Assessment of aflatoxins in pigmented rice using a validated immunoaffinity column method with fluorescence HPLC

Abstract

The determination of aflatoxins B₁, B₂, G₁ and G₂ in a variety of rice types was carried out using an immunoaffinity column (IAC) clean-up procedure followed by liquid chromatography (LC) with fluorescence detection. The sample was blended at high speed to form a fine powder and a test portion was extracted with methanol–water (60 + 40) using a high-speed blender. The filtered sample extract was diluted with 15% Tween 20 in phosphate buffered saline solution, and applied to an immunoaffinity column. Tween 20 is an essential part of the method to improve recoveries which were otherwise adversely affected by pigments in the rice. Aflatoxins were removed with neat methanol, and then directly analysed by reversed-phase LC with fluorescence detection using post-column bromination (Kobra cell). Test portions of black and polished rice were spiked with a mixture of aflatoxins, giving recoveries for individual and total aflatoxins ranging from 75.2 to 94.7%. Triplicate inter-day analysis at two levels gave relative standard deviations for repeatability (RSD_r) averaging 1.6% for total aflatoxins and 1.8% for aflatoxin B₁. A small survey of rice obtained from local markets in China found only one sample of red kojic rice which contained 2.9 μg/kg of aflatoxin B₁.

Introduction

Concern about human exposure to aflatoxins has tended inevitably to focus on high risk commodities such corn, nuts and dried fruit, where levels of aflatoxins in individual units can be both variable and relatively high (Andrade et al., 2013, Sugita-Konishi et al., 2010). Although it is known that cereals, other than corn, can be contaminated with aflatoxins, the emphasis for example, for wheat has tended towards monitoring and control of Fusarium toxins, where toxins such as deoxynivalenol commonly and frequently occur at mg/kg levels (Montes et al., 2012). Although rice is not immediately thought of as a high risk commodity, in terms of contamination levels of aflatoxins, there is substantial evidence indicating endemic low μg/kg occurrence of aflatoxin B₁ contamination in rice (Trucksess et al., 2011, Rahmani et al., 2011). In view of the fact that rice is a staple food item in much of the world, low level contamination can nevertheless be of concern leading to long-term exposure above recommended levels.

The analysis of foods containing pigments can be difficult due to low recoveries and interference from pigments in some food matrices. One strategy has been to use graphitised carbon in the clean-up column to remove pigments (Hu et al., 2006) or with immunoaffinity column (IAC) methods to use 0.5% polyoxyethylene–sorbitan monolaurate (Tween 20) to wash the IAC prior to elution (Stroka et al., 2000). For the analysis of aflatoxins in black sesame seeds, acceptable recoveries could only be achieved when Tween 20 was included in the solution used to load the extract onto the IAC (Liu et al., 2012). Others (Bansal et al., 2011, Mazaheri, 2009) have published validation data for IAC methods for aflatoxins in rice without using Tween 20, but the rice samples used for spiking have been brown (Bansal et al., 2011) or white rice (Mazaheri, 2009), where there are no problems of pigmentation. Although surveys have included black rice samples (Bansal et al., 2011) it is not clear whether the recoveries were specifically checked for the highly pigmented matrices. In the proposed method we have adopted a standardised approach of including Tween 20 for all samples of rice, as the incorporation of the surfactant ensures good recovery, adds only marginally to cost and there is no additional work to undertake by including this precautionary step.

Surveys for aflatoxins have been carried out using a variety of analytical methods which have included using an indirect competitive Enzyme Linked Immunosorbant Assay (ELISA) screening assay developed in the authors’ own laboratory (Reddy et al., 2009a), as well as commercial ELISA kits such as the Ridascreen^® AFB₁ test (Fredlund et al., 2009, Aydin et al., 2011) and AgraQuant^® (Zheng et al., 2005). Liquid chromatagraphy (LC) methods have involved extraction with acetonitrile/water (Fredlund et al., 2009, Nguyen et al., 2007) or methanol/water (Reiter et al., 1997, Bansal et al., 2011, Mazaheri, 2009, Rahmani et al., 2011) and clean-up using liquid/liquid extraction (Nguyen et al., 2007), florisil columns (Zuoxin et al., 2006), multi-functional columns (Fredlund et al., 2009), silica based monolithic columns (Khayoon et al., 2012), dispersive liquid–liquid microextraction (Campone et al., 2011) or commercial IAC from various manufacturers (Reiter et al., 1997, Bansal et al., 2011, Mazaheri, 2009, Rahmani et al., 2010, Rahmani et al., 2011, Trucksess et al., 2011).

Detection by instrumental methods has invariably employed fluorescence detection using pre-column trifluoroacetic acid (TFA) derivatisation (Zuoxin et al., 2006), post-column bromination with the Kobra cell (Fredlund et al., 2009, Reiter et al., 1997, Mazaheri, 2009) or pyridinum hydrobromide perbromide (PBPB) (Bansal et al., 2011), or alternatively post-column photochemical derivatisation (Rahmani et al., 2011, Trucksess et al., 2011). Liquid chromatography coupled with tandem mass spectrometry (LC–MS/MS) has also been employed for monitoring (Beltran et al., 2009) or for confirmation of the presence of aflatoxins in rice (Bansal et al., 2011, Trucksess et al., 2011). Various degrees of validation have been reported for these survey methods varying from minimal indications of method performance (Fredlund et al., 2009, Mazaheri, 2009, Nguyen et al., 2007) through to the generation of fairly comprehensive validation data (Reiter et al., 1997, Bansal et al., 2011, Rahmani et al., 2011).

Despite rice being a low-risk commodity, EU regulations controlling aflatoxin levels in cereals includes rice and imposes limits of 2.0 μg/kg and 4.0 μg/kg for aflatoxin B₁ and total aflatoxins respectively (European Commission, 2006a, European Commission, 2006b). Elsewhere in the world, limits for aflatoxin B₁ range from 5.0 μg/kg in Russia to 10.0 μg/kg in China and Japan and for total aflatoxins in rice from 20.0 μg/kg in the USA to 30.0 μg/kg in Brazil and India (Reddy et al., 2009b). The use of IAC for clean-up for aflatoxin analysis is now widely established and IAC methods have been used for a range of commodities (Ghali et al., 2009). There are validated official methods covering aflatoxins in nuts, dried fruit and spices (CEN, 2007, AOAC, 2010), but there are no CEN standards or AOAC First Action methods for aflatoxins in rice. The existence of regulatory limits demonstrates the need for an Official validated method for aflatoxins in rice which can be used for the purposes of food control. In this paper we have taken an IAC method which has been successfully validated for aflatoxins in sesame seeds (Liu et al., 2012), and applied the same approach to a variety of different pigmented rice samples including black rice. The use of a surfactant (Tween 20) as part of the clean-up proved to be effective in handling pigments in black sesame seeds (Liu et al., 2012). In this paper we have similarly employed a surfactant to improve recoveries as pigments in some types of rice can otherwise lead to significant losses. The use of surfactants is a unique important part of the reported method.