Current Issue

All pesticide standards were of high purity (>90%) and were purchased from Dr. Ehrenstofer (Augsburg, Germany) and Sigma-Aldrich (Buchs, Switzerland). HPLC grade ethyl acetate, methanol, acetone, and n-hexane were purchased from Merck Inc. (Darmstadt, Germany). Anhydrous magnesium sulfate (MgSO₄), sodium chloride, and octadecylsilane (C₁₈) were purchased from Sigma-Aldrich and Waters (Milford, MA, USA), respectively. A filter of 0.22- μm polytetrafluoroethylene (PTFE) was acquired from Teknokroma (Barcelona, Spain).

The stock solution of individual analyte (approximately 1000 μg/mL) was prepared in a 50-mL volumetric flask using acetonitrile, methanol, acetone, and n-hexane as solvents. For working standard mixtures, a range of final target concentrations was prepared in acetone from the above stock solution by serial dilution. All stock solutions were stored at -20°C in amber glass bottles to prevent photolysis.

Aquatic animal product samples were purchased from local markets in Korea. The de-skinned fillets (over 500 g) were homogenized and then stored at -20°C. The blank samples were tested to ensure that it did not contain any of the target pesticides before use as a negative control. The aquatic animal samples (over 500 g) were prepared for analysis using matrix-matched calibration and monitoring. The homogenized samples (2 g) of aquatic animal samples were transferred into a 50 mL centrifuge tube. Thereafter, 10 mL of ethyl acetate was added to each sample, shaken vigorously by hand for 30 s. This was followed by the addition of 500 mg of NaCl and 1g of anhydrous MgSO₄ to each sample, which was then vortexed for 5 min. After vortexing, the extracts were put into a freezer at -20°C for 15 min and then centrifuged at 4500 × g at a temperature of 4°C for 10 min. The supernatant was transferred into a 50 mL centrifuge tube. The organic phase was evaporated under nitrogen stream at 50°C and diluted in 10 mL of acetonitrile, after which C₁₈ (200 mg), PSA (200 mg), and anhydrous magnesium sulfate (500 mg) were added. The mixture was shaken for 5 min and centrifuged at 4500 × g, 4°C for 10 min. The supernatant was trans- ferred into a 15-mL centrifuge tube and evaporated using nitrogen stream at 50°C and reconstituted with 1mL of 20% acetone in hexane. Finally, the extracts were filtered through a 0.22-μm PTFE syringe filter. The final extracts (5 μL) were injected into the GC-MS/MS system for further analysis.

An Agilent 7890 GC system coupled with an Agilent 7010 GC/MS Triple Quadrupole (Agilent Technologies, Santa Clara, CA, USA) and a Rxi^®-5Sil MS (0.25mm i.d. × 30 m, 0.50 μm film thickness) capillary column was used for the GC-MS/MS analysis. Electron impact ionization (EI) mass spectra was obtained at 70 eV and monitored from 100 to 600 m/z for full scan mode analysis. The working parameters were as follows: injector temperature was set at 280°C and the carrier gas (He) at 1.0 mL/min. The optimized GC oven temperature was initially 70°C (held for 3 mins), increased to 180°C at a rate of 20°C/min, and then finally to 300°C at 5°C/min (held for 7 mins). The mass selective detector transfer line was set at 280°C and the ion source at 230°C. The injection mode was splitless, and the injection volume was 1μ L. Data collection was performed in the multiple reaction monitoring (MRM) mode, and the optimized MRM parameters are listed in Table 1.

The method was validated according to the Codex guideline (CAC/GL 71, 2009). The blank samples (flatfish, eel, shrimp, and Manila clam) were tested to ensure that they did not contain any interferences and/or target compounds. The measured parameters were the linearity, limits of detection (LOD), limits of quantification (LOQ), accuracy, and precision. The validation study was carried out using tissue samples previously checked to be free of residual pesticides. The LOD was calculated at a signal-to-noise ratio (S/N) of 3, whereas the LOQ value was calculated using an S/N ratio of 10. The linearity was tested using matrix-matched calibrations (blank, 10, 20, 50, 100, 150 μg/kg) that were prepared by adding the appropriate amount (200 μL) of standard mixtures in the solvent into the fish and shrimp samples. The accuracy and precision (expressed as recovery and relative standard deviation, respectively) were determined by analyzing all samples spiked at 10, 20, and 100 μg/kg. The accuracy and precision were validated based on three target concentrations (10, 20, and 100 μg/kg). The accuracy and precision were determined at the three levels in the blank samples in five replicate analyses.

GC-MS/MS is a valuable approach for the determination of highly hydrophobic and volatile organochlorine pesticides (Hernández et al., 2013; Chatterjee et al., 2016; FSIS, 2018). In the current study, GC-amenable pesticides (organophosphorus, pyrethroids, carbamates insecticides, herbicides, and fungicides) were selected based on their potential use and contamination in fishery products and the aquaculture industry. GC-MS/MS based analytical methods have been preferred for the determination of pesticide residues in fish due to their high sensitivity and selectivity with low interferences (Munaretto et al., 2013; Sapozhnikova & Lehotay, 2013; Manuelmolina-Ruiz et al., 2014; Sahu & Nelapati, 2018; Colazzo et al., 2019).

The precursor ions, product ions, and collision energies were optimized for the best intensity of target compounds (Table 1). Based on a full scan spectrum, precursor ions were selected; then, the collision energy was adjusted to generate the product ions. MRM transitions with the highest intensities with related collision energies as well as retention times for all the pesticides were selected for quantification. The most abundant precursor ion with the highest m/z value was designated as the quantification ion, whereas the least intense product ion was designated as the qualifier ion. Due to the co-eluting sample interfering with the analytes, two precursor or additional product ions were used as qualifiers to prevent possible false-positives.

The QuEChERS (quick, easy, cheap, effective, rugged, and safe) approach was applied to this method because of its versatility (de Oliveira et al., 2019). The analytical method was developed and validated using GC-MS/MS based on QuEChERS. The optimization of purification was carried out using a salting-out solvent extraction step and a d-SPE clean-up step to remove matrix components (e.g., fatty acid). For the extraction step, salts that are easily electrolyzed in an aqueous solution were used as reagents to achieve separation of the ethyl acetate of nonpolar pesticides in an organic solvent (Sapozhnikova, 2014; Cao et al., 2015; FSIS, 2018). For the purification step, MgSO₄, PSA, and C₁₈ were used. MgSO₄ was used for moisture removal (Perović et al., 2018). PSA provided polar adsorption and weak anion exchange, which removed polar compounds such as organic acids, fatty acids, carbohydrates, and sugars, whereas the C₁₈ hydrocarbon chains eliminated fatty acids and nonpolar interfering substances (Sapozhnikova & Lehotay, 2013; Shin et al., 2018; Kim et al., 2020). Based on previous studies, the combination of MgSO₄ (500mg), C₁₈ (200 mg), and PSA (200 mg) was adopted for multi-pesticide detection in fishery products.

Specificity was evaluated through the analysis of the four different fishery product samples against a reagent blank. No interference was observed at the same retention time as the analyte. The validation process was performed by determining the linearity, LOD, LOQ, accuracy, and precision based on the CODEX guidelines (CODEX,2014). The chromatograms of the target compounds are shown in Figure 1. The linearity (expressed as correlation coefficients, r²) of the matrix calibration curves was >0.98 for all target compounds. Our results showed good linearity and allowed for the coverage of all target compounds. The LOD a nd LOQ w ere ≤3 a nd ≤ 1 0μg/kg, respectively. T he accuracy (expressed as recovery, %) and precision (expressed as RSD, %) of the target compounds were evaluated in spiked blank samples at three concentrations (10, 20, and 100 μg/kg). The overall recoveries for all the target compounds ranged from 62.4% to 120%. The precision was observed at 20.7% (Table 2). Three compounds (i.e., chlorothalonil, iprodione, and terbufos) were excluded before the start of method validation because of their inconsistent recoveries and/or unsatisfactory linearity of the calibrations. Some pesticides cannot be appropriately assessed using the buffered QuEChERS method (Lehotay et al., 2005; Cho et al., 2016).

Fig. 1. 
GC-MS/MS Chromatogram of a spiked sample with 51 pesticides at 10 μg/kg.

The applicability of the method was evaluated through the analysis of the target pesticides in 79 fishery product samples purchased from the local markets in Korea. Trifluralin was detected in one sample (1%) at a concentration of 67μg/kg in the Manila clam, while its residue in flatfish was below LOQ. Trifluralin is frequently detected in aquatic animal samples. The residue of trifluralin was reported to be above 1μg /kg in shrimp produced in Asian countries (Chan et al., 2012). Trifluralin residues (35−204 μg/kg) were detected in 11 pangasius fillet imported from Vietnam in 2011 (Chan et al., 2013). Previous studies have revealed that the trifluralin residues in Manila clam and flatfish (<LOQ) indicated the presence of pesticide runoff into the aquaculture environment (Shin et al., 2011). Trifluralin has been reported to mostly appear in runoff water in agricultural fields (Antoniouds, 2012). Furthermore, trifluralin residues in shrimp are associated with its use in the control of fungi and parasites in aquaculture farms and the surrounding environment. Further studies are needed to more clearly interpret the pesticide residues found in aquatic animal species.

The aquaculture industry has been overwhelmed by a wide range of parasitic and bacterial diseases affecting cultured species (Bondad-Reantaso et al., 2005). In order to prevent or treat these diseases, several chemicals have been used in high-density aquatic farms (Kang et al., 2018). Moreover, non-compliant samples in farmed aquatic animals are increasing due to the unintended and overuse of chemical compounds (Park et al., 2020). Further investigations are required to assess the dietary exposure to ethoxyquin residues and their health risks associated with the dietary intake of the farmed aquatic animals (Choi et al., 2020).