Quantitative Determination of Tenuazonic Acid in Pig and Broiler

Sep 15, 2015 - However, absorption was deemed to be slower in broiler chickens (mean tmax 0.32 h in ... Tissue residue studies and further research to...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/JAFC

Quantitative Determination of Tenuazonic Acid in Pig and Broiler Chicken Plasma by LC-MS/MS and Its Comparative Toxicokinetics Sophie Fraeyman,*,† Mathias Devreese,† Nathan Broekaert,† Thomas De Mil,† Gunther Antonissen,†,‡ Siegrid De Baere,† Patrick De Backer,† Michael Rychlik,§ and Siska Croubels† †

Department of Pharmacology, Toxicology and Biochemistry and ‡Department of Pathology, Bacteriology and Avian Diseases, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, 9820 Merelbeke, Belgium § Chair of Analytical Food Chemistry, Technische Universität München, Alte Akademie 10, 85354 Freising, Germany ABSTRACT: A liquid chromatography−tandem mass spectrometry (LC-MS/MS) method to quantitate tenuazonic acid (TeA) in pig and broiler chicken plasma was successfully developed and validated. Linear matrix-matched calibration curves ranged between 5 and 200 ng/mL. Correlation coefficients, goodness-of-fit coefficients, and within-day and between-day precision and accuracy fell well within the acceptance criteria. The limit of quantitation was 5.0 ng/mL in both pig and broiler chicken plasma. The LC-MS/MS method was applied in a comparative toxicokinetic study in both pigs and broiler chickens. TeA was completely bioavailable after oral administration in both animal species. However, absorption was deemed to be slower in broiler chickens (mean tmax 0.32 h in pigs vs 2.60 h in chickens). TeA was more slowly eliminated in broiler chickens (mean t1/2el 0.55 h in pigs vs 2.45 h in chickens after oral administration), mainly due to the significantly lower total body clearance (mean Cl 446.1 mL/h/kg in pigs vs 59.2 mL/h/kg in chickens after oral administration). Tissue residue studies and further research to elucidate the biotransformation and excretion processes of TeA in pigs, broiler chickens, and other animal species are imperative. KEYWORDS: tenuazonic acid, LC-MS/MS, toxicokinetics, oral bioavailability, broiler chicken, pig

■

daily esophageal intubations of 1.25 or 2.5 mg TeA/kg BW.9 In the chicken embryo assay, the LD50 of TeA was 548 μg/egg, whereas no teratogenic effects were observed.10 Notwithstanding the high prevalence of this Alternaria mycotoxin in feed and food, and consequently the exposure of both animals and humans, little is known about the possible effects of TeA on animal or human health. This was acknowledged by the European Food Safety Authority (EFSA) stating that “at present, the knowledge on the possible effects of Alternaria toxins on animals (...) is scarce and not sufficient to assess the risk regarding Alternaria toxins for animal health”.11 The first aim of the present study was to develop a liquid chromatography−tandem mass spectrometry (LC-MS/MS) method to quantitate TeA in pig and broiler chicken plasma. To the best of our knowledge, no such analytical methods have been reported for plasma. With regard to food, different analytical, primarily LC, methods have been developed for the detection of Alternaria mycotoxins.12 To improve chromatographic properties of TeA, modifiers were added to the mobile phases. However, these modifiers were not compliant with MS detection.5,12,13 Derivatization of TeA overcame the chromatographic difficulties without the need for modifiers and resulted in the development of LC-MS/MS methods to quantitate TeA in different food matrices.4,13 However, derivatization is labor intensive, and derivatization yields may vary. Consequently, some LC-UV and LC-MS/MS methods for the quantification

INTRODUCTION Alternaria fungi can infect a variety of crops including wheat, sorghum, barley, and several fruits and vegetables. These fungi can produce several mycotoxins, although the most prevalent ones are alternariol (AOH), alternariol monomethyl ether (AME), and tenuazonic acid (TeA).1 In a 10 year survey, TeA was detected in 30.3% (322/1064) of German wheat samples with a maximum concentration of 4223.6 μg/kg in the year 2007.2 In a recent survey, TeA was detected in 65% of 83 feed and feed raw material samples analyzed, with a median concentration of 63 μg/kg.3 Furthermore, Asam et al. showed the high prevalence of TeA in a variety of foodstuffs. More specifically, 86% of the fruit juices, 92% of the cereals, and 87% of the spices investigated were contaminated with median concentrations of 1.8, 16, and 500 μg/kg, respectively.4 Next, TeA was also detected in tomato- and pepper-derived samples, with concentrations ranging from 3 to 2330 μg/kg.5 With regard to human exposure, Asam et al. demonstrated the presence of TeA in the urine of six human volunteers, with concentrations ranging from 1.3 to 17.3 μg/L. Regarding toxicokinetics, 87−93% of orally consumed TeA was excreted as parent compound in the urine of two human volunteers within 24 h after ingestion.6 In vitro data suggest that TeA exerts its toxic effect by suppressing chain termination or release of newly formed proteins from the ribosomes.7 To date, only a few papers are available regarding the in vivo toxicity of TeA. After ingestion of 10 mg TeA/kg body weight (BW), dogs displayed significant toxic effects, that is, retching, vomiting, flushed skin, bloody diarrhea, and systemic dehydration. Necropsy revealed hemorrhages in the lungs and the gastrointestinal tract.8 Hemorrhages have also been detected in broilers administered © XXXX American Chemical Society

Received: February 5, 2015 Revised: August 31, 2015 Accepted: September 7, 2015

A

DOI: 10.1021/acs.jafc.5b02828 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 1. Chemical structure of tenuazonic acid (TeA, left) and LC-MS/MS chromatograms of (A) a blank pig plasma sample, (B) a spiked pig plasma sample (TeA concentration = 100 ng/mL), and (C) an incurred pig plasma sample that was taken at 0.5 h pa of a single oral dose of 0.05 mg TeA/kg body weight (TeA concentration = 129 ng/mL). The upper trace presents TeA, and the lower trace presents the internal standard ([13C6, 15 N]-TeA). After an acclimatization period of at least 3 days, each pig or broiler chicken received either an oral (po) or intravenous (iv) dose of 0.05 mg TeA/kg BW. The analytical standard of TeA was dissolved in ethanol at a concentration of 1.75 or 1.00 mg TeA/mL for pigs or chickens, respectively. The stock solution was further diluted with water (po) or saline (iv) up to a final concentration that corresponded with a dose of 0.05 mg/kg BW. Animals were deprived of feed from 12 h before until 4 h after administration of TeA. Pigs and broiler chickens had ad libitum access to drinking water at all times. Oral bolus administrations were given in the crop (chickens) or by gastric gavage (pigs), whereas iv injections were performed in the wing vein (vena basilica, chickens) or through a double-lumen catheter (vena jugularis externa, pigs).16 About 1 mL of blood was sampled via the catheter (pigs) or the leg vein (vena metatarsalis plantaris superf icialis, broilers) and collected in EDTAcoated tubes just before (0 h) and at different time points postadministration (pa), that is, at 0.08, 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, and 10 h pa for pigs and at 0.08, 0.16, 0.32, 0.50, 1, 2, 4, 6, 8, and 10 h pa for broiler chickens. The blood samples were centrifuged (2095g, 10 min, 4 °C), and plasma samples (aliquots of 250 μL) were stored at ≤ −15 °C until analysis. The animal experiments were approved by the Ethical Committee of the Faculty of Veterinary Medicine and Bioscience Engineering of Ghent University (EC 2013/145). Plasma Sample Preparation. To 250 μL of plasma was added 12.5 μL of the IS WS (1 μg/mL). After a vortex-mixing step, 750 μL of ACN was added followed by vortex-mixing and centrifugation (8517g, 10 min, 4 °C). The supernatant was evaporated under a gentle nitrogen (N2) stream (45 ± 5 °C). The dry residue was dissolved in 262.5 μL of water/MeOH (80:20, v/v) for pig plasma and in 5 mM ammonium formate in water (pH 9)/MeOH (95:5, v/v) for chicken plasma. After a vortex-mixing step, the sample was filtered through a Millex-GV PVDF filter (0.22 μm) and transferred to an autosampler vial. A 10 μL aliquot was injected onto the LC-MS/MS instrument. Plasma Protein Binding. Plasma protein binding of TeA was assessed by spiking fresh blank pig and broiler chicken plasma at 10, 50, and 100 ng/mL. At each concentration level, one aliquot (aliquot 1) followed the plasma sample treatment (vide supra), but without filtering through a Millex-GV filter. A second aliquot and a third aliquot (aliquots 2 and 3) were incubated in a hot water bath, set at 37 and 41 °C for pig and broiler chicken plasma, respectively. After 1 h, aliquots 2 and 3 were transferred onto Amicon Ultra-0.5 centrifugal filter devices. Following centrifugation (8517g, 15 min, 37 or 41 °C), aliquots 2 and 3 were treated in a similar way as aliquot 1.

of TeA in food without the need for derivatization have been described.5,13,14 Therefore, the present study aimed to develop a LC-MS/MS method without a derivatization step. Next, toxicokinetic and plasma protein binding studies in both animal species were performed. The absolute oral bioavailability (F) and the main toxicokinetic parameters were calculated: area under the plasma concentration−time curve (AUC), maximum concentration after oral administration (Cmax), plasma concentration at time of intravenous injection (C0), time to maximal plasma concentration (tmax), elimination rate constant (kel), elimination half-life (t1/2el), total body clearance (Cl), volume of distribution (Vd), and mean residence time (MRT).

■

MATERIALS AND METHODS

Chemicals, Products, and Reagents. The analytical standard of TeA (purity ≥ 99%) was purchased from Fermentek (Jerusalem, Israel). The chemical structure of this compound is presented in Figure 1. The internal standard (IS) [13C6, 15N]-TeA was synthesized according to the method of Asam et al.15 TeA and the IS were stored at ≤ −15 °C. Water, methanol (MeOH), and acetonitrile (ACN) were of LC-MS grade and obtained from Biosolve (Valkenswaard, The Netherlands). Ammonium bicarbonate (purity ≥ 99.5%) and ammonium formate (purity > 99.0%) were of analytical grade and purchased from Sigma-Aldrich (Diegem, Belgium). Absolute ethanol was of analytical grade and obtained from VWR (Leuven, Belgium). Millex-GV PVDF filters (0.22 μm) and Amicon Ultra-0.5 mL centrifugal filter devices (molecular weight cutoff = 30 kDa) were purchased from Merck-Millipore (Overijse, Belgium). Preparation of Stock and Working Solutions. Stock solutions of TeA (1 mg/mL) and IS (100 μg/mL) were prepared in MeOH. Working solutions (WS) of 1, 0.1, and 0.01 μg TeA/mL were prepared daily by appropriate dilution of the stock solution and WS with water. A WS (1 μg/mL) of IS was made by further dilution of the stock solution in MeOH. Stock solutions of TeA and stock and WS of IS were stored at ≤ −15 °C. Animal Experiments. Toxicokinetic studies with TeA were performed in eight 10-week-old male piglets (Seghers hybrids, Ratterlow Seghers Holding, Lokeren, Belgium) and five 4-week-old broiler chickens (Ross 308, Geluveld, Belgium) of mixed gender. The mean BW was 33.44 ± 1.71 and 1.56 ± 0.20 kg for pigs and broilers, respectively. Both animal trials were designed as a two-way crossover. B

DOI: 10.1021/acs.jafc.5b02828 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry Corresponding calibration curves were used to quantify TeA in each aliquot. Plasma protein binding was calculated at each level (10, 50, and 100 ng/mL) using the following equation:17

plasma protein binding =

∑ (%difference)2 n−1

g=

(Caliquot 1 − Cav in aliquots 2 and 3) × 100

with % difference =

Caliquot 1

(2)

xback‐calculated − xnominal × 100 xnominal

Accuracy and Precision. Within-run accuracy and precision (repeatability) were determined by analyzing six blank samples that were spiked at 5, 10, and 100 ng/mL in the same run. The betweenrun accuracy and precision (reproducibility) were determined by analyzing two blank samples per concentration level (5, 10, and 100 ng/mL) on three different days (n = 6). The acceptance criteria for accuracy were −30% to +10% and −20% to +10% for concentration levels between 1 and 10 ng/mL and ≥10 ng/mL, respectively. The precision was evaluated by the determination of the relative standard deviation (RSD), which had to be below the RSDmax value. For the within-day precision, RSDmax is fixed at 25, 15, and 10% for concentrations of 5, 10, and 100 ng/mL respectively,20 whereas for between-run precision, the RSD had to be below the RSDmax calculated by the Horwitz equation:

(1) Liquid Chromatography. The LC system consisted of a quaternary, low-pressure mixing pump with vacuum degassing (type Surveyor MSpump Plus, ThermoFisher Scientific, Breda, The Netherlands) connected to an autosampler with a temperaturecontrolled tray and column oven (type Autosampler Plus, ThermoFisher Scientific). Chromatographic separation was achieved on a Zorbax Eclipse Plus column (C18; i.d., 3.0 × 100 mm; d.p., 3.5 μm) in combination with a guard column of the same type (i.d., 2.1 × 12.5 mm; d.p., 5 μm), both from Agilent Technologies (Sint-KatelijneWaver, Belgium). Mobile phase A consisted of 0.1 mM ammonium bicarbonate in water/MeOH (95:5, v/v) or 5 mM ammonium formate in water (pH 9) for pig and chicken plasma analysis, respectively. Mobile phase B consisted of MeOH. For the analysis of pig plasma samples, the following gradient elution program was run: 0.0−0.5 min (70% A, 30% B), 0.5−2.0 min (linear gradient to 20% A, 80% B), 2.0−7.0 min (20% A, 80% B), 7.0−7.5 min (linear gradient to 70% A, 30% B), 7.5−12.5 min (70% A, 30% B). For the analysis of broiler plasma samples, the gradient elution program was as follows: 0.0−2.0 min (90% A, 10% B), 2.0−2.5 min (linear gradient to 2% A, 98% B), 2.5−8.5 min (2% A, 98% B), 8.5−9.0 min (linear gradient to 90% A, 10% B), 9.0−12.5 min (90% A, 10% B). The flow rate was set at 300 μL/min. The column oven and tray temperature were set at 50 and 5 °C, respectively. Mass Spectrometry. The LC column effluent was interfaced to a TSQ Quantum Ultra triple-quadrupole mass spectrometer, equipped with a heated electrospray ionization (h-ESI) probe operating in the negative ionization mode (all from ThermoFisher Scientific). Instrument parameters were optimized by syringe infusion of a WS (1 μg/ mL) of TeA and the IS. The following MS/MS parameters were used: spray voltage, 4000 V; vaporizer temperature, 250 °C; sheath gas pressure, 40 arbitrary units (au); ion sweep gas pressure, 2.0 au; auxiliary gas pressure, 23 au; capillary temperature, 300 °C; source CID collision energy, 5 eV; and collision pressure, 1.5 mTorr. The resolution for Q1 and Q3 was set at 0.7 full width at half-maximum (fwhm). Acquisition was performed in the selected reaction monitoring (SRM) mode. The following transitions (m/z) were monitored for qualification and quantification, respectively, for TeA, 196 → 112 and 196 → 139, and for IS, 203 → 142 and 203 → 113. Tube lens offset was set at 98 and 84 V for TeA and IS, respectively. Method Validation. The method was validated according to a validation protocol previously described by De Baere et al., using spiked blank plasma samples obtained from healthy, untreated animals.18 A set of parameters that were in compliance with the recommendations and guidelines defined by the European Community and with criteria described in the literature were evaluated.18−22 These include evaluation of linearity, within- and between-run accuracy and precision, limit of quantitation (LOQ), limit of detection (LOD), specificity, carry-over, extraction recovery (RE), and matrix effects (signal suppression or enhancement, SSE). Linearity. Linearity was evaluated by preparing three matrixmatched calibration curves on three different days. Additionally, freshly prepared calibration curves were included during the analysis of the samples obtained from the toxicokinetic study. Calibration curve samples (range of 5−200 ng/mL) were prepared by applying standard working solutions directly onto the blank plasma samples, followed by a vortex mixing step. After 5 min of equilibration, the calibration curve samples were treated in a similar way as the unknown samples. The correlation coefficients (r) and goodness-of-fit coefficients (g) were calculated and had to be ≥0.99 and ≤20%, respectively.18,21,22 The following equation was used to calculate g:

RSDmax = 2(1 − 0.5log(concn))

(3)

Limit of Quantitation. The LOQ was the lowest concentration of the analyte for which the method was validated with an accuracy and precision that fell within the recommended ranges (see Accuracy and Precision). The LOQ was also established as the lowest point of the calibration curve. The LOQ was determined by analyzing six samples spiked at a TeA concentration level of 5 ng/mL, on the same day. Limit of Detection. The LOD was defined as the lowest concentration that could be recognized by the detector with a signal-to-noise (S/N) ratio of ≥3. The LOD values were calculated using the samples spiked at the LOQ level. Specificity. The specificity of the method was verified by analyzing 20 different blank plasma samples from both pigs and broiler chickens. The eventual analyte concentration of a peak in the elution zone of TeA or the IS had to be below the LOD. Carry-over. The absence of carry-over was verified by analyzing the reconstitution solvent injected after the highest calibration sample (200 ng/mL). If a peak was observed in the elution zone of TeA or the IS, it had to be below the LOD. Extraction Recovery and Matrix Effects. The slopes of three calibration curves were used to determine the extraction recovery (RE) and the matrix effects (SSE) as described by Matuszewski et al.23 The first calibration curve was made by spiking blank plasma samples before extraction at concentration levels of 1, 2, 5, 10, 20, 50, 100, and 200 ng/mL (= spiked samples). The second calibration curve was made by spiking blank plasma samples after extraction (= spiked extracts), whereas the third curve was prepared using standard solutions in the reconstitution solvent (= standard solutions). No IS was added, and absolute areas were used to calculate the RE and SSE through the following equations:

RE =

slope curve spiked samples slope curve spiked extracts

SSE =

(4)

slope curve spiked extracts slope curve standard solutions

(5)

Toxicokinetic Analysis. Noncompartmental toxicokinetic analysis was performed with WinNonlin 6.3 (Pharsight, St. Louis, MO, USA). The following main toxicokinetic parameters were calculated for iv and po administration in both animal species: area under the plasma concentration−time curve from time 0 to last concentration (AUC0−t, ng·h/mL), area under the plasma concentration−time curve from time 0 to infinity (AUC0−inf, ng·h/mL), maximum concentration after oral administration (Cmax, ng/mL), plasma concentration at time of iv injection (C0, ng/mL), time to maximal plasma concentration (tmax, h), elimination rate constant (kel, h−1), elimination half-life (t1/2el, h), total body clearance (Cl, mL/h/kg), volume of distribution (Vd, mL/kg), C

DOI: 10.1021/acs.jafc.5b02828 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 2. LC-MS/MS chromatograms of (A) a blank broiler plasma sample, (B) a spiked broiler plasma sample (tenuazonic acid (TeA) concentration = 100 ng/mL), and (C) an incurred broiler plasma sample that was taken at 2 h pa of a single oral dose of 0.05 mg TeA/kg BW (TeA concentration = 139 ng/mL). The upper trace presents TeA, and the lower trace presents the internal standard ([13C6, 15N]-TeA).

Table 1. Results of the Evaluation of Linearity (Goodness-of-fit Coefficient (g), Correlation Coefficient (r)), Extraction Recovery (RE), Matrix Effects (Signal Suppression or Enhancement, SSE), Limit of Quantitation (LOQ), and Limit of Detection (LOD) for the Analysis of Tenuazonic Acid in Plasma from Pigs and Broiler Chickensa

a

matrix

calibration range (ng/mL)

g (%)

r

RE (%)

SSE (%)

LOQ (ng/mL)

LOD (ng/mL)

pig plasma broiler chicken plasma

5−200 5−200

3.17 5.96

0.9996 0.9992

91.8 86.2

37.8 102.5

5.0 5.0

0.01 0.22

Acceptance criteria: g ≤ 20%, r ≥ 0.99.

and mean residence time (MRT, h). The absolute oral bioavailability (F) was calculated using the following equation: F=

ance and a more robust method for TeA in broiler plasma (Figure 2). Table 1 summarizes the validation results obtained for TeA in pig and chicken plasma. Linear matrix-matched calibration curves were reached (range of 5−200 ng/mL) with g and r values falling well within the acceptance criteria. The LOQ was 5.0 ng/mL in both pig and broiler plasma, whereas the LODs were 0.01 and 0.22 ng/mL in pig and chicken plasma, respectively. Furthermore, the protein precipitation sample treatment yielded high extraction recoveries, that is, 91.8% in pig plasma and 86.2% in broiler chicken plasma. There was a strong suppression of the signal of TeA in pig plasma (SSE = 37.8%), but not in broiler chicken plasma (SSE = 102.5%). The within-day and between-day precision and accuracy fell within the acceptance ranges and are presented in Table 2. There was no statistically significant difference in within-day and between-day precision for both pig and broiler chicken plasma at the concentration levels tested (5, 10, and 100 ng/ mL). The method was specific in both pig and broiler plasma because no interfering peaks could be detected in any of the blank samples at the retention time of TeA. Moreover, no significant carry-over was present, because no TeA was detected with a concentration higher than the LOD in the reconstitution solvent injected after the highest calibration sample (data not shown). Toxicokinetic Studies. No general side effects were observed after the po or iv administration of 0.05 mg TeA/ kg BW as a single dose. The plasma concentration−time profile of TeA in pigs and broiler chickens is shown in panels A and B of Figure 3, respectively. Table 3 summarizes the main

AUC0 − inf po AUC0 − inf iv

(6)

Cl and Vd after po administration were corrected for F. The mean absorption time (MAT, h) was calculated using the following equation: MATpo = MRTpo − MRTiv

(7)

Statistical Analysis. To assess the repeatability and reproducibility, a one-way analysis of variance (ANOVA, SPSS 21, IBM, Armonk, NY, USA) was performed at each calibrator concentration level (5, 10, and 100 ng/mL) for both broiler chicken and pig plasma. Furthermore, ANOVA was performed on all toxicokinetic parameters in both animal species and between both routes of administration. AUC0−inf iv, AUC0−t iv, t1/2el iv, MRTiv, tmax po, t1/2el po, and MRTpo were compared between broilers and pigs using a Kruskal−Wallis test. The level of significance was set at 0.05.

■

RESULTS LC-MS/MS Method. In the present study, the use of a double end-capped Zorbax Eclipse Plus column resulted in good chromatographic properties because TeA and the IS eluted as a sharp peak in both pig and broiler plasma (Figures 1 and 2). The use of ammonium bicarbonate and MeOH as mobile phases in a gradient elution program gave reproducible results in pig plasma. For the analysis of chicken plasma, the mobile phase ammonium bicarbonate in water/MeOH was replaced by ammonium formate in water and the gradient was adjusted. This resulted in a better chromatographic performD

DOI: 10.1021/acs.jafc.5b02828 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

toxicokinetic parameters. Mean AUC0−t values were 108.9 and 151.7 ng·h/mL in pigs and 805.6 and 893.2 ng·h/mL in broiler chickens after iv and po administration, respectively. Consequently, TeA was completely absorbed after oral administration in pigs (mean F = 1.47) and broiler chickens (mean F = 1.24). Mean maximum concentration after oral administration was reached at 0.32 h pa in pigs and at 2.60 h in broiler chickens. Furthermore, TeA was rapidly eliminated after po administration in pigs, but not in broiler chickens. In pigs, t1/2el was 0.51 and 0.55 h after iv and po administration, respectively. In contrast, t1/2el was much longer in broiler chickens, that is, 2.03 and 2.45 h after iv and po administration, respectively. The mean Cl values were 448.4 and 446.1 mL/h/kg in pigs and 59.3 and 59.2 mL/h/kg in broiler chickens after iv and po administration, respectively. The mean Vd values were 325.8 and 358.4 mL/kg in pigs and 164.6 and 207.2 mL/kg in broiler chickens after iv and po administration, respectively. The mean MRT values were 0.70 and 0.94 h in pigs and 4.46 and 5.52 h in broiler chickens after iv and po administration, respectively. In both animal species, the kel, t1/2el, Cl, and Vd did not statistically differ after iv or po administration. Plasma protein binding did not significantly differ between both animal species and was 72.8 ± 3.4 and 58.1 ± 9.2% in broiler chickens and pigs, respectively (p = 0.06).

Table 2. Results of the Within- and Between-Run Precision and Accuracy Evaluation for the Analysis of Tenuazonic Acid in both Pig and Broiler Chicken Plasmaa matrix

theor concn (ng/mL)

mean concn ± SD (ng/mL)

precision, RSD (%)

accuracy (%)

pig plasma

5b 5c 10b 10c 100b 100c

4.6 4.7 10.5 10.0 103.7 101.0

± ± ± ± ± ±

0.55 0.44 0.51 0.96 3.78 5.93

12.0 9.2 4.9 9.6 3.6 5.9

−8.8 −5.4 4.9 −0.5 3.7 1.0

broiler chicken plasma

5b 5c 10b 10c 100b 100c

5.2 5.1 9.2 10.0 101.2 100.8

± ± ± ± ± ±

0.63 0.77 0.74 0.78 7.50 4.17

12.1 15.0 8.1 7.7 7.4 4.1

4.6 2.1 −8.1 0.1 1.2 0.8

a

SD, standard deviation; RSD, relative standard deviation. Acceptance criteria: accuracy, 1−10 ng/L, −30% to +10%, and ≥10 ng/mL, −20% to +10%. Within-run precision (RSDmax): 5 ng/mL, 25.0%; 10 ng/mL, 15.0%; and 100 ng/mL, 10.0%. Between-run precision: 5 ng/mL, 35.5%; 10 ng/mL, 32.0%; and 100 ng/mL, 22.6%. bWithin-run. c Between-run.

■

DISCUSSION LC-MS/MS Method. First, a sensitive and specific LC-MS/ MS method was developed and validated to quantify TeA in pig plasma. Next, minor modifications were made for quantification of TeA in chicken plasma. The sample pretreatment consisted of a plasma protein precipitation method using ACN. Initially, a rather extensive sample preparation procedure for chicken plasma was evaluated. Three solid-phase extraction (SPE) columns were tested, that is, HybridSPE-Phospholipid cartridges (SigmaAldrich), C18 Bond elut (Agilent Technologies), and Oasis HLB columns (Waters, Zellik, Belgium). Compared to the protein precipitation method, the signal was reduced by factors of approximately 100 and 10 when using the Bond elut and HLB columns, respectively. The signal was completely lost when using the HybridSPE-Phospholipid cartridges. Therefore, the protein precipitation method was preferred as the final sample preparation method in both chicken and pig plasma. Protein precipitation is a frequently used sample preparation method for the LC-MS/MS analysis of several mycotoxins in pig and broiler chicken plasma.24−27 Despite several benefits of this extraction procedure, such as high throughput and cost effectiveness, a main drawback is the sustainability of the LCMS equipment due to remaining matrix compounds such as phospholipids in chicken plasma. Clotting was prevented by adjusting the mobile phase and gradient elution program for the analysis of TeA in chicken plasma, which resulted in a delayed retention time. Also, the divert valve was set to send as much effluent as possible to the waste. With regard to chromatographic performance, TeA has been described as a troublesome compound because of its acidic and metal-chelating properties. To obtain better chromatographic properties, modifiers such as Zn(II)SO4 were added to the mobile phase. However, these modifiers were not suited for MS.5,12,13 Shephard et al. improved the chromatographic performance of TeA by using an end-capped C18 column, which prevented interactions between the amine moiety of TeA and active silanol groups on the column.14 In the present study,

Figure 3. Plasma concentration−time profile of tenuazonic acid in (A) pigs (n = 8) and (B) broiler chickens (n = 5) after oral (po) and intravenous (iv) administration. Values are presented as mean ± SD.

E

DOI: 10.1021/acs.jafc.5b02828 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Table 3. Main Toxicokinetic Parameters of Tenuazonic Acid after Intravenous (iv) and Oral (po) Administration (0.05 mg/kg BW) to Pigs (n = 8) and Broiler Chickens (n = 5)a pig

broiler

toxicokinetic parameter

iv

po

iv

po

AUC 0−t (ng·h/mL) AUC 0−inf (ng·h/mL) F Cmax (ng/mL) C0 (ng/mL) tmax (h) kel (h−1) t1/2el (h) Cl (mL/h/kg) Vd (mL/kg) MRT (h) MAT (h)

108.9 ± 21.5 a 117.2 ± 23.8 a

151.7 ± 28.6 b 169.8 ± 36.0 b 1.47 ± 0.28 a 170.6 ± 32.2 a

805.6 ± 148.1 c 874.1 ± 182.2 c

893.2 ± 275.3 c 1026.7 ± 353.1 c 1.24 ± 0.51 a 133.3 ± 20.6 b

216.6 ± 53.0 a 1.38 ± 0.14 a 0.51 ± 0.05 a 448.4 ± 107.1 a 325.8 ± 75.1 a 0.70 ± 0.08 a

218.5 ± 53.5 a 0.32 ± 0.21 a 1.29 ± 0.23 a 0.55 ± 0.10 a 446.1 ± 110.8 a 358.4 ± 118.3 a 0.94 ± 0.20 b 0.24 ± 0.24 a

0.41 ± 0.17 b 2.03 ± 1.04 b 59.3 ± 13.6 b 164.6 ± 62.8 b 4.46 ± 1.06 c

2.60 ± 2.38 b 0.30 ± 0.09 b 2.45 ± 0.68 b 59.2 ± 13.2 b 207.2 ± 62.0 b 5.52 ± 1.23 c 1.06 ± 1.53 a

Values are presented as the mean ± SD. AUC0−t, area under the plasma concentration−time curve from time 0 to last concentration > LOQ; AUC0−inf, area under the plasma concentration−time curve from time 0 to infinity; F, oral bioavailability; Cmax, maximal plasma concentration; C0, plasma concentration at time 0; tmax, time to maximal plasma concentration; kel, elimination rate constant; t1/2el, elimination half-life; Cl, total body clearance corrected for F; Vd, volume of distribution corrected for F; MRT, mean residence time; MAT, mean absorption time. Values sharing a lower case online letter for the same parameter do not differ from one another at the 5% global significance level.

a

procedure, to exclude analytical variation. The percentage extrapolated AUCt−inf was 100% is regularly reported.30 In the study presented, the dose and Cl, two factors influencing AUC, did not differ between both routes of administration. Furthermore, the plasma samples obtained after iv and po administration of one animal were analyzed on the same day, using the same F

DOI: 10.1021/acs.jafc.5b02828 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

■

unchanged in the urine.6 Additionally, considering complete oral bioavailability, no first-pass effect by the liver was seen in pigs or broiler chickens. Because, to our knowledge, no in vitro nor in vivo information about the biotransformation of TeA is available, further research is needed to elucidate phase I and II metabolites in pigs and broiler chickens. Few toxicity studies have been performed in chickens. After chronic exposure, Giambrone et al. noted a decreased weight gain and lowered feed efficiency when TeA was given at 10 mg/ kg feed or by daily esophageal intubation (1.25 or 2.5 mg/kg BW) for 3 weeks. Other symptoms in the chickens receiving TeA via intubation were an enlarged and mottled spleen, erosions of the gizzard, and hemorrhages.9 Because the maximum feed contamination level detected by Streit et al. was