Enzymatic Catalysis at Nanoscale: Enzyme-Coated Nanoparticles as Colloidal Biocatalysts for Polymerization Reactions.

Enzyme-catalyzed controlled radical polymerization represents a powerful approach for the polymerization of a wide variety of water-soluble monomers. However, in such an enzyme-based polymerization system, the macromolecular catalyst (i.e., enzyme) has to be separated from the polymer product. Here, we present a compelling approach for the separation of the two macromolecular species, by taking the catalyst out of the molecular domain and locating it in the colloidal domain, ensuring quasi-homogeneous catalysis as well as easy separation of precious biocatalysts. We report on gold nanoparticles coated with horseradish peroxidase that can catalyze the polymerization of various monomers (e.g., N-isopropylacrylamide), yielding thermoresponsive polymers. Strikingly, these biocatalyst-coated nanoparticles can be recovered completely and reused in more than three independent polymerization cycles, without significant loss of their catalytic activity.

Introduction

Enzymes represent a family of nontoxic, environmentally friendly catalysts, which have proven to be a powerful tool in the polymerization of a wide variety of monomers and macromonomers.[1−3] Recent publications even demonstrate the use of peroxidases such as horseradish peroxidase (HRP), catalase from bovine liver, and laccase from trametes versicolor as biocatalysts in controlled radical polymerizations.[4−8] For the separation of the two macromolecular species, centrifugal filters with different molecular weight cutoffs can be employed, given that the molecular weight of the catalytic enzyme is different from that of the polymer product.[6,7]

One alternative and highly compelling approach for the separation of the two macromolecular species is to take one of the protagonists, preferentially the catalyst, out of the molecular domain and locate it in the colloidal domain, ensuring quasi-homogeneous catalysis as well as easy separation. This can be achieved by immobilizing the enzymes onto nanoparticles (NPs). In particular, metal and metal oxide NPs in the size range of 5–200 nm fulfill most of the desired requirements: (1) quasi-homogeneous catalysis as their dimension is on the nanoscale and their physical properties, such as diffusion constants, lie between molecular and colloidal domains,[9−12] thus ensuring higher reaction rates than heterogeneous catalysis; (2) high surface/volume ratio, ensuring relatively high catalyst concentration in the immobilized state;[13,14] and the last but not least (3) a simple and straightforward separation from the product by various techniques like centrifugation,[15−17] or in the case of magnetic NPs by magnetic separation.[18]

In general, proteins and enzymes can be immobilized onto solid supports using various techniques,[13,19,20] such as (a) “multipoint” covalent bonding via click chemistry[20−22] or coupling reactions, yielding amide[23−25] or imine bonds,[26−28] (b) cross-linking,[29] (c) affinity immobilization,[30,31] (d) entrapment,[32,33] and (e) adsorption.[15,16,34,35] Depending on the immobilization technique, the immobilization can enhance the properties of enzymes in terms of thermal stability, tolerance to extreme pH, and organic solvents.[36] Therefore, immobilized enzymes can appear to have higher activities than native enzymes under drastic conditions due to the enhanced stability.[34,36−42] On the other hand, the apparent activity of conventional immobilized enzymes can also be lower than that of their native counterparts, mainly because of the hindered substrate accessing or unfavorable conformational transition of the enzyme on the support.[13,43−46] Particularly in the case of colloidal supports, the catalytic performance of the immobilized enzymes also strongly depends on the colloidal stability of the colloidal support.[9,34] The catalytic performance of the enzymes immobilized onto NPs can decrease drastically, due to hindered accessibility of the substrates to their active sites,[9] when the NPs are not stable and aggregate or sediment in the reaction media. Hence, to create a functional enzymatically active colloidal system, a robust enzyme immobilization on the NPs as well as a colloidally stable system is required.

Recently, we have reported a quite simple approach for a highly robust immobilization of proteins and enzymes onto metal and metal oxide NPs, simply via physisorption in a ligand-exchange process.[15,16,34,35,47−52] Different enzymes, such as horseradish peroxidase, glucose oxidase, laccase, and catalase, were directly immobilized onto plasmonic gold NPs and superparamagnetic iron oxide (Fe3O4) NPs, yielding colloidally stable enzyme-coated NP systems. The enzyme coating on the NP surface is highly robust and enzymatically active. The colloidal stability and enzymatic performance (Vmax, Km) of the enzyme-coated NPs (Au@enzyme) are strongly dependent on the pH of the dispersion and the physicochemical properties of the respective enzyme. Hence, gold NPs are particularly suitable for the immobilization of proteins and enzymes because of their plasmonic properties, allowing fast detection of NP aggregation upon color change of dispersions, owing to the plasmonic coupling.[15,16,35,47,49,51]

The present work represents a proof-of-principle study for an enzymatic colloidal system, based on HRP-coated gold NPs (Figure ), being capable of catalyzing polymerization of different monomers, such as N-isopropylacrylamide (NIPAAM) and di(ethylene glycol)methyl ether methacrylate (DEGMA), yielding thermosensitive polymers (Figure ). Furthermore, we show that the Au@HRP NPs can be recovered almost completely after each polymerization reaction simply via centrifugation or via sedimentation in case of larger particles (100 nm Au@HRP NPs) and then be reused in two further (or even more) successful polymerization cycles. To the best of our knowledge, a highly stable enzymatic colloidal system, which is capable of multiple use and facile recovery, based on immobilized enzymes on metal NPs as a complete new and sustainable approach in the field of polymer synthesis and colloidal enzymatic catalysis, has not been reported so far.

Figure 1

Preparation of catalytically (re-)active HRP-coated gold nanoparticles via physisorption (ligand-exchange reaction). (A) Schematic illustration of coating citrate-stabilized gold nanoparticles with HRP (Au@HRP) and the following enzymatic reaction. The Au@HRP can be recovered multiple times. (B) UV–vis spectra of Au@citrate (black), Au@HRP (red), and Au@HRP recovered after enzymatic reaction (light blue). The inset displays the Au@HRP particle dispersion as prepared (red box) and recovered Au@HRP particle dispersion after an enzymatic reaction (light blue box). (C) UV–vis spectrum of oxidized 3,3′,5,5′-tetramethylbenzidine (TMB) with the characteristic adsorption peaks of the blue one-electron oxidation product TMB•+ at 370 and 652 nm.

Figure 2

Au@HRP as recyclable nanobiocatalyst for polymerization reaction. (A) Examplified multicycle atom transfer radical polymerization (ATRP) process of NIPAAM using HRP-coated gold NPs as nanobiocatalyst with 2-hydroxyethyl 2-bromoisobutyrate (HEBIB) as ATRP initiator. Because the gold NPs can be recovered via centrifugation, they can be reused multiple times. (B) Visible evidence of polymer formation over all three polymerization cycles. Representative photographs of poly(N-isopropylacrylamide) (PNIPAAM) (upper row) and poly(di(ethylene glycol)methyl ether methacrylate) (PDEGMA) (lower row) in aqueous solution after removing the Au@HRP via centrifugation. The temperatures of the solutions are displayed on the vials (left: below lower critical solution temperature (LCST) = room temperature; right: above LCST = 50 °C for PNIPAAM and 30 °C for PDEGMA). (C) UV–vis spectra of the original Au@HRP (pH 7.4) and the recovered gold NPs during the individual three polymerization cycles of NIPAAM. The inset shows photographs of the original Au@HRP (red box, pH 7.4) and the recovered Au@HRP (blue box, pH 7.4) after the third polymerization of NIPAAM.

Results and Discussion

Enzymatically (re-)active HRP-coated gold NPs (Au@HRP) of different sizes as reusable nanobiocatalysts were synthesized by coating the enzyme directly onto ∼15 and ∼100 nm citrate-stabilized gold NPs[34] (Figures and S1–S5 in the Supporting Information (SI)). Upon enzyme immobilization, the localized surface plasmon resonance (LSPR) of the 15 nm NP (Au15@HRP) dispersions red shifts from 519 to 523.5 nm (Figure B) and from 554 to 557.5 nm for the larger NP (Au100@HRP), owing to refractive index changes in the vicinity of the NP surface. The occurring redshift of 4.5 nm (3.5 nm for larger Au100@HRP NPs) and no significant change in color or LSPR band broadening indicate a successful enzyme coating without any particle aggregation during the entire coating process. In contrast to the systems reported in the literature,[45,53] the HRP-coated gold NPs (Au@HRP) were purified from the free enzyme via several purification steps (i.e., five centrifugation/redispersion cycles).

In accordance with our previous work on protein and enzyme-coated NPs,[15,16,34,35] the final purified Au@HRP NPs exhibited high colloidal stability at pH 7.4 (Figure B), and the enzyme coating on the gold NPs was very robust and did not detach from the NP surface during the five purification cycles (the fifth supernatant shows no enzymatic activity). The HRP loading was experimentally determined, being ca. 21 HRP molecules per particle (for 15 nm NPs) and theoretically confirmed, as reported previously.[34] The catalytic activity of the Au@HRP NPs was checked by the oxidation of 3,3′,5,5′-tetramethylbenzidine (TMB) in the presence of H2O2 (Figure C). TMB is oxidized by the immobilized HRP in the presence of H2O2, yielding a blue-colored product (TMB•+) upon one-electron oxidation and/or a yellow-colored product (tetramethylbenzidine diimine) upon two-electron oxidation. An exemplary UV–vis spectrum of the oxidized TMB via Au@HRP NPs is shown in Figure C. A detailed and comparative study of the enzymatic activity parameters, that is, the Michaelis–Menten constants Vmax and Km, for the catalytic oxidation of TMB for both the immobilized enzyme (Au@HRP) and the native HRP can be found in our recent publication.[34]

The Au@HRP NPs were used as a reusable colloidal nanobiocatalyst in the polymerization of NIPAAM and DEGMA, yielding thermosensitive polymers with lower critical solution temperatures (LCST) of 32.5 °C[54,55] (PNIPAAM) and 28.0 °C[55] (PDEGMA). The successful formation of polymers can be easily identified at elevated temperatures due to precipitation of insoluble polymers above their LCST. The reaction conforms to the enzyme/ATRPase-catalyzed polymerization of such water-soluble monomers, as reported previously by Bruns et al.[6,7]

The catalytically (re-)active Au15@HRP NPs were used as nanobiocatalyst in the polymerization of NIPAAM and DEGMA under ATRP conditions, in the presence of sodium ascorbate as reducing agent and 2-hydroxyethyl 2-bromoisobutyrate (HEBIB) as typical ATRP initiator. After the reaction was quenched by exposing the solution to air, the Au15@HRP NPs were recovered by centrifugation (60 min, 8000 rcf) and reused in two further polymerization cycles under the same conditions, as illustrated in Figure A. Even after three polymerization cycles, the Au@HRP NPs remain catalytically active and retain their colloidal stability as proven by the color of the NP dispersion (Figure , blue-framed cuvette) and the LSPR band, which did not change in comparison to the as-prepared Au@HRP NPs (red-framed), before the polymerization reaction. Particle loss in the range of approximately 7–11% was recorded after three polymerization cycles (7.25% for NIPAAM, 10.75% for DEGMA), which is attributed to the various centrifugation steps for the recovery. Because both polymers exhibit an LCST, polymer formation was affirmed after each cycle by inducing polymer precipitation upon heating the final purified polymer solution above the respective LCST, which can be observed by the bare eye (Figure B).

To gain further insight into the performance of the Au@HRP NPs in the polymerization reaction over various cycles, the evolution of the molecular weight and molecular weight distribution was studied by gel permeation chromatography (GPC). The GPC measurements (Figure ) revealed monomodal polymer weight distribution for both polymers PNIPAAM and PDEGMA. Furthermore, their molecular weights (Mn and Mw) and polydispersity indices (PDIs) are largely constant for the individual three polymerization cycles, indicating that the catalytic performance of the immobilized HRP molecules could be maintained during the individual polymerization cycles. The exact values of Mn, Mw, and PDI are summarized in Table S1 in the Supporting Information. The results for molecular weight and PDI vary from the results reported in the literature.[6] One possible explanation could be the pH of 7.4 of the reaction medium, which is known to be not optimal for HRP-mediated reactions,[34] including (activators regenerated by electron transfer) ATRP.[6] Nevertheless, pH for the polymerization mixture was set to 7.4, to ensure colloidal stability (Figure C) of the nanobiocatalyst throughout the reaction.[34]

Figure 3

GPC results of polymerization of PNIPAAM and PDEGMA for three polymerization cycles using Au@HRP as nanobiocatalyst (error bars were determined via standard deviation of the three measurements). Left column shows the average values of Mn (A), Mw (B), and PDI (C) of PNIPAAM for all three polymerization cycles. Right column shows the average values of Mn (D), Mw (E), and PDI (F) of PDEGMA for all three polymerization cycles.

To gain a better insight into the polymerization reaction, reaction kinetics via 1H NMR spectroscopy can be followed. However, for monitoring the kinetics via 1H NMR spectroscopy, a total volume of 250 μL (which corresponds to 20 mg of specimen) per NMR sample is required. Because NIPAAM and DEGMA are rather small monomers (Mw = 113 g/mol for NIPAAM and Mw = 188 g/mol for DEGMA), high amount of monomers and catalyst (i.e., ca. 20-fold more) would be required to obtain enough polymer product for a systematic study. However, these upscaling challenges were avoided by using a macromonomer with a similar chemical structure, such as poly(ethylene glycol)methyl ether acrylate (PEGA, Mw: 480 g/mol). By using a high-molecular-weight monomer, the requirements of the 1H NMR measurements could be somewhat fulfilled, without altering the monomer/catalyst ratio exorbitantly. Although, the Au@HRP NP-catalyzed polymerization of PEGA shows a linear correlation between reaction time and the negative natural logarithm of the conversion (for at least the first 5 h, Figure S6 in SI), suggesting that the polymerization follows a first-order kinetics. This indicates that the Au@HRP NP-catalyzed polymerization of PEGA follows somewhat the mechanism of a controlled radical polymerization, which is roughly in accordance with the results reported by Bruns et al. on pure HRP-mediated ATRP of NIPAAM.[6] Also the evolution of Mn and PDI follows roughly the trend in accordance with the pure HRP-mediated ATRP.[6] In view of these results and the complexity of the system, it becomes clear that reaction conditions for a controlled polymerization reaction using Au@HRP NPs as colloidal catalysts have not been found yet. However, detailed studies with variation of various reaction parameters, such as the concentration of ingredients, type of ingredients, and environmental parameters (temperature, pH, buffer, salts), will be required to indentify the reaction mechanism and eventually achieve a well-controlled polymerization reaction. These studies are the subject of ongoing work.

Conclusions

We have demonstrated a simple and straightforward synthesis route of catalytically active HRP-coated gold nanoparticles that can be used as a reusable nanobiocatalyst in several polymerization cycles of vinyl monomers under reaction conditions that are typical for ATRP reactions. However, the Mw and PDI results indicate that the polymerization reactions do not follow a controlled polymerization pathway but rather a free-radical polymerization. Finding the right conditions for achieving a higher degree of control would be the focus of the upcoming reports. Nevertheless, the outstanding characteristic within this system is the simple recovery via centrifugation and reusability of the Au@HRP over multiple cycles. The colloidal stability of the enzyme-coated gold NPs was provided throughout the process, that is, during the enzyme-coating process and the three subsequent polymerization cycles. Furthermore, we were able to limit the particle loss down to 7.25% for the three polymerization cycles of NIPAAM (10.75% for the three polymerization cycles of DEGMA). The molecular weight and PDI, measured by GPC, of the obtained polymers show no significant changes within the individual cycles, indicating that the catalytic activity of the immobilized HRP molecules could be maintained during all polymerization cycles. In view of these results, the use of HRP-coated NPs as a reusable colloidal catalyst is highly feasible in more than just three polymerization cycles. Hence, the enzyme-coated NPs take the enzyme-catalyzed polymerization reactions from the regime of homogeneous catalysis to the regime of quasi-homogeneous catalysis, thus enabling facile separation of the macromolecular catalyst from the macromolecular product.

Materials and Methods

Materials

Peroxidase from horseradish (HRP, lyophilized, powder; Sigma-Aldrich), bovine serum albumin (BSA, ≥98%), N-isopropylacrylamide (NIPAAM, 97%; Sigma-Aldrich), di(ethylene glycol)methyl ether methacrylate (DEGMA, 95%; Sigma-Aldrich), and poly(ethylene glycol)methyl ether acrylate (PEGA, average Mn = 480 g/mol; Sigma-Aldrich) were passed over aluminum oxide (neutral) prior to use to remove the inhibitor. 2-Hydroxyethyl 2-bromoisobutyrate (HEBIB, 95%; Sigma-Aldrich), (+)-sodium l-ascorbate (crystalline, ≥98%; Sigma-Aldrich), gold(III) chloride trihydrate (≥99.9% trace metal basis; Sigma-Aldrich), sodium citrate tribasic dihydrate (ACS reagent, ≥99.0%; Sigma-Aldrich), aluminum oxide (activated, neutral, Brockmann I; Sigma-Aldrich), phosphate-buffered saline (PBS, tablet; Sigma-Aldrich), sodium hydroxide (Grüssing GmbH), and 3,3′,5,5′-tetramethylbenzidine (Sigma-Aldrich) were used as received.

Methods

Gel Permeation Chromatography (GPC)

The apparent number-average molecular weight (Mn) and weight-average molecular weight (Mw) based on polystyrene standards and the polydispersity index (PDI) of the yielded polymers were determined by GPC. Measurements of PNIPAAM, PDEGMA, and PEGA were performed on an Agilent 1260 Infinity series equipped with an autosampler, a diode array detector, and a refractive index detector. They were measured with a precolumn (GRAM 10 μm (8.0 × 50 mm2)) and two main columns (GRAM 10 μm, 100 Å (8.0 × 300 mm2) and GRAM 10 μm, 3000 Å (8.0 × 300 mm2)) thermostatted to room temperature, using dimethylformamide (DMF, +5 g/L lithium bromide) as the mobile phase with a flow rate of 0.5 mL/min. The column setup was provided by Polymer Standards Service GmbH. GPC samples were prepared by dissolving the lyophilized polymer in DMF (polymer’s concentration amounts to 2 mg/mL). After homogenization overnight, 20 μL of the samples was injected into the GPC device. The measurement time was 65 min.

1H NMR

1H NMR spectroscopy was carried out with a Bruker Avance 300 (300 MHz) instrument. All measurements were performed at room temperature using deuterated dimethyl sulfoxide as solvent.

UV–Vis Spectroscopy

UV–vis spectra were acquired with a Specord Plus 250 (Analytik Jena) spectrophotometer. For measurements, 1.5 mL of the corresponding solution was placed in a cell and spectral analysis was performed in the 300–1100 nm range at room temperature.

Transmission Electron Microscopy (TEM)

The average size, distribution, and morphology of citrate-stabilized AuNP and HRP-coated AuNP before and after every polymerization cycle were analyzed by TEM using a LEO 922 A EFTEM microscope (OMEGA). TEM samples were prepared by placing one drop of diluted aqueous dispersion of AuNP in a carbon-coated copper grid and allowing the solvent to evaporate at room temperature. The average particle diameter and distribution were determined by evaluating at least 300 individual particles using the software ImageJ.

Preparation of 15 nm Gold Nanoparticles (Au15@citrate)

Gold NPs of about 15 nm in size were prepared by the citrate reduction method.[56] In a typical procedure, 750 mL of 0.25 mM HAuCl4 solution was put in an Erlenmeyer flask and heated to boiling. Under vigorous stirring, 18 mL of a 1 wt % citrate solution was added, resulting in a color change of the solution from a grayish-blue to deep wine-red within 10 min. Now the heating is stopped and the solution cools down with stirring overnight. The resulting Au15@citrate was stored at 4 °C.

Preparation of 100 nm Gold Nanoparticles (Au100@Citrate)

Quasi-spherical citrate-stabilized AuNPs with an average particle size of ∼100 nm were synthesized by the citrate reduction-based seeded growth method developed by Puntes et al.[57] In a 250 mL three-neck round-bottom flask, 150 mL of a 2.2 mM aqueous solution of sodium citrate was heated in an oil bath for 15 min under reflux and vigorous stirring at permanent control of the solution temperature. At 100 °C, 1 mL of a 25 mM HAuCl4 solution was immediately injected, by which the color of the solution changed from yellow to bluish-gray and then to soft pink within 10 min. The resulting citrate-stabilized AuNPs were used as seeds for the next seeded growth steps by successive addition of Au precursor, citrate solution, Milli-Q water, and dilution steps as reported elsewhere.[57] After 11 consecutive growth steps, citrate-stabilized 100 nm AuNPs were obtained. The resulting NPs were stored at 4 °C.

Immobilization of HRP onto Gold Nanoparticles

For preparation of an HRP solution, 2 mg of HRP is dissolved in 10 mL of a 0.1 wt % citrate solution. Subsequently, 90 mL of a filtered (Nylon, 0.1 μm) particle solution (9.1 × 10–5 M, pH = 9) is added dropwise under stirring to the protein solution (final HRP concentration of 0.02 mg/mL). The protein/particle solution is stirred overnight and afterward purified by five centrifugation steps (60 min, 8000 rcf; 4 °C). After every centrifugation step, the gold NPs are redispersed in basic water (pH = 9). The supernatant of the fifth centrifugation cycle was checked on enzymatic activity and enzyme desorption from the NPs surface via TMB oxidation.

Activity Measurements of Au15@HRP

Activity measurements of Au@HRP NPs were performed following the procedures reported elsewhere.[34]

ATRP of NIPAAM and DEGMA Using Au15@HRP as Nanobiocatalyst

The polymerization was carried out in 100 mM PBS solution at pH 7.4. HEBIB (2.4 μL, 3.5 mg, 16.5 μmol) and monomer (NIPAAM: 127.5 mg, 1.127 mmol; DEGMA: 121.1 mg, 207.9 μL, 1.127 mmol; NIPAAM was used as received; DEGMA was passed over aluminum oxide (neutral) prior to use to remove the inhibitor) were weighed in a flask and dissolved in 3.75 mL of PBS buffer. In two separate flasks, sodium ascorbate (7.5 mg, 37.9 μmol) was dissolved in 2.5 mL of PBS buffer and 1 mL of an HRP-coated gold NP solution ([Au0] = 4 mM) was weighed. Au15@HRP was obtained from a stock solution, and its concentration was measured via UV–vis spectroscopy. Oxygen was removed from all three flasks by purging the solutions with nitrogen for 20–30 min. The monomer/initiator mixture (750 μL) was syringed to the gold NP solution and then the polymerization was started by transferring 250 μL of the reducing agent solution into the Au15@HRP/NIPAAM/HEBIB solution in the vial by means of a syringe. The ratio of reactants in the reaction mixture was 1/68/1.1/1212 (HEBIB/monomer/sodium ascorbate/Au15@HRP (# NPs)). The reaction was stirred at room temperature under nitrogen atmosphere for 2.5 h. Subsequently, the polymerization was quenched by exposing the reaction to air, and the Au15@HRP NPs were removed from the solution by several centrifugation steps (60 min, 8000 rcf). After every centrifugation cycle, the Au15@HRP NPs were rinsed with PBS. The recovered gold NPs were used in two further polymerization cycles under the same conditions. The successful polymer formation can be easily confirmed due to precipitation of insoluble polymer above its LCST. Molecular weight and PDI of the polymer chains were investigated with GPC. Every data point was calculated out of three individual polymerizations. The uncertainty of the GPC device was in the range of 12–12.9%.

Polymerization of PEGA Using Au15@HRP as Nanobiocatalyst

The polymerization of PEGA using Au15@HRP as biocatalyst was carried out analogously to the polymerization of NIPAAM and DEGMA. HEBIB (4.19 μL, 6.1 mg, 28.89 μmol) and PEGA (1.08 g, 0.992 mL, 2.25 mmol, PEGA was passed over aluminum oxide (neutral) prior to use to remove the inhibitor) were weighed in a flask and dissolved in 3.0 mL of PBS buffer (pH = 7.4). In two separate flasks, sodium ascorbate (1.73 mg, 158.3 μmol) was dissolved in 2.5 mL of PBS buffer (pH = 7.4) and 2 mL of an Au15@HRP solution ([Au0] = 4 mM) was weighed. Oxygen was removed from all three flasks by purging the solutions with nitrogen for 30 min. The monomer/initiator mixture (2.5 mL) was syringed to the AuNP solution. Subsequently, polymerization was started by transferring 1 mL of the reducing agent solution into the Au15@HRP/PEGA/HEBIB solution by means of a syringe. The ratio of the reactants was 1/79/1.4/332 (HEBIB/PEGA/sodium ascorbate/Au15@HRP). Samples (250 μL) were taken in periodical time intervals to track the conversion rate via 1H NMR spectroscopy.