1199

J. Sep. Sci. 2014, 37, 1199–1205

Dexian Kong1,2 Qinglu Li1 Lichan Chen2 Yuwu Chi2 Guonan Chen2 1 School

of Life Sciences, Fujian Agriculture and Forestry University, Fuzhou, Fujian, China 2 Key Laboratory of Analysis and Detection Technology for Food Safety (Ministry of Education and Fujian Province), College of Chemistry, Fuzhou University, Fuzhou, Fujian, China Received November 20, 2013 Revised February 23, 2014 Accepted February 24, 2014

Research Article

Capillary electrophoresis coupled with electrochemiluminescence detection for the separation and determination of thyreostatic drugs in animal feed A rapid, simple, and practical method for the determination of four of the most used thyreostatic drugs (methimazole, 2-thiouracil, 6-methyl-2-thiouracil, and 6-propyl-2-thiouracil) using CE coupled to electrochemiluminescence detection has been established, based on the electrochemiluminescence enhancement of tris(2,2-bipyridyl)ruthenium(II) with these analytes. Parameters that affect separation and detection were optimized. Under the optimum experimental conditions, the four analytes could be well separated within 11 min at the separation voltage of 16 kV in a running solution containing 20 mM phosphate buffer (pH 9.0) and 1.0 × 10−4 M Ru(bpy)3 2+ , with a solution of 20 mM phosphate buffer (pH 12.0) containing 1.0 × 10−4 M Ru(bpy)3 2+ in the electrochemiluminescence detection cell. The detection limits for methimazole, 6-methyl-2-thiouracil, 6-propyl-2-thiouracil, and 2thiouracil were 0.1, 0.05, 0.05, and 0.01 ␮M, respectively. The proposed method was applied to analyze these drugs in spiked animal feed samples. The recoveries were 88.299.0 and 86.498.7% for the intraday and interday analyses, respectively. The RSDs were 2.74.8 and 1.85.0% for the intraday and interday analyses, respectively. The results demonstrate that the proposed method has promising applications in the detection of thyreostatic drugs in animal feeds. Keywords: Bipyridine / Capillary electrophoresis / Electrochemiluminescence / ruthenium / Thyreostatic drugs DOI 10.1002/jssc.201301254

1 Introduction Thyreostatic drugs (TDs) are a class of substances that can be used to regulate the production of hormones by the thyroid gland [1]. They usually include methimazole (TAP), 2-thiouracil (TU), and TU derivatives, such as 6-methyl-2thiouracil (MTU) and 6-propyl-2-thiouracil (PTU). The chemical structures of TDs are shown in Fig. 1. Because of the effect of enhancing water retention in tissues and thus increasing the weight of animals before slaughter, TDs are now illegally used in animal feeds, leading to drug residues in the meats [2]. Once ingested, these drugs may inhibit the normal

Correspondence: Professor Yuwu Chi, Key Laboratory of Analysis and Detection Technology for Food Safety (Ministry of Education and Fujian Province), College of chemistry, Fuzhou University, Fuzhou, Fujian 350108, China E-mail: [email protected] Fax: +86-591-22866137

Abbreviations: CIPtWE, capillary-integrated Pt ring working electrode; ECL, electrochemiluminescence; MTU, 6-methyl2-thiouracil; PBS, phosphate/NaOH solution; PTU, 6-propyl2-thiouracil; TAP, methimazole; TDs, thyreostatic drugs; TU, 2-thiouracil

 C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

metabolism and lower the gastrointestinal motility of human beings. In addition, it has also been found that TDs readily incorporate into nucleic acids and thus display antivirus and antitumor activities [3–5]. Since residues of TDs in edible animal tissue can pose a risk to human health, their uses in animal production were banned in the European Union in 1981 [6]. However in China, the process is still on the way, so it is very important to find sensitive and dependable methods to detect them. Many methods have been found to analyze thyreostatic residues in different samples. These methods include highperformance thin layer chromatography [7], HPLC with UV [8], or electrochemical or MS detectors [9–11], GC with nitrogen–phosphorus [12], flame photometric [13], or MS detection [14]. Although modern instrumental analysis such as GC and HPLC are powerful, but there are some limitations, such as relatively high instrumental cost, complicated experimental procedures, and large sample and solvent consumption, which may increase environmental burden and time consumption. Recently, focus has been transferred to CE because it is one of the most powerful separation techniques providing an alternative choice to GC and HPLC in sample separation for its many merits, such as extremely high separation efficiency, ultra-small sample and solvent

www.jss-journal.com

1200

D. Kong et al.

J. Sep. Sci. 2014, 37, 1199–1205

method, we employed a gravity injection technique. The proposed method has been successfully applied to analyze TDs in spiked animal feed sample, and the results indicate that the developed method has the advantages of rapidity, high sensitivity, and good reproducibility.

2 Materials and methods 2.1 Reagents and solutions

Figure 1. Structural formulae of TDs

volume requirement, low time consumption and ease removal of contaminants [15–17]. Up to now, there are only four articles reporting the applications of CE for the determination of thyreostatics [18–21]. Krivankova et al. analyzed TDs such as TU, MTU, and PTU in urine samples of humans, cows, and horses by CE with UV detection [18]. EsteveRomero et al. applied a MEKC–UV system in the analysis of all TDs in manufactured dried animal feed [19]. Perez-Ruiz et al. employed CE coupled to LIF detection to detect TU and 6-phenyl-2-thiouracil in spiked feed samples and urine [20]. Recently, we developed a CE with electrochemical detection method for the determination of all TDs in animal feed [21], but as far as we know, electrochemiluminescence (ECL) detection has not been applied for the determination of TDs using CE. Tris(2,2-bipyridyl)ruthenium(II) (Ru(bpy)3 2+ ) ECL has been applied to analyze many compounds and ions as a simple and sensitive detection technique [22]. It provides another excellent means of detection and has been applied to analyze a large range of analytes of biological and clinical interest, such as amino acids and pharmaceutical preparations [23–25]. At the same time, CE–ECL based on Ru(bpy)3 2+ has been applied for the determination of various analytes containing amine groups and their derivatives [26–34]. In our lab, a flowinjection analysis system equipped with an ECL detector has been developed for the analysis of TU [35]. From the research work, we found that TDs could enhance the ECL intensity of Ru(bpy)3 2+ over a wide pH range (pH 4.0–12.0), giving us a inspiration that there is a possibility to employ a CE–ECL method in the analysis of TDs. In this paper, a CE system coupled to a homemade ECL detection cell [36] was applied to the separation and detection of four TDs. After studying the influences of the detection potential, Ru(bpy)3 2+ concentration, pH of running buffer, and separation voltage, optimum conditions for the analysis were established. In order to increase the sensitivity of the  C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Ru(bpy)3 Cl2 ·6H2 O was obtained from Aldrich. TAP, PTU, MTU, and TU were purchased from Sigma. Other chemicals were all analytically pure or better. Double-distilled water was used to prepare sample solutions. A 1.0 × 10−2 M stock solution of Ru(bpy)3 Cl2 was prepared by dissolving 0.0749 g of Ru(bpy)3 Cl2 ·6H2 O with 10 mL of double-distilled water, and the resultant mixture was stored in a refrigerator. Meanwhile, 1.0 × 10−2 M stock solutions of above four TDs were prepared by dissolving appropriate amounts of solid samples with double-distilled water and kept in a refrigerator too. Testing solutions were freshly prepared before experiments by diluting the stock solutions with appropriate running buffer solutions. The running buffer solutions were prepared by mixing H3 PO4 solutions with concentrated NaOH solution and pH values of the running buffer solutions were measured by a pH meter. Before electrophoresis experiments, all solutions were filtered through a 0.22 ␮m polypropylene filter film. 2.2 Apparatus and electrophoresis conditions All CE–ECL experiments were carried out with a commercial CE–ECL Analyzer System (model: MPI-A, Xi’an Remax Electronic and Technological, China). The CE–ECL system integrated a CE high-voltage power together with an electrochemical unit and an ECL unit, and included a data acquisition unit [29]. The separation capillary used was an unmodified fused-silica capillary with 55 cm length × 50 ␮m id × 360 mm od (Yongnian Optical Fiber Factory, Hebei, China). In this study, we used a unique ECL detection cell containing a capillary-integrated Pt ring working electrode (CIPtWE), which was designed in our previous research [36]. The diagram of the ECL cell and the undersurface structure of the CIPtWE are shown in Fig. 2. The ECL cell included a reservoir for ECL reactions, a CIPtWE, an Ag/AgCl reference electrode, a Pt auxiliary electrode, and a Pt cathode. The detection cell body was composed of a bigger Plexiglas block and a smaller Plexiglas block tightly fixed to each other with two screws. The smaller Plexiglas block (0.5 cm thick) was used as a transparent window and the bigger one was the main part of the detection cell. The detail fabrication procedure of CIPtWE can be found in ref. [36]. After the CIPtWE was fabricated, the separation capillary attached was flushed with 0.1 M NaOH for 4 h, and then sequentially flushed for 10 min with water, 0.1 M HCl, water, and the running buffer solutions. Between every five sample injections, the capillary www.jss-journal.com

J. Sep. Sci. 2014, 37, 1199–1205

Electrodriven Separations

1201

PTU, MTU, and TU, respectively) and ground with a mortar and pestle until a homogeneous mixture was obtained. Then, 0.1 g of the homogeneous sample was mixed with 10.0 mL methanol in a centrifuge tube, and mechanically shaken for 5 min. After centrifugation at 3600 rpm for 10 min, the supernatant solution was collected and evaporated to dryness at 40⬚C under a gentle stream of nitrogen. The dry residue was dissolved with 5 mL pentane for subsequent purification by SPE. An AccuBond II Florisil SPE Cartridge (1 g, Agilent Technologies) was previously conditioned with 10 mL of methanol, and then the 5 mL sample was passed through the Florisil SPE Cartridge. After rinsing the column with 5 mL pentane, elution was performed by 2 mL methanol/dichloromethane (20:80 v/v) and 5 mL methanol successively. The collected eluent was evaporated to dryness at 40⬚C under a gentle stream of nitrogen. Finally, the residue was dissolved with 2.0 mL water containing 1.0 × 10−4 M Ru(bpy)3 2+ to obtain a final sample solution. At the same time, 10 g of nonspiked animal feed samples were pretreated with the same procedure to get blank sample solutions.

Figure 2. Schematic diagram of ECL detection cell for CE. (A) The ECL detection system, (B) the undersurface structure of the CIPtWE.

should be activated and reproduced by flushing sequentially for 10 min with 0.1 M HCl, water, 0.1 M NaOH, and water. The Pt working electrodes for cyclic voltammetry and CE– ECL measurements were well polished with fine aluminum powder and cleaned in an ultrasonic bath. In CE–ECL experiments, 10 ␮L of all the four thyreostatic drug stock solutions were mixed and diluted in 5.00 mL 1.0 × 10−4 M Ru(bpy)3 2+ water solution to get standard solutions. A solution of 1.0 × 10−4 M Ru(bpy)3 2+ in 20 mM phosphate buffer (pH 12.0) was directly injected into the reaction reservoir. Running buffer solution was 20 mM phosphate buffer (pH 9.0) containing 1.0 × 10−4 M Ru(bpy)3 2+ . Samples were injected with gravity injection, in which we placed the sample cell to a height of 45 cm (versus the ECL cell) for 40 s. The separation voltage was 16 kV. The photomultiplier tube was biased at −800 V. The four analytes can be separated satisfactorily in 11 min.

2.3 Determination of TDs in spiked animal feeds The extraction procedure of these drugs used here was a modification of the procedure reported in our previous work [21, 35]. First, 10 g of animal feed samples were spiked, respectively, with TAP, PTU, MTU, and TU at the given concentration levels (level 1: 0.1 ␮mol/g for all drugs; level 2: 0.2, 0.4, 0.4, and 0.3 ␮mol/g for TAP, PTU, MTU, and TU, respectively, level 3: 0.3, 0.8, 0.8, and 0.5 ␮mol/g for TAP,  C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

3 Results and discussion 3.1 Optimization of experimental conditions 3.1.1 ECL behavior of TDs in the presence of Ru(bpy)3 2+ TDs include TAP and thiourea pyrimidine compounds such as TU, MTU, and PTU (Fig. 1). Since the molecular structures of TU, MTU, and PTU are quite similar to each other but rather different from that of TAP, we took TAP and TU as the model analytes to study the ECL behaviors of TDs in the presence of Ru(bpy)3 2+ . On a Pt disk electrode, Ru(bpy)3 2+ in 0.1 M and pH 12.0 phosphate buffer solution began to exhibit ECL activity at +1.0 V, and the ECL intensity increased to 275 counts when the potential was close to +1.2 V (Fig. 3a). When TAP and TU were added separately into the Ru(bpy)3 2+ solutions, the ECL intensities were enhanced to about 1618 counts (Fig. 3b) and 1795 counts (Fig. 3c), respectively. These results indicated that TDs could react with Ru(bpy)3 2+ in the ECL process and enhance the ECL intensity of Ru(bpy)3 2+ . It is well known that most ECL responses of organic species in Ru(bpy)3 2+ system result from the reaction of strongly reducing intermediates (usually neutral radical species produced by electrochemical oxidation of the organics) with Ru(bpy)3 3+ (electrogenerated from Ru(bpy)3 2+ ) to generate the excited-state luminophore, Ru(bpy)3 2+ * that gives light emission [37, 38]. The ECL reactions of these drugs in the presence of Ru(bpy)3 2+ probably include a similar neutral radical mechanism. All of the four TDs studied are sulfhydryl compounds, they are easily oxidized to form highly reducing radicals (labeled as RS·) [39–44]. The oxidation peak potentials on a Pt disk electrode are +0.25, +0.8, +0.45, and www.jss-journal.com

1202

J. Sep. Sci. 2014, 37, 1199–1205

D. Kong et al.

3.1.3 Effect of Ru(bpy)3 2+ concentration

Figure 3. Dependence of the ECL intensity on cyclic voltammetry of 1.0 × 10−4 M Ru(bpy)3 2+ alone (a, solid line) and in the presence of 1.0 × 10−4 M TAP (b, dashed line), and 1.0 × 10−4 M TU (c, dotted line) in phosphate buffer (pH 12.0). Scan rate: 100 mV/s; photomultiplier tube biased at −800 V.

+0.5 V for TAP, TU, PTU, and MTU, respectively [21]. These oxidation potentials of TDs are much lower than that of Ru(bpy)3 2+ , +1.2 V [45]. As shown in Fig. 3, the ECL intensities of Ru(bpy)3 2+ systems in the absence (Fig. 3a) and presence of TDs (Fig. 3b and c) all reach to their maximum values at a potential of +1.2 V, where both RS· and Ru(bpy)3 3+ are produced. So, the possible mechanism for the ECL behavior of TDs in the Ru(bpy)3 2+ system may be described as follows: RSH − e− − H+ → RS·

(1)

3+ − Ru(bpy)2+ 3 − e → Ru(bpy)3

(2)

2+∗ · + product Ru(bpy)3+ 3 + RS → Ru(bpy)3

(3)

→ Ru(bpy)2+ Ru(bpy)2+∗ 3 3 + hv

(4)

3.1.2 Effect of detection voltage The ECL responses of TDs were strongly dependent on the voltage of the potential applied at the PtWE. In order to select an optimum detection potential, we investigated the ECL responses of TDs under different potentials. The results indicated that all analytes could enhance the ECL intensity of Ru(bpy)3 2+ visibly when the applied voltage was >+1.20 V. The enhanced ECL intensity (⌬I) increased with the detection potential in the range from +1.20 to +1.40 V. However, the baseline noise increased greatly, whereas ⌬I rose slightly when potential was >+1.40 V, causing a significant decrease in the ratio of S/N. Therefore, +1.40 V was selected as the detection potential in subsequent experiments for obtaining a suitable compromise of high sensitivity and low background intensity.  C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

In the study, Ru(bpy)3 2+ was used as the ECL reagent and it was supplied both in the ECL cell and running buffer. The previous report had demonstrated that the concentration of Ru(bpy)3 2+ had a great influence on the ECL detection sensitivity [31]. Thus, the influence of Ru(bpy)3 2+ concentration was studied when the concentration of all four analytes were fixed at 2.0 × 10−5 M. First, we fixed the concentration of Ru(bpy)3 2+ in the ECL cell to be 1.0 × 10−4 M and changed the Ru(bpy)3 2+ concentration in running buffer. When there was no Ru(bpy)3 2+ in running buffer, we could not see any ECL response on the electropherogram. When the Ru(bpy)3 2+ concentration increased from 1.0 × 10−5 to 1.0 × 10−4 M, the ECL intensities of all analytes increased. But at the same time, further increase in Ru(bpy)3 2+ concentration induced an obvious background noise. In this work, 1.0 × 10−4 M Ru(bpy)3 2+ in phosphate buffer was adopted for considering both detection sensitivity and economy in the use of reagents. Likewise, when we fixed the running buffer’s Ru(bpy)3 2+ concentration and changed the concentration of Ru(bpy)3 2+ in the ECL cell, the results were as mentioned above. Considering the sensitivity, consumption of reagents and background noise, we chose 1.0 × 10−4 M Ru(bpy)3 2+ in the ECL cell too.

3.1.4 Effect of running buffer Separation of TDs was studied in different buffer systems including phosphate/NaOH solution (PBS), acetate, Tris-HCl, and borate buffers. Finally, PBS was chosen in terms of the stable baseline, lower noise, shorter analysis time, and better peak shape. Effects of pH and concentration of the PBS running buffer on electrophoresis were investigated. We studied the separation degree of the four analytes when the pH changed from 5.0 to 11.0 at intervals of 1.0 pH unit. When the pH of the running buffer was too low (pH < 7.0), the separation resolution of the four analytes was poor. When the pH was increased to 8.0, the four analytes could be separated completely. Subsequently, when the pH was changed from 8.0 to 11.0, the separation resolution remained almost unchanged, and it was quite different with the result we got when we employed CE coupled to electrochemical detector to analyze those drugs in our previous study [21]. In the previous study, the running buffer pH should be settled exactly at pH 9.20, whereas in this work, the pH of running buffer could be freely changed from 8.0 to 11.0. In our opinion, it may be ascribed to the absorption of positive Ru(bpy)3 2+ onto the inner wall of the capillary or Ru(bpy)3 2+ , which formed ion pairs with negative analytes and thus changed the separation mechanism of the analytes. Although the exact mechanism requires further study, the separation could be conducted in a wider pH range. In the following study, we chose the pH of the running buffer to be 9.0. www.jss-journal.com

J. Sep. Sci. 2014, 37, 1199–1205

Electrodriven Separations

1203

After the pH of running buffer was defined, we investigated the concentration of phosphoric acid in the running buffer. When the concentration of phosphoric acid changed from 10 to 30 mM, the ECL intensity of these drugs changed slightly, but the migration time was increased obviously with the increase of phosphoric acid concentration. Furthermore, too high phosphate concentration (>20 mM) might cause a significant increase of Joule heat and thus decrease the separation resolution and make the baseline unstable. Accordingly, 20 mM phosphate buffer (pH 9.0) was chosen in order to get the minimum analysis time and best ECL sensitivity. 3.1.5 Effect of separation voltage Separation voltage simultaneously had an impact on migration time and ECL intensity. By adjusting the separation voltage from 10 to 20 kV with an interval of 2 kV, it was found that high separation voltage could shorten the analysis time. However, the separation voltage >16 kV could result in poor resolution and decrease in ECL intensity. Taking both the resolution and ECL detection sensitivity into account, we selected 16 kV as the optimum separation voltage. 3.1.6 Effect of injection method and injection time In order to increase the sensitivity of the method, besides ordinary electrokinetic injection, we studied some special injection methods such as electrical field magnified injection and gravity injection. In terms of electric discrimination to the analytes, the first two injection methods produced low ECL intensity, and then the gravity injection method was chosen. We placed the sample cell to a height (versus ECL cell) of 45 cm for 40 s to achieve the best ECL intensity and peak shape for all four analytes. Moreover, the effect of sample solution composition on injection was investigated. Sample solutions of analytes were prepared by diluting stock solutions with 1.0 × 10−4 M Ru (bpy)3 2+ aqueous solutions containing: 0, 2, 5, and 10 mM phosphate (pH 9.0). The injection of sample solutions diluted with 1.0 × 10−4 M Ru(bpy)3 2+ aqueous solution (without phosphate) resulted in the highest ECL intensity. Therefore, 1.0 × 10−4 M Ru(bpy)3 2+ aqueous solution was used for preparing sample solutions. Finally, a typical electropherogram for a standard mixture solution of TAP, PTU, MTU, and TU under the abovementioned optimum conditions was shown in Fig. 4A, which indicated that a baseline separation of all the four analytes was achieved within 11 min.

3.2 Repeatability, linearity, and detection limits A standard mixture solution containing 2 ␮M of each analyte was analyzed seven times to estimate the repeatability of the ECL intensity and migration time for the analytes under the optimum experimental conditions mentioned above.

 C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Figure 4. Electropherogram of standard mixture solution (20 ␮M of for all analytes; A) and extract of spiked sample (final concentration of 5.0 ␮M for all analytes) and blank sample (B). Fusedsilica capillary: 50 ␮m id × 55 cm length; running buffer: 20 mM PBS (pH 9.0) containing 1.0 × 10−4 M Ru(bpy)3 2+ ; ECL cell solution: 1.0 × 10−4 M Ru(bpy)3 2+ in 20 mM PBS (pH 12.0); working potential: +1.4 V (versus Ag/AgCl); separation voltage: 16 kV; injection method and time: gravity injection/40 s.

The results showed that the RSDs of ECL intensity and migration time were 6.2 and 1.02% (for TAP), 7.8 and 2.1% (for PTU), 5.4 and 2.5% (for MTU), 6.4 and 1.7% (for TU), respectively. A series of the standard mixture solutions were tested to determine the response linearity of the four analytes under the optimum experimental conditions. The results of regression analysis on calibration curves and detection limits (S/N = 3) are summarized in Table 1. The calibration curves exhibited excellent linearity over the concentration range of about three orders of magnitude. LODs of 0.1 ␮M (11.4 ng/mL) for TAP, 0.05 ␮M (8.5 ng/mL) for PTU, 0.05 ␮M (7.1 ng/mL) for MTU, and 0.01 ␮M (1.3 ng/mL) for TU were found.

www.jss-journal.com

1204

J. Sep. Sci. 2014, 37, 1199–1205

D. Kong et al.

Table 1. Results of regression analysis on calibration and the detection limitsa)

Compound

Regression equation y = a + bxb)

Correlation coefficient

Linear range (␮M)

Detection limitc) (␮M (ng/mL))

TAP PTU MTU TU

y = 222 x + 65 y = 368x + 141 y = 378.8 x + 184 y = 495.5x + 172

0.9953 0.9964 0.9950 0.9991

120 0.550 0.550 0.230

0.1(11.4) 0.05(8.5) 0.05(7.1) 0.01(1.3)

a) Working potential: +1.40V (versus Ag/AgCl); other conditions as in Fig. 4. b) Here, y and x are the ECL intensity (counts) and concentration of the analytes (␮M), respectively. c) The detection limits corresponding to concentrations giving S/N of 3.

3.3 Recovery studies in animal feed samples To evaluate the applicability in real samples, the presently developed CE–ECL method was applied to the determination of these drugs in spiked animal feed samples under the optimized conditions. The recovery experiments were explored for the samples with three different concentration levels, i.e., 5.0, 10.0, and 15.0 ␮M for TAP, 5.0, 20.0, and 40.0 ␮M for PTU, 5.0, 20.0, and 40.0 ␮M for MTU, and 5.0, 15.0, and 25.0 ␮M for TU, respectively. After being purified by SPE, the extracted samples were tested by the proposed method. One of the electropherograms for spiked sample (5.0 ␮M for each drug, respectively) and blank sample is shown in Fig. 4B. It is clear that these drugs were well separated and detected, and significant matrix effects were not observed in the determination. Table 2 shows the recoveries and RSDs of intraday and interday determination of spiked animal feed sample. The intraday investigation was performed by analyzing three parallel spiked samples with an interval of 2 h in one day, while the interday analysis was performed by analyzing spiked samples each day on three consecutive days. Each spiked sample (three concentration levels) was pretreated as described in Section 2.3, and each sample was repeated six times. The re-

coveries were calculated using the regression equation from calibration mentioned in Section 3.2. The recoveries were ranging from 88.2 to 99.0 and 86.4 to 98.7% for intraday and interday analyses, respectively, and the RSDs for the intraday and interday analysis were 2.74.8 and 1.85.0%, respectively. The recovery data obtained using the pretreatment method mentioned in Section 2.3 were acceptable according to the established criteria of validation and harmonization of analytical methods for veterinary drug residues [46].

4 Concluding remarks In this study, we demonstrated that the use of CE coupled to Ru(bpy)3 2+ ECL detection for the separation and analysis of TDs was possible. The new method showed good performance in terms of selectivity, sensitivity, simplicity, short analysis time, and good linearity. The proposed method gave LODs (S/N = 3) of 0.010.1 ␮M (1.311.4 ng/mL). Compared with other CE reports concerning TDs analysis, the LODs of present CE–ECL were obviously lower than those of CE–UV (50 ng/mL) [18], MEKC–UV (250400 ng/mL) [19], and CE–ED (660 ng/mL) [20], but were a little higher than those of the CE–LIF method (0.30.43 ng/mL) [21]. The

Table 2. The recoveries of TDs under different concentrations

Compound

TAP

PTU

MTU

TU

Sample concentration (␮M)

5.0 10.0 15.0 5.0 20.0 40.0 5.0 20.0 40.0 5.0 15.0 25.0

Intraday

Inter-day

Founda) (␮M)

Recovery (%)

RSD (%)

Founda) (␮M)

Recovery (%)

RSD (%)

4.41 ± 0.14 8.98 ± 0.36 14.8 ± 0.71 4.67 ± 0.19 18.66 ± 0.56 36.7 ± 1.15 4.82 ± 0.21 19.6 ± 0.81 37.4 ± 1.58 4.63 ± 0.17 14.98 ± 0.35 24.02 ± 0.66

88.2 89.8 98.7 93.4 93.3 91.8 96.4 98.0 93.5 92.6 99.9 96.1

3.2 4.0 4.8 4.1 3.0 3.1 4.4 4.1 4.2 3.7 2.3 2.7

4.32 ± 0.17 9.11 ± 0.45 14.6 ± 0.54 4.74 ± 0.22 18.75 ± 0.34 39.2 ± 1.95 4.79 ± 0.12 19.7 ± 0.76 37.4 ± 1.39 4.68 ± 0.21 14.8 ± 0.31 23.94 ± 0.76

86.4 91.1 97.3 94.8 93.8 98.0 95.8 98.5 93.5 93.6 98.7 95.8

3.9 4.9 3.7 4.6 1.8 5.0 2.5 3.9 3.7 4.5 2.1 3.2

a) Mean values of 18 determinations.

 C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.jss-journal.com

J. Sep. Sci. 2014, 37, 1199–1205

method showed its usefulness in spiked animal feed with satisfactory recoveries. These results suggest that the proposed method has promising applications in the detection of TDs in animal feed samples. This study was financially supported by National Natural Science Foundation of China (21375020), Specialized Research Fund for the Doctoral Program of Higher Education (20133514110001), Program for New Century Excellent Talents in Chinese University (NCET-10–0019), and the Program for Changjiang Scholars and Innovative Research Team in University (no. IRT1116). The authors have declared no conflict of interest.

5 References

Electrodriven Separations

1205

[19] Esteve-Romero, J., Escrig-Tena, I., Simo-Alfonso, E. F., Ramis-Ramos, G., Analyst 1999, 124, 125–128. ´ [20] Perez-Ruiz, T., Mart´ınez-Lozano, C., Sanz, A., Galera, R., Electrophoresis 2005, 26, 2384–2390. [21] Kong, D., Chi, Y., Chen, L., Dong, Y., Zhang, L., Chen, G., Electrophoresis 2009, 30, 3489–3495. [22] Li, J., Wang, E., Chem. Rec. 2012, 12, 177–187. ˜ A. M., Velasco, A. I., Elec[23] Lara, F. J., Garc´ıa-Campana, trophoresis 2010, 31, 1998–2027. [24] Miao, W., Chem. Rev. 2008, 108, 2506–2553. [25] Du, Y., Wang, E., J. Sep. Sci. 2007, 30, 875–890. [26] Liu, Y.-M., Cao, J.-T., Zheng, Y.-L., Chen, Y.-H., J. Sep. Sci. 2008, 31, 2463–2469. [27] Fu, Z., Wang, L., Li, C., Liu, Y., Zhou, X., Wei, W., J. Sep. Sci. 2009, 32, 3925–3929. [28] Deng, B., Xie, F., Li, L., Shi, A., Liu, Y., Yin, H., J. Sep. Sci. 2010, 33, 2356–2360.

[1] Cooper, D. S. N., Engl. J. Med. 2005, 352, 905–917.

[29] Sun, H., Li, L., Su, M., J. Clin. Lab. Anal. 2010, 24, 327– 333.

[2] Owen, N. V., Worth, H. M., Kiplinger, G. F., Food Cosmet. Toxicol. 1973, 11, 649–653.

[30] Zhou, M., Li, Y.-J., Ma, Y.-J., Wang, W.-F., Mi, J., Chen, H., Luminescence 2011, 26, 319–323.

[3] Yu, M.-Y. W., Sedlak, J., Lindsay, R. H., Arch. Biochem. Biophys. 1973, 155, 111–119.

[31] Bao, Y., Yang, F., Yang, X., Electroanalysis 2012, 24, 1597– 1603.

[4] Lindsay, R. H., Romine, C. J., Wong, M. Y., Arch. Biochem. Biophys. 1968, 126, 812–820.

[32] Wang, J., Zhang, X., Pi, F., Wang, X., Yang, N., Electrochim. Acta 2009, 54, 2379–2384.

[5] Bretner, M., Kulikowski, T., Dzik, J. M., Balinska, M., Rode, W., Shugar, D., J. Med. Chem. 1993, 36, 3611–3617.

[33] Xu, Y., Gao, Y., Wei, H., Du, Y., Wang, E., J. Chromatogr. A 2006, 1115, 260–266.

[6] E.C. Directive 81/602, Publications of the European Communities, Document No. L 222 32–33, August 7, 1981.

[34] Wang, J., Peng, Z., Yang, J., Wang, X., Yang, N., Talanta 2008, 75, 817–823.

[7] De Wasch, K., De Brabander, H. F., van Ginkel, L. A., Spaan, A., Sterk, S. S., Meiring, H. D., J. Chromatogr. A 1998, 819, 99–111.

[35] Chi, Y., Duan, J., Lin, S., Chen, G., Anal. Chem. 2006, 78, 1568–1573.

[8] Buick, R. K., Barry, C., Traynor, I. M., McCaughey, W. J., Elliott, C. T., J. Chromatogr. B 1998, 720, 71–79. [9] Abuinb, S., Centrich, F., Rubies, A., Companyo, R., Prat, M. D., Anal. Chim. Acta 2008, 617, 184–191. [10] Lohmus, M., Kallaste, K., Le Bizec, B., J. Chromatogr. A 2009, 1216, 8080–8089. [11] Wozniak, B., Zuchowska, I. M., Zmudzki, J., Jedziniak, P., Korycinska, B., Sielska, K., Witek, S., Klopot, A., Anal. Chim. Acta 2011, 700, 155–166. [12] Schilt, R., Weseman, J. M., Hooijerink, H., Korbee, H. J., Traag, W. A., Van Steenbergen, M. J., Haasnoot, W., J. Chromatogr. B 1989, 489, 127–137. [13] Laitem, L., Gaspar, P., J. Chromatogr. A 1977, 140, 266– 269. [14] Liu, Y., Zou, Q.-H., Xie, M.-X., Han, J., Rapid Commun. Mass Spectrom. 2007, 21, 1504–1510. [15] Boone, C. M., Waterval, J. C., Lingeman, H., Ensing, K., Underberg, W. J., J. Pharm. Biomed. Anal. 1999, 20, 831– 863. [16] Watzig, H., Dette, C., Pharmazie 1994, 49, 83–96. ˜ [17] Castaneda, G., Rodr´ıguez-Flores, J., R´ıos, A., J. Sep. Sci. 2005, 28, 915–924. [18] Krivankova, L., Krasensky, S., Bocek, P., Electrophoresis 1996, 17, 1959–1963.

 C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

[36] Chen, G., Chi, Y., Wu, X., Duan, J., Li, N., Anal. Chem. 2003, 75, 6602–6607. [37] Richter, M. M., Chem. Rev. 2004, 104, 3003–3036. [38] Yin, X.-B., Dong, S., Wang, E., TrAC, Trends Anal. Chem. 2004, 23, 432–441. [39] Aslanoglu, M., Peker, N., J. Pharm. Biomed. Anal. 2003, 33, 1143–1147. [40] Xi, X., Ming, L., Liu, J., J. Appl. Electrochem. 2010, 40, 1449–1454. [41] Goyal, R. N., Singh, U. P., Abdullah, A. A., Bioelectrochemistry 2005, 67, 7–13. [42] Po, H. N., Lo, C.-F., Jones, N., Lee, R. W., Inorg. Chim. Acta 1980, 46, 185–189. [43] Holzer, K. P., Wrona, M. Z., Bioelectrochem. Bioenerg. 1983, 10, 199–212. [44] Holzer, K. P., Wrona, M. Z., Bioelectrochem. Bioenerg. 1983, 11, 3–13. [45] Wu, X., Huang, F., Duan, J., Chen, G., Talanta 2005, 65, 1279–1285. [46] Ellis, R. L., Validation and harmonization of analytical methods for residue detection at the international level, in: Moats W.A., Medina M.B. (Eds.), Veterinary Drug Residues. Food Safety, ACS Symposium Series 636, Washington, DC, 1995, pp. 41–62.

www.jss-journal.com

Capillary electrophoresis coupled with electrochemiluminescence detection for the separation and determination of thyreostatic drugs in animal feed.

A rapid, simple, and practical method for the determination of four of the most used thyreostatic drugs (methimazole, 2-thiouracil, 6-methyl-2-thioura...
202KB Sizes 2 Downloads 3 Views

Recommend Documents