Rapid and sensitive UHPLC-MS/MS methods for dietary sample analysis of 43 mycotoxins in China total diet study.

INTRODUCTION: Mycotoxins are toxic metabolites produced by fungi that commonly contaminate foods. As recommended by the World Health Organization, total diet study (TDS) is the most efficient and effective way to estimate the dietary intakes of certain chemical substances for general populations. It requires sensitive and reliable analytical methods applicable to a wide range of complex food matrices and ready-to-eat dishes.
OBJECTIVES: A novel strategy with high selectivity and sensitivity, incorporating three methods based on ultra-high-performance liquid chromatography coupled with tandem mass spectrometry (UHPLC-MS/MS), was designed for measuring 43 mycotoxins in dietary samples in a China TDS.
METHODS: The 43 mycotoxins were divided into 3 groups for analysis to achieve better performance. For each group, an UHPLC-MS/MS method was developed to determine the target compounds after clean-up by solid phase extraction. A total of 21 isotope internal standards were employed for accurate quantitation. Method validation in terms of linearity, selectivity, sensitivity, accuracy, and precision was performed for all the 43 mycotoxins in 12 complex food matrices.
RESULTS: The limits of detection (LODs) and limits of quantitation (LOQs) were 0.002-1 ng mL-1 and 0.006-3 ng mL-1, respectively. The method recoveries of the 43 mycotoxins spiked in 12 food categories were in the range of 60.3%-175.9% after internal standard correction, with relative standard deviations (RSDs) below 13.9%. For practical application, this method was utilized for 72 dietary samples collected from 6 provinces in the 6th China TDS. More than 80% of the samples were found contaminated by mycotoxins. DON, SMC, FB1, ZEN, BEA, ENNB1, and ENNB were most detected.
CONCLUSIONS: The proposed methods with high sensitivity, accuracy, and robustness provide powerful tools for multi-mycotoxin monitoring and dietary exposure assessment, allowing 43 mycotoxins, including some emerging mycotoxins, to be accurately investigated in a total diet study for the first time.

Introduction

Mycotoxins represent toxic secondary metabolic products synthesized by filamentous fungal species. They are naturally occurring and widely found in various food types. Plant-based foods can be directly infested by the mycotoxin producing fungi. Other food products may become contaminated because of carry-over from feeds or raw materials. The general population is primarily exposed to common mycotoxins through their diet [1], [2]. Consumption of mycotoxin-contaminated foods may lead to acute or chronic effects, such as cytotoxicity and immunosuppression, as well as hepatic, gastrointestinal, and carcinogenic diseases [3], [4]. The World Health Organization (WHO) has described mycotoxins as one of the major causes of foodborne illnesses that pose a potential threat to animal and human health [5].

A total diet study (TDS) is jointly recommended by the WHO, the Food and Agriculture Organization (FAO), and the European Food Safety Authority (EFSA) as the most efficient and cost-effective method for evaluating dietary intakes of certain chemical compounds for population groups through cooked and ready-to-eat diets, [5], [6]. Incorporating the impact of cooking and preparation on less stable chemicals and on the formation of new ones, a TDS gives a more accurate estimation of dietary exposure. The analytical methods required to conduct a TDS must meet high requirements not only for sensitivity and reliability, but also for the practical applicability to a wide range of complex food matrices and ready-to-eat dishes.

Mycotoxins have been investigated in several TDSs conducted around the world, such as the French TDS (21 mycotoxins) [7], [8], [9], Netherlands TDS (37 mycotoxins) [10], [11], [12], Spanish TDS (18 mycotoxins) [3], Lebanese TDS (4 mycotoxins) [13], Canada TDS (1 mycotoxin) [14], Australia New Zealand TDS (11 mycotoxins) [15], Vietnam TDS (3 mycotoxins) [16], Ireland TDS (16 mycotoxins) [17], Regional Sub-Saharan Africa TDS (14 mycotoxins) [18], [19] and Hong Kong TDS (13 mycotoxins) [20]. A variety of analytical methods have been employed in TDSs, among which the LC-MS/MS technique is increasingly applied as a highly selective and a sensitive tool for multi-mycotoxin analysis in complex food matrices.

China has successfully conducted five TDSs since 1990 [21], [22]. The 6th China TDS (2016–2020) was further expanded to 24 provinces, representing the dietary habits of multiple geographical regions and covering most of the population (greater than2/3). In the 6th China TDS, daily consumed foods were classified into 13 categories: cereals and their products, legumes and their products, potatoes and their products, meats and their products, eggs and their products, aquatic foods, milk and dairy products, vegetables and their products, fruits and their products, sugar, water and beverages, alcohols, and condiments (including cooking oils). Cooking oil and condiments were put into the other 12 categories during the preparation and cooking of TDS samples, which further complicated the chemical compositions versus raw products, requiring an advanced analytical method.

The present study developed a sensitive, accurate, and robust strategy for detecting 43 mycotoxins, i.e. aflatoxin B1 (AFB1), aflatoxin B2 (AFB2), aflatoxin G1 (AFG1), aflatoxin G2 (AFG2), aflatoxin M1 (AFM1), aflatoxin M2 (AFM2), ochratoxin A (OTA), ochratoxin B (OTB), deoxynivalenol (DON), nivalenol (NIV), 3-acetyldeoxynivalenol (3A-DON), 15-acetyldeoxynivalenol (15A-DON), fusarenon-X (Fus X), 3-glucose-deoxynivalenol (DON-3-G), deepoxy-deoxynivalenol (DOM-1), HT2 toxin, T2 toxin, 4,15-diacetoxyscirpenol (DAS), neosolaniol (NEO), sterigmatocystin (SMC), citrinin (CIT), cyclopiazonic acid (CPA), moniliformin (MON), zearelenone (ZEN), zearalenone (ZAN), α-zearalenol (α-ZOL), β-zearalenol (β-ZOL), α-zearalanol (α-ZAL), β-zearalanol (β-ZAL), fumonisin B1 (FB1), fumonisin B2 (FB2), fumonisin B3 (FB3), patulin (PAT), beauvericin (BEA), enniatin A (ENNA), enniatin A1 (ENNA1), enniatin B (ENNB), enniatin B1 (ENNB1), tenuazonic acid (TeA) alternariol (AOH), altenuene (ALT), alternariol monomethyl ether (AME), and tentoxin (TEN), in all 12 food categories from the 6th China TDS by isotope dilution UHPLC/MS/MS. Considering the diversity of their physicochemical properties, the 43 mycotoxins were classified into three groups, with specific sample preparations and instrumental conditions for each group, to achieve the best performance. The methods were validated in terms of linearity, specificity, accuracy, LOD, LOQ, and intra- and inter-day variability, and then practically used to test 72 samples covering all 12 food categories. This versatile multi-mycotoxin and multi-matrix strategy with high sensitivity and broad applicability enables 43 mycotoxins, including Alternaria toxins and emerging toxins, to be included in a TDS for the first time and serves as a potent tool that will help monitor mycotoxins and assess dietary exposure.

Materials and methods

Chemicals and materials

LCMS-grade acetonitrile, methanol, ammonia acetate, formic acid, ammonia water, ammonium hydrogen carbonate and ammonium dihydrogen phosphate were commercially obtained from Fisher Scientific (USA). Mycotoxin standards for AFB1, AFB2, AFG1, AFG2, AFM1, AFM2, HT2, T2, 3A-DON, 15A-DON, FB1, FB2, FB3, OTA, OTB, AOH, AME, TeA, BEA, TEN, ALT, MON, DON, NIV, Fus X, ZEN, ZAN, α-ZOL, β-ZOL, α-ZAL, β-ZAL, PAT, DON-3-G, DOM-1, NEO, DAS, SMC, CIT, and CPA were obtained from Biopure (Austria). ENNs (ENNA1, ENNA, ENNB1, and ENNB) were supplied by PriboLab (Singapore). A total of 21 labeled internal standards were utilized, with 13C-DON, 13C-3-ADON, 13C-NIV, 13C-AFB1, 13C-AFB2, 13C-AFG1, 13C-AFG2, 13C-AFM1, 13C-ZEN, 13C-T2, 13C-HT2, 13C-PAT, 13C-DAS, 13C-SMC, 13C-FB1, 13C-FB2, 13C-FB3, 13C-OTA, and 13C-CIT provided by Biopure, 13C-TeA supplied by Fluka (USA), and TEN-d3 obtained from TRC (Canada). Ultrapure water was obtained on a Milli-Q system (Millipore, USA). The MycoSep 226 Aflazon+ multifunctional cartridge and MultiSep 211 Fum column were supplied by Romer Labs (Austria). The Oasis HLB SPE column (200 mg, 6 mL) was purchased from Waters (USA).

Preparation of standard solution

The choice of mycotoxin concentration is based on their sensitivity on the instrument and initial concentrations in commercial standard solutions. The mixed standard solution A contained 5 μg mL−1 of MON, PAT, ZEN, NIV, Fus X, DON, 3A-DON, 15A-DON, T2, and HT2; 2.5 μg mL−1 of NEO, DAS, DON-3-G, and DOM-1; 0.5 μg mL−1 of ZAN, α-ZOL, α-ZAL, β-ZOL, β-ZAL, and SMC; 0.1 μg mL−1 of AFB1 and AFG1; 0.025 μg mL−1 of AFB2, AFM1, AFM2, and AFG2. The mixed internal standard solution A contained 25 ng mL−1 of 13C-AFB1, 13C-AFB2, 13C-AFG1, 13C-AFG2, and 13C-AFM1; 50 ng mL−1 of 13C-T2; 0.5 μg mL−1 of 13C-HT2, 13C-DON, 13C-3-ADON, 13C-NIV, 13C-PAT, 13C-DAS, and 13C-SMC; 0.15 μg mL−1 of 13C-ZEN.

The mixed standard solution B contained 2.5 μg mL−1 of FB1, FB2, and FB3 along with 0.5 μg mL−1 of OTA and OTB. The mixed internal standard solution B contained 0.5 μg mL−1 of 13C-FB3 and 13C-OTA along with 0.25 μg mL−1 of 13C-FB1 and 13C-FB2.

The mixed standard solution C contained 5 μg mL−1 of Alternaria Toxins, CPA and CIT along with 0.5 μg mL−1 of ENNs and BEA. The mixed internal standard solution C contained 25 ng mL−1 of 13C-AFB2 and 0.5 μg mL−1 of 13C-TeA and TEN-d3.

Stock solutions A and C were prepared in acetonitrile, and stock solution B was prepared in acetonitrile–water (1:1, v/v). All the solutions were stored at −40 °C in the dark and diluted with the initial solvents for UHPLC-MS/MS analysis.

Food samples

In total, 72 dietary samples in the 12 categories were collected from six provinces (Hebei, Beijing, Jilin, Hubei, Guangdong, and Guizhou) in the 6th China TDS. Food sampling was designed similar to previous China TDSs [21], [22]. The 12 types of dietary samples were clustered in accordance with the local dietary recipes and consumption of local residents. After cooking, the prepared food was mixed to form a provincial composite sample for each food category. All samples were transferred to the laboratory as soon as possible through the cold chain and stored at −20 °C prior to analysis. The 72 samples were only used for method development and pre-screening purposes. The result is not sufficient to present the contamination level of studied mycotoxins in China.

Sample preparation

The 43 mycotoxins were assigned to 3 groups for detection. Group A included 26 mycotoxins (AFB1, AFB2, AFG1, AFG2, AFM1, AFM2, HT2, T2, 3A-DON, 15A-DON, MON, DON, NIV, Fus X ,ZEN, ZAN, α-ZEL, β-ZEL, α-ZAL, β-ZAL, PAT, DON-3-G, DOM-1, NEO, DAS, and SMC). Group B included 5 mycotoxins (FB1, FB2, FB3, OTA, and OTB). Group C included 12 mycotoxins (AOH, AME, TeA, ALT, TEN, BEA, ENNA1, ENNA, ENNB1, ENNB, CIT, and CPA).

Group A

Homogenized food samples of exactly 2 g (2 mL of water or beverage) in a 50-mL centrifuge tube were added to 40 μL of mixed isotope internal standard A (13C-AFB1, 13C-AFB2, 13C-AFG1, 13C-AFG2, 13C-AFM1, 13C-T2, 13C-HT2, 13C-DON, 13C-3-ADON, 13C-NIV, 13C-PAT, 13C-DAS, 13C-SMC, and 13C-ZEN) and 9 mL of extraction solution (acetonitrile/water solution; 84:16, v/v). The mixture was incubated at room temperature for 0.5 h, ultrasonicated for 0.5 h, and centrifuged at 9000 rpm for 10 min. The supernatant was then obtained for purification.

Exactly 5 mL of the supernatant was purified by passing through a MycoSep 226 Aflazon+ multifunctional cartridge. Then additional 3 mL of acetonitrile solution (acetonitrile/water; 84:16 v/v) was added to the cartridge to push out the remaining sample solution. All the effluent was collected, nitrogen-dried at 40 °C, and reconstituted in 1 mL of acetonitrile/0.2% formic acid in water (1:9 v/v). After vortex mixing for 30 s, the solution underwent centrifugation at 20000 rpm for 30 min for sample injection.

Group B

Food samples of exactly 2 g (2 mL of water or beverage) in a 50-mL centrifuge tube were added to 40 μL of mixed isotope internal standard B (13C-FB1, 13C-FB2, 13C-FB3, and 13C-OTA) and 10 mL extraction solution (acetonitrile/water; 1:1 v/v), shaken for 1 h at room temperature, and centrifuged at 9000 rpm for 10 min. Precisely, 5 mL of the resulting supernatant was adjusted to pH 6–9 with 0.1 mol L−1 NaOH solution, and added to 10 mL methanol/water (3:1 v/v).

The resulting mixture was allowed to pass through a MultiSep 211 Fum cartridge, which was subsequently washed with 10 mL methanol/water (3:1 v/v) to remove interfering compounds. After drying the cartridge, the analytes were eluted with 10 mL 0.1% formic acid methanol, nitrogen-dried at 40 °C, and reconstituted in 1 mL acetonitrile/0.2% formic acid in water (1:4 v/v). After vortex mixing for 30 s, the solution was centrifuged at 20000 rpm for 30 min prior to analysis.

Group C

Sample preparation of group C was the same as described in our previous publication [23]. Briefly, food samples of 2 g (2 mL of water or beverage) in a 50-mL centrifuge tube were added to 40 μL of mixed isotope internal standard C (13C-TeA, TEN-d3, and 13C-AFB2) and 9 mL extraction solution (acetonitrile/methanol/water; 45:10:45 v/v/v, pH 3.0 NaH2PO4), incubated for 0.5 h at room temperature, ultrasonicated for 0.5 h, and centrifuged at 9000 rpm for 10 min. Then, 5 mL of the resulting supernatant was added to 15 mL of 0.05 mol L−1 phosphate buffer (pH 3). After vortexing for 30 s, the supernatant was obtained for purification.

The resulting mixture was loaded onto an Oasis HLB SPE column that had been preconditioned with 5 mL of methanol/acetonitrile (1:1 v/v) followed by 5 mL of phosphate buffer (0.05 mol L−1, pH 3). The cartridge was then washed with 5 mL of methanol/water solution (1:4 v/v) and eluted with 5 mL each of methanol and acetonitrile. The eluate was nitrogen-dried at 40 °C, and reconstituted in 1 mL acetonitrile /water (1:9 v/v). After vortexing for 30 s, the solution underwent centrifugation (20000 rpm, 30 min) for sample injection.

UHPLC-MS/MS

UHPLC-MS/MS analysis was performed on an Exion LC AD™ System (SCIEX, USA) coupled with a Triple Quad 6500+ mass spectrometer (SCIEX, USA). The Analyst®1.6.3 and MultiQuant™3.0.2 were utilized for instrument operation and data processing.

The 26 major mycotoxins were separated on a CORTECS™ UPLC® C18 Column (2.1 × 100 mm, 1.6 μm, Waters). Water (A) and methanol/acetonitrile (1:1 v/v) (B) were used as the eluent with the following gradient: 5% B (initial), 5%–11% B (1–3 min), 11% B (3–12 min), 11%–28% B (12–12.1 min), 28% B (12.1–17 min), 28%–42% B (17–19 min), 42%–48% B (19–26 min), 48%–100 % B (26–27 min), 100% B (27–30 min), 100%–5% B (30–30.1 min), and 5% B (30.1–32 min). The flow rate was 0.4 mL/min. The column temperature was kept at 50 °C, and 5 µL of each sample was injected for analysis.

The MS/MS parameters in multi-reaction monitoring (MRM) mode under positive or negative ionization were optimized for each analyte as listed in Table 1. Other settings were as follows: ion spray voltages, −4500 V and +5500 V, respectively; source temperature, 550 °C; curtain gas, 20 psi; sheath gas, 50 psi; drying gas, 40 psi; collision gas (nitrogen), medium.

Table 1

MS/MS parameters on the precursor, quantification and confirmation daughter ion, declustering potential, and collision energy of 43 mycotoxins in the MRM mode.

Analyte	Precursor	Quantification ion	DP/CEa	Confirmation ion	DP/CEa
15A-DON	337.1(-H)	150.1	−20/−20	277.1	−20/−12
3A-DON	339.1(+H)	137.1	60/15	231.1	60/20
AFB1	313.1(+H)	240.9	120/55	284.9	120/28
AFB2	315.0(+H)	287.1	120/40	259.2	120/38
AFG1	329.1(+H)	311.1	125/33	243.3	125/38
AFG2	331.0(+H)	245.0	120/44	257.0	120/42
AFM1	329.2(+H)	259.1	135/35	273.2	135/32
AFM2	331.0(+H)	257.0	120/42	245.0	120/44
ALT	292.9(+H)	275.1	30/13	257.0	30/25
AME	270.9(-H)	256.0	−110/−29	228.0	−110/−39
AOH	258.8(+H)	185.1	150/43	213.0	150/37
BEA	784.5(+H)	244.2	220/38	262.3	220/34
CIT	250.9(+H)	232.8	50/40	205.1	50/37
CPA	334.9(-H)	140.0	−120/−36	180.1	−120/−37
DAS	384.2(+NH4+)	307.1	20/15	247.1	20/20
DOM-1	281.1(+H)	233.0	30/20	109.0	30/17
DON	297.1(+H)	231.0	40/20	249.0	40/15
DON-3-G	297.1(-C6H11O6)	231.0	40/20	249.0	40/15
ENNA	682.3(+H)	210.0	220/34	228.2	220/37
ENNA1	668.2(+H)	210.0	200/32	228.2	200/33
ENNB	640.3(+H)	196.4	180/34	214.2	180/33
ENNB1	654.4(+H)	196.0	180/33	214.1	180/35
FB1	722.3(+H)	704.2	40/41	334.3	40/55
FB2	707.2(+H)	689.3	50/40	337.4	50/52
FB3	707.2(+H)	337.3	50/50	355.3	50/46
Fus X	353.4(-H)	262.9	−50/−15	204.6	−50/−18
HT2	447.1(+Na)	345.0	100/25	285.1	100/28
MON	97.0(-Na)	40.8	−40/−21	–	–
NEO	400.1(+NH4+)	305.1	30/17	215.2	30/23
NIV	311.2(-H)	281.0	−20/−20	205.0	−20/−13
OTA	404.1(+H)	239.1	50/34	358.1	50/20
OTB	371.1(+H)	205.9	40/31	188.1	40/35
PAT	152.8(-H)	109.0	−60/−12	81.0	−60/−16
SMC	325.1(+H)	310.0	120/35	280.9	120/52
T2	489.1(+Na)	387.2	150/30	245.1	150/36
TeA	196.2(-H)	139.0	−50/−28	112.2	−50/−34
TEN	415.3(+H)	312.2	120/29	301.9	120/19
ZAN	319.3(-H)	275.0	−130/−25	205.0	−130/−28
ZEN	317.2(-H)	175.1	−140/−40	131.3	−140/35
α-ZAL	321.2(-H)	277.0	−150/−30	303.2	−150/−30
α-ZOL	319.3(-H)	274.9	−135/−30	160.0	−135/−39
β-ZAL	321.2(-H)	277.3	−155/−30	303.1	−155/−30
β-ZOL	319.2(-H)	275.3	−150/−25	159.8	−150/−40
13C-AFB1	330.3(+H)	301.2	115/31	255.2	115/57
13C-AFB2	332.0(+H)	303.2	100/38	273.1	100/45
13C-AFG1	346.3(+H)	328.2	70/30	257.1	70/40
13C-AFG2	348.2(+H)	330.2	55/39	259.3	55/45
13C-AFM1	346.1(+H)	288.1	100/32	273.0	100/30
13C-CIT	264.2(+H)	246.2	60/24	217.1	60/38
13C-3A-DON	356.3(+H)	245.2	60/15	145.2	60/45
13C-DAS	403.2(+NH4+)	244.3	30/23	213.2	30/24
13C-DON	312.2(+H)	263.2	60/17	245.1	60/15
13C-FB1	756.3 (+H)	738.5	50/56	356.4	50/43
13C-FB2	740.4 (+H)	358.4	50/53	722.4	50/42
13C-FB3	740.4 (+H)	358.3	75/53	376.4	75/47
13C-HT2	469.3(+Na)	362.2	120/29	300.3	120/26
13C-NIV	326.1(-H)	295.1	−67/−15	183.2	−67/−45
13C-OTA	424.1 (+H)	250.0	50/34	377.3	50/20
13C-PAT	160.1(-H)	115.0	−160/−13	86.2	−160/−15
13C-SMC	343.3(+H)	297.2	100/35	327.0	100/50
13C-T2	513.1(+Na)	406.3	163/33	334.2	163/32
13C-TeA	198.2(-H)	141.0	−50/−28	114.0	−50/−36
13C-ZEN	335.0(-H)	185.2	−150/−30	140.0	−150/−35
TEN-d3	418.2(+H)	314.9	140/30	305.4	140/19

DP, declustering potential (V); CE, collision energy (eV).

Chromatographic separation of FBs (B1, B2, and B3) and OTs (OTA and OTB) was performed on a CORTECS™ UPLC® C18 Column (2.1 × 100 mm, 1.6 μm, Waters) with a mobile phase consisting of 0.1% formic acid (A) and acetonitrile (B) at a flow rate of 0.4 mL/min. The following elution gradient was applied: 30% B (initial), 30%–45% B (0–2 min), 45%–55% B (2–5 min), 55%–100% B (5–6 min), 100% B (6–8 min), 100%–30% B (8–8.1 min), and 30% B (8.1–10 min). The column temperature was kept at 50 °C, and the injection volume was 5 µL.

The analytes were detected in positive MRM mode with parameters shown in Table 1. Other settings were as follows: ion spray voltage, +5500 V; source temperature, 550 °C; curtain gas, 25 psi; sheath gas, 55 psi; drying gas, 65 psi; collision gas (nitrogen) medium.

Chromatographic separation of the 12 mycotoxins in group C was carried out with a CORTECS™ UPLC® C18 Column (2.1 × 100 mm, 1.6 μm, Waters) as reported in our previous paper [23]. The eluent was composed of 0.01% aqueous ammonia with 5 mmol L−1 ammonium acetate (A) and acetonitrile (B). The gradient elution was performed as follows: 10% B (0–1 min), 10%–35% B (1–4 min), 35%–76% B (4–6 min), 76% B (6–7.5 min), 76%–100% B (7.5–8 min), 100% B (8–10 min), 100%–10% B (10–10.1 min), and 10% B (10.1–12 min). The flow rate was set at 0.4 mL min−1. The column temperature was kept at 50 °C, and 5 µL of each sample was injected for analysis.

The MS/MS parameters in multi-reaction monitoring (MRM) mode under positive or negative ionization were optimized for each analyte as listed in Table 1. Other settings were as follows: ion spray voltages of −4500 V and +5500 V, respectively; source temperature, 450 °C; curtain gas, 20 psi; sheath gas, 60 psi; drying gas 55 psi; collision gas (nitrogen), medium.

Method validation

Method validation was carried out following Commission Decision 2002/657/EC [24], EMEA [25], and FDA [26] guidelines. Validation parameters included selectivity, carry-over, linearity, accuracy (method recovery, RM), precision (intra- and inter-day variabilities), and sensitivity (LOD and LOQ).

The selectivity of the method was investigated by comparing the chromatograms of 12 distinct blank food samples with samples spiked with a mixture of analytes. The carry-over was carried out by injecting blank samples after injection of a high concentration calibration standard; residues were not greater than the respective analyte LODs.

The LOD and LOQ of each analyte were evaluated using both the standard solution and spiked blank samples in low amounts. Signal-to-noise (S/N) ratios for LOD and LOQ were above 3 and 10, respectively. Calibration standard curves of analytes were prepared using internal standard method and 1/x weighted linear regression with at least six concentration points for each analyte, assessed by calculating the regression coefficient (R2).

Apparent recovery (RA) and matrix effects (ME) were assessed using three sets of calibration curves without correction by internal standards. The values were determined as described below [23], [27]:

where A represents the slope of a calibration curve prepared in neat solvent; B represents the slope of a matrix-matched calibration curve prepared by spiking blank samples after sample preparation; and C represents the slope of a matrix-matched calibration curve prepared by spiking blank samples before sample preparation.

The accuracy and precision of the method were evaluated by recovery experiments. RM was investigated by spiking low, medium, and high concentrations of mycotoxin standards into 12 blank dietary matrices, which were corrected by internal standards. Method precision, expressed as intra-day and inter-day RSDs, was calculated using data from three different days in six replicates.

Results and discussion

This study aimed to develop a set of three methods for measuring 43 mycotoxins in 12 food categories with acceptable recoveries. This method combination can detect multiple mycotoxins with distinct characteristics in various food matrices. Applying the same sample preparation and analysis method to study all compounds is challenging because of the diverse and complex set of molecules that could potentially be found in ultra-trace amounts in dietary samples. Therefore, the 43 mycotoxins were divided into three groups for analysis.

MS/MS condition optimization

Optimization of MS/MS conditions was performed by direct infusion of individual standard of each analyte. Ionization mode, ion spray voltage, declustering potential (DP), curtain gas, source temperature, sheath gas, and drying gas were optimized stepwise to obtain the highest signal intensity of the precursor ion. ESI in negative and positive modes with ion pray voltages of −4.5 kV and +5.5 kV, respectively, were chosen. The collision energy (CE), a major factor that affects MRM transition, was optimized individually for each analyte to achieve the most sensitive and stable product ions. Two different MRM transitions per analyte were selected and optimized, except for MON. MON as a small molecule (molecular weight of 98 Da) yielded only one strong product ion. Thus, only one MRM transition (m/z 97 to m/z 41) can be programmed, as has been done in previous studies [28], [29]. The MRM transitions together with their corresponding DP and CE are presented in Table 1.

Chromatographic separation

The main factors affecting chromatographic separation were assessed, including UPLC column, eluent, additives (e.g., formic acid, ammonium acetate, acetic acid, and ammonium hydroxide), elution gradient, flow rate, and column temperature. A CORTECS™ UPLC® C18 column (2.1 mm × 100 mm, 1.6 μm, Waters) was selected based on its performance in achieving good resolution and peak morphology for the analytes within a short runtime.

Mobile phase composition (organic modifier and additives) was evaluated to obtain higher sensitivity and separation efficiency for the analytes. As shown in Fig. 1a, the peak areas of the 26 mycotoxins are differently affected by the mobile phase composition. Aflatoxins had similar peak areas in acid, neutral and alkaline conditions. For T2, HT2, SMC, and DAS, the peak areas slightly increased in 0.1% formic acid, but dropped significantly in 0.1% aqueous ammonia. For DON, ZEN and their derivatives, the peak areas decreased apparently in both the acid and alkaline media. Therefore, mobile phase without any additives was chosen for group A.

Fig. 1

Evaluation of the effects of additives in the mobile phase on the peak areas for (a) 26 analytes of group A, (b) 5 analytes of group B and (c) 12 analytes of group C (n = 3), with MON, PAT, ZEN, NIV, Fus X, DON, 3A-DON, 15A-DON, T2, HT2, CIT, CPA, TeA, AME, AOH, ALT, and TEN at 100 ng mL−1; NEO, DAS, DON-3-G, DOM-1, and FBs at 50 ng mL−1; ZAN, α-ZOL, α-ZAL, β-ZOL, β-ZAL, SMC, OTA, OTB, ENNs, and BEA at 10 ng mL−1; AFB1 and AFG1 at 2 ng mL−1; AFB2, AFM1, AFM2, and AFG2 at 0.5 ng mL−1. Abbreviations: HCOOH, formic acid; CH3COOH, acetic acid; CH3COONH4, ammonium acetate; NH4OH, ammonia water solution; CH3CN, acetonitrile; CH3OH, methanol.

The organic modifier (methanol or acetonitrile) in the mobile phase markedly affected chromatographic separation of analytes in group A. When a single organic modifier was used, adequate separation was hardly achieved for AFM1 and AFM2, ZAN and α-ZEL, and 3A-DON and 15A-DON, due to the high similarity of their structures and properties. Particularly, 3A-DON and 15A-DON as positional isomers presented a common precursor ion (m/z 337) and similar product ions. Complete separation was necessary to avoid peak overlapping and achieve accurate quantification of the two compounds. Different proportions of methanol and acetonitrile were tested for their separation efficiencies. By selecting methanol/acetonitrile (1:1 v/v) as the organic mobile phase, the satisfactory separation of the 26 mycotoxins in group A was achieved. Meanwhile, the isocratic elution at 11% B (3–12 min) and 28% (12.1–17 min), and a mild gradient elution program of 42–48% B (19–26 min) were essential to give optimal separation for DON derivatives, aflatoxins, and ZEN derivatives, respectively. MON and PAT are highly polar molecules having low interaction with the hydrophobic C18 column. Therefore, a gradient elution started from 95% water was applied to get acceptable retention of MON and PAT. A representative chromatogram of a standard mixture is shown in Fig. 2a.

Fig. 2

Extracted ion chromatograms of UPLC separation of the 43 mycotoxins. (a) 26 analytes of group A, (b) 5 analytes of group B, (c) 12 analytes of group C. The concentration of each analyte is the same as in Fig. 1.

For group B, formic acid greatly enhanced the sensitivity of ochratoxins and fumonisins (Table 1b), because it generates highly abundant hydrogen ions to assist positive ionization. Fortunately, the peak shapes of fumonisins were also improved in an acidic mobile phase. Acetonitrile was preferred as the organic modifier, due to the reduced background signals and higher elution ability compared with methanol. The complete separation of the 5 mycotoxins could be easily achieved with a mobile phase consisting of 0.1% formic acid (A) and acetonitrile (B) under a gradient elution (Fig. 2b).

Methods for detecting emerging mycotoxins, including Alternaria toxins, enniatins, and BEA have been optimized in our previous work [23]. Based on our study, inappropriate pH of the mobile phase and the use of methanol instead of acetonitrile can cause two peaks for TeA and peak splitting for AOH and ALT. Ammonia acetate as a modifier can result in the best peak shapes. In addition, additives in the mobile phase significantly affected the sensitivity of Alternaria toxins (Fig. 1c). Ultimately, 0.01% aqueous ammonia with 5 mmol L-1 ammonium acetate (solvent A) and acetonitrile (solvent B) were used as the mobile phase, which markedly enhanced the signals of TeA and AOH [23]. In addition, although CIT and CPA showed tailing peaks under neutral and acidic conditions, sharp peaks and good separation were obtained using the same chromatographic conditions as the emerging mycotoxins (Fig. 2c). Consequently, CIT and CPA were assigned to the group C.

Different column temperatures were tested, including 30, 40, and 50 °C. With the increase in column temperature, the analysis time decreased, and the separation efficiency and peak shape were improved. Therefore, 50 °C was selected for further experiments.

Different flow rates were also tested. No significant difference in sensitivity was observed when the flow rates were 0.3 mL min−1 and 0.4 mL min−1. Further increase in flow rate caused high back-pressure. Therefore, a flow rate of 0.4 mL min−1 was selected, as the separation could be completed in a shorter time.

The sensitivity of the compounds increased with the increase of injection volume from 1 to 5 µL. Greater volume caused peak broadening because of sample dispersion and column saturation.

Sample preparation

Extraction and cleanup are critical steps in separating multi-mycotoxins, notably in complex food matrices, which comprise proteins, fats, pigments and carbohydrates with very different compositions.

The optimal extraction solvents were selected also according to the solubility of mycotoxins and the solvent recommended by the manufacturer of the cartridges. Water, methanol, acetonitrile, and their combinations were tested, and typical extraction solvents in the literature were also considered. For the 26 mycotoxins in group A, 84:16 (v/v) acetonitrile aqueous solution was the optimal extraction solvent. It has been widely used for the extraction of trichothecenes [30], [31], zearalenone and its metabolites [32], and multi-mycotoxins [33]. Moreover, Malone et al. studied the extraction efficiency for mycotoxins in naturally contaminated commodities in detail [34], concluding that acetonitrile/water (84:16) was the most efficient solvent for aflatoxins extraction, and that 50%–80% organic solvent had similar efficiencies for extractions of OTA and fumonisins. Acetonitrile/methanol/water (45:10:45 v/v/v) was the optimal extraction solvent for emerging mycotoxins, and was also used to extract Alternaria toxins from vegetables [35], [36]. To improve the recoveries of compounds possessing carboxyl groups, such as TeA and CPA, pH 3.0 NaH2PO4 was added to an acetonitrile/methanol/water (45:10:45 v/v/v) solution as the extraction solvent.

Solid-phase extraction (SPE) are commonly employed for reducing the matrix effect and obtaining a satisfactory recovery. Four cartridges, including MycoSep 226, MultiSep 211 Fum, Oasis HLB, and Oasis C18, were assessed using the mixed standard solution of target compounds. MycoSep and MultiSep cartridges represent multi-functional columns comprising adsorbents specifically designed for mycotoxin purification. Fig. 3 depicts the recoveries by using various cartridges for 43 compounds.

Fig. 3

Recovery by using Mycosep226 cartridge, Multisep 211Fum cartridge, Oasis HLB SPE column for 43 mycotoxins (n = 3). The concentration of each analyte is the same as in Fig. 1.

The MycoSep 226 AflaZon+ cartridge used in this study is a one-step clean-up cartridge that retain interfering substances from complex samples and let target compounds through. According to the operating instructions, sample extract in acetonitrile/water (84:16) can be directly loaded onto the cartridge without any preconditioning before use. It is quite convenient for mycotoxin determination. After evaluation, 26 mycotoxins (group A) were found to have good recoveries using this cartridge.

Ochratoxins and fumonisins contain carboxyl groups, with strong water solubility and sensitivity to pH changes. Therefore, the clean-up of these toxins is mainly based on ion-exchange mechanism. MultiSep 211 Fum cartridge packed with mixed sorbent materials including ion-exchange resin exhibits strong retention of these compounds. According to the operating instructions, the loading buffer was Methanol/Water (3:1, pH 6–9). Under this condition, ochratoxins and fumonisins existed in ionic form and could be well retained in the cartridge. A wash step with the same buffer (Methanol/Water, 3:1) was applied to remove sample matrices and interferences. Methanol containing 0.1% formic acid was used as eluting solvent. Under the acidic condition, these toxins changed into neutral form and thereby could be eluted from the cartridge and collected for further analysis.

CIT and CPA are acidic mycotoxins with strong polarity and are easily adsorbed by the MycoSep 226 cartridge, thereby resulting in poor recoveries. However, improved recovery was obtained for CIT and CPA using the Oasis HLB column. The 10 emerging mycotoxins (Alternaria mycotoxins, enniatins and beauvericin) have been optimally assessed in our previous work [23]. The performances of MycoSep 226, Oasis C18, and Oasis HLB were comparatively assessed, as well as their abilities to enrich these 10 analytes. As shown in Fig. 3, Oasis HLB column yielded improved recoveries and was selected for clean-up of the 12 mycotoxins in group C.

Different centrifugation speeds (5000 rpm, 7000 rpm, 9000 rpm) and durations (10 min, 15 min) were tested during sample extraction. A speed of 7000 rpm was not enough to completely separate the food matrix, especially for foods containing large amounts of fat and oil. After optimization, centrifugation at 9000 rpm for 10 min was selected to get a clear supernatant in a shorter time.

Filtration with a 0.22 μm filter is commonly required to remove the particles in tested sample solution before UHPLC-MS/MS analysis. However, some mycotoxins can partially bind to several types of filter membrane. Alternatively, a high-speed centrifugation at 20000 rpm for 30 min was applied prior to injection into the instrument.

Method validation

As shown in Fig. 4, there were no peak interferences at the same retention times and m/z channels of the 43 target compounds, which indicated the absence of interfering endogenous substances, as well as good selectivity for the developed method. Carry-over experiments were conducted by injecting blank samples after a high concentration standard, and no sample-to-sample carryover was observed.

Fig. 4

LC-MS/MS extracted ion chromatograms of blank food samples and food matrices fortified with (a) AFB1 and AFG1 at 0.4 ng mL−1, AFB2, AFM1, AFM2, and AFG2 at 0.1 ng mL−1, ZEN and MON at 20 ng mL−1, β-ZAL, β-ZOL, ZAN, α-ZAL, α-ZOL, and SMC at 2 ng mL−1, NEO at 10 ng mL−1; (b) DON, DON-3-G, 15A-DON, 3A-DON, Fus X, T2, HT2, PAT, CIT, and CPA at 20 ng mL−1, DOM-1 and DAS at 10 ng mL−1, NIV, OTA, and OTB at 2 ng mL−1; and (c) FBs at 10 ng mL−1, AME, AOH, ALT, TeA, and TEN at 20 ng mL−1, ENNs and BEA at 2 ng mL−1.

The LOD and LOQ of each analyte were measured first with the standard solution, giving the results as shown in Table S1. In addition, the LOD and LOQ were also evaluated with blank samples spiked at low amounts, and subjected to the whole analytical procedure. The results are shown in Table 2. In general, satisfactory values were achieved for all the 43 mycotoxins in the 12 food matrices. LOQs ranged between 0.006 μg kg−1 (AFB1, AFB2, and SMC) and 3 μg kg−1 (PAT). LODs between 0.002 μg kg−1 (AFB1, AFB2, and SMC) and 1 μg kg−1 (PAT).

Table 2

Matrix effect, apparent recoveries, LODs and LOQs of the 43 mycotoxins. Matrix effect and apparent recoveries are measured without internal standard correction. LODs and LOQs are measured using spiked blank samples in low amounts.

Analyte	Creals and their products				Legume and their products
	Matrix Effect (%)	RA (Apparent recovery, %)	LOQ (μg kg−1)	LOD (μg kg−1)	Matrix Effect (%)	RA (Apparent recovery, %)	LOQ (μg kg−1)	LOD (μg kg−1)
15A-DON	92.0	82.2	0.3	0.1	82.6	83.0	0.3	0.1
3A-DON	87.3	66.1	0.6	0.2	66.5	66.8	0.6	0.2
AFB1	99.1	62.8	0.006	0.002	90.6	75.0	0.006	0.002
AFB2	67.9	56.7	0.006	0.002	62.4	54.0	0.006	0.002
AFG1	87.0	61.7	0.01	0.004	77.8	62.9	0.01	0.004
AFG2	85.9	60.2	0.01	0.004	68.5	60.0	0.01	0.004
AFM1	102.7	47.1	0.01	0.004	60.0	51.6	0.01	0.004
AFM2	75.8	50.0	0.01	0.004	75.2	51.4	0.01	0.004
ALT	105.1	90.1	0.6	0.2	101.3	77.3	0.6	0.2
AME	117.0	79.6	0.06	0.02	71.4	55.8	0.06	0.02
AOH	108.3	58.2	0.6	0.2	73.9	51.5	0.6	0.2
BEA	127.3	87.6	0.06	0.02	162.8	107.3	0.06	0.02
CIT	78.3	67.0	0.1	0.04	103.3	77.2	0.1	0.04
CPA	101.9	78.6	0.06	0.02	86.8	63.6	0.06	0.02
DAS	78.3	66.1	0.1	0.04	77.1	78.8	0.1	0.04
DOM-1	65.2	75.3	0.6	0.2	50.6	43.8	0.6	0.2
DON	150.2	61.4	0.6	0.2	43.6	58.3	0.6	0.2
DON-3-G	69.5	70.3	0.3	0.1	63.1	68.9	0.3	0.1
ENNA	127.1	80.6	0.06	0.02	145.1	95.6	0.06	0.02
ENNA1	188.2	131.6	0.06	0.02	144.1	91.4	0.06	0.02
ENNB	164.1	111.3	0.01	0.004	166.1	89.7	0.01	0.004
ENNB1	156.2	139.0	0.06	0.02	177.8	102.8	0.06	0.02
FB1	113.7	87.1	0.03	0.01	68.9	113.6	0.03	0.01
FB2	99.3	81.0	0.06	0.02	99.5	94.5	0.06	0.02
FB3	105.5	89.2	0.06	0.02	80.5	101.5	0.06	0.02
Fus X	82.3	70.6	0.6	0.2	81.1	69.5	0.6	0.2
HT2	76.4	70.9	0.2	0.08	63.3	65.7	0.2	0.08
MON	58.8	56.7	2.0	0.8	57.2	67.2	2.0	0.8
NEO	84.2	67.9	0.06	0.02	71.4	70.0	0.06	0.02
NIV	42.4	87.2	0.3	0.1	42.6	79.0	0.3	0.1
OTA	100.2	68.9	0.01	0.004	101.7	82.8	0.01	0.004
OTB	95.8	67.5	0.01	0.004	115.6	80.6	0.01	0.004
PAT	42.2	68.5	3.0	1.0	49.8	80.3	3.0	1.0
SMC	85.2	50.1	0.006	0.002	85.4	73.2	0.006	0.002
T2	66.6	63.0	0.1	0.04	81.9	73.0	0.1	0.04
TeA	122.0	96.1	0.6	0.2	90.7	81.1	0.6	0.2
TEN	99.7	95	0.1	0.05	121.4	94.9	0.1	0.05
ZAN	148.3	116.2	0.06	0.02	126.2	70.0	0.06	0.02
ZEN	105.0	69.1	0.06	0.02	92.3	66.3	0.06	0.02
α-ZAL	96.2	76.0	0.06	0.02	91.5	59.0	0.06	0.02
α-ZOL	68.9	63.9	0.03	0.01	68.6	68.1	0.03	0.01
β-ZAL	84.1	79.2	0.03	0.01	84.9	70.3	0.03	0.01
β-ZOL	95.0	76.5	0.1	0.04	90.6	66.8	0.1	0.04

Linearity was assessed with at least six concentration points for each analyte on three consecutive days (Table S1). The regression coefficients (R2) of calibration curves were in the range of 0.9902–0.9995, except for BEA which had a R2 of 0.9778. These findings indicated that the method had good linearity for the totality of analytes.

Matrix effects (ME) and apparent recovery (RA) were also investigated. Because of matrix complexity, ME values ranged from 32.5% to 196.7%, and RA ranged from 32.1% to 192.6% (Table 2). These results demonstrated that internal standard compensation is required to effectively analyze the compounds. Therefore, internal standards with comparable RA values were chosen as reference internal standards for analytes without commercially available internal standards.

The accuracy and precision of the method were also examined by assessing blank food samples at three levels on 3 distinct days with six replicates performed per day. Accuracy ranged from 60.3% to 175.9%; intra- and inter-day precisions (RSD) were 0.2%–12.2% and 2.1%–13.9%, respectively (Table 3).

Table 3

Accuracy and precision data for determination of 43 mycotoxins at three levels in one day (n = 6) and three distinct days (n = 18). Method recoveries are corrected by internal standards.

Analyte	Spiked level(μg kg−1)	Cereals and their products				Legume and their products
		Measured value (μg kg−1)	RM (Method recovery, %)	RSD (%)		Measured value (μg kg−1)	RM (Methodrecovery, %)	RSD (%)
				Intra-day (n = 6)	Inter-day (n = 18)			Intra-day (n = 6)	Inter-day (n = 18)
15-ADON	2	2.33	116.8	3.2	2.6	2.22	110.9	4.2	6.5
	20	23.2	116.0	1.8	3.8	22.23	111.1	1.6	3.3
	200	233	116.5	2.7	6.5	252.9	126.4	4.3	6.9
3A-DON	2	1.96	97.8	1.4	6.2	2.07	103.4	6.0	4.5
	20	19.5	97.5	2.9	10.2	18.73	93.6	2.8	5.1
	200	183.4	91.7	4.8	2.8	186.8	93.4	10.5	6.9
AFB1	0.04	0.038	94.9	1.2	3.9	0.038	94.8	5.0	8.1
	0.4	0.37	92.7	3.1	6.7	0.39	96.7	1.8	6.2
	4	3.68	92.0	5.4	10.1	3.71	92.7	4.7	7.2
AFB2	0.01	0.009	91.0	1.6	3.6	0.009	92.8	8.2	10.9
	0.1	0.09	91.3	5.3	10.2	0.09	90.6	3.9	6.1
	1	0.94	94.9	5.1	6.4	0.97	97.4	8.3	6.3
AFG1	0.04	0.038	95.0	9.4	10.1	0.04	91.1	6.0	8.1
	0.4	0.37	91.3	6.0	7.5	0.38	95.3	2.8	4.3
	4	3.81	95.3	6.6	7.8	3.71	92.8	10.5	6.8
AFG2	0.01	0.01	100.0	7.5	9.2	0.007	69.7	7.9	6.1
	0.1	0.096	95.7	9.1	10.3	0.09	87.0	6.0	4.5
	1	0.96	96.1	1.9	5.7	0.94	94.4	5.5	6.9
AFM1	0.01	0.01	100.4	6.6	6.2	0.008	89.1	8.0	9.1
	0.1	0.095	95.5	6.0	7.6	0.09	91.8	9.9	5.5
	1	0.98	98.1	4.7	6.9	0.98	98.7	5.4	6.2
AFM2	0.01	0.011	110.1	10.4	11.1	0.008	80.0	7.3	9.1
	0.1	0.11	107.5	4.6	6.1	0.095	95.0	8.7	6.7
	1	1.06	106.4	2.6	3.8	0.94	94.2	10.8	11.2
ALT	2	2.25	112.4	2.4	5.6	2.28	114.0	1.6	5.8
	20	25.26	126.3	3.8	6.3	17.26	86.3	1.6	6.6
	200	257.2	128.6	4.2	6.8	156.6	78.3	6.5	7.8
AME	2	1.26	62.8	2.9	4.2	1.66	83.0	11.8	12.2
	20	14.47	72.3	1.2	5.2	14.79	74.0	4.8	6.6
	200	110	94.0	2.4	5.2	131.0	65.5	7.2	8.2
AOH	2	1.77	88.5	5.2	11.2	2.78	139.1	4.1	6.2
	20	14.72	73.6	8.5	10.9	13.65	68.3	11.1	12.3
	200	20.02	100.1	1.6	4.2	151.8	75.9	4.7	6.8
BEA	0.2	0.3	149.1	2.8	4.8	0.25	123.9	9.6	11.2
	2	1.96	97.8	2.5	4.1	1.37	68.4	7.7	9.6
	20	14.07	70.4	3.6	6.2	12.31	61.5	5.6	6.2
CIT	2	1.46	72.8	4.7	7.2	1.69	84.5	3.8	7.5
	20	17.96	89.8	5.5	8.3	15.79	79.0	4.7	8.1
	200	180.2	90.0	3.4	6.2	163.9	82.0	5.6	9.5
CPA	2	1.78	89.1	3.6	6.2	1.64	82.4	5.1	8.5
	20	16.54	82.7	2.5	4.3	17.66	88.3	3.2	5.9
	200	173.8	86.9	5.2	6.3	176.2	88.1	4.6	8.2
DAS	1	0.95	94.8	7.3	6.9	1.03	102.6	5.2	6.9
	10	9.39	93.9	4.5	6.1	9.07	90.7	6.5	8.3
	100	94.73	94.7	2.8	3.6	103.7	103.7	7.6	8.9
DOM-1	1	1.16	116.0	4.3	8.1	1.13	112.8	13.2	10.2
	10	11.85	118.5	1.1	2.6	11.11	111.1	3.9	5.3
	100	111.3	111.3	2.4	8.4	114.9	114.9	3.3	6.2
DON	2	1.97	98.5	2.0	2.5	1.95	97.5	2.0	2.6
	20	20.08	100.4	4.4	6.8	18.75	93.7	1.8	3.8
	200	201.9	101	1.0	7.5	193.3	96.7	4.7	5.5
DON-3-G	1	1.07	107.1	10.0	11.5	0.88	88.3	5.8	6.2
	10	10.26	102.6	5.4	6.9	10.27	102.7	7.6	10.3
	100	108.6	108.6	3.2	4.1	107.4	107.4	4.2	6.5
ENNA	0.2	0.33	164.4	6.5	8.8	0.13	64.3	5.4	9.5
	2	2.57	128.6	4.5	5.2	1.23	61.6	2.7	6.4
	20	18.72	93.6	10.1	12.0	13.96	69.8	9.9	12.1
ENNA1	0.2	0.31	154.3	3.3	6.3	0.17	84.0	8.1	9.8
	2	2.05	102.4	2.6	7.8	1.41	70.2	3.5	3.9
	20	20.86	104.3	2.3	5.6	12.24	61.2	4.6	5.6
ENNB	0.2	0.28	138.8	2.6	4.2	0.15	72.8	4.2	9.5
	2	2.43	121.5	3.5	8.3	1.48	73.9	7.5	7.9
	20	25.37	126.9	5.8	8.5	12.16	60.8	1.9	2.3
ENNB1	0.2	0.24	121.9	4.2	8.8	0.15	72.7	1.1	3.8
	2	2.2	110.0	9.1	10.2	1.55	77.5	1.7	3.6
	20	20.36	101.8	7.4	10.6	12.94	64.7	3.2	5.6
FB1	1	0.86	85.9	5.4	7.3	0.86	85.7	4.3	2.5
	10	9.19	91.9	3.9	4.9	9.72	97.2	7.4	6.6
	100	98.19	98.2	9.7	5.6	91.46	91.5	5.7	7.3
FB2	1	0.86	86.3	1.2	4.3	1.04	104.3	2.5	5.2
	10	10.24	102.0	1.3	5.2	9.13	91.3	3.3	4.1
	100	97.91	97.9	1.9	2.8	103.8	103.8	2.4	3.9
FB3	1	1.06	106.2	1.1	3.9	1.05	104.5	2.1	7.1
	10	11.62	116.2	4.2	7.6	9.73	97.3	0.2	2.6
	100	98.42	98.4	3.2	4.8	103.4	103.4	2.4	4.3
Fus X	2	2.08	103.9	3.6	2.5	1.92	95.8	4.3	6.2
	20	20.06	100.3	4.5	5.8	18.74	93.7	2.3	7.3
	200	199.8	99.9	1.7	9.6	209	104.5	2.1	5.1
HT2	2	2.22	110.8	10	11.2	2.04	102.1	5.3	8.9
	20	20.22	101.1	1.4	2.4	20.75	103.8	3.5	6.1
	200	184.2	92.1	2.6	3.1	185.3	92.6	3.8	6.5
MON	2	1.79	89.5	8.2	10.8	2.06	103.5	10.2	11.8
	20	17.78	88.9	1.5	6.3	22.66	113.3	7.6	8.1
	200	183.2	91.6	7.7	6.1	187	93.5	10.0	6.5
NEO	1	0.95	95.3	0.7	2.9	0.97	96.9	4.8	5.1
	10	9.89	98.9	1.4	5.1	8.55	85.5	2.9	3.6
	100	100.3	100.3	3.2	4.5	94.57	94.6	3.3	4.9
NIV	2	2.044	102.0	5.6	6.8	1.74	87.0	8.4	7.4
	20	19.2	96.1	4.4	6.9	17.7	88.5	7.4	10.1
	200	182.5	91.3	2.1	7.2	181.1	90.6	5.9	6.8
OTA	0.2	0.18	92.6	1.1	2.5	0.2	100.8	2.0	3.5
	2	1.96	94.6	3.5	4.9	1.88	94.0	1.1	4.2
	20	18.12	94.2	3.8	5	19.05	95.3	4.2	8.3
OTB	0.2	0.17	85.0	0.7	2.6	0.18	90	2.9	6.3
	2	1.81	92.3	0.2	3.6	2.02	101	7.6	8.0.2
	20	18.17	90.2	2	8.2	19.87	99.4	1.1	2.1
PAT	2	1. 92	96.0	8.6	10.2	1.86	93.2	2.5	5.1
	20	19.05	95.3	11.8	12.5	18.96	94.8	4.0	10.8
	200	188.5	94.2	11.6	12.9	182.0	92.0	6.0	11.3
SMC	0.2	0.18	87.8	7.8	8.1	0.18	90.0	11.4	10.2
	2	1.92	96.2	5.4	6.3	1.92	96.1	4.7	6.5
	20	17.65	88.2	7.2	9.1	18.81	94.1	3.6	8.3
T2	2	2.16	108	9.2	10.2	2.26	113.1	1.2	3.3
	20	21.82	109.1	9.4	10.5	22.35	111.8	2.2	4.1
	200	214.6	107.3	9.8	11.8	214.4	107.2	2.7	3.5
TeA	2	2.45	122.5	3.4	5.6	1.93	96.3	4.3	6.5
	20	18.38	91.9	1.5	3.3	18.6	93.0	1.5	3.2
	200	180.6	90.3	0.9	3.2	176.5	88.2	5.2	6.6
TEN	2	2.49	124.6	5.9	10.8	2.02	101.1	9.5	11.8
	20	20.48	102.4	2.0	5.3	19.18	95.9	2.9	5.6
	200	184.8	92.4	1.9	3.4	168.8	84.4	1.9	2.3
ZAN	0.2	0.2	102	4.0	4.3	0.2	100.5	7.7	6.1
	2	1.96	98.0	2.6	4.9	2.04	101.8	2.1	3.5
	20	20.52	102.6	3.1	5.1	20.5	102.5	3.3	4.2
ZEN	2	2.02	101.0	3.4	5.2	1.94	97.4	7.2	8.2
	20	19.11	95.5	3.5	7.3	18.72	93.6	0.5	3.8
	200	199.3	99.1	4.2	6.8	198.9	99.5	2.8	4.5
α-ZAL	0.2	0.23	117.6	2.5	3.5	0.17	85.0	3.7	4.9
	2	2.15	107.4	2.5	4.9	1.63	81.5	5.4	8.5
	20	21.71	108.6	1.4	2.6	17.96	89.8	1.2	3.6
α-ZOL	0.2	0.16	80.0	10.2	11.6	0.18	90.6	6.5	8.1
	2	1.88	94.0	4.8	6.9	1.96	98.0	5.4	6.3
	20	18.6	93.0	2.2	3.5	19.01	95.1	1.2	3.8
β-ZAL	0.2	0.23	114.0	4.3	2.8	0.19	95.0	11.3	10.9
	2	2.21	110.4	5.2	8.2	2.03	101.5	2.4	3.6
	20	22.49	112.4	6.0	6.4	21.01	105.1	11.6	10.1
β-ZOL	0.2	0.22	110.0	6.9	8.2	0.21	105.0	9.2	11.2
	2	2.22	111.0	2.6	5.6	2.02	101.2	5.5	6.3
	20	23.23	116.2	3.1	6.8	20.19	100.9	6.1	5.1

Application to dietary samples

For practical application, the developed methods were utilized for detecting 43 mycotoxins in dietary samples collected in six provinces (Hebei, Beijing, Jilin, Hubei, Guangdong, and Guizhou) in the 6th China TDS. The levels of the 43 mycotoxins were obtained, and the results are presented in Table 4. Representative MRM chromatograms of naturally polluted samples are shown in Fig. 5. Among 72 dietary samples, 60 (83.3%) contained at least one mycotoxin. Overall, mycotoxin contamination varied significantly among food categories. ZEN (45.8%), FB1 (30.6%), DON (26.4%), AFB1 (22.2%), and emerging mycotoxins had high rates of detection, with values above 20%. Meanwhile, 15A-DON, DOM-1, AFM1, AFM2, AFG2, OTA, AOH, MON, PAT, T2, HT2, DAS, NEO, CPA and CIT were not detected in any samples. Cereals, legumes, and their respective products were the main contaminated food types. In contrast, sugar, beverages and water were barely contaminated.

Table 4

The occurrence of 43 mycotoxins in 72 food samples from 12 food categories collected in six provinces for the 6th China TDS.

Food composites		Cereals	Legume	Potatoes	Meats	Eggs	Aquatic foods	Dairy products	Vegetables	Fruits	Sugar	Beverages and water	Alcohol beverages
15A-DON	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
3A-DON	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	1/6	0/6	0/6	0/6	1/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	1.65	ND	ND	ND	2.74	ND	ND	ND	ND
AFB1	Positive samples/ Samples analyzed (n/n)	3/6	3/6	1/6	3/6	1/6	3/6	0/6	2/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	0.02–0.07	0.11–1.38	0.04	0.02–0.05	0.05	0.02–0.08	ND	0.04–0.07	ND	ND	ND	ND
AFB2	Positive samples/ Samples analyzed (n/n)	0/6	2/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	0.03–0.21	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
AFG1	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	1/6	0/6	0/6	0/6	0/6	1/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	0.02	ND	ND	ND	ND	0.02
AFG2	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
AFM1	Positive samples/ Samples analyzed (n/n)	0/6	2/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	0.02–0.03	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
AFM2	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
ALT	Positive samples/ Samples analyzed (n/n)	1/6	0/6	0/6	0/6	0/6	0/6	0/6	1/6	1/6	0/6	0/6	0/6
	Range (μg kg−1)	7.90	ND	ND	ND	ND	ND	ND	8.76	1.71	ND	ND	ND
AME	Positive samples/ Samples analyzed (n/n)	6/6	2/6	3/6	0/6	1/6	2/6	0/6	2/6	0/6	1/6	1/6	2/6
	Range (μg kg−1)	0.14–10.53	0.23–1.77	0.08–2.41	ND	0.72	0.17–0.38	ND	1.00–1.10	ND	0.12	0.16	0.27–1.00
AOH	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
BEA	Positive samples/ Samples analyzed (n/n)	4/6	5/6	6/6	3/6	4/6	4/6	3/6	2/6	1/6	0/6	0/6	1/6
	Range (μg kg−1)	0.22–1.77	0.17–5.46	0.19–2.58	0.16–1.07	0.27–6.70	0.32–1.02	0.16–6.31	1.85–1.93	1.21	ND	ND	4.38
CIT	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0	0	0	0	0	0	0	0
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
CPA	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
DAS	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
DOM-1	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0	0	0	0	0	0
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
DON	Positive samples/ Samples analyzed (n/n)	6/6	2/6	3/6	1/6	0/6	2/6	0/6	1/6	0/6	0/6	0/6	4/6
	Range (μg kg−1)	4.22–75.56	2.13–4.65	1.67–22.52	2.51	ND	0.94–1.84	ND	0.71	ND	ND	ND	0.73–11.79
DON-3-G	Positive samples/ Samples analyzed (n/n)	2/6	0/6	0/6	1/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	1/6
	Range (μg kg−1)	5.04–12.45	ND	ND	0.45	ND	ND	ND	ND	ND	ND	ND	3.33
ENNA	Positive samples/ Samples analyzed (n/n)	2/6	3/6	3/6	1/6	3/6	2/6	2/6	0/6	0/6	0/6	0/6	1/6
	Range (μg kg−1)	0.12–0.16	0.26–1.26	0.17–1.13	0.24	0.38–1.55	0.19–0.30	0.15–1.38	ND	ND	ND	ND	0.16
ENNA1	Positive samples/ Samples analyzed (n/n)	2/6	3/6	2/6	1/6	3/6	2/6	1/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	0.10–0.22	0.22–0.29	0.15–0.18	0.23	0.24–0.51	0.18–0.21	0.25	ND	ND	ND	ND	ND
ENNB	Positive samples/ Samples analyzed (n/n)	5/6	6/6	5/6	5/6	5/6	6/6	2/6	5/6	2/6	0/6	0/6	1/6
	Range (μg kg−1)	0.21–0.69	0.05–2.10	0.21–2.38	0.05–3.45	0.07–3.75	0.07–4.88	0.02–0.09	0.11–0.29	0.05–0.83	ND	ND	0.13
ENNB1	Positive samples/ Samples analyzed (n/n)	2/6	4/6	5/6	2/6	4/6	3/6	1/6	2/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	0.21–0.35	0.14–0.73	0.12–0.53	0.11–0.65	0.09–0.87	0.22–0.89	0.87	0.15–0.26	ND	ND	ND	ND
FB1	Positive samples/ Samples analyzed (n/n)	5/6	4/6	4/6	3/6	1/6	2/6	0/6	0/6	0/6	0/6	0/6	3/6
	Range (μg kg−1)	0.06–7.54	0.13–1.72	0.07–0.58	0.13–2.08	0.09	0.11–0.30	ND	ND	ND	ND	ND	0.22–2.16
FB2	Positive samples/ Samples analyzed (n/n)	2/6	0/6	1/6	1/6	1/6	1/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	0.90–4.39	ND	0.22	0.14	0.10	0.80	ND	ND	ND	ND	ND	ND
FB3	Positive samples/ Samples analyzed (n/n)	2/6	2/6	1/6	1/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	0.85–9.24	0.14–0.17	0.23	0.46	ND	ND	ND	ND	ND	ND	ND	ND
Fus X	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	1/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	6.92	ND	ND
HT2	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
MON	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
NEO	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
NIV	Positive samples/ Samples analyzed (n/n)	1/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	4.84	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
OTA	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
OTB	Positive samples/ Samples analyzed (n/n)	0/6	2/6	0/6	0/6	0/6	0/6	0/6	1/6	1/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	0.05–0.12	ND	ND	ND	ND	ND	0.09	0.25	ND	ND	ND
PAT	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
SMC	Positive samples/ Samples analyzed (n/n)	5/6	3/6	4/6	3/6	1/6	1/6	0/6	4/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	0.02–0.04	0.05–1.87	0.10–0.13	0.03–0.09	0.07	0.04	ND	0.09–0.17	ND	ND	ND	ND
T2	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
TeA	Positive samples/ Samples analyzed (n/n)	6/6	1/6	0/6	1/6	1/6	2/6	2/6	1/6	0/6	0/6	0/6	4/6
	Range (μg kg−1)	2.17–15.17	0.86	ND	1.29	14.38	0.68–1.57	1.54–2.06	1.14	ND	ND	ND	2.45–17.62
TEN	Positive samples/ Samples analyzed (n/n)	4/6	2/6	2/6	1/6	2/6	2/6	0/6	3/6	1/6	1/6	1/6	3/6
	Range (μg kg−1)	0.73–7.70	0.28–0.43	0.15–0.35	0.32	0.15–0.26	0.66–0.68	ND	0.18–0.83	0.16	0.12	0.16	0.16–1.18
ZAN	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
ZEN	Positive samples/ Samples analyzed (n/n)	5/6	6/6	6/6	5/6	3/6	4/6	0/6	4/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	0.27–0.83	0.16–0.51	0.16–1.02	0.12–1.23	0.18–0.83	0.11–1.99	ND	0.23–0.99	ND	ND	ND	ND
α-ZAL	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
α-ZOL	Positive samples/ Samples analyzed (n/n)	1/6	0/6	0/6	1/6	0/6	0/6	0/6	1/6	1/6	0/6	0/6	0/6
	Range (μg kg−1)	0.07	ND	ND	0.06	ND	ND	ND	1.62	1.23	ND	ND	ND
β-ZAL	Positive samples/ Samples analyzed (n/n)	0/6	0/6	0/6	0/6	0/6	0/6	0/6	0/6	4/6	0/6	0/6	0/6
	Range (μg kg−1)	ND	ND	ND	ND	ND	ND	ND	ND	0.09–1.05	ND	ND	ND
β-ZOL	Positive samples/ Samples analyzed (n/n)	0/6	1/6	0/6	2/6	0/6	1/6	0/6	0/6	0/6	0/6	0/6	1/6
	Range (μg kg−1)	ND	0.17	ND	0.92–0.92	ND	0.93	ND	ND	ND	ND	ND	0.28

ND: level below LOD; positive sample refers to a sample with a result above the LOD.

Fig. 5

Representative MRM-chromatograms of food samples with natural contaminations of mycotoxins.

Among ZEN and derivatives, ZEN showed the highest detection rate of 45.8%, followed by β-ZOL (6.9%), β-ZAL (5.6%), and α-ZOL (5.6%). ZAN and α-ZAL were not detected. ZEN was most frequently detected in cereals, legumes, potatoes, meats, eggs, aquatic foods and vegetables, with more than half of the samples testing positive.

Among DON and its derivatives, the detection rate of DON was 26.4%, with amounts ranging from 0.71 μg kg−1 to75.56 μg kg−1. Cereals, legumes, potatoes, meats, aquatic foods, and alcohols, all showed mycotoxin presence, but the rates of positive samples were quite different among food categories. Cereals was the highest, with all the samples testing positive. DON contamination was greatly affected by climate, especially in the hot-humid area and the middle-lower reach of the Yangtze River, where the rain season was conducive to growth of mold and toxin production. The most elevated DON amounts were detected in cereals from Hubei at 75.56 μg kg−1. DON-3-G (5.6%) and 3A-DON (2.8%) were rarely detected. Fus X and NIV were found only in one sample, and 15A-DON and DOM-1 were not detected in any samples.

AFB1 was found in 22.2% of the tested samples, at levels ranging between 0.02 and 1.38 μg kg−1. Cereals, legumes, potatoes, meats, eggs, aquatic foods and vegetables, as well as their respective products, all showed aflatoxin contamination, at very low concentrations. AFB2 was detected only in legumes (2.8%). AFG1 was found in eggs and alcohols (2.8%). AFM1, AFM2, and AFG2 were not detected in any samples.

Fumonisins were frequently detected in cereals, legumes, potatoes, meats, eggs and aquatic foods, with detection rates of 30.6%, 8.3%, and 8.3% for FB1, FB2, and FB3, respectively. Among ochratoxins, OTB had a low rate of detection (5.6%), and OTA was not detected.

Emerging mycotoxins had high rates of detection, except AOH that was not detected. Enniatins occurred in more than half (58.3%) of the samples. BEA was found in 45.8% samples. Among Alternaria mycotoxins, detection rates were 30.6% for TEN, 27.8% for AME, 23.6% for TeA, and 4.2% for ALT. Cereals, legumes, potatoes and eggs showed the highest incidence rates, with most samples being positive.

Among the remaining mycotoxins, SMC was found positive in 29.2% of the samples, at concentrations below 2 μg kg−1. MON, PAT, T2, HT2, DAS, NEO, CPA, CIT were not detected.

Serious contamination occurred for given food categories from certain mycotoxins. Overall, more than 80% of the samples were found contaminated by mycotoxins. DON, SMC, FB1, ZEN, BEA, ENNB1, and ENNB were most detected. These findings indicate the major points for further investigation.

Comparison

Table 5 summarizes the mycotoxin analytical methods in TDSs carried out in different countries, such as France, the Netherlands, Spain, Lebanon, Canada, New Zealand, Vietnam, Ireland, regional Sub-Saharan Africa, and Hong Kong. The number of mycotoxins analyzed varies from one mycotoxin to 37 mycotoxins. The study presented here developed a sensitive, accurate, and robust method for detecting 43 mycotoxins in the 6th China TDS, and the number of mycotoxins was the most studied at once. Compared with the 4th and 5th China TDSs, 10 emerging mycotoxins (AOH, AME, TeA, ALT, TEN, BEA, ENNA1, ENNA, ENNB1 and ENNB) were added into the 6th China TDS for the first time. Among the TDSs conducted in other countries, ENNs were investigated only in the Netherlands, while ATs, BEA, and TEN remain unexplored in other countries.

Table 5

A summary of the mycotoxin analytical methods used in the different TDSs.

Country	Analyzed mycotoxins	Analyzed food	Food preparation	Analytical technique	Reference
France	AFB1, AFB2, AFG1, AFG2, AFM1, OTA, PAT, ZEN, FB1, FB2, DON, NIV,3AcDON, 15AcDON, T-2, T-2 triol, HT-2, NEO, FUS-X, DAS, MAS	Vegetarians food; biscuits; breakfast cereals; breads; pasta; rice; cakes; chocolates; desserts; nuts and oilseeds; vegetables; pulses; eggs; sugars; breads, buns; butter; dairy products; coffee; meat; offal; fruits; soft drinks; alcoholic beverages; pizzas, salt cakes, quiches; sandwiches; soup; prepared dishes; salads; compotes	AFBG, AFM1, OTA, ZEN, FB1,FB2: IAC;PAT: Sodium carbonate solution;Trichothecenes: Celite/carbon column.	AFBG, OTA,FB1,FB2: HPLC; PAT:HPLC-UV; Trichothecenes: GC–MS; AFM1, OTA, ZEN: HPLC-FD	[7]
France	AFB1, AFB2, AFG1, AFG2, AFM1, OTA, OTB, PAT, T-2, HT-2, NIV, DON, 3-Ac-DON, 15-Ac-DON, ZEN, α-ZAL, β-ZAL, α-ZOL, β-ZOL, FB1,FB2	Breads; breakfast cereals; pasta; rice; croissants; pastries; biscuits; cakes; milk; dairy products; eggs; butter; offal; delicatessen meat; vegetables; fruits; dried fruits; nuts and seeds; chocolate; non-alcoholic beverages; alcoholic beverages; coffee; pizzas; sandwiches; snacks; mixed dishes; desserts; compotes	AFM1: IAC;FB1, FB2, OTA, PAT, TCTs A and B, ZEA: extraction without purification	AFM1: IAC–LC–FD; FB1, FB2, OTA, PAT, TCTs A and B, ZEA: LC-MS/MS	[8], [9]
Netherlands	AFB1, AFB2, AFG1, AFG2, AFM1, AOH, AME, BEA, CIT, ENNA, ENNA1, ENNB, ENNB1, OTA, PAT, ZEN, α-ZOL, β-ZOL, STE, FB1, FB2, FB3, DON, DON-3G,FUS-X, NEO, DAS, NIV, 3A-DON, 15A-DON, T-2, HT-2, MON, MPA, NPA, PeA, ROC and 13 Ergot alkaloids	Grains and grain-based products; legumes; meat and offal, nuts and seeds; oils and fats; soy products; tuber; vegetables	PAT: extraction without purification;AFB1, AFB2, AFG1, AFG2, AFM1: IAC;Trichothecenes: SPE;Other mycotoxins: extraction without purification	PAT: HPLC-MS/MS; AFM1: HPLC-FLD; AFB1, AFB2, AFG1, AFG2: HPLC-FLD; Trichothecenes: GC–MS/MS; Other mycotoxins: LC-MS/MS	[10], [11], [12]
Hong Kong(China)	AFB1, AFB2, AFG1, AFG2, OTA, FB1, FB2, FB3, DON, AcDONs, ZEN, α-ZOL, β-ZOL	Cereals and their products, Vegetables and their products, Legumes, nuts and seeds and their products, Fruits, Meat, poultry and game and their products, Fats and oils, Beverages, alcoholic, Mixed dishes, Snack foods, Sugars and confectionery, Condiments, sauces and herbs	extraction without purification	UPLC-MS/MS	[20]
Spain	AFM1	Milk; dairy products	IAC	HPLC-FD	[38], [39]
Spain	AFB1, AFB2, AFG1, AFG2, OTA, ZEN, FB1, FB2, DON, NIV, 3A-DON, 15A-DON,T-2, HT-2, T-2 triol, NEO, Fus-X, DAS	Cereal and cereal products; olives; pickles; apple; pear; eggs; milk; milk shakes; custards; soya products; cheeses; grapes; alcoholic beverages; juices; oils	extraction without purification	UHPLC	[3]
Lebanon	AFB1, AFM1, OTA, DON	Bread and toast; biscuits and croissants; cakes and pastries, pasta and other cereal products; pizza and pies; rice and rice-based products; pulses; olive oil, sesame oil, and other oils; nuts, seeds, olives and dried dates; cheese; milk and milk-based beverages; milk-based ice cream and pudding; yogurt and yogurt-based products; caffeinated beverages; alcoholic beverages	Food samples (except for ‘‘Olive oil, sesame oil and other oils’’) extraction without purification;For ‘‘Olive oil, sesame oil and other oils’’, liquid–liquid extraction and IAC	LC-FD	[13]
Canada	OTA	Cereal and cereal products; alcohol drinks; coffee; tea; beans; fruits; sugars; chocolate; cheese; milk; eggs; dessert; meat; herb and spices; dried fruits; soya products; mixed dishes	IAC	LC-MS/MS	[14]
Australia New Zealand	AFB1, AFB2, AFG1, AFG2, AFM1, AFM2	Alcoholic and non-alcoholic beverages; cereal and cereal products; condiments; dairy products; eggs; fats; oils; fish; seafood; fish products; fruits; meat products; nuts and seeds; snacks; sugars; vegetables; infant food	—	HPLC-UV	[37]
Australia New Zealand	AFB1, AFB2, AFG1, AFG2, OTA	Alcoholic non-alcoholic beverages; cereal and cereal products; condiments; dairy products; eggs; fats; oils; fish; seafood; fish products; fruits; meat products; nuts and seeds; snacks; sugars; vegetables; infant food	—	—	[40]
Australia New Zealand	AFB1, AFB2, AFG1, AFG2, AFM1, OTA, PAT, ZEA, FB1, FB2, DON	Alcoholic and non-alcoholic beverages; cereal products; condiments; dairy products; eggs; fats; oils; fish; fruits; meat; nuts; seeds; snacks; sugars; vegetables; infant food, beverages; fast food	—	—	[15]
Viet Nam	AFB1, OTA, FBs	Rice and products; Wheat and products; Other cereals; Tubes, root and products; Beans and products; Tofu; Oily seeds; Vegetables; Sugar, confectionary; Seasoning; Oil, fat; Meat and products; Egg and milk; Fish; Other aquatic products	extraction without purification	ELISA	[16]
Ireland	AFB1, AFB2, AFG1, AFG2, AFM1, OTA, FB1, FB2, DON, 3A-DON, 15A-DON, DAS, T-2, HT-2, ZEN, PAT	Cereals, dairy, eggs, meat, fish, potatoes, vegetables, fruit, fruit dried, nuts seeds, herbs spices, soups,.sauces, sugar and preserves, confectionery, beverages, fats oils, snacks, composite	—	AFB1, AFB2, AFG1, AFG2, AFM1, OTA, FB1, FB2, ZEN, PAT: HPLCDON, 3A-DON, 15A-DON, DAS, T-2, HT-2: LC/MS	[17]
Sub-Saharan Africa	AFB1, AFB2, AFG1, AFG2, FB1, FB2, FB3, FB4, STC, OTA, CIT, ZEN, Ergot Alkaloids, T2, HT2	cereals, tubers, legumes, vegetables, nuts and seeds, dairy, oils, beveragesand miscellaneous	extraction without purification	LC–MS/MS	[18], [19]

A variety of analytical methods have been employed in TDSs, including ELISA [16], LC-UV [7], [37], LC-FLD [7], [8], [13], [38], GC–MS [7], GC–MS/MS [10], and LC-MS/MS [8], [10], [14], [18], [20]. Among them, LC-MS/MS is increasingly applied as a highly selective and sensitive tool for multi-mycotoxin analysis in complex food matrices. A combination of different methods is still necessary in TDS to achieve satisfactory sensitivity and accuracy, especially when multiple mycotoxins were considered. Similarly, in this study, the 43 mycotoxins were classified into three groups due to their diverse properties, with specific testing methods for each group, to achieve the best performance. Recently, we reported a UHPLC-MS/MS method for analyzing 10 emerging mycotoxins (AOH, AME, TeA, ALT, BEA, TEN, ENNA1, ENNA, ENNB1, and ENNB) [23], presenting our staged research progress on mycotoxin determination in China TDS. The 10 emerging mycotoxins were also considered in this work (classified in group C), and after further investigation, two more mycotoxins (CPA and CIT) were included in group C.

Regarding the food categories, the foodstuffs involved in the China TDS were more complicated, which included not only 12 categories of goods, but also the preparation and cooking of TDS samples, further complicating chemical compositions versus raw products.

Since mycotoxins may occur in trace amounts in dietary samples, sensitivity plays a critical role in a TDS. In previous studies of TDSs in China and abroad, for some mycotoxins, the LODs were relatively high, and therefore, a considerable number of “not detected” values were obtained. In the Irish TDS, fusarium toxins were not detected in any of the samples tested; however, the respective LODs were relatively high (20 µg kg−1 for fumonisins, 10 µg kg−1 for zearalenone, and 50 µg kg−1 for all remaining fusarium toxins) [17]. When conducting exposure assessment, the non-detects are set to 0, LOD and LOD/2 to estimate the lower bound (LB), upper bound (UB), and medium bound (MB) of exposure, respectively. Therefore, a high value of LOD may affect the accuracy of exposure assessment. In the second French TDS [8], the mean daily exposure to T2 and HT2 ranged 8.93 ng kg−1 bw (LB) to 51.8 ng kg−1 bw (UB) for adults, and 14.5 ng kg−1 bw (LB) to 91.1 ng kg−1 bw (UB) for children. These UB estimates were very close to or exceeded the group provisional maximum tolerable daily intake (PMTDI) of 60 ng kg−1 bw day−1 due to uncertainty in analytical results (with LODs of 3 µg kg−1 for T2 and HT2). Similarly, in the 4th and 5th China TDSs, the detection rates of HT2 were 2.8% and 0% (with LOD of 0.8 µg kg−1), but the MB of the exposure to T-2 and HT-2 was calculated to be 70 ng kg−1 bw day−1 and 52 ng kg−1 bw day−1, respectively, representing 116.7% and 87% of the PMTDI value, which could not accurately reflect the real exposure level. In this study, by choosing the [M + Na]+ as the precursor ion, the LOD of HT2 was greatly improved (0.08 µg kg−1). The proposed methods achieved a significant increase in sensitivity of multi-mycotoxins and could contribute to a more accurate estimation of dietary exposure.

Conclusions

The present work developed a highly sensitive and reliable strategy incorporating three UHPLC-MS/MS methods to determine 43 mycotoxins in dietary samples. The method recoveries for target compounds were 60.3–175.9%, with inter-day RSD below 13.9%, which are acceptable for analysis of multi-mycotoxins at trace levels. Upon optimization, LOQs were 0.0006–3 μg kg−1, indicating high sensitivity. The method was validated and applied to the 6th China TDS with success. Of the 72 food samples, more than 80% were found contaminated by mycotoxins. The most detected mycotoxins were DON, SMC, FB1, ZEN, BEA, ENNB1, and ENNB. Based on these results, the screening of mycotoxins that are of high level and high detection rate can be considered higher priority in future risk assessment. This novel strategy provides a basis for monitoring mycotoxins in various foods and will help to assess dietary exposure to mycotoxins.

CRediT authorship contribution statement

Nannan Qiu: Methodology, Formal analysis, Writing – original draft, Writing – review & editing, Funding acquisition. Danlei Sun: Formal analysis, Validation. Shuang Zhou: Methodology, Investigation, Validation, Writing – review & editing. Jingguang Li: Conceptualization, Investigation, Project administration. Yunfeng Zhao: Project administration, Resources, Supervision. Yongning Wu: Resources, Project administration.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.