¹ Student Research Committee, Tabriz University of Medical Sciences, Tabriz, Iran.
² Department of Pharmaceutical and Food Control, School of Pharmacy, Tabriz University of Medical Sciences, Tabriz, Iran.
³ Department of Human Ecology, School of Public Health, Tehran University of Medical Sciences, Tehran, Iran^.
4 Department of Technology Processing of Fruits, Vegetables and Milk, Kharkov State University of Food Technology and Trade, Kharkiv, Ukraine.
⁵ Halal Research Center of IRI, FDA, Tehran, Iran.

 
Introduction

Mycotoxins are the secondary metabolites produced by different types of fungi (Flores-Flores et al., 2015). Their presence in food depends on a variety of factors, such as temperature, humidity, storage, and Processing (Rawat, 2015). Aflatoxin is one of the most toxic known mycotoxins produced by genus Aspergillus, mostly aspergillus flavus
and aspergillus parasiticus species (Yu et al., 2004).
Aflatoxin B1 (AFB1), aflatoxin B2 (AFB2), aflatoxin G1 (AFG1), and aflatoxin G2 (AFG2) are the most important types of aflatoxin. Furthermore, aflatoxin M1 (AFM1) and aflatoxin  M2 (AFM2) are metabolites of AFB1 and AFB2 in the livestock body (Kumar et al., 2017), which are generally found in a wide range of foods such as grains, pistachio, milk, and nuts (Alshannaq and Yu, 2017). Aflatoxins are highly toxic, mutagenic, teratogenic, and carcinogenic compounds that are known as the causative agents of liver carcinogenesis. The most toxic aflatoxin is AFB1 that was determined by the International Agency for Research on Cancer (IARC) as group 1 carcinogen types (Abrar et al., 2013).
The allowed limits of AFB1 and total aflatoxin  are 2 and 4 µg/kg in food products in the European :union:, respectively (Authority, 2007, Kilicel et al., 2017). Various methods were introduced to detect aflatoxins, such as ELISA and HPLC using mass detection and several other methods. In this regard, HPLC with fluorescence detection (FLD) is one of the most commonly used methods to detect aflatoxin (Espinosa-Calderón et al., 2011).
The dispersive liquid–liquid microextraction (DLLME) technique was introduced in 2006, which is a quick, low-cost, and low-solvent consumption method. This method is used to extract a small amount of the material from aqueous samples. This technique is based on injecting a small amount of a solvent extraction agent (like C2Cl4) and a dispersed solvent (like acetone) into an aqueous solution. The resulting solution is then centrifuged to separate the organic and aqueous layers. This process is useful in extracting organic compounds from aqueous samples (Berijani et al., 2006, Rezaee et al., 2006).
In previous studies over analysis of mycotoxins using the development and rapid sample preparation methods, the suitability of DLLME was investigated for sensitive determination of aflatoxins in samples. HPLC_FLD was used as a determination technique (Campone et al., 2011).
In this study, the aim was to measure the amount of aflatoxin extracted using DLLME method in tomato paste samples collected from the northwest of Iran. Later, the amount of aflatoxin contamination in Iran was compared with the international standards.
Materials and Methods
Chemicals and reagents: Sodium chloride (NaCl), n-hexane, chloroform (CHCL3), and Formic acid (CH2O2) included high purity LC grade provided by Merck (Darmstadt, Germany). The stock solution of AFB1 at a concentration of 1000.0 ng/ml was supplied by Sigma (St.Louis, MO, USA). Methanol and acetonitrile were prepared from Duksan Pure Chemicals (Kyungkido, Korea), deionized extra pure water was also purchased from Shahid Ghazi pharmaceutical company (Tabriz-Iran). To construct a calibration curve, 6 dilutions of AFB1 standards were prepared from 0.6, 1.25, 2.5, 5, 10, and 20 ng/ml by adding appropriate volumes of methanol to obtain these concentrations. The Stock solution of AFB1was kept at 4 ºC in the dark.
Equipment and chromatographic conditions: The HPLC analyses were carried out in a KNAUER HPLC system with a K-1000 pump (Berlin, Germany) coupled with a fluorescence detector. Biotech 2003 degasser (United State) and the analytical column included C18 column 4.6x250 mm (Knauer, Berlin, Germany).
The mobile phase was acetonitrile: methanol: water (24:25:51), de-aerated by an ultrasonic bath for 10 min at a flow rate of 1.5 ml/min in isocratic mode. The fluorescence detector was set at wavelengths of 373 and 450 nm for excitation and emission, respectively.
Method of validation: The developed analytical method was validated based on repeatability (intra-day precision) and reproducibility (inter-day precision), which is also called RSD, linearity, accuracy (as recoveries), selectivity, as well as limits of detection (LODs) and quantification (LOQs). The LOD was calculated as the concentration of analyte, whose peak area was three times the area of the noise of a blank sample (S/N ≥ 3). LOQ was calculated by taking three replicates of the detection limits when S/N ≥ 10. The acceptable value for RSD is below 20% and the acceptable range for recovery is 70-120% (Fontana et al., 2016). In measuring linearity, a value of higher than 0.99 is acceptable.
Tomato paste sampling: Tomato paste samples were collected from different stores in 2019 within two months. A total of 30 samples were collected from different brands, including 10 famous Iranian brands, 12 non-famous Iranian brands, 5 non-Iranian brands, and 3 homemade tomato pastes.
Extraction procedure: The extraction procedure of analyte from tomato paste samples was carried out according to a previous study method (Amirkhizi et al., 2015) after applying some modifications. To this end, 5 g well-homogenized tomato paste samples were mixed with 0.1 g NaCl and 20 ml acetonitrile-water (80:20, V/V). Later, they were blended by a magnetic stirrer for 10 minutes; 1 g diatomaceous earth was added and mixed for 5 min and the mixture was clarified with Whatman filter paper No.1 (Whatman Inc., Clifton, NJ, USA).
In the next stage, 3 ml of n-Hexane was added to 10 ml of the extracted solution and shortly shaken by Vortex. The n-Hexane was separated by centrifugation for 1 min and cleaned up by DLLME (Amirkhizi et al., 2015).
Dispersive liquid–liquid microextraction procedure: A volume of 260 µl chloroform (extracting solvent) was added to 1 ml of the obtained extract (disperser solvent) and the pH was adjusted to about 5.8 by formic acid. The mixture was quickly injected into a 10 ml centrifuge test tube with a conical bottom containing 2.5 ml of distilled water. It was then vortexed for a few seconds to obtain a stable and cloudy solution; so, AFB1 was entrapped into the fine CHCl3 droplets. This ternary component solution was centrifuged at 5000 RPM for 3 minutes after the aqueous upper layer was removed. The precipitated chloroform was transferred to a small tube by a 220 µl sampler and dried at a mild flow of nitrogen gas. The residue was re-dissolved in 20 µl methanol and then injected into the HPLC system (Afzali et al., 2012, Amirkhizi et al., 2015).
Results
Method validation: In order to verify the linearity, six concentrations of the standard B1 0.6, 1.25, 2.5, 5, 10, and 20 ng/ml were injected into the device in three replications and obtained with a calibration curve R2 = 0.9974, which is acceptable. The LOD and LOQ were 0.14 and 0.44 µg/kg, respectively. The recovery and RSD rates were acceptable (Table 1).
Testing the method for evaluating the natural occurrence of AFB1 in tomato pastes:
The analytical results of AFB1 in samples are reported in Table 2. AFB1 was detected in all samples and quantified in 14 samples. In this study, concentration ranged from 0.47 µg/l to 7.7 µg/l. In this study, 46.6% of the samples were contaminated with AFB1. The average contamination rate was 1.1 ± 0.02 µg/kg. A total of 6 samples contained more than 2 µg/kg AFB1, which is within the limit allowed by the Europe :union:. As expected, none of them was famous commercial brands.