Amplified Peroxidase-Like Activity in Iron Oxide Nanoparticles Using

Feb 24, 2017 - After centrifugation at 12 000 rpm for 10 min, the protein-free filtrates in the filtrate receiver were used as a blank. ..... Farwell ...
0 downloads 4 Views 2MB Size
Download PDF
Research Article www.acsami.org

Amplified Peroxidase-Like Activity in Iron Oxide Nanoparticles Using Adenosine Monophosphate: Application to Urinary Protein Sensing Ya-Chun Yang,† Yen-Ting Wang,† and Wei-Lung Tseng*,†,‡,§ †

Department of Chemistry, National Sun Yat-sen University, Taiwan School of Pharmacy, College of Pharmacy, Kaohsiung Medical University, Taiwan § Center for Nanoscience and Nanotechnology, National Sun Yat-sen University, Taiwan ‡

S Supporting Information *

ABSTRACT: Numerous compounds such as protein and double-stranded DNA have been shown to efficiently inhibit intrinsic peroxidase-mimic activity in Fe3O4 nanoparticles (NP) and other related nanomaterials. However, only a few studies have focused on finding new compounds for enhancing the catalytic activity of Fe3O4 NPrelated nanomaterials. Herein, phosphate containing adenosine analogs are reported to enhance the oxidation reaction of hydrogen peroxide (H2O2) and amplex ultrared (AU) for improving the peroxidase-like activity in Fe3O4 NPs. This enhancement is suggested to be a result of the binding of adenosine analogs to Fe2+/Fe3+ sites on the NP surface and from adenosine 5′-monophosphate (AMP) acting as the distal histidine residue of horseradish peroxidase for activating H2O2. Phosphate containing adenosine analogs revealed the following trend for the enhanced activity of Fe3O4 NPs: AMP > adenosine 5′-diphosphate > adenosine 5′-triphosphate. The peroxidaselike activity in the Fe3O4 NPs progressively increased with increasing AMP concentration and polyadenosine length. The Michaelis constant for AMP attached Fe3O4 NPs is 5.3-fold lower and the maximum velocity is 2.7-fold higher than those of the bare Fe3O4 NPs. Furthermore, on the basis of AMP promoted peroxidase mimicking activity in the Fe3O4 NPs and the adsorption of protein on the NP surface, a selective fluorescent turn-off system for the detection of urinary protein is developed. KEYWORDS: adenosine monophosphate, iron oxide nanoparticles, human serum albumin, peroxidase mimic, amplex ultrared

■

nanowires,20 and graphene quantum dots. 21 Catalase-, phosphatase-, superoxide dismutase-, esterase-, and ferroxidase-mimic activities were reported for bare Co3O4 NPs,22 CeO2-related hydrogels,23 NiO nanoflowers,24 histidine-containing peptide-coated gold nanoparticles,25 and Pt nanodotsdeposited gold nanorods,26 respectively. Additionally, several nanomaterials exhibit multiple enzyme-like activities, such as peroxidase- and catalase-like activities of γ-Fe2O3 NPs27 and peroxidase- and catalase-mimic activities of apo-ferritin-coated Pt NPs.28 Among these nanomaterials, Fe3O4 NP-related nanomaterials as peroxidase mimetics have been widely used in place of horseradish peroxidase (HRP) in enzyme-linked immunosorbent assay.29 Also, they were extensively implemented as a catalyst for colorimetric and fluorescent detection of H2O2 and H2O2 product-related enzyme systems14,30−36 based on the combination of H2O2 mediated oxidation of the chromogenic substrate and enzyme-catalyzed production of H2O2. Additionally, phosphate and catechol containing compounds could coordinate with Fe3+ ions on the surface of the Fe3O4 NPs,

INTRODUCTION Enzymes are one of the essential macromolecules in biological systems because of their unique advantages as biocatalysts with a high reaction selectivity, high reaction efficiency, and milder reaction conditions.1,2 They are being used in an increasing number of applications in a wide range of fields, such as biofuel production,3 enantioselective syntheses,4 macromolecular syntheses,5 and drug screening.6 However, the limitations of enzymes in widespread application include high cost, low stability, short-term storage, pH-sensitive activity, less adaptability to harsh environments, and difficulties in recovery and recycling. To respond to these drawbacks, numerous nanomaterials have been developed to mimic a series of natural enzymes such as oxidase, peroxidase, catalase, phosphatase, superoxide dismutase, esterase, and ferroxidase.7,8 Oxidase-like nanomaterials are exemplified by citrate-capped gold nanoparticles (NPs),9,10 poly(acrylic acid)-coated cerium oxide NPs,11 lysozyme-stabilized Pt nanoclusters,12 and graphene oxidegold nanocluster hybrids.13 Peroxidase mimetics include bare Fe3O4 NPs,14,15 Fe2O3/SiO2 nanoparticles,16 graphene oxide functionalized with hemin,17 graphene oxide coated with lysozyme-stabilized gold nanopclusters,13 aptamer-functionalized gold nanoparticles,18 CeO 2 nanoparticles,19 V 2O5 © 2017 American Chemical Society

Received: December 6, 2016 Accepted: February 24, 2017 Published: February 24, 2017 10069

DOI: 10.1021/acsami.6b15654 ACS Appl. Mater. Interfaces 2017, 9, 10069−10077

Research Article

ACS Applied Materials & Interfaces therefore inhibiting their catalytic activity.37 Thus, target analyte-induced shielding against peroxidase-like activity in the Fe3O4 NPs was well-suited for the sensitive and selective detection of catecholamines,37 phosphate ions,38 and doublestranded DNA.39 Electrostatic attraction between protein and surface capping ligands significantly suppressed the peroxidaselike activity of the Fe3O4 NPs.40 However, only Liu and coworker observed that the adsorption of polycytosine on the surface of the Fe3O4 NPs efficiently promoted their catalytic activity during H2O2-mediated oxidation of 3,3′,5,5′-tetramethylbenzidine (TMB).41 In the present study, we inferred that adenosine 5′-monophosphate (AMP) served as an activity enhancer to boost peroxidase-like activity in the Fe3O4 NPs, which accelerated H2O2-induced oxidation of amplex ultrared (AU). When AMP attaches onto the NP surface through coordination between its phosphate groups and Fe2+/Fe3+ sites, AMP mimics the distal histidine residue of HRP activating the H2O2 bounded to Fe2+/Fe3+ sites. As a result, the presence of AMP improved the catalytic activity of the Fe3O4 NPs. Similarly, a previous study reported that adenosine triphosphate (ATP) efficiently enhanced the activity in HRP-mimicking Gquadruplex-hemin DNAzyme.42,43 Furthermore, the AMPpromoted peroxidase-like activity in the Fe3O4 NPs was successfully applied for the identification of human serum albumin (HSA) in urine.

■

(Model ASAP 2020, Micromeritics, U.S.A.), and the pore-size distribution was calculated based on the Barrett−Joyner−Halenda method. Sample Preparation. The Fe3O4 NPs (∼10 mg/mL) were prepared following a previously described procedure33 and described in the Supporting Information for the present work. Prior to the catalytic reaction, the Fe3O4 NPs, AMP, and AU were prepared in a solution of Tris-HCl (10 mM, pH 7.0). Subsequently, we incubated the Fe3O4 NPs (10 mg/mL, 50 μL) with a concentration of AMP (6− 1000 μM, 50 μL) at ambient temperature for 1 min. After 1 min, AU (50 μM, 50 μL), H2O2 (25 mM, 50 μL), and Tris-HCl (1.67 mM, 300 μL; pH 7.0) were sequentially added; the resulting solution was placed in a water bath at 37 °C for 25 min. By applying an external magnetic field, the Fe3O4 NPs were removed from the solution. We transferred the obtained solution to a 1 mL cuvette and measured their fluorescence spectra at an excitation wavelength of 568 nm. To investigate the effect of nucleotide analogs on the catalytic activity of Fe3O4 NPs, AMP was replaced with other nucleotides. A series of AU concentrations (0.6−8 μM) were used as substrates at fixed concentrations of 1 mg/mL Fe3O4 NPs and 2.5 mM H2O2 to investigate the steady-state mechanism of the Fe3O4 NP-induced catalytic reaction. The zero order rate constants were estimated with linear plots of the maximal fluorescence as a function of time. Steady state kinetic assays were performed in 10 mM Tris-HCl (pH 7.0). The kinetic parameters were determined by the Michaelis−Menten equation. To determine the adsorption efficiency of Fe3O4 NPs, they were incubated with AMP at ambient temperature for 25 min. After collecting the Fe3O4 NPs by an external magnetic field, the obtained supernatant was quantified by UV−visible absorption spectroscopy. The absorption efficiency was determined through the following equation:

EXPERIMENTAL SECTION

Chemical. 30% H2O2 was purchased from Showa Chemical Industry Co., Ltd. (Tokyo, Japan) and used without further purification. The following chemicals were ordered from SigmaAldrich Co. (St. Louis, MO, U.S.A.): FeCl3·6H2O, FeCl2·4H2O, Na3PO4, Na2HPO4, NaH2PO4, sodium pyrophosphate tetrabasic decahydrate, tris(hydroxymethyl)aminomethane (Tris), hydrochloric acid, adenosine 3′,5′-cyclic monophosphate (cyclic-AMP), adenosine 5′-triphosphate (ATP), adenosine 5′-diphosphate (ADP), AMP, cytidine 5′-triphosphate (CTP), cytidine 5′-diphosphate (CDP), cytidine 5′-monophosphate (CMP), guanosine 5′-triphosphate (GTP), guanosine 5′-diphosphate (GDP), guanosine 5′-monophosphate (GMP), thymidine 5′-triphosphate (TTP), thymidine 5′diphosphate (TDP), thymidine 5′-monophosphate (TMP), adenosine, cytidine, guanosine, thymidine, benzyl alcohol, immunoglobulin (IgG; human serum), human serum albumin (HSA; human plasma), transferrin (human plasma), protamine (salmon sperm), lysozyme (chicken egg white), trypsin (bovine pancreas), hemoglobin (bovine blood), myoglobin (bovine heart), and 2,2′-azino-bis-3-ethylbenzthiazoline-6-sulfonic acid (ABTS). AU and aminophenyl fluorescein (APF) were obtained from Thermo Fisher Scientific (Waltham, MA, U.S.A.) Instrumentation. Absorption and fluorescence spectra were recorded on a double-beam UV−visible spectrophotometer (V-530, Jasco, Tokyo, Japan) and a fluorescence spectrophotometer (F-7000, Hitachi, Tokyo, Japan), respectively. The zeta potential was determined using a DelsaNano zeta potential and a submicrometer particle size analyzer (Beckman Coulter, Inc., Pasadena, CA). The capillary electrophoresis (CE) system comprised a commercial UV absorbance detector (Spectra System UV2000 HR, Thermo Separation Products), a high-voltage power supply (Bertan, Hicksville, NY), a 60 cm-long fused-silica capillary (Polymicro Technologies, Phoenix, AZ; 75-μm inner diameter, and 360 μm outer diameter). The reversed-phase high-performance liquid chromatography (HPLC) system included a Spectra Series P100isocratic pump (Thermo Separation Products, Waltham, MA, U.S.A.), a manual injector, a separation column (Shodex C18−4E, Showa Denko America, Inc. New York, U.S.A.; 4.6 mm ID × 250 mm; 5 μm), and a Spectra Series UV100 detector (Thermo Separation Products, Waltham, MA, U.S.A.). The measurement of specific surface area was conducted using a multipoint Brunauer−Emmett−Teller (BET) technique

adsorption efficiency (%) = (C0 − C)/C0 where C0 is the initial concentration of AMP, and C is the concentration of the supernatant. To study the effect of the type of adenosine analog on the adsorption efficiency, AMP was replaced by adenosine, ADP or ATP one at a time. Sensing of HSA. The Fe3O4 NPs (10 mg/mL, 50 μL) were incubated with HSA (0.4−120 μM, 250 μL) at ambient temperature for 20 min, followed by incubation with AMP (1 mM, 50 μL) at ambient temperature for 1 min. After 1 min, AU (50 μM, 50 μL) and H2O2 (25 mM, 50 μL) were sequentially added, and the resulting solution was incubated in a water-bath at 37 °C for 25 min. After the removal of the Fe3O4 NPs by a magnet, the fluorescence spectra of the supernatant was collected under an excitation of 568 nm. In this study, we replaced HSA with other proteins to test the selectivity of the proposed system. To test the proposed system’s practicability, samples of urine (50 μL) were collected from a 24-year-old healthy adult female. The collected samples were treated using the Nanosep Centrifugal Devices (Pall filtron, Ann Arbor, MI, U.S.A.) with a 3 kDa molecular weight cutoff. After centrifugation at 12 000 rpm for 10 min, the protein-free filtrates in the filtrate receiver were used as a blank. In addition, urine samples were spiked with a series of standard HSA (50 μL) concentrations. The mixture (100 μL) of HSA and the Fe3O4 NPs was kept at ambient temperature for 20 min. The protein-adsorbed Fe3O4 NPs were isolated from the urine samples using an external magnetic field. After the supernatant was discarded, the remaining solution was incubated with a solution containing AMP (1 mM, 50 μL) and Tris-HCl (10 mM, 50 μL; pH 7.0) at ambient temperature for 25 min. The procedure for including the addition of AU and H2O2, Fe3O4 NP-catalyzed oxidation of AU, the collection of the Fe3O4 NPs, and the measurement of fluorescence spectra was the same as the aforementioned method for HSA detection. Combur-10-Test M strips (Roche Diagnostics GmbH, Mannheim, Germany) were used for the semiquantitative measurement of urinary protein. 10070

DOI: 10.1021/acsami.6b15654 ACS Appl. Mater. Interfaces 2017, 9, 10069−10077

Research Article

ACS Applied Materials & Interfaces

Figure 1. AMP-promoted peroxidase-like activity of the Fe3O4 NPs in 10 mM Tris-HCl (pH 7.0). (A) Fluorescence spectra and photographs of the oxidized AU obtained from the use of (a) Fe3O4 NPs and (b) AMP-Fe3O4 NPs as catalysts. (B) Effect of the adenosine concentration on the AUH2O2 reaction system catalyzed by AMP-Fe3O4 NPs. (C) Double reciprocal plot of the activities of (a) Fe3O4 NPs and (b) AMP-Fe3O4 NPs at a fixed NP concentration. The velocity (V) obtained from the incubation of (a) Fe3O4 NPs and (b) AMP-Fe3O4 NPs with different concentrations of AU at pH 7.0 (D) Effect of the number of bases of polyadenosine concentration on the AU-H2O2 reaction system catalyzed by polyadenosineadsorbed Fe3O4 NPs. The Fe3O4 NPs were incubated with adenosine analogs for 1 min in 10 mM Tris-HCl (pH 7.0). The concentrations of NPs, AMP, AU, and H2O2 are 1 mg/mL, 0,1 mM, 5 μM, and 2.5 mM, respectively. The error bars represent the standard deviation based on three independent measurements.

■

RESULTS AND DISCUSSION Role of AMP in Fe3O4-Catalyzed Reaction of AU and H2O2. The Fe3O4 NPs were prepared by coprecipitation of Fe2+ and Fe3+ in an ammonia solution.44 Previous studies indicated that the Fe3O4 NPs possessed a mixed valence state between Fe2+ and Fe3+, and their particle size was around 13 nm.44 Because the average hydrodynamic diameter of the assynthesized Fe3O4 NPs was about 570 nm (Figure S1), we speculated that the Fe3O4 NPs were self-assembled into spherical aggregates in an aqueous solution.37 When AMP molecules attached onto the NP surface, the Fe3O4 NPs retained a similar average hydrodynamic diameter and diameter distribution (Figure S2). The effects of AMP and other nucleotides on the catalytic activity of the Fe3O4 NPs were investigated using H2O2-mediated oxidation of AU. Since this catalytic reaction has been reported to be optimal at pH 7.0, our experiments were performed in a solution of Tris-HCl (pH 7.0).33 Compared to the bare Fe3O4 NPs as peroxidase mimetics, the attachment of AMP onto the surface of the Fe3O4 NPs caused an approximately 3-fold increase in the fluorescence of the oxidized AU (Figure 1A). In addition, an AMP-induced improvement of the Fe3O4 NP activity was observed directly from the color intensity of the oxidized AU (inset in Figure 1A). Control experiments showed that AMP was incapable of reacting with AU and H2O2 in the absence of Fe3O4 NPs (Figure S3). The fluorescence intensity of the oxidized AU at 590 nm progressively increased with increasing

AMP concentration (Figure 1B), indicating that the activity of the Fe3O4 NPs can be tuned by controlling the AMP concentration. The error bars in Figure 1B show that the relative standard deviation of the signal was smaller than 9% at each concentration of AMP. At any concentration of AMP, the Fe3O4 NP-catalyzed reaction of AU and H2O2 was completed after 30 min. Figure 1C shows the catalytic activity of the Fe3O4 NPs and the AMP-attached Fe3O4 NPs (AMP-Fe3O4 NPs). H2O2-mediated AU oxidation in both types of NPs follows the Michaelis−Menten mechanism. The Michaelis constant (Km) and the maximal velocity (Vmax) of the nanomaterials were determined by the Michaelis−Menten equation. In contrast to Fe3O4 NPs (Km, 0.17 μM; Vmax, 1.55 μM/s), AMP-Fe3O4 NPs exhibit a relatively small Km (0.036 μM) and a relatively large Vmax (4.15 μM/s). These results reflect that the AMP-Fe3O4 NPs are more efficient in converting a substrate into the product than the Fe3O4 NPs. As indicated in Figure 1D, the activity of the Fe3O4 NPs was gradually amplified by increasing the number of polyadenosine bases. Obviously, more AMP molecules were present on the surface of the Fe3O4 NPs, which induces the activation of more H2O2 in situ. To investigate the effect of storage time on the AMP-induced improvement of the Fe3O4 NP activity, the same batch of the catalytic activity of the Fe3O4 NPs was monitored in the presence of AMP over four months. The peroxidase-like activity of the Fe3O4 NPs remained over four months and their activity was still greatly improved upon the addition of 0.1 mM AMP (Figure S4). This 10071

DOI: 10.1021/acsami.6b15654 ACS Appl. Mater. Interfaces 2017, 9, 10069−10077

Research Article

ACS Applied Materials & Interfaces

Figure 2. (A) Chemical structures of adenosine analogs. (B) Effect of adenosine analogs on the AU-H2O2 reaction system catalyzed by the Fe3O4 NPs mixed with (a) AMP, (b) ADP, (c) ATP, (d) cyclic-AMP, (e) adenosine, (f) inosine, and (g) deionized water (control). The concentration of each adenosine analog is 0.1 mM. (C) Fluorescence spectra of the fluorescent product obtained from the addition of (a) Fe3O4 NPs, (b) Fe3O4 NPs and H2O2, (c) AMP-Fe3O4 NPs and H2O2, (d) ADP-Fe3O4 NPs and H2O2, (e) ATP-Fe3O4 NPs and H2O2, and (f) Fe3O4 NPs, H2O2, and adenosine to APF. (D) Adsorption efficiency of ATP, ADP, AMP, and adenosine on the Fe3O4 NPs. The Fe3O4 NPs were incubated with adenosine analogs for 1 min in 10 mM Tris-HCl (pH 7.0). The concentrations of NPs, AMP, AU, and H2O2 are 1 mg/mL, 0,1 mM, 5 μM, and 2.5 mM, respectively. The error bars represent the standard deviation based on three independent measurements.

AMP by the Fe3O4 NPs could improve their peroxidase-like activity. To test this hypothesis, reversed-phase HPLC was used to detect AMP before and after the treatment of the Fe3O4 NPs (Figure S6). After collecting the Fe3O4 NPs using an external magnetic field, the analysis of the supernatant by reversedphase HPLC showed a significant decrease in the peak intensity of the AMP without the production of adenosine. This result indicates that AMP was affixed to the surface of the Fe3O4 NPs. Consequently, we excluded that the Fe3O4 NPs improved activity by AMP is attributable to the NP-induced hydrolysis of AMP. Finally, the effect of adenosine analogs on the Fe3O4 NP activity was used to elucidate the mechanism associated with the AMP-induced enhancement of the Fe3O4 NP activity. The chemical structures of adenosine analogs are shown in Figure 2A. By comparing the oxidized AU fluorescence intensity, adenosine analogs exhibited the following trend in the Fe3O4 NP activity improvement: AMP > ADP > ATP (Figure 2B). Additionally, the activity of the Fe3O4 NPs was almost unchanged in the presence of adenosine, cyclic AMP, and inosine. These results clearly reflect that the coordination between the phosphate group of the adenosine analog and the Fe3+/Fe2+ on the Fe3O4 NP surface was a determining factor in improving the Fe3O4 NP activity. The hydroxyl radical, APF, was further used to quantify the concentration of hydroxyl radicals under different catalytic conditions.47 Curves a and b in Figure 2C show that the reaction between the Fe3O4 NPs and H2O2 produced hydroxyl radicals, inducing the conversion of

result indicates that the improved activity of the Fe3O4 NPs with 0.1 mM AMP is insensitive to ambient conditions. Mechanism for AMP-Induced Improvement of the Activity of the Fe3O4 NPs. To clarify the aforementioned observation, we infer that the improved activity of the Fe3O4 NPs with AMP could be due to three possible mechanisms. First, a previous study demonstrated that the adsorption of DNA on the surface of the Fe3O4 NPs enhanced the electrostatic attraction with positively charged TMB increasing the activity of the Fe3O4 NPs..41 Thus, we examined whether the interaction between AMP-Fe3O4 NPs and AU is crucial for the AMP-induced improvement of the Fe3O4 NP activity. CE was used to determine the electrophoretic mobility of AU, which is highly related to the effective charges of AU in an aqueous solution.45 Figure S5 shows that the migration time of AU is similar to neutral benzyl alcohol, signifying that AU is a neutral organic substrate in 10 mM Tris-HCl (pH 7.0). Under the same buffer condition, the zeta potential (−24 mV) of AMP-Fe3O4 NPs was almost unchanged in the presence of AU, signifying that the hydrophobic interaction does not force the adsorption of neutral AU molecules on the surface of AMPFe3O4 NPs. On the basis of these results, we ruled out that the enhanced activity of AMP-Fe3O4 NPs was a result of the electrostatic interaction between them and the AU. Second, a previous study reported that nucleoside triphosphates greatly enhanced the nanoceria oxidase-like activity as a result of oxidative reaction coupling with the nucleoside triphosphate hydrolysis.46 Thus, we also explored whether the hydrolysis of 10072

DOI: 10.1021/acsami.6b15654 ACS Appl. Mater. Interfaces 2017, 9, 10069−10077

Research Article

ACS Applied Materials & Interfaces

Figure 3. Schematic representation of the mechanism for the AMP-induced enhancement of the peroxidase-like activity of the Fe3O4 NPs.

Figure 4. (A) Turn-off sensing of HSA based on the adsorption of HSA on the NPs, the repulsion force between HSA and AMP, and AMPpromoted activity of the Fe3O4 NPs. (B, C) The ΔIF value obtained from the AU-H2O2 reaction system catalyzed by (B) Fe3O4 NPs and (C) AMPFe3O4 NPs in the presence of 1 μM (a) IgG, (b) HSA, (c) transferrin, (d) protamine, (e) lysozyme, (f) trypsin, (g) hemoglobin, and (h) myoglobin. The Fe3O4 NPs were incubated with protein for 20 min in 10 mM Tris-HCl (pH 7.0), following by the addition of 0.1 mM adenosine. The concentrations of NPs, AMP, AU, and H2O2 are 1 mg/mL, 0,1 mM, 5 μM, and 2.5 mM, respectively. The error bars represent the standard deviation based on three independent measurements.

APF to the fluorescent product. With the exception of adenosine, the binding of the adenosine analog to the Fe3O4 NP surface efficiently improved the production of hydroxyl radicals from H2O2 (curves c−f in Figure 2C). The effect of the adenosine analog on the NP-mediated conversion of H2O2 to hydroxyl radicals decreased as follows: AMP > ADP > AMP. Thus, the attachment of AMP to the NP surface generates the greatest amount of hydroxyl radicals, leading to the fastest oxidation of AU. Figure 2D presents the Fe3O4 NP adsorption efficiency for adenosine analogs in 10 mM Tris-HCl (pH 7.0). The order of the adenosine analog adsorption by the Fe3O4 NPs is AMP > ADP > ATP > adenosine. Obviously, AMP is the most effective in amplifying the Fe3O4 NP activity, because the Fe3O4 NPs provide the most binding sites for AMP; more AMP molecules are bounded to the NP surface, allowing the activation of more H2O2 molecules. Additionally, the BET

specific surface area of the Fe3O4 NPs was determined to be 91.4 m2/g and their pore volume was 0.32 cm3/g (Figure S7). Slight changes in BET specific surface area and pore volume (80.9 m2/g, 0.29 cm3/g) were observed in the AMP-Fe3O4 NPs. Evidently, the specific surface area is not the dominant factor for the AMP-induced improvement of the Fe3O4 NP activity. From these results, we concluded that three processes happened during the AMP-Fe3O4 NP-catalyzed reaction of H2O2 and AU (Figure 3): (1) AMP molecules directly attached to the surface of the Fe3O4 NPs through the coordination of the AMP phosphate group and the Fe3+/Fe2+ sites on the NP surface; (2) AMP on the NP surface imitates the distal histidine residue of HRP, which is effective in activating H2O2 molecules bound to the surface at Fe3+/Fe2+ sites; and (3) the generated hydroxyl radicals trigger the oxidation of AU, generating the fluorescent products. 10073

DOI: 10.1021/acsami.6b15654 ACS Appl. Mater. Interfaces 2017, 9, 10069−10077

Research Article

ACS Applied Materials & Interfaces

Figure 5. (A, B) Quantification of HSA by the proposed system (A) Fluorescence spectra of the proposed system in the presence of increasing HSA concentration. The arrow indicates the signal changes as increases in analyte concentration: 0, 0.2, 0.6, 1, 2, 4, 6, 8, 10, 20, 40, and 60 μM. Quantification based on AMP-Fe3O4 NPs probe. (B) A plot of the ΔIF value of the proposed system versus the concentration of HSA. (C, D) Determination of urinary protein by the proposed system. (C) Fluorescence spectra obtained from the addition (a) protein-free urine, (b) untreated urine, (c−g) HSA-spiked urine sample to the proposed system. The spiked concentration of HSA: (c) 1, (d) 2, (e) 4, (f) 6, and (g) 8 μM. (B) A plot of the ΔIF1 value of the proposed system versus the spiked concentration of HSA. The other conditions are the same as in Figure 4. The error bars represent standard deviations based on three independent measurements.

Development of the Sensor for Quantification of HSA. Determining the level of HSA in biological fluids is of interest because it is implicated in many diseases, such as renal disease, diabetes, and liver disease.48 Thus, this study applied AMPFe3O4 NPs for the detection of HSA based on the precise control of Fe3O4 NP activity (Figure 4A). When HSA molecules are adsorbed onto the NP surface at a pH of 7.0, the Fe3O4 NP surface charges became negative. This is attributed to the fact that HSA exhibits negative charges at pH levels above its isoelectric point (pI, 5.3). Electrostatic repulsion between the HSA-modified Fe3O4 NPs and AMP forces AMP away from the NP surface. As a result, AMP molecules are unable to enhance the activity of the Fe3O4 NPs. To verify this hypothesis, we initially measured the Fe3O4 NP zeta potentials in the absence and presence of HSA. Upon the addition of HSA, the zeta potential of the Fe3O4 NPs varied from +5 to −11 mV. Evidently, HSA molecules are indeed adsorbed on the surface of the Fe3O4 NPs, thereby creating a negative surface charge. Next, this study tested the selectivity of the proposed system toward HSA. Figure 4B shows changes in fluorescence intensity (590 nm) difference (ΔIF = IF0 − IF) of the proposed system after separately adding bovine serum albumin (pI = 4.7), transferrin (pI = 5.5), protamine (pI = 12− 13), lysozyme (pI = 11), trypsin (pI = 10), hemoglobin (pI = 6.8), and myoglobin (pI = 6.8). IF0 and IF represent the fluorescence intensity of the proposed system at 590 nm before and after the addition of the protein. ΔIF is the difference between the fluorescence intensity (590 nm) from the proposed system in the absence and presence of the protein.

Only low-pI proteins (>6.0) caused a noteworthy increase in the ΔIF value, reflecting that high-pI proteins on the NP surface were incapable of repelling the AMP attachment. Thus, the proposed system is highly selective toward low-pI proteins. However, in the absence of AMP, the Fe3O4 NPs provided poor selectivity of HSA, indicating that the use of AMP as an activity enhancer is effective in improving the Fe3O4 NP selectivity. As the concentration of HSA increased at fixed concentrations of AMP, H2O2, and AU, the fluorescence of the oxidized AU progressively decreased (Figure 5A). The linear relationship (R2 = 0.9920) of the ΔIF value against the HSA concentration was from 0.2 to 10 μM (approximately 13−660 mg/L; Figure 5B). The relative standard deviation in the ΔIF value at three concentration levels (0.2, 1, and 10 μM) was less than 5%, signifying that this probe is reproducible in detecting HSA. The limit of detection (LOD) at a signal-to-noise ratio of 3 for HSA was estimated to be 70 nM, which is comparable to LODs measured from squaraine dye,49 3-amino-N-alkylcarbazole-based organic lumino-materials,50 viscosity sensitive fluorescent dye,51 Terbium(III) complex,52 and aggregationinduced emission-based fluorescent dye.53 By taking advantage of the proposed system’s selectivity and sensitivity, its feasibility in determining urinary protein was validated. Note that the abundant proteins in urine include HSA (