Article pubs.acs.org/Langmuir
Assembling Paramagnetic Ceruloplasmin at Electrode Surfaces Covered with Ferromagnetic Nanoparticles. Scanning Electrochemical Microscopy in the Presence of a Magnetic Field Edyta Matysiak,† Alexander J. R. Botz,‡ Jan Clausmeyer,‡ Barbara Wagner,§ Wolfgang Schuhmann,‡ Zbigniew Stojek,† and Anna M. Nowicka*,† †
Faculty of Chemistry, University of Warsaw, ul. Pasteura 1, PL-02-093 Warsaw, Poland Analytical Chemistry-Center for Electrochemical Sciences (CES), Ruhr-Universität Bochum, Universitätsstrasse 150, D-44780 Bochum, Germany § Biological and Chemical Research Centre, Faculty of Chemistry, University of Warsaw, ul. Zwirki i Wigury 101, PL-02-093 Warsaw, Poland ‡
ABSTRACT: Adsorption of ceruloplasmin (Cp) at a gold electrode modified with ferromagnetic iron nanoparticles encapsulated in carbon (Fe@C Nps) leads to a successful immobilization of the enzyme in its electroactive form. The proper placement of Cp at the electrode surface on top of the nanocapsules containing an iron core allowed a preorientation of the enzyme, hence allowing direct electron transfer between the electrode and the enzyme. Laser ablation coupled with inductively coupled plasma mass spectrometry indicated that Cp was predominantly located at the paramagnetic nanoparticles. Scanning electrochemical microscopy measurements in the sample-generation/tip-collection mode proved that Cp was ferrooxidative inactive if it was immobilized on the bare gold surface and reached the highest activity if it was adsorbed on Fe@C Nps in the presence of a magnetic field. metabolism.18 In oxidized human ceruloplasmin, the T1PR copper site is reduced and cannot be oxidized, at least in part because of its high reduction potential, which is not catalytically useful or appropriate for electrocatalysis.18 It is known that an efficient direct bioelectrocatalysis with a redox protein requires a good electrical contact between the protein and the electrode surface. The efficiency of bioelectrocatalysis strongly depends on the orientation of the enzyme with respect to the electrode surface. The electron transfer between the redox centers within a rather large threedimensional structure of a protein and an electrode is still a challenging task. The electroactive centers in redox proteins are deeply embedded in the protein shell, and as a consequence, the electron transfer rate is usually rather slow.19 Moreover, the direct adsorption of a large protein onto a metallic electrode surface may lead to its denaturation.20 This has to be considered when a method for enzyme immobilization on a specific electrode surface is selected. There are only a few papers reporting on direct electron transfer (DET) between Cp and an electrode surface.16,21−24 Because the immobilization of active enzymes on electrode surfaces is an important aspect of bioelectrochemistry and
1. INTRODUCTION Ceruloplasmin (Cp), like other multinuclear copper oxidases, exhibits ferroxidase activity and has an important relevant role in iron metabolism.1−3 The ferroxidase activity of plasma Cp can be considered as the most important antioxidant property, because after Fe(II) had been removed from the plasma, the production of reactive oxygen species (ROS) generated in the noncatalyzed autoxidation of Fe(II) is blocked.4 Cp plays an essential role in the mobilization of stored iron: it oxidizes toxic ferrous iron to its nontoxic ferric form5,6 and controls the rate of efflux of iron from cells.7 Also, as in the cell environment, under close-to-physiological conditions, Cp can efficiently catalyze the oxidation of Fe(II) to Fe(III).8−13 In addition to Fe(II), Cp can also oxidize various organic compounds, e.g., aminophenols and 5-hydroxyindoles.14 The biological and electrocatalytical activity of ceruloplasmin is based on the presence of six Cu sites in its structure. Three Cu centers are located in the T2/T3 cluster, and three others are present at the three T1-binding sites.15 The three Cu ions building the T1 sites are mononuclear centers called T1Remote, T1CysHis, and T1PR.16 The T2/T3 cluster is connected with the T1CysHis center via the highly efficient and reliable Cys-2His electron transfer (ET) pathway. This center is also able to accept electrons from other sources, e.g., Fe(II).16−18 The T1Remote center is not needed for the oxidase activity of Cp but might be important in increasing the overall efficiency of Fe © 2015 American Chemical Society
Received: March 30, 2015 Revised: June 26, 2015 Published: July 3, 2015 8176
DOI: 10.1021/acs.langmuir.5b01155 Langmuir 2015, 31, 8176−8183
Article
Langmuir
were performed using Ar as the carrier gas. An open ablation cell was used for this study with an effective volume of ∼4.5 cm3.34 2.3. Scanning Electrochemical Microscopy (SECM). Scanning electrochemical microscopy measurements were performed using a modified Sensolytics SECM setup and an analog bipotentiostat (IPS/ Jaissle PG-100). SECM measurements were performed in a fourelectrode setup using the ceruloplasmin-modified electrode (Au/Fe@ C Nps/Cp) as working electrode 1 (WE1) and the 10 μm Pt microelectrode SECM tip as WE2. A Pt wire was used as a counter electrode (CE), and a miniaturized double junction Ag/AgCl/3 M KCl/0.1 M KCl electrode was used as the reference electrode (RE). The double junction was used to prevent leakage of chloride ions to the electrolyte and the corresponding deactivation of the oxidase activity of ceruloplasmin. To optimize the impact of the magnetic field while the SECM approach curves were being recorded and to position the Cp-modified sample electrode in an upward-facing position, an electrochemical cell was specifically designed. At the top end of the sample electrode, a 2 cm Teflon tube was attached and placed inside a circular Fe14Nd2B magnet. The magnetic field intensity was ∼40 mT (400 Gs). SECM approach curves were performed above at least four different spots of the Cp-modified surface in a pH 7.0 solution containing 20 μM K4[Fe(CN)6], 100 mM PB, and 150 mM K2SO4. To generate [Fe(CN)6]3− at the Cp-modified sample electrode which is then detected at the SECM tip by its reduction into initial [Fe(CN)6]4−, an Esample of 250 mV and an Etip of 80 mV versus Ag/AgCl/3 M KCl/0.1 M KCl were applied. 2.4. Electrochemical Measurements. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were performed using an Autolab PGSTAT 12 potentiostat (Metrohm-Autolab). A three-electrode configuration was used in all measurements with a 2 mm diameter gold disk electrode (BAS) as the WE. All potentials are referenced against the Ag/AgCl/3 M KCl/0.1 M KCl RE, and a Pt wire was used as the CE. Before being used, the Au electrode was polished with 1.0 μm and later with 0.3 μm Al2O3 powders on a wet pad. After being polished, the electrode was rinsed with a direct stream of ultrapure water to remove the alumina. The electrode surface was inspected electrochemically by voltammetric cycling between −0.3 and 1.5 V with a scan rate of 50 mV s−1 in a 100 mM H2SO4 solution until a stable voltammogram, typical for a clean gold electrode, was obtained.35 The roughness factor of the electrodes was always below 1.2.
electroanalysis, there is a need for a reliable strategy allowing the characterization of the immobilized enzymes. In this respect, scanning electrochemical microscopy was found to be a powerful tool for the investigation of enzyme-modified electrodes.25−27 In contrast to electrochemical methods typically used for the evaluation of enzyme-modified electrodes using the underlying electrode, the local catalytic activity of the immobilized protein is examined from the solution side of the biointerface. When the potential at the microelectrode scanning electrochemical microscopy (SECM) tip is controlled, the products or the reactants of the enzymatic reaction are detected with high specificity. The sample-generation/tip-collection (SG/TC) mode of SECM is commonly used to investigate enzyme-modified surfaces28−31 because of its high sensitivity in combination with the intrinsic microelectrode properties such as low capacitance, high mass transport rate, and small parasitic currents. In this work, we demonstrate that a layer of ferromagnetic nanoparticles covering the gold electrode surface guarantees the immobilization of Cp in a preferred orientation to allow the electrocatalytic activity of the paramagnetic Cp. If simultaneously a magnet is localized beneath the modified electrode, the activity of Cp will increase as confirmed by a combination of SECM and laser ablation coupled with inductively coupled plasma mass spectrometry (LA ICPMS).
2. MATERIALS AND METHODS 2.1. Materials. All chemicals were of the highest purity available. KH2PO4, K2HPO4, K2SO4, K4[Fe(CN)6]·3H2O, and KF were obtained from Sigma-Aldrich. Human ceruloplasmin was also purchased from Sigma-Aldrich and used as received. The initial 150 μg mL−1 Cp solution was prepared in 10 mM phosphate buffer (PB) (pH 6.0). The concentration of the enzyme in the stock solution [100 mM PB with 150 mM K2SO4 (pH 7.0)] was determined as suggested in ref 32. The enzyme solution was stored at 4 °C until it was used. Microelectrodes were fabricated by sealing an Ø 10 μm Pt wire (Goodfellow) into borosilicate glass capillaries [length of 100 mm; Øouter = 1.5 mm; Øinner = 0.75 mm (Hilgenberg)] and connected using soldering tin. Prior to each measurement, the SECM tip was polished using 3 and 1 μm lapping film (3 M) and ultrasonicated in ultrapure water for at least 3 min. Phosphate buffer solution (PB; 100 mM, pH 7.0) was prepared by mixing solutions of KH2PO4 and K2HPO4 with subsequent addition of 150 mM K2SO4. All solutions were prepared using Milli-Q water with a conductivity of 0.056 μS cm−1. The ferromagnetic electrode modifier, carbon-encapsulated iron nanoparticles (Fe@C Nps), was synthesized according to a procedure described elsewhere.33 2.2. ICPMS Measurements with Laser Ablation. An inductively coupled plasma mass spectrometer (PerkinElmer NexION 300) equipped with a laser ablation system (LSX-200+, CETAC) was used. The laser ablation system combines a stable, environmentally sealed 213 nm UV laser (Nd:YAG, solid state) with a high sampling efficiency, a variable 1−20 Hz pulse repetition rate, and a maximal energy of 5.0 mJ/pulse. After careful optimization, the LA setup was operated at a constant 10 Hz repetition rate, an energy of 1.25 mJ/ pulse, and a spot size of 50 μm. The signal intensities were registered for seven selected isotopes, using peak hopping mode and variable dwell times, depending on the abundance and content of the particular element in the sample: 12C (10 ms), 23Na (5 ms), 31P (10 ms), 49Ti (5 ms), 57Fe (5 ms), 65Cu (10 ms), and 197Au (3 ms). All measurement cycles consisted of the signal intensities registered during the multiline ablation (n = 10) over the area selected on the surface of the sample. The blank was registered for 10 s between each line ablation and for 20 s before the start, as well as after the multiline ablation. For each selected isotope, the registered blank values were subtracted from the signals recorded during the ablation of the samples. All experiments
3. RESULTS AND DISCUSSION 3.1. Synthesis of Carbon-Coated Iron Nanoparticles (Fe@C Nps). Ferromagnetic carbon-encapsulated iron nanoparticles (Fe@C Nps) were used as a modifier of the electrode surface. Fe@C Nps were synthesized using a carbon arc route according to the methods described in the literature.36,37 The simultaneous vaporization of Fe and C followed by rapid subsequent cooling was employed in the synthesis. The process was conducted under an Ar−H2 atmosphere [50:50 (v/v)] at a pressure of 30.0 kPa. To isolate the carbon-coated particles from uncoated iron particles, refluxing of the mixture in boiled 3 M HCl was performed for 24 h. Next the products were washed in an excess of water and ethanol and finally dried in air at 70 °C. The treatment with HCl was conducted to dissolve all Fe particles not completely coated by protective carbon layers. The morphology of the particles is shown in the inset of Figure 1. The TEM image demonstrates that individual Fe particles are coated by a thin and continuous carbon layer (a few nanometers thick). The diameter of the carbon-coated iron nanoparticles was in the range of 5−65 nm; however, the major fraction of the nanoparticles (∼70%) was between 5 and 30 nm in diameter. The mean value of the diameter of Fe@C Nps was estimated to be ∼20 nm (see Figure 1). The used nanoparticles had the maximal achievable magnetic moment (121 emu g−1) 8177
DOI: 10.1021/acs.langmuir.5b01155 Langmuir 2015, 31, 8176−8183
Article
Langmuir
couple for a bare Au electrode modified with Fe@C Nps and a C Np layer. A simple Ershler−Randles equivalent circuit (see the inset in Figure 2)38,39 was employed to extract relevant parameters following procedures described previously.38,40 The influence of the modifier on the electrochemical parameters of the electrode should be visible in the changes in the charge transfer resistance, Rct. The data presented in Table 1 clearly show that the electrode modification did not affect the electrochemical properties; the Rct values were particularly stable, and the exponent parameter ϕdl only slightly decreased (by 6 and 18% for nanoparticles with an iron core and pure carbon, respectively) compared to that of the bare gold electrode. These ϕdl data (small deviation from the value for the bare gold) indicate that the electrode surface was rather evenly coated with ferromagnetic layers of Fe@C Nps.39 The solution resistance, Rs, was equal to ∼191 Ω for all measurements. The Warburg parameter was nearly constant, suggesting that electrode modification did not affect the transport of the electroactive species. 3.3. Immobilization of Cp on an Fe@C Nps Layer. Cp was adsorbed physically on the Fe@C Nps layers covering the gold surface by placing a 6 μL drop of Cp [1 μg mL−1 in 100 mM PB with 150 mM K2SO4 (pH 7.0)] on the Au surface. It should be mentioned that in human ceruloplasmin ∼45% of copper ions are paramagnetic. Correspondingly, there are several signals in corresponding EPR spectra of the purified human ceruloplasmin: the signal from Cu T2 with a g∥ of 2.25 and an A∥ of 18 mT and two dissimilar signals from Cu T1 centers with different values of the hyperfine splitting constant (7.2 and 9 mT) but similar g values (2.20 and 2.21).41,42 The layer of magnetic nanoparticles formed by adsorption was not perfectly continuous, while the particles were evenly distributed.43 Correspondingly, adsorption of ceruloplasmin may lead to deposition also on the bare metal. Hence, to definitively prove that the ferromagnetic electrode modifier in combination with a magnetic field leads to the precise location of Cp molecules on the electrode modifier LA ICPMS measurements were performed aiming at the visualization of the distribution of 57Fe and 65Cu at the electrode surface. The signal intensities were registered during multiline ablation, while the sample/electrode was moving with a scan rate of 50 μm s−1 (Figure 3). The obtained data clearly show that the adsorption of paramagnetic Cp molecules takes place predominantly on the ferromagnetic electrode modifier. Within each laser beam scan, the 57Fe and 65Cu signals are co-located. On the contrary, at a bare gold surface, the adsorption is rather poor, preventing a substantial amount of active immobilized Cp on the Au surface. Apparently, the adsorption of Cp on the Fe@C Nps-modified electrode surface does not lead to enzyme deactivation. Hence, we anticipated that Cp adsorbed in its native active state on Fe@C Nps is capable of catalyzing the oxidation of Fe2+ to Fe3+. 3.4. Voltammetry of Cp Adsorbed on an Fe@C Nps Layer. Representative cyclic voltammograms of Fe@C Npsmodified Au electrodes with adsorbed Cp in the presence and absence of an external magnetic field are shown in Figure 3. An oxidation process at ∼250 mV versus Ag/AgCl/3 M KCl/0.1 M KCl (this potential is equivalent to 460 mV vs NHE) is seen. On the contrary, in the case of Cp adsorbed on the surface of a bare Au disk electrode, no electron transfer with the enzyme was observed even in the presence of an external magnetic field (solid black line in Figure 4). Obviously, direct electron transfer
Figure 1. Histogram of the nanoparticle size distribution on the TEM image of an electrode surface covered with Fe@C Nps.
at a relatively low magnetic field (∼8 kOe). The remnant magnetization is only 6 emu g−1, and this value corresponds to
