Residual levels and dietary intake risk assessment of 11 pesticides in apricots from different ecological planting regions in China

Chromatographic separation and mass spectrometric optimization

To obtain the best monitoring conditions for each compound, a 0.5 mg/L mixed standard solution of 11 pesticides was mixed with the mobile phase through a syringe pump and then injected into the mass spectrometer for tuning. The precursor ion of the compound to be tested was determined by the primary mass spectrometry scan under ESI⁺ and ESI^– modes, and then the product ion was scanned by the secondary mass spectrometry. Two groups of ion pairs with the best sensitivity were selected for detection; one group was used for quantification, and another, for qualitative analysis. The optimization results showed high sensitivity of all the 11 pesticides under the ESI⁺ mode. Among them, abamectin (B1a), β-cypermethrin, deltamethrin, fenpropathrin, and bifenthrin were [M + NH₄]⁺, and other compounds were [M + H]⁺. MS parameters of 11 pesticides are mentioned in Table S2.

Formic acid and ammonium acetate are commonly used reagents to enhance the ionization of target compounds [M + H]⁺ and [M + NH₄]⁺ under the ESI⁺ mode, and they can effectively improve the peak pattern, making the peak sharper and more symmetrical; therefore, they need to be added during gradient elution³⁸. To improve work efficiency, it is necessary to separate and complete the monitoring of 11 pesticides in the shortest possible time; therefore, we selected two different types of chromatographic columns (ACQUITY UPLC HSS C₁₈ and ACQUITY UPLC HSS T3) and three different mobile phases (Ι: 0.1% formic acid aqueous solution—ACN, II: 0.05% formic acid aqueous solution—ACN, and III: 0.1% formic acid/5 mmol/L ammonium acetate aqueous solution—ACN) for optimization experiments. We observed that when using the HSS T3 chromatographic column, β-cypermethrin, deltamethrin, fenpropathrin, and bifenthrin did not show a good retention effect under the three mobile phase systems, and there was substantial tailing of the chromatographic peak. The shape of the chromatographic peak and sensitivity of the target compound were used as evaluation indicators. Compared with Ι and II, mobile phase III produced better sensitivity for all target compounds (Fig. 1), with sharper and more symmetrical peaks of β-cypermethrin, deltamethrin, fenpropathrin, and bifenthrin. This may be because the addition of 5 mmol/L ammonium acetate improved the retention performance of the HSS C₁₈ chromatography columns without affecting the ionization efficiency of all target compounds. In summary, we selected the HSS C₁₈ column for chromatographic separation and used 0.1% formic acid/5 mmol/L ammonium acetate aqueous solution—ACN as the mobile phase to further optimize the gradient elution procedure and effectively separate and detect all the target compounds within 8 min.

Optimization of purification materials

The flesh of apricot contains sugar, protein, calcium, phosphorus, carotene, thiamine, riboflavin, niacin, and vitamin C. Due to these diverse impurities, the analysis of the sample matrix becomes highly complex. Therefore, these impurities need to be removed from the matrix samples before analysis. Currently, PSA, C₁₈, and MWCNTs are widely used to adsorb to the fruit substrate³⁹. PSA has a strong adsorption capacity for metal ions, fatty acids, sugars, and fat-soluble pigments, C₁₈ has a strong adsorption capacity for non-polar impurities (such as fat, sterol, and volatile oil), while MWCNTs have a strong adsorption capacity for pigments, which can effectively remove chlorophyll, lutein, and carotene. However, C₁₈ and MWCNTs can also simultaneously adsorb pesticides, resulting in poor recovery. Nano-ZrO₂ has a large specific surface area and good adsorption stability and has recently been used to purify substrates. It can selectively remove fats and pigments from samples compared to conventional C₁₈ fillers.

In the current study, different purification materials were combined for the analysis of 11 pesticide residues and to propose the best purification strategy in the pretreatment of apricot samples. As displayed in Fig. 2, the average recovery of 11 pesticides in the apricot was higher using the C₁₈/nano-ZrO₂/MWCNTs than other combinations. Nano-ZrO₂ showed better adsorption than PSA in purifying fatty acids, organic acids, polar pigments, and sugars in apricot, owing to its larger specific surface area, better adsorption capacity, and stability. To conclude, the combination of 10 mg C₁₈, 30 mg nano-ZrO₂, and 5 mg MWCNTs demonstrated the best recovery for 11 pesticides, with recovery in the range of 72% to 114%, at a pesticide spiking level of 0.01 mg/kg. In summary, we finally determined that among the tested combinations, C₁₈/nano-ZrO₂/MWCNTs (10 mg/ 30 mg/5 mg) is the best purification combination for the pre-treatment of apricot samples.

Linearity, matrix effects, limit of detection and limit of quantification

The standard curve obtained from the standard working solutions of 11 pesticides and the calibration curve from blank apricot matrix spiked with 11 pesticides showed good linearity (0.001, 0.005, 0.01, 0.05, 0.1, and 0.5 mg/L), with R² ≥ 0.9959 for all tested samples (Table 1).

To evaluate MEs, the slopes of matching 11 pesticide standards with solvent and apricot matrix were calculated at the same concentration. According to the derived slope of the matrix-matched calibration curve, MEs of 11 pesticides in apricot were between 89 and 113% (Table 1), well within the range of 80% to 120%, indicating that the MEs could be ignored. It also suggests that the current pre-treatment method has a good purification effect and eliminates the matrix effect very well, laying a robust foundation for the subsequent step of quantitative analysis of samples. We next used the standard solution curve to quantify the 11 pesticide residues in apricot.

The LOD refers to the minimum concentration or minimum amount of a component to be tested that can be detected from a test sample under a given confidence level by an analytical method. Its physical meaning is the amount of the measured component when the signal is 3 times the standard deviation (S = 3σ) of the reagent blank signal (background signal). Sometimes it also refers to the amount of the measured component corresponding to when the signal is three times the background signal generated by the reagent blank (S = 3 N). The LOQ refers to the minimum amount of the analyte in the sample that can be quantitatively determined, and the determination result should have a certain accuracy⁴⁰. The LOQ reflects whether the analytical method has the sensitive quantitative detection ability. The LOQ is the lowest validated level with sufficient recovery and precision, which was estimated to be 0.001 mg/L, while the LOD is the lowest calibration level, which was 2 µg/kg, according to SANTE/12,682/2020.

Accuracy and precision

In the matrix, 11 pesticides were spiked at four levels (0.002, 0.02, 0.1, and 1 mg/kg), and for each spiked sample, there were six replicates. The recoveries of 11 pesticides in apricot at all levels ranged between 72 and 119%. The inter- and intra-level relative standard deviations (RSDs, %) of 11 pesticides in apricot were < 11.9% (Fig. 3), suggesting that the method was reliable within reproducibility in the laboratory according to NY/T 788—2018³³ and SANTE/12,682/2020³⁴.

Sample stability

The stability of samples during pre-treatment and UPLC-MS/MS was investigated by spiking the tested apricot samples with 11 pesticides at two concentrations. Under two storage conditions (room temperature for 8 h and at − 20 °C for 30 d), at the end of the test period, the recoveries of 11 pesticides at the two levels were 82%—117% (short-term) and 73%–103% (long-term), respectively (Table 2). Thus, the 11 pesticides showed good stability in the apricot matrix, making them suitable for subsequent analyses.

Terminal residues of 11 pesticides in apricots

The terminal residues of 11 pesticides in apricot samples collected from seven provinces or cities of China are listed in Table 3. Using the proposed method, four of the 11 pesticides, including acetamiprid, bifenthrin, fenpropathrin, and imidacloprid, were detected in apricot samples. The pesticide residues were detected in the samples from all sampling sites (Shandong, Xinjiang, Hebei, Gansu, Shanxi, Henan, and Beijing), with levels lower than the MRLs specified in China⁷. The research of Li et al.⁸ showed that imidacloprid was detected in Xinjiang apricots and below the MRLs⁷, which was consistent with our study. Therefore, imidacloprid is the high-frequency detection pesticide in apricot, which should be paid enough attention. Furthermore, only 41 out of the total 210 samples contained pesticide residues, with a detection rate of 19.5%. Therefore, to ensure food safety, the local government should strengthen monitoring and guidance of these detected pesticides specifically in those provinces or cities.

Dietary risk assessment

To assess the possible exposure routes and levels of the pesticides, dietary exposure was conducted to clarify the actual/expected exposure and potential harm to sensitive groups⁴¹. For the dietary risk assessment, six groups (male and female aged 2–4 years, 18–30 years, and 60–70 years) were selected considering significant differences in the ratio of food intake to body weight among people of different genders and ages⁴². Considering the dietary needs of people and following the principle of maximizing risk, the intake of apricot was calculated based on the fruit intake by Chinese people (six groups). The average daily intake of apricots in the different groups is mentioned in Table 4. The NEDI and RQ were calculated according to the dietary intake and weight survey data combined with the pesticide residues (four pesticides in total) detected in apricots in our study. The estimated NEDI values of the four pesticides detected in apricots were in the range of 1.0 × 10⁻⁴ mg/d^.bw to 25.2 × 10⁻⁴ mg/d^.bw and the RQs of the four pesticides in apricot were 0.003–1.184% for Chinese people (Table 5). The sums of the RQs% of the four pesticides in apricot for Chinese people of different age groups and genders were 1.469% (2–4 yrs, male), 1.626% (2–4 yrs, female), 0.127% (18–30 yrs, male), 0.168% (18–30 yrs, female), 0.128% (60–70 yrs, male), and 0.162% (60–70 yrs, female). Thus, the RQ values were less than 100% and indicated an acceptable level of the four pesticides detected in apricots. Concurrently, in terms of gender, we found a higher risk of dietary exposure in women than in men; with increase in age, dietary exposure was observed to gradually decrease, while children (2–4 years old) had the highest dietary exposure. Meanwhile, the evaluation results showed that value of RQ of bifenthrin was highest, followed by acetamiprid, fenpropathrin and imidacloprid. Therefore, regulatory departments of government should strengthen the monitoring, supervision and regulation of bifenthrin to prevent the occurrence of events harmful to dietary health. However, the dietary risk assessment results of apricot samples collected from seven provinces or cities of China in this study were obtained from the total RQ values of detected pesticides. Obviously, this assessment method is still insufficient. In the future, we will study and develop a scientific model of dietary risk assessment of multiple pesticides that includes multiple factors.

Source: Ecology - nature.com