Multi-residue Method for Determination of 238 Pesticides in Chinese

Nov 7, 2014 - In cucumber, 203 pesticides were in the 70–120% recovery range with good reproducibility by PSA mini-cartridge column cleanup at a spi...
0 downloads 7 Views 2MB Size
Download PDF
Article pubs.acs.org/JAFC

Multi-residue Method for Determination of 238 Pesticides in Chinese Cabbage and Cucumber by Liquid Chromatography−Tandem Mass Spectrometry: Comparison of Different Purification Procedures Mei-Ai Zhao,†,§ Ya-Nan Feng,‡,§ Yong-Zhe Zhu,*,‡ and Jeong-Han Kim∥ †

College of Life Science, and ‡College of Chemistry and Pharmacy, Qingdao Agricultural University, Changcheng Road, Chengyang, Qingdao, Shandong 266-109, People’s Republic of China ∥ Department of Agricultural Biotechnology, Seoul National University, 599 Gwanak-ro, Silim-dong, Gwanak-gu, Seoul 151-742, Republic of Korea S Supporting Information *

ABSTRACT: This paper describes the comparison of five sample cleanup procedures for the determination of 238 pesticides via triple quadrupole liquid chromatography−tandem mass spectrometry (LC−MS/MS, with only 10 min of chromatographic running time) in Chinese cabbage and cucumber. Samples were extracted with a quick, easy, cheap, effective, rugged, and safe (QuECHERS) preparation method and cleanup with different sorbents, including primary secondary amine (PSA), multi-walled carbon nanotubes (MWCNTs), and polystyrene (PLS), to find out the most suitable cleanup methods for Chinese cabbage and cucumber. The recovery and matrix effect were evaluated by monitoring the main parameters in one group of 238 pesticides at the spiked level of 8 and 40 μg/kg. In Chinese cabbage, when PSA dispersive solid-phase extraction (D-SPE) was applied, recoveries of 183 pesticides ranged between 70 and 120% with relative standard deviation (RSD) values lower than 20% at a spiked level of 40 μg/kg, indicating the effectiveness of the purification step. In cucumber, 203 pesticides were in the 70−120% recovery range with good reproducibility by PSA mini-cartridge column cleanup at a spiked level of 40 μg/kg and RSD values were generally below 20%. The limits of quantitation [LOQs; signal-to-noise (S/N) = 10] were in the range of 0.16−10.20 μg/kg for Chinese cabbage and 0.06−21.06 μg/kg for cucumber, while the limits of detection (LODs; S/N = 3) were between 0.05 and 3.06 μg/kg and between 0.02 and 6.32 μg/kg in Chinese cabbage and cucumber, respectively. The proposed methods that might be applied for the multi-residue analysis in Chinese cabbage and cucumber are contributed to their rapid speed and good recoveries. KEYWORDS: multi-residue, pesticides, purification, Chinese cabbage, cucumber, liquid chromatography−tandem mass spectrometry (LC−MS/MS)

■

INTRODUCTION Pesticides are part of a large group of organic compounds that present extremely diverse physicochemical properties and are widely used around the world in the control or prevention of weeds or crop diseases.1 The increasing vegetable intake could significantly increase pesticide exposure and health risk in humans.2 The current problems in food safety and phytosanitar barriers demand a rigorous control in correctly identifying and quantifying residues of pesticides present in vegetables and fruits as well as the absence of banned pesticides for certain crops. Thus, there is a growing requirement from companies and governmental agencies to improve the analytical performance in pesticide residue monitoring, making it necessary to increase the efficiency and develop cost-effective and less timeconsuming methods for determining multi-residue pesticides. Chromatographic techniques, including high-performance liquid chromatography (HPLC) and gas chromatography (GC), have usually been applied for the determination of pesticide residues in food samples.3−6 Liquid chromatography− tandem mass spectrometry (LC−MS/MS) or its modified procedures were employed for the analyses of a wide range of pesticides in fruits, vegetables, and cereals. In comparison of the LC−MS/MS analytical results to those from gas chromatog© 2014 American Chemical Society

raphy−mass spectrometry (GC−MS), LC−MS/MS has emerged as a preferred technique, especially for those pesticides that are compatible in both analytical systems, especially for polar, thermoplastic instability, low-volatile compounds.7 In addition, the combination of LC−MS/MS and GC−MS can also provide the effective and efficient means to both identify and quantify hundreds of pesticide analytes in a variety of complex matrices. Pang et al. employed both LC−MS/MS and GC−MS for the analysis of 446 pesticides in fruits and vegetables and 450 pesticides in honey, fruit juice, and wine.8,9 Recently, triple quadrupole LC−MS/MS has been applied to detect multi-class pesticide residues in olive oil and rapeseed with high acquisition speed, selectively, and detectability.10,11 In 2003, the quick, easy, cheap, effective, rugged, and safe (QuEChERS) method was introduced by Anastassuades and Lehoaty et al.12,13 This technique has been widely used as a pesticide multi-residue method in vegetables, fruits, and many other matrices because of its advantages of high extraction Received: Revised: Accepted: Published: 11449

July 30, 2014 November 6, 2014 November 7, 2014 November 7, 2014 dx.doi.org/10.1021/jf504570b | J. Agric. Food Chem. 2014, 62, 11449−11456

Journal of Agricultural and Food Chemistry

Article

“C”. The mixture was shaken vigorously for 2 min and centrifuged at 13 000 rpm for 2 min. Then, 400 μL of supernatant was transferred to an autosampler vial, followed by the addition of 100 μL of acetonitrile [0.5% (v/v) formic acid], and this was then filtered through a 0.2 μm polytetrafluoroethylene (PTFE) filter. Moreover, the 30 mg PSA mini-cartridge column for method “D” and the 20 mg PLS mini-cartridge column for method “E” were also used for sample purification. The columns were preconditioned with 2 mL of acetone and activated with 2 mL of acetonitrile. After loading 1 mL of acetonitrile layer, the column was eluted with 1 mL of methanol [0.1% (v/v) formic acid] as the elution solvent. Then, the sample was dissolved in 1 mL of acetonitrile after evaporating the methanol, and 400 μL of solution was transferred to an autosample vial, to which 100 μL of acetonitrile [containing 0.5% (v/v) formic acid] was added for analysis. LC−MS/MS Analysis. Triple quadrupole LC−MS/MS was a Shimadzu LCMS-8030 (Japan) equipped with an electrospray ionization (ESI) source, operating in positive and negative ionization modes. LC was carried out with two Shimadzu LC-30AD pumps (Japan), a Sil-30AC autosampler, a CTO-30AC column oven, and a CBM-20A controller. A Phenomenex Kinetex C18 column (Phenomenex, Torrance, CA), 2.6 μm, 100 × 2.1 mm, was used. Nitrogen was used in the ion source and as the collision gas. The LC−MS/MS system parameters were set as follows: flow rate, 0.5 mL/min; interface voltage, 4.5 kV; desolvation line temperature, 250 °C; nebulizing gas flow, 3 L/min; heat block temperature, 400 °C; performed gas flow, 15 L/min; and collision-induced ionization (CID) gas, 230 kPa. The temperature of the column was 40 °C, and a 10 μL amount of the sample extract was injected in each run. Mobile Phase. Mobile phase A was water, and mobile phase B was methanol, both containing 5 mM ammonium formate and 0.1% (v/v) formic acid. The gradient program was as follows: 100% A (0% B) for 0.5 min from the start, ramped to 55% B linearly for 0.5 min, followed by a linear gradient up to 95% B over 7 min, and maintained for 2 min at 95% B. A constant flow rate of 0.5 mL/min was used, and the total running time was 10 min. Optimization of MS/MS Parameters. The optimization of MS conditions was performed for each pesticide by introducing individual standard solutions directly into MS. The two most intense precursor ion−fragment ion transitions of each pesticide were chosen, and the optimal MS/MS parameters for all transitions were recorded. The final method was constructed on the basis of these data (see Table S1 of the Supporting Information). Matrix Effect Study. Change of ionization efficiency in the presence of other compounds is call the matrix effect. An average matrix effect exceeding 25% is generally considered to significantly affect quantitative analytical results,17,18 depending upon the relative standard deviation (RSD) values. To an aliquot of blank extract in methanol, a pesticide mixture, 78% of the extract volume, was added. Matrix effect values were calculated for all samples using matrix-matched calibration and a methanol calibration slope with the following formula: ME (%) = 100 × ((Sm/ Ss) − 1), where ME is the matrix effect and Sm and Ss are the slopes of the calibration curves obtained in the matrix and methanol, respectively. Recovery Experiments. A recovery study was performed by spiking each sample with a mixture of a working standard solution (10 mg/L) containing 238 pesticides to reach a final concentration of 8 and 40 μg/kg, and each spike level was repeated 3 times.

efficiency, smaller volume of organic solvents, simplicity of operation, simple and convenient equipment, and low cost per sample.14 In recent years, modified QuEChERS methods for sample preparation with LC−MS/MS or GC−MS detection obtained high recoveries and reproducibility.1,5,15−17 The above methods were approved and mainly employed in governments, food industry, and related laboratories, and much time was consumed because of complex procedures as well as the expensive cost. In the case of vegetables, the pesticide residue analysis was more different and difficult, corresponding to complex components in the matrix, and the pesticides sprayed were unknown during cultivation. In this paper, different cleanup methods [primary secondary amine (PSA), multi-walled carbon nanotubes (MWCNTs), and polystyrene (PLS) as sorbents] based on QuEChERS sample preparation were evaluated for Chinese cabbage and cucumber. The analysis was carried out by tandem mass spectrometry within 10 min. The aim of this study was to develop rapid and robust analytical methods for the simultaneous extraction and determination of 238 pesticides in Chinese cabbage and cucumber.

■

MATERIALS AND METHODS

Chemicals and Reagents. Most of the pesticide standards used in this work were purchased from Dr. Ehrenstorfer (Augsburg, Germany), Chem Service (West Chester, PA), and Fluka (Buchs, Germany), with the highest available purity. Some pesticide standards were generously provided by the Korean Food and Drug Administration. HPLC-grade acetonitrile, acetone, and methanol were obtained from Honeywell (Morristown, NJ). LC−MS-grade formic acid and ammonium formate were ordered from Fluka (Buchs, Germany). The DisQuE QuEChERS pouch as in the EN method 15662 sample preparation kit, was acquired from Waters (Milford, MA) and contained 4 g of anhydrous MgSO4, 1 g of NaCl, 1.5 g of sodium citrate dehydrate, and 0.5 g of disodium citrate sesquihydrate. Dispersive solid-phase extraction (D-SPE) 5982-5021 was purchased from Agilent Technologies (Santa Clara, CA) and contained 25 mg of PSA and 150 mg of anhydrous MgSO4. The 30 mg PSA cartridge column and 20 mg PLS cartridge column were purchased from Aisti Science (Wakayama, Japan). MWCNTs (15−20 nm) were purchased from Hanwha Nanotech (Seoul, Korea). Preparation of Standard Solutions. Individual standard solutions were prepared in acetone, acetonitrile, or methanol with a concentration of approximately 1000 mg/L and stored at −20 °C. A working standard mixture was prepared by diluting each individual standard solution with methanol, to obtain a concentration of 10 mg/L for each compound. The linearity in the response was studied with standard solution prepared in both blank matrix extra and acetonitrile solvent. Sample Preparation. Chinese cabbage (Brassica rapa) and cucumber (Cucumis sativus) were obtained as organic products from a Seoul local market in Korea, and no pesticide residue was detected in the vegetables. An approximately 200 g portion of sample was weighed and homogenized at 10 000 rpm for 1 min. A 10 g sample of homogenized Chinese cabbage (or cucumber) was weighed into a 50 mL Teflon centrifuge tube, followed by the addition of 10 mL of acetonitrile, and the tube was vigorously shaken for 1 min manually. A DisQuE QuEChERS pouch was added to each sample, which was then shaken again. The tube was centrifuged at 3500 rpm for 5 min at 4 °C after cooling on ice. Finally, the acetonitrile layer was prepared for the cleanup step. Sample Purification. A total of 1 mL of the acetonitrile layer was transferred to a 2 mL microcentrifuge tube containing different sorbents, including D-SPE 5982-5021 containing 25 mg of PSA and 150 mg of MgSO4 for method “A”. A total of 2.5 mg of MWCNTs was added to the sorbent of method “A” as method “B”. Only 10 mg of MWCNTs was used as the cleanup sorbent with D-SPE for method

■

RESULTS AND DISCUSSION LC−MS/MS Method Validation Using Standard Solutions. The linearity of each pesticide in Chinese cabbage and cucumber samples was studied in the range of 1−200 μg/L with seven calibration points (1, 5, 10, 20, 50, 100, and 200 μg/ L) by a matrix-matched standard calibration. Summary results are presented in Table S1 of the Supporting Information. These data indicated that the LC−MS/MS procedure had good linearity and high sensitivity. Approximately 96% of the 11450

dx.doi.org/10.1021/jf504570b | J. Agric. Food Chem. 2014, 62, 11449−11456

Journal of Agricultural and Food Chemistry

Article

Figure 1. Chromatogram of 238 pesticides at a spiked level of 8 μg/kg in Chinese cabbage samples. The original chromatogram (A) was divided into panels B (1.5−3.5 min), C (3.5−5.5 min), and D (5.5−7.5 min) by running time.

Figure 2. Chromatogram of 238 pesticides at a spiked level of 8 μg/kg in cucumber samples. The original chromatogram (A) was divided into panels B (1.5−3.5 min), C (3.5−5.5 min), and D (5.5−7.5 min) by running time.

thion, and isoprothidane, had r2 < 0.99 in the cucumber matrix solution. In Chinese cabbage (see Table S2 of the Supporting Information), the limits of detection (LODs) for 211 of 238 pesticides were below 1 μg/kg, the LODs for 20 pesticides ranged between 1 and 2 μg/kg, and seven pesticides had LOD values above 2 μg/kg, including benthiavalicarb-isopropyl (2.15 μg/kg), fenpiclnoil (2.02 μg/kg), hexythiazox (2.92 μg/kg), malathion (2.02 μg/kg), orthosulfamuron (2.40 μg/kg), pyrazosulfuron-ethyl (2.00 μg/kg), and pyrazoxyfen (3.06 μg/

pesticides (238 pesticides), in either the methanol or the matrix, analyzed by ESI in positive or negative mode, exhibited r2 ≥ 0.99 and a linear range from 1 to 200 μg/kg. For three pesticides, bentazone, fluoroxypyr, and trichlorfon, r2 was less than 0.99 in both the methanol standard solution and the matrix solution. Seven pesticides, including boscalid, bromacil, cyazofamid, diafenthiuron, formothion, terbuthylazine, and triflumuron, showed r2 < 0.99 in the Chinese cabbage matrix solution. In cucumber, seven pesticides, including bromacil, clothianidin, diafenthiuron, ethoxyquin, fludioxonil, formo11451

dx.doi.org/10.1021/jf504570b | J. Agric. Food Chem. 2014, 62, 11449−11456

Journal of Agricultural and Food Chemistry

Article

Figure 3. Percentage of pesticides within each recovery range with four different cleanup procedures by LC−MS/MS for the (A) Chinese cabbage and (B) cucumber samples. The number of pesticides is depicted above the bars.

former study, in which adequate recoveries were reported with the use of this method.5,19−22 When PSA mini-cartridge column (method “D”) replaced PSA D-SPE, as observed, the performance of method “D” reached the highest recoveries in cucumber; there was 203 pesticide recoveries in range of 70− 120% (Table 2). In the case of Chinese cabbage, recoveries of method “D” were lower than those obtained in method “A”; 173 pesticide recoveries were in the range of 70−120% at a spiked level of 40 μg/kg. The effect of MWCNTs (methods “B” and “C”) for cleanup was investigated in target vegetables. 143 and 138 pesticide recoveries ranged from 70 to 120% at a spiked level of 40 μg/kg for Chinese cabbage, whereas 193 and 170 pesticide recoveries ranged from 70 to 120% at the same spiked level for cucumber. It meant that the lowest recoveries were achieved in method “C” (PSA sorbent was excluded) at a spiked level of 40 μg/kg in Chinese cabbage. Less acceptable recoveries were obtained using a PLS mini-cartridge column (method “E”); just 167 and 187 of the detected pesticide recoveries were in the acceptable range (70−120%) at a spiked level of 40 μg/kg in Chinese cabbage and cucumber, respectively. On the basis of the above results in the recovery test, it could be concluded that the PSA D-SPE was the most suitable for Chinese cabbage and the PSA-mini cartridge column was the most suitable for cucumber out of the five investigated methods. Accuracy and Reproducibility Using Spiked Samples. Both retention time and the presence of all ions in the correct ratio were necessary (see Table S1 of the Supporting Information)23 for identification. In this study, there was no interference observed; the retention time and ion ratio should correspond to those obtained from a standard; and it showed good stability of relative selected reaction monitoring (SRM) responses. From the recovery study, PSA D-SPE and PSA mini-cartridge column were selected as the best methods for matrix purification of Chinese cabbage and cucumber, respectively. The RSD results demonstrated that the methods were repeatable. In Chinese cabbage, 162 and 183 of the 238 pesticides in method “A” had recoveries in the range of 70− 120% at the spiked levels of 8 and 40 μg/kg; the number of RSDs less than or equal to 10% was 84 and 139 pesticides; and

kg). In the case of cucumber (see Table S3 of the Supporting Information), the LODs for 209 of 238 pesticides were below 1 μg/kg, the LODs for 17 pesticides ranged between 1 and 2 μg/ kg, and 12 pesticides had LOD values above 2 μg/kg, containing bifenazate (4.69 μg/kg), boscalid (2.15 μg/kg), carbaryl (2.36 μg/kg), famoxadone (2.02 μg/kg), fipronil (2.30 μg/kg), hexythiazox (2.75 μg/kg), MCPA (2.04 μg/kg), mecarbam (2.02 μg/kg), metconazole (3.34 μg/kg), pyrazosulfuron-ethyl (4.04 μg/kg), tebuconazole (3.76 μg/kg), and triflumuron (6.32 μg/kg). There were no significant differences among the estimated instrument LOD values, which were calculated from the results obtained with standard solutions prepared with the matrix extract and solvent. Matrix Effect. The matrix effect is very compounddependent, often probably because of co-eluting matrix components, which interact with the pesticide in the ionization step in the interface.6 Decreasing or increasing of the ion amount formed results in a corresponding positive or negative matrix effect, respectively. The matrix effects were determined for 238 pesticides in two vegetables using five different cleanup methods at two concentration levels (8 and 40 μg/kg). The observed matrix effect values are presented in Table S1 of the Supporting Information. It could be concluded that the mean values of the matrix effect for 238 pesticides were 14.89 and 20.5% in Chinese cabbage and cucumber, respectively, which were less than 25%. However, 92 and 42 pesticides of the matrix effect values were still greater than 25% in two vegetables; it meant that there was a significant influence on quantitative detection that could not be ignored. Therefore, matrix-matched calibration curves were used for decreasing the influence of the matrix. Optimization of the Purification Procedure. The recoveries and repeatability of the spiking experiments at the spiked levels of 8 and 40 μg/kg are listed in Tables S1 and S2 of the Supporting Information. The chromatograms are presented in Figures 1 and 2. Relevant histograms by five different cleanup steps with LC−MS/MS are shown in Figure 3. Considerably higher recoveries were obtained when PSA DSPE was applied in method “A”. Recoveries of 183 and 193 of pesticides were in the range of 70−120% at a spiked level of 40 μg/kg in Chinese cabbage and cucumber, respectively (Table 1). The results were consistent with those presented in a 11452

dx.doi.org/10.1021/jf504570b | J. Agric. Food Chem. 2014, 62, 11449−11456

10−20%

120%

70−120%