- Email: [email protected]
Synthesis and characterization of enzyme–magnetic nanoparticle complexes: effect of size on activity and recovery
Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Synthesis and characterization of enzyme–magnetic nanoparticle complexes: effect of size on activity and recovery Hee Joon Park a , Joshua T. McConnell a , Soheil Boddohi b , Matt J. Kipper b,c , Patrick A. Johnson a,∗ a b c
Department of Chemical and Petroleum Engineering, University of Wyoming, Laramie, WY 82071, USA Department of Chemical and Biological Engineering, Colorado State University, Fort Collins, CO 80523, USA School of Biomedical Engineering, Colorado State University, Fort Collins, CO 80523, USA
a r t i c l e
i n f o
Article history: Received 29 March 2010 Received in revised form 5 October 2010 Accepted 9 November 2010 Available online 1 December 2010 Keywords: Magnetic nanoparticles Biocatalysis Glucose oxidase X-ray photoelectron spectroscopy
a b s t r a c t The influence of particle size on the activity and recycling capabilities of enzyme conjugated magnetic nanoparticles was studied. Co-precipitation and oxidation of Fe(OH)2 methods were used to fabricate three different sizes of magnetic nanoparticles (5 nm, 26 nm and 51 nm). Glucose oxidase was covalently bound to the magnetic nanoparticles by modifying the surfaces with 3-(aminopropyl)triethoxysilane (APTES) and a common protein crosslinking agent, glutaraldehyde. Analysis by Transmission Electron Microscopy (TEM) showed that the morphology of the magnetic nanoparticles to be spherical and sizes agreed with results of the Brunauer, Emmett, and Teller (BET) method. Magnetic strength of the nanoparticles was analyzed by magnetometry and found to be 49 emu g−1 (5 nm), 73 emu g−1 (26 nm), and 85 emu g−1 (51 nm). X-ray photoelectron spectroscopy (XPS) confirmed each step of the magnetic nanoparticle surface modification and successful glucose oxidase binding. The immobilized enzymes retained 15–23% of the native GOx activity. Recycling stability studies showed approximately 20% of activity loss for the large (51 nm) and medium (26 nm) size glucose oxidase-magnetic nanoparticle (GOxMNP) bioconjugate and about 96% activity loss for the smallest GOx-MNP bioconjugate (5 nm) after ten cycles. The bioconjugates demonstrated equivalent total product conversions as a single reaction of an equivalent amount of the native enzyme after the 5th cycle for the 26 nm nanoparticles and the 7th cycle for the 51 nm nanoparticles. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Enzymes are versatile proteins with great potential for applications in research and industry due to their myriad biocatalytic transformations with chemo-, regio-, and stereospecificity in mild process conditions. Evolving in cellular systems, enzymes are not well adapted to continuous, high throughput, single-product industrial processes. Enzymes lack long-term stability, prove difficult to separate and reuse, and usually require extensive downstream processing [1]. Research into how to adapt enzymes for more industrial uses has led to the development of various carrier-bound immobilized enzymes, which facilitate the use of enzymes in continuous processes and help overcome the cost constraints by aiding in efficient separation and recycling of costly enzymes [2]. Through the enzyme immobilization process, enzymes can be widely dispersed in reaction solutions without aggregation and enzyme stability can
∗ Corresponding author at: Department of Chemical and Petroleum Engineering, University of Wyoming, 1000 E. University Avenue, Dept. 3295, Laramie, WY 82071, USA. Tel.: +1 307 766 6524; fax: +1 307 766 6777. E-mail address: [email protected] (P.A. Johnson). 0927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2010.11.006
be enhanced with multivalent covalent attachment [3–5]. Among possible carriers, magnetic nanoparticles (Fe3 O4 ) show promise due to their simple, non-chemical separation method [5–11] and despite the non-porous structure, they permit high quantities of enzyme loading due to the large surface area [12]. In the last decade, there has been growing interest in magnetic nanoparticle bioconjugates for use in applications from drug delivery, hyperthermia treatment, and cell separation, to biosensors and enzymatic assays [13,14], biocatalysts [15] and environmental remediation [16]. Magnetic nanoparticles have been used as support materials for immobilization of enzymes such as yeast alcohol dehydrogenase [6] and lipase [7], with various surface modifications. Various attempts have been made to immobilize glucose oxidase on different types of solid supports using glutaraldehyde as a cross-linking agent for biosensor and biofuel cell applications [17,18]. In our experiments, glucose oxidase (GOx, EC 1.1.3.4) was immobilized on three sizes of magnetic particles. Glucose oxidase is a homodimer flavoprotein containing two active sites per molecule [19,20], catalyzing the oxidation of -d-glucose to gluconic acid concomitant with the reduction of oxygen to hydrogen peroxide [8]. Glucose oxidase has been used as a model enzyme to test
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203
various enzyme immobilization techniques and also shows great potential for biosensor and biofuel cell applications [5,21,22]. In this work, three sizes of magnetic nanoparticles were fabricated (5 nm, 26 nm and 51 nm) using two different methods, co-precipitation and ferrous hydroxide oxidation. The magnetic nanoparticles were modified with 3-(aminopropyl)triethoxysilane (APTES) and glutaraldehyde to covalently immobilize enzymes to the surface. Magnetic nanoparticles were characterized using Transmission Electron microscopy (TEM), the Brunauer, Emmett and Teller (BET) method and physical properties measurement system (PPMS). Each layer of surface modification was confirmed with X-ray photoelectron spectroscopy (XPS). Long-term enzyme and recycling stability were also examined.
199
2.4. Glucose oxidase immobilization
2. Experimental
10 mg of magnetic nanoparticles modified with APTES were dispersed by sonication in 8 mL of 8% glutaraldehyde solution (pH 7.4) and vortexed for 1 h. The magnetic nanoparticles were separated by magnetic decantation and washed 3 times with 10 mL of PBS (pH 7.4, 137 mM NaCl, 2.7 mM KCl, 8 mM Na2 HPO4 , and 2 mM KH2 PO4 ). The magnetic nanoparticles were redispersed in 2 mL of PBS. 10 mL of glucose oxidase solution (1 mg/mL in PBS) was added to the magnetic nanoparticles and gently vortexed overnight. The GOx-MNP bioconjugate was separated by magnetic decantation, washed 3 times with 10 mL of PBS, dispersed in 10 mL PBS and stored at 4 ◦ C. The supernatant from the Gox-MNP bioconjugate and last wash solutions were kept in order to determine enzyme loading efficiency, using BCA assay kit.
2.1. Materials
2.5. Determination of glucose oxidase activity
Glucose oxidase type II-S solid from Aspergillus niger (-d-glucose: oxygen 1-oxidoreductase, EC 1.1.3.4), 3(aminopropyl)triethoxysilane (APTES) (99%), iron (II) chloride (99%), iron (III) chloride (99%), and glutaraldehyde (8% aqueous) were purchased from Sigma–Aldrich (St. Louis, MO). Hydrogen peroxide (30%) and ammonium hydroxide (29.5%) were purchased from Fisher Scientific (Pittsburgh, PA). Anhydrous methanol (ACS grade) was purchased from EMD Chemicals (Gibbstown, NJ). Phosphate-buffered saline (PBS) solution (pH 7.4, 10×) was purchased from invitrogen (Carlsbad, CA). A glucose oxidase assay kit was purchased from Megazyme (Wicklow, Ireland). A BCA assay kit was purchased from Thermo Scientific (Rockford, IL). Neodymium magnets (DY0X0-N52; 14,800 Gauss) were purchased from K&J Magnetics, Inc. Deionized water was used in the preparation of aqueous solutions.
Glucose oxidase activity was measured using a glucose assay kit purchased from Megazyme. Glucose oxidase catalyzes the oxidation of -d-glucose to d-glucono-␦-lactone, which then hydrolyzes to gluconic acid and hydrogen peroxide. The resultant hydrogen peroxide then reacts with p-hydroxybenzoic acid and 4-aminoantipyrine in the presence of peroxidase to form the quinoneimine dye complex which has a strong absorbance at 510 nm. The GOx-MNP bioconjugate suspension was sonicated and 0.5 mL of suspension was added to chemical mixture according to glucose assay kit preparation procedures.
2.2. Magnetic nanoparticle fabrication Three sizes of magnetic nanoparticles were prepared by two different precipitation methods. Method A is co-precipitation in alkaline media. A solution of Fe2+ /Fe3+ with a molar ratio of 0.5 (0.0125 M/0.025 M) in 51 mL of deionized water was stirred and ammonium hydroxide (29.5%) added at a rate of 0.2 mL/min to adjust the pH to 10 (approximately 2.8–2.9 mL). The precipitation process was carried out at room temperature. Method B is the oxidative alkaline hydrolysis of ferrous ions. 0.05 M of FeCl2 was dissolved in 51 ml of deionized water. An aqueous suspension of Fe(OH)2 was prepared by adding 1 M KOH, adjusting to pH 7.9–8 with stirring. Precipitation proceeded for 2 h at room temperature. Method C is the oxidative alkaline hydrolysis of ferrous ions at high temperature. 0.05 M of FeCl2 was dissolved in 51 ml of deionized water while heated to 90 ◦ C. The aqueous suspension of Fe(OH)2 was prepared by the same procedure as Method B. Precipitation was carried out for 2 h at 90 ◦ C with stirring. The black precipitates from all three synthetic methods were strongly attracted by a permanent magnet and separated by magnetic decantation. The MNPs were washed 3 times with 51 mL of deionized water, 2 times with 51 mL of ethanol, and dried at room temperature. 2.3. Surface modification of magnetic nanoparticles with APTES 10 mg of magnetic nanoparticles were dispersed in 10 mL of 10% APTES solution (anhydrous methanol base) using sonication. The solution was vortexed overnight at room temperature then separated by magnetic decantation. MNPs were washed 3 times with 10 mL of anhydrous methanol and stored at 4 ◦ C for the enzyme binding process.
2.6. Recycling stability test Recycling stability of the GOx-MNP bioconjugate was determined by measuring the activity of the three GOx-MNP bioconjugates 10 consecutive times. -d-Glucose, p-hydroxybenzoic acid, 4-aminoantipyrine and peroxidase were mixed in one cuvette and added to 0.5 mL of GOx-MNP bioconjugate. After 17 min the GOxMNP bioconjugate was collected at the bottom of the cuvette for 3 min. Then dye solution was removed by pipette, transferred to a cuvette and measured at 510 nm. After each activity measurement, the GOx-MNP bioconjugate was washed 3 times with PBS (pH 7.4) and collected using magnetic decantation. The washing solutions from each of the 10 consecutive cycles were collected in a beaker to avoid losing GOx-MNP bioconjugate. After the last cycle, the activity of the particles collected in the beaker was measured to calculate the true activity loss from recycling as compared to enzyme deactivation. 2.7. Characterization The size and morphology of magnetic nanoparticles were determined by Transmission Electron Microscopy (TEM, Hitachi H-7000) at 75 kV. The samples were prepared on a formvar support film with a copper grid (300 mesh, Electron Microscopy Science, USA) by dropping magnetic nanoparticles in ethanol solution on the support and then evaporating in air at room temperature. The mean diameter of each sample was also determined by the Brunauer, Emmett, and Teller (BET) theory using a BET instrument. The binding of APTES and glucose oxidase on magnetic nanoparticles was confirmed by surface analysis with X-ray Photoelectron Spectroscopy (XPS, 5800 series, Multi-Technique ESCA systems). A sample from each stage of bioconjugate synthesis method B (magnetic nanoparticle, magnetic nanoparticle w/APTES, magnetic nanoparticle w/APTES and glutaldehyde, Gox-MNP bioconjugate) was immobilized on double-sided tape before putting in the specimen chamber under ultra high vacuum. Survey and high resolution results were collected and analyzed by XPSPeak 4.1 software. The
200
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203
Fig. 1. Transmission electron microscopy (TEM) images of Fe3 O4 nanoparticles. (A) Co-precipitation method, 5 nm; (B) oxidizing ferrous hydroxide at 25 ◦ C, 26 nm; (C) oxidizing ferrous hydroxide at 90 ◦ C, 50 nm. Table 1 Physical properties of magnetic nanoparticles for each synthetic method (AS , surface area; DBET , particle size based on BET; DTEM , particle size based on TEM; MS , saturation magnetization). Method
As (m2 g−1 )
DBET (nm)
DTEM (nm)
MS (emu g−1 )
A B C
218 44 23
5 26 51
5 24 48
48.5 72.8 84.5
of 72.8 emu g−1 , which is lower than that of bulk magnetite (92 emu g−1 ) [24]. The largest magnetic nanoparticles, 51 nm, were obtained from Method C. The largest magnetic nanoparticles showed the strongest magnetization among samples (84.5 emu g− 1), which is close to the magnetization value of bulk magnetite. Diameter measurements from TEM were consistent with those from BET measurements. The particle size and magnetization results are summarized in Table 1.
magnetic properties were measured by magnetometry in the Physical Property Measurement System (PPMS) from Quantum Design. 3. Results and discussion 3.1. Particle size and physical properties of magnetic nanoparticles Characteristic TEM images of magnetic nanoparticles from the three different methods are shown in Fig. 1. Observed sizes of the magnetic nanoparticles were consistent with previously reported research [8,21,23,24]. The smallest particles, 5 nm diameter, were obtained from ferrous/ferric co-precipitation method at room temperature (Method A) and showed the lowest magnetization (48.5 emu g−1 ) among the samples. Particles with a diameter of 26 nm were synthesized by Method B and showed a magnetization
Fig. 2. XPS survey scan of bare magnetic nanoparticles (A) and after APTES modification on the surface of magnetic nanoparticels (B). (Magnetic nanoparticles were synthesized using method B.)
Fig. 3. High resolution XPS spectra of (A) C 1s and (B) O 1s for each step of bioconjugate synthesis using method B (MNP, MNP w/APTES, MNP-APTES w/Glutaraldehyde and Gox-MNP bioconjugate).
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203 Table 2 Chemical composition (atom %) of MNPs based on high resolution XPS analysis. (Magnetic nanoparticles were synthesized using method B.).
O1s Fe2p C 1s N 1s Si 2p a
Bare MNP
MNP-APTES
MNP-APTESglutaraldehyde
GOx-MNP conjugate
45.5 32 22.5a N/A N/A
39.6 21.6 28.4 1.6 8.8
42.4 25.3 23 2.2 7.1
36.5 16.8 34.3 4.8 7.6
Impurity.
Size distributions of the three types of magnetic nanoparticles varied depending on the synthetic method. Particles from Method A showed the smallest size distribution, with particle sizes ranging from 3 nm to 9 nm. Method B had a slightly larger distribution with particle sizes range from 13 nm to 36 nm. Particles from Method C had the largest size range, from 8 nm to 133 nm.
201
the presence of cross linker derived species C O (288.5 eV). After glucose oxidase modification, elevated C–N and C N (285.4 eV) and C O and C–O (287.7 eV) peaks were observed. The peaks were significantly higher than the glutaraldehyde modified magnetic nanoparticles, characteristic of peptide bonds and supporting the presence of glucose oxidase immobilized on the nanoparticle surface. Fig. 3(B) shows the high resolution peak shift of O 1s throughout the bioconjugate fabrication process. The largest peak contribution for bare magnetic nanoparticles was from oxygen in the crystal structure with binding energy of 529.6 eV, which is typical of oxide crystal structures [26]. A hydroxyl functional group on the surface of magnetic nanoparticles was indicated at 530.9 eV. After APTES modification, the hydroxyl peak shifted to 531.6 eV due to covalent bonding with silicon at the surface (Fig. 3(B)). Glucose oxidase binding to the magnetic nanoparticle caused significant peak shifts due to the OH, O–C, O–H, O–N and O–C–N functional groups in protein structure.
3.2. Binding confirmation and surface analysis
3.3. Binding efficiency, activity measurement
XPS survey spectra and high resolution spectra of the Fe2p1, Fe2p3, C 1s, O 1s, N 1s and Si 2p orbitals were obtained for each step of the fabrication processes (magnetic nanoparticles, magnetic nanoparticles w/APTES, magnetic nanoparticles w/APTES and glutaraldehyde, and GOx-MNP bioconjugate). Table 2 gives an overview of the high resolution spectra atom % data for each step of the fabrication process. Advantageous carbon or rubbish carbon was observed on the bare magnetic nanoparticles, an impurity remaining from the fabrication processes. Shown in Fig. 2, the survey scan confirms the presence of nitrogen and silicon after APTES modification. The binding of APTES was also supported by the presence of an amine functional group confirmed by the presence of a C–N peak for C 1s in the high resolution spectra at 285.2 eV (Fig. 3 and Table 3) [25]. Glutaraldehyde modification was confirmed by
After the immobilization of the enzyme to the MNPs, the unbound protein in the supernatant was assayed using the BCA assay kit. All three types of MNP showed between 7.3% and 7.6% binding efficiency, despite the relatively large differences in surface area (Table 1). The enzymatic activity of GOx-MNP bioconjugate suspensions was measured using the glucose oxidase kit from Megazyme. Activity measurements were taken for three months after preparation and no significant decrease in enzymatic activity was noticed (results not shown). Enzyme activity after immobilization on MNPs was measured and compared to the native enzyme to determine the percent of activity retained. The 5 nm MNPs demonstrated the highest activity retention (23%), while the largest size MNPs (51 nm) showed the least, at 15% of the native enzyme activity.
Table 3 XPS peak-fitted parameters of Fe2p1, Fe2p3, C 1s, O 1s, N 1s and Si 2p on surface of magnetic nanoparticle. (Magnetic nanoparticles were synthesized using method B.).
MNP (Bare)
MNP w/APTES
MNP-APTES w/glutaraldehyde
MNP-APTES-Glu w/GOx
Element
Binding energy (eV)
FWHM (eV)
Fe2p1 Fe2p3 C 1s O 1s Fe2p1 Fe2p3 C 1s O 1s N 1s Si 2p Fe2p1 Fe2p3 C 1s
723.87 711.16 284.39 529.60 530.90 723.86 710.91 283.95 285.21 529.77 531.61 399.48 101.30 723.31 710.66 284.07 285.84 288.51 529.51 531.34 399.47 100.80 723.98 710.87 284.29 285.44 287.71 529.55 531.01 399.73 101.42
11.68 4.37 2.18 1.85 2.44 8.19 4.11 1.85 1.71 2.16 2.09 2.70 2.32 8.45 4.26 2.20 2.03 2.96 2.09 1.76 3.65 2.69 7.21 3.87 1.94 2.28 3.03 1.73 2.82 2.62 2.08
O 1s N 1s Si 2p Fe2p1 Fe2p3 C 1s O 1s N 1s Si 2p
Area 15,797.16 15,137.04 2692.96 7603.96 6691.90 13,770.37 11,430.19 3592.41 897.15 10,444.57 5382.42 563.98 1458.03 11,128.58 7965.67 1701.25 535.15 250.04 8206.39 2350.07 497.72 665.96 5344.11 5451.44 2050.11 1462.45 335.43 2496.07 6653.37 911.35 828.93
Chemical bond Iron oxide Iron oxide C–C Iron oxide O–H Iron oxide Iron oxide C–C C–N Iron oxide O–Si N–C Si–O Iron oxide Iron oxide C–C C N C O Iron oxide O C/O–Si N–C/N C Si–O/Si–C Iron oxide Iron oxide C–C C N C O/C–O Iron oxide O–H/O–C/O C/O–Si N–C/N C Si–O
C/Fe
0.09
0.18
0.13
0.36
202
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203
Fig. 5. Specific activity comparison (A) Specific activity of bioconjugates with 5 nm, 26 nm and 51 nm MNP in every cycle and specific activity of native GOx. (B) Total product conversion of bioconjugates and native GOx.
Fig. 4. Enzymatic activity of glucose oxidase–magnetic nanoparticle bioconjugate for 10 consecutive cycles. (A) Bioconjugate with 50 nm MNP; (B) bioconjugate with 26 nm MNP; (C) bioconjugate with 5 nm MNP. (Standard error (SE) is indicated as error bars.)
3.4. Recycling stability The possibility of recycling the GOx-MNP bioconjugate for successive activity measurements was investigated. The procedure was conducted 10 consecutive times for each of the three difference sizes of magnetic nanoparticles (Fig. 4). To account for particle loss from cycling, the supernatant from each cycle rinse was col-
lected in a beaker and the GOx-MNP bioconjugate was separated by prolonged exposure to the magnet. The activity for each cycle, the activity from the supernatant residues and the activity loss during the ten consecutive cycling tests is shown in Fig. 4. The largest sized GOx-MNP bioconjugate (51 nm) retained 57% of original activity, and counting the activity from the supernatant residues, retained a total activity around 80%. The medium sized GOx-MNP bioconjugate (26 nm) retained the highest activity after 10 cycles with 72%. With 9% activity from residues, the total retained activity was 81% of the overall activity. The smallest GOx-MNP bioconjugate (5 nm) showed a dramatic decrease in activity. Only 4% of original activity was retained by the 10th cycle, with the supernatant residues exhibiting the highest activity of all three bioconjugates at 20%, and an overall activity loss of 75%. The losses for the smallest bioconjugate can be explained by the difference of magnetization with particle size. The smallest bioconjugate particle had the lowest magnetization (48.5 emu g−1 ) and 3 min between the cycling tests might not be sufficient time to allow all the particles in solution to be acted on by the magnet, and settle to the bottom of cuvette, so loss of GOx-MNP bioconjugate during each cycle resulted in a large decrease in activity. A comparison of the 51 nm and 26 nm bioconjugates from Fig. 4 shows the 51 nm bioconjugate lost almost 25% of its original activity at fourth cycling, compared to 12% for the 26 nm bioconjugate. A 25% decrease in activity is not observed for the 26 nm bioconjugate until the 10th cycle. Also, the decrease in activity of the 26 nm bioconjugate was relatively steady throughout all 10 cycles. The 51 nm
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203
bioconjugate was synthesized using particle fabrication Method C and had a wide size distribution, from 8 nm to 133 nm. Particles less than 10 nm likely were lost throughout recycling process, as was seen with the 5 nm bioconjugate. By contrast, the 26 nm bioconjugate showed a relatively narrow particle distribution (13–36 nm) and a relatively small decrease in activity. 3.5. Specific activity of bioconjugates and total product conversion Using binding efficiency and activity measurements, the specific activity in each cycle of the three types of bioconjugate was calculated (Fig. 5(A)). The highest initial specific activity was found in 5 nm bioconjugates (5770 U/g protein) with dramatic decreases evident over the 10 cycles. The 26 nm bioconjugate showed an initial specific activity of 5070 U/g protein and the lowest specific activity (3810 U/g protein) was shown for the 51 nm bioconjugate. A comparison of the total product conversion (Fig. 5(B)) shows the quantity of product produced by the enzymatic reaction. Native GOx showed very high product conversion (507 mmol/g over 20 min). For the GOx-MNP bioconjugate, productivity by the 5th cycle for the 26 nm bioconjugate and the 7th cycle for the 51 nm bioconjugate was equivalent to a single cycle of the native GOx. Additional productivity above the native GOx was achieved by subsequent recycling for the two GOx-MNP bioconjugates. The 5 nm bioconjugate was not able to attain the equivalent of the native enzyme product conversion until the 10th recycling test. 4. Conclusion Three different sizes of magnetic nanoparticles (5 nm, 26 nm, and 51 nm) were fabricated by coprecipitation and oxidation of Fe(OH)2 , and characterized by BET and TEM to analyze the diameter and morphology of the particles. Magnetization of the particles was measured by PPMS. Surface modification of magnetic nanoparticles was carried out with APTES and the enzyme glucose oxidase was covalently immobilized on the magnetic nanoparticles with glutaraldehyde. The enzymes retained their activity after immobilization and stability was tested by recycling GOx-MNP bioconjugate for 10 consecutive cycles. Successful surface modification was confirmed by XPS studies and results showed strong evidence of APTES, glutaraldehyde (C O 288.5 eV) and protein modification (C–N 285.4 eV and C O 287.7 eV) on the magnetic nanoparticle surface. 0.54–0.56 mg of GOx enzyme was immobilized on the three different types of MNPs with 15–22.7% of the original enzyme activity retained. From the recycling stability studies, it was observed that the 26 nm size MNPs showed the highest activity retention after 10 cycles and the least amount of particle loss between cycles, while the 5 nm MNPs had the highest initial activity but were the most difficult to recover. Increasing the size of the magnetic core to 51 nm did not further improve the recoverability over the 26 nm MNPs. Evidence of the benefit of using the recyclable biocatalysts was realized after the 5th cycle for the 26 nm bioconjugates and the 7th cycle for the 51 nm bioconjugates. This
203
work shows the potential of recyclable biocatalysts for practical applications in chemical reaction processes and can be a model for the use of a wider variety of enzymes. Acknowledgements This research was supported by funding from the North Central Regional Sun Grant Center at South Dakota State University through a grant provided by the US Department of Transportation, Office of the Secretary, Grant No. DTOS59-07-G-00054. Additional funding was provided by the School of Energy Resources (SER) at the University of Wyoming. The authors would also like to thank the help of Dr. Jinke Tang for the assistance with the Physical Property Measurement System. References [1] U.T. Bornscheuer, Angewandte Chemie International Edition 42 (2003) 3336–3337. [2] R.A. Sheldon, R. Schoevaart, L.M. Van Langen, Biocatalysis & Biotransformation 23 (2005) 141–147. [3] M.Y. Arica, Journal of Applied Polymer Science 77 (2000) 2000–2008. [4] C. Mateo, O. Abian, R. Fernandez-Lafuente, J.M. Guisan, Enzyme and Microbial Technology 26 (2000) 509–515. [5] J. Huang, R. Zhao, H. Wang, W. Zhao, L. Ding, Biotechnology Letters 32 (2010) 817–821. [6] M. Liao, D. Chen, Biotechnology Letters 23 (2001) 1723–1727. [7] S. Huang, M. Liao, D. Chen, Biotechnology Progress 19 (2003) 1095–1100. [8] G. Kouassi, J. Irudayaraj, G. McCarty, BioMagnetic Research and Technology 3 (2005) 1. ´ M. Antalík, M. Timko, C.N. Ramchand, D. Lobo, [9] M. Koneracká, P. Kopcansky, R.V. Mehta, R.V. Upadhyay, Journal of Magnetism and Magnetic Materials 201 (1999) 427–430. [10] Y. Liu, R. Hamby, R.D. Colberg, Powder Technology 64 (1991) 3–41. [11] T. Bahar, S.S. C¸elebi, Enzyme and Microbial Technology 26 (2000) 28–33. [12] L. Betancor, M. Fuentes, G. Dellamora-Ortiz, F. López-Gallego, A. Hidalgo, N. Alonso-Morales, C. Mateo, J.M. Guisán, R. Fernández-Lafuente, Journal of Molecular Catalysis B: Enzymatic 32 (2005) 97–101. [13] Q.A. Pankhurst, J. Connolly, S.K. Jones, J. Dobson, Journal of Physics D: Applied Physics 36 (2003) R167–R181. [14] P. Tartaj, M.D.P. Morales, S. Veintemillas-Verdaguer, T. Gonzalez-Carreno, C.J. Serna, Journal of Physics D: Applied Physics 36 (2003) R182–R197. [15] L. Zeng, K. Luo, Y. Gong, Journal of Molecular Catalysis B: Enzymatic 38 (2006) 24–30. [16] D. Horák, M. Babic, H. Macková, M.J. Benes, Journal of Separation Science 30 (2007) 1751–1772. [17] S. Libertino, A. Scandurra, V. Aiello, F. Giannazzo, F. Sinatra, M. Renis, M. Fichera, Applied Surface Science 253 (2007) 9116–9123. [18] V. Bulmus, H. Ayhan, E. Piskin, Chemical Engineering Journal 65 (1997) 71–76. [19] K.C. Gulla, M.D. Gouda, M.S. Thakur, N.G. Karanth, Biosensors and Bioelectronics 19 (2004) 621–625. ´ J. Zeremski, D. Periˇcin, V. Leskovac, Molecular and Cel[20] G. Wohlfahrt, S. Trivic, lular Biochemistry 260 (2004) 69–83. [21] L. Rossi, A. Quach, Z. Rosenzweig, Analytical and Bioanalytical Chemistry 380 (2004) 606–613. [22] J. Kim, H. Jia, P. Wang, Biotechnology Advances 24 (2006) 296–308. [23] J. Mürbe, A. Rechtenbach, J. Töpfer, Materials Chemistry and Physics 110 (2008) 426–433. [24] M. Tada, S. Hatanaka, H. Sanbonsugi, N. Matsushita, M. Abe, Journal of Applied Physics 93 (2003) 7566–7568. [25] S. Delpeux, F. Beguin, R. Benoit, R. Erre, N. Manolova, I. Rashkov, European Polymer Journal 34 (1998) 905–915. [26] J.T. Kloprogge, L.V. Duong, B.J. Wood, R.L. Frost, Journal of Colloid and Interface Science 296 (2006) 572–576.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Synthesis and characterization of enzyme–magnetic nanoparticle complexes: effect of size on activity and recovery Hee Joon Park a , Joshua T. McConnell a , Soheil Boddohi b , Matt J. Kipper b,c , Patrick A. Johnson a,∗ a b c
Department of Chemical and Petroleum Engineering, University of Wyoming, Laramie, WY 82071, USA Department of Chemical and Biological Engineering, Colorado State University, Fort Collins, CO 80523, USA School of Biomedical Engineering, Colorado State University, Fort Collins, CO 80523, USA
a r t i c l e
i n f o
Article history: Received 29 March 2010 Received in revised form 5 October 2010 Accepted 9 November 2010 Available online 1 December 2010 Keywords: Magnetic nanoparticles Biocatalysis Glucose oxidase X-ray photoelectron spectroscopy
a b s t r a c t The influence of particle size on the activity and recycling capabilities of enzyme conjugated magnetic nanoparticles was studied. Co-precipitation and oxidation of Fe(OH)2 methods were used to fabricate three different sizes of magnetic nanoparticles (5 nm, 26 nm and 51 nm). Glucose oxidase was covalently bound to the magnetic nanoparticles by modifying the surfaces with 3-(aminopropyl)triethoxysilane (APTES) and a common protein crosslinking agent, glutaraldehyde. Analysis by Transmission Electron Microscopy (TEM) showed that the morphology of the magnetic nanoparticles to be spherical and sizes agreed with results of the Brunauer, Emmett, and Teller (BET) method. Magnetic strength of the nanoparticles was analyzed by magnetometry and found to be 49 emu g−1 (5 nm), 73 emu g−1 (26 nm), and 85 emu g−1 (51 nm). X-ray photoelectron spectroscopy (XPS) confirmed each step of the magnetic nanoparticle surface modification and successful glucose oxidase binding. The immobilized enzymes retained 15–23% of the native GOx activity. Recycling stability studies showed approximately 20% of activity loss for the large (51 nm) and medium (26 nm) size glucose oxidase-magnetic nanoparticle (GOxMNP) bioconjugate and about 96% activity loss for the smallest GOx-MNP bioconjugate (5 nm) after ten cycles. The bioconjugates demonstrated equivalent total product conversions as a single reaction of an equivalent amount of the native enzyme after the 5th cycle for the 26 nm nanoparticles and the 7th cycle for the 51 nm nanoparticles. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Enzymes are versatile proteins with great potential for applications in research and industry due to their myriad biocatalytic transformations with chemo-, regio-, and stereospecificity in mild process conditions. Evolving in cellular systems, enzymes are not well adapted to continuous, high throughput, single-product industrial processes. Enzymes lack long-term stability, prove difficult to separate and reuse, and usually require extensive downstream processing [1]. Research into how to adapt enzymes for more industrial uses has led to the development of various carrier-bound immobilized enzymes, which facilitate the use of enzymes in continuous processes and help overcome the cost constraints by aiding in efficient separation and recycling of costly enzymes [2]. Through the enzyme immobilization process, enzymes can be widely dispersed in reaction solutions without aggregation and enzyme stability can
∗ Corresponding author at: Department of Chemical and Petroleum Engineering, University of Wyoming, 1000 E. University Avenue, Dept. 3295, Laramie, WY 82071, USA. Tel.: +1 307 766 6524; fax: +1 307 766 6777. E-mail address: [email protected] (P.A. Johnson). 0927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2010.11.006
be enhanced with multivalent covalent attachment [3–5]. Among possible carriers, magnetic nanoparticles (Fe3 O4 ) show promise due to their simple, non-chemical separation method [5–11] and despite the non-porous structure, they permit high quantities of enzyme loading due to the large surface area [12]. In the last decade, there has been growing interest in magnetic nanoparticle bioconjugates for use in applications from drug delivery, hyperthermia treatment, and cell separation, to biosensors and enzymatic assays [13,14], biocatalysts [15] and environmental remediation [16]. Magnetic nanoparticles have been used as support materials for immobilization of enzymes such as yeast alcohol dehydrogenase [6] and lipase [7], with various surface modifications. Various attempts have been made to immobilize glucose oxidase on different types of solid supports using glutaraldehyde as a cross-linking agent for biosensor and biofuel cell applications [17,18]. In our experiments, glucose oxidase (GOx, EC 1.1.3.4) was immobilized on three sizes of magnetic particles. Glucose oxidase is a homodimer flavoprotein containing two active sites per molecule [19,20], catalyzing the oxidation of -d-glucose to gluconic acid concomitant with the reduction of oxygen to hydrogen peroxide [8]. Glucose oxidase has been used as a model enzyme to test
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203
various enzyme immobilization techniques and also shows great potential for biosensor and biofuel cell applications [5,21,22]. In this work, three sizes of magnetic nanoparticles were fabricated (5 nm, 26 nm and 51 nm) using two different methods, co-precipitation and ferrous hydroxide oxidation. The magnetic nanoparticles were modified with 3-(aminopropyl)triethoxysilane (APTES) and glutaraldehyde to covalently immobilize enzymes to the surface. Magnetic nanoparticles were characterized using Transmission Electron microscopy (TEM), the Brunauer, Emmett and Teller (BET) method and physical properties measurement system (PPMS). Each layer of surface modification was confirmed with X-ray photoelectron spectroscopy (XPS). Long-term enzyme and recycling stability were also examined.
199
2.4. Glucose oxidase immobilization
2. Experimental
10 mg of magnetic nanoparticles modified with APTES were dispersed by sonication in 8 mL of 8% glutaraldehyde solution (pH 7.4) and vortexed for 1 h. The magnetic nanoparticles were separated by magnetic decantation and washed 3 times with 10 mL of PBS (pH 7.4, 137 mM NaCl, 2.7 mM KCl, 8 mM Na2 HPO4 , and 2 mM KH2 PO4 ). The magnetic nanoparticles were redispersed in 2 mL of PBS. 10 mL of glucose oxidase solution (1 mg/mL in PBS) was added to the magnetic nanoparticles and gently vortexed overnight. The GOx-MNP bioconjugate was separated by magnetic decantation, washed 3 times with 10 mL of PBS, dispersed in 10 mL PBS and stored at 4 ◦ C. The supernatant from the Gox-MNP bioconjugate and last wash solutions were kept in order to determine enzyme loading efficiency, using BCA assay kit.
2.1. Materials
2.5. Determination of glucose oxidase activity
Glucose oxidase type II-S solid from Aspergillus niger (-d-glucose: oxygen 1-oxidoreductase, EC 1.1.3.4), 3(aminopropyl)triethoxysilane (APTES) (99%), iron (II) chloride (99%), iron (III) chloride (99%), and glutaraldehyde (8% aqueous) were purchased from Sigma–Aldrich (St. Louis, MO). Hydrogen peroxide (30%) and ammonium hydroxide (29.5%) were purchased from Fisher Scientific (Pittsburgh, PA). Anhydrous methanol (ACS grade) was purchased from EMD Chemicals (Gibbstown, NJ). Phosphate-buffered saline (PBS) solution (pH 7.4, 10×) was purchased from invitrogen (Carlsbad, CA). A glucose oxidase assay kit was purchased from Megazyme (Wicklow, Ireland). A BCA assay kit was purchased from Thermo Scientific (Rockford, IL). Neodymium magnets (DY0X0-N52; 14,800 Gauss) were purchased from K&J Magnetics, Inc. Deionized water was used in the preparation of aqueous solutions.
Glucose oxidase activity was measured using a glucose assay kit purchased from Megazyme. Glucose oxidase catalyzes the oxidation of -d-glucose to d-glucono-␦-lactone, which then hydrolyzes to gluconic acid and hydrogen peroxide. The resultant hydrogen peroxide then reacts with p-hydroxybenzoic acid and 4-aminoantipyrine in the presence of peroxidase to form the quinoneimine dye complex which has a strong absorbance at 510 nm. The GOx-MNP bioconjugate suspension was sonicated and 0.5 mL of suspension was added to chemical mixture according to glucose assay kit preparation procedures.
2.2. Magnetic nanoparticle fabrication Three sizes of magnetic nanoparticles were prepared by two different precipitation methods. Method A is co-precipitation in alkaline media. A solution of Fe2+ /Fe3+ with a molar ratio of 0.5 (0.0125 M/0.025 M) in 51 mL of deionized water was stirred and ammonium hydroxide (29.5%) added at a rate of 0.2 mL/min to adjust the pH to 10 (approximately 2.8–2.9 mL). The precipitation process was carried out at room temperature. Method B is the oxidative alkaline hydrolysis of ferrous ions. 0.05 M of FeCl2 was dissolved in 51 ml of deionized water. An aqueous suspension of Fe(OH)2 was prepared by adding 1 M KOH, adjusting to pH 7.9–8 with stirring. Precipitation proceeded for 2 h at room temperature. Method C is the oxidative alkaline hydrolysis of ferrous ions at high temperature. 0.05 M of FeCl2 was dissolved in 51 ml of deionized water while heated to 90 ◦ C. The aqueous suspension of Fe(OH)2 was prepared by the same procedure as Method B. Precipitation was carried out for 2 h at 90 ◦ C with stirring. The black precipitates from all three synthetic methods were strongly attracted by a permanent magnet and separated by magnetic decantation. The MNPs were washed 3 times with 51 mL of deionized water, 2 times with 51 mL of ethanol, and dried at room temperature. 2.3. Surface modification of magnetic nanoparticles with APTES 10 mg of magnetic nanoparticles were dispersed in 10 mL of 10% APTES solution (anhydrous methanol base) using sonication. The solution was vortexed overnight at room temperature then separated by magnetic decantation. MNPs were washed 3 times with 10 mL of anhydrous methanol and stored at 4 ◦ C for the enzyme binding process.
2.6. Recycling stability test Recycling stability of the GOx-MNP bioconjugate was determined by measuring the activity of the three GOx-MNP bioconjugates 10 consecutive times. -d-Glucose, p-hydroxybenzoic acid, 4-aminoantipyrine and peroxidase were mixed in one cuvette and added to 0.5 mL of GOx-MNP bioconjugate. After 17 min the GOxMNP bioconjugate was collected at the bottom of the cuvette for 3 min. Then dye solution was removed by pipette, transferred to a cuvette and measured at 510 nm. After each activity measurement, the GOx-MNP bioconjugate was washed 3 times with PBS (pH 7.4) and collected using magnetic decantation. The washing solutions from each of the 10 consecutive cycles were collected in a beaker to avoid losing GOx-MNP bioconjugate. After the last cycle, the activity of the particles collected in the beaker was measured to calculate the true activity loss from recycling as compared to enzyme deactivation. 2.7. Characterization The size and morphology of magnetic nanoparticles were determined by Transmission Electron Microscopy (TEM, Hitachi H-7000) at 75 kV. The samples were prepared on a formvar support film with a copper grid (300 mesh, Electron Microscopy Science, USA) by dropping magnetic nanoparticles in ethanol solution on the support and then evaporating in air at room temperature. The mean diameter of each sample was also determined by the Brunauer, Emmett, and Teller (BET) theory using a BET instrument. The binding of APTES and glucose oxidase on magnetic nanoparticles was confirmed by surface analysis with X-ray Photoelectron Spectroscopy (XPS, 5800 series, Multi-Technique ESCA systems). A sample from each stage of bioconjugate synthesis method B (magnetic nanoparticle, magnetic nanoparticle w/APTES, magnetic nanoparticle w/APTES and glutaldehyde, Gox-MNP bioconjugate) was immobilized on double-sided tape before putting in the specimen chamber under ultra high vacuum. Survey and high resolution results were collected and analyzed by XPSPeak 4.1 software. The
200
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203
Fig. 1. Transmission electron microscopy (TEM) images of Fe3 O4 nanoparticles. (A) Co-precipitation method, 5 nm; (B) oxidizing ferrous hydroxide at 25 ◦ C, 26 nm; (C) oxidizing ferrous hydroxide at 90 ◦ C, 50 nm. Table 1 Physical properties of magnetic nanoparticles for each synthetic method (AS , surface area; DBET , particle size based on BET; DTEM , particle size based on TEM; MS , saturation magnetization). Method
As (m2 g−1 )
DBET (nm)
DTEM (nm)
MS (emu g−1 )
A B C
218 44 23
5 26 51
5 24 48
48.5 72.8 84.5
of 72.8 emu g−1 , which is lower than that of bulk magnetite (92 emu g−1 ) [24]. The largest magnetic nanoparticles, 51 nm, were obtained from Method C. The largest magnetic nanoparticles showed the strongest magnetization among samples (84.5 emu g− 1), which is close to the magnetization value of bulk magnetite. Diameter measurements from TEM were consistent with those from BET measurements. The particle size and magnetization results are summarized in Table 1.
magnetic properties were measured by magnetometry in the Physical Property Measurement System (PPMS) from Quantum Design. 3. Results and discussion 3.1. Particle size and physical properties of magnetic nanoparticles Characteristic TEM images of magnetic nanoparticles from the three different methods are shown in Fig. 1. Observed sizes of the magnetic nanoparticles were consistent with previously reported research [8,21,23,24]. The smallest particles, 5 nm diameter, were obtained from ferrous/ferric co-precipitation method at room temperature (Method A) and showed the lowest magnetization (48.5 emu g−1 ) among the samples. Particles with a diameter of 26 nm were synthesized by Method B and showed a magnetization
Fig. 2. XPS survey scan of bare magnetic nanoparticles (A) and after APTES modification on the surface of magnetic nanoparticels (B). (Magnetic nanoparticles were synthesized using method B.)
Fig. 3. High resolution XPS spectra of (A) C 1s and (B) O 1s for each step of bioconjugate synthesis using method B (MNP, MNP w/APTES, MNP-APTES w/Glutaraldehyde and Gox-MNP bioconjugate).
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203 Table 2 Chemical composition (atom %) of MNPs based on high resolution XPS analysis. (Magnetic nanoparticles were synthesized using method B.).
O1s Fe2p C 1s N 1s Si 2p a
Bare MNP
MNP-APTES
MNP-APTESglutaraldehyde
GOx-MNP conjugate
45.5 32 22.5a N/A N/A
39.6 21.6 28.4 1.6 8.8
42.4 25.3 23 2.2 7.1
36.5 16.8 34.3 4.8 7.6
Impurity.
Size distributions of the three types of magnetic nanoparticles varied depending on the synthetic method. Particles from Method A showed the smallest size distribution, with particle sizes ranging from 3 nm to 9 nm. Method B had a slightly larger distribution with particle sizes range from 13 nm to 36 nm. Particles from Method C had the largest size range, from 8 nm to 133 nm.
201
the presence of cross linker derived species C O (288.5 eV). After glucose oxidase modification, elevated C–N and C N (285.4 eV) and C O and C–O (287.7 eV) peaks were observed. The peaks were significantly higher than the glutaraldehyde modified magnetic nanoparticles, characteristic of peptide bonds and supporting the presence of glucose oxidase immobilized on the nanoparticle surface. Fig. 3(B) shows the high resolution peak shift of O 1s throughout the bioconjugate fabrication process. The largest peak contribution for bare magnetic nanoparticles was from oxygen in the crystal structure with binding energy of 529.6 eV, which is typical of oxide crystal structures [26]. A hydroxyl functional group on the surface of magnetic nanoparticles was indicated at 530.9 eV. After APTES modification, the hydroxyl peak shifted to 531.6 eV due to covalent bonding with silicon at the surface (Fig. 3(B)). Glucose oxidase binding to the magnetic nanoparticle caused significant peak shifts due to the OH, O–C, O–H, O–N and O–C–N functional groups in protein structure.
3.2. Binding confirmation and surface analysis
3.3. Binding efficiency, activity measurement
XPS survey spectra and high resolution spectra of the Fe2p1, Fe2p3, C 1s, O 1s, N 1s and Si 2p orbitals were obtained for each step of the fabrication processes (magnetic nanoparticles, magnetic nanoparticles w/APTES, magnetic nanoparticles w/APTES and glutaraldehyde, and GOx-MNP bioconjugate). Table 2 gives an overview of the high resolution spectra atom % data for each step of the fabrication process. Advantageous carbon or rubbish carbon was observed on the bare magnetic nanoparticles, an impurity remaining from the fabrication processes. Shown in Fig. 2, the survey scan confirms the presence of nitrogen and silicon after APTES modification. The binding of APTES was also supported by the presence of an amine functional group confirmed by the presence of a C–N peak for C 1s in the high resolution spectra at 285.2 eV (Fig. 3 and Table 3) [25]. Glutaraldehyde modification was confirmed by
After the immobilization of the enzyme to the MNPs, the unbound protein in the supernatant was assayed using the BCA assay kit. All three types of MNP showed between 7.3% and 7.6% binding efficiency, despite the relatively large differences in surface area (Table 1). The enzymatic activity of GOx-MNP bioconjugate suspensions was measured using the glucose oxidase kit from Megazyme. Activity measurements were taken for three months after preparation and no significant decrease in enzymatic activity was noticed (results not shown). Enzyme activity after immobilization on MNPs was measured and compared to the native enzyme to determine the percent of activity retained. The 5 nm MNPs demonstrated the highest activity retention (23%), while the largest size MNPs (51 nm) showed the least, at 15% of the native enzyme activity.
Table 3 XPS peak-fitted parameters of Fe2p1, Fe2p3, C 1s, O 1s, N 1s and Si 2p on surface of magnetic nanoparticle. (Magnetic nanoparticles were synthesized using method B.).
MNP (Bare)
MNP w/APTES
MNP-APTES w/glutaraldehyde
MNP-APTES-Glu w/GOx
Element
Binding energy (eV)
FWHM (eV)
Fe2p1 Fe2p3 C 1s O 1s Fe2p1 Fe2p3 C 1s O 1s N 1s Si 2p Fe2p1 Fe2p3 C 1s
723.87 711.16 284.39 529.60 530.90 723.86 710.91 283.95 285.21 529.77 531.61 399.48 101.30 723.31 710.66 284.07 285.84 288.51 529.51 531.34 399.47 100.80 723.98 710.87 284.29 285.44 287.71 529.55 531.01 399.73 101.42
11.68 4.37 2.18 1.85 2.44 8.19 4.11 1.85 1.71 2.16 2.09 2.70 2.32 8.45 4.26 2.20 2.03 2.96 2.09 1.76 3.65 2.69 7.21 3.87 1.94 2.28 3.03 1.73 2.82 2.62 2.08
O 1s N 1s Si 2p Fe2p1 Fe2p3 C 1s O 1s N 1s Si 2p
Area 15,797.16 15,137.04 2692.96 7603.96 6691.90 13,770.37 11,430.19 3592.41 897.15 10,444.57 5382.42 563.98 1458.03 11,128.58 7965.67 1701.25 535.15 250.04 8206.39 2350.07 497.72 665.96 5344.11 5451.44 2050.11 1462.45 335.43 2496.07 6653.37 911.35 828.93
Chemical bond Iron oxide Iron oxide C–C Iron oxide O–H Iron oxide Iron oxide C–C C–N Iron oxide O–Si N–C Si–O Iron oxide Iron oxide C–C C N C O Iron oxide O C/O–Si N–C/N C Si–O/Si–C Iron oxide Iron oxide C–C C N C O/C–O Iron oxide O–H/O–C/O C/O–Si N–C/N C Si–O
C/Fe
0.09
0.18
0.13
0.36
202
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203
Fig. 5. Specific activity comparison (A) Specific activity of bioconjugates with 5 nm, 26 nm and 51 nm MNP in every cycle and specific activity of native GOx. (B) Total product conversion of bioconjugates and native GOx.
Fig. 4. Enzymatic activity of glucose oxidase–magnetic nanoparticle bioconjugate for 10 consecutive cycles. (A) Bioconjugate with 50 nm MNP; (B) bioconjugate with 26 nm MNP; (C) bioconjugate with 5 nm MNP. (Standard error (SE) is indicated as error bars.)
3.4. Recycling stability The possibility of recycling the GOx-MNP bioconjugate for successive activity measurements was investigated. The procedure was conducted 10 consecutive times for each of the three difference sizes of magnetic nanoparticles (Fig. 4). To account for particle loss from cycling, the supernatant from each cycle rinse was col-
lected in a beaker and the GOx-MNP bioconjugate was separated by prolonged exposure to the magnet. The activity for each cycle, the activity from the supernatant residues and the activity loss during the ten consecutive cycling tests is shown in Fig. 4. The largest sized GOx-MNP bioconjugate (51 nm) retained 57% of original activity, and counting the activity from the supernatant residues, retained a total activity around 80%. The medium sized GOx-MNP bioconjugate (26 nm) retained the highest activity after 10 cycles with 72%. With 9% activity from residues, the total retained activity was 81% of the overall activity. The smallest GOx-MNP bioconjugate (5 nm) showed a dramatic decrease in activity. Only 4% of original activity was retained by the 10th cycle, with the supernatant residues exhibiting the highest activity of all three bioconjugates at 20%, and an overall activity loss of 75%. The losses for the smallest bioconjugate can be explained by the difference of magnetization with particle size. The smallest bioconjugate particle had the lowest magnetization (48.5 emu g−1 ) and 3 min between the cycling tests might not be sufficient time to allow all the particles in solution to be acted on by the magnet, and settle to the bottom of cuvette, so loss of GOx-MNP bioconjugate during each cycle resulted in a large decrease in activity. A comparison of the 51 nm and 26 nm bioconjugates from Fig. 4 shows the 51 nm bioconjugate lost almost 25% of its original activity at fourth cycling, compared to 12% for the 26 nm bioconjugate. A 25% decrease in activity is not observed for the 26 nm bioconjugate until the 10th cycle. Also, the decrease in activity of the 26 nm bioconjugate was relatively steady throughout all 10 cycles. The 51 nm
H.J. Park et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 198–203
bioconjugate was synthesized using particle fabrication Method C and had a wide size distribution, from 8 nm to 133 nm. Particles less than 10 nm likely were lost throughout recycling process, as was seen with the 5 nm bioconjugate. By contrast, the 26 nm bioconjugate showed a relatively narrow particle distribution (13–36 nm) and a relatively small decrease in activity. 3.5. Specific activity of bioconjugates and total product conversion Using binding efficiency and activity measurements, the specific activity in each cycle of the three types of bioconjugate was calculated (Fig. 5(A)). The highest initial specific activity was found in 5 nm bioconjugates (5770 U/g protein) with dramatic decreases evident over the 10 cycles. The 26 nm bioconjugate showed an initial specific activity of 5070 U/g protein and the lowest specific activity (3810 U/g protein) was shown for the 51 nm bioconjugate. A comparison of the total product conversion (Fig. 5(B)) shows the quantity of product produced by the enzymatic reaction. Native GOx showed very high product conversion (507 mmol/g over 20 min). For the GOx-MNP bioconjugate, productivity by the 5th cycle for the 26 nm bioconjugate and the 7th cycle for the 51 nm bioconjugate was equivalent to a single cycle of the native GOx. Additional productivity above the native GOx was achieved by subsequent recycling for the two GOx-MNP bioconjugates. The 5 nm bioconjugate was not able to attain the equivalent of the native enzyme product conversion until the 10th recycling test. 4. Conclusion Three different sizes of magnetic nanoparticles (5 nm, 26 nm, and 51 nm) were fabricated by coprecipitation and oxidation of Fe(OH)2 , and characterized by BET and TEM to analyze the diameter and morphology of the particles. Magnetization of the particles was measured by PPMS. Surface modification of magnetic nanoparticles was carried out with APTES and the enzyme glucose oxidase was covalently immobilized on the magnetic nanoparticles with glutaraldehyde. The enzymes retained their activity after immobilization and stability was tested by recycling GOx-MNP bioconjugate for 10 consecutive cycles. Successful surface modification was confirmed by XPS studies and results showed strong evidence of APTES, glutaraldehyde (C O 288.5 eV) and protein modification (C–N 285.4 eV and C O 287.7 eV) on the magnetic nanoparticle surface. 0.54–0.56 mg of GOx enzyme was immobilized on the three different types of MNPs with 15–22.7% of the original enzyme activity retained. From the recycling stability studies, it was observed that the 26 nm size MNPs showed the highest activity retention after 10 cycles and the least amount of particle loss between cycles, while the 5 nm MNPs had the highest initial activity but were the most difficult to recover. Increasing the size of the magnetic core to 51 nm did not further improve the recoverability over the 26 nm MNPs. Evidence of the benefit of using the recyclable biocatalysts was realized after the 5th cycle for the 26 nm bioconjugates and the 7th cycle for the 51 nm bioconjugates. This
203
work shows the potential of recyclable biocatalysts for practical applications in chemical reaction processes and can be a model for the use of a wider variety of enzymes. Acknowledgements This research was supported by funding from the North Central Regional Sun Grant Center at South Dakota State University through a grant provided by the US Department of Transportation, Office of the Secretary, Grant No. DTOS59-07-G-00054. Additional funding was provided by the School of Energy Resources (SER) at the University of Wyoming. The authors would also like to thank the help of Dr. Jinke Tang for the assistance with the Physical Property Measurement System. References [1] U.T. Bornscheuer, Angewandte Chemie International Edition 42 (2003) 3336–3337. [2] R.A. Sheldon, R. Schoevaart, L.M. Van Langen, Biocatalysis & Biotransformation 23 (2005) 141–147. [3] M.Y. Arica, Journal of Applied Polymer Science 77 (2000) 2000–2008. [4] C. Mateo, O. Abian, R. Fernandez-Lafuente, J.M. Guisan, Enzyme and Microbial Technology 26 (2000) 509–515. [5] J. Huang, R. Zhao, H. Wang, W. Zhao, L. Ding, Biotechnology Letters 32 (2010) 817–821. [6] M. Liao, D. Chen, Biotechnology Letters 23 (2001) 1723–1727. [7] S. Huang, M. Liao, D. Chen, Biotechnology Progress 19 (2003) 1095–1100. [8] G. Kouassi, J. Irudayaraj, G. McCarty, BioMagnetic Research and Technology 3 (2005) 1. ´ M. Antalík, M. Timko, C.N. Ramchand, D. Lobo, [9] M. Koneracká, P. Kopcansky, R.V. Mehta, R.V. Upadhyay, Journal of Magnetism and Magnetic Materials 201 (1999) 427–430. [10] Y. Liu, R. Hamby, R.D. Colberg, Powder Technology 64 (1991) 3–41. [11] T. Bahar, S.S. C¸elebi, Enzyme and Microbial Technology 26 (2000) 28–33. [12] L. Betancor, M. Fuentes, G. Dellamora-Ortiz, F. López-Gallego, A. Hidalgo, N. Alonso-Morales, C. Mateo, J.M. Guisán, R. Fernández-Lafuente, Journal of Molecular Catalysis B: Enzymatic 32 (2005) 97–101. [13] Q.A. Pankhurst, J. Connolly, S.K. Jones, J. Dobson, Journal of Physics D: Applied Physics 36 (2003) R167–R181. [14] P. Tartaj, M.D.P. Morales, S. Veintemillas-Verdaguer, T. Gonzalez-Carreno, C.J. Serna, Journal of Physics D: Applied Physics 36 (2003) R182–R197. [15] L. Zeng, K. Luo, Y. Gong, Journal of Molecular Catalysis B: Enzymatic 38 (2006) 24–30. [16] D. Horák, M. Babic, H. Macková, M.J. Benes, Journal of Separation Science 30 (2007) 1751–1772. [17] S. Libertino, A. Scandurra, V. Aiello, F. Giannazzo, F. Sinatra, M. Renis, M. Fichera, Applied Surface Science 253 (2007) 9116–9123. [18] V. Bulmus, H. Ayhan, E. Piskin, Chemical Engineering Journal 65 (1997) 71–76. [19] K.C. Gulla, M.D. Gouda, M.S. Thakur, N.G. Karanth, Biosensors and Bioelectronics 19 (2004) 621–625. ´ J. Zeremski, D. Periˇcin, V. Leskovac, Molecular and Cel[20] G. Wohlfahrt, S. Trivic, lular Biochemistry 260 (2004) 69–83. [21] L. Rossi, A. Quach, Z. Rosenzweig, Analytical and Bioanalytical Chemistry 380 (2004) 606–613. [22] J. Kim, H. Jia, P. Wang, Biotechnology Advances 24 (2006) 296–308. [23] J. Mürbe, A. Rechtenbach, J. Töpfer, Materials Chemistry and Physics 110 (2008) 426–433. [24] M. Tada, S. Hatanaka, H. Sanbonsugi, N. Matsushita, M. Abe, Journal of Applied Physics 93 (2003) 7566–7568. [25] S. Delpeux, F. Beguin, R. Benoit, R. Erre, N. Manolova, I. Rashkov, European Polymer Journal 34 (1998) 905–915. [26] J.T. Kloprogge, L.V. Duong, B.J. Wood, R.L. Frost, Journal of Colloid and Interface Science 296 (2006) 572–576.