Multiclass mycotoxin analysis in Silybum marianum by ultra high performance liquid chromatography–tandem mass spectrometry using a procedure based on QuEChERS and dispersive liquid–liquid microextraction

Abstract

Ultra high performance liquid chromatography–tandem mass spectrometry (UHPLC–MS/MS) has been proposed for the determination of 15 mycotoxins in milk thistle (Silybum marianum), including aflatoxins, fumonisins, trichothecenes, ochratoxin A, citrinin, sterigmatocystin and zearalenone. The mycotoxins were detected by electrospray ionization in positive ion mode and multiple reaction monitoring (MRM), achieving the separation in about 4 min. Sample treatment consisted of a modified method based on a first step using a QuEChERS-based procedure which allowed the determination of fumonisin B₁, fumonisin B₂, nivalenol, deoxynivalenol and fusarenon-X, and a subsequent clean-up based on dispersive liquid–liquid microextraction (DLLME) for the determination of the rest of mycotoxins. The method has been validated in extract and seeds of milk thistle, obtaining limits of quantification lower than those usually permitted by legislation in food matrices, with precisions lower than 10%. Recoveries were between 62.3% and 98.9%, except for zearalenone in seeds samples and citrinin in extract. The method was also applied for studying the occurrence of these mycotoxins in market samples (six samples of seeds, three of them purchased in bulk in a street vendor, and one natural extract of milk thistle), and HT-2, T-2 and zearalenone have been found in some of the samples. To the best of our knowledge, this is the first time that this type of treatment has been used for these complex food matrices, allowing the analyses of the most important mycotoxins.

Introduction

During the last decade, both offer and consumption of products with specific nutritional and/or functional characteristics, declared to have positive effects on our health have significantly increased. They can be found in pharmacies, supermarkets, food shops, and even advertised on the radio/TV or by internet, through web-pages located inside but also outside Europe. Among these products, food supplements are intended to provide a concentrated source of nutrients and other substances and contain, in many cases, herbal products and/or their derivatives as ingredients (i.e. ginseng, echinacea, gingko, green tea, etc.). Previous studies have demonstrated that these materials can suffer from fungi and mycotoxin contamination [1], [2], [3], [4], which could lead to diverse human health problems. The principal fungi genera and species that can produce toxins and the mycotoxicosis symptoms of some of the most common mycotoxins are reported in the article of Capriotti et al. [5].

Within European Union (EU), herbals are registered either as food ingredients or as drugs depending on the Member State. This fact hampers the standardization of the legislation to regulate the production and commercialization of these products within the common market. Although the European Food Safety Authority (EFSA) has published a series of documents to facilitate the safe evaluation of plants intended to be used in food supplements [6], regarding contaminant content, including mycotoxins, EFSA only specifies that herbal products must comply with the current food legislation of the EU. In this sense, EU has set maximum levels for aflatoxins, ochratoxin A, patulin, deoxynivalenol, zearalenone and fumonisins for several foodstuffs derived from fruits and cereals by the Commission Regulation (CE) No. 1881/2006 [7] and subsequent amended. However, it must be pointed out that none of these Regulations set maximum levels for mycotoxins in food supplements. Moreover, recently the EFSA gave a “Call for continuous collection of chemical contaminants occurrence data in food and feed”, including mycotoxins and plant toxins between the target substances to be controlled [8].

Regarding the contamination of herbal products used as drug ingredients, the European Pharmacopeia only sets a maximum level for aflatoxin B1 (2 μg kg⁻¹) and for the total content of aflatoxins B₁ + B₂ + G₁ + G₂ (4 μg kg⁻¹) [9]. For other mycotoxins, the European Pharmacopeia neither recommends nor establishes any allowed maximum content. Therefore, there seems to be evidences of a legal loophole in this field. This all suggests that developing efficient, high sensitive, fast and multianalyte methods to control these residues in dietary supplements based on herbal products, fulfilling the established legislation [10] is indispensable.

Different analytical methods have been proposed for mycotoxin determination in food, such as TLC [11], [12], enzyme-linked immunosorbent assay (ELISA) [13], [14], GC–MS [15] or capillary electrophoresis (CE) [16]. However, the most popular techniques for determining mycotoxins in foods are HPLC with UV/Vis or fluorescence (FL) detection [17], [18], [19], [20], [21] or MS detection [22], [23], [24], [25]. UHPLC–MS/MS has become very popular in the last years for the multiclass analysis of mycotoxins [3], [26], [27], [28] and for their determination with other contaminants [29], [30], [31], [32]. For extraction and clean-up, different approaches have been proposed, the most common methodologies being solid–liquid extraction followed by solid phase extraction (SPE) and immunoaffinity columns (IACs) containing specific antibodies to the analyte of interest [33]. Recent reviews present an overview of the different methodologies proposed for the determination of mycotoxins in food, including the most frequent sample treatments [5], [34], [35], [36], [37], [38], [39], [40].

The scarce methods proposed for the determination of mycotoxins in herbal products are based mainly on LC-FL for aflatoxins (involving post-column derivatization) and ochratoxin A [41], [42], [43], [44], [45] or LC–MS for multiresidue analysis [3], [46]. Most of these methods use solid–liquid extraction and IAC for clean-up, including the method recommended by Pharmacopeia for the determination of aflatoxin B₁, which has been validated only for devil's claw roots, ginger and senna pods [9]. However, IAC is an expensive and complex purification system which suffers from low recoveries for some mycotoxins, due to the complexity of these matrices. Moreover, the multiclass analysis is quite limited with this selective extraction. As a consequence, simpler, more efficient, multiclass and environmentally friendly extraction systems are demanding. An increasingly popular treatment is the so called QuEChERS (Quick, Easy, Cheap, Effective, Rugged and Safe), which has been widely used in the last years, mainly for the extraction of pesticides, but also for other compounds [47]. QuEChERS methodology presents some advantages, such as its simplicity, minimum steps, and effectiveness for cleaning-up complex samples [48]. It involves two steps: the first one is an extraction step based on partitioning via salting-out extraction involving the equilibrium between an aqueous and an organic layer, and the second one is a dispersive SPE (DSPE) step that involves further clean-up using combinations of MgSO₄ and different sorbents, such as C18 or primary and secondary amine (PSA), to remove interfering substances. QuEChERS based methods have been recently reported for the extraction of different mycotoxins in cereal products [15], [23], [24], [26], [49], [50], [51], eggs [27], wine [20], or beer-based drinks [52], and in the multiresidue extraction of different contaminants, including mycotoxins, in organic food products and milk [29], [30], [31]. However, to the best of our knowledge, this methodology has not been used for the determination of mycotoxins in botanicals.

Dispersive liquid–liquid microextraction (DLLME) is another emerging technique introduced for treatment of liquid samples [53], [54], [55]. This technique is based on the use of a ternary component solvent system, where an appropriate mixture of a few microliters of an organic extraction solvent (usually with a density higher than water), and a small volume of a disperser solvent (miscible with the extraction solvent and with water), is injected rapidly into an aqueous sample, resulting in the formation of a stable emulsion. The organic analytes present in the aqueous sample are rapidly extracted into the extraction solvent as a result of the large contact surface between the organic and the aqueous phases. Phase separation is performed by centrifugation and an organic phase with the analytes of interest is settled in the bottom of a conical tube and subsequently analyzed by an appropriate technique. DLLME has been used in the determination of ochratoxin A in wine [20], [21], [25] and aflatoxins in cereal [18] with excellent results.

Considering the above described situation about the determination of mycotoxins in herbal products, we propose a multiclass method based on UHPLC–MS/MS for the simultaneous determination of 15 mycotoxins in Silybum marianum, commonly known as milk thistle. This botanical is one of the most consumed food supplements due to its protective effects on the liver and to greatly improve its function; moreover, although analytical methods for studying the occurrence of mycotoxins in other herbal products (such as ginseng, ginger and others) have been previously reported, this is not the case of milk thistle. The studied mycotoxins are those included in regulation (EC) 1881/2006; in the document “Call for continuous collection of chemical contaminants occurrence data in food and feed”; and some others considered by the International Agency for Research on Cancer (IARC) [56]: aflatoxin B₁ (AFB₁), aflatoxin B₂ (AFB₂), aflatoxin G₁ (AFG₁) and aflatoxin G₂ (AFG₂), fumonisin B₁ (FB₁), fumonisin B₂ (FB₂), nivalenol (NIV), deoxynivalenol (DON), fusarenon-X (F-X), HT-2 toxin (HT-2), T-2 toxin (T-2), ochratoxin A (OTA), citrinin (CIT), sterigmatocystin (STE) and zearalenone (ZEN). In addition, an extraction method based on QuEChERS procedure for determination of FB₁, FB₂, NIV, DON and F-X with an additional clean-up DLLME step for determination of AFB₁, AFB₂, AFG₁, AFG₂, OTA, T-2, HT-2, STE, CIT and ZEN, has been proposed. To the best of our knowledge, this is the first time that this type of treatment has been used for these food matrices.