Development of rapid analytical methods for detecting mycotoxins in cereal grains


Cereals & Oilseeds
Project code:
01 July 2004 - 30 June 2006
AHDB Cereals & Oilseeds.
AHDB sector cost:
£53,000 from HGCA (Project No. 3008)
Project leader:
Richard Luxton University of the West of England Frenchay Campus, Coldharbour Lane, Bristol, Avon, BS16 1QY


pr412-abstract-and-summary pr412-final-project-report

About this project


Mycotoxins are naturally occurring toxic compounds produced by fungal growth. In excess of 200 mycotoxins have been identified, although most analytical laboratories can only assay for up to 20 of these.

The potential health risk of exposure to mycotoxins in foods depends upon the type of mycotoxin due their differing degrees of toxicity. Some of the more common mycotoxins are genotoxic and carcinogenic. Acute effects require that large amounts of mycotoxin are ingested. Traditionally the analysis of mycotoxins uses affinity columns to purify the extract prior to high pressure liquid chromatographic (HPLC) detection. These assays are time-consuming and relatively expensive (~£100).

The increasing consumer demand for safe food is accompanied by the need for rapid and cost-effective analytical techniques capable of screening for the presence of mycotoxins to allow for the positive release of food ingredients before they enter the food chain. This requirement was addressed in the project by developing an analytical system using disposable biosensors capable of identifying and quantifying a pre-selected suite of mycotoxin residues.

Six mycotoxins were selected by the project consortium members based on food industry analytical requirements and availability of antibodies. The six mycotoxins were: ochratoxin A, deoxynivalanol (DON) and aflatoxins B1, B2, G1, G2. Zearalenone and fumonisins FB1 and FB2 were identified for inclusion if any of the selected mycotoxins were problematic.

The overall aim of the FQS 61 project was to investigate the feasibility of developing an automated detection system for mycotoxins in foodstuffs and feeds.  This involved combining a novel rapid extraction method with biosensor technology integrated in a twelve channel measurement system allowing simultaneous measurement of up to six different mycotoxins, in duplicate.

Rapid extraction techniques developed at CCFRA allowed mycotoxins to be extracted from grain in less than 10 minutes using a non-toxic, Pytosol solvent.  The extraction method compared well with conventional methods.  The measurement of mycotoxins was performed on an instrument developed by Uniscan Instrument which incorporates a competitive immunoassay, combined with an electrochemical detection step.  Development of the mycotoxin immunoassays was undertaken at the University of the West of England. 

The immunoassays were performed using a biosensor which immobilised the mycotoxin antibodies on the biosensor surface.  An array of 12 biosensors was produced by a screen printing method using conductive inks.  Each biosensor could have a different antibody immobilised on the surface giving the potential of 12 separate measurements. 

Gwent Electronic Materials were responsible for designing and fabricating the arrays of biosensors and Applied Enzyme Technologies developed stabilisers for the antibodies immobilised on the biosensor surface.  The electrochemical measurement was made by the instrument giving the concentration of mycotoxins present in the sample using neural network.  The software can also evaluate any cross-reactivity between the specific antibodies and other mycotoxins.

During the course of the project a large number of antibodies were characterised for the different mycotoxin assays. Problems obtaining suitable antibodies for all six mycotoxins proposed resulted in only assays for total aflatoxin, aflatoxin B1, ochratoxin A and DON being fully characterised.  The instrument developed was able to automate the immunoassays, allowing 12 simultaneous measurements to be made.

The neural network showed that there was no cross-reactivity between the assays used on the instrument, indicating that the antibodies were giving a high degree of specificity to each of the mycotoxins.  The assays were optimised to cover the concentration ranges that are required to discriminate levels of mycotoxin above and below the MRL.

The evaluation stage of the project showed that the antibodies on the biosensor surface were stable at room temperature for at least six months and a preliminary end-user study of ochratoxin A extracted from grain samples showed that the system could in principle be used to measure mycotoxins.  However, further optimisation is required to reduce the assay time from the current 77 minutes. 

The project successfully demonstrated that arrays of biosensors can be integrated into an instrument allowing the automated measurement of multiple mycotoxins to be performed.