Silver nanoparticles as a bifunctional probe for label-free and reagentless colorimetric hydrogen peroxide chemosensor and cholesterol biosensor

ob id n b 1st D c a r t i c l e i n f o Article history: tion in food considerably affects cholesterol concentration in the human body. Therefore, the proper determination of cholesterol and/or treatment st cholesterol bio- ol oxidase (ChOx), ocapture probe for tion:  3 one (1) analysed using peroxidase. Horseradish peroxidase (HRP) enzyme is typically used with [H2O2]. The analysed result reflects the cholesterol level of a sample, which is determined via electro- chemical [9,10,16e18] or colorimetric [1,6,19e21] measurement. However, the activity of HRP enzyme is significantly affected by various factors, such as temperature, pH and ionic strength. Therefore, various peroxidase-like nanomaterials have been used in the development of cholesterol biosensors, including graphene * Corresponding author. E-mail address: hoang.tranvinh@hust.edu.vn (H.V. Tran). Contents lists availab Journal of Science: Advance .e l Journal of Science: Advanced Materials and Devices 5 (2020) 385e391Peer review under responsibility of Vietnam National University, Hanoi.rosis, myocardial infarction, hypertension, cerebral thrombosis and lipid metabolism disorders [2e8]. Notably, cholesterol concentra- The hydrogen peroxide (H2O2) product from Eq. (1) can be1. Introduction Cholesterol (C27H46O) is an essential component of mammalian cell membranes that plays a role in establishing appropriate membrane permeability and fluidity. It is also involved in the im- mune system and the brain synapses and exhibits anticancer effect [1,2]. The total blood cholesterol level of a healthy person is approximately 5.2 mM L1 (200 mg dL1); however, higher cholesterol levels can lead to coronary heart diseases, arterioscle- levels in body fluids is crucial for early diagnosis of the aforementioned diseases [5,6,9e15]. Mo sensors have been constructed using cholester which is an enzyme specific to cholesterol, as a bi cholesterol detection through the following reac CholesterolþO2 þH2O / ChOx Cholest 4 en þ H2O2Received 29 December 2019 Received in revised form 10 June 2020 Accepted 15 June 2020 Available online 20 June 2020 Keywords: Silver nanoparticles (AgNPs) Oxidation-etching reaction Plasmonic Cholesterol detection Cholesterol oxidase (ChOx) Human blood Enzymehttps://doi.org/10.1016/j.jsamd.2020.06.001 2468-2179/© 2020 The Authors. Publishing services b ( b s t r a c t A colorimetric chemosensor for hydrogen peroxide (H2O2) and biosensor for cholesterol is developed in this work following a ‘label-free and reagentless’ concept. In this strategy, silver nanoparticles (AgNPs) are used as a bifunctional probe, namely, a chemical capture probe for H2O2 recognition and a signal probe for displaying detection results. The working principle of these colorimetric sensors is based on the redox reaction between AgNPs and H2O2. This reaction converts AgNPs into Agþ ions, which can induce a visible color change from yellow to pinkish or colorless depending on the amount of H2O2. Therefore, H2O2 concentration can be measured. The developed colorimetric H2O2 chemosensor exhibits a highly selective and sensitive property towards H2O2 with a limit of detection (LOD) of 3.5 mM. Subsequently, a colorimetric biosensor is developed for cholesterol detection by combining AgNPs and cholesterol oxi- dase (ChOx) enzyme. Here, ChOx oxidizes cholesterol into cholest-4-en-3-one and H2O2. Then, the detected as-generated H2O2 concentration reflects the cholesterol level in the sample. This cholesterol biosensor exhibits high selectivity and sensitivity for cholesterol detection with a LOD of 40 mM. The developed sensor is successfully applied to the detection of cholesterol levels in serum samples. Our results show that the AgNPebased colorimetric method is cost-effective, sensitive and selective. It has potential applications in H2O2 and cholesterol detection and in clinical and medical diagnoses. © 2020 The Authors. Publishing services by Elsevier B.V. on behalf of Vietnam National University, Hanoi. This is an open access article under the CC BY license ( of Electrical Engineering, Hanoi University of Science and Technology (HUST), 1st Dai Co Viet Road, Hanoi, Viet NamOriginal Article Silver nanoparticles as a bifunctional pr reagentless colorimetric hydrogen perox cholesterol biosensor Hoang V. Tran a, *, Tuan V. Nguyen a, Lan T.N. Nguye a School of Chemical Engineering, Hanoi University of Science and Technology (HUST), b Nguyen Hue High School, Tam Diep Town, Ninh Binh Province, Viet Nam journal homepage: wwwy Elsevier B.V. on behalf of Vietname for label-free and e chemosensor and , Hong S. Hoang c, Chinh D. Huynh a ai Co Viet Road, Hanoi, Viet Nam le at ScienceDirect d Materials and Devices sevier .com/locate/ jsamdNational University, Hanoi. This is an open access article under the CC BY license quantum dots [7,20], gold nanoparticles (AuNPs) [8,19,22], zinc oxide (ZnO) nanorods [10], cupric oxide nanoparticles (CuONPs) [6] and boron nitride (BN) nanosheet/copper monosulphide (CuS) nanocomposites [15]. In the literature, most artificial peroxidases are used to develop colorimetric H2O2 sensors and cholesterol biosensors by following the ‘labelling concept’ [1,6e8,10,15,19e22]. Here, we propose a novel ‘label-free and reagentless’ approach for the development of colorimetric H2O2 chemosensor and cholesterol biosensor by using silver nanoparticles (AgNPs) as a bifunctional probe. In this strategy, AgNPs are used to replace the HRP enzyme and its chromogenic substrate, simplifying the DA
Að%Þ ¼ A0  Ac A0 x100% (2) 2.3. Cholesterol detection A 10 mL cholesterol solution with different concentrations ranging from 0mM to 3 mM,180 mL of PBS (pH 7.4) and 10 mL of the ChOx solution was pipetted into a 1.5 mL Eppendorf tube. The ts f ol so 1 H.V. Tran et al. / Journal of Science: Advanced Materials and Devices 5 (2020) 385e391386detection procedure. The experimental results show that the pro- posed tests are highly selective and sensitive for H2O2 and cholesterol detection. 2. Materials and methods 2.1. Materials and reagents Citric acid monohydrate (C6H8O7.H2O, AR, >99 wt.%), urea (CH4N2O, AR, >99.5 wt.%), silver nitrate (AgNO3, AR, >99,9 wt.%), sodium hydroxide (NaOH, AR, >99 wt.%), hydrochloric acid (HCl) solution (37 v/v.%), H2O2 solution (30 v/v.%), phosphate-buffered saline (PBS) tablets, cholesterol (AR, >99 wt.%), sucrose, ascorbic acid, galactose, fructose and lactose were purchased from Sigma- eAldrich. A cholesterol solution was prepared by dissolving cholesterol into ethanol with 0.1 wt.% Triton X-100. ChOx enzyme from Streptomyces sp. (lyophilised powder, 20 units/mg protein) was purchased from SigmaeAldrich. A ChOx enzyme solution was prepared by dissolving ChOx into PBS solution at a concentration of 0.25 mg mL1. An AgNP solution was synthesised in our laboratory by following the steps in a previous report [23] at a concentration of 100mM. A AgNP bifunctional probe solution (10mM)was prepared by diluting 25 mL of AgNP stock solution into 250 mL of distilled water (the pH of the probe solution was adjusted to 7 by using 1 M acetic acid solution). 2.2. H2O2 detection Subsequently, 200 mL of H2O2 solutionwas pipetted into a 1.5mL Eppendorf tube containing 1 mL of AgNP solution. The mixture was vortexed and incubated in a water bath at 40 C for 45 min. Then, the Eppendorf tube was cooled to room temperature, and the so- lution was transferred to a cuvette for ultravioletevisible (UVeVis) measurement. The working conditions of the H2O2 sensor were optimised, including a contact time of 30 min, a pH of 7 and a re- action temperature at 30 C. The optical densities (ODs) at a wavelength of 425 nm (OD425 nm) before (A0) and after (Ac) H2O2 addition were used to quantify H2O2 concentration (C) in the sample. The percentage difference of the OD at 425 nm (DA/A, given in %) was selected as the signal for this H2O2 chemosensor by using the following equation: Table 1 Compositions of the ‘blank sample’, ‘standard sample’, ‘sample’ and ‘sample plus’ tes Sample name Compositions ChOx solutiona (mL) PBS solution (mL) Cholester Blank 10 190 e Standard 10 180 10 Sample 10 180 e Sample Plus 10 170 10 a A ChOx concentration of 0.25 mg mL1 was used. b A cholesterol solution was first prepared in ethanol solvent standing in Triton X-100 c Human plasma samples were diluted 5 by using 1 PBS solution.mixtures were stirred and subsequently incubated in a water bath at 50 C for 60 min. After cooling to room temperature, 1 mL of AgNP solution was injected into the Eppendorf tube, and the so- lution was incubated again at 30 C for 45 min. Finally, all the Eppendorf tubes were cooled to room temperature and the solu- tions in the tubes were transferred to cuvettes for UVeVis measurements. 2.4. Determination of cholesterol concentration in human serum samples Human sera were collected from blood samples in a local hos- pital (M.D. Sera) and then diluted fivefold with 1 PBS. The cholesterol concentrations in these diluted sera were determined using the proposed method, and a standard cholesterol solution was used as the control sample [24]. Four independent experiments were conducted for each serum sample, namely, ‘blank’, ‘standard’, ‘sample’ and ‘sample plus’, which are listed in Table 1. UVeVis measurements were performed as described previously, and the cholesterol concentration in each serum sample was calculated using the following equation [24]: ½Cholesterol ðmMÞ ¼ ΔA=Að%Þsample ΔA=Að%Þstandard x 5:2 (3) where: DA/A(%)sample and DA/A(%)standard refer to the percentage differences of the optical absorption density of the ‘sample’ and the ‘standard sample’, respectively. The number ‘5.2’ denotes a diluted factor. 3. Methods The morphologies and crystal structures of the NPs were char- acterised via transmission electron microscopy (TEM) using JEOL JEM-1010. The average particle size and distribution were obtained via dynamic light scattering (DLS) using a Malvern Zetasizer Nano ZS (UK). UVeVis spectra were collected using an Agilent 8453 UVeVis spectrophotometer system with a wavelength range of 200e1200 nm. The X-ray diffraction (XRD) patterns of AgNPs were obtained at room temperature by using a D8 Advance, Bruker ASX with CuKa radiation (l ¼ 0.15406 nm) within the range of 2q ¼ 10e70 and a scanning rate of 0.02 s1. or cholesterol determination in human blood samples. lutionb (mL) Diluted human plasma solutionc (mL) Total volume (mL) e 200 e 200 10 200 10 200. The concentration used for the experiments was 0.25 mg mL . Fig. 1. Working principle of the colorimetric (a) H2O2 chemosensor and (b) cholesterol biosensor using AgNPs as a bifunctional probe. Inset (i, ii and iii) TEM images of the AgNP solution (i) before and (ii) after incubation with 0.8 mM H2O2 and (iii) after incubation with a reacted mixture of 0.25 mg mL1 of ChOx and 1 mM of cholesterol. Fig. 2. (a, b, c) TEM images of AgNP solution (a) before and after incubation with (b) 0.8 mM H2O2 and (c) a mixture of 0.25 mg mL1 of ChOx þ 1 mM of cholesterol, respectively. (d to f) Lateral size distribution histograms of the AgNP bifunctional probe (d) before and after incubation with (e) 0.8 mM of H2O2 and (f) a reacted mixture of ChOx enzyme þ 1 mM of cholesterol. H.V. Tran et al. / Journal of Science: Advanced Materials and Devices 5 (2020) 385e391 387 cedH.V. Tran et al. / Journal of Science: Advan3884. Results and discussion 4.1. Working principle of label-free and reagentless colorimetric H2O2 chemosensor and cholesterol biosensor The working principle of the colorimetric H2O2 chemosensor is presented in Fig. 1a. Given that the standard reduction potential of the Agþ/Ag couple (ε0Agþ=Ag ¼ 0:8 VÞ is lower than that of the H2O2/ H2O couple (ε0H2O2=H2O ¼ 1:77 V) at pH 7, the AgNPs are oxidised by H2O2 in the following reaction: 2Ag0 þH2O2/2Agþ þ 2OH (4) The reaction in Eq. (4) converts AgNPs (in Ag0 state) into Agþ ions. Therefore, AgNPs are dissolved into Agþ ions by H2O2, exhibiting the first role of AgNPs, namely, as a chemical capture probe for recognising H2O2 in a solution. Correspondingly, the conversion of AgNPs into Agþ ions can induce a visible color change from yellow to pinkish or colorless, depending on the amount of added [H2O2]. Therefore, [H2O2] can be quantitatively measured [25]. The high equilibrium constant of the reaction in Eq. (4) (Kp ¼ 7.59  103) [23] is responsible for the rapid fading of the color of the AgNP solution, which can be directly observed by the Fig. 3. (a) UVeVis spectra of the developed H2O2 colorimetric chemosensor after the addition of (i) distilled water (control sample), (ii) 0.1 mM of H2O2, (iii) 0.1 mM of ascorbic acid, (iv) 0.1 mM of fructose, (v) 0.1 mM of galactose, (vi) 0.1 mM of lactose and (vii) 0.1 mM of sucrose (Inset: color of corresponding samples). (b) DA/A (%) of the corresponding samples from (a).Materials and Devices 5 (2020) 385e391naked eyes or via UVeVis measurement. This reaction indicates the second role of AgNPs, i.e. as a signal probe for reporting H2O2 concentration in a sample. The color change in the AgNP solution can be attributed to the etching effect of H2O2 on AgNPs. Accord- ingly, AgNPs are anisotropically sculptured by H2O2, leading to changes in the size and shape of AgNPs and a reduction in AgNP concentration [23,25e32]. The TEM image (Fig. 2a) shows that the original AgNP particles exhibit a spherical shape with particle sizes of approximately 20e40 nm. Fig. 2d shows a narrow size Fig. 4. (a) UVeVis spectra of AgNPs-based colorimetric H2O2 detection with concen- trations ranging from 0 mM to 0.1 mM (Inset: color of corresponding samples). (b, c) Calibration curves for H2O2 sensing with different modes: (b) in the relationship of DA/ A (%) vs. [H2O2] and (c) in a relationship of Dlmax vs. [H2O2]. in AgNP size can be determined by comparing it with the original size of the AgNPs, i.e. approximately 40 nm in Fig. 2a. This result implies that AgNP particles have been etched by the as-generated H2O2 from Eq. (1). The DLS of this sample (Fig. 2f) shows a new distribution zone with a small size ranging from 1 nm to 10 nm. This zone can be attributed to the etching of the AgNP particles. The obtained DLS results are fitted with the TEM data in Fig. 2c. In addition, this reaction can also be monitored in UVeVis mea- surements (Fig. 6a) through the reduction in OD425 nm of the surface plasmon resonance of the AgNP solution versus the added cholesterol concentration in the sample. All these results validate the proposed working principle of the cholesterol biosensor, as illustrated in Fig. 1b. 4.2. AgNP-based colorimetric chemosensor for H2O2 determination Fig. 3a presents the UVeVis spectra of AgNP solutions after incubation with 0.1 mM ascorbic acid, fructose, galactose, sucrose and lactose. No changes occurred in the intensity of the specific absorption at 425 nm (OD425 nm) when distilled water (the control sample; Fig. 3a, curve i), ascorbic acid, fructose, galactose, lactose or sucrose (Fig. 3a, curves iii to vii, respectively) was added to the AgNP solutions. Meanwhile, a strong reduction in OD425 nm was ced Materials and Devices 5 (2020) 385e391 389H.V. Tran et al. / Journal of Science: Advandistribution (20 nme110 nm) of the original AgNPs with an average size of 40 nm, which are in good agreement with the TEM data (Fig. 2a). The TEM image of the AgNP solution after incubationwith 0.8 mM H2O2 (Fig. 2b) clearly shows that H2O2 has etched AgNPs, leading to a strong reduction in the size of AgNPs ranging from ~40 nm (Fig. 2a) to ~2 nm. Moreover, the density of AgNPs after incubation with 0.8 mM H2O2 (Fig. 2b) is considerably lower than that in Fig. 2a (i.e. the original AgNP solution), implying a decrease in AgNP concentration. Similarly, the DLS of the AgNP solution after incubation with 0.8 mM H2O2 (Fig. 2e) exhibits an extremely nar- row distribution (~0.8 nme~2.5 nm) of the size of AgNPs, which is remarkably small compared with that of the original AgNPs (~20 nme~110 nm; Fig. 2d). The TEM and DLS results imply a strong etching reaction between H2O2 and AgNPs as proposed by the working principle illustrated in Fig. 1a. Fig. 1b presents the proposed working principle of the colori- metric cholesterol biosensor that was developed by combining the AgNP solution with ChOx enzyme. ChOx is well-known for exhibiting extremely high specificity towards cholesterol (Eq. (1)). ChOx oxidises cholesterol into cholest-4-en-3-one and releases H2O2, which is immediately recognized by the AgNP solution- based H2O2 colorimetric sensor. The TEM image of the AgNP so- lution after incubation with a reacted mixture containing ChOx enzyme þ 1 mM cholesterol (Fig. 2c) presents diameters of approximately 3 nme5 nm for the AgNP particles. This reduction observed in the AgNP solutions incubated with H2O2 (curve ii). The percentage difference in OD425 nm (DA/A, %; Fig. 3b) exhibited small values for DA/A upon the addition of water (~0%), ascorbic Fig. 5. (a) UVeVis spectra of the AgNP solution (i) before and (ii) after incubation with 0.25 mg mL1 of ChOx. (b) UVeVis spectra of the mixture of AgNPs and 0.25 mg mL1 of ChOx in the presence of 1 mM of (i) glucose, (ii) lactose, (iii) sucrose, (iv) ascorbic acid, (v) water (blank sample) and (vi) cholesterol (Inset: DA/A (%) of the corre- sponding samples).Fig. 6. (a) UVeVis spectra of the cholesterol biosensor with different cholesterol concentrations. (b) Calibration curve for cholesterol detection. tific Line 0.1e 0.5. 0.1e 0.5e 0.02 0.02 0.00 2.10 0.02 0.01 osen this cedacid (~0.79%), fructose (~4.17%), galactose (~2.04%), lactose (~2.88%) or sucrose (~1.14%). Meanwhile, the DA/A value reached 98.48% when H2O2 was added. These data demonstrated the high selectivity of the AgNP-based H2O2 colorimetric chemosensor. Notably, the color of the AgNP solution faded when H2O2 was added. By contrast, it did not change when nonspecific targets were added (Fig. 3a, inset), implying that this colorimetric change can be observed by the naked eyes for the immediate qualitative detection of H2O2. Fig. 4a shows the UVeVis spectra of the AgNP solutions after incubation with different H2O2 concentrations. A decrease in OD425 nm can be observed by increasing the added [H2O2] concentration. The calibration curve in Fig. 4b exhibits a linear relationship be- tween the DA/A (%) of OD425 nm and [H2O2] (0 mMe0.05 mM) with R2 ¼ 0.965. The limit of detection (LOD) was estimated to be 3.5 mM (S/N ¼ 3). In Fig. 4b, a redshift on the specific peak position (Dlmax, nm) of the AgNP solution (425 nm for the original AgNP solution) can be seen when the added [H2O2] was increased Table 2 LOD of the developed cholesterol biosensor based on the use of nanomaterials as ar Materials Enzyme used AgNPs þ ChOx ChOx ZnO NPs/CNTs ChOx CuO/graphene nanosphere ChOx p(HEMA)/polypyrrole ChOx CuO NPs ChOx (PDDAe[MWCNTseChOx]) ChOx AuNPs ChOx MoS2 nanosheets ChOx GQDs ChOx ChOxeCS/HbeCS ChOx Table 3 Results of cholesterol determination from serum samples by using the developed bi Sample Added cholesterol (mM) Recognised by the biosensor developed in No. 1 0.0 9.14 0.5 10.44 No. 2 0.0 7.02 0.5 8.04 H.V. Tran et al. / Journal of Science: Advan390(Fig. 4a). This finding suggests thatDlmax can be utilized as a second signal for H2O2 monitoring, with a linear relationship between Dlmax (nm) and [H2O2], as shown in Fig. 4c (R2 ¼ 0.964). The LOD was estimated to be 3.5 mM (S/N ¼ 3). 4.3. Colorimetric biosensor for cholesterol determination Fig. 5a shows the UVeVis of the AgNP solutions before (curve i) and after inc