Journal of Chromatography B, 945–946 (2014) 110–114
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Simultaneous determination of -lactam antibiotics and -lactamase inhibitors in bovine milk by ultra performance liquid chromatography-tandem mass spectrometry Nasi Li a,b , Feng Feng b , Bingcheng Yang a , Pingping Jiang b , Xiaogang Chu b,∗ a b
School of Pharmacy, East China University of Science and Technology, Shanghai 200237, China Institute of Food Safety, Chinese Academy of Inspection and Quarantine, Beijing 100123, China
a r t i c l e
i n f o
Article history: Received 19 August 2013 Received in revised form 15 November 2013 Accepted 21 November 2013 Available online 26 November 2013 Keywords: -Lactam antibiotics -Lactamase inhibitors Ultra-high performance liquid chromatography Tandem mass spectrometry Bovine milk
a b s t r a c t An ultra performance liquid chromatography-tandem mass spectrometric (UPLC-MS/MS) method has been developed for the simultaneous determination of four -lactam antibiotics (amoxicillin, ampicillin, cefotaxime, and cefoperazone) and two -lactamase inhibitors (tazobactam, sulbactam) in bovine milk. The analytes were extracted with water from bovine milk and purified with Oasis HLB solid phase extraction (SPE) cartridges. The analytes were determined in less than 3 min by UPLC-MS/MS in positive and negative electrospray ionization (ESI) modes, separately. The method was linear over the range of 1–100 g/L for tazobactam, sulbactam, ampicillin, and cefoperazone, and 2–100 g/L for amoxicillin and cefotaxime. The recoveries for all six analytes in bovine milk ranged from 82.5 to 98.3%. The limits of detection and the limits of quantitation were 0.1–0.2 g/L and 0.3–0.5 g/L, respectively. The intra- and inter-day precisions were less than 6% for each compound. © 2013 Elsevier B.V. All rights reserved.
1. Introduction -Lactam antibiotics, such as amoxicillin (AMOX), ampicillin (AMP), cefoperazone (CFP), and cefotaxime (CFT), are widely used for the treatment of bacterial infections in human beings and animals. Their wide application in animals represents a potential hazard because residues of these antibiotics may persist in edible tissues or foodstuffs, such as bovine milk [1–3]. -Lactamase inhibitors, such as sulbactam (SUL) and tazobactam (TAZ), are molecules used in conjunction with -lactam antibiotics to extend their spectrum of activity [4–7]. For example, the commercial veterinary drug composed of AMOX and SUL has been used for the treatment of bovine mastitis [8]. To keep these drugs under regulation and ensure the safety of bovine milk, it is necessary to develop analytical methods for the determination of -lactam antibiotics and -lactamase inhibitors in bovine milk. Several methods have been developed for the determination of -lactams in bovine milk [1,9–12]. The maximum residue limits (MRLs) for -lactams in bovine milk have been regulated in several countries. The European Union (EU) has established an MRL of 4, 4, and 50 g/L for amoxicillin (AMOX), amipicillin (AMP), and
∗ Corresponding author. Tel.: +86 10 85770775; fax: +86 10 85770775. E-mail address: [email protected] (X. Chu). 1570-0232/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jchromb.2013.11.044
cefoperazone (CFP) in bovine milk, respectively [13]. The USA has adopted an MRL of 10 g/L for AMOX in bovine milk [14]. However, methods developed for the determination of -lactamase inhibitors are limited with the few literatures focus on human plasma [6,7]. It is known that the matrix of human plasma is much different from that of bovine milk which contains significant amounts of proteins, saturated fat, and calcium. These matrix compounds are serious interferences to the effective determination of the analytes, and thus have to be removed. In this paper, a sensitive method based on UPLC-MS/MS has been developed for the simultaneous determination of -lactam antibiotics and -lactamase inhibitors in bovine milk. After the optimization of the sample pretreatment and the detection parameters, satisfactory results were obtained in terms of specificity, detection limit, precision, and recovery. 2. Experimental 2.1. Materials and reagents All chemicals and solvents were of analytical grade. AMOX, AMP, CFP (sodium salt), CFT (sodium salt), SUL, and TAZ (sodium salt) were obtained from Dr. Ehrenstorfer (Augsburg, Germany). The chemicals were stored in a dry atmosphere at −40 ◦ C. Acetonitrile and formic acid were obtained from Fisher (Baer-Bel,
N. Li et al. / J. Chromatogr. B 945–946 (2014) 110–114
France). Lead acetate, potassium oxalate-sodium hydrogen phosphate, and sodium dihydrogen phosphate were purchased from Beijing Chemical Company (Beijing, China). Ultrapure water was from Milli-Q System (Millipore, Eschborn, Germany). The cartridges used for solid phase extraction (SPE) were Oasis HLB cartridges (5 mL/500 mg, Waters Corp. Milford, USA). Different types of sorbents described were studied, i.e., C18 cartridges from Waters (Milford, MA, USA), and PSA and SAX cartridges from Varian (Sint-Katelijine-Waver, Belgium). Filter membranes (0.22 m) for filtering the extracts before the injection were from Agilent Corp. (Palo Alto, CA, USA). The lead acetate solution was prepared by diluting 20.00 g of lead acetate in 100 mL of Milli-Q water. Potassium oxalate-sodium hydrogen phosphate solution was prepared by diluting 3.00 g of potassium oxalate and 7.00 g of sodium hydrogen phosphate in 100 mL of Milli-Q water. The phosphate buffer (0.05 M) was prepared by diluting 7.80 g of sodium dihydrogen phosphate in 1 L of Milli-Q water. The pH was adjusted to 8.5 with 10 M NaOH solution using an ORION 710A pH/ISE meter (Beverly, MA, USA). The solutions are stored at room temperature and kept stable for 15 days. Individual stock solutions of chemicals (1 mg/mL) were prepared in Milli-Q water and stored in amber-colored bottles at 4 ◦ C for one month. Standard solutions were prepared by diluting individual stock solution with water prior to use. The working solutions kept in amber-colored bottles at 4 ◦ C were stable for at least one month.
111
the mobile phase was 0.25 mL/min and the injection volume was 5 L. The UPLC system was coupled with a hybrid triple-quadruple linear ion trap mass spectrometer (AB Sciex QTrap® 5500, Toronto, Canada), equipped with an ESI ion source. The optimization of MS condition was achieved by direct infusion of each compound separately at 7 L/min. For TAZ, SUL, CFT, and CFP, the source block temperature was set at 400 ◦ C in the negative ion mode with a capillary voltage of −4.5 kV. For AMOX and AMP, the source block temperature was set at 500 ◦ C in the positive ion mode with a capillary voltage of 5.5 kV. Nitrogen gas was used as the ion source gas 1 (Gas 1, 60 psi), ion source gas 2 (Gas 2, 50 psi), collision gas (6 psi), and curtain gas (20 psi). Detection was operated in the MRM mode and the acquisition of two transitions made it possible to obtain at least three identification points as required by the guideline of 2002/657/EC [15]. The higher intensity transition was selected for the quantitation purpose and the resolution was set at unit (0.7 units). Mass parameters for each analyte, including the precursor ion (Q1 ), product ion (Q3 ), dwell time, declustering potential (DP), and collision energy (CE), are summarized in supplementary material, SI-Table 1. 2.4. Method validation The method was validated according to the FDA guideline [16] and the analytical parameters were: specificity, linearity, limit of detection (LOD), limit of quantification (LOQ), recovery, precision, and stability.
2.2. Sample preparation 5 mL of the milk sample was added into 10 mL of water and the sample was vortex-mixed for 3 min. After adding 1 mL of potassium oxalate-sodium hydrogen phosphate solution, the sample was again vortex-mixed for 1 min. Subsequently, 1 mL of lead acetate solution was added, followed by vortex-mixing for 3 min. The sample was then centrifuged at 5000 rpm for 10 min at 4 ◦ C, and the upper layer was carefully transferred into a centrifugation tube. Such extraction procedure was repeated three times with 10 mL of water each time. Finally, the water supernatants were combined in a centrifuge tube containing 5 mL of 0.05 M phosphate buffer (pH 8.5), and the pH value of the extraction solution was adjusted to 8.5 with 0.1 M NaOH. The extraction solution (37 mL) was loaded onto an SPE cartridge preconditioned consecutively with 5 mL each of acetonitrile and 50 mM phosphate buffer (pH 8.5). The cartridge was then rinsed rapidly with 3 mL of 50 mM phosphate buffer (pH 8.5) and 2 mL of water and suctioned to “dryness” for 2 min. The analytes were eluted with 3 mL of acetonitrile at 1 mL/min. The effluent was evaporated to dryness under a gentle stream of nitrogen in a heater block adjusted to 45 ◦ C. The volume of the final extract was quantified to 1 mL with water and was finally filtered through a 0.22 m filter membrane prior to the injection. 2.3. LC-MS/MS conditions LC was performed using a Waters Acquity UPLC system (Waters Corp., MA, USA), equipped with an automatic degasser, a quaternary pump, and an auto-sampler. Chromatographic separation was performed using a Waters Acquity UPLC BEH C18 column (50 mm length × 2.1 mm i.d., 1.7 m) at 35 ◦ C. The mobile phase A was water containing 0.1% formic acid and the mobile phase B was acetonitrile containing 0.1% formic acid. A gradient elution program was started with 5% of B, increased to 50% at 3 min and then increased to 95% at 3.1 min, maintained at 95% of B for 1 min, returned to initial composition at 4.1 min, and then held constantly for 0.9 min to equilibrate the column. The flow rate of
2.4.1. Specificity To verify the absence of interfering substances around the retention time of analytes, twenty blank milk samples were analyzed. 2.4.2. Linearity, LOD, and LOQ Linearity was determined on matrix-matched calibration curves that were created using blank sample extracts spiked before LCMS/MS at six concentrations (1, 2, 5, 10, 50, and 100 g/L). Limit of detection (LOD) and limit of quantification (LOQ) were obtained by spiking blank samples before sample preparation and referred to concentrations that have signal to noise ratio (S/N) of 3 and 10, respectively. For LOD and LOQ, the acceptable criterion for accuracy was ±15% of the true concentration and the acceptable precision was ±20%. 2.4.3. Recovery The method recovery of all compounds was determined with milk samples in six replicates fortified at three levels (2, 4, and 6 g/L) before sample preparation using calibration curves established with milk extracts spiked prior to LC-MS/MS analysis. 2.4.4. Precision The precision of the entire method was evaluated in terms of intra-day precision and inter-day precision. Intra-day precision was determined at three fortification levels (2, 4, and 6 g/L) by the analysis of spiked samples in six replicates on a single day. Intraday precision was determined for the same fortification levels of spiked samples in six replicates on three different days. 2.4.5. Stability Stability experiments were performed to evaluate the analyte stability in milk sample extracts at room temperature and in stock solutions at 4 ◦ C. The stability of the analytes in milk sample extracts (spiked at 6 ppb) stored at room temperature was evaluated as follows: the spiked milk sample extracts were analyzed after storage of 0, 6, 12, and 24 h, respectively, to get a calculated concentration. Then, the calculated concentration was compared with the theory
112
N. Li et al. / J. Chromatogr. B 945–946 (2014) 110–114
Fig. 1. Structures of six analytes. (A) Amoxicillin; (B) Ampicillin; (C) Tazobactam sodium; (D) Sulbactam; (E) Cefotaxime sodium; (F) Cefoperazone sodium.
concentration. The stability of the stock solution stored in dark at 4 ◦ C was evaluated as follows: the stock solution was analyzed at 0, 4, 8, 16, 32 days, respectively, to get a calculated concentration. Then, the calculated concentration was compared with the theory concentration.
3. Results and discussion 3.1. Consideration of sample preparation Sample pretreatment is a crucial step for the simultaneous determination of the six analytes in bovine milk. The extraction efficiency is highly influenced by operation parameters, such as pH value and organic solvents [11]. -Lactams are readily degraded under strong acidic or basic conditions as a result of the hydrolysis of nucleophilic -lactam rings (Fig. 1) [1]. The use of acetonitrile for both deproteinization and extraction resulted in high fat-containing extract, even after defatting the extract with nhexane and evaporation of the acetonitrile layer under a nitrogen stream at 40 ◦ C [12]. To address these problems, no organic solvents in our study were involved to extract chemicals or to precipitate fats and proteins. Water was selected as the extractant and lead acetate solution was used for the precipitation. Such strategy was found to be effective and satisfactory recoveries ranging from 85.0 to 98.0% for TAZ, AMOX, SUL, AMP, CFT, and CFP were obtained. An SPE procedure was performed to further remove the matrix interferences and to enrich the extracts by replacing high volumes of water with low volumes of organic solvent, thus lowering the detection limit. Several kinds of SPE cartridges commonly used were explored including Oasis HLB, C18 , PSA, and SAX. It was found that all analytes tested had no retention in PSA and SAX cartridges. Good retention behavior for the six analytes could be observed using HLB and C18 cartridges. However, poor recoveries were observed using C18 cartridge, which was in agreement with previous observation [9]. By contrast, HLB behaved much better in terms of good recovery and good precision (shown in Fig. 2), and thus it was selected in the next experiment.
The retention behavior of -lactam analytes using the sorbents is known to be pH dependent [1,9] . The effect of sample pH on extraction recoveries was explored by using 40 mL of standard solutions with different pH values of 6.5, 8.5, and 10. The data showed that the best recoveries ranging from 90 to 98% for all six analytes were achieved for the solution at a pH of 8.5, and thus it was selected as the best pH condition. 3.2. Consideration of MS response mode A typical chromatographic separation of the six analytes is shown in Fig. 3. The coupling of UPLC and MS/MS is a powerful tool for the accurate confirmation of -lactam antibiotics. The formation of at least two fragment ions from the protonated molecule enables the acquisition of two or more MRM transitions for each analyte which provides added confidence in the confirmation of analytes and decreases the risk of false positives. To select the ionization mode, standard solutions of each analyte were injected with a concentration of 1 g/mL. The data indicated that AMOX and AMP had higher response in ESI+ mode while TAZ, SUL, CFT, and CFP had higher response in ESI− mode. Because a simultaneous
Fig. 2. Comparison of C18 and Oasis HLB for the determination of six analytes.
N. Li et al. / J. Chromatogr. B 945–946 (2014) 110–114
113
Table 1 Performance characteristics of the proposed method for the detection of the six analytes in bovine milk.
Linear dynamic range (g/L) R2 LOD (g/L) LOQ (g/L) MRL (g/L)
TAZ
AMOX
SUL
AMP
CFT
CFP
1–100 0.9991 0.10 0.30 –
2–100 0.9992 0.20 0.50 4
1–100 0.9989 0.10 0.30 –
1–100 0.9993 0.10 0.30 4
2–100 0.9995 0.20 0.50 –
1–100 0.9996 0.10 0.30 50
scan in both ESI+ and ESI− modes might be detrimental to both the MS equipment and the sensitivity, ESI− and ESI+ mode were selected separately, providing a more sensitive and machinefriendly method. 3.3. Matrix effect It is known that the presence of matrix components can affect the ionization of the analytes when ESI is used. In this study, the matrix effect was evaluated by comparing the slope ratios of the matrix matched calibration curves and the solvent calibration curves. 0.78–0.95 of the slope ratios were obtained for all six analytes. According to what is mentioned previously by Frenich et al. [17], signal suppression or enhancement effect was considered to be tolerable if the value was between 0.8 and 1.2. In other words, the matrix effect can be acceptable. From this point, it can be concluded that there was low interference from matrix for the six analytes tested. 3.4. Performance characteristics of the method 3.4.1. Specificity The specificity of the method was firstly evaluated by analyzing twenty blank milk samples. No interference peaks from coeluted compounds were observed by comparing the chromatograms of blank samples (Fig. 3A) and spiked samples at LOQ level (Fig. 3B).
3.4.2. Linearity, LOD, and LOQ The performance characteristics of the method were shown in Table 1. It can be seen that excellent linear relationships were obtained with the coefficients in the range of 0.9989–0.9996. The LOQs for AMOX, AMP, and CFT in bovine milk meet the MRLs established by the EU. No MRLs for TAZ, SUL, and CFP, or studies for the determination of them in bovine milk have been established. However, the LOQs obtained in this study were much lower than the LOQs, 500 g/L for TAZ and CFT, and 20 g/L for SUL, obtained in plasma in previous studies [5,6]. 3.4.3. Recovery and precision To test the efficiency of the method, recoveries at three concentrations levels were performed, as provided in Table 2. 82.5–98.3% of recoveries were achieved for all six analytes, indicating the suitability of the proposed method for the simultaneous determination of the six analytes tested. In addition, the results of intra- and interday reproducibility were also explored and the data were given in Table 3. The intra- and inter-day relative standard deviations (RSDs) for the spiked samples at three concentration levels were in the range of 3.2–5.5%, indicating the good precision of the proposed method. 3.4.4. Stability All six analytes were found to be stable in 24 h at room temperature in milk sample extracts as proved by the recoveries ranging from 95.4 to 99.8%. Stock solutions of the six analytes were stable
Fig. 3. Chromatograms of six analytes in bovine milk spiked at LOQ level. (A) Blank milk sample; (B) Spiked milk sample at LOQ level.
114
N. Li et al. / J. Chromatogr. B 945–946 (2014) 110–114
Table 2 Recovery (R) and relative standard deviation (RSD, %) of the six analytes in bovine milk (n = 6). Fortification level (g/L)
2 4 6
TAZ
AMOX
SUL
AMP
CFT
CFP
R
RSD
R
RSD
R
RSD
R
RSD
R
RSD
R
RSD
98.2 97.8 98.3
6.0 5.7 4.3
82.5 84.6 85.0
5.6 5.2 4.3
93.7 95.4 95.7
5.7 4.8 4.2
84.0 85.2 85.5
5.3 4.7 4.0
92.3 92.1 90.5
5.8 5.2 4.5
91.0 90.5 90.7
6.2 5.3 4.7
Table 3 Intra-day study results (RSD%) (n = 6) and Inter-day precision study results (RSD%) (n = 18) of the six analytes in bovine milk. Spiked level (g/L)
TAZ AMOX SUL AMP CFT CFP a b
2
4
6
Intra-daya
Inter-dayb
Intra-day
Inter-day
Intra-day
Inter-day
4.3 4.7 4.2 4.5 4.5 4.4
5.3 4.8 4.6 5.5 4.8 4.9
3.8 3.7 3.5 3.8 4.0 3.8
4.8 4.7 4.6 4.8 4.4 4.8
3.2 3.5 3.0 3.3 3.5 3.4
4.3 4.5 4.3 4.2 4.3 4.4
Intra-day study was performed in six replicates on a single. day. Inter-day study was performed in six replicates on three separate days.
in 30 days when stored at 4 ◦ C, and the recoveries ranged from 98.5 to 101.5%. 4. Conclusion A sensitive and specific UPLC-MS/MS method has been developed for the simultaneous determination of four -lactam antibiotics and two -lactamase inhibitors in bovine milk. The sample extraction method was proved to be effective, avoiding the damage of target analytes and the poor extraction efficiency encountered in the common methods. The proposed method demonstrated good accuracy and good precision, and it can be applied as a routine procedure to identify and quantify -lactam antibiotics and -lactamase inhibitors in bovine milk for food quality and safety control. Acknowledgement The research was sponsored by Research Special Funds for Public Welfare Project (No. 2012104002). Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jchromb. 2013.11.044.
References ˜ A.I. Partal-Rodera, M.E. León-González, Anal. Chim. Acta 556 [1] E. Benito-Pena, (2006) 415. [2] L.K. Sørensen, L.K. Snor, J. Chromatogr. A 882 (2000) 145. ˜ M. del Olmo-Iruela, L. Gámiz-Gracia, [3] M.I. Bailón-Pérez, A.M. García-Campana, C. Cruces-Blanco, J. Chromatogr. A 1216 (2009) 8355. [4] P. Wang, M.L. Qi, Y. Sun, J. Pharm. Biomed. Anal. 36 (2004) 565. [5] T.Li. Tsou, Y.C. Huang, C.W. Lee, A.R. Lee, H.J. Wang, S.H. Chen, J. Sep. Sci. 30 (2007) 2413. [6] Y.J. Zhou, J. Zhang, B.N. Guo, J. Chromatogr. B 878 (2010) 3119. [7] M. Carlier, V. Stove, J.A. Roberts, Int. J. Antimicrob. Ag. 40 (2012) 416. [8] H. Li, X. Xia, Y.N. Xue, J. Chromatogr. B 900 (2012) 59. [9] M. Martínez-Huelamo, E. Jiménez-Gámez, M.P. Hermo, J. Sep. Sci. 32 (2009) 2385. [10] S. Bogialli, V. Capitolino, R. Curini, J. Agric. Food Chem. 52 (2004) 3286. [11] S. Riediker, R.H. Stadler, Anal. Chem. 73 (2001) 1614. [12] T. Reyns, S. De Boever, S. De Baere, P. De Backer, S. Croubels, Anal. Chim. Acta 597 (2007) 282. [13] Council Directive 96/23/EEC of 29 April 1996 on measures to monitor certain substances and residues thereof in live animals and animal products and repealing Directives 85/358/EEC and Decisions 89/187/EEC and 91/664/EEC. Off. J. Eur. Commun. L125 (1996) 10. [14] Code of Federal Regulation, Title 21 (Food and Drugs), vol. 6, Part 556 (July 28, 2011). [15] Commision Decision 2002/657/EC, Off. J. Eur. Commun. L221 (2002) 23. [16] Guidance for Industry: Bioanalytical Method Validation, US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER) and Center for Veterinary Medicine (CVM), May 2001. [17] A.G. Frenich, R. Romero-González, M.L. Gómez-Pérez, J.L. Martinez Vidal, J. Chromatogr. A 1218 (2011) 4349.
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Simultaneous determination of -lactam antibiotics and -lactamase inhibitors in bovine milk by ultra performance liquid chromatography-tandem mass spectrometry Nasi Li a,b , Feng Feng b , Bingcheng Yang a , Pingping Jiang b , Xiaogang Chu b,∗ a b
School of Pharmacy, East China University of Science and Technology, Shanghai 200237, China Institute of Food Safety, Chinese Academy of Inspection and Quarantine, Beijing 100123, China
a r t i c l e
i n f o
Article history: Received 19 August 2013 Received in revised form 15 November 2013 Accepted 21 November 2013 Available online 26 November 2013 Keywords: -Lactam antibiotics -Lactamase inhibitors Ultra-high performance liquid chromatography Tandem mass spectrometry Bovine milk
a b s t r a c t An ultra performance liquid chromatography-tandem mass spectrometric (UPLC-MS/MS) method has been developed for the simultaneous determination of four -lactam antibiotics (amoxicillin, ampicillin, cefotaxime, and cefoperazone) and two -lactamase inhibitors (tazobactam, sulbactam) in bovine milk. The analytes were extracted with water from bovine milk and purified with Oasis HLB solid phase extraction (SPE) cartridges. The analytes were determined in less than 3 min by UPLC-MS/MS in positive and negative electrospray ionization (ESI) modes, separately. The method was linear over the range of 1–100 g/L for tazobactam, sulbactam, ampicillin, and cefoperazone, and 2–100 g/L for amoxicillin and cefotaxime. The recoveries for all six analytes in bovine milk ranged from 82.5 to 98.3%. The limits of detection and the limits of quantitation were 0.1–0.2 g/L and 0.3–0.5 g/L, respectively. The intra- and inter-day precisions were less than 6% for each compound. © 2013 Elsevier B.V. All rights reserved.
1. Introduction -Lactam antibiotics, such as amoxicillin (AMOX), ampicillin (AMP), cefoperazone (CFP), and cefotaxime (CFT), are widely used for the treatment of bacterial infections in human beings and animals. Their wide application in animals represents a potential hazard because residues of these antibiotics may persist in edible tissues or foodstuffs, such as bovine milk [1–3]. -Lactamase inhibitors, such as sulbactam (SUL) and tazobactam (TAZ), are molecules used in conjunction with -lactam antibiotics to extend their spectrum of activity [4–7]. For example, the commercial veterinary drug composed of AMOX and SUL has been used for the treatment of bovine mastitis [8]. To keep these drugs under regulation and ensure the safety of bovine milk, it is necessary to develop analytical methods for the determination of -lactam antibiotics and -lactamase inhibitors in bovine milk. Several methods have been developed for the determination of -lactams in bovine milk [1,9–12]. The maximum residue limits (MRLs) for -lactams in bovine milk have been regulated in several countries. The European Union (EU) has established an MRL of 4, 4, and 50 g/L for amoxicillin (AMOX), amipicillin (AMP), and
∗ Corresponding author. Tel.: +86 10 85770775; fax: +86 10 85770775. E-mail address: [email protected] (X. Chu). 1570-0232/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jchromb.2013.11.044
cefoperazone (CFP) in bovine milk, respectively [13]. The USA has adopted an MRL of 10 g/L for AMOX in bovine milk [14]. However, methods developed for the determination of -lactamase inhibitors are limited with the few literatures focus on human plasma [6,7]. It is known that the matrix of human plasma is much different from that of bovine milk which contains significant amounts of proteins, saturated fat, and calcium. These matrix compounds are serious interferences to the effective determination of the analytes, and thus have to be removed. In this paper, a sensitive method based on UPLC-MS/MS has been developed for the simultaneous determination of -lactam antibiotics and -lactamase inhibitors in bovine milk. After the optimization of the sample pretreatment and the detection parameters, satisfactory results were obtained in terms of specificity, detection limit, precision, and recovery. 2. Experimental 2.1. Materials and reagents All chemicals and solvents were of analytical grade. AMOX, AMP, CFP (sodium salt), CFT (sodium salt), SUL, and TAZ (sodium salt) were obtained from Dr. Ehrenstorfer (Augsburg, Germany). The chemicals were stored in a dry atmosphere at −40 ◦ C. Acetonitrile and formic acid were obtained from Fisher (Baer-Bel,
N. Li et al. / J. Chromatogr. B 945–946 (2014) 110–114
France). Lead acetate, potassium oxalate-sodium hydrogen phosphate, and sodium dihydrogen phosphate were purchased from Beijing Chemical Company (Beijing, China). Ultrapure water was from Milli-Q System (Millipore, Eschborn, Germany). The cartridges used for solid phase extraction (SPE) were Oasis HLB cartridges (5 mL/500 mg, Waters Corp. Milford, USA). Different types of sorbents described were studied, i.e., C18 cartridges from Waters (Milford, MA, USA), and PSA and SAX cartridges from Varian (Sint-Katelijine-Waver, Belgium). Filter membranes (0.22 m) for filtering the extracts before the injection were from Agilent Corp. (Palo Alto, CA, USA). The lead acetate solution was prepared by diluting 20.00 g of lead acetate in 100 mL of Milli-Q water. Potassium oxalate-sodium hydrogen phosphate solution was prepared by diluting 3.00 g of potassium oxalate and 7.00 g of sodium hydrogen phosphate in 100 mL of Milli-Q water. The phosphate buffer (0.05 M) was prepared by diluting 7.80 g of sodium dihydrogen phosphate in 1 L of Milli-Q water. The pH was adjusted to 8.5 with 10 M NaOH solution using an ORION 710A pH/ISE meter (Beverly, MA, USA). The solutions are stored at room temperature and kept stable for 15 days. Individual stock solutions of chemicals (1 mg/mL) were prepared in Milli-Q water and stored in amber-colored bottles at 4 ◦ C for one month. Standard solutions were prepared by diluting individual stock solution with water prior to use. The working solutions kept in amber-colored bottles at 4 ◦ C were stable for at least one month.
111
the mobile phase was 0.25 mL/min and the injection volume was 5 L. The UPLC system was coupled with a hybrid triple-quadruple linear ion trap mass spectrometer (AB Sciex QTrap® 5500, Toronto, Canada), equipped with an ESI ion source. The optimization of MS condition was achieved by direct infusion of each compound separately at 7 L/min. For TAZ, SUL, CFT, and CFP, the source block temperature was set at 400 ◦ C in the negative ion mode with a capillary voltage of −4.5 kV. For AMOX and AMP, the source block temperature was set at 500 ◦ C in the positive ion mode with a capillary voltage of 5.5 kV. Nitrogen gas was used as the ion source gas 1 (Gas 1, 60 psi), ion source gas 2 (Gas 2, 50 psi), collision gas (6 psi), and curtain gas (20 psi). Detection was operated in the MRM mode and the acquisition of two transitions made it possible to obtain at least three identification points as required by the guideline of 2002/657/EC [15]. The higher intensity transition was selected for the quantitation purpose and the resolution was set at unit (0.7 units). Mass parameters for each analyte, including the precursor ion (Q1 ), product ion (Q3 ), dwell time, declustering potential (DP), and collision energy (CE), are summarized in supplementary material, SI-Table 1. 2.4. Method validation The method was validated according to the FDA guideline [16] and the analytical parameters were: specificity, linearity, limit of detection (LOD), limit of quantification (LOQ), recovery, precision, and stability.
2.2. Sample preparation 5 mL of the milk sample was added into 10 mL of water and the sample was vortex-mixed for 3 min. After adding 1 mL of potassium oxalate-sodium hydrogen phosphate solution, the sample was again vortex-mixed for 1 min. Subsequently, 1 mL of lead acetate solution was added, followed by vortex-mixing for 3 min. The sample was then centrifuged at 5000 rpm for 10 min at 4 ◦ C, and the upper layer was carefully transferred into a centrifugation tube. Such extraction procedure was repeated three times with 10 mL of water each time. Finally, the water supernatants were combined in a centrifuge tube containing 5 mL of 0.05 M phosphate buffer (pH 8.5), and the pH value of the extraction solution was adjusted to 8.5 with 0.1 M NaOH. The extraction solution (37 mL) was loaded onto an SPE cartridge preconditioned consecutively with 5 mL each of acetonitrile and 50 mM phosphate buffer (pH 8.5). The cartridge was then rinsed rapidly with 3 mL of 50 mM phosphate buffer (pH 8.5) and 2 mL of water and suctioned to “dryness” for 2 min. The analytes were eluted with 3 mL of acetonitrile at 1 mL/min. The effluent was evaporated to dryness under a gentle stream of nitrogen in a heater block adjusted to 45 ◦ C. The volume of the final extract was quantified to 1 mL with water and was finally filtered through a 0.22 m filter membrane prior to the injection. 2.3. LC-MS/MS conditions LC was performed using a Waters Acquity UPLC system (Waters Corp., MA, USA), equipped with an automatic degasser, a quaternary pump, and an auto-sampler. Chromatographic separation was performed using a Waters Acquity UPLC BEH C18 column (50 mm length × 2.1 mm i.d., 1.7 m) at 35 ◦ C. The mobile phase A was water containing 0.1% formic acid and the mobile phase B was acetonitrile containing 0.1% formic acid. A gradient elution program was started with 5% of B, increased to 50% at 3 min and then increased to 95% at 3.1 min, maintained at 95% of B for 1 min, returned to initial composition at 4.1 min, and then held constantly for 0.9 min to equilibrate the column. The flow rate of
2.4.1. Specificity To verify the absence of interfering substances around the retention time of analytes, twenty blank milk samples were analyzed. 2.4.2. Linearity, LOD, and LOQ Linearity was determined on matrix-matched calibration curves that were created using blank sample extracts spiked before LCMS/MS at six concentrations (1, 2, 5, 10, 50, and 100 g/L). Limit of detection (LOD) and limit of quantification (LOQ) were obtained by spiking blank samples before sample preparation and referred to concentrations that have signal to noise ratio (S/N) of 3 and 10, respectively. For LOD and LOQ, the acceptable criterion for accuracy was ±15% of the true concentration and the acceptable precision was ±20%. 2.4.3. Recovery The method recovery of all compounds was determined with milk samples in six replicates fortified at three levels (2, 4, and 6 g/L) before sample preparation using calibration curves established with milk extracts spiked prior to LC-MS/MS analysis. 2.4.4. Precision The precision of the entire method was evaluated in terms of intra-day precision and inter-day precision. Intra-day precision was determined at three fortification levels (2, 4, and 6 g/L) by the analysis of spiked samples in six replicates on a single day. Intraday precision was determined for the same fortification levels of spiked samples in six replicates on three different days. 2.4.5. Stability Stability experiments were performed to evaluate the analyte stability in milk sample extracts at room temperature and in stock solutions at 4 ◦ C. The stability of the analytes in milk sample extracts (spiked at 6 ppb) stored at room temperature was evaluated as follows: the spiked milk sample extracts were analyzed after storage of 0, 6, 12, and 24 h, respectively, to get a calculated concentration. Then, the calculated concentration was compared with the theory
112
N. Li et al. / J. Chromatogr. B 945–946 (2014) 110–114
Fig. 1. Structures of six analytes. (A) Amoxicillin; (B) Ampicillin; (C) Tazobactam sodium; (D) Sulbactam; (E) Cefotaxime sodium; (F) Cefoperazone sodium.
concentration. The stability of the stock solution stored in dark at 4 ◦ C was evaluated as follows: the stock solution was analyzed at 0, 4, 8, 16, 32 days, respectively, to get a calculated concentration. Then, the calculated concentration was compared with the theory concentration.
3. Results and discussion 3.1. Consideration of sample preparation Sample pretreatment is a crucial step for the simultaneous determination of the six analytes in bovine milk. The extraction efficiency is highly influenced by operation parameters, such as pH value and organic solvents [11]. -Lactams are readily degraded under strong acidic or basic conditions as a result of the hydrolysis of nucleophilic -lactam rings (Fig. 1) [1]. The use of acetonitrile for both deproteinization and extraction resulted in high fat-containing extract, even after defatting the extract with nhexane and evaporation of the acetonitrile layer under a nitrogen stream at 40 ◦ C [12]. To address these problems, no organic solvents in our study were involved to extract chemicals or to precipitate fats and proteins. Water was selected as the extractant and lead acetate solution was used for the precipitation. Such strategy was found to be effective and satisfactory recoveries ranging from 85.0 to 98.0% for TAZ, AMOX, SUL, AMP, CFT, and CFP were obtained. An SPE procedure was performed to further remove the matrix interferences and to enrich the extracts by replacing high volumes of water with low volumes of organic solvent, thus lowering the detection limit. Several kinds of SPE cartridges commonly used were explored including Oasis HLB, C18 , PSA, and SAX. It was found that all analytes tested had no retention in PSA and SAX cartridges. Good retention behavior for the six analytes could be observed using HLB and C18 cartridges. However, poor recoveries were observed using C18 cartridge, which was in agreement with previous observation [9]. By contrast, HLB behaved much better in terms of good recovery and good precision (shown in Fig. 2), and thus it was selected in the next experiment.
The retention behavior of -lactam analytes using the sorbents is known to be pH dependent [1,9] . The effect of sample pH on extraction recoveries was explored by using 40 mL of standard solutions with different pH values of 6.5, 8.5, and 10. The data showed that the best recoveries ranging from 90 to 98% for all six analytes were achieved for the solution at a pH of 8.5, and thus it was selected as the best pH condition. 3.2. Consideration of MS response mode A typical chromatographic separation of the six analytes is shown in Fig. 3. The coupling of UPLC and MS/MS is a powerful tool for the accurate confirmation of -lactam antibiotics. The formation of at least two fragment ions from the protonated molecule enables the acquisition of two or more MRM transitions for each analyte which provides added confidence in the confirmation of analytes and decreases the risk of false positives. To select the ionization mode, standard solutions of each analyte were injected with a concentration of 1 g/mL. The data indicated that AMOX and AMP had higher response in ESI+ mode while TAZ, SUL, CFT, and CFP had higher response in ESI− mode. Because a simultaneous
Fig. 2. Comparison of C18 and Oasis HLB for the determination of six analytes.
N. Li et al. / J. Chromatogr. B 945–946 (2014) 110–114
113
Table 1 Performance characteristics of the proposed method for the detection of the six analytes in bovine milk.
Linear dynamic range (g/L) R2 LOD (g/L) LOQ (g/L) MRL (g/L)
TAZ
AMOX
SUL
AMP
CFT
CFP
1–100 0.9991 0.10 0.30 –
2–100 0.9992 0.20 0.50 4
1–100 0.9989 0.10 0.30 –
1–100 0.9993 0.10 0.30 4
2–100 0.9995 0.20 0.50 –
1–100 0.9996 0.10 0.30 50
scan in both ESI+ and ESI− modes might be detrimental to both the MS equipment and the sensitivity, ESI− and ESI+ mode were selected separately, providing a more sensitive and machinefriendly method. 3.3. Matrix effect It is known that the presence of matrix components can affect the ionization of the analytes when ESI is used. In this study, the matrix effect was evaluated by comparing the slope ratios of the matrix matched calibration curves and the solvent calibration curves. 0.78–0.95 of the slope ratios were obtained for all six analytes. According to what is mentioned previously by Frenich et al. [17], signal suppression or enhancement effect was considered to be tolerable if the value was between 0.8 and 1.2. In other words, the matrix effect can be acceptable. From this point, it can be concluded that there was low interference from matrix for the six analytes tested. 3.4. Performance characteristics of the method 3.4.1. Specificity The specificity of the method was firstly evaluated by analyzing twenty blank milk samples. No interference peaks from coeluted compounds were observed by comparing the chromatograms of blank samples (Fig. 3A) and spiked samples at LOQ level (Fig. 3B).
3.4.2. Linearity, LOD, and LOQ The performance characteristics of the method were shown in Table 1. It can be seen that excellent linear relationships were obtained with the coefficients in the range of 0.9989–0.9996. The LOQs for AMOX, AMP, and CFT in bovine milk meet the MRLs established by the EU. No MRLs for TAZ, SUL, and CFP, or studies for the determination of them in bovine milk have been established. However, the LOQs obtained in this study were much lower than the LOQs, 500 g/L for TAZ and CFT, and 20 g/L for SUL, obtained in plasma in previous studies [5,6]. 3.4.3. Recovery and precision To test the efficiency of the method, recoveries at three concentrations levels were performed, as provided in Table 2. 82.5–98.3% of recoveries were achieved for all six analytes, indicating the suitability of the proposed method for the simultaneous determination of the six analytes tested. In addition, the results of intra- and interday reproducibility were also explored and the data were given in Table 3. The intra- and inter-day relative standard deviations (RSDs) for the spiked samples at three concentration levels were in the range of 3.2–5.5%, indicating the good precision of the proposed method. 3.4.4. Stability All six analytes were found to be stable in 24 h at room temperature in milk sample extracts as proved by the recoveries ranging from 95.4 to 99.8%. Stock solutions of the six analytes were stable
Fig. 3. Chromatograms of six analytes in bovine milk spiked at LOQ level. (A) Blank milk sample; (B) Spiked milk sample at LOQ level.
114
N. Li et al. / J. Chromatogr. B 945–946 (2014) 110–114
Table 2 Recovery (R) and relative standard deviation (RSD, %) of the six analytes in bovine milk (n = 6). Fortification level (g/L)
2 4 6
TAZ
AMOX
SUL
AMP
CFT
CFP
R
RSD
R
RSD
R
RSD
R
RSD
R
RSD
R
RSD
98.2 97.8 98.3
6.0 5.7 4.3
82.5 84.6 85.0
5.6 5.2 4.3
93.7 95.4 95.7
5.7 4.8 4.2
84.0 85.2 85.5
5.3 4.7 4.0
92.3 92.1 90.5
5.8 5.2 4.5
91.0 90.5 90.7
6.2 5.3 4.7
Table 3 Intra-day study results (RSD%) (n = 6) and Inter-day precision study results (RSD%) (n = 18) of the six analytes in bovine milk. Spiked level (g/L)
TAZ AMOX SUL AMP CFT CFP a b
2
4
6
Intra-daya
Inter-dayb
Intra-day
Inter-day
Intra-day
Inter-day
4.3 4.7 4.2 4.5 4.5 4.4
5.3 4.8 4.6 5.5 4.8 4.9
3.8 3.7 3.5 3.8 4.0 3.8
4.8 4.7 4.6 4.8 4.4 4.8
3.2 3.5 3.0 3.3 3.5 3.4
4.3 4.5 4.3 4.2 4.3 4.4
Intra-day study was performed in six replicates on a single. day. Inter-day study was performed in six replicates on three separate days.
in 30 days when stored at 4 ◦ C, and the recoveries ranged from 98.5 to 101.5%. 4. Conclusion A sensitive and specific UPLC-MS/MS method has been developed for the simultaneous determination of four -lactam antibiotics and two -lactamase inhibitors in bovine milk. The sample extraction method was proved to be effective, avoiding the damage of target analytes and the poor extraction efficiency encountered in the common methods. The proposed method demonstrated good accuracy and good precision, and it can be applied as a routine procedure to identify and quantify -lactam antibiotics and -lactamase inhibitors in bovine milk for food quality and safety control. Acknowledgement The research was sponsored by Research Special Funds for Public Welfare Project (No. 2012104002). Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jchromb. 2013.11.044.
References ˜ A.I. Partal-Rodera, M.E. León-González, Anal. Chim. Acta 556 [1] E. Benito-Pena, (2006) 415. [2] L.K. Sørensen, L.K. Snor, J. Chromatogr. A 882 (2000) 145. ˜ M. del Olmo-Iruela, L. Gámiz-Gracia, [3] M.I. Bailón-Pérez, A.M. García-Campana, C. Cruces-Blanco, J. Chromatogr. A 1216 (2009) 8355. [4] P. Wang, M.L. Qi, Y. Sun, J. Pharm. Biomed. Anal. 36 (2004) 565. [5] T.Li. Tsou, Y.C. Huang, C.W. Lee, A.R. Lee, H.J. Wang, S.H. Chen, J. Sep. Sci. 30 (2007) 2413. [6] Y.J. Zhou, J. Zhang, B.N. Guo, J. Chromatogr. B 878 (2010) 3119. [7] M. Carlier, V. Stove, J.A. Roberts, Int. J. Antimicrob. Ag. 40 (2012) 416. [8] H. Li, X. Xia, Y.N. Xue, J. Chromatogr. B 900 (2012) 59. [9] M. Martínez-Huelamo, E. Jiménez-Gámez, M.P. Hermo, J. Sep. Sci. 32 (2009) 2385. [10] S. Bogialli, V. Capitolino, R. Curini, J. Agric. Food Chem. 52 (2004) 3286. [11] S. Riediker, R.H. Stadler, Anal. Chem. 73 (2001) 1614. [12] T. Reyns, S. De Boever, S. De Baere, P. De Backer, S. Croubels, Anal. Chim. Acta 597 (2007) 282. [13] Council Directive 96/23/EEC of 29 April 1996 on measures to monitor certain substances and residues thereof in live animals and animal products and repealing Directives 85/358/EEC and Decisions 89/187/EEC and 91/664/EEC. Off. J. Eur. Commun. L125 (1996) 10. [14] Code of Federal Regulation, Title 21 (Food and Drugs), vol. 6, Part 556 (July 28, 2011). [15] Commision Decision 2002/657/EC, Off. J. Eur. Commun. L221 (2002) 23. [16] Guidance for Industry: Bioanalytical Method Validation, US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER) and Center for Veterinary Medicine (CVM), May 2001. [17] A.G. Frenich, R. Romero-González, M.L. Gómez-Pérez, J.L. Martinez Vidal, J. Chromatogr. A 1218 (2011) 4349.
