Determination of 11 quinolones in bovine milk using immunoaffinity stir bar sorptive microextraction and liquid chromatography with fluorescence detection

Accepted Manuscript Title: Determination of 11 quinolones in bovine milk using immunoaffinity stir bar sorptive microextraction and liquid chromatography with fluorescence detection Author: Kai Yao Wei Zhang Linyan Yang Jianfang Gong Liuan Li Tianming Jin Cun Li PII: DOI: Reference:

S1570-0232(15)30183-5 http://dx.doi.org/doi:10.1016/j.jchromb.2015.09.008 CHROMB 19617

To appear in:

Journal of Chromatography B

Received date: Revised date: Accepted date:

6-8-2015 7-9-2015 9-9-2015

Please cite this article as: Kai Yao, Wei Zhang, Linyan Yang, Jianfang Gong, Liuan Li, Tianming Jin, Cun Li, Determination of 11 quinolones in bovine milk using immunoaffinity stir bar sorptive microextraction and liquid chromatography with fluorescence detection, Journal of Chromatography B http://dx.doi.org/10.1016/j.jchromb.2015.09.008 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Determination of 11 quinolones in bovine milk using immunoaffinity stir bar sorptive microextraction and liquid chromatography with fluorescence detection Kai Yao, Wei Zhang, Linyan Yang, Jianfang Gong, Liuan Li; Tianming Jin, Cun Li* [email protected] College of Animal Science and Veterinary Medicine, Tianjin Agricultural University, 300384 Tianjin, People’s Republic of China *

Corresponding author. Tel.: +86-13920589691; fax: +86-22-23784590.

1

Highlights  Quantification of 11 QNs in bovine milk.  Monoclonal antibody to QNs was immobilized to glass bar.  An immunoaffinity SBSME-LC-FLD method was firstly developed and validated.  Different conditions during sample preparation were optimized.  The column capacity and stability of immunoaffinity sitr bar was investigated.

2

ABSTRACT A sensitive, selective and reproducible immunoaffinity stir bar sorptive microextraction (SBSME) coupled with liquid chromatography-fluorescence method for determination of 11 quinolones (QNs) in bovine milk was developed and validated. It is first report of a broad-specificity monoclonal antibody to QNs that has been immobilized to glass bar for preparation of a re-usable immunoaffinity stir bar. Analytes were extracted by placing stir bar in milk and shaking on a rotary shaker for 30 min at 30 rpm, followed by liquid chromatography and fluorescence detection. The newly developed method has limits of detection for each QN from 0.05 to 0.1 ng/g with intra-day and inter-day precision ranging from 3.2% to 11.9% and from 5.2% to 12.5%, respectively. This allowed us to quantitatively analyze drugs in bovine milk with the advantage of significantly simplified sample preparation. The proposed method was successfully applied to the bovine milk samples analyses with QNs, demonstrating its rare application in animal food safety analysis. Keywords: Immunoaffinity stir bar; Quinolone; Monoclonal antibody; Bovine milk; Liquid chromatography

3

Introduction Quinolones (QNs) are synthetic antibacterial agents that have been widely used in human and animals for treatment of intestinal, urinary and respiratory infections [1]. Their antimicrobial mechanism against gram-negative bacteria and gram-positive bacteria can be explained as follows: QNs can affect bacterial reproduction by inhibiting topoisomerase and DNA gyrase enzymes [2]. In order to promote production and maintain animal health, the improper use of antibacterials are becoming more common, which leads to drug residues in edible animal products and poses serious threats to consumers’ health, including development of resistant bacterial strains, liver damage, allergic hypersensitivity and gastrointestinal reactions. The European Union (EU) has set maximum residue limits of quinolones in milk and other tissues to minimize the incidence of toxicity and to regulate the use of QNs in animals [3]. MRLs for several QNs such as enrofloxacin, ciprofloxacin and marbofloxacin in bovine milk ranged from 75 to100 ng/g. Thus, there is an urgent need to develop a highly specific, sensitive and efficient method to quantify QNs in milk. A number of methods has been developed for determination of QNs residues, many

of

them

based

on

polarization

immunoassay

[4], enzyme-linked

immunosorbent assay [5], capillary electrophoresis with laser-induced fluorescence [6,7], micellar electrokinetic capillary with indirect laser-induced fluorescence [8] and liquid chromatography (LC) with DAD [9], MS/MS [10-12] or fluorescence 4

[13,14]. Bovine milk is considered a complex matrix and contains many types of interfering substances, such as lipid and protein. These substances often interfere with analyte determination and become a major problem when analytes are at trace concentration, so a cleanup step to remove their interference is inevitable part of sample preparation prior to the measurement of analytes in bovine milk [15]. The majority of current clean-up methods are based on solid-phase extraction [16-18] and liquid-liquid extraction [19,20]. These methods of analytes separation and concentration require additional cleanup procedures and large numbers of organic solvents, which are laborious, expensive and time-consuming due to many steps and the lack of selectivity. Immunoaffinity solid phase microextraction (SPME) was first reported for determining theophylline in serum samples in 2001 [21]. It is a form of immunoaffinity chromatography (IAC) in which highly selective antibodies are bound to a solid support material such as glass bar, fused silica capillary or stainless steel rod. To date, immunoaffinity SPME or immunoaffinity in-tube SPME devices have been mainly used for detecting theophylline [21], 7-aminoflunitrazepam [22], benzodiazepines [23], fluoxetine [24], interferon alpha [25], and penicillin binding protein 2a [26]. Immunoaffinity column is one of the most powerful method for purifying QNs in complicated matrices [27-29]. However, the preparation process for immunoaffinity column requires large amount of monoclonal antibodies (mAbs), and the cleanup procedure is complex. As far as we know, no method about immunoaffinity SPME has been published for determination of a class of veterinary 5

drugs in animal-derived food. Therefore, the developed method in this study is mainly focused on construction of an immunoaffinity stir bar by immobilizing mAbs on a glass bar for purification and extraction of 11 QNs from bovine milk which need little amount of mAbs and organic solvent. The extraction is followed by the quantification of QNs with high performance liquid chromatography (HPLC) with programmable fluorescence detection (FLD). The method was validated according to the cross-reactivity, binding capacity, linearity, recovery, accuracy, precision and the stability of immunoaffinity stir bar, and was expected to be easy, sensitive, reproducible and environmental-friendly.

1. Materials and methods 2.1. Reagents and materials All chemicals used were of analytical grade. Acetonitrile was obtained from Merck (Darmstadt, Germany). Sulfuric acid (98%), hydrogen peroxide (30%), sodium hydroxide, sodium chloride, potassium chloride, potassium phosphate monobasic and sodium phosphate dibasic were purchased from Tianjin Chemical Reagent No. 3 Plant (Tianjin, China). Formic acid was purchased from Tianjin Fuyu Fine Chemical Co. Ltd. (Tianjin, China). Methanol (MeOH) was obtained from Tianjin Kangchao Biopharmaceutical Tech Co. Ltd. (Tianjin, China). Ethanol absolute was purchased from Tianjin Jinke Fine Chemical Research Institute (Tianjin, China). (3-aminopropyl)-triethoxysilane (APTES), glutaraldehyde grade II (25% aqueous solution) and ethanolamine were obtained from Sigma-Aldrich (St. Louis, 6

MO, USA). All water was obtained from a Milli-Q system (Bedford, MA, USA) and was collected at 18 MΩ or higher. Borosilicate glass bars (5 mm × 10 cm) were obtained from Beijing Glass Instrument Factory (Beijing, China). Marbofloxacin (MAR), danofloxacin mesylate (DAN), flumequine (FLU), enrofloxacin (ENR) and orbifloxacin (ORB) were purchased from Dr. Ehrenstorfer (Augsburg, Germany). The ciprofloxacin (CIP) was purchased from Sigma-Aldrich (St. Louis, MO). Ofoxacin (OFL), norfloxacin (NOR) and levofloxacin (LEV) were purchased from National Institutes for Food and Drug Control (Beijing, China). Lomefloxacin

(LOM)

and

Fleroxacin

(FLE)

were

purchased

from

National Institute For The Control Of Pharmaceutic (Beijing, China). The purity of all drugs was greater than 99%. The mAb to QNs, named XXQ-6-2-C1 were produced by our laboratory against CIP. The mAb showed cross-reactivities with 11 QNs in the range of 47.7%-128% (Table 1). Individual QNs stock solutions (100 μg/mL) were prepared by dissolving 10 mg of standard in 2 mL 0.03% sodium hydroxide and make a final volume of 100 mL with MeOH. Working solutions were prepared daily by a step-by-step dilution with MeOH to a final concentration of 1μg/mL. All stock solutions were stored in amber glass bottles at 4 C in the dark for 3 months. Phosphate buffered saline (PBS) consisted of 1.8 mM phosphate monobasic, 11.4 mM sodium phosphate dibasic, 014 M sodium chloride and 2.7 mM potassium chloride and the pH was adjusted to 7.4 with 1 M sodium hydroxide. The PBS containing 0.01% sodium azide was prepared by dissolving 0.2 g sodium azide to a final volume of 1 L with PBS. 7

2.2.Instruments The vortex mixer was purchased from North-Biotechnology Co. Ltd. (Beijing, China), and the centrifuge was purchased from Hunan Cence Instrument Co. Ltd. (H1650, Hunan, China). The heating magnetic stirrer was from Tianjin Honour Instrument Co. Ltd. (EMS-9A, Tianjin, China). Samples were agitated during extraction and desorption process using a rotary shaker purchased from Junyi Electrophoresis Instrument Co. Ltd. (JY15B, Beijing, China). The 12-sample nitrogen evaporator with a heating bath was purchased from Organomation Associates Inc. (Berlin, MA, USA). The HPLC system was from Agilent Technologies 1260 Infinity (Santa Clara, USA), and was equipped with a quaternary pump and fluorescence detector. The chromatographic separation of the QNs was carried out on an Inert Sustain C18 column (150 × 4.6 mm i.d., 6 μm, Shimadzu, Tokyo, Japan) with gradient elution. Mobile phase consisted of water containing 0.1% formic acid (A) and acetonitrile (B). The flow rate was 0.8 mL min-1 and the column temperature was set to 35 C. The injection volume was 20 μL. The gradient program and wavelengths for excitation and emission are shown as follows: 0-20 min, 15% B; 20-27 min, 15%-50% B; 27-35 min, 50% B; 35-45 min, 50%-90% B; 45-55 min, 90%-15% B. The excitation/emission wavelengths were programmed at 297/515 nm for MAR (0-7.3 min), at 280/450 nm for NOR, LEV, OFL, FLE, CIP, LOM, DAN, ENR and ORB (7.3-20 min), and at 320/365 nm for FLU (20-35 min). The retention time of each QN was shown in Table 1.

8

2.3.Immunoaffinity stir bar preparation The mAbs were bound to glass bars using glutaraldehyde (GA) as the coupling reagent according to Yuan et al [21]. and Lord et al [22, 23]. Briefly, the lower halves of glass bars were cleaned and etched by immersing in a mixture of 70 mL 98% sulfuric acid and 30 mL 30% hydrogen peroxide. After the mixture lowered to room temperature, heating mixture in water bath at 80 C for 1 h. Bars were then thoroughly rinsed with water, absolute ethanol and water. Particular care was taken in case of contamination. The cleaned parts of bars were silanized with ethanolic APTES solution (5 mL APTES, 5 mL water and 90 mL absolute ethanol) for 24 h at room temperature, followed by rinsing with water and absolute ethanol. The bars were placed in a vacuum oven with nitrogen flushed at 70 C overnight or 15 h. Bars were activated by placing in 100 mL of PBS containing 2.5 mL GA for 7 h, followed by rinsing with water. After rinsing, the GA activated surface were immersed in 10 mL mAb PBS solution (0.5 mg/mL) to a depth of 3 cm and stirred for 24 h at 4 C using a magnetic stirrer. It was found that the specific binding of bars and antibody reached saturation when antibody concentration over 0.6 ng/mL[23]. Subsequently, the bars were washed with water and the unreacted aldehyde groups were removed with 0.3 M aqueous ethanolamine solution (adjusted to pH 7.4 with HCl). Finally, bars were stored in PBS containing 0.01% sodium azide and 0.2 mg/mL sodium cyanoborohydride at 4 C for 48 h in order to strength covalent bond. The same procedure was used to produce a control immunoaffinity stir bar without antibody. 9

2.4.Column capacity determination A relatively large amount (1 μg) of each QN was mixed with 2 mL PBS containing 300 μL MeOH. Bars were immersed in solutions described above and agitated at 30 rpm on a rotary shaker for 30 min. Then the QNs saturated bars were washed with water to remove the carryover of drug to the desorption solution. After this procedure, the bars were put in 900 μL MeOH and PBS mixture (8:2, v/v) to elute the analytes, which was gently stirred at room temperature for 20 min. The elute was dried under a stream of high purity grade nitrogen at 50 C. The residue was reconstituted in 0.2 mL of mobile phase and monitored by LC with FLD detection. Immunoaffinity stir bar was regenerated by rinsing with water and returned to PBS containing 0.1% sodium azide at 4 C when not in use. 2.5.Sample preparation procedure for immunoaffinity stir bar The immunoaffinity stir bar coupled with LC-FLD procedure was optimized for each condition, including extraction time (10, 20, 25, 30, 35 and 40 min); stirring rate (15, 30, 45 and 60 rpm); various proportions of elution solution (MeOH: PBS, absolute MeOH, 9:1, 8:2, 7:3, v/v); desorption time (5, 10, 15, 20, 25 and 30 min) and volume of desorption solution (700, 800, 900 and 1000 μL). The following is the immunoaffinity stir bar method for analysis of QNs in milk after optimization. Bovine milk was obtained from different supermarkets (Tianjin, China). 4.0 g sample was accurately weighed into a 10 mL polypropylene centrifuge tube and spiked with standard working solution. Then, the samples stand in the dark for 30 min at room temperature before centrifugation for 10 min at 10000 rpm, the upper fat 10

layer was removed using a Pasteur pipette and the procedure was repeated twice. Prior to use, the bars were taken out of 4 C, and warmed up to room temperature. After sample treatment, the bars were placed in milk and the tubes were shaken on a rotary shaker for 30 min at 30 rpm. The bars were washed water to remove the carryover of drug to the desorption solution, the compound was eluted in another 2 mL polypropylene centrifuge tube with V-bottom containing 900 μL MeOH and PBS mixture (8:2, v/v), which was gently shaken at room temperature for 20 min. The elution was dried at 50 C under a stream of high purity grade nitrogen and was reconstituted in 0.2 mL of mobile phase. Then, filtered through 0.22 μm nylon membrane filters, 20 μL was injected into the chromatographic system and monitored by FLD.

3. Results and discussion 3.2.Preparation of immunoaffinity stir bar This study was carried out to develop a rapid, convenient and sensitive method to determine QNs using immunoaffinity stir bar. Borosilicate glass bars were easy to access and were chosen as solid support material due to its good solidity, ideal mechanical stability and low cost. The most important part in preparing immunoaffinity stir bar was the selection of the antibody and amount of the immunosorbent. Two types of antibodies, mAb and polyclonal antibody (pAb) were used as immunosorbent. The pAb is a mixture of antibodies, which is easy to produce, but has a disadvantage of binding with various immunogens, and shows less selectivity in preparing immunoaffinity stir bar for detecting analytes [22]. The mAb 11

has homogeneous binding property and only binds to a single binding site. Although mAb is expensive and difficult to produce, the mAb preparation is reproducible when a hybridoma line is developed [27]. Therefore, mAb is the best choice for immunoaffinity stir bar preparation. In this study, only a small amount (5 mg) of mAb was needed in the whole immunoaffinity stir bar process compared with IAC process published before (28 mg) [28-30], but limit of detection (LOD) and limit of quantity (LOQ) were not influenced by using less mAbs. The immobilization method to immobilize antibodies to glass bars was by means of covalent linking using GA, which was first employed by Yuan et al [21]. and improved by Lord et al [22, 23]. The amino groups from the antibodies were bound to GA activated silica surfaces of glass bars. The advantage of covalent linking over non-covalent linking is that, covalent linking can improve binding capacity and stability of antibody on glass bars. The resulting immunoaffinity stir bar shows good performance in detecting QNs in bovine milk samples. 3.3.Optimization of the extraction and elution conditions. Organic solvents, such as MeOH [29, 31, 32], ACN [12, 33, 34], trichloroacetic acid [31, 35] and dichloromethane [36], are used for extracting QNs in previously reported papers. Most of the organic solvents are toxic and volatile, which is not only harmful to researchers’ health but are also bad for environment, for example some chlorinated solvents are difficult to decompose even for a long time. To simplify and develop an environment-friendly sample pretreatment for determination of QNs, we tried to use organic solvents as less as possible. 12

Optimization of extraction time, stirring rate, proportions of elution solution, desorption time and volume of desorption solvent, and is very important for analyzing QNs in bovine milk because these parameters can have a great influence on antibody performance during extraction and desorption procedure. Extraction was performed on a shaker with 30 rpm by placing immunoaffinity stir bar in 4 g bovine milk sample which spiked with DAN at a concentration of 0.1 ng/g. Extraction time to reach equilibrium between analytes and antibody was optimized from 10 to 40 min at six points (Fig.1A). It can be seen in Fig 1A when extraction time was 30 min, the peak area tends to plateau, indicating that the stir bar extracted maximum amounts of DAN. Therefore, 30 min was selected in further analyses. Different levels of stirring rates (15, 30, 45, 60 rpm) were evaluated (Fig.1B). High agitating speed can facilitate mass transfer of DAN towards immunosorbent and reach equilibrium time faster. Figure 1B shows that with stirring rates beyond 30 rpm, no improvement in adsorption was found, so 30 rpm was chosen to perform the following research. After extraction, the bars were washed with water for about 5 s to minimize carryover of DAN into desorption solvent [22]. The selection of elution solvent is of great importance to working life of immunoaffinity stir bar. MeOH is usually used as elution solution, but previous studies show that absolute MeOH can result in a low recovery [28]. However, when PBS is added in MeOH as elution solvent, it shows good recoveries. Therefore, MeOH and PBS mixture in various proportions (absolute MeOH, 9:1, 8:2, 7:3, v/v) was optimized in this study (Fig.1C). It was found that 13

optimal elution solution was MeOH and PBS mixture (8:2, v/v). This may be explained by the fact that MeOH in PBS leads to loss of hydrogen or ionic bonds between analyte and adsorption phase [37]. To minimize analyses time, six desorption time points from 5 min to 30 min at interval of 5 min were evaluated. Fig.1D shows that peak area of DAN was highest at 20 min and then decreased. The reason may be that a large amount of MeOH can denature antibodies and may cause serious damage to the immunoaffinity stir bar when staying in MeOH for a relatively long time [38]. The effect of elution volume (700, 800, 900, 1000 μL) was also investigated, as presented in Figure.1E. The response of DAN increased with addition of elution volume until 900 μL. Hence, 900 μL was chosen for subsequent research, that was evaporated easily and environment-friendly. At the same concentration, we found that peak area value of DAN was higher than any other QNs in this study. It shows more directly to readers that the difference of results in optimization process due to DAN’s high peak area value, that’s the reason why we choose DAN as a typical model in optimization process. The results for the other QNs are giving the similar trend with those obtained from DAN studies. 3.4.Column capacity determination After a number of parameters were optimized, 1000 ng each of the 11 QNs standard solutions was individually extracted by immunoaffinity stir bar with gentle shaking (30 rpm) to determine column capacity. Then, washed with water and eluted with 900 μL MeOH and PBS mixture (8:2, v/v). The column capacity for the 11 QNs 14

are presented in Table 1. It shows that CIP, NOR, LOM and DAN have a relative higher column capacity of at least 0.57 pmol/cm2 than that of other QNs. The specificity of immunoaffinity stir bar was observed. A control stir bar was used to measure column capacity of QNs, no analytes were found to be extracted and eluted. An immunoaffinity stir bar was used to measure column capacity of other drugs (tetracycline and chloramphenicol), which was found to be approximately zero. The results above demonstrated that the QNs can be specifically retained by immunoaffinity stir bar through antibody-antigen interactions not the adsorption on stir bars. The preparation of immunoaffinity stir bar is relatively expensive and time consuming. Therefore, the reusability of the stir bar was evaluated once every 3 days for 45 days by column capacity determination. The obtained changing curve of the column capacity over 45 days was shown in Fig 2. The column capacity for QNs declined rapidly at first use, and then decreased slowly with the increasing cycles of use. It can be explained by the fact that MeOH and PBS mixture (8:2, v/v) was used to elute analytes that can cause loss of antibodies which are not firmly immobilized on stir bar at first use. High MeOH concentration can gradually decrease column capacity through irreversible denaturation of the immobilized mAbs during the elution step, and the denatured mAbs didn’t participate in extraction for subsequent uses. Rashid discard the elute when using IAC to purify clenbuterol in plasma at first use, and collect data from the second use in order to eliminate the influence of capacity loss at first use on research data [39]. The immunoaffinity stir bar was found to have lost about 80% of its column capacity after 25 extraction/elution cycles, but 15

recoveries of QNs were not affected. The sensitivity of SBSME method is directly determined by the sensitivity of the LC detector employed [22]. Thus, when using a more sensitive MS or MS/MS as detector, immunoaffinity stir bar could potentially be used to analyze forbidden veterinary medicine more than 25 cycles , because these medicine only needs to be qualitatively detected. 3.4. Method validation The developed method was validated in terms of linearity, recovery, accuracy, precision, LOD and LOQ according to the analytical method validation criteria. The matrix effects can cause interference in analysis. It has been reported that selective extraction methods could eliminate or minimize matrix effects problem [40]. Therefore, a matrix-matched calibration was not used in this study due to the limited column capacity. External calibration curve was created by plotting peak area for each of QNs standard solutions versus different concentration over the range from 0.1 to 100 ng/mL, each concentration level was injected in triplicate (instrumental replicates). The calibration curve for each analyte shows that good linearity with a close correlation between the concentration and the peak area ranging between 0.9993 and 1. The LOD and LOQ of the method were defined as signal-to-noise (S/N) ratio of 3:1 and 10:1, respectively. As shown in Table 2, the LOD was 0.05 ng/g for DAN and 0.1 ng/g for CIP, NOR, LOM, ENR, FLE, MAR, LEV, OFL, ORB and FLU, respectively. LOQ was 0.2 ng/g for DAN and 0.3 ng/g for CIP, NOR, LOM, ENR, FLE, MAR, LEV, OFL, ORB and FLU, respectively. These values are significantly 16

lower than those obtained with previously published methods [33, 41], indicating that the developed method is sensitive enough. The main reason is that previous cleanup methods used SPE or liquid-liquid extraction which lack specificity and employ large amounts of organic solvents. The precision and accuracy was evaluated at a spiked concentration of 0.1 ng/g for each blank bovine milk sample by determining recoveries and relative standard deviation (RSD). As shown in Table 3, the mean recoveries of the 11 QNs ranged from 11.8% to 40%. The recoveries are higher than the state-of-art immunoaffinity SPME method by Lord et al.22 developed immunoaffinity solid phase microextraction probes for analysis of 7-aminoflunitrazepam in urine with recoveries ranging from 1% to 27%. It may be explained by the fact that the stir bar capacity is small, because the limited amounts of antibody was covalently bound to glass bar and covalent linking was weaken by organic solvents in clean-up procedure. Though recovery was low, a low recovery doesn’t mean a poor precision, the intra-day and inter-day RSDs were in the range of 3.2%-11.9% and 5.2%-12.5%, respectively. Figure 3 shows that chromatograms of a milk sample spiked with 0.1 ng/g of MAR using the immunoaffinity SBSME-LC-FLD and blank samples of bovine milk. No interfering peaks were found at the retention time of MAR, illustrating the cleanup was successful.

4. Conclusion The immunoaffinity SBSME-LC-FLD method was developed for the determination of 11 QNs in bovine milk samples. The whole sample preparation 17

procedure was significantly simplified and less than 900 μL organic solvent was used. Moreover, the immunoaffinity stir bar has a high degree of stability that can be used many times, and only small amounts of antibody was needed in stir bar preparation compared to IAC preparation. The method is environment friendly and economical. Though lower recoveries were achieved in this study, the method can meet the demand for selective extraction QNs from bovine milk. It is feasible to make immunoaffinity stir bars specific for other veterinary drugs which are prohibited by authorities

in

food

sample

of

animal-origin,

such

as

chloramphenicol,

diethylestilbestrol and clenbuterol. Its column capacity also has great potential to be improved by increasing the active binding sites on glass bar or using other forms of solid support which has large surface area in further research.

Acknowledgment We thank Dr. Peiqiang Yu (Department of Animal and Poultry Science, College of Agriculture and Bioresources, University of Saskatchewan, Canada) for assistance in preparing the manuscript. This work was supported by grants from the National Science Fund of China (No. 31072172 and No. 31372482), the Veterinary Biotechnology Scientific Research Innovation Team of Tianjin, China (Grant No.TD12-5019), and the Veterinary “Young and Middle-aged Backbone of Innovation Talent Culture Project” of Tianjin, China.

18

References 1. B. Suh, B. Lorber, Quinolones, Med. Clin. N Am. 79 (1995) 869-894. 2. J. Z. Shen, L. J. Xie, Synthetic antibacterial, in: Veterinary pharmacology, China Agricultural University Press, Beijing, China, 2000, pp. 162-167. 3. Commission Regulation (EC) No. 508/1999 of 4 March 1999, Off. J. Eur. Communities L60 (1999) 20-47. 4. T. J. Mi, Z. H. Wang, S. A. Eremin, J. Z. Shen, S. X. Zhang, Simultaneous determination of multiple (fluoro)quinolone antibiotics in food samples by a one-step fluorescence polarization immunoassay, J. Agric. Food. Chem. 61(2014) 9347-9355. 5. A. C. Huet, C. Charlier, S. A. Tittlemier, G. Singh, S. Benrejeb, P. Delahaut, Simultaneous determination of (fluoro)quinolone antibiotics in kidney, marine products, eggs, and muscle by enzyme-linked immunosorbent assay (ELISA), J. Agric. Food Chem. 54 (2006) 2822-2827. 6. A. Rusu, G. Hancu, G. Völgyi, G. Tóth, B. Noszál, A. Gyéresi, Separation and determination of quinolone antibacterials by capillary electrophoresis, J. Chromatogr. Sci. 52 (2014) 919-925. 7. M. Lombardo- Agüí, L. Gámiz-Gracia, A. M. García-Campaña, C. Cruces-Blanco, Sensitive determination of fluoroquinolone residues in waters by capillary electrophoresis with laser-induced fluorescence detection. Anal. Bioanal. Chem. 396 (2010) 1551-1557. 8. H. L. Meng, G. H. Cheng, X. Guo, P. Chen, Q. H. Cai, Y. F. Tian, Determination of 19

five quinolone antibiotic residues in foods by micellar electrokinetic capillary chromatography with quantum dot indirect laser-induced fluorescence, Anal. Bioanal. Chem. 406 (2014) 3201-3208. 9. E. N. Evaggelopoulou, V. F. Samanidou, B. Michaelidis, I. Papadoyannis, Development and validation of an LC-DAD method for the routine analysis of residual quinolones in fish edible tissue and fish feed. Application to farmed gilthead sea bream following dietary administration, J. Liq. Chromatogr. Related Technol. 37 (2014) 2142-2161. 10. J. W. Jang, K. S. Lee, K. Kwon, S. C. Bae, H. S. Kim, Simultaneous Determination of thirteen quinolones in livestock and fishery products using ultra performance LC with electrospray ionization tandem mass spectrometry, Food Sci. Biotechnol. 22 (2013) 1-9. 11. N. Dorival-García, A. Zafra-Gómez, S. Cantarero, A. Navalón, J. L. Vílchez, Simultaneous determination of 13 quinolone antibiotic derivatives in wastewater samples

using

solid-phase

extraction

and

ultra

performance

liquid

chromatography-tandem mass spectrometry, Microchem. J. 106 (2013) 323-333. 12. A. Gajda, A. Posyniak, J. Zmudzki, M. Gbylik, T. Bladek, Determination of (fluoro) quinolones in eggs by liquid chromatography with fluorescence detection and confirmation by liquid chromatography-tandem mass spectrometry, J. Food Chem. 135 (2012) 430-439. 13. S. Lee, Y. Shim, H. Kim, D. Shin, Multiresidue determination of quinolones in porcine, chicken, and bovine muscle using liquid chromatography with fluorescence 20

detection, Food Sci. Biochem. 18 (2009) 978-984. 14. M. Kanda, T. Kusano, S. Kanai, H. Hayashi, Y. Matushima, T. Nakajima, Rapid determination of fluoroquinolone residues in honey by a microbiological screening method and liquid chromatography, J. AOAC Int. 93 (2010) 1331-1339. 15. D. Nathalie, P. Valérie, H. Marie-Claire, Immunoaffinity solid-phase extraction for the trace-analysis of low-molecular-mass analytes in complex sample matrices, J. Chromatogr. B 745 (2000) 15-37. 16. X. L. Wu, L. Xiang, C. H. Mo, Y. W. Li, Y. N. Jiang, Q. Y. Yan, X. Lu, X. P. Huang, Determination of quinolones in vegetables using ultra performance liquid chromatography-electrospray ionization tandem mass spectrometry, Chin. J. Chem . 41 (2013) 876-881. 17. X. S. Meng, Q. Ma, Q. Zhang, Q. Lv, H. Bai, C. Wang, W. T. Li, Simultaneous determination of 16 fluoroquinolone antibiotics in cosmetics by ultra-performance liquid chromatography/triple quadrupole mass spectrometry with ultrasound-assisted extraction and solid-phase extraction, Anal. Methods 7 (2015), 675-683. 18. N. Dorival-Garcia, A. Zafra-Gómez, S. Cantarero, A. Navalon, J. L. Vilchez, Simultaneous determination of 13 quinolone antibiotic derivatives in waste water samples

using

solid-phase

extraction

and

ultra

performance

liquid

chromatography-tandem mass spectrometry, Microchem. J. 106 (2013) 323-333. 19. C. W. Tsai, C. S. Lin, W. H. Wang, Multi-residue determination of sulfonamide and

quinolone

residues

in

fish

tissues

by

high

performance

liquid

chromatography-tandem mass spectrometry (LC-MS/MS), J. Food. Drug. Analysis 21

20 (2012) 674-680. 20. A. Spielmeyer, J. Ahlborn, G. Hamscher, Simultaneous determination of 14 sulfonamides and tetracyclines in biogas plants by liquid-liquid-extraction and liquid chromatography tandem mass spectrometry, Anal. Bioanal. Chem. 406 (2014) 2513-2524. 21. H. D. Yuan, W. M. Mullett, J. Pawliszyn, Biological sample analysis with immunoaffinity solid-phase microextraction, Analyst 126 (2001) 1456-1461. 22. H. L. Lord, M. Rajabi, S. Safari, J. Pawliszyn, Development of immunoaffinity solid phase microextraction probes for analysis of sub ng/mL concentrations of 7-aminoflunitrazepam in urine. J. Pharm. Biomed. Anal. 40 (2006) 769-780. 23. H. L. Lord, M. Rajabi, S. Safari, J. Pawliszyn, A study of the performance characteristics of immunoaffinity solid phase microextraction probes for extraction of a range of benzodiazepines. J. Pharm. Biomed. Anal. 44 (2007) 506-519. 24. M. E. C. Queiroz, E. B. Oliveira, F. Breton, J. Pawliszyn, Immunoaffinity in-tube solid phase microextraction coupled with liquid chromatography-mass spectrometry for analysis of fluoxetine in serum samples, J. Chromatogr. A 1174 (2007) 72-77. 25. A. R. Chaves, M. E. C. Queiroz, Immunoaffinity in-tube solid phase microextraction coupled with liquid chromatography with fluorescence detection for determination of interferon alpha in plasma samples. J. Chromatogr. B 928 (2013) 37-43. 26. Y. H. Liu, H. Lord, M. Maciazek-Jurczyk, S. Jolly, M. A. Hussain, J. Pawliszyn, Development of an immunoaffinity solid phase microextraction method for the 22

identification of penicillin binding protein 2a. J. Chromatogr. A 1364 (2014) 64-73. 27. R. R. Burgess, N. E. Thompson, Advances in gentle immunoaffinity chromatography. Curr. Opin. Biotechnol. 13 (2002) 304-308. 28. S. J. Zhao, X. L. Li, Y. K. Ra, C. Li, H. Y. Jiang, J. C. Li, Z. N. Qu, S. X. Zhang, F. Y. He, Y. P. Wan, C. W. Feng, Z. R. Zheng, J. Z. Shen, Developing and optimizing an immunoaffinity cleanup technique for determination of quinolones from chicken muscle. J. Agric. Food. Chem. 57 (2009) 365-371. 29. C. Li, Z. H. Wang, X. Y. Cao, R. C. Beier, S. X. Zhang, S. Y. Ding, X. W. Li, J. Z. Shen, Development of an immunoaffinity column method using broad-specificity monoclonal antibodies for simultaneous extraction and cleanup of quinolone and sulfonamide antibiotics in animal muscle tissues. J. Chromatogr. A 1209 (2008) 1-9. 30. B. Y. Li, C. Li, H. Y. Jiang, Z. H. Wang, X. Y. Cao, S. J. Zhao, S. X. Zhang, J. Z. Shen, Purification of nine sulfonamides from chicken tissues by immunoaffinity chromatography using two monoclonal antibodies, J. AOAC Int. 91 (2008) 1488-1493. 31. J. A. R. Paschoal, F. G. R. Reyes, S. Rath, Determination of quinolone residues in tilapias (Orechromis niloticus) by HPLC-FLD and LC-MS/MS QToF. Food Addit. Contam. A 26 (2009) 1331-1340. 32. C. Robert, N. Gillard, P. Y. Brasseur, N. Ralet, M. Dubois, P. Delahaut, Rapid multiresidue and multi-class screening for antibiotics and benzimidazoles in feed by ultra high performance liquid chromatography coupled to tandem mass spectrometry, Food Control 50 (2015) 509-515. 23

33. C. Blasco, Y. Pico, Development of an improved method for trace analysis of quinolones in eggs of laying hens and wildlife species using molecularly imprinted polymers, J. Agric. Food. Chem. 60 (2012) 11005-11014. 34. X. L. Hou, G. Chen, Li. Zhu, T. Yang, J. Zhao, L. Wang, Y. L. Wu, Development and validation of an ultra high performance liquid chromatography tandem mass spectrometry method for simultaneous determination of sulfonamides, quinolones and benzimidazoles in bovine milk, J. Chromatogr. B 962 (2014) 20-29. 35. E. Verdon, P. Couedor, B. Roudaut, P. Sandérs, Multiresidue method for simultaneous

determination

of

ten

quinolone

anti-bacterial

residues

in

multimatrix/multispecies animal tissues by liquid chromatography with fluorescence detection: single laboratory validation study, J. AOAC. Int. 88 (2005) 1179-1192. 36. O. R. Idowu, J. O. Peggins, Simple, rapid determination of enrofloxacin and ciprofloxacin in bovine milk and plasma by high-performance liquid chromatography with fluorescence detection, J. Pharm. Biomed. Anal. 35 (2004) 143-153. 37. M. A. Firer, Efficient elution of functional proteins in affinity chromatography, J. Biochem. Bioph. Meth. 49 (2001) 433–442. 38. E. Watanabe, Y. Yoshimura, Y. Yuasa, H. Nakazawa, Immunoaffinity column clean-up for the determination of imazalil in citrus fruits, J. Anal. Chim. Acta 433 (2001) 199-206. 39. B. A. Rashid, P. Kwasowski, D. Stevenson, Solid phase extraction of clenbuterol from plasma using immunoaffinity followed by HPLC, J. Pharm. Biomed. Anal. 21 (1999) 635-639. 24

40. B. K. Matuszewski, M. L. Constanzer, C. M. Chavez-Enq, Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLC-MS/MS, Anal. Chem. 75 (2003) 3019-3030. 41. A. Zafra-Gómez, A. Garballo, O. Ballesteros, A. Navalón, L. E. García-Ayuso, Simultaneous determination of quinolone antibacterials in bovine milk by liquid chromatography-mass spectrometry, Biomed. Chromatogr. 22 (2008) 1186-1193.

25

Figure Captions

Fig.1. Effect of different parameters on immunoaffinity stir bar sorptive microextraction (SBSME) performance. (A) Extraction time; (B) Stir rate; (C) Various proportions of elution solution; (D) Desorption time; (E) Elution volume.

26

Fig.2. Variation curve of immunoaffinity stir bar capacity for DAN with 15 cycles of use over 45 days.

27

Fig.3. Chromatograms of (A) MAR standard (0.8 ng/mL), (B) spiked bovine milk sample (0.1 ng/g) and (C) blank bovine milk sample.

28

Tables Table 1. Retention time, column capacities, and cross-reactivities for the QNs Analytes MWa CCb (pmol/cm2) CRc (%) TRd (min) CIP 332 2.15 128.0 8.753 LOM 352 1.25 52.5 10.464 DAN 358 0.57 71.1 10.500 FLE 370 0.48 55.2 7.986 OFL 362 0.35 68.1 7.888 ENR 360 0.54 114.3 12.212 NOR 320 1.69 100.0 7.782 MAR 363 0.46 54.2 6.648 LEV 371 0.43 69.6 7.825 ORB 396 0.38 56.1 13.958 FLU 262 0.43 47.7 29.808 a b c MW, molecular weight. CC, column capacity. CR, cross-reactivity. d

tR, retention time

29

Table 2. Calibration curve, LOD and LOQ for each QNs drug Analytes

a

Linear range (ng/mL)

CIP

0.1-100

LOM

0.1-100

DAN

0.1-100

FLE

0.1-100

OFL

0.1-100

ENR

0.1-100

NOR

0.1-100

MAR

0.1-100

LEV

0.1-100

ORB

0.1-100

FLU

0.1-100

Working standard (ng/mL) 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100 0.1, 0.5, 1,10, 50,100

Calibration equation

R2 (n=3)

LODc (ng/g)

LOQd (ng/g)

ya=0.224cb+0.0653

1

0.1

0.3

y=0.1644c-0.0667

0.9997

0.1

0.3

y=2.0511c+0.3254

0.9999

0.05

0.2

y=0.1672c-0.2095

0.9995

0.1

0.3

y=0.0249c+0.0107

0.9998

0.1

0.3

y=0.2524c+0.5261

0.9999

0.1

0.3

y=0.3025c+0.3467

0.9993

0.1

0.3

y=0.0568c+0.0865

0.9997

0.1

0.3

y=0.0255c+0.0301

0.9996

0.1

0.3

y=0.2243c+0.0203

0.9999

0.1

0.3

y=0.0496c+0.0123

0.9996

0.1

0.3

y, peak area. bc, analyte concentration (ng/mL). cLOD, limit of detection. dLOQ,

limit of quantity.

30

Table 3. Recoveries and RSD for 11 QNs in bovine milk Analytes

Spiked concentration (ng/g)

CIP LOM DAN FLE OFL ENR NOR MAR LEV ORB FLU a

0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1

a

Mean recovery (%)

intra-day (n=5) 13.2 28.3 20.1 35.9 11.8 15.7 32.4 40.0 23.4 29.6 21.5

RSD, relative standard deviation.

31

3.2 4.8 11.9 7.8 6.6 7.0 5.9 4.6 7.4 4.4 9.1

RSD (%) inter-day (n=3) 6.8 6.3 12.5 9.5 8.7 7.5 10.4 6.1 8.7 5.2 7.5