Ionic liquid‐based dispersive liquid–liquid microextraction of anthelmintic drug residues in small‐stock meat followed by LC‐ESI‐MS/MS detection

Abstract An ionic liquid‐based dispersive liquid–liquid microextraction (IL‐DLLME) of 20 anthelmintic drugs followed and detected by liquid chromatography–tandem mass spectrometry (LC–MS/MS) has been developed, optimized, and validated. The parameters affecting the anthelmintic extraction efficiencies such as selection of extraction solvent (ionic liquids), selection of disperser solvent, volume of extraction solvent, volume of disperser solvent, pH of the aqueous phase, extraction time, salt addition, and centrifugation time were optimized. Validation was conducted according to ISO/IEC 17025:2017 and Commission Implementing Regulation (EU) 2021/808 of 22 March 2021. Validation parameters such as calibration function, matrix effect, limit of detection (LOD), limit of quantification (LOQ), decision limit (CCα), accuracy, and precision were established. Coefficient of determination (R 2) values ranging from .99938 to .99995 were obtained using the matrix calibration curve spiked at 0, 0.25, 1.0, 1.5, and 2.0 times MRL. The LODs and LOQs were calculated using the standard deviation of the response and the slopes of the calibration curves ranged from 0.35 to 26.1 μg/kg and from 1.2 to 87.0 μg/kg, respectively, and were dependent on calibration range. The CCα values ranged from 23 to 1022.0 μg/kg and are also dependent on the MRL concentration levels. The coefficient of variation (CV) values calculated are within the reproducibility range of 16%–30% adapted from the Horwitz Equation CV = 2(1–0.5 log C) and ranged from 1.7% to 16.9%. The developed and validated and the standard QuEChERS method were compared. The IL‐DLLME LC–MS/MS method was applied to 32 small stock (18 caprine [goat] and 14 ovine [sheep]) liver samples received from municipal abattoirs at Botswana National Veterinary Laboratory for the analysis of anthelmintic drug residues. The results obtained indicated that the anthelmintic drug residues were all below the detection capability, and therefore, the samples were passed as fit for human consumption.


| INTRODUC TI ON
Anthelmintic drugs are used to treat livestock and humans infected with parasitic worms or helminths (helminthiasis).These drugs are classified based on similar chemical structures or mode of action.
In addition to challenges of drug residue in meat, humans are also exposed to drug-resistant parasites.
The EU established protocols for testing and monitoring veterinary drugs which involved setting maximum residue levels (MRLs) to ensure that meat destined for local and export markets is safe for human consumption and public health.Anthelmintic drugs (allowed substances) and their maximum residue limits in foodstuff of animal origin are listed in annex of Commission Regulation (EU) No 37/2010 of 22 December 2009(Commission Regulation, 2010).The MRLs of the marker residues have been set in all ruminant livers for anthelmintic drug as listed in Table 1.

| Apparatus
A

| Preparation of stock and working standard solutions
Stock standard solutions (1 mg/mL) for individual compounds were prepared by dissolving each standard in DMSO/MeOH, (50/50) v/v (anthelmintic drug standards dissolve best DMSO and do not precipitate during storage.DMSO freezes fast and takes long time to thaw before use, methanol was added to reduce the thawing time).

| Isolation of anthelmintic drug residue from liver samples
Isolation of anthelmintic drug residues from liver samples was done by chopping up about 10 g of liver sample and homogenizing the chopped liver in a food processor.Thereafter, 2.0 ± 0.02 g of the previously homogenized liver sample was weighed into a 50-mL polypropylene centrifuge tube.Three blank samples were fortified with working standard solution at 1.0 MRL (equivalent to 200 μL of working standard) for quality controls.Five-point pre-extraction spiked matrix standards (PrEMS), fortified with working standard solution at spiking levels of 0.0, 0.5, 1.0, 1.5, and 2.0 MRL (equivalent to 0, 100, 200, 300, and 400 μL of working standard, respectively), were used for calibration.To all the samples, control samples and calibration samples, 100 μL of stable isotope labeled/internal standard was added.After fortification, the samples were allowed to stand for a minimum time of 30 min before extraction.Using a bottle-top dispenser, 8 mL of acetonitrile was then added to all the samples.
Thereafter, the samples were homogenized for about 20 s using a probe homogenizer and then mixed vigorously on a vortex mixer for 10 min.The tube was then centrifuged at 2795 g for 10 min, and the supernatant was collected and concentrated to about 100 μL in a stream of nitrogen in the water bath set at 50°C.The concentrated solution was diluted with 5 mL of water and filtered through 0.45μm PVDF syringe filters.The sample was then ready for the IL-DLLME extraction procedure.

| IL-DLLME procedure
The 5-mL sample extract was transferred into a 15-mL conical bottom centrifuge tube, and 0.4 mL of methanol containing 60 μL of was quickly injected into the sample extract and the tube was shaken immediately for 30 s and allowed to stand for another 40 s prior to centrifugation at 2795 g for 5 min.The fine droplets of the ionic liquid will settle at the bottom of the centrifuge tube.The upper aqueous phase is discarded, the remaining ionic liquid phase is diluted with 200 μL of DMSO, and 5 μL of the diluted ionic liquid phase is injected into the LC-MS/MS system (Wang et al., 2015).

| LC-MS/MS analysis
ExionLC™ Series ultra-high-performance liquid chromatographic system coupled with the QTRAP® 6500+ MS system from AB Sciex was used for LC-MS/MS analysis.LC-MS/MS system was controlled by Analyst® Software version 1.7 and the results were shown in Table 3.

| Optimization of the extraction method
In the extraction method using IL-DLLME, the main factors that af-

CHROMATOGRAM +VE MODE PRE EXTRACTION MATRIX STANDARD IL-DLLME
extraction effect.Figure 2 shows that [C6MIM][PF6] was found to be the best extraction solvent among the three ionic liquids evaluated and it was, therefore, used in all subsequent experiments.

| Selection of disperser solvent
A disperser solvent plays an essential role in the formation of a cloudy solution in the IL-DLLME extraction method.It is important that a suitable solvent is identified for the extraction process.The disperser solvent must be miscible with the extraction solvent (organic phase) and also with the sample solution (aqueous phase) to ensure the formation of a cloudy solution (Hwang et al., 2018).In All the disperser solvents produced a cloudy solution when injected into the aqueous phase but methanol was more cloudy as compared to others; however, the highest extraction recovery (>80%) of all the anthelmintic drugs was observed using methanol compared with the  other two solvents and this is probably due to its higher compatibility of methanol with aqueous solution than acetone and acetonitrile.
Methanol was, therefore, selected for all subsequent experiments; in addition, it was greener than acetonitrile.

| Determination of volume of [C6MIM][PF6]
For the determination of the optimum volume of the ionic liquid [C6MIM][PF6], the disperser solvent was kept constant at (0.3 mL) containing different volumes of [C6MIM][PF6] (30, 40, 50, 60, 70, 80, 90, and 100 μL).Results showed that the extraction recoveries of the anthelmintic drug residues increased when the volume of [C6MIM][PF6] was increased from 30 to 70 μL, and then declined from 80 μL with an increase in the volume of IL (Figure S3).When the volume of the ionic liquid is small, the extraction is not efficient, the ionic liquid was not enough to be dispersed in the aqueous phase.In addition, when the ionic liquid volume was high, more of the ionic liquid will sediment after centrifugation, this will result in the dilution of the analyte content in the sedimented ionic liquid.The extraction recovery with [C6MIM][PF6] at 60 μL was above 87% on average of all anthelmintic drugs.Therefore, a volume of 60 μL of [C6MIM][PF6] was selected for all subsequent experiments.

| Determination of the optimum volume of disperser solvent
The disperser volume was determined with the volume of [C6MIM] [PF6] maintained constant at 60 μL and varying the volume of the disperser solvent (methanol) at 0.2, 0.3, 0.5, 0.6, 0.8, and 1.0 mL.
When a small volume of disperser solvent was used, the cloudy solution formed was less tense, hence lower extraction recoveries.
The results showed that the best extraction recoveries (ERs) of all anthelmintic drug residues were found to be at volumes of around 0.3 and 0.4 mL and then declined with an increase in the volume of methanol from 0.5 to 1 mL.When a large volume of disperser solvent was used, the ionic liquid could be dissolved in the disperser solvent, resulting in low volume of sedimented ionic liquid phase or no sedimented ionic liquid phase after centrifugation (Wang Gao et al., 2016).The results showed that the highest ERs of the anthelmintic drug residues were obtained when the disperser solvent volume was 0.4 mL.Thus, a volume of 0.4 mL of methanol was selected for all subsequent experiments (Figure S4).

| Determination of the optimum pH of aqueous phase
The anthelmintic drugs mixture solution to be extracted was adjusted to pH 2.0-9.0 with hydrochloric acid or ammonium hydroxide prior to the IL-DLLME procedure (Figure 4).The results showed that the extraction recoveries >80% were fairly constant between pH 5 and 7 and decreased with an increase in pH.Most of the anthelmintic drug compounds are amphoteric, having the potential to act both as an acid and a base.Because most of the compounds are amphoteric, the variation of the pH of the aqueous sample extract had a slight positive impact on extraction recoveries with a slight deviation from pH 7, but with a higher impact at low pH values (2-4) and high pH values (8 and 9).The pH was found to have a minimal positive impact, and hence in this study, the pH of the aqueous extracts was maintained at pH 7.
F I G U R E 4 Effect of pH of aqueous blank extracts on extraction recoveries of 20 anthelmintic drugs (blank extracts, 5.0 mL ( pH 2, 3, 4, 5,  6, 7, 8, and 9  dissolved by vortex mixing.The results showed an increase of 4% from 90% to 94% (on average for all analytes) in the extraction recoveries of anthelmintic drugs at NaCl concentrations of between 0% and 0.5%, but extraction recoveries decreased to 84% when NaCl concentrations were at 1.0% and further decreased with increase in NaCl percentage (Figure S5).The salt addition (NaCl) in an extraction system can improve the extraction efficiency due to the salting out effect, but when the percentage of the salt increases, the ionic strength in the system increases, and the high ionic strength in an IL-DLLME system can enhance the solubility of an ionic liquid in aqueous phase, consequently decrease the extraction performance (Wang et al., 2015).The addition of NaCl was, therefore, not used for the subsequent work due to its minimal impact on the extraction recovery and to minimize the extraction time.

| Effect of extraction time
In an IL-DLLME procedure, the maximum quantity of analyte was transferred into the IL phase when the extraction equilibrium is obtained (Wang et al., 2015).The work was carried out by fortifying the aqueous blank extracts with anthelmintic drug standard mixture at MRL and shaking the mixture for 1 min.The effect of extraction time on the yield was evaluated by varying the extraction times (3, 5, 10, 20, 30, 40, 50, and 60 s) prior to centrifugation (Figure 5).The results showed that an extraction recovery of 77% for the anthelmintic drugs was observed when the extraction time was 20 s.To ensure maximum extraction efficiency, the cloudy sample solution was obtained at 40 s prior to centrifugation.Forty seconds was, therefore, selected for all subsequent experiments in this study.

| Effect of the centrifugation time
The cloudy solution formed is centrifuged and the fine IL droplets settle at the bottom of the centrifuge tube; however, it is important to establish the optimum centrifugation time with respect to the settlement IL volume and the analyte concentration in the IL phase (Wang et al., 2015).MRL.The cloudy solutions were allowed to stand for 40 s prior to centrifugation at varying times (1, 3, 5, 10, and 15 min) and at 2795 g.
Results showed that the extraction recoveries >77% of anthelmintic drugs were achieved within the centrifugation time of 5 min (Figure S6).Therefore, a centrifugation time of 5 min for the formed cloudy solution was selected for all subsequent experiments.

| Linearity
The linearity of the mass spectrometry using a 5-point calibration curve in the range of 0.0, 0.5, 1.0, 1.5, and 2.0 times MRLs of compounds was achieved (Figure S1).The coefficient of determination (r 2 ) values for the calibration curves used in the study were ≥.99 and the r 2 values ranged from .9980 to .9999 as listed in Table 4.

| Selectivity/specificity
In this experiment, 10 different lots of caprine and 10 different lots  3.2.4| The decision limit (CCα) The decision limit for confirmation (CCα) means the limit at and above which it can be concluded with an error probability of α that a sample is noncompliant and the value 1 − α means statistical certainty in percentage that the permitted limit has been exceeded (Commission, 2021).The CCα values were determined by analyzing at least 21 blank materials per matrix fortified with the analyte (s) at MRL concentrations.The mean concentration at MRL plus 1.64 times the corresponding standard deviation equals the decision limit (α = 5%).
matrices requires an extraction step that releases analytes from the sample matrix.Complex biological samples contain the analyte alongside a diverse range of chemicals that can have an adverse impact on the accurate and precise quantification of the analyte.Caprine and TA B L E 1 Anthelmintics and their MRLs as in Commission Regulation (EU) No 37/2010.
this study, acetone, methanol, and acetonitrile were evaluated as possible disperser solvents.The study was carried out by fortifying the blank extracts with anthelmintic drug standard mixture at MRL concentrations with 0.300 mL of each disperser solvent (methanol, acetone, and acetonitrile) containing 50 μL of the selected extraction solvent [C6MIM][PF6] and analyzed in LC-MS/MS to calculate the extraction recoveries.Figure3shows the results of the disperser solvent optimization.These three solvents all performed well as disperser solvents with extraction recoveries >45% for all analytes.
Effect of the addition of salt(NaCl)    In this work, the disperser solvent at 0.4 mL containing 60 μL of [C6MIM][PF6] was injected in aqueous liver extracts fortified with anthelmintic drug at MRL concentrations, the extracts contained various amounts of NaCl salt (0%, 0.5%, 1%, 2%, 4%, 6%, 8% [w/v]) extraction method was validated according to Commission Implementing Regulation (EU) 2021/808 of 22 March 2021 and ISO/ IEC 17025:2017 (Wang et al., 2015).The validation parameters carried out were specificity/selectivity, LOD, LOQ, linearity, decision limit (CCα), within laboratory reproducibility, and repeatability.The validation study was carried out using seven independent blank liver samples fortified at 0.25, 1.0, and 1.5 times MRL levels of individual analytes, and the experiments were repeated twice on different days bringing the total number of samples to 20 per fortification level (n = 21).The samples were quantified with pre-extraction spiked matrix standards (PrEMS) fortified at 0.0, 0.25, 1.0, 1.5, and 2.0 times MRLs.The calculated LOD, LOQ, coefficient of determination (R 2 ), and CCα values based on interbatch reproducibility data are given in Table of ovine blank samples were analyzed together with 10 different lots of caprine and 10 different lots of ovine blank samples fortified at MRL concentrations with the anthelmintic drugs standard to check for any interferences of signals, peaks or ion traces in the region of interest where the target analyte is expected to elute.The fortified liver and blank liver samples were analyzed using mass spectrometry (Figure S2).The anthelmintic drug standards fortified at MRLs were extracted using IL-DLLME and analyzed using LC-MS/MS.A typical ion chromatogram is shown in Figure 1.No interfering peaks were observed at the retention time for the respective anthelmintic analytes.3.2.3 | Trueness/accuracyTrueness means the closeness of agreement between the average value obtained from a large series of test results and an accepted reference value.For the determination of accuracy/trueness, caprine and ovine liver samples were fortified at 0.25 × MRL, 1.0 × MRL, and 1.5 × MRL for each anthelmintic compound.Mean recovery (n = 21) for all analytes, determined on three separate occasions, was within the recommended range of 70%-120% for values >1 to 10 μg/ kg and 80%-120% for spiking values <10 μg/kg as in Commission Implementing Regulation (EU) 2021/808 of 22 March 2021.The percentage recoveries for all the compounds are shown in

Anthelmintics Marker residue Animal species Matrix MRL (μg/kg)
The drawback of the classical LLE method is (Rawa-Adkonis et al., 2006) its own limitations which include the use of large organic solvents, possibility of low recoveries due to interaction between sorbent and sample matrix, low reproducibility due to differences between sorbent batches and sorbent bed clogging (cartridges as well as disks) by particles of sample suspended matter(Rawa-Adkonis et al., 2006).In the past two decades, a greener sample preparation approach, liquid-phase microextraction (LPME), emerged as an attractive alternative to the traditional extraction and Calculated LOD, LOQ, coefficient of determination (R 2 ), and CCα values based on interbatch reproducibility data.
[C8MIM][PF6] in water was 18.8, 7.5, and 2.0 mg L −1 , respectively(Ionic Liquid, 2022).According to the experimental results, the solubility of ionic liquids in water affected their extraction recoveries.An increase in extraction recoveries was observed with an increase in the alkyl chain length of imidazole-based ionic liquids and the highest recoveries were achieved with imidazole-based ionic liquids containing six or more alkyl chain lengths.Poor extraction was observed with the [C4MIM][PF6] ionic liquid in comparison with those with the longer alkyl chain length (six to eight alkyl chain lengths).The higher the solubility of the ionic liquid in water, the lower the TA B L E 4 The work was carried out by fortifying the aqueous blank extracts with anthelmintic drug standard mixture at Effect of extraction time on extraction recoveries of 20 anthelmintic drugs (spiked aqueous blank extracts, 5.0 mL with 400 μL containing 60 μL of [C6MIM][PF6]) left to stand for 3, 5, 10, 20, 30, 40, 50, and 60 s after shaking.Examples of reproducibility CVs for quantitative methods over a range of analyte mass fractions [10].Intra-and interbatch variation for recovery of anthelmintic drug residues from liver samples.
Table 4 above shows the CCα values for the anthelmintic drugs, using the interbatch reproducibility data.Comparison of IL-DLLME method with QuEChERS method.