A novel electrochemiluminescence sensor coupled with capillary electrophoresis for simultaneous determination of quinapril hydrochloride and its metabolite quinaprilat hydrochloride in human plasma

Talanta 179 (2018) 213–220

Contents lists available at ScienceDirect

Talanta journal homepage: www.elsevier.com/locate/talanta

A novel electrochemiluminescence sensor coupled with capillary electrophoresis for simultaneous determination of quinapril hydrochloride and its metabolite quinaprilat hydrochloride in human plasma Shuangjiao Suna,b, Yanfen Weia, Hao Wanga, Yupin Caoa, Biyang Denga,

MARK

⁎

a

Key Laboratory for the Chemistry and Molecular Engineering of Medicinal Resources (Ministry of Education of China), School of Chemistry and Pharmaceutical Sciences, Guangxi Normal University, Guilin 541004, China b Shaoyang University, Shaoyang 422000, China

A R T I C L E I N F O

A B S T R A C T

Keywords: Sensor Electrochemiluminescence Capillary electrophoresis Quinapril Human plasma Zinc oxide nanoparticle

A novel electrochemiluminescence (ECL) sensor with composite consisted of silica-sol, Zinc oxide nanoparticles (ZnO NPs), polyvinylpyrrolidone (PVP) and tris(2, 2′-bipyridine) ruthenium (II) was constructed. A new method for simultaneous determination of quinapril hydrochloride (QHCl) and its metabolite quinaprilat hydrochloride (QTHCl) in human plasma was developed using the ECL sensor coupled with capillary electrophoresis (CE). ECL intensities of QHCl and QTHCl increased dramatically when the ECL sensor was used as working electrode. The running buffer contains 14 mmol/L phosphate (pH 8.0) and 20% n-propyl alcohol. Under optimized experimental conditions, the linearity ranges of the method are 0.007–8.0 μg/mL for QHCl and 0.009–8.3 μg/mL for QTHCl. The detection limits of QHCl and QTHCl (S/N=3) are 3.6 ng/mL and 3.9 ng/mL, respectively. The method was applied for the simultaneous determination of QHCl and QTHCl in human plasma with satisfactory results.

1. Introduction In ECL study, ECL sensors were constructed based on ECL reagent immobilized on the electrode. It did not only reduce the luminous reagent consumption and analysis cost, but also can simplify experimental design [1,2]. Nano-particles can significantly improve the determination sensitivity of ECL sensors due to their unique electrical and optical properties [3]. At present, nano-modified ECL sensors have become a hot topic of ECL research [4–8]. Zinc oxide nanoparticles (ZnO NPs) were applied on photoelectro-catalysis [9], gas sensor [10], modified electrode due to their good physic properties, such as good electrical conduction, catalytic activity, optical performance, and adsorption ability [11–13]. In recent years, the application of sol-gel technology has gradually increased in electrochemistry analysis, especially in the preparation of composite electrode and film modified electrodes because of its good characteristics in conductivity, photovoltaic cells and electrocatalysis [14–16]. Silica-sol is a translucent colloidal solution and could form a three-dimensional network structure which has the properties of good biocompatibility, physical rigidity, chemically inertness, thermal stability and almost no swelling properties in water [14]. The applications of silica sol-gel technology in preparing modified electrode had been ⁎

Corresponding author. E-mail address: [email protected] (B. Deng).

https://doi.org/10.1016/j.talanta.2017.10.050 Received 5 June 2017; Received in revised form 17 October 2017; Accepted 24 October 2017 Available online 04 November 2017 0039-9140/ © 2017 Elsevier B.V. All rights reserved.

reported recently [17–21]. Silica-sol has film-forming properties. However, the existence of high shrinkage in the film-forming process could result in membrane cracks easily. Therefore, the water-soluble organic polymer can be added as the auxiliary film. Polyvinylpyrrolidone (PVP) is a kind of water-soluble amino polymer with flexible chain structure, which provided with high solubility, film-formation, complexation property and surface activity can be used as coating agent, dispersant, thickener, and binder, etc. [14]. Quinapril (Fig. S1A) hydrochloride (QHCl) is an active angiotensin converting enzyme inhibitor used as a safe and high-performance antihypertensive drug in treating hypertension and heart failure [22,23]. Its active metabolite, quinaprilat (Fig. S1B) hydrochloride (QTHCl) showed higher angiotensin converting enzyme activity on heart, arteries and kidney than QHCl used as angiotensin converting enzyme inhibitor [24,25]. Until now, some methods have been developed for the simultaneous determinations of QHCl and QTHCl, including high performance liquid chromatography (HPLC) coupled with UV detection [26], solid phase extraction HPLC method [27], ultra-performance liquid chromatography–electrospray ionization mass spectrometry (UHPLC-ES-MS/MS) [28], HPLC-MS/MS [29]. These methods have drawbacks, such as costly instrumentation, large reagent use and time-consuming procedures. Capillary electrophoresis with electrochemiluminescence (CE-

Talanta 179 (2018) 213–220

S. Sun et al.

then merged the ethyl acetate extracts and evaporated to dryness under a stream of dry nitrogen at 80 °C. The dry residue was dissolved in 200 μL double distilled water (DDW). Prior to analysis, sample solutions were filtered through 0.45 µm membrane filters.

ECL) possesses the advantages of high sensitivity, wide linearity range, less sample volume, good selectivity, and low equipment cost. It has been widely applied to pharmaceutical analysis in recent years [30–33]. However, CE-ECL for the simultaneous determination of QHCl and QTHCl has not been reported. In this paper, a silica-sol/ZnO NPs/PVP/ Ru(bpy)32+ ECL sensor was prepared for the simultaneous determination of QHCl and QTHCl based on CE separation. ZnO NPs were applied for enhancing electron translation rate and adsorption ability of Ru (bpy)32+. Silica-sol was used to fix Ru(bpy)32+ on the electrode surface. PVP was added into silica-sol to prevent the film rupture. The electrochemical properties and detection performances of the sensor were investigated in detail. The method selectivity was greatly improved due to CE powerful separation.

2.3. Preparation of ZnO NPs ZnO NPs were prepared according to reference and some improvements were made [39]. Under the rapid magnetic stirring, 2.0 mL NaOH solution (0.5 mol/L) was dropwise added into 50 mL ZnSO4 solution (0.10 mol/L). Continued to stir for 0.5 h and formed a translucent gel-like solution, which was alkaline zinc carbonate precursor solution. ZnO NPs were obtained by solid state reaction after the precursor solution being dried at 80 °C, then roasted at 300 °C for 2 h. ZnO nanoparticles were dispersed in 5 mL DDW by using ultrasonic waves and were purified by dialysis for 12 h using dialysis bag with molecular weight cut off 6000–8000 in DDW. Dried the purified ZnO NPs solution at 60 °C and pure ZnO NPs with particle size about 20 nm were obtained.

2. Experimental 2.1. Apparatus and reagents The MPI-B CE–ECL system was produced by Xi’an Remex Electronic Science-Tech Co., Ltd. (Xi’an, China). It consists of four main parts: a numerical control capillary electrophoresis high-voltage power supply, a multifunction chemiluminescence detector, a multichannel data collection analyzer and a numerical control flow injection sample injector [34–36]. To meet the needs of the measurement, an end-column ECL cell designed in reference was used and it is composed of a threeelectrode system: a silica-sol/PVP/ZnO NPs/Ru(bpy)32+ sensor as the working electrode, a Pt wire as the auxiliary electrode and Ag/AgCl (saturated KCl) as the reference electrode [37]. A 50 cm length of uncoated fused-silica capillary (75 µm i.d.) was obtained from Yongnian Optical Conductive Fiber Plant (Hebei, China). A Model HSJ-4A pH meter was produced by Shanghai Precision and Scientific Instrument Corporation (Shanghai, China). A XW-80A vortex mixer was obtained from Haimen Kylin-Bell Lab Instruments Co., Ltd. (Jiangsu, China). JEM-2100F transmission electron microscope (JEOL, Tokyo, Japan); Tris(2,2′-bipyridyl) ruthenium(II) chloride hexahydrate was purchased from Alfa Aesar (Johnson Matthey, Ward Hill, MA, USA). QHCl was obtained from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China) and QTHCl was produced by Shanghai Zhenzhun Biological Technology Co., Ltd. (Shanghai, China). QHCl tablets were produced by Harbin Pharmaceutical Group Co., Ltd. (Harbin, China); NH4HCO3 was produced Tianjin Guangfu Technology Development Co., Ltd. (Tianjin, China); ZnSO4, Na2HPO4, Na3PO4, NaH2PO4, NaOH, Tween 80 (TW-80), n-propyl alcohol were purchased from Xilong Chemical Co., Ltd. (Guangdong, China). Silica-sol was produced by Jining Huide Chemical Co., Ltd. (Jining, China). Polyvinylpyrrolidone (PVP), β-cyclodextrin (β-CD) and sodium dodecyl sulfate (SDS) were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All reagents used were of analytical grade. Double-distilled water (DDW) was used throughout.

2.4. Preparation of modified sensor The glassy carbon electrode (GCE) was treated by being polished with 0.30 µm and 0.050 µm Al2O3 polishing powder and cleaned by ultrasound in a mixed solution of ethanol and DDW with the volume ratio of 1:1, then waited to dry naturally. Added 1.5 mg ZnO NPs into 0.3 mL silica-sol and mixed it with ultrasound for 30 min, then added 0.15 mL PVP (0.05%) and 0.15 mL Ru(bpy)32+ (2.0 mmol/L), continued to mix with ultrasound for 30 min and obtained a silica-sol/PVP/ ZnO NPs/ Ru(bpy)32+ compound solution. Immersed the processed glassy carbon electrode in the compound solution for 30 s, then took out and placed it at 4 °C for 12 h due to gaining stable sensor. After being washed three times by DDW, the silica-sol/PVP/ZnO NPs/Ru (bpy)32+ sensor was prepared for experiment. The schematic fabrication process of ECL sensor and CE-ECL detection was shown in Fig. 1. 2.5. Procedures A new capillary was activated by 0.1 mol/L NaOH for 12 h. Before each run, the activated capillary was flushed with 0.1 mol/L NaOH for 10 min and DDW for 10 min, and then the corresponding running buffer for 10 min. Adjust the distance of silica-sol/PVP/ZnO NPs/Ru(bpy)32+ sensor and capillary outlet end about 200 µm. A 350 μL 55 mmol/L (pH=8.0) phosphate solution was added into the ECL detection cell before the experiment. In all experiments, samples were introduced into the capillary by electrokinetic injection at 12 kV for 10 s and separated in the capillary at 12 kV. Detection potential was fixed at 1.20 V. The running buffer (pH 8.0) contained 14 mmol/L phosphate and 20% npropyl alcohol. The potential of the photomultiplier tube (PMT) was operated at 800 V.

2.2. Sample preparation 3. Results and discussion 3.0 mL of blank blood sample was drawn from a healthy adult male volunteer with weight 62 kg and height 172 cm. The volunteer was not taking any medications in the past month. QHCl tablets (containing 20 mg QHCl)) were taken with 250 mL lukewarm water, and then 3.0 mL blood sample was obtained from elbow vein after 1 h. The blood samples were extracted according to the reference [38]. Firstly, they were set in centrifugal test tubes which were treated by heparin sodium, then centrifuged at 3500 rpm for 10 min in order to separate the plasma. 200 μL plasma was accurately transferred into a centrifugal tube, then 50 μL of 0.10 mol/L NaOH and 2 mL ethyl acetate were added into the centrifugal tube. After being shaken on a vortex mixer for 5 min, the mixtures were centrifuged at 12,000 rpm for 10 min. The upper layer was transferred into another tube and the lower residue was extracted twice again with ethyl acetate according to the previous step,

3.1. Effect of modification agents on ECL intensity The prepared ZnO NPs were characterized by transmission electron microscopy. As shown in Fig. 2A, the particle size was homogeneous and it was about 20 nm The XRD patterns of the prepared ZnO NPs were shown in Fig. 2B. All the detectable peaks could be indexed as the ZnO hexagonal wurtzite structure found in the standard reference data (JCPDS: 36e1451). Effect of modification agent on ECL intensity was investigated. The dosage of different reagents in modifier has great influence on ECL intensity. When silica-sol volume was 0.3 mL, the effects of ZnO NPs, PVP and Ru(bpy)32+ amount on ECL intensity were studied in the paper. The ECL intensity obviously increased with the increasing of ZnO NPs use level and reached the maximum when ZnO 214

Talanta 179 (2018) 213–220

S. Sun et al.

Fig. 1. Schematic fabrication process of ECL sensor and CE-ECL detection.

reaction system, Ru(bpy)3 2+ is first oxidized to Ru(bpy)33+, then Ru (bpy)3 3+ reacts with QHCl and QTHCl and produces Ru(bpy)32+* after a series of reaction steps. Ru(bpy)32+*, which is not stable, releases the energy and causes ECL emission. As shown in Fig. 4A, the sharp ECL peaks with almost constant intensity was obtained in the experiment. The ECL intensity of the sensor is 95% of original intensity when the sensor was sealed with plastic wrap and was placed at 4 °C for 3 weeks. The experimental results showed that the silica-sol/PVP/Nano-ZnO/Ru (bpy)32+ sensor had good stability.

NPs dosage was 1.5 mg, then weakened gradually with the increasing of ZnO NPs (Fig. 3A). It could be explained that the moderate increase of ZnO NPs may result in the enhance on electron transmission rate between Ru(bpy)32+ and GCE, however, excessive ZnO NPs would reduce the contact between Ru(bpy)32+ and QHCl, therefore, the ECL intensity downed. The PVP use level was investigated (Fig. 3B). It indicated that the ECL intensity increased with the increasing of PVP when PVP volume is less than 1.5 mL then became weaker when PVP volume is more than 1.5 mL. It could be explained that PVP could not cover the electrode surface entirely when PVP volume is less than 1.5 mL and the silica gel, ZnO NPs and Ru (bpy)32+ would shell from the electrode surface, which leaded to the decrease of ECL intensity. When PVP volume is more than 1.5 mL, the increasing of PVP film's thickness would affect the entrance of drugs into the membrane and reduce to react with Ru(bpy)32+, therefore, it decrease the ECL intensity. Considering the ECL intensity and the stability of colloidal silica film, 0.15 mL PVP (0.05%) was chosen in the experiment. The effect of Ru(bpy)32+ dosage on ECL intensity of QHCl were also researched, As shown in Fig. S2, the ECL intensity increased with the increasing of Ru(bpy)32+ use level, however, the blank ECL intensity of the sensor obviously enhanced and the system had an excellent signal to noise ratio when Ru(bpy)32+ (2.0 mmol/L) use level was 0.15 mL. 0.15 mL Ru(bpy)32+ (2.0 mmol/ L) was chosen for further experiment.

3.3. Sensor repeatability The repeatability of the same sensor was investigated by continuously detecting the ECL intensity of 2.0 μg/mL QTHCl under the same conditions and the RSD of peak heights was 2.8% (n=12) (Fig. 4B). The repeatability of the five sensors of the same method preparation was also investigated by detection of 2.0 μg/mL QTHCl under the same conditions. The experimental results showed the RSD of the peak heights was 3.6% in 15 times measurement. Therefore, the experimental results showed that the silica-sol/PVP/Nano-ZnO/Ru (bpy)32+ sensor had good repeatability. 3.4. Electrochemical and ECL behaviors of sensor

3.2. Sensor stability

The voltammetric behaviors of silica-sol/PVP, silica-sol/PVP/Ru (bpy)32+, and silica-sol/Nano-ZnO/PVP/Ru(bpy)32+ sensor for 2.0 μg/ mL QHCl and 2.0 μg/mL QTHCl were studied at a scan rate of 0.1 V s−1. As shown in Fig. 5A, silica-sol/PVP did not appear redox peaks, silica-sol/PVP/Ru(bpy)32+ had weak redox peaks and silica-sol/

The stability of the silica-sol/PVP/Nano-ZnO/Ru(bpy)32+ sensor was examined by scanning continuous CV cycles for 12 min in solution of 0.05 mol/L PBS (pH 7.5) containing 2.0 μg/mL QTHCl. In this

Fig. 2. TEM image (A) and XRD pattern (B) of ZnO NPs.

215

Talanta 179 (2018) 213–220

S. Sun et al.

Fig. 4. ECL intensity-time curve for continuous 35 cycles in 12 min (A) and repeatability (B) using the constructed sensor. 2.0 μg/mL QTHCl in 0.05 M PBS (pH 7.5) at a scan rate of 0.1 V s−1. Fig. 3. Effects of ZnO-NPs mass (A) and PVP volume (B) on ECL intensities. Detection conditions for A: scan rate, 100 mV/s; sensor: 0.30 mL Silica-sol, 0.20 mL PVP, 0.20 mL Ru(bpy)32+ (2.0 mmol/L); solution: 50 mmol/L pH7.5 PBS, 2.0 μg/mL QHCl. Detection conditions for B: scan rate, 100 mV/s; sensor: 0.30 mL Silica-sol, 1.5 mg ZnO NPs, 0.20 mL Ru(bpy)32+ (2.0 mmol/L); solution: 50 mmol/L pH7.5 PBS, 2.0 μg/mL QHCl.

Nano-ZnO/PVP/Ru(bpy)32+ sensor had the highest oxidation current peak and it could be explained that ZnO nanoparticles can increase reaction sites. The electrochemical behaviors of silica-sol/PVP/Ru (bpy)32+ and silica-sol/PVP/Ru(bpy)32+ in 50 mmol/L PBS at pH 7.5 were also studied before and after adding 2.0 μg/mL QTHCl and QHCl. As shown in Fig. 5B, ECL intensity enhanced after QHCl and QTHCl were added into the ECL detection cell. Compared with the silica-sol/ PVP/Ru(bpy)32+sensor, silica-sol/Nano-ZnO/PVP/Ru(bpy)32+ sensor had obvious ECL sensitization for 2.0 μg/mL QHCl and QTHCl and the ECL intensity had enhanced 4 times for QHCl and 5 times for QTHCl. 3.5. Selection of separation additives QTHCl is a metabolic product of QHCl. Compared with QHCl, it does not have the ethyl bonded to carbonyl oxygen. The similar structure of QHCl and QTHCl brings the difficulty of separation. As shown in Fig. 6A(a), electrophoresis peaks of the two analytes overlapped completely when running buffer did not have separation additive. In order to obtain good resolution, the effect of PVP, SDS, β-CD, TW-80 and n-propyl alcohol on the separation of the two analytes were studied, respectively. The results showed that PVP, SDS, β-CD, and TW80 had little effect on the separation. However, n-propyl alcohol could distinctly improve the separation degree. The effect of n-propyl alcohol volume fraction on separation was studied. As shown in Fig. 6A, the two analytes had good separation, short retention time and high sensitivity when the volume fraction of n-propyl alcohol was 20%. In following experiment, 20% n-propyl alcohol was added into the running buffer solution as additive. Under the same conditions, the CE-ECL of the two

Fig. 5. CVs (A) and ECL (B) profiles. (a) silica-sol/PVP；(b) silica-sol/PVP/Ru (bpy)32+；(c) b+2.0 μg/mL QTHCl；(d) b + 2.0 μg/mLQHCl; (e) silica-sol/Nano-ZnO/ PVP/Ru(bpy)32+；(f) e+2.0 μg/mL QTHCl; (g) e +2.0 μg/mLQHCl; Scan rate for 0.1 V s−1 and the buffer solution in detection cell for 50 mmol/L PBS at pH7.5.

216

Talanta 179 (2018) 213–220

S. Sun et al.

effect of buffer concentration in the ECL cell on ECL intensity was also studied. As shown in Fig. 7B, the ECL intensities of the two analytes reached the maximum when the buffer concentration was 55 mmol/L. Therefore, 55 mmol/L phosphate buffer was selected for further experiments.

3.8. Effects of pH and concentration of running buffer The effect of pH and concentration of running buffer (containing 20% n-propanol) on the ECL intensity was investigated. As shown in Fig. 7C, changed the pH from 5.5 to 9.0 and the ECL intensities of the two analytes reached their maximum values at pH 8.0 when the buffer concentration was kept at 12 mmol/L. When pH value of the running buffer was kept at 8.0, the buffer concentrations were changed from 6.0 to 18 mmol/L. As shown in Fig. S4, the highest ECL signal was obtained for QHCl and QTHCl when the buffer concentration was 14 mmol/L. With the continuous increasing of phosphate buffer concentration, ECL intensity decreased. This is because the increasing of ionic strength in the buffer would increase the electrophoretic current and Joule heat which would broaden the electrophoresis peak width and reduce ECL intensity. Therefore, 14 mmol/L phosphate buffer containing 20% npropanol at pH 8.0 was chosen for following experiments.

3.9. Effect of separation voltage Separation voltage directly influenced the migration time and the ECL intensity of QHCl and QTHCl. High separation voltage would increase the current and shorten the analysis time. However, too high separation voltage produced the Joule's heat, which affected the separation of the tested analytes. The influence of CE separation voltage on ECL intensity and separation was investigated from 8.0 to 16 kV. As shown in Fig. 7D, the ECL intensity of both QHCl and QTHCl increased with the increasing of separation voltage and reached their maximum values when the separation voltage was 12 kV. The effect of separation voltage on the separation of QHCl and QTHCl was shown in Fig. 7D (curve R), the separation of the two analytes decreased with the increase of separation voltage and their resolution are greater than 1.5 when the separation voltage was 12 kV. Therefore, the 12 kV was selected as the separation voltage considering the intensities and resolution of the two analytes.

Fig. 6. CE-ECL electropherograms of QHCl and QTHCl. A: 0% n-propanol (a), 10% npropanol (b), 20% n-propanol (c), and 30% n-propanol (d) containing 1.0 µg/mL QHCl and 1.0 µg/mL QTHCl; B: 1.0 µg/mL QTHCl (e); blank plasma (f), plasma sample (g), c + 1 µg/mL QHCl and 1 µg/mL-QTHCl (h). Detection conditions: Silica-sol/Nano-ZnO/PVP/ Ru(bpy)32+ sensor as working electrode; Detection potential, 1.25 V; A: Electrokinetic injection for 10 kV×10 s; B: Electrokinetic injection for 12 kV×10 s; Separation voltage, 12 kV; A: Running buffer of 12 mmol/L PBS (pH 7.5) and detection cell solution of 50 mmol/L PBS (pH7.5); B: Running buffer of 14 mmol/L PBS (pH 8.0) and detection cell solution of 55 mmol/L PBS (pH8.0).

analytes was studied in isolation and it could be determined that peak 1 was QTHCl and peak 2 was QHCl.

3.10. Effect of injection voltage and injection time

3.6. Effect of detection potential

Injection voltage and injection time are other two important factors affecting ECL intensity and separation. The effect of injection voltage on ECL intensity and separation was investigated by changing injection voltage from 6.0kV to 16 kV while keeping injection time constant. As shown in Fig. S5, the ECL intensity of the two analytes were weaker when the injection voltage were 8–10 kV. When the injection voltage was 12 kV, both QHCl and QTHCl had stronger ECL intensity and it increased slowly when the injection voltage continued to increase. When the injection voltage was 12 kV, the two analytes could be separated completely and the resolution was above 1.5. Therefore, the 12 kV was selected as the injection voltage considering the sensitivity and separation of the two analytes. The effect of injection time on ECL intensity and resolution also was investigated. ECL intensity of the two analytes became stronger with the increasing of injection time. However, longer injection time causes the electrophoresis peaks to broaden. Considering the peak width and the peak shape, 10 s was selected as the optimized injection time, the resolution of the two analytes decreased with the injection time increasing and it was above 1.5 when the injection time was 10 s. Therefore, 10 s was selected as the optimum injection time considering the sensitivities and resolution of the two analytes.

The detection potential was the important factor influencing ECL intensity of the QHCl and QTHCl and the effect of different detection potential changed from 1.10 to 1.35 V on the ECL intensity of the two analytes were studied. As shown in Fig. S3, when the potential was 1.10 V and 1.15 V, the ECL intensity of the two analytes were weak and reached the maximum values at 1.20 V. When the detection potential is more than 1.20 V, water is markedly oxided to produce oxygen on the working electrode surface. The oxygen will block the contact between Ru(bpy)33+ and the analyte, therefore, ECL intensity decreases. So 1.20 V was selected as the optimum detection potential. 3.7. Effect of pH in the ECL cell The pH in the ECL cell would directly affect the ECL intensity because the reaction between Ru(bpy)32+ and analyte depends on the pH. Fixed the buffer solution concentration at 50 mmol/L, the effect of pH in the ECL cell on ECL intensity was investigated with pH value range from 5.5 to 9.0. As shown in Fig. 7A, the ECL intensity of the two analytes reached the maximum at pH 8.0. Therefore, the optimal pH value in the ECL cell was fixed at 8.0 in following experiments. The 217

Talanta 179 (2018) 213–220

S. Sun et al.

Fig. 7. Effects of detection buffer pH (A), buffer concentration in detection cell (B), running buffer pH (C) and separation voltage (D) on ECL intensity. Detection conditions of A: detection potential, 1.20 V; electrokinetic injection, 10 kV ×10 s; separation voltage, 12 kV; running buffer, 12 mmol/L PBS containing 20% n-propanol at pH 7.5; detection solution, 50 mmol/L PBS at pH7.5. Detection conditions of B: pH of detection solution, 8.0; other conditions as (A). Detection conditions of C: buffer concentration in detection cell for 55 mmol/L; other conditions as B. Detection conditions of D: running buffer, 14 mmol/L PBS at pH 8.0; other conditions as C. a. 2.0 μg/mL QHCl; b. 2.0 μg/mL QTHCl; R: Resolution between QHCl and QTHCl.

calibration curves of QHCl and QTHCl were shown in Fig. 8. The linearity ranges are from 0.007 to 8.0 μg/mL for QHCl and from 0.009 to 8.3 μg/mL for QTHCl. The regression equations were y = 484.6x + 36.84 for QHCl and y = 453.6x + 34.16 for QTHCl with the correlation coefficients (R2) 0.9986 and 0.9981, respectively. The detection limits (S/N=3) of QHCl and QTHCl were 3.6 ng/mL and 3.9 ng/mL, respectively. The comparison with other methods was shown in Table 1. As seen in Table 1, CE-ECL in the paper has the lowest LODs. A standard mixture solution containing 2.0 μg/mL QHCl and QTHCl each was consecutively determined six times. The RSDs of the peak height and the migration time were 3.3% and 3.1% for QHCl, 2.4% and 2.7% for QTHCl, respectively. The six replicates containing 1.0 µg/mL QHCl and 2.0 µg/mL QTHCl were investigated within 6 days. The experimental

3.11. Analytical characteristics According to Section 2.2, a series of standard mixture solutions containing QHCl and QTHCl were tested. The peak heights of QHCl and QTHCl showed linear relationship with their concentrations. The

Table 1 The LODs of different methods for measuring QHCl and QTHCl. Method

QHCl LOD (ng/mL)

HPLC -UV HPLC UHPLC-ES-MS/MS HPLC-MS/MS CE-ECL

Fig. 8. The calibration curves of QHCl and QTHCl a for curve of QHCl, b for curve of QTHCl.

218

10 60 5.010 5 3.6

QTHCl LOD (ng/mL)

Ref.

20 50 10.012 5 3.9

[26] [27] [28] [29] This Method

Talanta 179 (2018) 213–220

S. Sun et al.

results showed that the interday RSDs of peak heights were less than 3.7%. The experimental results indicated that the method for the determination of quinapril hydrochloride and quinaprilat hydrochloride has good reproducibility.

[4]

[5]

3.12. Sample analysis [6]

Under the optimal conditions, electropherograms obtained from the standard solution of 1.0 µg/mL QHCl (e), blank extract of human plasma (f), extract of human plasma after 1.0 h of oral administration (g) and spiked extract of human plasma with 1 µg/m L QHCl and 1 µg/ m L QTHCl (h) were illustrated in Fig. 6B. As shown in Fig. 6B, blank plasma had a weak ECL peak and the plasma samples had two strong peaks and a weak peak. The two strong peaks became stronger while the weak peak showed little change after being spiked with 1 µg/m L QHCl and 1 µg/m L QTHCl. It could be explained that the two strong peaks were produce by QHCl and QTHCl. Used curve (e) as references, the peak 1 represented QTHCl and the peak 2 represented QHCl. According to the peak height of curve (g), the concentrations of QHCl and QTHCl in human plasma were 0.276 µg/mL and 0.641 µg/mL, respectively. The recoveries of QHCl spiked with 0.250 and 0.500 µg/mL were 98.0% and 101%, respectively. The recoveries of QTHCl spiked with 0.500 and 1.00 µg/mL were 98.9% and 102%, respectively. The recoveries was roughly identical to recovery levels of analytes when only DI water using as a mixture standard. The results showed that the substrate in the samples did not interfere with the determination of QHCl and QTHCl. The RSDs of recoveries were less than 3.1% (n=6).

[7]

[8]

[9]

[10] [11]

[12] [13]

[14]

4. Conclusions

[15]

Using ZnO NPs, silica-sol, PVP and Ru(bpy)32+ as modifying reagents, a silica-sol/ZnO NPs/PVP/ Ru(bpy)32+ sensor was prepared and a new method for simultaneous ECL determination of QHCl and its metabolite QTHCl coupled with CE was developed using the silica-sol/ ZnO NPs/PVP/Ru(bpy)32+ sensor as working electrode and adding npropanol into the running buffer as separation additive. Compared with bare glassy carbon electrode, silica-sol/ZnO NPs/PVP/ Ru(bpy)32+ sensor as working electrodes had applied for the determination of QHCl and QTHCl in human plasma with satisfactory result. The sensor would be expected to be used for the determination and pharmacokinetics of other drugs which have ECL behavior with Ru(bpy)32+.

[16] [17]

[18] [19]

[20]

Acknowledgements

[21]

This work was financially supported by the National Natural Science Foundation of China (Grant numbers 21365006 and 21765004) and by the Guangxi Science Foundation of China (Grant numbers 2014GXNSFDA118004 and 1598025-4). The research fund of State Key Laboratory for Chemistry and Molecular Engineering of Medicinal Resources (Guangxi Normal University) (CMEMR2017-A5) is gratefully acknowledged.

[22]

[23]

[24]

Appendix A. Supporting information

[25]

Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2017.10.050.

[26]

References [27] [1] X. Yue, Z. Zhu, M. Zhang, Z. Ye, Reaction-based turn-on electrochemiluminescent sensor with a ruthenium(II) complex for selective detection of extracellular hydrogen sulfide in rat brain, Anal. Chem. 87 (2015) 1839–1845. [2] L. Liu, X. Wang, Q. Ma, Z. Lin, S. Chen, Y. Li, L. Lu, H. Qu, X. Su, Multiplex electrochemiluminescence DNA sensor for determination of hepatitis B virus and hepatitis C virus based on multicolor quantum dots and Au nanoparticles, Anal. Chim. Acta 916 (2016) 92–101. [3] S. Bozorgzadeh, B. Haghighi, L. Gorton, Fabrication of a highly efficient solid state

[28]

[29]

219

electrochemiluminescence sensor using Ru(bpy)32+ incorporated nano ZnOMWCNTs-nafion composite film, Electrochim. Acta 164 (2015) 211–217. W. Gao, Z. Liu, L. Qi, J. Lai, S.A. Kitte, G. Xu, Ultrasensitive glutathione detection based on lucigenin cathodic electrochemiluminescence in the presence of MnO2 nanosheets, Anal. Chem. 88 (2016) 7654–7659. Y. Sha, X. Zhang, W. Li, W. Wu, S. Wang, Z. Guo, J. Zhou, X. Su, A label-free multifunctionalized graphene oxide based electrochemiluminscence immunosensor for ultrasensitive and rapid detection of Vibrio parahaemolyticus in seawater and seafood, Talanta 147 (2016) 220–225. Y. Liu, H. Wang, C. Xiong, Y. Yuan, Y. Chai, R. Yuan, A sensitive electrochemiluminescence immunosensor based on luminophore capped pd@Au core-shell nanoparticles as signal tracers and ferrocenyl compounds as signal enhancers, Biosens. Bioelectron. 81 (2016) 334–340. L. Zhao, J. Li, Y. Liu, Y. Wei, J. Zhang, J. Zhang, X. Chen, A novel ECL sensor for determination of carcinoembryonic antigen using reduced graphene Oxide-BaYF 5:Yb, Er upconversion nanocomposites and gold nanoparticles, Sens. Actuators B 232 (2016) 484–491. T. Dong, L. Hu, K. Zhao, A. Deng, J. Li, Multiple signal amplified electrochemiluminescent immunoassay for brombuterol detection using gold nanoparticles and polyamidoamine dendrimers-silver nanoribbon, Anal. Chim. Acta 945 (2016) 85–94. W. Shen, Z. Li, H. Wang, Y. Liu, Q. Guo, Y. Zhang, Photocatalytic degradation for methylene blue using zinc oxide prepared by codeposition and sol-gel methods, J. Hazard. Mater. 152 (2008) 172–175. X. Niu, W. Du, W. Du, K. Jiang, Preparation of nano-ZnO and its gas sensitivity, Chin. J. Appl. Chem. 20 (2003) 968–971. S.A. Kumar, H. Cheng, S. Chen, Electroanalysis of ascorbic acid (vitamin C) using nano-ZnO/poly(luminol) hybrid film modified electrode, React. Funct. Polym. 69 (2009) 364–370. Z. Liu, Y. Liu, G. Shen, R. Yu, Nano-ZnO/chitosan composite film modified electrode for voltammetric detection of DNA hybridization, Anal. Lett. 41 (2008) 1083–1095. W. Sun, Z. Zhai, D. Wang, S. Liu, K. Jiao, Electrochemistry of hemoglobin entrapped in a Nafion/ nano-ZnO film on carbon ionic liquid electrode, Bioelectrochem 74 (2009) 295–300. X. Li, T. Ren, N. Wang, X. Ji, Gold nanoparticles-enhanced amperometric tyrosinase biosensor based on three-dimensional sol-gel film-modified gold electrodes, Anal. Sci. 29 (2013) 473–477. H. Peng, Z. Huang, Y. Zheng, W. Chen, A. Liu, X. Lin, A novel nanocomposite matrix based on graphene oxide and ferrocene-branched organically modified sol–gel/ chitosan for biosensor application, J. Solid State Electrochem. 18 (2014) 1941–1949. R.N. Dansby-Sparks, R. Ouyang, Z. Xue, Optical and electrochemical sol-gel sensors for inorganic species, Sci. Chin. 52 (2009) 1777–1788. Y. Li, Y. Luo, L. Li, H. Cheng, W. Huang, Preparation and application of electrochemiluminescence sensor by immobilizing tris (2, 2′-bipyridine) ruthenium (II) on the surface of gold electrode with silica sol/nano-Au/PVP/L-cysteine, Electrochem 83 (2015) 155–160. E.I. Morosanova, Silica and silica–titania sol-gel materials: synthesis and analytical application, Talanta 102 (2012) 114–122. P. Raghu, B.K. Swamy, T.M. Reddy, B.N. Chandrashekar, K. Reddaiah, Sol–gel immobilized biosensor for the detection of organophosphorous pesticides: a voltammetric method, Bioelectrochemistry 83 (2012) 19–24. A. Safavi, A.R. Banazadeh, Highly efficient and stable palladium nanoparticles supported on an ionic liquid silica Sol-Gel modified electrode, Electroanalysis 23 (2011) 1536–1542. P. Su, W. Shiu, M.S. Tsai, Flexible humidity sensor based on Au nanoparticles/ graphene oxide/thiolated silica sol-gel film, Sens. Actuators B 216 (2015) 467–475. S. Funda, E. Esin, S. Ramazan, The effect of quinapril treatment on insulin resistance leptin and high sensitive creactive protein in hypertensive patients, Clin. Exp. Hypertens. 33 (2011) 548–554. S.G. Kanorskiĭ, V.G. Tregubov, V.M. Pokrovskiĭ, Advantages of quinapril therapy in patients with arterial hypertension and functional class III chronic heart failure with preserved left ventricular ejection fraction, Kardiologiia 52 (2012) 31–37. J.L. Davis, K. Kruger, D.H. LaFevers, B.M. Barlow, J.M. Schirmer, B.A. Breuhaus, Effects of quinapril on angiotensin converting enzyme and plasma renin activity as well as pharmacokinetic parameters of quinapril and its active metabolite, quinaprilat, after intravenous and oral administration to mature horse, J. Exp. Psychol. Appl. 46 (2014) 729–733. M. Stolarczyk, A. Maslanka, A.N. Apola, J. Krzek, Determination of losartan potassium, quinapril hydrochloride and hydrochlorothiazide in pharmaceutical preparations using derivative spectrophotometry and chromatographic-densitometric method, Acta Pol. Pharm. 70 (2013) 967–976. C. Abbara, G. Aymard, S. Hinh, B. Diquet, Simultaneous determination of quinapril and its active metabolite quinaprilat in human plasma using high-performance liquid chromatography with ultraviolet detection, J. Chromatogr. B 766 (2002) 199–207. J.A. Prieto, R.M. Alonso, R.M. Jimenez, A. Blanco, Solid-phase extraction and highperformance liquid chromatography applied to the determination of quinapril and its metabolite quinaprilat in urine, J. Chromatogr. Sci. 39 (2001) 153–159. B. Dasandi, S. Shah, Determination of quinapril and quinaprilat in human plasma by ultraperformance liquid chromatography–electrospray ionization mass spectrometry, Biomed. Chromatogr. 23 (2009) 492–498. S.A. Parekh, A. Pudage, S.S. Joshi, V.V. Vaidya, N.A. Gomes, S.S. Kamat, Simultaneous determination of hydrochlorothiazide, quinapril and quinaprilat in human plasma by liquid chromatography-tandem mass spectrometry, J. Chromatogr. B 837 (2008) 59–69.

Talanta 179 (2018) 213–220

S. Sun et al.

Electrophoresis 23 (2002) 3683–3691. [35] X. Yin, E. Wang, Capillary electrophoresis coupling withelectrochemilumin- escence detection: a review, Anal. Chim. Acta 533 (2005) 113–120. [36] L. Guo, F. Fu, G. Chen, Capillary electrophoresis with electrochemilu- minescence detection: fundamental theory, apparatus, and applications, Anal. Bioanal. Chem. 399 (2011) 3323–3343. [37] Y. Wei, H. Wang, S. Sun, L. Tang, Y. Cao, B. Deng, An ultrasensitive electrochemiluminescence sensor based on reduced graphene oxide-copper sulfide composite coupled with capillary electrophoresis for determination of amlodipine besylate in mice plasma, Biosens. Bioelectron. 86 (2016) 714–719. [38] B. Deng, Y. Kang, X. Li, Q. Xu, Determination of erythromycin in rat plasma with capillary electrophoresis–electrochemiluminescence detection of tris(2,2′- bipyridyl) ruthenium (II), J. Chromatogr. B 857 (2007) 136–141. [39] Y. Wang, C. Zhang, S. Bi, G. Luo, Preparation of ZnO nanoparticles using the direct precipitation method in a membrane dispersion micro-structured reactor, Powder Technol. 25 (2010) 130–136.

[30] D. An, Z. Chen, J. Zheng, S. Chen, L. Wang, Z. Huang, L. Weng, Determination of biogenic amines in oysters by capillary electrophoresis coupled with electrochemiluminescence, Food Chem. 168 (2015) 1–6. [31] H. Guo, X. Wu, A. Wang, X. Luo, Y. Ma, M. Zhou, Separation and detection of tropane alkaloids in Anisodus tanguticus by capillary electrophoresis-electrochemiluminescence, New J. Chem. 39 (2015) 8922–8927. [32] S. Sun, C. Long, C. Tao, S. Meng, B. Deng, Ultrasonic microdialysis coupled with capillary electrophoresis electrochemiluminescence study the interaction between trimetazidine dihydrochloride and human serum albumin, Anal. Chim. Acta 851 (2014) 37–42. [33] S. Sun, Y. Wei, C. Long, B. Deng, Capillary electrophoresis with end-column electrochemiluminescence for ultrasensitive determination of urapidil hydrochloride in rat plasma and its application to pharmacokinetics study, J. Chromatogr. B 1006 (2015) 146–150. [34] W. Cao, J. Liu, X. Yang, E. Wang, New technique for capillary electro- phoresisdirectly coupled with end-column electrochemiluminescence detection,

220