Vue au microscope d'une moisissure se développant sur l'alimentation
31/05/2023
Research
3 min

Mould in food: identifying mutagenic and carcinogenic toxins

The combined use of several different software applications could speed up the acquisition of knowledge on toxins known as mycotoxins, which are produced by certain types of mould growing in food. This innovative approach has been used by ANSES scientists to identify mycotoxins that may be mutagenic and/or carcinogenic.

Mould in foodstuffs, particularly those of plant origin, can produce toxins known as mycotoxins, some of which can cause genetic mutations or cancer in humans or animals: "Mould can grow throughout the food production process, from cultivation of crops through to the finished product," explains Denis Habauzit, project manager in the Toxicology of Contaminants Unit at the ANSES Fougères Laboratory. Many foods – such as cereals, fruit and vegetables – can potentially be contaminated by mycotoxins.

European regulations strictly limit the maximum authorised quantities for the main mycotoxins potentially found in foodstuffs placed on the market, but there is a lack of data on the others: "Research has shown that food can contain mycotoxins for which we have little or no information on their toxicity, and which are neither regulated nor monitored," says Valérie Fessard, head of the unit.

Computer tools to identify the most toxic compounds

The unit's scientists turned to computer modelling to identify mycotoxins with mutagenic or carcinogenic activity. They used a combination of quantitative structure-activity relationship (QSAR) software, which predicts the effects of compounds on living organisms according to their structure. This method has the advantage of enabling an initial characterisation of mycotoxins, which are difficult to synthesise and purify. It avoids the need to test them and in particular to resort to animal testing. The results were published in the journal Environmental Pollution in April 2023.

The team selected the most effective combinations of software by testing them on mycotoxins with known carcinogenic or mutagenic potential. All the software applications selected were free, to facilitate their use by other research teams. The scientists then analysed 904 mycotoxins and mycotoxin metabolites from the database created by the team. They found that 127 had mutagenic potential and 548 may be carcinogenic.

Initial sorting still to be refined

 "The software is still being developed," warns Denis Habauzit, "so there is a risk of error. But it has drawn attention to certain compounds and identified those on which experimental toxicology studies should be carried out as a priority." Ninety-five of these mycotoxins seem to be both mutagenic and carcinogenic. They could therefore pose a health risk even in small quantities.

At the same time, the potential effects of the mycotoxins should be examined in relation to the actual quantities of these compounds found in food and feed, to determine the risk they actually pose. These quantities are currently only known for certain mycotoxins, in particular via the Total Diet Studies (TDSs) carried out by ANSES, with data on other toxins still needing to be obtained. 

This research is the most extensive to date on mycotoxins using QSAR software. It demonstrates the value of these IT tools when making an initial selection from among the hundreds of mycotoxins, in order to identify those to be assessed as a priority. These predictions are especially important given that climate change and fungicide use restrictions could promote mould development and the contamination of foodstuffs by emerging mycotoxins.