Development and validation of a modified QuEChERS method coupled with LC-MS/MS for simultaneous determination of difenoconazole, dimethoate, pymetrozine, and chlorantraniliprole in brinjal collected from fields and markets places to assess human health risk

Abstract

An effective and sensitive analytical method was developed to quantify the most common pesticide residues (difenoconazole, dimethoate, pymetrozine, and chlorantraniliprole) used for brinjal cultivation in Bangladesh. The quantification of the analytes was done using liquid chromatography-tandem mass spectrometry (LC-MS/MS). The samples were extracted using a modified QuEChERS method and followed by purification with dispersive solid phase extraction (d-SPE) sorbents (PSA, GCB, and C18). Matrix-matched calibration with a regression coefficient R² ≥ 0.9964 were used to minimize the brinjal matrix effect. The method was validated in quintuple (n = 5) at five different spiked levels (8–400 μg/kg) having recoveries in the range of 70.3–113.2% with relative standard deviations RSDs ≤6.8%, limits of detection (LOD) and limits of quantification (LOQ) was in the range of 0.15–0.66 μg/kg and 0.4–2.0 μg/kg, respectively, for the four analytes. A total 100 samples (50 samples directly from fields of Jessore district, Bangladesh and 50 samples from local market of Dhaka, Bangladesh) were collected to analyse the pesticides residue. The result showed that pesticides residue was found in both the field and market collected samples, 54% and 38%, respectively. The overall mean residue levels of four pesticides in field samples were significantly higher than those of market samples. Moreover, 20% of the field samples and 10% of the market samples had dimethoate residues, which were the most abundant among the four analytes and it ranged from 0.017 to 0.252 mg/kg. In terms of health risk assessments, dimethoate showed the highest estimated daily intake (EDI) and hazard quotient (HQ) values that are 3.02 × 10⁻⁵ mg/kg/day and 1.51%, respectively, in field samples. Till now, there have been no regulations or guidelines for the maximum admissible pesticide residue in Bangladesh. Therefore, the above findings will be an initial step for the regulatory authorities of Bangladesh to implement regulations and guidelines for pesticide usage.

Graphical abstract

Highlights

• •Simultaneous analysis of four pesticides residue in brinjal by LC-MS/MS.
• •Development of modified QuEChERS method with optimized d-SPE cleanup.
• •Pesticides residue was in both the field (54%) and market (38%) samples.
• •The results showed that there is no significant human health risk.

1. Introduction

Brijal is a very common crop that belongs to the solanaceae family. A total of 5,87,000 metric tons of brinjal production on 1,32,000 acres of land area during the summer and winter seasons in 2021 made this crop the second-highest grown vegetable in Bangladesh [1]. It is rich in total water-soluble sugars, free-reducing sugars, and proteins, among other nutrients [2]. It is also a good source of fiber, vitamin B6, and thiamine essential for a healthy life. In addition, brinjal is enriched with antioxidants that stimulate the elimination of free radicals and unstable molecules in our body [3].

Brinjal suffers from a number of diseases caused by fungi, bacteria, nematode and phytoplasma during cultivation [4]. To control disease as well as pests during production and storage, it is urgent to use pesticides [5,6]. Moreover, to prevent food from rotting, pesticides are also used before placing it on the market [7]. Therefore, the use of pesticides in agriculture has brought a wide range of benefits. However, excessive and inappropriate use of these pesticides can result in the high residue which exceed prescribed maximum residue limits (MRL) [[8], [9], [10]]. Sometimes, pesticides are used as overdoses and adulterated formulas without following the prescribed pre-harvest interval (PHI) [11]. Moreover, inadequate knowledge of farmers about good agriculture practices (GAP) is another reason for overdose and several pesticide application which results in a large amount of pesticide residues in crops.

The toxicity of pesticides may become severe depending on the type, amount, and route of exposure [12]. But chronic toxicity is caused by long-term pesticide residue intake in food or water. It may cause congenital disabilities, genotoxicity, endocrine disruption, neurodegenerative disorders, and various types of cancer. Children and pregnant women are the most vulnerable to these toxic effects [13]. As a result, it has become a great headache for analytical researchers to determine residues in crops with an effective and reliable method.

Brinjal is sprayed with more than 16 common pesticides over 40 times during the whole cultivation period in Bangladesh [14]. From these, difenoconazole (DFN), dimethoate (DMT), pymetrozine (PYM), and chlorantraniliprole (CLP) are four of the most frequently used pesticides (Fig. 1). Among these, DFN is a fungicide that is used to control the fungal diseases on fruits, vegetables, and other field crops. It is a 1,2,4-triazole fungicide that acts by interfering with ergosterol biosynthesis [15]. DMT is a systemic organophosphorus insecticide. It is generally used on field grown crops like leafy greens and tree-crop vegetables [16]. PYM, an insecticide with a remarkable selectivity for plant-sucking insects, such as aphids, whiteflies and plant hoppers, due to its systemic action [17]. Vegetables, brinjals, potatoes, and some other crops are the main field of application against insects [18]. It has tremendous effects on resistance management and good control of piercing and sucking mouthpart pests [19]. CLP is an anthranilic diamide insecticide that acts as a ryanodine receptor activator. This insecticide interrupts the normal muscle contraction in insects that lead to their death. It is highly active against Hemipteran, Thysanopteran, and Lepidopteran insects. CLP shows selectivity toward nontarget organisms as well as target pest species. Therefore, it is recommended as an alternative to pyrethroids for vegetables [20].

Fig. 1.

Chemical structures of dimethoate, difenoconazole, chlorantraniliprole and pymetrozine.

In pesticide residue analysis, HPLC/GC with several traditional detectors (HPLC-UVD/DAD/PDA, GC-ECD/FTD/FID/FPD), etc. are generally used due to their low installation cost and availability [[21], [22], [23]]. But their sensitivity, as well as selectivity, is comparatively low. Moreover, operating cost of these instruments is high and may end up with false positive or negative results [24]. On the contrary, the high detectability and selectivity of mass spectrometry gives better results in multi-residue analysis, allowing the determination of trace amount of pesticide residue [25,26]. Liquid chromatography coupled with tandem mass spectrometry (LC–MS/MS) has become more popular due to its effective separation, identification, and quantification of the pesticide residue [27,28]. Based on this, LC–MS/MS with electrospray ionization and triple quadruple detection in the multiple reaction monitoring (MRM) modes was used.

Several extraction procedures have been proposed to extract pesticides from different matrices, such as liquid–liquid extraction (LLE), solid-phase extraction (SPE), dispersive solid-phase extraction (d-SPE), magnetic solid phase extraction (MSPE), solid phase micro-extraction (SPME), and supercritical fluid extraction (SFE), [23]. Although LLE and SPE are very common for extraction, LLE needs more solvent, which is costly and time-consuming and in SPE, SPE cartridge is very expensive. Most laboratories cannot afford these techniques [29,30]. To overcome the limitation, a quick, easy, cheap, effective, rugged, and safe (QuEChERS) method is a better choice for extracting pesticides from different matrices [[31], [32], [33], [34]].

Various modifications of the QuEChERS have been developed to analyse pesticide residues in different matrices [35]. Usually, modifications are considered to get better recoveries and minimize the matrix effect [[36], [37], [38]]. In QuEChERS, extraction suffers from various matrices that are very difficult to clean up. However, using different d-SPE sorbents can be beneficial for cleanup with better recovery and less matrix effect [[39], [40], [41]]. Therefore, this study does not only limit QuEChERS extraction but also pays enough attention to optimize the d-SPE cleanup for maximum recoveries and minimum matrix effect.

In this study, a simple, rapid, and effective protocol using a modified QuEChERS method followed by optimized d-SPE (PSA, GCB, and C18) cleanup coupled with Liquid chromatography-tandem mass spectrometry (LC–MS/MS) was developed for the simultaneous determination of DFN, DMT, PYM, and CLP. The optimized method was assayed for all the target analytes in 100 real samples (50 samples directly from the field in Jessore, Bangladesh and 50 samples from the local market in Dhaka, Bangladesh) to evaluate its applicability. Finally, the acquired data was used to assess human health risk.

2. Experimental

2.1. Reagents and chemicals

All required pesticide standards, difenoconazole (purity ≥95.5%, CAS: 119446-68-3), dimethoate (purity ≥98.0%, CAS: 60-51-5), pymetrozine (purity≥98.0%, CAS: 123312-89-0), and chlorantraniliprole (purity ≥96.6%, CAS: 500008-45-7) were supplied by Sigma-Aldrich (St Louis, MO, USA). LC-MS grade acetonitrile (ACN) (CAS: 75-05-8), analytical-grade sodium chloride (NaCl) (≥99.9%, CAS: 7647-14-5), anhydrous magnesium sulphate (MgSO4) (≥99.0%, CAS: 7487-88-9), ammonium acetate (NH₄OAc) (≥99.0%, CAS: 631-61-8), and formic acid (FA) (99.98%, CAS: 64-18-6), were purchased from AppliChem GmbH, Ottoweg, D-64291 Darmstadt, Germany. The d-SPE sorbents, primary secondary amine (PSA) (CAS: 1318259-33-4), graphitized carbon black (GCB) (CAS: 7440-44-0), and octadecylsilane (C18) (CAS: 71889-02-6), were collected from CNW Technologies, ANPEL Laboratory Techonlogies (Shanghai), China. Ultrapure deionized (DI) water (18 MΩ cm) was used to prepare the mobile phase each time.

2.2. Instrumentation

LC-MS/MS was used to carry out chromatographic analysis with an Agilent 1290 LC system (1290 Infinity II) coupled to a triple quadruple mass spectrometer (6420LC/TQ). The chromatographic separation was carried out through a ZORBAX RRHD Eclipse plus C18 column (2.1 × 100 mm, 1.8 μm particle size). The column oven is maintained at 30 °C. The solvent system consisted of 0.1% formic acid +10 mmol/L ammonium acetate in water (A): ACN (B): The mobile phase started at 50% B (0–3.1 min), increased to 90% B (3.1–7.1 min), and then again decreased to 50% B (7.1–10.0 min), the flow rate was 0.45 mL/min, with a total run time of 10 min and the injection volume was 5 μL. The retention times of PYM, DMT, DFN, and CLP were at 1.28, 1.89, 5.32 and 8.40 min, respectively. Positive electrospray ionization (ESI+) interface in multiple reaction monitoring (MRM) modes was conducted. Two mass transitions were used for all the target analytes where precursor ions yielded as [M + H⁺]. For quantitation, the most intense peak of the two mass transitions was used, while another was used for confirmation. The dwell voltage was 200 eV, and the fragmentation energy was 135 eV for all the products ions but the collision energies were different. The instrumental conditions are listed in Table 1.

Table 1.

Fragmentation pattern of four pesticides in MRM mode.

Pesticides name	Precursor ion, (M + H+)	Product ions	Collision energy, eV	Dwell energy, eV	Fragmentation energy, eV
PYM	218.1	105a	20	200	135
		79b	20
DMT	230	125a	29
		199b	13
DFN	406	251a	30
		337b	30
CLP	484.2	453.1a	11
		286.1b	11

^aQuantifier.

^bQualifier.

2.3. Sample collection and preservation

A total of number of 100 samples were collected among them, 50 samples were collected from the field of Jessore district, and the other 50 samples were collected from different markets of Dhaka, Bangladesh (Fig. 2) during November–December 2021. Each sample weight around 1 kg was collected and kept in properly labeled (with unique code) zipper bag to protect from contamination. Then the samples were kept in an ice box for temporary preservation and immediately carried to the laboratory. Then all the samples were finely chopped, blended, and taken in properly labeled zipper bags to be store at −20 °C for further analysis.

Fig. 2.

Map showing the location of Jessore district and Dhaka city, Bangladesh from where the brinjal samples were collected.

2.4. Sample extraction and cleanup

Sample preparation was done using a modified QuEChERS method followed by optimized d-SPE cleanup. For the extraction procedure, 5 g of homogenized samples were taken in 50 mL Teflon tubes. After that, 10 mL of ACN was added to the sample, followed by 1 min handshaking. Then QuEChERS extraction salt (3 g of anhydrous MgSO₄ and 1 g NaCl was added to the sample and mixed vigorously using a vortex mixture (Model: AVM-2000-C, Brand: Digisystem, Origin: Taiwan) for 1 min. Then the tubes were centrifuged (Model: DSC-200A-2, Brand: Digisystem, Origin: Taiwan) at 4000 rpm for 5 min. After that 1.5 mL clear supernatant into a 2 mL micro-centrifuge tube previously filled with an optimum amount of d-SPE sorbents (50 mg PSA, 5 mg GCB and 3 mg C18). Then the micro-centrifuge tube was mixed with vortex vigorously for 1 min and centrifuged (Model: D3024, Brand: Scilogex, Origin: Germany) at 5000 rpm for 5 min. At last, the clean supernatant was transferred through a 0.24 μm filter into a 2 ml autosampler vial for LC-MS/MS analysis.

2.5. Preparation of standard in solvent

A 100 mg/L standard stock solution of DFN, DMT, PYM and CLP were prepared as exactly 10.0 mg of each standard was taken separately in a 100 mL volumetric flask (Brand: Pyrex). The volumetric flask was filled up to the mark with LC-MS grade ACN solvent. From 100 mg/L stock solution, 10 mg/L working solution was prepared in separate 100 mL volumetric flask (Brand: Pyrex) with the identical solvent. From the 10 mg/L working solution and intermediate working standard solution of four mixed standard of 2.0 mg/L with same solvent. Finally, from the 2.0 mg/L mixed standard solution, a serial dilution of 0.2, 0.1, 0.05, 0.02, 0.01, 0.005, 0.002, 0.001, 0.0005, 0.0002, and 0.0001 mg/L was prepared. All the standard stock solutions were kept at −4 °C for further analysis.

2.6. Preparation of standard in the sample matrix

From the 2.0 mg/L mixed standard solution, similar serial dilution of 0.2, 0.1, 0.05, 0.02, 0.01, 0.005, 0.002, 0.001, 0.0005, 0.0002, and 0.0001 mg/L were prepared by using brinjal extract instead of ACN solvent. All standard solutions were stored in a refrigerator at −4 °C -before analysis.

2.7. Health risk assessment

The estimated short-term intake (ESTI) was calculated to assess the health risk from the consumption of brinjal using the following Eq. (1) [42].

$$ESTI=P\times \frac{\mathrm{C}\mathrm{R}}{\mathrm{B}\mathrm{W}}$$ (1)

here P = highest residue level of pesticides in binjal samples (mg/kg), CR = 9.74 g/person/day; brinjal consumption rate in Bangladesh (BBS 2019) [43] and BW = 60 kg; average body weight (WHO 2015) [44].

The short-term health risk /acute health indices (aHI) were calculated using the following Eq. (2) [45].

$$aHI=ESTI/\mathrm{A}\mathrm{R}\mathrm{f}\mathrm{D}\times 100\%$$ (2)

here, ARfD is an acute reference dose.

The estimated daily intake (EDI) was calculated to evaluate the health risks using the following Eq. (3) [11].

$$EDI=M\times \frac{\mathrm{C}\mathrm{R}}{\mathrm{B}\mathrm{W}}$$ (3)

here, M = mean residue level of pesticide in binjal (mg/kg).

The hazard quotient (HQ) or long-term health risk was measured using the following Eq. (4) [46].

$$HQ=EDI/\mathrm{A}\mathrm{D}\mathrm{I}\times 100\%$$ (4)

here, ADI is accepted daily intake.

If HQ becomes 100%, that indicates a potential risk. The higher the HQ value, is higher the health risk [47]. For the above calculation, the ARfD and ADI values were obtained from the Joint FAO/WHO Meeting on Pesticide Residues (JMPR) database (http://apps.who.int/pesticide-residues-jmpr-database) (JMPR, 2004; FAO/WHO, 2019) (access date: March 10, 2022) for a 60 kg person for the targeted pesticides.

2.8. Method validation

Linearity, specificity, accuracy, precision, limit of detection (LOD), and quantification (LOQ) were all performed to validate the method. The correlation coefficient value (R²) of different point calibration curves was evaluated. The specificity was evaluated by observing any unwanted peak in the blank sample at the same retention time of the analyte.

The accuracy of the method was evaluated from the recovery (%) by spiking the analyte in the control sample, relative standard deviation, RSD (%) was calculated by spiking quintuples (n = 5) which indicates the precision. The analyte that gives a response, signal-to-noise ratio (S/N) of 3, which was expressed as the limit of detection (LOD), and the analyte gives a response signal-to-noise ratio (S/N) of 9 which was expressed as the limit of quantification (LOQ). The matrix effect (ME) was evaluated using the following Eq. (5) [48].

$$\mathrm{M}\mathrm{E}\%=\left(\frac{\mathrm{S}\mathrm{l}\mathrm{o}\mathrm{p}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{m}\mathrm{a}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{x}\mathrm{m}\mathrm{a}\mathrm{t}\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{d}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{s}\mathrm{c}\mathrm{u}\mathrm{r}\mathrm{v}\mathrm{e}-\mathrm{S}\mathrm{l}\mathrm{o}\mathrm{p}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{s}\mathrm{c}\mathrm{u}\mathrm{r}\mathrm{v}\mathrm{e}}{\mathrm{S}\mathrm{l}\mathrm{o}\mathrm{p}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{s}\mathrm{c}\mathrm{u}\mathrm{r}\mathrm{v}\mathrm{e}}\right)\times 100\%$$ (5)

The positive and negative values of the ME% indicate enhancement and suppression effects of the matrix respectively.

3. Optimization of QuEChERS

The matrix interferences of samples and the chemical properties of pesticides are mainly responsible for difficulties during the sample extraction procedure. ACN as an extraction solvent is the most suitable in liquid chromatography [23]. Therefore, we used ACN for the extraction of target analytes. However, the matrix content of brinjal includes sugars, vitamins, minerals, and amide proteins, due to their solubility in organic solvents, these also come out in the extraction solvent (ACN) as matrix. So the interfering matrix should be minimized through proper cleanup before analysis [49].

From the previous studies, we found that better recoveries can be achieved using MgSO4 only [50]. Still, water content in the ACN phase remains an interference, resulting in low selectivity. By separating ACN from the aqueous phase, NaCl promotes the partitioning of polar compounds into ACN [50]. For this reason, 3 g anhydrous MgSO4 along with 1 g NaCl is used to facilitate phase separation and partitioning of pesticides into ACN. As mentioned earlier, more sensitive LC-MS/MS analysis requires the highest cleanup of interfering matrices. This cleanup can be done by SPE (solid phase extraction) or d-SPE (dispersive solid phase extraction) sorbents. Even though cleanup with d-SPE sorbents needs more time than SPE, we used d-SPE to reduce the cost. Usually C18 or C8, florisil, graphitized carbon black (GCB), primary secondary amine (PSA), etc, are used in d-SPE cleanup to remove of lipids, polar and fattty co-extracts, pigments, and some fatty acids, saccharides, and organic acids from the extract [51]. For the removal of co-extract matrix PSA, GCB, and C18 were used by optimizing the amounts, as these sorbents can remove analytes in higher amounts. For optimization, we analyzed the recovery of pesticides using 20–70 mg PSA (20, 30, 40, 50, 60, and 70 mg), we got minimum of 95% recovery for all the analytes at 50 mg (Fig. 3a). Similarly for GCB, 1–10 mg (1, 2, 3, 5, 7, and 10 mg), we got minimum of 90% recovery for all the analytes at 5 mg (Fig. 3b). Lastly, we analyzed the recovery for C18, using 1–10 mg (1, 2, 3, 5, 7, and 10 mg), we got minimum of 95% recovery for all the analytes at 3 mg (Fig. 3c).

Fig. 3.

Recovery optimization of (a) PSA, (b) GCB, and (c) C18 amounts.

4. Results and discussion

4.1. Method performance

4.1.1. Specificity

The specificity of the method was estimated by the analysis of a blank brinjal sample in quintuples. Not any peak was found for the sample matrix at the retention time of the target analytes. Therefore, the current method of analysis is specific.

4.1.2. Linearity

To check the linearity, ten-point matrix-matched calibration was done for DFN using (0.2–200 μg/L) concentrations with an R² value of 0.9987. Similarly, for PYM, CLP nine-point matrix-matched calibration was done from (0.5–200 μg/L) concentrations with R² values of 0.9982 and 0.9995. Finally, eight-point matrix-matched calibration was carried out from (1.0–200 μg/L) concentrations with an R² value 0.9964 for DMT (Table 2).

Table 2.

Recoveries, LODs and LOQs for individual pesticides in brinjal samples.

Pesticides name	Linear range (μg/L) and linearity (R2)	Recovery at different spike levels		LOD (μg/kg)	LOQ (μg/kg)
		Spiked levels (n = 5) (μg/kg)	Recoveries [Mean ± RSD] %
PYM	0.5–200 (0.9982)	400	72.4 ± 0.5	0.33	1.0
		200	70.3 ± 4.7
		100	78.5 ± 2.5
		20	70.9 ± 5.2
		8	107.2 ± 6.3
DMT	1.0–200 (0.9964)	400	82.5 ± 5.3	0.66	2.0
		200	83.7 ± 6.8
		100	91.3 ± 3.8
		20	83.2 ± 3.1
		8	113.2 ± 6.1
DFN	0.2–200 (0.9987)	400	78.5 ± 6.7	0.15	0.4
		200	76.2 ± 4.6
		100	87.9 ± 2.3
		20	81.9 ± 4.6
		8	105.8 ± 6.0
CLP	0.5–200 (0.9995)	400	76.7 ± 2.2	0.33	1.0
		200	77.4 ± 4.8
		100	79.2 ± 5.2
		20	85.4 ± 5.5
		8	95.5 ± 3.6

4.1.3. Matrix effect

Three of the four pesticides (DMT, DFN, and CLP) showed matrix effects as matrix enhancements of ME% 7.70%, 12.60%, and 26.41% respectively. Hence, their matrix-matched calibration curves came out with better linearity in the matrix (R² = 0.9964, R² = 0.9987, and R² = 0.9995) than in solvent (R² = 0.9899, R² = 0.9909, and R² = 0.9527). In contrast, only one pesticide PYM was affected by the matrix suppression (ME% = −28.36%) and therefore, linearity for matrix-matched calibrations (R² = 0.9982) was a little less than solvent calibrations (R² = 0.9992). Finally, the matrix-matched calibrations were used to quantify all the analytes in this study (Table 2).

4.1.4. LOD and LOQ

The LODs of PYM, DMT, DFN, and CLP were 0.33, 0.66, 0.15, and 0.33 μg/kg respectively (Table 2). For both PYM and CLP the LOQs were set at 1.0 μg/kg, for DMT and DFN LOQs at 2.0 and 0.4 μg/kg (Table 2). the LODs we found for all target pesticides were enough low.

4.1.5. Recovery

The recovery studies were done for all the analytes at different spike levels of 8, 20, 100, 200, and 400 μg/kg. The recovery obtained at different spiking level, in the range of 70.3–113.2% for PYM, DMT, DFN, and CLP. The relative standard deviations (RSDs) was in the range of ±0.5 to ±6.8% (Table 2). Representative chromatogram has been in supplementary data. The recoveries studies are good enough according to SANTE guideline [52].

4.2. Pesticides residue in brinjal

To detect and quantify the residue levels of four selected pesticides (PYM, DMT, DFN, and CLP) from the most frequently used pesticides in brinjal cultivation practices, 100 samples were collected. An effective and sensitive analytical method was developed. To observe the differences in residual status between field and market samples, 50 samples were collected directly from the farmers at the farming field sides of Jessore, and 50 samples were collected from the local markets of Dhaka, Bangladesh.

10 (20%) samples collected from field were positive detected of DMT residues from 0.018 to 0.252 mg/kg and mean residue 0.186 mg/kg and all these residues were over the MRLs (0.01 mg/kg) (Table 3). Kaium et al. reported, similar result as DMT residue in three (6%) of the brinjal samples collected from markets of Dhaka, Bangladesh [11]. M. Prodhan et al. reported, three out of 72 samples collected from different market places in Thessaloniki, Greece, were detected as positive of DMT residues ranging from 0.01 to 0.022 mg/kg [53]. Moreover, 6 (12%) samples contained DFN residues and 4 (8%) samples contained PYM residues with a high amount over MRLs (0.01 mg/kg) but 7 (14%) samples contained CLP residues lower than MRL value (0.6 mg/kg) (Table 3).

Table 3.

Residue analysis data of four pesticides in brinjal samples.

vegetable	Sample source	Pesticide	No of sample		Detected samples, %	MRL (mg/kg)	No of samples > MRL (mg/kg)	Residual range	Mean residue
			Analyzed	Detected
Brinjal	Field	PYM	50	4	8	0.01	4	0.056–0.381	0.241
		DMT	50	10	20	0.01	10	0.018–0.252	0.186
		DFN	50	6	12	0.01	6	0.032–0.351	0.195
		CLP	50	7	14	0.6	0	0.024–0.256	0.156
	Market	PYM	50	2	4	0.01	2	0.019–0.254	0.137
		DMT	50	5	10	0.01	5	0.017–0.208	0.122
		DFN	50	4	8	0.01	4	0.025–0.266	0.107
		CLP	50	8	16	0.6	0	0.014–0.332	0.198

Out of 50 samples collected from different markets, although 8 (16%) samples were detected positive for CLP residues ranging from 0.014 to 0.332 mg/kg, residues were not above the MRL (0.6 mg/kg). On the contrary, 2 (4%) samples were detected positive for PYM, 5 (10%) samples showed positive results for DMT residues, and 4 (8%) samples were positively detected for DFN residues, and these three residues were over the corresponding MRLs too (Table 3).

4.3. Comparison of method with the previous studies

A list of previous studies is shown in Table 4 with LODs, recoveries, and residues of target pesticides using different instruments and extraction techniques in several matrices. Comparing the LODs of the present method with some close similar methods reported in previous literature, it was clear that, the LODs for all target pesticides in our new developed method is significantly lower than the previously reported methods. Therefore, the method to analyse pesticides was more sensitive compared to the previous studies.

Table 4.

Comparison of the LODs of the present method with similar methods reported in previous literature.

Sample matrix	Instruments	Extraction techniques	Pestisides	LODs	Recoveries	Residues	Ref.
Brinjal	LC-MS/MS	d-SPE (PSA, GCB, C18)	PYM, DMT, DFN and CLP	0.15–0.66 μg/kg	70.3–113.2%	0.014–0.381 mg/kg	This study
Hyacinth bean and eggplant	GC-FTD	d-SPE (PSA,Charcoal, MgSO4)	DMT			0.032–.217 mg/kg	[55]
Brinjal, cabbage, capsicum, cauliflower, okra, and tomato	HPLC-PDA		CLP	0.03 mg/kg	85–96%	–	[56]
Eggplant and cauliflower	GC-FTD	d-SPE (PSA, MgSO4)	DMT	0.01 mg/kg LOQ	80–109%	0.052–0.132 mg/kg	[42]
Country bean and bitter gourd	GC-FTD	d-SPE (PSA, MgSO4)	DMT			–	[11]
Eggplant	GC-FTD	d-SPE (PSA, MgSO4)	DMT			0.048–0.058 mg/kg.	[57]
Poitned gourd	GC-FTD	d-SPE (PSA, MgSO4)	DMT	0.002–0.003 mg/kg	85–106%	0.080–0.105 mg/kg	[45]
Country bean and yard long bean	GC-FTD	d-SPE (PSA, MgSO4)	DMT		85–110%	0.008–0.321 mg/kg	[58]
Strawberries	LC-MS/MS	d-SPE (PSA)	PYM		84.2–101.4%	<0.010–0.019 mg/kg	[19]

4.4. Risks assessment

The estimated health risks of pesticides in collected brinjal samples are represented in Table 5. In terms of field samples, acute or short-term risk (aHIs) analysis reveals that all ESTI values were much lower compared to the JMPR's ARfD values. Only dimethoate with aHI 0.20% showed the highest risk factor (Table .5). Since each of the aHIs was much lower than 100%, there was a negligible short-term risk with the brinjal consumption that are exposed to the analyzed pesticides. Moreover, the estimation of the chronic or long-term risk factors disclosed with the hazard quotient (HQs) is relatively higher than that of acute or short-term risk. The HQs values were higher for PYM, DMT, and DFN than those of CLP (Table 5).

Table 5.

The short-term and long-term risks due to the average daily intake (ADI) of pesticides through brinjal consumption. ARfD and ADI were adopted from the JMPR database.

source	Pesticide	Short term risk			Long term risk
		ESTI mg/kg/day	ARfD mg/kg/bw/day	aHI %	EDI mg/kg/day	ADI mg/kg/day	HQ %
Field	PYM	6.18E-05	0.1	0.06	3.91E-05	0.03	0.13
	DMT	4.09E-05	0.02	0.20	3.02E-05	0.002	1.51
	DFN	5.70E-05	0.3	0.02	3.17E-05	0.01	0.32
	CLP	4.16E-05	–	–	2.53E-05	2	0.00
Market	PYM	4.12E-05	0.1	0.04	2.22E-05	0.03	0.07
	DMT	3.38E-05	0.02	0.17	1.98E-05	0.002	0.99
	DFN	4.32E-05	0.3	0.01	1.74E-05	0.01	0.17
	CLP	5.39E-05	–	–	3.21E-05	2	0.00

In market samples, acute or short-term risk assessment gave every ESTI value a much lower than ARfD value, similar to field samples. Besides, the estimated aHI and HQ values were comparatively lower than the field ones (Table 5).

Overall, the comparisons suggest that the collected field samples are more contaminated than the local market samples. This may be rationalized by the fact that the market samples are washed several times through the procedure of collection from the field, transportation, and placing in the retail and local markets. Besides, it should be considered that the samples were analyzed, directly without washing. People in Bangladesh usually wash vegetables after buying them. Thus washing, peeling and cooking practices will reduce the amount of pesticide residues to a significant level [54]. Moreover, each of the HQs was much lower than 100%, indicating a negligible health risk. Hence, we can conclude that people who have no significant health risk from the brinjal consumption are exposed to the analyzed pesticides.

5. Conclusion

To determine the residue of PYM, DMT, DFN, and CLP pesticides in brinjal collected from field (50 samples) and market (50 samples). A modified QuEChERS method was developed including optimized dSPE clean-up followed by LC-MS/MS analysis. The developed analytical method was effective and sensitive with low LOD, LOQ and good recoveries (70.3–113.2%) at different spiking (8, 20, 100, 200, and 400 μg/kg) with RDS≤6.8%.Compared o market places samples field incurred samples contained little high residue. The residue data was used to assess the human health risk. We obtained a maximum hazard quotient of 1.51%, much lower than 100%. Our study result can be concluded the consumption of brinjal not much harmful depending on the limited (100) samples data. To the best of our knowledge, this is the first modified QuEChERS method with optimized d-SPE cleanup to compare field incurred and market brinjal samples.

Disclaimer

Our current study for the first time developed and validate a modified QuEChERS method coupled with LC-MS/MS for simultaneous determination of PYM, DMT, DFN and CLP in brinjal samples collected from field and market place to assess the human health risk. In this study, we considered only 100 brinjal samples (50 colleced from fields and 50 collected from markets of Dhaka) to evaluate the human health risk of detected pesticides. Moreover, we analyzed only four of the frequently used pesticides for brinjal cultivation using a limited number of samples. Therefore, further studies are essenial to determine the pesticides contamination levels in brinjal from both sources and to assess the health risk of the detected pesticides.

Author contribution statement

Tajnin Jahan: Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Sabina Yasmin: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Md Aftab Ali Shaikh: Contributed reagents, materials, analysis tools or data; Wrote the paper.

Md Jubayer Ibn Yousuf, Md. Saidul Islam: Performed the experiments; Analyzed and interpreted the data.

Md Tazul Islam Choudhury: Analyzed and interpreted the data.

Humayun Kabir: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Data availability statement

Data will be made available on request.

Declaration of interest's statement

The authors declare no conflict of interest.