Research article Received: 17 June 2014,
Revised: 30 September 2014,
Accepted: 19 November 2014
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/bio.2832
Highly sensitive and selective electrochemiluminescence determination of cholesterol utilizing a functional electrode with a core–shell nanostructure Wei Li, Junying Ge and Changzhi Zhao* ABSTRACT: A highly sensitive and selective method for the determination of cholesterol is required to evaluate trace amounts of cholesterol in test samples. In this work, selected gold nanoparticles (AuNPs) and 5-amino-2-mercapto-1,3,4thiadiazole (AMT) were used and a thin film of three-dimensional gold–AMT core–shell nanoparticles (p-AMT–AuNPs) was prepared using an electrochemical method. Cholesterol oxidase was then bonded to the film surface to give a functional electrode. Based on catalysis by the electrode functionalized for cholesterol and a luminol–H2O2 electrochemiluminescence (ECL) system, a highly sensitive and selective ECL method was developed for the determination of cholesterol. Under optimized conditions, ECL intensity showed a good linear relationship with cholesterol over the concentration range 0.05– 11.0 μg/ml, with a correlation coefficient of 0.999 and a limit of detection of 0.02 μg/ml. The proposed method was used to determine cholesterol in dairy products with a relative standard deviation of < 1.8% and recovery rates of 98.1–104%. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: cholesterol; electrochemiluminescence; core–shell nanoparticles; Au nanoparticles; 5-amino-2-mercapto-1,3,4-thiadiazole
Introduction Cholesterol is a natural steroid compound. It is widely present in animal cell membranes, and is also an important precursor of hormones and bile in humans (1). Although cholesterol is important and necessary for human health, an abnormal cholesterol intake may cause some diseases, such as metabolic disorders, hypertension, arteriosclerosis, coronary heart disease, brain thrombosis and myocardial infarction (1,2). As an essential nutrient in humans, dietary intake is the main cause of elevated blood cholesterol, and comes mainly from animal products (e. g. dairy). The estimation of cholesterol levels in food has attracted more and more attention in order to control cholesterol intake. Various methods have been reported for the determination of cholesterol, including spectrophotometry (3), fluorescent spectrometry (4,5), chemiluminescence (6), high-performance liquid chromatography (7) and electrochemical methods (8–11). Some of these have certain disadvantages, such as scarce reagents (3,5), expensive equipment (7), poor selectivity or sensitivity (8,9) or excessive operation times (10,11). A few of the analytical methods reported in recent years have led to advancements in the determination of cholesterol in specific samples, however, special pretreatment or techniques are usually needed (12). Thus, the development of a highly sensitive and selective, convenient and cost-efficient technique for the estimation of cholesterol in foods is important. Electrochemiluminescence (ECL) is the production of light from an electrode surface when a suitable potential is applied to the ECL reagents (13). Among its numerous applications, ECL has been successfully used in cholesterol analysis by
Luminescence 2015
applying luminol (LH2) as an ECL reagent (14). The mechanism involved is shown in equations (1)-(5). ChE
Cholesteryl acetate þ H2 O → Cholesterol þ Acetic acid ChOx
Cholesterol þ O2 → Δ-Cholestenone þ H2 O2 OH–
(1) (2)
LH2 → LH– e– →L•– þ Hþ
(3)
L•– þ H2 O2 →AP2• þ H2 O
(4)
AP2• →AP2þ hv ð425 nmÞ
(5)
A functional working electrode in favor of the analyte is needed to develop a highly sensitive and selective ECL method. In the modified electrode technique, an electropolymerized polymer film might be a good choice, because it can improve the efficiency of the conduction electrons, shorten response time, decrease the overpotential of the analytes in the redox process and show good stability (15). 5-Amino-2-mercapto1,3,4-thiadiazole (AMT), a heterocyclic thiol compound, has been successfully used in the modification of electrodes and has been * Correspondence to: C. Zhao, College of Chemistry & Molecular Engineering, Qingdao University of Science & Technology, Qingdao 266042, China. E-mail: [email protected] College of Chemistry & Molecular Engineering, Qingdao University of Science & Technology, Qingdao, China
Copyright © 2015 John Wiley & Sons, Ltd.
W. Li et al. applied in electrochemical analysis (16). In addition to pure polymers, electrode modification using composites of polymers containing gold nanoparticles (AuNPs) has attracted a great deal of attention because of the interesting optical and electrocatalytic properties of AuNPs (17,18). Among nano-structured materials, AuNPs often gather as a core within a polymer shell, forming a stable three-dimensional (3D) structure on the electrode. Based on the excellent performance of core–shell nano-structured materials, Terzi et al. successfully developed a glucose sensor consisting of a poly (3,4-ethylenedioxythiophene)/citrate anion-encapsulated AuNPmodified GC electrode (19). Although core–shell NPs have been used in electrochemical analysis, no report was available on the ECL determination of cholesterol using a core–shell NP-modified electrode. In the present study, we prepared a thin film of 3D gold–AMT core–shell nanoparticles (p-AMT–AuNPs) using AuNPs and AMT in an electrochemical method. Then, via glutaraldehyde connected to the free –NH2 group on the surface of p-AMT–AuNPs, cholesterol oxidase (ChOx) was bonded onto the shell to make a functional electrode. Integration of the functionalized electrode and ECL system of luminol–H2O2 (Fig. 1) allowed the development of a highly sensitive and selective ECL method for the determination of cholesterol.
Experimental Materials and reagents AMT and luminol were obtained from J&K Chemical Ltd (Beijing, China). ChOx (EC 1.1.3.6, 10 units/mg protein) and cholesterol esterase (ChE, EC 3.1.1.13, 10,000 units/g protein) were from Shanghai Lanji Science and Technology Development Co., Ltd
(Shanghai, China). Cholesterol, HAuCl4 and glutaraldehyde (50%) were from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). All other chemicals were of analytical grade or higher. A stock solution (5.00 × 103 mol/l) of cholesterol was prepared, as described previously (0.0484 g cholesterol in 25 ml water with 1 ml isopropanol and 0.5 ml Triton X-100) (8). Working solutions of cholesterol were freshly prepared by diluting the stock solution with PBS (pH 7.0) containing 1% (v/v) Triton X-100. A borax solution (pH 8.0) was used as the electrolyte solution. Deionized water (> 15 MΩ) was used throughout the study and was acquired from a KLUP-III water treatment system (Kang Ning Water Industry, China). Full-fat flavored yogurt, De yi sour milk and skimmed milk were purchased from a supermarket for use as samples. The three dairy products were processed simply. A 1.0 g sample of each product was put into a 10 ml volumetric flask, with 3 ml of 1% Triton X-100 phosphate buffer and 3.0 ml of ChE (1.0 mg/ml), the temperature of the volumetric flask was maintained at 37 °C for 30 min in a water bath; the volume was then made up to 10 ml with 1% Triton X-100 PBS. After pretreatment, the dairy products were used as test solutions. Apparatus and measurements Voltammetric experiments were performed with a CHI 660B electrochemical workstation (Shanghai Chenhua Instrument, Co., Ltd, China). For ECL measurements, a MPI-B multifunctional ECL analyzer (Xi’an Remex Analytical Instrument Ltd, Co.) was used at room temperature. The standard three-electrode arrangement consisted of a CHI 111 Ag/AgCl reference electrode, a CHI 115 Pt counter electrode and a modified electrode as working electrode. A glassy carbon disk (3.0 mm diameter)
Figure 1. Preparation of the functionalized electrode and ECL determination of cholesterol.
wileyonlinelibrary.com/journal/luminescence
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015
Highly sensitive and selective determination for cholesterol electrode (GCE) was used as the base for fabrication of the modified electrode. The emission window was placed in front of a photomultiplier tube (detection range 300–650 nm) biased at 600 V. The cholesterol sample was put into the electrolytic cell, and the functionalized electrode was immersed in the electrolytic cell for 1 min, the ECL spectrum was then obtained by generating a step potential (SP) from 0 to 0.65 V. All measurements were repeated at least five times and the means of measurements are presented with the relative standard deviation (RSD). Scanning electron microscopy (SEM) images were obtained on a Hitachi S-4300 SEM (Japan).
ChOx, and a functionalized electrode with ChOx bonding to p-AMT–AuNPs was successfully prepared.
Results and discussion Characteristics of the functionalized electrode
HAuCl4 · 3H2O (0.1%; 5.0 ml) and 0.25 ml AMT (1.0 mmol/l) were added to 45 ml of water in a round-bottomed flask with constant stirring for 30 min. Then, 5.0 ml NaBH4 (0.125%) was added and stirred for another 30 min. The color of the solution turned light red indicating the formation of AMT–AuNPs (20). The GCE was first polished to a mirror-like finish using 3000mesh emery paper, followed by 1.0 and 0.05 μm alumina slurry. The GCE was then washed successively with 1 : 1 nitric acid and alcohol, sonicated in a deionized water bath to remove any residual alumina and dried. The electropolymerization of AMT– AuNPs on GCE was performed using a cycling potential between 0.40 and 1.70 V at a scan rate of 50 mV/s. When the scanning cycle had been repeated for 15 rounds, a thin film was obtained on the surface of the GCE. The electrode was then washed thoroughly in PBS to remove the adsorbed AMT–AuNPs and dried in air; a GCE covered with poly-AMT–AuNPs (p-AMT–AuNPs/GCE) was obtained. As shown in Fig. 2, homogeneous p-AMT–AuNPs with irregular sphericity accumulated on the surface of the GCE, and an aggregated configuration of nanoparticles is clearly seen in the SEM image, indicating that the p-AMT–AuNPs were successfully immobilized on the surface of the electrode. This result is similar to that reported in the literature (20). Finally, the p-AMT–AuNPs/GCE was immersed in a mixture of ChOx (1.0 mg/ml) and glutaraldehyde (1.5%) for 8 h. The electrode was then washed thoroughly in PBS to remove the adsorbed
Because electrochemical impedance spectroscopy (EIS) is an effective tool for characterizing the interface properties of electrodes, it is generally used to characterize the results of electrode modification. Figure 3 shows EIS images of the bare GCE, p-AMT–AuNPs/GCE and the functionalized electrode in 1.0 mmol/l [Fe(CN)6]3/4 solution. The surface electron-transfer resistance, Ret, derived from the semicircle diameter of EIS = 800 Ω for the bare GCE (Fig. 3, curve a). After the AMT– AuNPs had been electropolymerized on GCE, Ret was found to be 1250 Ω (Fig. 3, curve b). This increase indicated that the charge-transfer rate of [Fe(CN)6]3/4 was reduced, which was attributed to the hindrance effect on this redox couple of the deposited AMT–AuNPs layer on the GCE surface. For the functionalized electrode, Ret (3400 Ω) increased substantially (Fig. 3, curve c). This increase in diameter indicated that the chargetransfer rate of [Fe(CN)6]3/4 was reduced obviously, which was attributed to the hindrance effect on this redox couple, caused by the bonding of protein ChOx to p-AMT–AuNPs on the electrode surface. Linear sweep voltammetric (LSV) and ECL curves of the functionalized electrode in the absence and presence of cholesterol were obtained in an alkaline solution of luminol (Fig. 4). The relevant curves for ECL intensity are given versus the applied potential at the functionalized electrode. When the potential scan pointed in a positive direction, LSV in the absence of cholesterol (Fig. 4A, a) showed an oxidation peak for luminol at ~ 570 mV versus Ag/AgCl, but ECL was not obvious because of the absence of cholesterol. Very weak ECL (Fig. 4B, a) is suggested from the electro-oxidation of luminol anion (LH‾) to the luminol radical anion (L•‾) (equation (2)). L•‾ )was then oxidized by the dissolved O2 in the solution to produce excited state 3-aminophthalate species (AP2‾•) and ECL. In the presence of cholesterol, the anodic peak of luminol (Fig. 4A, b) increased obviously, which was attributed to the diffusion of luminol towards the electrode due to the
Figure 2. SEM of electropolymerized AMT–AuNPs film on the GCE surface.
Figure 3. EIS of various electrodes in 0.1 mol/l KCl containing 1.0 mmol/l [Fe(CN)6] : (a) bare GCE, (b) p-AMT–AuNPs/GCE and (c) the functionalized electrode. The frequency range is between 0.01 and 100,000 Hz with applied voltage of 5 mV. (Inset) Equivalent circuit.
Synthesis of AMT-capped AuNPs and the preparation of functionalized of electrode
3/4
Luminescence 2015
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/luminescence
W. Li et al.
Figure 5. Three consecutive ECL amplitudes at linear sweep (a) and step potential (b) from 0 to 0.65 V in the electrolyte solution containing 4.00 μg/ml cholesterol on the functionalized electrode.
Figure 4. Linear sweep voltammograms (A) and their corresponding ECL (B) for 4 1.00 × 10 mol/l luminol containing 0.1 mol/l borax solution (pH 8.0) and NaNO3 in the absence (a) and presence (b) of cholesterol on the functionalized electrode at 100 mV/s.
oxidation of L•‾ by H2O2 (equation (3)), which was generated by catalytic reaction of the functionalized electrode to cholesterol. Simultaneously, a strong ECL was obtained, because L•‾ was oxidized to the excited state AP2‾• (equation (4)), which subsequently emitted 425 nm light (equation (5)). Based on selective catalysis of the functionalized electrode for cholesterol and the high sensitivity of ECL, cholesterol could be determined with highly sensitivity and selectivity.
cholesterol at the functionalized electrode, the amounts of protein deposited on p-AMT–AuNPs/GCE were investigated. As the ChOx deposition time increased from 2 to 10 h, ECL responses increased continually, and reached a maximum at 8 h. With further increases in the protein deposition time, the ECL intensity began to decrease. This could be attributed to an overload of proteins hindering the charge-transfer of luminol to the electrode. Considering the above factors, we chose 8 h as the immobilization time for the enzyme. Effect of the surfactant concentration. Triton X-100 was used in the preparation of the sample solution to increase the solubility of Ch. Because the surfactant might form a micelle that affects the determination of Ch, the luminous intensity of the modified electrode for Ch was investigated at different Triton concentrations. When the Triton concentration was < 1%, Ch was not completely dissolved and the luminous intensity was weak. When the concentration was > 5%, the surfactant formed micelles that might interfere with the determination of Ch. Because Ch can dissolve completely in a 1% surfactant solution, and light emission was strong enough for the determination of Ch, a 1% ratio was selected as the Triton concentration.
Selection of the generated method for ECL A variety of electrochemical methods are available to stimulate ECL. In addition to the most commonly used LSV, other methods are step potential (SP), constant potential electrolysis and pulse potential voltammetry. As a method for generating ECL, SP has the advantage of lower reagent consumption, good repeatability and saving time. In SPs, multi-potential steps can be applied and cycled, and the obtained response waveforms are recorded as a function of time. Figure 5 shows ECL amplitudes at LSV (a) and SP (b) from 0 to 0.65 V for 4.00 μg/ml cholesterol on the functionalized electrode. It can be seen (a and b) that the amplitude generated by SP was approximately four times that of LSV, and the response time was reduced by a quarter. However, the amplitude generated by SP is almost unchanged, showing that SP generated ECL has good reproducibility. Based on comparison of the experimental results, the SP method was used to generate ECL. Optimization of experimental conditions Effect of the amount on immobilizing enzyme. The amount of immobilized ChOx noticeably affects the sensitivity of the cholesterol determination. To obtain the best response to
wileyonlinelibrary.com/journal/luminescence
Effect of electrolyte pH. The pH of the electrolyte plays an important role in enzyme activity and the ECL reaction. In general, the ECL reaction of luminol needed alkaline conditions, but the best pH for ChOx was 7.0 (14). In order to obtain an optimal pH for the electrolyte solution, the effect of electrolyte pH on ECL intensity was investigated using various buffer solutions, such as NaAc–HAc (pH = 6.0, 6.5), PBS (pH = 7.0, 7.5) and borax solution (pH = 8.0, 8.5, 9.0). At a fixed concentration of luminol and an applied potential of 0.570 V, the resulting profile (Fig. 5) shows an optimum pH of 8.0, although some catalytic activity of the enzyme could be bound at this pH. Thus, the electrolyte pH was maintained at 8.0 in the experiment. Effect of luminol concentration. Because luminol is a lightemitting reagent, its concentration has a distinct influence on ECL. By varying the luminol concentration from 1.00 × 106 to 1.00 × 103 mol/l, its effect on ECL intensity was investigated. Initial tests showed that the ratio of the enhanced ECL intensity of the sample to the blank ECL signal was dependent upon the luminol concentration. When the luminol concentration exceeded 1.00 × 104 mol/l, there was almost no further increase in the ECL, indicating that 1.00 × 104 mol/l is a suitable concentration of luminol.
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015
Highly sensitive and selective determination for cholesterol
Figure 6. ECL responses on the functionalized electrode in the electrolyte containing 0, 0.05, 0.80, 2.00, 3.50, 4.00, 5.50, 7.00, 8.00, 9.00, 10.0, 11.0 and 12.0 μg/ml cholesterol under optimized experimental conditions.
Effect of step potential. The SP provides the oxidation potential to luminol, and strong influences ECL intensity. From Fig. 4 (Aa), it can be seen that the luminol was oxidized at 0.570 V on the surface of the functionalized electrode, but considering the effect of overpotential, optimization of SP is necessary. Optimization of SP was performed over a potential range from 0.40 to 0.80 V. The results show that a positive shift in SP obviously affects ECL intensity, which increased with the positive shift in SP. However, when SP was above +0.65 V, the ECL intensity showed no significant increase. In addition, other chemicals might be oxidized at more positive potentials, thus interfering with determination of the substrate. Hence, SP was set at 0.65 V in this study. Range and sensitivity The linear range for the determination of cholesterol was investigated using the functionalized electrode. As shown in Fig. 6, ECL responses increased with increasing substrate concentration from 0 to 12.0 μg/ml, and ECL intensities were proportional to cholesterol concentrations over the range 0.05–11.0 μg/ml. The calibration equation was IECL = 1.00 × 102 C + 557.15 (where C is the concentration of cholesterol in μg/ml) with correlation coefficients of 0.999. The limit of detection was estimated to be
0.02 μg/ml at a signal-to-noise ratio of 3. The low detection limit shows that the presented method allows the determination of trace amounts of cholesterol in the samples. Accuracy, repeatability, stability and interference test The use of Certified Reference Materials (CRMs) to establish the accuracy of a new analytical method has gradually become regular practice. Following to the proposed method, CRMs containing 8.00 μg/ml cholesterol were determined. The results obtained for seven measurements were 7.92 ± 0.11 (95% confidence interval), giving an acceptable level of accuracy and precision. The functionalized electrode also shows good reproducibility. The RSD for five successive determinations of the substrate was 2.5%. Five functionalized electrodes were prepared using the same conditions, and their RSD was 4.1% for cholesterol. Selectivity was always restricted for the determination of cholesterol in actual samples using electrochemical methods (14). To confirm the selectivity of the presented method, possible interferences were investigated by measuring 0.80 μg/ml cholesterol in electrolyte containing a specified concentration of the interfering compounds. Addition of 0.50 μg/ml lactose, vitamin B2, vitamin D and vitamin A resulted in ~ 1.3% deviation. After
Table 1. Determination for total cholesterol in dairy products Sample
Determined cholesterol (mg/100 g)
RSD (n = 7, %)
Amount of cholesterol (mg/100 g) Added
Full-fat flavored yogurt Deyi sour milk Yili skimmed milk
Luminescence 2015
7.82 5.73 4.41
1.82 0.83 1.01
1.00 1.00 4.00
Copyright © 2015 John Wiley & Sons, Ltd.
Recovery (%)
Determined 8.65 7.02 7.91
98.1 104 98.9
wileyonlinelibrary.com/journal/luminescence
W. Li et al. adding 0.20 μg/ml lactic acid, glucose and histidine, the deviation is < 1.2%. So the effect of interference on the ECL response is negligible under the test conditions because of enzymatic selectivity. The outstanding selectivity of the method is attributed to the selectivity of the enzyme on the electrode. Determination of total cholesterol in dairy products Cholesterol in dairy products is mainly in the form of cholesterol esters, and must be converted to free cholesterol with ChE for determination. To avoid deterioration, the samples were freshly prepared. After the ECL responses of cholesterol standard solutions had been determined, sample solutions were measured under the same conditions. All the cholesterol concentrations in the detection solutions were within the linear response range. Recovery tests were performed by adding various concentrations of standard substrates to the test solution. The results listed in Table 1 show that the amounts of cholesterol found in the dairy products were fairly close to the labeled values. As summarized in Table 1, RSD was < 1.8% in seven parallel measurements of the same sample, and the recovery was in the range of 98.1–104%.
Conclusions This study showed that a functionalized electrode has high sensitivity and high selectivity for the determination of cholesterol. In particular, the functionalized electrode has the advantages of simple preparation, good repeatability and stability. The electrode-based system could be conveniently applied as a platform in the determination of other analytes using other enzymesubstituted electrodes for the testing of biological samples. Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 21475072), the Natural Science Foundation of Shandong Province, China (No. 2009ZRB01393), and the Open-end Fund of State Key Laboratory of Electroanaytical Chemistry (No. SKLEAC201106).
References 1. Arya SK , Datta M Malhotra BD . Recent advances in cholesterol biosensor. Biosens Bioelectron 2008;23:1083–100. 2. Shih WC , Yang MC Lin MS . Development of disposable lipid biosensor for the determination of total cholesterol. Biosens Bioelectron 2009;24:1679–84. 3. Li R , Xiong C , Xiao Z Ling L . Colorimetric detection of cholesterol 2 with G-quadruplex-based DNAzymes and ABTS . Anal Chim Acta 2012;724:80–5.
wileyonlinelibrary.com/journal/luminescence
4. Cheng Y , Jiang P , Lin S , Li Y Dong X . An imprinted fluorescent chemosensor prepared using dansyl-modified β-cyclodextrin as the functional monomer for sensing of cholesterol with tailor-made selectivity. Sensor Actuat B-Chem 2014;193:838–43. 5. Duan M , Peng Y , Zhang L , Wang X , Ge J , Jiang J , et al. DNA-stabilized silver nanoclusters with guanine-enhanced fluorescence as a novel indicator for enzymatic detection of cholesterol. Anal Method 2013;5:2182–7. 6. Hong L , Liu A-L , Li G-W , Chen W Lin X-H . Chemiluminescent cholesterol sensor based on peroxidase-like activity of cupric oxide nanoparticles. Biosens Bioelectron 2013;43:1–5. 7. Bauer LC , Santana DA , Macedo MS , Torres AG , Souza NE Simionato JI . Method validation for simultaneous determination of cholesterol and cholesterol oxides in milk by RP-HPLC–DAD. J Braz Chem Soc 2014;25:161–8. 8. Zhao C , Wan L , Jiang L , Wang Q Jiao K . Highly sensitive and selective cholesterol biosensor based on direct electron transfer of hemoglobin. Anal Biochem 2008;383:25–30. 9. Matharu Z , Solanki PR , Gupta V Malhotra BD . Mediator free cholesterol biosensor based on self-assembled monolayer platform. Analyst 2012;137:747–53. 10. Ruecha N , Rangkupan R , Rodthongkum N Chailapakul O . Novel paper-based cholesterol biosensor using graphene/polyvinyl pyrrolidone/polyaniline nanocomposite. Biosens Bioelectron 2014;52:13–9. 11. Srivastava M , Srivastava SK , Nirala NR Prakash R . A chitosan-based polyaniline–Au nanocomposite biosensor for determination of cholesterol. Anal Method 2014;6:817–24. 12. Xiao H-Y , Chen J , Wei G-F , Hu Q-Q , Ai Y-Q , Han Y , et al. Determination of cholesterol in hogwash oil by flow injection chemiluminescence with bis(2,4,6-trichlorophenyl) oxalate-hydrogen peroxide system. Chinese J Anal Chem 2013;41:432–5. 13. Forster RJ , Bertoncello P Keyes TE . Electrogenerated chemiluminescence. Annu Rev Anal Chem 2009;2:359–85. 14. Zhang M , Yuan R , Chai Y , Chen S , Zhong H , Wang C , et al. A biosensor for cholesterol based on gold nanoparticles-catalyzed luminol electrogenerated chemiluminescence. Biosens Bioelectron 2012;32:288–92. 15. Cosnier S , Holzinger M . Electrosynthesized polymers for biosensing. Chem Soc Rev 2011;40:2146–56. 16. Kalimuthu P , John SA . Nanostructured electropolymerized film of 5amino-2- mercapto-1,3,4-thiadiazole on glassy carbon electrode for the selective determination of L-cysteine. Electrochem Commun 2009;11:367–70. 17. Xiao YH , Li CM . Nanocomposites: from fabrications to electrochemical bioapplications. Electroanalysis 2008;20:648–62. 18. Riskin M , Tel-Vered R , Lioubashevski O Willner I . Ultrasensitive surface plasmon resonance detection of trinitrotoluene by a bis-anilinecross-linked Au nanoparticles composite. J Am Chem Soc 2009;131:7368–78. 19. Terzi F , Zanfrognini B , Zanardi C , Pigani L Seeber R . Poly(3,4ethylenedioxythiophene)/Au-nanoparticles composite as electrode coating suitable for electrocatalytic oxidation. Electrochim Acta 2011;56:3575–9. 20. Kannan P , John SA . Fabrication of conducting polymer-gold nanoparticles film on electrodes using monolayer protected gold nanoparticles and its electrocatalytic application. Electrochim Acta 2011;56:7029–37.
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015
Revised: 30 September 2014,
Accepted: 19 November 2014
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/bio.2832
Highly sensitive and selective electrochemiluminescence determination of cholesterol utilizing a functional electrode with a core–shell nanostructure Wei Li, Junying Ge and Changzhi Zhao* ABSTRACT: A highly sensitive and selective method for the determination of cholesterol is required to evaluate trace amounts of cholesterol in test samples. In this work, selected gold nanoparticles (AuNPs) and 5-amino-2-mercapto-1,3,4thiadiazole (AMT) were used and a thin film of three-dimensional gold–AMT core–shell nanoparticles (p-AMT–AuNPs) was prepared using an electrochemical method. Cholesterol oxidase was then bonded to the film surface to give a functional electrode. Based on catalysis by the electrode functionalized for cholesterol and a luminol–H2O2 electrochemiluminescence (ECL) system, a highly sensitive and selective ECL method was developed for the determination of cholesterol. Under optimized conditions, ECL intensity showed a good linear relationship with cholesterol over the concentration range 0.05– 11.0 μg/ml, with a correlation coefficient of 0.999 and a limit of detection of 0.02 μg/ml. The proposed method was used to determine cholesterol in dairy products with a relative standard deviation of < 1.8% and recovery rates of 98.1–104%. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: cholesterol; electrochemiluminescence; core–shell nanoparticles; Au nanoparticles; 5-amino-2-mercapto-1,3,4-thiadiazole
Introduction Cholesterol is a natural steroid compound. It is widely present in animal cell membranes, and is also an important precursor of hormones and bile in humans (1). Although cholesterol is important and necessary for human health, an abnormal cholesterol intake may cause some diseases, such as metabolic disorders, hypertension, arteriosclerosis, coronary heart disease, brain thrombosis and myocardial infarction (1,2). As an essential nutrient in humans, dietary intake is the main cause of elevated blood cholesterol, and comes mainly from animal products (e. g. dairy). The estimation of cholesterol levels in food has attracted more and more attention in order to control cholesterol intake. Various methods have been reported for the determination of cholesterol, including spectrophotometry (3), fluorescent spectrometry (4,5), chemiluminescence (6), high-performance liquid chromatography (7) and electrochemical methods (8–11). Some of these have certain disadvantages, such as scarce reagents (3,5), expensive equipment (7), poor selectivity or sensitivity (8,9) or excessive operation times (10,11). A few of the analytical methods reported in recent years have led to advancements in the determination of cholesterol in specific samples, however, special pretreatment or techniques are usually needed (12). Thus, the development of a highly sensitive and selective, convenient and cost-efficient technique for the estimation of cholesterol in foods is important. Electrochemiluminescence (ECL) is the production of light from an electrode surface when a suitable potential is applied to the ECL reagents (13). Among its numerous applications, ECL has been successfully used in cholesterol analysis by
Luminescence 2015
applying luminol (LH2) as an ECL reagent (14). The mechanism involved is shown in equations (1)-(5). ChE
Cholesteryl acetate þ H2 O → Cholesterol þ Acetic acid ChOx
Cholesterol þ O2 → Δ-Cholestenone þ H2 O2 OH–
(1) (2)
LH2 → LH– e– →L•– þ Hþ
(3)
L•– þ H2 O2 →AP2• þ H2 O
(4)
AP2• →AP2þ hv ð425 nmÞ
(5)
A functional working electrode in favor of the analyte is needed to develop a highly sensitive and selective ECL method. In the modified electrode technique, an electropolymerized polymer film might be a good choice, because it can improve the efficiency of the conduction electrons, shorten response time, decrease the overpotential of the analytes in the redox process and show good stability (15). 5-Amino-2-mercapto1,3,4-thiadiazole (AMT), a heterocyclic thiol compound, has been successfully used in the modification of electrodes and has been * Correspondence to: C. Zhao, College of Chemistry & Molecular Engineering, Qingdao University of Science & Technology, Qingdao 266042, China. E-mail: [email protected] College of Chemistry & Molecular Engineering, Qingdao University of Science & Technology, Qingdao, China
Copyright © 2015 John Wiley & Sons, Ltd.
W. Li et al. applied in electrochemical analysis (16). In addition to pure polymers, electrode modification using composites of polymers containing gold nanoparticles (AuNPs) has attracted a great deal of attention because of the interesting optical and electrocatalytic properties of AuNPs (17,18). Among nano-structured materials, AuNPs often gather as a core within a polymer shell, forming a stable three-dimensional (3D) structure on the electrode. Based on the excellent performance of core–shell nano-structured materials, Terzi et al. successfully developed a glucose sensor consisting of a poly (3,4-ethylenedioxythiophene)/citrate anion-encapsulated AuNPmodified GC electrode (19). Although core–shell NPs have been used in electrochemical analysis, no report was available on the ECL determination of cholesterol using a core–shell NP-modified electrode. In the present study, we prepared a thin film of 3D gold–AMT core–shell nanoparticles (p-AMT–AuNPs) using AuNPs and AMT in an electrochemical method. Then, via glutaraldehyde connected to the free –NH2 group on the surface of p-AMT–AuNPs, cholesterol oxidase (ChOx) was bonded onto the shell to make a functional electrode. Integration of the functionalized electrode and ECL system of luminol–H2O2 (Fig. 1) allowed the development of a highly sensitive and selective ECL method for the determination of cholesterol.
Experimental Materials and reagents AMT and luminol were obtained from J&K Chemical Ltd (Beijing, China). ChOx (EC 1.1.3.6, 10 units/mg protein) and cholesterol esterase (ChE, EC 3.1.1.13, 10,000 units/g protein) were from Shanghai Lanji Science and Technology Development Co., Ltd
(Shanghai, China). Cholesterol, HAuCl4 and glutaraldehyde (50%) were from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). All other chemicals were of analytical grade or higher. A stock solution (5.00 × 103 mol/l) of cholesterol was prepared, as described previously (0.0484 g cholesterol in 25 ml water with 1 ml isopropanol and 0.5 ml Triton X-100) (8). Working solutions of cholesterol were freshly prepared by diluting the stock solution with PBS (pH 7.0) containing 1% (v/v) Triton X-100. A borax solution (pH 8.0) was used as the electrolyte solution. Deionized water (> 15 MΩ) was used throughout the study and was acquired from a KLUP-III water treatment system (Kang Ning Water Industry, China). Full-fat flavored yogurt, De yi sour milk and skimmed milk were purchased from a supermarket for use as samples. The three dairy products were processed simply. A 1.0 g sample of each product was put into a 10 ml volumetric flask, with 3 ml of 1% Triton X-100 phosphate buffer and 3.0 ml of ChE (1.0 mg/ml), the temperature of the volumetric flask was maintained at 37 °C for 30 min in a water bath; the volume was then made up to 10 ml with 1% Triton X-100 PBS. After pretreatment, the dairy products were used as test solutions. Apparatus and measurements Voltammetric experiments were performed with a CHI 660B electrochemical workstation (Shanghai Chenhua Instrument, Co., Ltd, China). For ECL measurements, a MPI-B multifunctional ECL analyzer (Xi’an Remex Analytical Instrument Ltd, Co.) was used at room temperature. The standard three-electrode arrangement consisted of a CHI 111 Ag/AgCl reference electrode, a CHI 115 Pt counter electrode and a modified electrode as working electrode. A glassy carbon disk (3.0 mm diameter)
Figure 1. Preparation of the functionalized electrode and ECL determination of cholesterol.
wileyonlinelibrary.com/journal/luminescence
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015
Highly sensitive and selective determination for cholesterol electrode (GCE) was used as the base for fabrication of the modified electrode. The emission window was placed in front of a photomultiplier tube (detection range 300–650 nm) biased at 600 V. The cholesterol sample was put into the electrolytic cell, and the functionalized electrode was immersed in the electrolytic cell for 1 min, the ECL spectrum was then obtained by generating a step potential (SP) from 0 to 0.65 V. All measurements were repeated at least five times and the means of measurements are presented with the relative standard deviation (RSD). Scanning electron microscopy (SEM) images were obtained on a Hitachi S-4300 SEM (Japan).
ChOx, and a functionalized electrode with ChOx bonding to p-AMT–AuNPs was successfully prepared.
Results and discussion Characteristics of the functionalized electrode
HAuCl4 · 3H2O (0.1%; 5.0 ml) and 0.25 ml AMT (1.0 mmol/l) were added to 45 ml of water in a round-bottomed flask with constant stirring for 30 min. Then, 5.0 ml NaBH4 (0.125%) was added and stirred for another 30 min. The color of the solution turned light red indicating the formation of AMT–AuNPs (20). The GCE was first polished to a mirror-like finish using 3000mesh emery paper, followed by 1.0 and 0.05 μm alumina slurry. The GCE was then washed successively with 1 : 1 nitric acid and alcohol, sonicated in a deionized water bath to remove any residual alumina and dried. The electropolymerization of AMT– AuNPs on GCE was performed using a cycling potential between 0.40 and 1.70 V at a scan rate of 50 mV/s. When the scanning cycle had been repeated for 15 rounds, a thin film was obtained on the surface of the GCE. The electrode was then washed thoroughly in PBS to remove the adsorbed AMT–AuNPs and dried in air; a GCE covered with poly-AMT–AuNPs (p-AMT–AuNPs/GCE) was obtained. As shown in Fig. 2, homogeneous p-AMT–AuNPs with irregular sphericity accumulated on the surface of the GCE, and an aggregated configuration of nanoparticles is clearly seen in the SEM image, indicating that the p-AMT–AuNPs were successfully immobilized on the surface of the electrode. This result is similar to that reported in the literature (20). Finally, the p-AMT–AuNPs/GCE was immersed in a mixture of ChOx (1.0 mg/ml) and glutaraldehyde (1.5%) for 8 h. The electrode was then washed thoroughly in PBS to remove the adsorbed
Because electrochemical impedance spectroscopy (EIS) is an effective tool for characterizing the interface properties of electrodes, it is generally used to characterize the results of electrode modification. Figure 3 shows EIS images of the bare GCE, p-AMT–AuNPs/GCE and the functionalized electrode in 1.0 mmol/l [Fe(CN)6]3/4 solution. The surface electron-transfer resistance, Ret, derived from the semicircle diameter of EIS = 800 Ω for the bare GCE (Fig. 3, curve a). After the AMT– AuNPs had been electropolymerized on GCE, Ret was found to be 1250 Ω (Fig. 3, curve b). This increase indicated that the charge-transfer rate of [Fe(CN)6]3/4 was reduced, which was attributed to the hindrance effect on this redox couple of the deposited AMT–AuNPs layer on the GCE surface. For the functionalized electrode, Ret (3400 Ω) increased substantially (Fig. 3, curve c). This increase in diameter indicated that the chargetransfer rate of [Fe(CN)6]3/4 was reduced obviously, which was attributed to the hindrance effect on this redox couple, caused by the bonding of protein ChOx to p-AMT–AuNPs on the electrode surface. Linear sweep voltammetric (LSV) and ECL curves of the functionalized electrode in the absence and presence of cholesterol were obtained in an alkaline solution of luminol (Fig. 4). The relevant curves for ECL intensity are given versus the applied potential at the functionalized electrode. When the potential scan pointed in a positive direction, LSV in the absence of cholesterol (Fig. 4A, a) showed an oxidation peak for luminol at ~ 570 mV versus Ag/AgCl, but ECL was not obvious because of the absence of cholesterol. Very weak ECL (Fig. 4B, a) is suggested from the electro-oxidation of luminol anion (LH‾) to the luminol radical anion (L•‾) (equation (2)). L•‾ )was then oxidized by the dissolved O2 in the solution to produce excited state 3-aminophthalate species (AP2‾•) and ECL. In the presence of cholesterol, the anodic peak of luminol (Fig. 4A, b) increased obviously, which was attributed to the diffusion of luminol towards the electrode due to the
Figure 2. SEM of electropolymerized AMT–AuNPs film on the GCE surface.
Figure 3. EIS of various electrodes in 0.1 mol/l KCl containing 1.0 mmol/l [Fe(CN)6] : (a) bare GCE, (b) p-AMT–AuNPs/GCE and (c) the functionalized electrode. The frequency range is between 0.01 and 100,000 Hz with applied voltage of 5 mV. (Inset) Equivalent circuit.
Synthesis of AMT-capped AuNPs and the preparation of functionalized of electrode
3/4
Luminescence 2015
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/luminescence
W. Li et al.
Figure 5. Three consecutive ECL amplitudes at linear sweep (a) and step potential (b) from 0 to 0.65 V in the electrolyte solution containing 4.00 μg/ml cholesterol on the functionalized electrode.
Figure 4. Linear sweep voltammograms (A) and their corresponding ECL (B) for 4 1.00 × 10 mol/l luminol containing 0.1 mol/l borax solution (pH 8.0) and NaNO3 in the absence (a) and presence (b) of cholesterol on the functionalized electrode at 100 mV/s.
oxidation of L•‾ by H2O2 (equation (3)), which was generated by catalytic reaction of the functionalized electrode to cholesterol. Simultaneously, a strong ECL was obtained, because L•‾ was oxidized to the excited state AP2‾• (equation (4)), which subsequently emitted 425 nm light (equation (5)). Based on selective catalysis of the functionalized electrode for cholesterol and the high sensitivity of ECL, cholesterol could be determined with highly sensitivity and selectivity.
cholesterol at the functionalized electrode, the amounts of protein deposited on p-AMT–AuNPs/GCE were investigated. As the ChOx deposition time increased from 2 to 10 h, ECL responses increased continually, and reached a maximum at 8 h. With further increases in the protein deposition time, the ECL intensity began to decrease. This could be attributed to an overload of proteins hindering the charge-transfer of luminol to the electrode. Considering the above factors, we chose 8 h as the immobilization time for the enzyme. Effect of the surfactant concentration. Triton X-100 was used in the preparation of the sample solution to increase the solubility of Ch. Because the surfactant might form a micelle that affects the determination of Ch, the luminous intensity of the modified electrode for Ch was investigated at different Triton concentrations. When the Triton concentration was < 1%, Ch was not completely dissolved and the luminous intensity was weak. When the concentration was > 5%, the surfactant formed micelles that might interfere with the determination of Ch. Because Ch can dissolve completely in a 1% surfactant solution, and light emission was strong enough for the determination of Ch, a 1% ratio was selected as the Triton concentration.
Selection of the generated method for ECL A variety of electrochemical methods are available to stimulate ECL. In addition to the most commonly used LSV, other methods are step potential (SP), constant potential electrolysis and pulse potential voltammetry. As a method for generating ECL, SP has the advantage of lower reagent consumption, good repeatability and saving time. In SPs, multi-potential steps can be applied and cycled, and the obtained response waveforms are recorded as a function of time. Figure 5 shows ECL amplitudes at LSV (a) and SP (b) from 0 to 0.65 V for 4.00 μg/ml cholesterol on the functionalized electrode. It can be seen (a and b) that the amplitude generated by SP was approximately four times that of LSV, and the response time was reduced by a quarter. However, the amplitude generated by SP is almost unchanged, showing that SP generated ECL has good reproducibility. Based on comparison of the experimental results, the SP method was used to generate ECL. Optimization of experimental conditions Effect of the amount on immobilizing enzyme. The amount of immobilized ChOx noticeably affects the sensitivity of the cholesterol determination. To obtain the best response to
wileyonlinelibrary.com/journal/luminescence
Effect of electrolyte pH. The pH of the electrolyte plays an important role in enzyme activity and the ECL reaction. In general, the ECL reaction of luminol needed alkaline conditions, but the best pH for ChOx was 7.0 (14). In order to obtain an optimal pH for the electrolyte solution, the effect of electrolyte pH on ECL intensity was investigated using various buffer solutions, such as NaAc–HAc (pH = 6.0, 6.5), PBS (pH = 7.0, 7.5) and borax solution (pH = 8.0, 8.5, 9.0). At a fixed concentration of luminol and an applied potential of 0.570 V, the resulting profile (Fig. 5) shows an optimum pH of 8.0, although some catalytic activity of the enzyme could be bound at this pH. Thus, the electrolyte pH was maintained at 8.0 in the experiment. Effect of luminol concentration. Because luminol is a lightemitting reagent, its concentration has a distinct influence on ECL. By varying the luminol concentration from 1.00 × 106 to 1.00 × 103 mol/l, its effect on ECL intensity was investigated. Initial tests showed that the ratio of the enhanced ECL intensity of the sample to the blank ECL signal was dependent upon the luminol concentration. When the luminol concentration exceeded 1.00 × 104 mol/l, there was almost no further increase in the ECL, indicating that 1.00 × 104 mol/l is a suitable concentration of luminol.
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015
Highly sensitive and selective determination for cholesterol
Figure 6. ECL responses on the functionalized electrode in the electrolyte containing 0, 0.05, 0.80, 2.00, 3.50, 4.00, 5.50, 7.00, 8.00, 9.00, 10.0, 11.0 and 12.0 μg/ml cholesterol under optimized experimental conditions.
Effect of step potential. The SP provides the oxidation potential to luminol, and strong influences ECL intensity. From Fig. 4 (Aa), it can be seen that the luminol was oxidized at 0.570 V on the surface of the functionalized electrode, but considering the effect of overpotential, optimization of SP is necessary. Optimization of SP was performed over a potential range from 0.40 to 0.80 V. The results show that a positive shift in SP obviously affects ECL intensity, which increased with the positive shift in SP. However, when SP was above +0.65 V, the ECL intensity showed no significant increase. In addition, other chemicals might be oxidized at more positive potentials, thus interfering with determination of the substrate. Hence, SP was set at 0.65 V in this study. Range and sensitivity The linear range for the determination of cholesterol was investigated using the functionalized electrode. As shown in Fig. 6, ECL responses increased with increasing substrate concentration from 0 to 12.0 μg/ml, and ECL intensities were proportional to cholesterol concentrations over the range 0.05–11.0 μg/ml. The calibration equation was IECL = 1.00 × 102 C + 557.15 (where C is the concentration of cholesterol in μg/ml) with correlation coefficients of 0.999. The limit of detection was estimated to be
0.02 μg/ml at a signal-to-noise ratio of 3. The low detection limit shows that the presented method allows the determination of trace amounts of cholesterol in the samples. Accuracy, repeatability, stability and interference test The use of Certified Reference Materials (CRMs) to establish the accuracy of a new analytical method has gradually become regular practice. Following to the proposed method, CRMs containing 8.00 μg/ml cholesterol were determined. The results obtained for seven measurements were 7.92 ± 0.11 (95% confidence interval), giving an acceptable level of accuracy and precision. The functionalized electrode also shows good reproducibility. The RSD for five successive determinations of the substrate was 2.5%. Five functionalized electrodes were prepared using the same conditions, and their RSD was 4.1% for cholesterol. Selectivity was always restricted for the determination of cholesterol in actual samples using electrochemical methods (14). To confirm the selectivity of the presented method, possible interferences were investigated by measuring 0.80 μg/ml cholesterol in electrolyte containing a specified concentration of the interfering compounds. Addition of 0.50 μg/ml lactose, vitamin B2, vitamin D and vitamin A resulted in ~ 1.3% deviation. After
Table 1. Determination for total cholesterol in dairy products Sample
Determined cholesterol (mg/100 g)
RSD (n = 7, %)
Amount of cholesterol (mg/100 g) Added
Full-fat flavored yogurt Deyi sour milk Yili skimmed milk
Luminescence 2015
7.82 5.73 4.41
1.82 0.83 1.01
1.00 1.00 4.00
Copyright © 2015 John Wiley & Sons, Ltd.
Recovery (%)
Determined 8.65 7.02 7.91
98.1 104 98.9
wileyonlinelibrary.com/journal/luminescence
W. Li et al. adding 0.20 μg/ml lactic acid, glucose and histidine, the deviation is < 1.2%. So the effect of interference on the ECL response is negligible under the test conditions because of enzymatic selectivity. The outstanding selectivity of the method is attributed to the selectivity of the enzyme on the electrode. Determination of total cholesterol in dairy products Cholesterol in dairy products is mainly in the form of cholesterol esters, and must be converted to free cholesterol with ChE for determination. To avoid deterioration, the samples were freshly prepared. After the ECL responses of cholesterol standard solutions had been determined, sample solutions were measured under the same conditions. All the cholesterol concentrations in the detection solutions were within the linear response range. Recovery tests were performed by adding various concentrations of standard substrates to the test solution. The results listed in Table 1 show that the amounts of cholesterol found in the dairy products were fairly close to the labeled values. As summarized in Table 1, RSD was < 1.8% in seven parallel measurements of the same sample, and the recovery was in the range of 98.1–104%.
Conclusions This study showed that a functionalized electrode has high sensitivity and high selectivity for the determination of cholesterol. In particular, the functionalized electrode has the advantages of simple preparation, good repeatability and stability. The electrode-based system could be conveniently applied as a platform in the determination of other analytes using other enzymesubstituted electrodes for the testing of biological samples. Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 21475072), the Natural Science Foundation of Shandong Province, China (No. 2009ZRB01393), and the Open-end Fund of State Key Laboratory of Electroanaytical Chemistry (No. SKLEAC201106).
References 1. Arya SK , Datta M Malhotra BD . Recent advances in cholesterol biosensor. Biosens Bioelectron 2008;23:1083–100. 2. Shih WC , Yang MC Lin MS . Development of disposable lipid biosensor for the determination of total cholesterol. Biosens Bioelectron 2009;24:1679–84. 3. Li R , Xiong C , Xiao Z Ling L . Colorimetric detection of cholesterol 2 with G-quadruplex-based DNAzymes and ABTS . Anal Chim Acta 2012;724:80–5.
wileyonlinelibrary.com/journal/luminescence
4. Cheng Y , Jiang P , Lin S , Li Y Dong X . An imprinted fluorescent chemosensor prepared using dansyl-modified β-cyclodextrin as the functional monomer for sensing of cholesterol with tailor-made selectivity. Sensor Actuat B-Chem 2014;193:838–43. 5. Duan M , Peng Y , Zhang L , Wang X , Ge J , Jiang J , et al. DNA-stabilized silver nanoclusters with guanine-enhanced fluorescence as a novel indicator for enzymatic detection of cholesterol. Anal Method 2013;5:2182–7. 6. Hong L , Liu A-L , Li G-W , Chen W Lin X-H . Chemiluminescent cholesterol sensor based on peroxidase-like activity of cupric oxide nanoparticles. Biosens Bioelectron 2013;43:1–5. 7. Bauer LC , Santana DA , Macedo MS , Torres AG , Souza NE Simionato JI . Method validation for simultaneous determination of cholesterol and cholesterol oxides in milk by RP-HPLC–DAD. J Braz Chem Soc 2014;25:161–8. 8. Zhao C , Wan L , Jiang L , Wang Q Jiao K . Highly sensitive and selective cholesterol biosensor based on direct electron transfer of hemoglobin. Anal Biochem 2008;383:25–30. 9. Matharu Z , Solanki PR , Gupta V Malhotra BD . Mediator free cholesterol biosensor based on self-assembled monolayer platform. Analyst 2012;137:747–53. 10. Ruecha N , Rangkupan R , Rodthongkum N Chailapakul O . Novel paper-based cholesterol biosensor using graphene/polyvinyl pyrrolidone/polyaniline nanocomposite. Biosens Bioelectron 2014;52:13–9. 11. Srivastava M , Srivastava SK , Nirala NR Prakash R . A chitosan-based polyaniline–Au nanocomposite biosensor for determination of cholesterol. Anal Method 2014;6:817–24. 12. Xiao H-Y , Chen J , Wei G-F , Hu Q-Q , Ai Y-Q , Han Y , et al. Determination of cholesterol in hogwash oil by flow injection chemiluminescence with bis(2,4,6-trichlorophenyl) oxalate-hydrogen peroxide system. Chinese J Anal Chem 2013;41:432–5. 13. Forster RJ , Bertoncello P Keyes TE . Electrogenerated chemiluminescence. Annu Rev Anal Chem 2009;2:359–85. 14. Zhang M , Yuan R , Chai Y , Chen S , Zhong H , Wang C , et al. A biosensor for cholesterol based on gold nanoparticles-catalyzed luminol electrogenerated chemiluminescence. Biosens Bioelectron 2012;32:288–92. 15. Cosnier S , Holzinger M . Electrosynthesized polymers for biosensing. Chem Soc Rev 2011;40:2146–56. 16. Kalimuthu P , John SA . Nanostructured electropolymerized film of 5amino-2- mercapto-1,3,4-thiadiazole on glassy carbon electrode for the selective determination of L-cysteine. Electrochem Commun 2009;11:367–70. 17. Xiao YH , Li CM . Nanocomposites: from fabrications to electrochemical bioapplications. Electroanalysis 2008;20:648–62. 18. Riskin M , Tel-Vered R , Lioubashevski O Willner I . Ultrasensitive surface plasmon resonance detection of trinitrotoluene by a bis-anilinecross-linked Au nanoparticles composite. J Am Chem Soc 2009;131:7368–78. 19. Terzi F , Zanfrognini B , Zanardi C , Pigani L Seeber R . Poly(3,4ethylenedioxythiophene)/Au-nanoparticles composite as electrode coating suitable for electrocatalytic oxidation. Electrochim Acta 2011;56:3575–9. 20. Kannan P , John SA . Fabrication of conducting polymer-gold nanoparticles film on electrodes using monolayer protected gold nanoparticles and its electrocatalytic application. Electrochim Acta 2011;56:7029–37.
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015
