Carbacos, E.; Lo´pez-Ruiz, B. Biosens. Bioelectron. 2005, 20, 2268– 2275. ¨ ktem, H. A.; O ¨ ktem, Z.; Tuncel, S. A. Polym. Int. (11) Arica, M. Y.; O 1999, 48, 879–884. (12) Duracher, D.; Elaissari, A.; Mallet, F.; Pichot, Ch. Langmuir 2000, 16, 9002–9008. (13) Grabstein, V.; Bianco-Peled, H. Biotechnol. Prog. 2003, 19, 17281733. (14) Huo, D.; Li, Y.; Qian, Q.; Kobayashi, T. Colloid Surf. B 2006, 50, 36–42. (15) Kawaguchi, H.; Fujimoto, K.; Mizuhara, Y. Colloid Polym. Sci. 1992, 270, 53–57. (16) Kawaguchi, H.; Kohki, K.; Takahashi, T.; Achiha, K.; Yasui, M.; Fujimoto, K. Macromol. Symp. 2000, 151, 591–598. (17) Taniguchi, T.; Duracher, D.; Delair, T.; Elaissari, A.; Pichot, Ch. Colloids Surf. B 2003, 29, 53–65. (18) Ortega, N.; Busto, M. D.; Perez-Mateos, M. Bioresour. Technol. 1998, 64, 105–111. (19) Tanaka, M.; Fillmore, D. J. J. Chem. Phys. 1979, 70, 1214–1218.

Activity of Enzymes in Core-Shell Microgels (20) Czeslik, C.; Winter, R. Phys. Chem. Chem. Phys. 2001, 3, 235– 239. (21) Vihola, H.; Laukkanen, A.; Valtola, L.; Tenhu, H.; Hirvonen, J. Biomaterials 2005, 26, 3055–3064. (22) Linse, S.; Cabaleiro-Lago, C.; Xue, W.-F.; Lynch, I.; Lindman, S.; Thulin, E.; Radford, S. E.; Dawson, K. A. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 8691–8696. (23) Cabaleiro-Lago, C.; Quinlan-Pluck, F.; Lynch, I.; Lindman, S.; Minogue, A. M.; Thulin, E.; Walsh, D. M.; Dawson, K. A.; Linse, S. J. Am. Chem. Soc. 2008, 130, 15437–15443. (24) Lu, Y.; Mei, Y.; Drechsler, M.; Ballauff, M. J. Phys. Chem. B. 2006, 110, 3930–9037. (25) Kim, J.-H.; Ballauff, M. Colloid Polym. Sci. 1999, 277, 1210–1214. (26) Wittemann, A.; Ballauff, M. Phys. Chem. Chem. Phys. 2006, 8, 5269–5275. (27) Haupt, B.; Neumann, Th.; Wittemann, A.; Ballauff, M. Biomacromolecules 2005, 6, 948–955. (28) Henzler, K.; Haupt, B.; Ballauff, M. Anal. Biochem. 2008, 378, 184–189.

J. Phys. Chem. B, Vol. 113, No. 49, 2009 16045 (29) Lindman, S.; Lynch, I.; Thulin, E.; Nilsson, H.; Dawson, K. A.; Linse, S. Nano Lett. 2007, 7, 914–920. (30) Shamim, N.; Liang, H.; Hidajat, K.; Uddin, M. S. J. Colloid Interface Sci. 2008, 320, 15–21. (31) Copeland, R. A. Enzymes - A practical introduction to structure, mechanism, and data analysis, 2nd ed.; Wiley-VCH: New York, 2000. (32) Weber, J. P.; Fink, A. L. J. Biol. Chem. 1980, 255, 9030–9032. (33) Valuev, L. I.; Zefirova, O. N.; Obydennova, I. V.; Plate, N. A. J. Bioact. Compat. Polym. 1994, 9, 55–65. (34) Wittemann, A.; Ballauff, M. Anal. Chem. 2004, 76, 2813–2819. (35) Keerl, M.; Smirnovas, V.; Winter, R.; Richtering, W. Angew. Chem., Int. Ed. 2008, 47, 338–341. (36) Dousseau, F.; Pe´zolet, M. Biochemistry 1990, 29, 8771–8779. (37) Jackson, M.; Mantsch, H. H. Crit. ReV. Biochem. Mol. Biol. 1995, 30, 95–120. (38) Zechel, D. L.; Withers, S. G. Acc. Chem. Res. 2000, 33, 11–18. (39) Grover, A. G.; Macmurchie, D. D.; Cushley, R. J. Biochim. Biophys. Acta 1977, 482, 98–108.

JP907508W