Research Article Received: 18 December 2013
Revised: 26 May 2014
Accepted article published: 20 June 2014
Published online in Wiley Online Library: 23 July 2014
(wileyonlinelibrary.com) DOI 10.1002/jsfa.6794
Determination of 15 sedative residues in mutton by rapid resolution liquid chromatography–tandem mass spectrometry Jinmei Wei, Yuzhu Luo,* Li Zhang and Suli Fang Abstract BACKGROUND: The use of xenobiotic compounds in animal husbandry has given rise to consumer anxieties regarding residual risk and food safety. Thus, animal tissues have become main samples for residue analysis and food safety for sedatives. In this study, a rapid resolution liquid chromatography–tandem mass spectrometry (RRLC-MS/MS) method was established for the determination of 15 sedatives residues in mutton. RESULTS: After enzymolysis, sedatives residues in mutton were extracted by ammonium hydroxide–acetonitrile (10:90, v/v) and determined by RRLC-MS/MS with quantification by standard curve method. The calibration curves showed good linearity within the concentrations of 0.5–50 𝛍g kg−1 with the correlation coefficients (r2 ) ranged from 0.9639 to 0.9984. The limits of detection (LODs) and quantification (LOQs) were 0.25–2.5 and 0.5–5 𝛍g kg−1 , respectively. The average recoveries of spikes samples were in the ranges of 74.1–116.8% with relative standard deviations of intra- and inter-day ranged from 2.6% to 11.2% and from 2.1% to 11.4%, respectively. CONCLUSION: This method is simple, sensitive and accurate in the determination of sedative residues. © 2014 Society of Chemical Industry Keywords: rapid resolution liquid chromatography–tandem mass spectrometry (RRLC-MS/MS); sedatives; residues; mutton
INTRODUCTION
598
The sedative hypnotics have been used for a long time as animal tranquillisers to prevent stress and weight loss during their transportation, and they are added to animal feeds occasionally to reduce the animal’s reaction, slowing down the metabolic processes for enhancing their growth rate.1 – 3 Thus, animal tissues have become the main samples for residue analysis and food safety for sedatives. Xylazine hydrochloride, azaperone, zolpidem tartrate, carazolol, chlordiazepoxide, droperidol, haloperidol, promethazine hydrochloride, acepromazine maleate, perphenazine, propionylpromazine hydrochlorid, chlorpromazine hydrochloride, nitrazepam, fluphenazine hydrochloride and clonazepam are the most commonly used drugs. These 15 compounds belong to four families in chemical structure: benzodiazepines, phenothiazines, butyrophenones and imidazopyridines. Their chemical structures are shown in Fig. 1. The use of xenobiotic compounds in animal husbandry has given rise to consumer anxieties regarding residual risk and food safety. These sedative hypnotics have been found to be the probable causes of over-dosage dependence, excessive sedation, anaesthesia, coma and death.4 The residues of these compounds in foodstuff of animal origin may represent a potential risk to consumers and many countries have regulated their use in veterinary medicine, and residues of some sedatives are prohibited in animal derived food.5,6 In China, the Ministry of Agriculture of the People’s Republic of China has introduced laws regarding the maximum residue limits of veterinary drug residues in food of animal origin. Chlorpromazine, methaqualone and diazepam were not allowed J Sci Food Agric 2015; 95: 598–606
in any animal food, and xylazine was not allowed in milk as well.7 There is no relevant national standard for the maximum residue limits of other sedatives in mutton yet. A number of methods have been reported for the identification and determination of these compounds in biological samples in clinical and forensic toxicology,8 – 11 whereas only a few methods have been published for analysis of these drugs in animal tissues. Cheng et al.2 and Wang et al.12 reported a gas chromatography–mass spectrometry (GC-MS) method to determine diazepam, methaqualone, chlorpromazine and promethazine in swine tissues and benzodiazepine residues in pork by using multi-walled carbon nanotube solid phase, respectively. The limits of detection and quantification in swine muscle and liver tissues ranged from 0.2 to 0.3 μg kg−1 and 0.5 to 1 μg kg−1 , respectively.2 The limits of detection were 2 μg kg−1 for diazepam and 10 μg kg−1 for estazolam, alprazolam and triazolam in pork.12 Sun et al.3 determined 10 sedative residues in pork and kidney by UPLC-MS/MS. The limits of detection of the 10 sedatives were 0.5 μg kg−1 , and the limit of quantification was 1 μg kg−1 . To the best of our knowledge, so far, no studies have been reported in which determined these drugs have been determined in mutton.
∗
Correspondence to: Yuzhu Luo, Gansu Key Laboratory of Herbivorous Animal Biotechnology, Gansu Agricultural University, 730070 Lanzhou,China. E-mail: [email protected] Gansu Agricultural University/Gansu Key Laboratory of Herbivorous Animal Biotechnology, 730070 Lanzhou, China
www.soci.org
© 2014 Society of Chemical Industry
Determination of Fifteen Sedative Residues in Mutton
www.soci.org
Figure 1. The chemical structures of the 15 sedatives tested.
The purpose of the present study was to develop a reliable and sensitive method for the simultaneous detection and determination of 15 sedative residues in mutton by RRLC-MS/MS.
MATERIALS AND METHODS
J Sci Food Agric 2015; 95: 598–606
Standard solutions All standard stock solutions were prepared in MeOH at a concentration of 200 μg mL−1 . Composite standard intermediate solutions were prepared via dilution of the stock solutions in 0.027 mol L−1 aqueous FA-MeCN (90:10, v/v) at 50, 30, 10, 5, 2, 1 and 0.5 μg L−1 . Matrix-matched standard working solutions were prepared in
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
599
Chemicals and reagents The 15 compounds tested were as follows: zolpidem tartrate (purity 99%), chlordiazepoxide (≥99%), haloperidol (99%), promethazine hydrochloride (≥98%), prochlorperazine (99%), chlorpromazine hydrochloride (99%), nitrazepam (99%), fluphenazine hydrochloride (≥98%) and clonazepam (99%) were purchased from the National Institute for Food and Drug Control (Beijing, China). Xylazine hydrochloride (≥99%), carazolol (≥99%) and acepromazine maleate (≥99%) were obtained from Dr. Ehrenstorfer GmbH (Augsburg, Germany). Propionylpromazine
hydrochloride (≥99%) and azaperone (≥99%) were supplied by Sigma-Aldrich (Taufkirchen, Germany). Droperidol (≥99%) was provided by US Pharmacopeial Convention (Rockville, MD, USA). Acetonitrile (MeCN) and methanol (MeOH) were purchased from Merck (Darmstadt, Germany), 𝛽-glucuronidase/aryl sulfatase, formic acid (FA) and ammonium hydroxide from Sigma-Aldrich. Ultrapure water was obtained from a Milli-Q water purification system (Billerica, MA, USA). All solvents and reagents used were of HPLC or analytical grade, unless otherwise stated.
www.soci.org blank sample extracts, which were extracted from blank mutton raised by our laboratory and confirmed in advance not to contain any of the tested analytes. All of the standard solutions were stored at −18 ∘ C in a dark amber bottle. Sample preparation Blank mutton was from sheep free of drugs raised by our laboratory, and actual mutton was bought from a local market (Lanzhou, China). Two types of sample were minced, mixed together, homogenised and stored at −18 ∘ C for later use, respectively. The samples were removed from the freezer and allowed to reach room temperature prior to analysis. A 5 g sample was weighed in a polypropylene centrifuge tube (50 mL) to which 15 mL ammonia–acetonitrile (10:90, v/v), and 30 μL 𝛽-glucuronidase/aryl sulfatase was added. The sample was vortexed and shaked for 10 h at 37 ∘ C in a dark place. Then it was extracted for 10 min at 30 ∘ C with ultrasonication and then centrifuged at 2810 × g for 5 min. The supernatant was transferred into a polypropylene centrifuge tube (50 mL). Another 15 mL 0.26 mol L−1 ammonium hydroxide–MeCN was added to the residue and extracted for 10 min at 30 ∘ C with ultrasonication. The extraction step was repeated twice and the ammonium hydroxide–MeCN extraction was collected into the polypropylene centrifuge tube above. Five grams of sodium chloride was added to the polypropylene centrifuge tube filled with the extraction. The sample was vortexed and allowed to stand for 10 min. Most of lipids in the extracts were removed and the rest extracts were evaporated to dryness at 37 ∘ C. The residue was redissolved in 1 mL 0.027 mol L−1 aqueous FA-MeCN (90:10, v/v) and filtered with a 0.22 μm organic membrane. The filtrate was subjected to analysis by RRLC-MS/MS. RRLC-MS/MS analysis RRLC was carried out using an Agilent Technologies 1290 series system equipped with a G1322A degasser, a G1312B SL binary pump, a G1357D high-performance autosampler (HiP ALS SL+) (Agilent Technologies, Palo Alto, CA, USA). Separation was achieved on an Acquity UPLC CSHTM C18 column (100 mm × 2.1 mm i.d., particle size 1.7 μm; Waters, Milford, MA, USA) that had been maintained at 30 ∘ C in a thermostated column
J Wei et al.
compartment (model G1316B SL, Agilent Technologies, Palo Alto, CA, USA). The binary solvent system consisted of MeCN (mobile phase A) and 0.027 mol L−1 FA in ultrapure water (mobile phase B).The linear mobile phase gradient started at 10% A, increased to 70% A (1–15 min) and ramped back down to 10% A (15–16 min). The flow rate was 0.3 mL min−1 , and the injection volume was 5 μL. A 6460 triple-quadrupole mass spectrometer (Agilent Technologies) was operated in multiple reaction monitoring (MRM) mode with the positive ESI source (Agilent G1948B) for all the target analytes during the chromatographic analysis. For the optimisation of MS/MS parameters, composite standard working solution (1 μg mL−1 ) diluted in a mixture of 0.027 mol L−1 aqueous FA and MeCN (90:10, v/v) was introduced into the ESI+ source at an infusion flow rate of 5 μL min−1 . All the parameters were duly applied in order to find the optimal conditions providing the highest intensities of the analytes. Finally, the instrument was operated with the capillary voltage at +3.2 kV, and charging voltage at +1 kV. Nitrogen was used as nebuliser gas of 0.31 MPa, a carrier gas of 10 L min−1 at 350 ∘ C, and a sheath gas of 11 L min−1 at 350 ∘ C. The conditions used for the ionisation source were set as follows: source temperature, 120 ∘ C; and desolvation temperature, 350 ∘ C. The cone and desolvation gas flows were 50 and 500 L h−1 , respectively. The parameters optimised of cone voltage, collision energy, dwell time, qualitative ion pair and quantitative ion pair for each analyte during infusion are listed in Table 1. An Agilent Mass Hunter workstation (Agilent Technologies) was used for the control of equipment, data acquisition and analysis.
Method validation The present study was validated using a within-laboratory protocol. To determine the limit of detection (LOD) and limit of quantification (LOQ), composite standard working solutions were prepared in blank sample extract to find the concentrations that would give signal-to-noise ratios of 3 and 10, respectively. The fortified samples were prepared by spiking the composite standard intermediate solutions into blank samples at different concentrations for matrix-matched calibration curves. The concentrations were:
Table 1. The parameters optimised of cone voltage (CV), collision energy (CE), dwell time (DT), qualitative (Q1) and quantitative (Q2) MRM transition for each analyte Analyte
600
Xylazine hydrochloride Azaperone Zolpidem tartrate Carazolol Nitrazepam Acepromazine maleate Perphenazine Promethazine hydrochloride Chlorpromazine hydrochloride Droperidol Chlordiazepoxide Haloperidol Propionylpromazine hydrochlorid Fluphenazine hydrochloride Clonazepam
wileyonlinelibrary.com/jsfa
Q1 (m/z) 221.1/163.75 328.4/121 308.3/263.1 299.5/116.2 282.3/179.9 327.4/57.9 404.38/143 285.2/85.8 319.11/57.99 380.21/165.01 300.32/227.03 376.4/122.8 341.2/57.7 438.33/143.04 316.1/214.03
Q2 (m/z) 221.1/163.75 328.4/121 308.3/235.1 299.5/116.2 282.3/236.2 327.4/85.91 404.38/171.14 285.2/198 319.11/86.06 380.21/194.22 300.32/282.07 376.4/165.4 341.2/85.7 438.33/171.1 316.1/214.03
© 2014 Society of Chemical Industry
DT (ms) 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05
CV (V) 30 35 50 35 35 30 45 20 34 26 30 34 35 35 34
CE (V) 35 30 35/25 30/30 35/25 35/20 30/25 15/30 33/20 25/15 23/21 57/35 35/20 27/21 37/23
J Sci Food Agric 2015; 95: 598–606
Determination of Fifteen Sedative Residues in Mutton
www.soci.org
Figure 2. MRM chromatogram of the 15 standard tranquillisers in a blank sample (A) and of the 15 tranquillisers spiked in a mutton sample at 30 μg kg−1 .
601
J Sci Food Agric 2015; 95: 598–606
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
www.soci.org
J Wei et al.
Figure 2. Continued.
602 wileyonlinelibrary.com/jsfa
© 2014 Society of Chemical Industry
J Sci Food Agric 2015; 95: 598–606
Determination of Fifteen Sedative Residues in Mutton
www.soci.org
Figure 2. Continued.
603
J Sci Food Agric 2015; 95: 598–606
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
www.soci.org
J Wei et al.
Table 2. Calibration curves, correlation coefficients, limit of detection (LOD) and limit of quantification (LOQ) of the method developed
Standard sample
RT (min)
Azaperone Xylazine hydrochloride Zolpidem tartrate Carazolol Chlordiazepoxide Droperidol Haloperidol Promethazine hydrochloride Acepromazine maleate Perphenazine Propionylpromazine hydrochloride Chlorpromazine hydrochloride Fluphenazine hydrochloride Nitrazepam Clonazepam
2.82 3.20 4.30 4.85 5.05 5.40 6.63 6.77 6.81 7.70 7.91 8.28 8.56 9.28 9.87
Spiked level (μg kg−1 ) 5–50 5–50 0.5–50 5–50 5–50 0.5–50 5–50 1–50 0.5–50 1–50 5–50 5–50 0.5–50 0.5–50 2–50
Linear equation y = 126.98x + 114.03 y = 41.8489x + 97.8894 y = 853.442x + 195.653 y = 92.67x + 53.114 y = 81.3144x + 138.444 y = 98.795x + 77.9818 y = 42.0712x + 55.6248 y = 783.963x + 517.991 y = 926.957x + 336.5 y = 577.043x + 860.76 y = 49.1929x + 9.7867 y = 97.5618x + 120.861 y = 1584.83x + 623.487 y = 140.141x + 213.812 y = 62.047x + 25.631
Correlation coefficient (r2 ) 0.9974 0.9756 0.9912 0.9978 0.9853 0.9960 0.9756 0.9970 0.9898 0.9864 0.9984 0.9821 0.9639 0.9860 0.9976
LOD (μg kg−1 ) 2.5 2.5 0.5 2.5 2 0.5 2 0.5 0.25 0.5 2.5 2.5 0.5 0.25 2
LOQ (μg kg−1 ) 5 5 1 4 4 1 5 1 0.5 1 4 5 1 1 4
RT, retention time. In each equation, y is the peak area of the quantitative MRM transition, and x is the spiked level.
• 5, 10, 20, 30 and 50 μg kg−1 for azaperone, xylazine hydrochloride, carazolol, chlorpromazine hydrochloride, chlordiazepoxide, propionylpromazine hydrochloride and haloperidol • 0.5, 1, 10, 30 and 50 μg kg−1 for nitrazepam, zolpidem tartrate, acepromazine maleate, droperidol and fluphenazine hydrochloride • 1, 5, 10, 30 and 50 μg kg−1 for perphenazine and promethazine hydrochloride • 2, 10, 20, 30 and 50 μg kg−1 for clonazepam Accuracy and precision were appraised through intra- and inter-day analyses, which were conducted using the fortified mutton samples at three different concentrations (10, 30 and 50 μg kg−1 ). Intra-day analysis was conducted in six replicates at each concentration level, whereas inter-day analysis was performed in three replicates at each concentration level on three consecutive days.
RESULTS AND DISCUSSION
604
Sample preparation The 15 sedative hypnotics belong to four types of drug with different chemical structures so it is very difficult to extract them efficiently by the same method. Several organic solvents were studied for extraction, such as acetonitrile, methanol, ethyl acetate, ammonia–dichloromethane (5:95, v/v), ammonia–ethyl acetate (5:95, v/v), ammonia–methanol (10:90, v/v), ammonia–acetonitrile (10:90, v/v), 0.1 mol L−1 HCl–acetonitrile (5:95, v/v) and water in acetonitrile or methanol in different proportions. Considering the recovery of the analytes and clean-up performance for the interferences, ammonia–acetonitrile (10:90, v/v) was chosen as extraction solution. But the recovery was less than 50% by only using extraction solution. The recovery was increased to over 80% by adding 30 μL 𝛽-glucuronidase/aryl sulfatase to the extraction solution when the spiked samples were extracted. These results indicated that the enzymolysis and alkalescent condition are proper for extracting these analytes. Some drugs which had amino or hydroxyl group in their structures were combined
wileyonlinelibrary.com/jsfa
with glucuronic acid or sulfuric acid with aryl glycosides or ester in the process of metabolism in animals. The bound analytes were dissociated and released by enzymatic hydrolysis before extract. Multiple reaction monitoring ion selection For the confirmation of sedative residues, a minimum of four identification points are required by the commission of the European Union.13 Two and 2.5 identification points were earned by a precursor ion and a transition product using RRLC-MS/MS, respectively. Azaperone, xylazine hydrochloride, carazolol and clonazepam earned 4.5 identification points at multiple reaction monitoring (MRM) mode, respectively. The others earned 7.0 identification points at the same condition, respectively. Thus, all of the analytes could be determined in MRM mode with the ESI+ source during the chromatographic analysis. Optimisation of RRLC-MS/MS parameters The standard working solution, column and mobile phase were examined to achieve the best peak symmetry on the liquid chromatographic analysis. First, 1 μg mL−1 of standard working solution prepared in 0.027 mol L−1 aqueous FA-MeCN (1:1, v/v) was injected into a ZORBAX Rx C18 (250 mm × 3 mm i.d., particle size 5 μm, pH 2.0–8.0, pore size 80 Å; Agilent) column in which a mobile phase (mobile phase A, MeCN; mobile phase B, 0.027 mol L−1 aqueous FA) flowed with a similar gradient to those of the present study and at 1 mL min−1 . However, fluphenazine hydrochloride and perphenazine, chlorpromazine hydrochloride, nitrazepam and clonazepam overlapped together, respectively. A greater portion of the 0.027 mol L−1 aqueous FA in the working solution solvent (0.027 mol L−1 aqueous FA-MeCN, 80:20, v/v) resulted in great improvement on the peak split. All of the analytes were separated completely except xylazine hydrochloride and azaperone. The polarity of solution was increased with the portion of the 0.027 mol L−1 aqueous FA raised in the mobile phase, while the pH was decreased. That suggested that the larger polarity and lower pH were fit for separating these analytes. So, the 0.027 mol L−1 aqueous FA and MeCN (80:20, v/v) were provisionally used as
© 2014 Society of Chemical Industry
J Sci Food Agric 2015; 95: 598–606
Determination of Fifteen Sedative Residues in Mutton
www.soci.org
Table 3. The within-laboratory accuracy (recovery, %) repeatability (RSD, %) and reproducibility (RSD, %) of the method developed, using mutton samples fortified with the various compounds at 10, 30 and 50 μg kg−1 Intra-day analysis, accuracy (repeatability, %, n = 6)
Compound 10 μg kg−1 Xylazine hydrochloride Azaperone Zolpidem tartrate Carazolol Nitrazepam Acepromazine maleate Perphenazine Promethazine hydrochloride Chlorpromazine hydrochloride Droperidol Chlordiazepoxide Haloperidol Propionylpromazine hydrochloride Fluphenazine hydrochloride Clonazepam
88.7 (2.6) 94.1 (10.6) 88.9 (7.8) 99.2 (10.7) 77.5 (11.2) 97.1 (9.8) 80.7 (7.6) 78.2 (8.5) 81.9 (5.5) 89.5 (9.9) 78.7 (7.9) 85.0 (9.1) 94.3 (7.7) 88.7 (6.6) 94.4 (9.3)
30 μg kg−1
50 μg kg−1
10 μg kg−1
30 μg kg−1
50 μg kg−1
88.9 (5.8) 93.7 (5.5) 81.9 (7.6) 87.4 (5.3) 74.1 (3.3) 88.7 (5.7) 78.2 (8.4) 80.7 (9.6) 97.0 (8.2) 88.8 (6.4) 88.0 (6.8) 78.7 (7.4) 109.5 (6.4) 123.9 (9.4) 88.9 (7.4)
85.9 (4.4) 109.5 (8.1) 90.7 (8.4) 87.6 (5.9) 84.4 (8.9) 95.3 (7.8) 88.9 (9.9) 88.9 (6.8) 96.7 (8.4) 97.2 (7.8) 84.4 (6.0) 94.3 (4.8) 93.4 (11.2) 116.8 (9.5) 101.2 (7.7)
88.5 (3.6) 99.5 (8.6) 80.9 (9.8) 90.1 (4.5) 80.4 (8.9) 100.7 (3.1) 85.4 (9.8) 78.7 (2.1) 91.9 (10.9) 86.4 (8.0) 92.8 (6.0) 87.7 (8.4) 91.4 (6.9) 95.5 (7.6) 91.9 (11.4)
90.4 (4.7) 93.3 (7.9) 85.2 (7.4) 94.5 (3.3) 85.5 (8.2) 90.7 (6.6) 82.2 (7.4) 90.4 (5.6) 110.7 (5.9) 79.9 (10.6) 100.3 (3.7) 108.8 (5.6) 92.5 (11.1) 90.0 (4.1) 82.7 (5.1)
82.2 (4.9) 112.8 (3.4) 94.4 (8.2) 85.9 (7.1) 90.4 (10.1) 99.4 (7.3) 89.1 (4.4) 90.0 (10.6) 92.5 (5.0) 102.4 (7.7) 96.2 (7.7) 90.6 (8.1) 97.6 (2.9) 89.1 (4.6) 85.6 (10.8)
the standard working solution solvent. In order to achieve better peak symmetry, the analytes prepared in 0.027 mol L−1 aqueous FA-MeCN (80:20, v/v) were detected using a Gemini-NX C18 (150 mm × 2 mm i.d., particle size 3 μm, pH 1.0–12.0, pore size 110 Å; Phenomenex, Torrance, CA, USA) and Acquity UPLC CSHTM C18 column (100 mm × 2.1 mm i.d., particle size 1.7 μm, pH 1.0-11.0, pore size 130 Å; Waters) columns eluted with the mobile phase (0.027 mol L−1 aqueous FA-MeCN) at 1 mL min−1 . Peaks shapes were better on the Acquity UPLC CSHTM C18 column for all analytes. The appearance of xylazine hydrochloride and azaperone was not ideal. The two compounds may be adsorbed on the column and could not be eluted completely. In order to optimise the chromatographic parameters, the composition of working standard solution was changed to 0.027 mol L−1 acqueous FA-MeCN (90:10, v/v). It was detected on the Acquity UPLC CSHTM C18 column for improving the poor peak shapes of xylazine hydrochloride and azaperone. Peaks sharpness of all analytes was dramatically increased upon elution with 10% MeCN increased to 70% (1–15 min). The working solution was suitable for all analytes seperated when the ratio of 0.027 mol L−1 aqueous AFand MeCN was 90:10. Therefore, an Acquity UPLC CSHTM C18 was chosen as the analytical column. Meanwhile, a mixed solution of 0.027 mol L−1 aqueous FA-MeCN (90:10, v/v) was used for dilution of the standard working solution and redissolution of sample residues. The MRM chromatogram of 15 standard tranquillisers in matrix solution is shown in Fig. 2.
Spiked levels, curves and correlation coefficients are listed in Table 2. The calibration curves showed good linearity within the concentrations of 0.5–50 μg kg−1 of zolpidem tartrate, acepromazine maleate, nitrazepam, fluphenazine hydrochloride and droperidol with the correlation coefficients r2 ≥ 0.9639. Azaperone, xylazine hydrochloride, carazolol, chlorpromazine hydrochloride, propionylpromazine hydrochloride, chlordiazepoxide and haloperidol had good linear relationship within the concentrations of 5–50 μg kg−1 , and the correlation coefficients were from 0.9756 to 0.9984. The linearity ranges of perphenazine and promethazine hydrochloride were 1–50 μg kg−1 . Clonazepam had good linear relationship within the concentrations of 2–50 μg kg−1 . Accuracy was assessed by a recovery % produced in intra- and inter-day analyses. Precision was calculated in terms of intra-day repeatability and inter-day reproducibility, which were expressed as relative standard deviation (RSD). The recovery % was calculated by comparing the results of the intra- and inter-day analyses with the matrix-matched standard working solutions (composite standard intermediate solutions) prepared at the same concentrations in blank extract. The results of intra- and inter-day analyses performed at 10, 30 and 50 μg kg−1 are presented in Table 3. The averages of intra- and inter-day accuracy ranged from 74.1 to 116.8% and from 78.7 to 112.8%, respectively. Repeatability and reproducibility were 2.6–11.2 and 2.1–11.4, respectively. Consequently, the developed method in this study is quite reliable, reproducible and accurate in determining sedatives in mutton. Applicability of the developed method To evaluate the reliability and practicability of the developed method, it was applied to the measurement of sedatives in nine mutton samples purchased from the local market. None of the targeted analytes were detected by the RRLC/MS-MS. The target compounds may not be used and if used, they could be metabolised below the LODs in mutton. Additionally, further researches, including a large number of sample analyses and animal-incurred and tracking experiments are necessary to demonstrate the usage of the targeted drugs and the residue in mutton.
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
605
Method validation The identification of the analytes was based on retention times and accurate mass measurements. Quantification was performed using the peak area of the MRM chromatogram. No interfering peaks were detected at the retention time of all analytes. As given in Table 2, the LODs and LOQs were determined as the analyte concentrations giving peak heights three and 10 times higher than the baseline noise, respectively, via the analysis of the target compounds in blank mutton extracts. All LOD and LOQ values were in ranges of 0.25–2.5 and 0.5–5 μg kg−1 , respectively, which were sufficiently sensitive for the analysis of real samples. Response linearity was evaluated by matrix-matched calibration curves. J Sci Food Agric 2015; 95: 598–606
Inter-day analysis, accuracy (reproducibility, %, n = 9)
www.soci.org
CONCLUSION A method was developed to determine sedatives using RRLC/MS-MS in mutton. RRLC/MS-MS provided enough analytical sensitivity and accuracy for the analytes in samples originating from mutton. The results of validation criteria supported the reliability and practicability of the method developed herein. The method can meet the analysis requirement of veterinary drug residues, providing a favourable tool for detecting other sedatives in animal derived food.
ACKNOWLEDGEMENT This study was financially supported by the Natural Science Foundation of Gansu (1208RJZA156, 1010RJZA202). We also thank Mr Zhou and Ms Xie (Gansu Entry-Exit Inspection and Quarantine Bureau, Gansu, China) for their enthusiastic help in providing us data analysis.
REFERENCES 1 Wu NP, Ban FG, Peng L, Zhou HX, Liu ZT and Liu HW, Screening method for 11 sedatives in feedstuff by ultra performance liquid chromatography–quadrupole time-of-flight mass spectrometry. J Chin Mass Spectrom Soc 33:94–99 (2012). 2 Cheng LL, Zhang YJ, Shen JZ, Wu CM and Zhang SX, GC–MS method for simultaneous determination of four sedative hypnotic residues in swine tissues. Chromatographia 71:155–158 (2010). 3 Sun L, Zhang L, Xu Q, Wang SH and Wang X, Determination of ten sedative residues in pork and kidney by ultra performance liquid chromatography–tandem mass spectrometry. Chin J Chromatogr 28:38–42 (2010).
J Wei et al.
4 Yan LJ, Zhang J, Pan CS, Lin LY, Zhang XY and Shen HQ, High throughput screening of tranquilizers in dairy products using ultra performance liquid chromatography coupled to high resolution time-of-flight mass spectrometry. Chin J Anal Chem 41:31–35 (2013). 5 Zhang LQ, Wu PG, Jin Q, Zhang YM, Wang XF and Ren R, Research progress on detection technique of chromatographic analysis of sedative residues. Chin J Health Lab Technol 23:1634–1648 (2013). 6 Wang ML, Guo B, Huang ZQ, Duan JM, Chen ZF, Chen B, et al, Improved compatibility of liquid chromatography with electrospray tandem mass spectrometry for tracing occurrence of barbital homologous residues in animal tissues. J Chromatogr A 1217:2821–2831(2010). 7 The Ministry of Agriculture of the People’s Republic of China, Notification No. 235 of MOA, Maximum residue limits of veterinary drug residues in food of animal origin (2002). 8 Zhang CM, Huang DM, Cheng QX and Huang ZG, Determination of sedative component in Chinese medicines by high resolution chromatography–electron impact–mass spectrometry. J Anal Sci 2:13–17 (2012). 9 Choi JH, Lamshoft M, Zuhlke S, Park KH, Shim JH and Spiteller M, Determination of sedatives and adrenergic blockers in blood meal using accelerated solvent extraction and Orbitrap mass spectrometry. J Chromatogr A 1260:111–119 (2012). 10 Liang XH, Dai ZY and Liu WE, Simultaneous determination of 5 sedative hypnotics in human plasma by reversed phase high-performance liquid chromatography. J Cent South Univ (Med Sci) 34:689–693 (2009). 11 Lee S, Han E, In S, Choi H, Chung H and Chung KH, Determination of illegally abused sedative-hypnotics in hair samples from drug offenders. J Anal Toxicol 35:312–315 (2011). 12 Wang LP, Li X, Sun Y, Zhao HX, Qiu YM, Zhong WK, et al., Determination of four benzodiazepines residues in pork using gas chromatography–mass spectrometry. Chin J Anal Chem 33:951–954 (2005). 13 European Communities, Commission Decision 2002/657/EC. Off J Eur Commun 221:8–36 (2002).
606 wileyonlinelibrary.com/jsfa
© 2014 Society of Chemical Industry
J Sci Food Agric 2015; 95: 598–606
Revised: 26 May 2014
Accepted article published: 20 June 2014
Published online in Wiley Online Library: 23 July 2014
(wileyonlinelibrary.com) DOI 10.1002/jsfa.6794
Determination of 15 sedative residues in mutton by rapid resolution liquid chromatography–tandem mass spectrometry Jinmei Wei, Yuzhu Luo,* Li Zhang and Suli Fang Abstract BACKGROUND: The use of xenobiotic compounds in animal husbandry has given rise to consumer anxieties regarding residual risk and food safety. Thus, animal tissues have become main samples for residue analysis and food safety for sedatives. In this study, a rapid resolution liquid chromatography–tandem mass spectrometry (RRLC-MS/MS) method was established for the determination of 15 sedatives residues in mutton. RESULTS: After enzymolysis, sedatives residues in mutton were extracted by ammonium hydroxide–acetonitrile (10:90, v/v) and determined by RRLC-MS/MS with quantification by standard curve method. The calibration curves showed good linearity within the concentrations of 0.5–50 𝛍g kg−1 with the correlation coefficients (r2 ) ranged from 0.9639 to 0.9984. The limits of detection (LODs) and quantification (LOQs) were 0.25–2.5 and 0.5–5 𝛍g kg−1 , respectively. The average recoveries of spikes samples were in the ranges of 74.1–116.8% with relative standard deviations of intra- and inter-day ranged from 2.6% to 11.2% and from 2.1% to 11.4%, respectively. CONCLUSION: This method is simple, sensitive and accurate in the determination of sedative residues. © 2014 Society of Chemical Industry Keywords: rapid resolution liquid chromatography–tandem mass spectrometry (RRLC-MS/MS); sedatives; residues; mutton
INTRODUCTION
598
The sedative hypnotics have been used for a long time as animal tranquillisers to prevent stress and weight loss during their transportation, and they are added to animal feeds occasionally to reduce the animal’s reaction, slowing down the metabolic processes for enhancing their growth rate.1 – 3 Thus, animal tissues have become the main samples for residue analysis and food safety for sedatives. Xylazine hydrochloride, azaperone, zolpidem tartrate, carazolol, chlordiazepoxide, droperidol, haloperidol, promethazine hydrochloride, acepromazine maleate, perphenazine, propionylpromazine hydrochlorid, chlorpromazine hydrochloride, nitrazepam, fluphenazine hydrochloride and clonazepam are the most commonly used drugs. These 15 compounds belong to four families in chemical structure: benzodiazepines, phenothiazines, butyrophenones and imidazopyridines. Their chemical structures are shown in Fig. 1. The use of xenobiotic compounds in animal husbandry has given rise to consumer anxieties regarding residual risk and food safety. These sedative hypnotics have been found to be the probable causes of over-dosage dependence, excessive sedation, anaesthesia, coma and death.4 The residues of these compounds in foodstuff of animal origin may represent a potential risk to consumers and many countries have regulated their use in veterinary medicine, and residues of some sedatives are prohibited in animal derived food.5,6 In China, the Ministry of Agriculture of the People’s Republic of China has introduced laws regarding the maximum residue limits of veterinary drug residues in food of animal origin. Chlorpromazine, methaqualone and diazepam were not allowed J Sci Food Agric 2015; 95: 598–606
in any animal food, and xylazine was not allowed in milk as well.7 There is no relevant national standard for the maximum residue limits of other sedatives in mutton yet. A number of methods have been reported for the identification and determination of these compounds in biological samples in clinical and forensic toxicology,8 – 11 whereas only a few methods have been published for analysis of these drugs in animal tissues. Cheng et al.2 and Wang et al.12 reported a gas chromatography–mass spectrometry (GC-MS) method to determine diazepam, methaqualone, chlorpromazine and promethazine in swine tissues and benzodiazepine residues in pork by using multi-walled carbon nanotube solid phase, respectively. The limits of detection and quantification in swine muscle and liver tissues ranged from 0.2 to 0.3 μg kg−1 and 0.5 to 1 μg kg−1 , respectively.2 The limits of detection were 2 μg kg−1 for diazepam and 10 μg kg−1 for estazolam, alprazolam and triazolam in pork.12 Sun et al.3 determined 10 sedative residues in pork and kidney by UPLC-MS/MS. The limits of detection of the 10 sedatives were 0.5 μg kg−1 , and the limit of quantification was 1 μg kg−1 . To the best of our knowledge, so far, no studies have been reported in which determined these drugs have been determined in mutton.
∗
Correspondence to: Yuzhu Luo, Gansu Key Laboratory of Herbivorous Animal Biotechnology, Gansu Agricultural University, 730070 Lanzhou,China. E-mail: [email protected] Gansu Agricultural University/Gansu Key Laboratory of Herbivorous Animal Biotechnology, 730070 Lanzhou, China
www.soci.org
© 2014 Society of Chemical Industry
Determination of Fifteen Sedative Residues in Mutton
www.soci.org
Figure 1. The chemical structures of the 15 sedatives tested.
The purpose of the present study was to develop a reliable and sensitive method for the simultaneous detection and determination of 15 sedative residues in mutton by RRLC-MS/MS.
MATERIALS AND METHODS
J Sci Food Agric 2015; 95: 598–606
Standard solutions All standard stock solutions were prepared in MeOH at a concentration of 200 μg mL−1 . Composite standard intermediate solutions were prepared via dilution of the stock solutions in 0.027 mol L−1 aqueous FA-MeCN (90:10, v/v) at 50, 30, 10, 5, 2, 1 and 0.5 μg L−1 . Matrix-matched standard working solutions were prepared in
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
599
Chemicals and reagents The 15 compounds tested were as follows: zolpidem tartrate (purity 99%), chlordiazepoxide (≥99%), haloperidol (99%), promethazine hydrochloride (≥98%), prochlorperazine (99%), chlorpromazine hydrochloride (99%), nitrazepam (99%), fluphenazine hydrochloride (≥98%) and clonazepam (99%) were purchased from the National Institute for Food and Drug Control (Beijing, China). Xylazine hydrochloride (≥99%), carazolol (≥99%) and acepromazine maleate (≥99%) were obtained from Dr. Ehrenstorfer GmbH (Augsburg, Germany). Propionylpromazine
hydrochloride (≥99%) and azaperone (≥99%) were supplied by Sigma-Aldrich (Taufkirchen, Germany). Droperidol (≥99%) was provided by US Pharmacopeial Convention (Rockville, MD, USA). Acetonitrile (MeCN) and methanol (MeOH) were purchased from Merck (Darmstadt, Germany), 𝛽-glucuronidase/aryl sulfatase, formic acid (FA) and ammonium hydroxide from Sigma-Aldrich. Ultrapure water was obtained from a Milli-Q water purification system (Billerica, MA, USA). All solvents and reagents used were of HPLC or analytical grade, unless otherwise stated.
www.soci.org blank sample extracts, which were extracted from blank mutton raised by our laboratory and confirmed in advance not to contain any of the tested analytes. All of the standard solutions were stored at −18 ∘ C in a dark amber bottle. Sample preparation Blank mutton was from sheep free of drugs raised by our laboratory, and actual mutton was bought from a local market (Lanzhou, China). Two types of sample were minced, mixed together, homogenised and stored at −18 ∘ C for later use, respectively. The samples were removed from the freezer and allowed to reach room temperature prior to analysis. A 5 g sample was weighed in a polypropylene centrifuge tube (50 mL) to which 15 mL ammonia–acetonitrile (10:90, v/v), and 30 μL 𝛽-glucuronidase/aryl sulfatase was added. The sample was vortexed and shaked for 10 h at 37 ∘ C in a dark place. Then it was extracted for 10 min at 30 ∘ C with ultrasonication and then centrifuged at 2810 × g for 5 min. The supernatant was transferred into a polypropylene centrifuge tube (50 mL). Another 15 mL 0.26 mol L−1 ammonium hydroxide–MeCN was added to the residue and extracted for 10 min at 30 ∘ C with ultrasonication. The extraction step was repeated twice and the ammonium hydroxide–MeCN extraction was collected into the polypropylene centrifuge tube above. Five grams of sodium chloride was added to the polypropylene centrifuge tube filled with the extraction. The sample was vortexed and allowed to stand for 10 min. Most of lipids in the extracts were removed and the rest extracts were evaporated to dryness at 37 ∘ C. The residue was redissolved in 1 mL 0.027 mol L−1 aqueous FA-MeCN (90:10, v/v) and filtered with a 0.22 μm organic membrane. The filtrate was subjected to analysis by RRLC-MS/MS. RRLC-MS/MS analysis RRLC was carried out using an Agilent Technologies 1290 series system equipped with a G1322A degasser, a G1312B SL binary pump, a G1357D high-performance autosampler (HiP ALS SL+) (Agilent Technologies, Palo Alto, CA, USA). Separation was achieved on an Acquity UPLC CSHTM C18 column (100 mm × 2.1 mm i.d., particle size 1.7 μm; Waters, Milford, MA, USA) that had been maintained at 30 ∘ C in a thermostated column
J Wei et al.
compartment (model G1316B SL, Agilent Technologies, Palo Alto, CA, USA). The binary solvent system consisted of MeCN (mobile phase A) and 0.027 mol L−1 FA in ultrapure water (mobile phase B).The linear mobile phase gradient started at 10% A, increased to 70% A (1–15 min) and ramped back down to 10% A (15–16 min). The flow rate was 0.3 mL min−1 , and the injection volume was 5 μL. A 6460 triple-quadrupole mass spectrometer (Agilent Technologies) was operated in multiple reaction monitoring (MRM) mode with the positive ESI source (Agilent G1948B) for all the target analytes during the chromatographic analysis. For the optimisation of MS/MS parameters, composite standard working solution (1 μg mL−1 ) diluted in a mixture of 0.027 mol L−1 aqueous FA and MeCN (90:10, v/v) was introduced into the ESI+ source at an infusion flow rate of 5 μL min−1 . All the parameters were duly applied in order to find the optimal conditions providing the highest intensities of the analytes. Finally, the instrument was operated with the capillary voltage at +3.2 kV, and charging voltage at +1 kV. Nitrogen was used as nebuliser gas of 0.31 MPa, a carrier gas of 10 L min−1 at 350 ∘ C, and a sheath gas of 11 L min−1 at 350 ∘ C. The conditions used for the ionisation source were set as follows: source temperature, 120 ∘ C; and desolvation temperature, 350 ∘ C. The cone and desolvation gas flows were 50 and 500 L h−1 , respectively. The parameters optimised of cone voltage, collision energy, dwell time, qualitative ion pair and quantitative ion pair for each analyte during infusion are listed in Table 1. An Agilent Mass Hunter workstation (Agilent Technologies) was used for the control of equipment, data acquisition and analysis.
Method validation The present study was validated using a within-laboratory protocol. To determine the limit of detection (LOD) and limit of quantification (LOQ), composite standard working solutions were prepared in blank sample extract to find the concentrations that would give signal-to-noise ratios of 3 and 10, respectively. The fortified samples were prepared by spiking the composite standard intermediate solutions into blank samples at different concentrations for matrix-matched calibration curves. The concentrations were:
Table 1. The parameters optimised of cone voltage (CV), collision energy (CE), dwell time (DT), qualitative (Q1) and quantitative (Q2) MRM transition for each analyte Analyte
600
Xylazine hydrochloride Azaperone Zolpidem tartrate Carazolol Nitrazepam Acepromazine maleate Perphenazine Promethazine hydrochloride Chlorpromazine hydrochloride Droperidol Chlordiazepoxide Haloperidol Propionylpromazine hydrochlorid Fluphenazine hydrochloride Clonazepam
wileyonlinelibrary.com/jsfa
Q1 (m/z) 221.1/163.75 328.4/121 308.3/263.1 299.5/116.2 282.3/179.9 327.4/57.9 404.38/143 285.2/85.8 319.11/57.99 380.21/165.01 300.32/227.03 376.4/122.8 341.2/57.7 438.33/143.04 316.1/214.03
Q2 (m/z) 221.1/163.75 328.4/121 308.3/235.1 299.5/116.2 282.3/236.2 327.4/85.91 404.38/171.14 285.2/198 319.11/86.06 380.21/194.22 300.32/282.07 376.4/165.4 341.2/85.7 438.33/171.1 316.1/214.03
© 2014 Society of Chemical Industry
DT (ms) 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05
CV (V) 30 35 50 35 35 30 45 20 34 26 30 34 35 35 34
CE (V) 35 30 35/25 30/30 35/25 35/20 30/25 15/30 33/20 25/15 23/21 57/35 35/20 27/21 37/23
J Sci Food Agric 2015; 95: 598–606
Determination of Fifteen Sedative Residues in Mutton
www.soci.org
Figure 2. MRM chromatogram of the 15 standard tranquillisers in a blank sample (A) and of the 15 tranquillisers spiked in a mutton sample at 30 μg kg−1 .
601
J Sci Food Agric 2015; 95: 598–606
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
www.soci.org
J Wei et al.
Figure 2. Continued.
602 wileyonlinelibrary.com/jsfa
© 2014 Society of Chemical Industry
J Sci Food Agric 2015; 95: 598–606
Determination of Fifteen Sedative Residues in Mutton
www.soci.org
Figure 2. Continued.
603
J Sci Food Agric 2015; 95: 598–606
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
www.soci.org
J Wei et al.
Table 2. Calibration curves, correlation coefficients, limit of detection (LOD) and limit of quantification (LOQ) of the method developed
Standard sample
RT (min)
Azaperone Xylazine hydrochloride Zolpidem tartrate Carazolol Chlordiazepoxide Droperidol Haloperidol Promethazine hydrochloride Acepromazine maleate Perphenazine Propionylpromazine hydrochloride Chlorpromazine hydrochloride Fluphenazine hydrochloride Nitrazepam Clonazepam
2.82 3.20 4.30 4.85 5.05 5.40 6.63 6.77 6.81 7.70 7.91 8.28 8.56 9.28 9.87
Spiked level (μg kg−1 ) 5–50 5–50 0.5–50 5–50 5–50 0.5–50 5–50 1–50 0.5–50 1–50 5–50 5–50 0.5–50 0.5–50 2–50
Linear equation y = 126.98x + 114.03 y = 41.8489x + 97.8894 y = 853.442x + 195.653 y = 92.67x + 53.114 y = 81.3144x + 138.444 y = 98.795x + 77.9818 y = 42.0712x + 55.6248 y = 783.963x + 517.991 y = 926.957x + 336.5 y = 577.043x + 860.76 y = 49.1929x + 9.7867 y = 97.5618x + 120.861 y = 1584.83x + 623.487 y = 140.141x + 213.812 y = 62.047x + 25.631
Correlation coefficient (r2 ) 0.9974 0.9756 0.9912 0.9978 0.9853 0.9960 0.9756 0.9970 0.9898 0.9864 0.9984 0.9821 0.9639 0.9860 0.9976
LOD (μg kg−1 ) 2.5 2.5 0.5 2.5 2 0.5 2 0.5 0.25 0.5 2.5 2.5 0.5 0.25 2
LOQ (μg kg−1 ) 5 5 1 4 4 1 5 1 0.5 1 4 5 1 1 4
RT, retention time. In each equation, y is the peak area of the quantitative MRM transition, and x is the spiked level.
• 5, 10, 20, 30 and 50 μg kg−1 for azaperone, xylazine hydrochloride, carazolol, chlorpromazine hydrochloride, chlordiazepoxide, propionylpromazine hydrochloride and haloperidol • 0.5, 1, 10, 30 and 50 μg kg−1 for nitrazepam, zolpidem tartrate, acepromazine maleate, droperidol and fluphenazine hydrochloride • 1, 5, 10, 30 and 50 μg kg−1 for perphenazine and promethazine hydrochloride • 2, 10, 20, 30 and 50 μg kg−1 for clonazepam Accuracy and precision were appraised through intra- and inter-day analyses, which were conducted using the fortified mutton samples at three different concentrations (10, 30 and 50 μg kg−1 ). Intra-day analysis was conducted in six replicates at each concentration level, whereas inter-day analysis was performed in three replicates at each concentration level on three consecutive days.
RESULTS AND DISCUSSION
604
Sample preparation The 15 sedative hypnotics belong to four types of drug with different chemical structures so it is very difficult to extract them efficiently by the same method. Several organic solvents were studied for extraction, such as acetonitrile, methanol, ethyl acetate, ammonia–dichloromethane (5:95, v/v), ammonia–ethyl acetate (5:95, v/v), ammonia–methanol (10:90, v/v), ammonia–acetonitrile (10:90, v/v), 0.1 mol L−1 HCl–acetonitrile (5:95, v/v) and water in acetonitrile or methanol in different proportions. Considering the recovery of the analytes and clean-up performance for the interferences, ammonia–acetonitrile (10:90, v/v) was chosen as extraction solution. But the recovery was less than 50% by only using extraction solution. The recovery was increased to over 80% by adding 30 μL 𝛽-glucuronidase/aryl sulfatase to the extraction solution when the spiked samples were extracted. These results indicated that the enzymolysis and alkalescent condition are proper for extracting these analytes. Some drugs which had amino or hydroxyl group in their structures were combined
wileyonlinelibrary.com/jsfa
with glucuronic acid or sulfuric acid with aryl glycosides or ester in the process of metabolism in animals. The bound analytes were dissociated and released by enzymatic hydrolysis before extract. Multiple reaction monitoring ion selection For the confirmation of sedative residues, a minimum of four identification points are required by the commission of the European Union.13 Two and 2.5 identification points were earned by a precursor ion and a transition product using RRLC-MS/MS, respectively. Azaperone, xylazine hydrochloride, carazolol and clonazepam earned 4.5 identification points at multiple reaction monitoring (MRM) mode, respectively. The others earned 7.0 identification points at the same condition, respectively. Thus, all of the analytes could be determined in MRM mode with the ESI+ source during the chromatographic analysis. Optimisation of RRLC-MS/MS parameters The standard working solution, column and mobile phase were examined to achieve the best peak symmetry on the liquid chromatographic analysis. First, 1 μg mL−1 of standard working solution prepared in 0.027 mol L−1 aqueous FA-MeCN (1:1, v/v) was injected into a ZORBAX Rx C18 (250 mm × 3 mm i.d., particle size 5 μm, pH 2.0–8.0, pore size 80 Å; Agilent) column in which a mobile phase (mobile phase A, MeCN; mobile phase B, 0.027 mol L−1 aqueous FA) flowed with a similar gradient to those of the present study and at 1 mL min−1 . However, fluphenazine hydrochloride and perphenazine, chlorpromazine hydrochloride, nitrazepam and clonazepam overlapped together, respectively. A greater portion of the 0.027 mol L−1 aqueous FA in the working solution solvent (0.027 mol L−1 aqueous FA-MeCN, 80:20, v/v) resulted in great improvement on the peak split. All of the analytes were separated completely except xylazine hydrochloride and azaperone. The polarity of solution was increased with the portion of the 0.027 mol L−1 aqueous FA raised in the mobile phase, while the pH was decreased. That suggested that the larger polarity and lower pH were fit for separating these analytes. So, the 0.027 mol L−1 aqueous FA and MeCN (80:20, v/v) were provisionally used as
© 2014 Society of Chemical Industry
J Sci Food Agric 2015; 95: 598–606
Determination of Fifteen Sedative Residues in Mutton
www.soci.org
Table 3. The within-laboratory accuracy (recovery, %) repeatability (RSD, %) and reproducibility (RSD, %) of the method developed, using mutton samples fortified with the various compounds at 10, 30 and 50 μg kg−1 Intra-day analysis, accuracy (repeatability, %, n = 6)
Compound 10 μg kg−1 Xylazine hydrochloride Azaperone Zolpidem tartrate Carazolol Nitrazepam Acepromazine maleate Perphenazine Promethazine hydrochloride Chlorpromazine hydrochloride Droperidol Chlordiazepoxide Haloperidol Propionylpromazine hydrochloride Fluphenazine hydrochloride Clonazepam
88.7 (2.6) 94.1 (10.6) 88.9 (7.8) 99.2 (10.7) 77.5 (11.2) 97.1 (9.8) 80.7 (7.6) 78.2 (8.5) 81.9 (5.5) 89.5 (9.9) 78.7 (7.9) 85.0 (9.1) 94.3 (7.7) 88.7 (6.6) 94.4 (9.3)
30 μg kg−1
50 μg kg−1
10 μg kg−1
30 μg kg−1
50 μg kg−1
88.9 (5.8) 93.7 (5.5) 81.9 (7.6) 87.4 (5.3) 74.1 (3.3) 88.7 (5.7) 78.2 (8.4) 80.7 (9.6) 97.0 (8.2) 88.8 (6.4) 88.0 (6.8) 78.7 (7.4) 109.5 (6.4) 123.9 (9.4) 88.9 (7.4)
85.9 (4.4) 109.5 (8.1) 90.7 (8.4) 87.6 (5.9) 84.4 (8.9) 95.3 (7.8) 88.9 (9.9) 88.9 (6.8) 96.7 (8.4) 97.2 (7.8) 84.4 (6.0) 94.3 (4.8) 93.4 (11.2) 116.8 (9.5) 101.2 (7.7)
88.5 (3.6) 99.5 (8.6) 80.9 (9.8) 90.1 (4.5) 80.4 (8.9) 100.7 (3.1) 85.4 (9.8) 78.7 (2.1) 91.9 (10.9) 86.4 (8.0) 92.8 (6.0) 87.7 (8.4) 91.4 (6.9) 95.5 (7.6) 91.9 (11.4)
90.4 (4.7) 93.3 (7.9) 85.2 (7.4) 94.5 (3.3) 85.5 (8.2) 90.7 (6.6) 82.2 (7.4) 90.4 (5.6) 110.7 (5.9) 79.9 (10.6) 100.3 (3.7) 108.8 (5.6) 92.5 (11.1) 90.0 (4.1) 82.7 (5.1)
82.2 (4.9) 112.8 (3.4) 94.4 (8.2) 85.9 (7.1) 90.4 (10.1) 99.4 (7.3) 89.1 (4.4) 90.0 (10.6) 92.5 (5.0) 102.4 (7.7) 96.2 (7.7) 90.6 (8.1) 97.6 (2.9) 89.1 (4.6) 85.6 (10.8)
the standard working solution solvent. In order to achieve better peak symmetry, the analytes prepared in 0.027 mol L−1 aqueous FA-MeCN (80:20, v/v) were detected using a Gemini-NX C18 (150 mm × 2 mm i.d., particle size 3 μm, pH 1.0–12.0, pore size 110 Å; Phenomenex, Torrance, CA, USA) and Acquity UPLC CSHTM C18 column (100 mm × 2.1 mm i.d., particle size 1.7 μm, pH 1.0-11.0, pore size 130 Å; Waters) columns eluted with the mobile phase (0.027 mol L−1 aqueous FA-MeCN) at 1 mL min−1 . Peaks shapes were better on the Acquity UPLC CSHTM C18 column for all analytes. The appearance of xylazine hydrochloride and azaperone was not ideal. The two compounds may be adsorbed on the column and could not be eluted completely. In order to optimise the chromatographic parameters, the composition of working standard solution was changed to 0.027 mol L−1 acqueous FA-MeCN (90:10, v/v). It was detected on the Acquity UPLC CSHTM C18 column for improving the poor peak shapes of xylazine hydrochloride and azaperone. Peaks sharpness of all analytes was dramatically increased upon elution with 10% MeCN increased to 70% (1–15 min). The working solution was suitable for all analytes seperated when the ratio of 0.027 mol L−1 aqueous AFand MeCN was 90:10. Therefore, an Acquity UPLC CSHTM C18 was chosen as the analytical column. Meanwhile, a mixed solution of 0.027 mol L−1 aqueous FA-MeCN (90:10, v/v) was used for dilution of the standard working solution and redissolution of sample residues. The MRM chromatogram of 15 standard tranquillisers in matrix solution is shown in Fig. 2.
Spiked levels, curves and correlation coefficients are listed in Table 2. The calibration curves showed good linearity within the concentrations of 0.5–50 μg kg−1 of zolpidem tartrate, acepromazine maleate, nitrazepam, fluphenazine hydrochloride and droperidol with the correlation coefficients r2 ≥ 0.9639. Azaperone, xylazine hydrochloride, carazolol, chlorpromazine hydrochloride, propionylpromazine hydrochloride, chlordiazepoxide and haloperidol had good linear relationship within the concentrations of 5–50 μg kg−1 , and the correlation coefficients were from 0.9756 to 0.9984. The linearity ranges of perphenazine and promethazine hydrochloride were 1–50 μg kg−1 . Clonazepam had good linear relationship within the concentrations of 2–50 μg kg−1 . Accuracy was assessed by a recovery % produced in intra- and inter-day analyses. Precision was calculated in terms of intra-day repeatability and inter-day reproducibility, which were expressed as relative standard deviation (RSD). The recovery % was calculated by comparing the results of the intra- and inter-day analyses with the matrix-matched standard working solutions (composite standard intermediate solutions) prepared at the same concentrations in blank extract. The results of intra- and inter-day analyses performed at 10, 30 and 50 μg kg−1 are presented in Table 3. The averages of intra- and inter-day accuracy ranged from 74.1 to 116.8% and from 78.7 to 112.8%, respectively. Repeatability and reproducibility were 2.6–11.2 and 2.1–11.4, respectively. Consequently, the developed method in this study is quite reliable, reproducible and accurate in determining sedatives in mutton. Applicability of the developed method To evaluate the reliability and practicability of the developed method, it was applied to the measurement of sedatives in nine mutton samples purchased from the local market. None of the targeted analytes were detected by the RRLC/MS-MS. The target compounds may not be used and if used, they could be metabolised below the LODs in mutton. Additionally, further researches, including a large number of sample analyses and animal-incurred and tracking experiments are necessary to demonstrate the usage of the targeted drugs and the residue in mutton.
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
605
Method validation The identification of the analytes was based on retention times and accurate mass measurements. Quantification was performed using the peak area of the MRM chromatogram. No interfering peaks were detected at the retention time of all analytes. As given in Table 2, the LODs and LOQs were determined as the analyte concentrations giving peak heights three and 10 times higher than the baseline noise, respectively, via the analysis of the target compounds in blank mutton extracts. All LOD and LOQ values were in ranges of 0.25–2.5 and 0.5–5 μg kg−1 , respectively, which were sufficiently sensitive for the analysis of real samples. Response linearity was evaluated by matrix-matched calibration curves. J Sci Food Agric 2015; 95: 598–606
Inter-day analysis, accuracy (reproducibility, %, n = 9)
www.soci.org
CONCLUSION A method was developed to determine sedatives using RRLC/MS-MS in mutton. RRLC/MS-MS provided enough analytical sensitivity and accuracy for the analytes in samples originating from mutton. The results of validation criteria supported the reliability and practicability of the method developed herein. The method can meet the analysis requirement of veterinary drug residues, providing a favourable tool for detecting other sedatives in animal derived food.
ACKNOWLEDGEMENT This study was financially supported by the Natural Science Foundation of Gansu (1208RJZA156, 1010RJZA202). We also thank Mr Zhou and Ms Xie (Gansu Entry-Exit Inspection and Quarantine Bureau, Gansu, China) for their enthusiastic help in providing us data analysis.
REFERENCES 1 Wu NP, Ban FG, Peng L, Zhou HX, Liu ZT and Liu HW, Screening method for 11 sedatives in feedstuff by ultra performance liquid chromatography–quadrupole time-of-flight mass spectrometry. J Chin Mass Spectrom Soc 33:94–99 (2012). 2 Cheng LL, Zhang YJ, Shen JZ, Wu CM and Zhang SX, GC–MS method for simultaneous determination of four sedative hypnotic residues in swine tissues. Chromatographia 71:155–158 (2010). 3 Sun L, Zhang L, Xu Q, Wang SH and Wang X, Determination of ten sedative residues in pork and kidney by ultra performance liquid chromatography–tandem mass spectrometry. Chin J Chromatogr 28:38–42 (2010).
J Wei et al.
4 Yan LJ, Zhang J, Pan CS, Lin LY, Zhang XY and Shen HQ, High throughput screening of tranquilizers in dairy products using ultra performance liquid chromatography coupled to high resolution time-of-flight mass spectrometry. Chin J Anal Chem 41:31–35 (2013). 5 Zhang LQ, Wu PG, Jin Q, Zhang YM, Wang XF and Ren R, Research progress on detection technique of chromatographic analysis of sedative residues. Chin J Health Lab Technol 23:1634–1648 (2013). 6 Wang ML, Guo B, Huang ZQ, Duan JM, Chen ZF, Chen B, et al, Improved compatibility of liquid chromatography with electrospray tandem mass spectrometry for tracing occurrence of barbital homologous residues in animal tissues. J Chromatogr A 1217:2821–2831(2010). 7 The Ministry of Agriculture of the People’s Republic of China, Notification No. 235 of MOA, Maximum residue limits of veterinary drug residues in food of animal origin (2002). 8 Zhang CM, Huang DM, Cheng QX and Huang ZG, Determination of sedative component in Chinese medicines by high resolution chromatography–electron impact–mass spectrometry. J Anal Sci 2:13–17 (2012). 9 Choi JH, Lamshoft M, Zuhlke S, Park KH, Shim JH and Spiteller M, Determination of sedatives and adrenergic blockers in blood meal using accelerated solvent extraction and Orbitrap mass spectrometry. J Chromatogr A 1260:111–119 (2012). 10 Liang XH, Dai ZY and Liu WE, Simultaneous determination of 5 sedative hypnotics in human plasma by reversed phase high-performance liquid chromatography. J Cent South Univ (Med Sci) 34:689–693 (2009). 11 Lee S, Han E, In S, Choi H, Chung H and Chung KH, Determination of illegally abused sedative-hypnotics in hair samples from drug offenders. J Anal Toxicol 35:312–315 (2011). 12 Wang LP, Li X, Sun Y, Zhao HX, Qiu YM, Zhong WK, et al., Determination of four benzodiazepines residues in pork using gas chromatography–mass spectrometry. Chin J Anal Chem 33:951–954 (2005). 13 European Communities, Commission Decision 2002/657/EC. Off J Eur Commun 221:8–36 (2002).
606 wileyonlinelibrary.com/jsfa
© 2014 Society of Chemical Industry
J Sci Food Agric 2015; 95: 598–606
