1824 Journal of Food Protection, Vol. 77, No. 10, 2014, Pages 1824–1829 doi:10.4315/0362-028X.JFP-14-103 Copyright G, International Association for Food Protection
Research Note
Lateral-Flow Assay for Rapid Quantitative Detection of Clorprenaline Residue in Swine Urine TAO PENG,1 FU S. ZHANG,2 WAN C. YANG,3 DONG X. LI,2 YUAN CHEN,3 YONG H. XIONG,1 HUA WEI,1 AND WEI H. LAI1* 1State
Key Laboratory of Food Science and Technology, Nanchang University, 235 Nanjing East Road, Nanchang 330047, People’s Republic of China; Agricultural Product Quality Safety and Inspection Center, 151 Wenjiao Road, Nanchang 330077, People’s Republic of China; and 3Jiangxi Zodolabs Bioengineering Co., Ltd., 235 Nanjing East Road, Nanchang 330047, People’s Republic of China
2Jiangxi
MS 14-103: Received 2 March 2014/Accepted 31 May 2014
ABSTRACT Clorprenaline (CLP), a b2-adrenergic agonist, was first found in veterinary drugs for cold treatment in China in 2013. It is a potential new lean meat-boosting feed additive because it can promote animal muscular mass growth and decrease fat accumulation. A competitive colloidal gold-based lateral flow immunoassay system with a portable strip reader was successfully developed for rapid quantitative detection of CLP residue in swine urine. The detection system was optimized so that the detection can be completed within 9 min with a limit of detection of 0.15 mg?liter21. The assay exhibited good linear range from 3.0 to 20.0 mg?liter21, with reliable correlation of 0.9970 and with no obvious cross-reaction with five other b2-agonist compounds. Twenty spiked swine urine samples were tested by lateral flow immunoassay and liquid chromatography–tandem mass spectrometry to confirm the accuracy of the system. Results show good correlation between the two methods. This method is rapid, sensitive, specific, and convenient. It can be applied in the field for on-site detection of CLP in urine samples.
Clorprenaline (CLP) [1-(2-chlorophenyl)-2-propan-2ylamino-ethanol] belongs to the b2-adrenergic agonist group. CLP has been clinically used for the treatment of bronchial asthma and asthmatic bronchitis (4, 10). In 2007, Cao et al. warned about CLP hydrochloride being illegally used as new growth promoter in livestock production (2). Like other b2-adrenergic agonists, CLP has been proven to function in nutrient redistribution in the metabolism of animals at higher dosage levels (11). Due to lack of public concern, the use of CLP as a lean meat-boosting additive has evaded the examination of authorities. After eating meat products containing CLP residues, consumers have exhibited various symptoms of poisoning, such as muscular tremors, weakness of limbs, fever, headache, and nausea (15). In 2013, Chinese police injuncted a pharmaceutical company that used harmful additives in animal medicines. The Ministry of Public Security stated that reports have been made by the agricultural authorities of Haiyan County of East China’s Zhejiang province about many of the county’s swine being tested for CLP hydrochloride, a banned lean meat-boosting additive (1). Current methods for detecting CLP, such as highpressure liquid chromatography (15), capillary zone electrophoresis (17), and near-infrared spectroscopy (16) are time consuming and also require expensive instruments and complicated operation. * Author for correspondence. Tel: 0086-791-83969526; Fax: 0086-79188157619; E-mail: [email protected].
Colloidal gold-based lateral-flow immunochromatographic assay (LFIA), which is presently widely applied in the area of food safety (5, 6), medical diagnosis (3), and environment monitoring (9), allows rapid, sensitive, stable, convenient, inexpensive, and user-friendly determination of analytes (13). Our study pioneered the development of a colloidal gold-based LFIA system, involving the use of a portable strip reader for the rapid and quantitative detection of CLP residues in swine urine. MATERIALS AND METHODS Materials. CLP, terbutaline, salbutamol, ritodrine, fenoterol, bovine serum albumin (BSA; fraction V), and goat anti-mouse antibody were obtained from Sigma (St. Louis, MO). Ractopamine was obtained from Dr. Ehrenstorfer GmbH (Augsburg, Germany). CLP-BSA conjugate antigen and anti-CLP monoclonal antibody were provided by Wuxi Zodoboer Biotech. Co., Ltd. (Wuxi, People’s Republic of China). Hydrogen tetrachloroaurate trihydrate was obtained from Aldrich (Milwaukee, WI). The sample pad, conjugate release pad, nitrocellulose membrane, and absorbent pad were purchased from Millipore, Inc. (Bedford, MA). All solvents and other chemicals were of analytical reagent grade. The BioDot XYZ Platform, which combines motion control with the BioJet Quanti3050k dispenser and AirJet Quanti3050k dispenser, was acquired from BioDot (Irvine, CA). The portable strip reader was introduced in our previous study (8). Preparation of colloidal gold. One milliliter of 1% (wt/vol) stock solution of hydrogen tetrachloroaurate trihydrate was added to 100 ml (vol) of distilled water and heated to boiling point. Then,
J. Food Prot., Vol. 77, No. 10
RAPID QUANTITATIVE DETECTION OF CLORPRENALINE RESIDUE
1825
FIGURE 1. The flow chart of colloidal gold-based lateral-flow immunoassay system for rapid quantitative detection of CLP. A 100-ml urine sample without pretreatment was pipetted into the ELISA well, incubated for 3 min, and added to the sample well of the lateral flow strip. After 6 min, the strip was scanned with the portable strip reader; the result is shown in the data display window.
1.3 ml of freshly prepared 1% sodium citrate solution was added to the gold solution under constant stirring. When the color of the mixture turned red, the solution continued boiling for another 5 min; the colloidal gold solution was obtained and stored at 4uC. Preparation of colloidal gold-monoclonal antibody. The colloidal gold-monoclonal antibody (CG-mAb) was prepared as follows: 10 ml of colloidal gold solution was adjusted to pH 6.0 with 0.2 M K2CO3. With gentle stirring, 1 ml of anti-CLP monoclonal antibody solution was added dropwise to the colloidal gold solution to a final concentration of 1.2 mg?ml21, after which the mixture was agitated for 60 min. Then, 1 ml of 1% (wt/vol) polyethylene glycol20000 solution was added to the mixture, which was agitated for 30 min. Afterward, 1 ml of BSA (10%, wt/vol) was added for further blocking for 30 min. The mixture was centrifuged at 8,000 g at 4uC for 30 min to remove the unlabeled free antibody. The resulting precipitate was dissolved in 1 ml of dilution buffer, and then 1.5 ml of the prepared CG-mAb was added to an enzyme-linked immunosorbent assay (ELISA) well and dried at 30uC for 2.5 h. Preparation of immunochromatographic test strips. The sample pad, which was treated with 50 mM borate buffer (pH 7.4, containing 1% BSA, 0.5% Tween-20, and 0.05% sodium azide), was dried at 60uC for 2 h. The CLP-BSA conjugation (0.10 mg?ml21) and rabbit anti-mouse antibody (1.0 mg?ml21) were applied to the test and control lines on the nitrocellulose membrane, respectively, and dried at 35uC for 12 h. The nitrocellulose membrane, absorption pad, glass fiber membrane, and pretreated sample pad were assembled as the test strip (Fig. 1). Lateral flow test procedure. As illustrated in the flow chart (Fig. 1), a 100-ml urine sample without pretreatment was pipetted into the ELISA well and incubated for 3 min. The sample solutionCG-mAb reaction system was added to the sample well of the lateral flow strip. After 6 min, the strip was scanned with the portable strip reader. The color intensity of test line (AT), control line (AC), and ratio of AT to AC (T/C) were recorded. Optimization of the reaction conditions for CLP detection. The colloidal gold lateral-flow immunoassay was optimized by varying the following parameters: amount of labeled antibody (1.0, 1.2, and 1.5 mg?ml21); volume of CG-mAb in the ELISA well (1.5, 2.5, and 4 ml); concentration of CLP-BSA on the test line (0.1,
0.15, and 0.2 mg?ml21); and reaction time. The optimum conditions were obtained from each experiment, which was performed with five replications. Quantitative calibration curve of the LFIA system. Forty swine urine samples that were ascertained to be free of CLP by liquid chromatography–tandem mass spectrometry (LC-MS/MS) were collected from Jiangxi province and mixed as negative sample. Portions of the sample mixture were spiked with CLP at concentrations of 1.0, 3.0, 5.0, 7.0, 10.0, 15.0, and 20.0 mg?liter21. The T/C ratio of negative control and positive samples were renamed as B0 and BX. The calibration curve was constructed by plotting the BX/B0 ratio against the logarithm of CLP concentrations. Information of the linear regression equation was established for quantitative analysis. LC-MS/MS method. One milliliter of ammonium acetate (2 mol?liter21) was added to 1 ml of the swine urine sample, followed by addition of 15 ml of the enzyme agent. After oscillating in a water bath at 37uC for 16 h, 5 ml of ion-pair reagent was added into the mixture. The solution was then transferred to an OG column, which was preconditioned with 3 ml of methanol and 6 ml of water. Finally, the analyte was eluted into a 15-ml polypropylene centrifuge tube containing 5 ml of methanol. The eluted solution was dried under a gentle nitrogen stream at 50uC. The residue was then dissolved in 500 ml of dilution buffer and filtered through two 0.45-mm nylon filters for LC-MS/MS (Agilent 1260/ 6410/6410, Agilent Technologies, Palo Alto CA) analysis. Liquid chromatography was performed under the following conditions: The liquid chromatography system included 3.5-mm ZORBAX Eclipse Plus C18 analytical column (2.1 by 150 mm, Agilent Technologies). The mobile phase contained 30% solvent A (0.1% formic acid methanol, vol/vol) and 70% solvent B (5% methanol and 0.1% formic acid aqueous solution). The flow rate and column temperature were 0.3 ml?min21 and 35uC, respectively. For the mass spectrometry, Micromass Quattro LC triplequadrupole mass spectrometer with an electrospray ionization source was interfaced to the LC system (Agilent Technologies). The MS system was tuned using 3 mg?liter21 of CLP solution in a mobile phase and directly infused into the electrospray ionization source at 10 liters?min21. The conditions were established for the LC-MS/MS selected ion recording mode at 214.20 m/z for CLP. Multiple reaction monitoring (MS) conditions for CLP were
1826
PENG ET AL.
J. Food Prot., Vol. 77, No. 10
TABLE 1. Results of the orthogonal L9 (3)3 test performed to optimize the parameters (the amount of labeled antibody, volume of colloidal gold-monoclonal antibody in the ELISA well, and concentration of CLP–BSA antigen on the test line)
No.
Amt (mg?ml21) of labeled antibody
Vol (ml) of CG-mAb in the ELISA well
Concn (mg?ml21) of CLP-BSA on the test line
1
1.0
1.5
0.1
2
1.0
2.5
0.15
3
1.0
4.0
0.2
4
1.2
1.5
0.1
5
1.2
2.5
0.15
6
1.2
4.0
0.2
7
1.5
1.5
0.1
8
1.5
2.5
0.15
9
1.5
4.0
0.2
a b
Color intensity of test (AT) and control (AC) linesa
81.4 24.4 88.7 38.7 105.5 75.7 171.6 103.2 210.5 144.1 280.4 168.1 185.3 154.7 236.1 171.7 283.6 187.7
¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡
10.45/AT 8.45/AC 9.56/AT 7.39/AC 11.68/AT 8.93/AC 9.82/AT 6.94/AC 18.47/AT 12.53/AC 17.39/AT 13.33/AC 15.65/AT 16.87/AC 19.08/AT 13.63/AC 18.70/AT 16.04/AC
Inhibition rate of positive sample (%)b
10.2 13.6 18.7 36.1 17.3 25.6 23.9 19.6 12.8
The mean values of AT and AC were obtained from control samples. The concentration of positive sample was 3 mg?liter21.
optimized for the transition from 214.20 to 154.20. Multiple reaction monitoring was used for the identification of CLP. Statistical analysis. The statistical analysis of the data in each figure was performed by analysis of variance and F test. P , 0.05 is considered statistically significant.
RESULTS AND DISCUSSION Optimization of the parameters. The amount of labeled antibody, volume of CG-mAb in the ELISA well, and concentration of CLP-BSA antigen on the test line were important for the sensitivity of the test strips. An orthogonal L9 (3)3 test was performed to optimize the parameters of the LFIA system (14). The results (Table 1) indicated that the optimal conditions are as follows: anti-CLP monoclonal antibody was 1.2 mg?ml21, volume of CG-mAb in ELISA well was 1.5 ml, and concentration of CLP-BSA antigen on the test line was 0.10 mg?ml21. Detection time of LFIA system. CG-mAb was sprayed onto the conjugate pad in a traditional test strip. An instantaneous reaction occurs between antigen and antibody when the tested sample migrates by capillary action through the test strips. In this study, the sample was initially incubated with CG-mAb in the ELISA well for 3 min, so the analyte in the sample could bind with CGmAb completely. Thus, the sensitivity of the method can be enhanced. Under the optimized parameters, the reaction time was determined by the samples spiked with different CLP concentrations. After 2 min of reaction on the strips, the T/C ratio of strips was recorded every 60 s using the strip reader, for a total of 20 min. Figure 2 shows that for the negative sample, the T/C ratio exhibited a decline in the first
3 min, then it increased and maintained a constant value of 2.02 ¡ 0.09 at 6 min. The T/C ratio of positive samples increased with the reaction time, and both reached a relative balance at 6 min. Therefore, with the addition of the 3-min incubation in the ELISA well, the total time of the LFIA system for CLP detection was 9 min. Establishment of the calibration curve for CLP. This system represents a competitive binding immunoassay. The competitors, which are the analyte in the sample and the antigen on the test line, compete for the limited binding sites on the CG-mAb. In this method, the value of AT is inversely proportional to the amount of CLP in the samples. Serially spiked samples were detected by the LFIA system to establish the calibration curve. The quantitative detection data are illustrated in Figure 3A. AT shows an obvious decline, with increasing CLP concentrations; conversely, AC shows a constant average value of 103.41 ¡ 5.84. As shown in Figure 3B, the calibration curve was constructed by plotting the BX/B0 ratio against the logarithm of CLP concentrations. It exhibits good linearity in the range of 3.0 to 20.0 mg?liter21 (R2 ~ 0.9970), and the coefficient of variation for each concentration is less than 10%. The limit of detection is calculated to be 0.15 mg?liter21, according to the mean measured value (0.092 mg?liter21) of 20 negative samples plus threefold standard deviations (3 | 0.021) (7). Specificity of the LFIA system. To evaluate the specificity of the colloidal gold-based LFIA for the detection of CLP, high concentrations (100 mg?liter21) of CLP, ractopamine, terbutaline, salbutamol, ritodrine, and fenoterol were analyzed. As shown in Fig. 4, the T/C ratio of five other b2-agonist compounds was as high as that of
J. Food Prot., Vol. 77, No. 10
RAPID QUANTITATIVE DETECTION OF CLORPRENALINE RESIDUE
1827
FIGURE 2. Immunoreaction dynamics of T/C ratio with different CLP concentrations (0, 5, and 10 mg?liter21). The mixed solution added to the sample well; the T/C ratio of strips was recorded every 60 s using the strip reader from 2 to 20 min, reaching a balance at 6 min. Error bars represent standard deviations from the means (n ~ 5). * P , 0.05 for the T/C ratio of 0 mg?liter21 versus that of 5.0 or 10.0 mg?liter21. # P , 0.05 for the T/C ratio of 5.0 mg?liter21 versus that of 0 or 10.0 mg?liter21. ** P , 0.05 for the T/C ratio of 10.0 mg?liter21 versus that of 0 or 5.0 mg?liter21.
the negative control sample. However, the T/C ratio of CLP was 0, indicating that the current system was able to specifically detect CLP from other b2-agonist. Stability of the LFIA system. To evaluate the stability of the strips, the accelerated aging test was carried out by storing the strips at 60uC for 10 days. AT, AC, and T/C ratios were recorded daily by using the portable strip reader. After 7 days, the T/C ratio of negative and positive (3 mg?liter21) were 1.648 ¡ 0.021 and 1.228 ¡ 0.016, respectively. Compared with that of the first day (1.628 ¡ 0.034 and 1.217 ¡ 0.028), it showed no significant change. However, after 10 days, the T/C ratio of negative sample and positive sample increased to 2.182 ¡ 0.179 and 1.491 ¡ 0.311, which shows the strips are invalid. According to the
FIGURE 3. Quantitative detection of CLP in urine sample. Serially spiked samples (1.0, 3.0, 5.0, 7.0, 10.0, 15.0, and 20.0 mg?liter21) were detected by the LFIA system to establish the calibration curve. (A) Alteration of AT and AC with increasing CLP concentrations. AT shows an obvious decline with increasing CLP concentrations; conversely, AC shows a constant average value of 103.41 ¡ 5.84. (B) Calibration curve (Y ~ 20.3744Ln(X) z 1.2233, R2 ~ 0.9970) was constructed by plotting the BX/B0 ratio against the logarithm of CLP concentrations. B0 and BX are the T/C ratio of negative control and positive samples, respectively. Error bars represent standard deviations from the means.
FIGURE 4. Specificity study of the LFIA system. The concentration of all the compounds spiked in urine sample was 100 mg?liter21. Besides CLP, the T/C ratio of five other b2-agonist compounds is as high as that of control sample. The system can specifically detect CLP residue in swine urine. Error bars represent standard deviations from the means (n ~ 5). * P , 0.05 for the T/C ratio of CLP versus that of the five other b2-agonist compounds.
J. Food Prot., Vol. 77, No. 10
Y ~ 0.973X z 0.406 R2 ~ 0.9998
Arrhenius equation (12), the stability (longevity of the test strips from their date of manufacture to the date of use) of the LFIA system is valid for 12 months of storage at room temperature.
b
a
Mean value of 20 samples at each concentration with five replicates. The equation was obtained by the mean values of the LFIA system against those of the LC-MS/MS.
2.33 1.19 4.57 98.93 ¡ 2.3 108.4 ¡ 2.03 100.45 ¡ 4.6 2.97 ¡ 0.07 5.42 ¡ 0.101 10.05 ¡ 0.46 111.14 ¡ 18.2 112.35 ¡ 9.2 102.00 ¡ 9.4 3.0 5.0 10.0
3.33 ¡ 0.55 5.62 ¡ 0.46 10.2 ¡ 0.94
15.01 9.02 11.81
Recovery (%) Measured (mg?liter21) Recovery (%) Measured (mg?liter21)a Spiked (mg?liter21)
LFIA system
Coefficient of variation (%)
LC-MS/MS
Coefficient of variation (%)
Regression equation of the two methodsb
PENG ET AL.
TABLE 2. Recovery of CLP in swine urine samples in different detection methods
1828
Accuracy evaluation of the LFIA system. To evaluate the LFIA system, the LFIA quantitative detection system was compared with LC-MS/MS. The calibration curve of LC-MS/MS was constructed using six different concentrations of CLP (0.5, 1, 3, 5, 7, and 10 mg?liter21), and it was plotted by the signals of peak area against each CLP concentration in the urine samples. Twenty swine urine samples, which were spiked with CLP at 3.0, 5.0, and 10.0 mg?liter21, were tested by the two methods. The measured values are shown in Table 2, and indicates good correlation between the measured values gained from the two methods, as reflected by a reliable correlation coefficient (R2 ~ 0.9998) and a slope of 0.973. The developed approach specifically detects as little as 0.15 mg?liter21 CLP within 9 min and exhibits no obvious cross-reactivity with five other b2-agonist compounds. The results gained from the LFIA indicate good correlation and strong agreement with the results obtained by use of LC-MS/ MS. In addition, accelerated aging test indicates possible stability of the test strips for 12 months However, compared with the LC-MS/MS method, the coefficient of variation of the assay is large (9 to 16%), and the accuracy is low. In conclusion, as a screening method, the protocol is thus a simple, rapid, and quantitative method to detect CLP residue under field conditions. ACKNOWLEDGMENTS We are grateful to the National Key Technology R&D Program of China for the 12th Five-Year Plan (No. 2012BAK17B02), the National Natural Science Foundation of China (No. 30960301), Jiangxi Province Main Science and Technology Leader Project (20113BCB22007), and the Jiangxi Education Bureau Technology Project (KJLD13009) for financial support.
REFERENCES 1. Anonymous. Drugmaker found to use banned additive for animal medicines. Available at: http://news.xinhuanet.com/english/china/ 2013-08/11/c_132619704.htm. Accessed 11 August 2013. 2. Cao, Y., S. X. Huang, and D. N. Li. 2007. Find out a new badrenergic agonist—chlorprenaline by LC-MS. Chin. J. Vet. Med. 41: 30–32. 3. Chou, S. F. 2013. Development of a manual self-assembled colloidal gold nanoparticle-immunochromatographic strip for rapid determination of human interferon-c. Analyst 138:2620–2623. 4. Kraft, B., and J. G. Armstrong. 1965. Clinical studies with CLP, a new bronchodilator with and without the antihistamine methapyrilene. Appl. Ther. 7:829. 5. Lai, W. H., Y. C. F. Daniel, Y. Xu, and Y. H. Xiong. 2008. Screening procedures for clenbuterol residue determination in raw swine livers using lateral-flow assay and enzyme-linked immunosorbent assay. J. Food Prot. 71:865–869. 6. Lai, W. H., Y. Xu, D. Y. C. Fung, and Y. H. Xiong. 2007. Development of a lateral-flow assay for rapid screening of the performance-enhancing sympathomimetic drug clenbuterol used in animal production; food safety assessments. Asia Pac. J. Clin. Nutr. 16(Suppl.):106–110. 7. Lei, Y. C., Y. F. Tsai, Y. T. Tai, C. Y. Lin, K. H. Hsieh, T. H. Chang, S. Y. Shen, and T. F. Kuo. 2008. Development and fast screening of
J. Food Prot., Vol. 77, No. 10
8.
9.
10.
11.
12.
RAPID QUANTITATIVE DETECTION OF CLORPRENALINE RESIDUE
salbutamol residues in swine serum by an enzyme-linked immunosorbent assay in Taiwan. J. Agric. Food Chem. 56:5494–5499. Li, C. H., W. Luo, H. Y. Xu, Q. Zhang, H. Xu, Z. P. Aguilar, W. H. Lai, H. Wei, and Y. H. Xiong. 2013. Development of an immunochromatographic assay for rapid and quantitative detection of clenbuterol in swine urine. Food Control 34:725–732. Liu, X., J. J. Xiang, Y. Tang, X. L. Zhang, Q. Q. Fu, J. H. Zou, and Y. Lin. 2012. Colloidal gold nanoparticle probe-based immunochromatographic assay for the rapid detection of chromium ions in water and serum samples. Anal. Chim. Acta 745:99–105. Lu, C., Z. Luo, L. Huang, and X. S. Li. 2011. The Ru-catalyzed enantioselective preparation of chiral halohydrins and their application in the synthesis of (R)-clorprenaline and (S)-sotalol. Tetrahedron Asymmetry 22:722–727. Sai, F., H. Miao, Y. F. Zhao, H. J. Chen, and Y. N. Wu. 2012. Simultaneous detection of residues of 25 b2-agonists and 23 bblockers in animal foods by high-performance liquid chromatography coupled with linear ion trap mass spectrometry. J. Agric. Food Chem. 60:1898–1905. Wang, D. B., B. Tian, Z. P. Zhang, J. Y. Deng, Z. Q. Cui, R. F. Yang, X. Y. Wang, H. P. Wei, and X. E. Zhang. 2013. Rapid detection of
13.
14.
15.
16.
17.
1829
Bacillus anthracisspores using a super-paramagnetic lateral-flow immunological detection system. Biosens. Bioelectron. 42:661–667. Xie, Q. Y., Y. H. Wu, Q. R. Xiong, H. Y. Xu, Y. H. Xiong, K. Liu, Y. Jin, and W. H. Lai. 2014. Advantages of fluorescent microspheres compared with colloidal gold as a label in immunochromatographic lateral flow assays. Biosens. Bioelectron. 54:262–265. Xu, W., X. L. Chen, X. L. Huang, W. C. Yang, C. M. Liu, W. H. Lai, H. Y. Xu, and Y. H. Xiong. 2013. Ru(phen)32z doped silica nanoparticle based immunochromatographic strip for rapid quantitative detection of b-agonist residues in swine urine. Talanta 114:160– 166. Yan, H., S. Liu, M. Gao, and N. Sun. 2013. Ionic liquids modified dummy molecularly imprinted microspheres as solid phase extraction materials for the determination of clenbuterol and clorprenaline in urine. J. Chromatogr. A 1294:10–16. Yin, J., X. H. Gu, H. Lu, L. P. Guo, and H. P. Xie. 2010. Fast analysis in quality control of compound clorprenaline by near infrared spectroscopy. Chem. Res. Appl. 1:004. Zou, D., X. J. Wu, and G. Z. Nan. 1998. Determination of three components in compound clorprenaline tablet by capillary zone electrophoresis. Chin. Pharm. J. 33:06.
Research Note
Lateral-Flow Assay for Rapid Quantitative Detection of Clorprenaline Residue in Swine Urine TAO PENG,1 FU S. ZHANG,2 WAN C. YANG,3 DONG X. LI,2 YUAN CHEN,3 YONG H. XIONG,1 HUA WEI,1 AND WEI H. LAI1* 1State
Key Laboratory of Food Science and Technology, Nanchang University, 235 Nanjing East Road, Nanchang 330047, People’s Republic of China; Agricultural Product Quality Safety and Inspection Center, 151 Wenjiao Road, Nanchang 330077, People’s Republic of China; and 3Jiangxi Zodolabs Bioengineering Co., Ltd., 235 Nanjing East Road, Nanchang 330047, People’s Republic of China
2Jiangxi
MS 14-103: Received 2 March 2014/Accepted 31 May 2014
ABSTRACT Clorprenaline (CLP), a b2-adrenergic agonist, was first found in veterinary drugs for cold treatment in China in 2013. It is a potential new lean meat-boosting feed additive because it can promote animal muscular mass growth and decrease fat accumulation. A competitive colloidal gold-based lateral flow immunoassay system with a portable strip reader was successfully developed for rapid quantitative detection of CLP residue in swine urine. The detection system was optimized so that the detection can be completed within 9 min with a limit of detection of 0.15 mg?liter21. The assay exhibited good linear range from 3.0 to 20.0 mg?liter21, with reliable correlation of 0.9970 and with no obvious cross-reaction with five other b2-agonist compounds. Twenty spiked swine urine samples were tested by lateral flow immunoassay and liquid chromatography–tandem mass spectrometry to confirm the accuracy of the system. Results show good correlation between the two methods. This method is rapid, sensitive, specific, and convenient. It can be applied in the field for on-site detection of CLP in urine samples.
Clorprenaline (CLP) [1-(2-chlorophenyl)-2-propan-2ylamino-ethanol] belongs to the b2-adrenergic agonist group. CLP has been clinically used for the treatment of bronchial asthma and asthmatic bronchitis (4, 10). In 2007, Cao et al. warned about CLP hydrochloride being illegally used as new growth promoter in livestock production (2). Like other b2-adrenergic agonists, CLP has been proven to function in nutrient redistribution in the metabolism of animals at higher dosage levels (11). Due to lack of public concern, the use of CLP as a lean meat-boosting additive has evaded the examination of authorities. After eating meat products containing CLP residues, consumers have exhibited various symptoms of poisoning, such as muscular tremors, weakness of limbs, fever, headache, and nausea (15). In 2013, Chinese police injuncted a pharmaceutical company that used harmful additives in animal medicines. The Ministry of Public Security stated that reports have been made by the agricultural authorities of Haiyan County of East China’s Zhejiang province about many of the county’s swine being tested for CLP hydrochloride, a banned lean meat-boosting additive (1). Current methods for detecting CLP, such as highpressure liquid chromatography (15), capillary zone electrophoresis (17), and near-infrared spectroscopy (16) are time consuming and also require expensive instruments and complicated operation. * Author for correspondence. Tel: 0086-791-83969526; Fax: 0086-79188157619; E-mail: [email protected].
Colloidal gold-based lateral-flow immunochromatographic assay (LFIA), which is presently widely applied in the area of food safety (5, 6), medical diagnosis (3), and environment monitoring (9), allows rapid, sensitive, stable, convenient, inexpensive, and user-friendly determination of analytes (13). Our study pioneered the development of a colloidal gold-based LFIA system, involving the use of a portable strip reader for the rapid and quantitative detection of CLP residues in swine urine. MATERIALS AND METHODS Materials. CLP, terbutaline, salbutamol, ritodrine, fenoterol, bovine serum albumin (BSA; fraction V), and goat anti-mouse antibody were obtained from Sigma (St. Louis, MO). Ractopamine was obtained from Dr. Ehrenstorfer GmbH (Augsburg, Germany). CLP-BSA conjugate antigen and anti-CLP monoclonal antibody were provided by Wuxi Zodoboer Biotech. Co., Ltd. (Wuxi, People’s Republic of China). Hydrogen tetrachloroaurate trihydrate was obtained from Aldrich (Milwaukee, WI). The sample pad, conjugate release pad, nitrocellulose membrane, and absorbent pad were purchased from Millipore, Inc. (Bedford, MA). All solvents and other chemicals were of analytical reagent grade. The BioDot XYZ Platform, which combines motion control with the BioJet Quanti3050k dispenser and AirJet Quanti3050k dispenser, was acquired from BioDot (Irvine, CA). The portable strip reader was introduced in our previous study (8). Preparation of colloidal gold. One milliliter of 1% (wt/vol) stock solution of hydrogen tetrachloroaurate trihydrate was added to 100 ml (vol) of distilled water and heated to boiling point. Then,
J. Food Prot., Vol. 77, No. 10
RAPID QUANTITATIVE DETECTION OF CLORPRENALINE RESIDUE
1825
FIGURE 1. The flow chart of colloidal gold-based lateral-flow immunoassay system for rapid quantitative detection of CLP. A 100-ml urine sample without pretreatment was pipetted into the ELISA well, incubated for 3 min, and added to the sample well of the lateral flow strip. After 6 min, the strip was scanned with the portable strip reader; the result is shown in the data display window.
1.3 ml of freshly prepared 1% sodium citrate solution was added to the gold solution under constant stirring. When the color of the mixture turned red, the solution continued boiling for another 5 min; the colloidal gold solution was obtained and stored at 4uC. Preparation of colloidal gold-monoclonal antibody. The colloidal gold-monoclonal antibody (CG-mAb) was prepared as follows: 10 ml of colloidal gold solution was adjusted to pH 6.0 with 0.2 M K2CO3. With gentle stirring, 1 ml of anti-CLP monoclonal antibody solution was added dropwise to the colloidal gold solution to a final concentration of 1.2 mg?ml21, after which the mixture was agitated for 60 min. Then, 1 ml of 1% (wt/vol) polyethylene glycol20000 solution was added to the mixture, which was agitated for 30 min. Afterward, 1 ml of BSA (10%, wt/vol) was added for further blocking for 30 min. The mixture was centrifuged at 8,000 g at 4uC for 30 min to remove the unlabeled free antibody. The resulting precipitate was dissolved in 1 ml of dilution buffer, and then 1.5 ml of the prepared CG-mAb was added to an enzyme-linked immunosorbent assay (ELISA) well and dried at 30uC for 2.5 h. Preparation of immunochromatographic test strips. The sample pad, which was treated with 50 mM borate buffer (pH 7.4, containing 1% BSA, 0.5% Tween-20, and 0.05% sodium azide), was dried at 60uC for 2 h. The CLP-BSA conjugation (0.10 mg?ml21) and rabbit anti-mouse antibody (1.0 mg?ml21) were applied to the test and control lines on the nitrocellulose membrane, respectively, and dried at 35uC for 12 h. The nitrocellulose membrane, absorption pad, glass fiber membrane, and pretreated sample pad were assembled as the test strip (Fig. 1). Lateral flow test procedure. As illustrated in the flow chart (Fig. 1), a 100-ml urine sample without pretreatment was pipetted into the ELISA well and incubated for 3 min. The sample solutionCG-mAb reaction system was added to the sample well of the lateral flow strip. After 6 min, the strip was scanned with the portable strip reader. The color intensity of test line (AT), control line (AC), and ratio of AT to AC (T/C) were recorded. Optimization of the reaction conditions for CLP detection. The colloidal gold lateral-flow immunoassay was optimized by varying the following parameters: amount of labeled antibody (1.0, 1.2, and 1.5 mg?ml21); volume of CG-mAb in the ELISA well (1.5, 2.5, and 4 ml); concentration of CLP-BSA on the test line (0.1,
0.15, and 0.2 mg?ml21); and reaction time. The optimum conditions were obtained from each experiment, which was performed with five replications. Quantitative calibration curve of the LFIA system. Forty swine urine samples that were ascertained to be free of CLP by liquid chromatography–tandem mass spectrometry (LC-MS/MS) were collected from Jiangxi province and mixed as negative sample. Portions of the sample mixture were spiked with CLP at concentrations of 1.0, 3.0, 5.0, 7.0, 10.0, 15.0, and 20.0 mg?liter21. The T/C ratio of negative control and positive samples were renamed as B0 and BX. The calibration curve was constructed by plotting the BX/B0 ratio against the logarithm of CLP concentrations. Information of the linear regression equation was established for quantitative analysis. LC-MS/MS method. One milliliter of ammonium acetate (2 mol?liter21) was added to 1 ml of the swine urine sample, followed by addition of 15 ml of the enzyme agent. After oscillating in a water bath at 37uC for 16 h, 5 ml of ion-pair reagent was added into the mixture. The solution was then transferred to an OG column, which was preconditioned with 3 ml of methanol and 6 ml of water. Finally, the analyte was eluted into a 15-ml polypropylene centrifuge tube containing 5 ml of methanol. The eluted solution was dried under a gentle nitrogen stream at 50uC. The residue was then dissolved in 500 ml of dilution buffer and filtered through two 0.45-mm nylon filters for LC-MS/MS (Agilent 1260/ 6410/6410, Agilent Technologies, Palo Alto CA) analysis. Liquid chromatography was performed under the following conditions: The liquid chromatography system included 3.5-mm ZORBAX Eclipse Plus C18 analytical column (2.1 by 150 mm, Agilent Technologies). The mobile phase contained 30% solvent A (0.1% formic acid methanol, vol/vol) and 70% solvent B (5% methanol and 0.1% formic acid aqueous solution). The flow rate and column temperature were 0.3 ml?min21 and 35uC, respectively. For the mass spectrometry, Micromass Quattro LC triplequadrupole mass spectrometer with an electrospray ionization source was interfaced to the LC system (Agilent Technologies). The MS system was tuned using 3 mg?liter21 of CLP solution in a mobile phase and directly infused into the electrospray ionization source at 10 liters?min21. The conditions were established for the LC-MS/MS selected ion recording mode at 214.20 m/z for CLP. Multiple reaction monitoring (MS) conditions for CLP were
1826
PENG ET AL.
J. Food Prot., Vol. 77, No. 10
TABLE 1. Results of the orthogonal L9 (3)3 test performed to optimize the parameters (the amount of labeled antibody, volume of colloidal gold-monoclonal antibody in the ELISA well, and concentration of CLP–BSA antigen on the test line)
No.
Amt (mg?ml21) of labeled antibody
Vol (ml) of CG-mAb in the ELISA well
Concn (mg?ml21) of CLP-BSA on the test line
1
1.0
1.5
0.1
2
1.0
2.5
0.15
3
1.0
4.0
0.2
4
1.2
1.5
0.1
5
1.2
2.5
0.15
6
1.2
4.0
0.2
7
1.5
1.5
0.1
8
1.5
2.5
0.15
9
1.5
4.0
0.2
a b
Color intensity of test (AT) and control (AC) linesa
81.4 24.4 88.7 38.7 105.5 75.7 171.6 103.2 210.5 144.1 280.4 168.1 185.3 154.7 236.1 171.7 283.6 187.7
¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡
10.45/AT 8.45/AC 9.56/AT 7.39/AC 11.68/AT 8.93/AC 9.82/AT 6.94/AC 18.47/AT 12.53/AC 17.39/AT 13.33/AC 15.65/AT 16.87/AC 19.08/AT 13.63/AC 18.70/AT 16.04/AC
Inhibition rate of positive sample (%)b
10.2 13.6 18.7 36.1 17.3 25.6 23.9 19.6 12.8
The mean values of AT and AC were obtained from control samples. The concentration of positive sample was 3 mg?liter21.
optimized for the transition from 214.20 to 154.20. Multiple reaction monitoring was used for the identification of CLP. Statistical analysis. The statistical analysis of the data in each figure was performed by analysis of variance and F test. P , 0.05 is considered statistically significant.
RESULTS AND DISCUSSION Optimization of the parameters. The amount of labeled antibody, volume of CG-mAb in the ELISA well, and concentration of CLP-BSA antigen on the test line were important for the sensitivity of the test strips. An orthogonal L9 (3)3 test was performed to optimize the parameters of the LFIA system (14). The results (Table 1) indicated that the optimal conditions are as follows: anti-CLP monoclonal antibody was 1.2 mg?ml21, volume of CG-mAb in ELISA well was 1.5 ml, and concentration of CLP-BSA antigen on the test line was 0.10 mg?ml21. Detection time of LFIA system. CG-mAb was sprayed onto the conjugate pad in a traditional test strip. An instantaneous reaction occurs between antigen and antibody when the tested sample migrates by capillary action through the test strips. In this study, the sample was initially incubated with CG-mAb in the ELISA well for 3 min, so the analyte in the sample could bind with CGmAb completely. Thus, the sensitivity of the method can be enhanced. Under the optimized parameters, the reaction time was determined by the samples spiked with different CLP concentrations. After 2 min of reaction on the strips, the T/C ratio of strips was recorded every 60 s using the strip reader, for a total of 20 min. Figure 2 shows that for the negative sample, the T/C ratio exhibited a decline in the first
3 min, then it increased and maintained a constant value of 2.02 ¡ 0.09 at 6 min. The T/C ratio of positive samples increased with the reaction time, and both reached a relative balance at 6 min. Therefore, with the addition of the 3-min incubation in the ELISA well, the total time of the LFIA system for CLP detection was 9 min. Establishment of the calibration curve for CLP. This system represents a competitive binding immunoassay. The competitors, which are the analyte in the sample and the antigen on the test line, compete for the limited binding sites on the CG-mAb. In this method, the value of AT is inversely proportional to the amount of CLP in the samples. Serially spiked samples were detected by the LFIA system to establish the calibration curve. The quantitative detection data are illustrated in Figure 3A. AT shows an obvious decline, with increasing CLP concentrations; conversely, AC shows a constant average value of 103.41 ¡ 5.84. As shown in Figure 3B, the calibration curve was constructed by plotting the BX/B0 ratio against the logarithm of CLP concentrations. It exhibits good linearity in the range of 3.0 to 20.0 mg?liter21 (R2 ~ 0.9970), and the coefficient of variation for each concentration is less than 10%. The limit of detection is calculated to be 0.15 mg?liter21, according to the mean measured value (0.092 mg?liter21) of 20 negative samples plus threefold standard deviations (3 | 0.021) (7). Specificity of the LFIA system. To evaluate the specificity of the colloidal gold-based LFIA for the detection of CLP, high concentrations (100 mg?liter21) of CLP, ractopamine, terbutaline, salbutamol, ritodrine, and fenoterol were analyzed. As shown in Fig. 4, the T/C ratio of five other b2-agonist compounds was as high as that of
J. Food Prot., Vol. 77, No. 10
RAPID QUANTITATIVE DETECTION OF CLORPRENALINE RESIDUE
1827
FIGURE 2. Immunoreaction dynamics of T/C ratio with different CLP concentrations (0, 5, and 10 mg?liter21). The mixed solution added to the sample well; the T/C ratio of strips was recorded every 60 s using the strip reader from 2 to 20 min, reaching a balance at 6 min. Error bars represent standard deviations from the means (n ~ 5). * P , 0.05 for the T/C ratio of 0 mg?liter21 versus that of 5.0 or 10.0 mg?liter21. # P , 0.05 for the T/C ratio of 5.0 mg?liter21 versus that of 0 or 10.0 mg?liter21. ** P , 0.05 for the T/C ratio of 10.0 mg?liter21 versus that of 0 or 5.0 mg?liter21.
the negative control sample. However, the T/C ratio of CLP was 0, indicating that the current system was able to specifically detect CLP from other b2-agonist. Stability of the LFIA system. To evaluate the stability of the strips, the accelerated aging test was carried out by storing the strips at 60uC for 10 days. AT, AC, and T/C ratios were recorded daily by using the portable strip reader. After 7 days, the T/C ratio of negative and positive (3 mg?liter21) were 1.648 ¡ 0.021 and 1.228 ¡ 0.016, respectively. Compared with that of the first day (1.628 ¡ 0.034 and 1.217 ¡ 0.028), it showed no significant change. However, after 10 days, the T/C ratio of negative sample and positive sample increased to 2.182 ¡ 0.179 and 1.491 ¡ 0.311, which shows the strips are invalid. According to the
FIGURE 3. Quantitative detection of CLP in urine sample. Serially spiked samples (1.0, 3.0, 5.0, 7.0, 10.0, 15.0, and 20.0 mg?liter21) were detected by the LFIA system to establish the calibration curve. (A) Alteration of AT and AC with increasing CLP concentrations. AT shows an obvious decline with increasing CLP concentrations; conversely, AC shows a constant average value of 103.41 ¡ 5.84. (B) Calibration curve (Y ~ 20.3744Ln(X) z 1.2233, R2 ~ 0.9970) was constructed by plotting the BX/B0 ratio against the logarithm of CLP concentrations. B0 and BX are the T/C ratio of negative control and positive samples, respectively. Error bars represent standard deviations from the means.
FIGURE 4. Specificity study of the LFIA system. The concentration of all the compounds spiked in urine sample was 100 mg?liter21. Besides CLP, the T/C ratio of five other b2-agonist compounds is as high as that of control sample. The system can specifically detect CLP residue in swine urine. Error bars represent standard deviations from the means (n ~ 5). * P , 0.05 for the T/C ratio of CLP versus that of the five other b2-agonist compounds.
J. Food Prot., Vol. 77, No. 10
Y ~ 0.973X z 0.406 R2 ~ 0.9998
Arrhenius equation (12), the stability (longevity of the test strips from their date of manufacture to the date of use) of the LFIA system is valid for 12 months of storage at room temperature.
b
a
Mean value of 20 samples at each concentration with five replicates. The equation was obtained by the mean values of the LFIA system against those of the LC-MS/MS.
2.33 1.19 4.57 98.93 ¡ 2.3 108.4 ¡ 2.03 100.45 ¡ 4.6 2.97 ¡ 0.07 5.42 ¡ 0.101 10.05 ¡ 0.46 111.14 ¡ 18.2 112.35 ¡ 9.2 102.00 ¡ 9.4 3.0 5.0 10.0
3.33 ¡ 0.55 5.62 ¡ 0.46 10.2 ¡ 0.94
15.01 9.02 11.81
Recovery (%) Measured (mg?liter21) Recovery (%) Measured (mg?liter21)a Spiked (mg?liter21)
LFIA system
Coefficient of variation (%)
LC-MS/MS
Coefficient of variation (%)
Regression equation of the two methodsb
PENG ET AL.
TABLE 2. Recovery of CLP in swine urine samples in different detection methods
1828
Accuracy evaluation of the LFIA system. To evaluate the LFIA system, the LFIA quantitative detection system was compared with LC-MS/MS. The calibration curve of LC-MS/MS was constructed using six different concentrations of CLP (0.5, 1, 3, 5, 7, and 10 mg?liter21), and it was plotted by the signals of peak area against each CLP concentration in the urine samples. Twenty swine urine samples, which were spiked with CLP at 3.0, 5.0, and 10.0 mg?liter21, were tested by the two methods. The measured values are shown in Table 2, and indicates good correlation between the measured values gained from the two methods, as reflected by a reliable correlation coefficient (R2 ~ 0.9998) and a slope of 0.973. The developed approach specifically detects as little as 0.15 mg?liter21 CLP within 9 min and exhibits no obvious cross-reactivity with five other b2-agonist compounds. The results gained from the LFIA indicate good correlation and strong agreement with the results obtained by use of LC-MS/ MS. In addition, accelerated aging test indicates possible stability of the test strips for 12 months However, compared with the LC-MS/MS method, the coefficient of variation of the assay is large (9 to 16%), and the accuracy is low. In conclusion, as a screening method, the protocol is thus a simple, rapid, and quantitative method to detect CLP residue under field conditions. ACKNOWLEDGMENTS We are grateful to the National Key Technology R&D Program of China for the 12th Five-Year Plan (No. 2012BAK17B02), the National Natural Science Foundation of China (No. 30960301), Jiangxi Province Main Science and Technology Leader Project (20113BCB22007), and the Jiangxi Education Bureau Technology Project (KJLD13009) for financial support.
REFERENCES 1. Anonymous. Drugmaker found to use banned additive for animal medicines. Available at: http://news.xinhuanet.com/english/china/ 2013-08/11/c_132619704.htm. Accessed 11 August 2013. 2. Cao, Y., S. X. Huang, and D. N. Li. 2007. Find out a new badrenergic agonist—chlorprenaline by LC-MS. Chin. J. Vet. Med. 41: 30–32. 3. Chou, S. F. 2013. Development of a manual self-assembled colloidal gold nanoparticle-immunochromatographic strip for rapid determination of human interferon-c. Analyst 138:2620–2623. 4. Kraft, B., and J. G. Armstrong. 1965. Clinical studies with CLP, a new bronchodilator with and without the antihistamine methapyrilene. Appl. Ther. 7:829. 5. Lai, W. H., Y. C. F. Daniel, Y. Xu, and Y. H. Xiong. 2008. Screening procedures for clenbuterol residue determination in raw swine livers using lateral-flow assay and enzyme-linked immunosorbent assay. J. Food Prot. 71:865–869. 6. Lai, W. H., Y. Xu, D. Y. C. Fung, and Y. H. Xiong. 2007. Development of a lateral-flow assay for rapid screening of the performance-enhancing sympathomimetic drug clenbuterol used in animal production; food safety assessments. Asia Pac. J. Clin. Nutr. 16(Suppl.):106–110. 7. Lei, Y. C., Y. F. Tsai, Y. T. Tai, C. Y. Lin, K. H. Hsieh, T. H. Chang, S. Y. Shen, and T. F. Kuo. 2008. Development and fast screening of
J. Food Prot., Vol. 77, No. 10
8.
9.
10.
11.
12.
RAPID QUANTITATIVE DETECTION OF CLORPRENALINE RESIDUE
salbutamol residues in swine serum by an enzyme-linked immunosorbent assay in Taiwan. J. Agric. Food Chem. 56:5494–5499. Li, C. H., W. Luo, H. Y. Xu, Q. Zhang, H. Xu, Z. P. Aguilar, W. H. Lai, H. Wei, and Y. H. Xiong. 2013. Development of an immunochromatographic assay for rapid and quantitative detection of clenbuterol in swine urine. Food Control 34:725–732. Liu, X., J. J. Xiang, Y. Tang, X. L. Zhang, Q. Q. Fu, J. H. Zou, and Y. Lin. 2012. Colloidal gold nanoparticle probe-based immunochromatographic assay for the rapid detection of chromium ions in water and serum samples. Anal. Chim. Acta 745:99–105. Lu, C., Z. Luo, L. Huang, and X. S. Li. 2011. The Ru-catalyzed enantioselective preparation of chiral halohydrins and their application in the synthesis of (R)-clorprenaline and (S)-sotalol. Tetrahedron Asymmetry 22:722–727. Sai, F., H. Miao, Y. F. Zhao, H. J. Chen, and Y. N. Wu. 2012. Simultaneous detection of residues of 25 b2-agonists and 23 bblockers in animal foods by high-performance liquid chromatography coupled with linear ion trap mass spectrometry. J. Agric. Food Chem. 60:1898–1905. Wang, D. B., B. Tian, Z. P. Zhang, J. Y. Deng, Z. Q. Cui, R. F. Yang, X. Y. Wang, H. P. Wei, and X. E. Zhang. 2013. Rapid detection of
13.
14.
15.
16.
17.
1829
Bacillus anthracisspores using a super-paramagnetic lateral-flow immunological detection system. Biosens. Bioelectron. 42:661–667. Xie, Q. Y., Y. H. Wu, Q. R. Xiong, H. Y. Xu, Y. H. Xiong, K. Liu, Y. Jin, and W. H. Lai. 2014. Advantages of fluorescent microspheres compared with colloidal gold as a label in immunochromatographic lateral flow assays. Biosens. Bioelectron. 54:262–265. Xu, W., X. L. Chen, X. L. Huang, W. C. Yang, C. M. Liu, W. H. Lai, H. Y. Xu, and Y. H. Xiong. 2013. Ru(phen)32z doped silica nanoparticle based immunochromatographic strip for rapid quantitative detection of b-agonist residues in swine urine. Talanta 114:160– 166. Yan, H., S. Liu, M. Gao, and N. Sun. 2013. Ionic liquids modified dummy molecularly imprinted microspheres as solid phase extraction materials for the determination of clenbuterol and clorprenaline in urine. J. Chromatogr. A 1294:10–16. Yin, J., X. H. Gu, H. Lu, L. P. Guo, and H. P. Xie. 2010. Fast analysis in quality control of compound clorprenaline by near infrared spectroscopy. Chem. Res. Appl. 1:004. Zou, D., X. J. Wu, and G. Z. Nan. 1998. Determination of three components in compound clorprenaline tablet by capillary zone electrophoresis. Chin. Pharm. J. 33:06.
