Chemical Contaminants in Plant Based Protein Survey

Food Standards Scotland (FSS) requested a survey of chemical contaminants in plant-based protein products (PBPP). There are various types of PBPP, these include meat alternative products such as tofu, meat free or meat substitute mince, non-meat pieces, and protein powders. PBPP can be made from a variety of different protein sources; soy, wheat or wheat gluten, pea, other legumes and vegetable proteins are amongst the most common found in supermarket plant-based meat replacement products. Plant-based protein powders are also made from a range of different proteins such as pea and soy protein. There is a risk of mycotoxin and plant alkaloid contamination in the raw materials and therefore FSS want to assess risk in the processed products to inform future policy. 

Contamination with mycotoxins depends upon many factors. The presence of mycotoxins in foods can occur in the field or during storage. Agricultural practices and climate effects such as drought, heavy rain and temperature can affect the potential risk for contamination. In addition, other factors such as post-harvest handling and storage conditions of crops and raw materials are potential risk factors in mycotoxin development. Although there are mycotoxin MLs within legislation for cereal products, unprocessed cereals and other commodities such as nuts there are no specific MLs for other raw materials such as soybeans or peas. There are also no MLs for plant-based alkaloids for soybeans or peas within current legislation, either in GB or the EU. 

Food law in the UK comprises European Union (EU) legislation and domestic legislation for each UK nation. All references to EU legislation should be read as either EU law in Northern Ireland (NI) or assimilated law in England, Wales and Scotland, as applicable. In NI, under the Windsor Framework, EU food law continues to apply, but for England, Wales and Scotland retained EU law was on 1 January 2024 under the Retained EU Law (Revocation and Reform) Act 2023 reclassified as assimilated law into national law.

The project was carried out in two phases, sampling and analysis were carried out at two different time points to ensure sampling was representative. 

Fera provided advice to FSS to help with the design of Phase 1 of the study and FSS purchased all samples and sent them to the laboratory. FSS requested the laboratory help with design of the testing regime, but specified the following analytes as mandatory and that all samples should be analysed for these unless there was a scientific basis to exclude them. 

Table 1. Priority List of Contaminants

There are currently no specific maximum levels in place for mycotoxins and plant alkaloids in PBPP, although any products on sale on the retail market must meet the general provisions of food safety law, i.e. that the food is not harmful or unfit for consumption and must be of the expected nature, substance, and quality (11). The mycotoxins listed in Table 1 are included in Commission Regulation (EC) No 1881/2006 (as amended) setting maximum levels for certain contaminants in foodstuffs for a range of foods (1, 2). In addition, Commission Regulation (EU) 2016/239 introduced MLs for atropine and scopolamine (tropane alkaloids) in cereal based foods for infants and young children (12). MLs are in force for ergot alkaloids in some cereal based foods in the European Union EU) (4) and also Northern Ireland, however these limits do not apply in Great Britain (GB). A comparison of MLs in force in GB and the EU / NI is given in Table 2. 

Table 2. Comparison of maximum levels in force for GB and EU/NI.

As well as mycotoxins and natural toxins there were also reports in the literature of the presence of acrylamide, a processing contaminant, in PBPP (13). Acrylamide is formed during processing that involves heat as a reaction product between some reducing sugars and amino acids. It has been commonly reported in heat treated starchy foods such as baked goods (bread, biscuits, crackers) and fried goods (potato and vegetable crisps, deep fried breadcrumbed or battered products). It’s occurrence has also been reported in a range of other foodstuffs including roasted coffee, olives, cocoa, dried fruit and extruded snacks (14). It was agreed that all products in Phase 1 would also be analysed for acrylamide as they were all processed foods.

The analytes to be tested in each type of protein in Phase 1 were agreed in consultation with FSS taking into account budgetary and time constraints. The final product list was compiled by FSS, the protein types were selected on the basis of market share. The three main types chosen (soy, pea and wheat) are those most frequently found for sale, making them the most commonly consumed. UK market data shows soy is the largest category with 31.1% of the UK market in 2025. Pea protein was identified as the fastest growing, with projected Compound Annual Growth Rate (CAGR) of 6.4%. Market studies confirm pea protein is one of the two leading categories although the exact market share in 2025 is not reported. Wheat protein is described as present but with a smaller UK market share but there is no so information on a quantified market share. Other plant proteins, e.g. rice, hemp and potato are described as emerging niches in the UK market (15). 

Additional sampling was discussed with FSA and products were prioritised based on data gaps around products such as protein powders, ready-to-eat protein shakes and protein-based cereal snack bars in addition to a range of fresh and frozen plant-based meat alternative products. It was decided to avoid products coated in breadcrumbs or other cereal based coatings as the cereal ingredients could contain mycotoxins. The product / analyte combinations agreed for Phase 1 are listed in Table 3. 

Table 3. Phase 1 Analyte / product combinations.

Ergot alkaloids were only tested in wheat based products as these contaminants specifically occur in cereals and no evidence of their occurrence in other commodities (e.g. soy) was found in the literature.

In addition, sterigmatocystin was also measured in all samples as the 11+ IAC multi-mycotoxin method was used was also able to measure this mycotoxin. 

FSS commissioned a follow on study to analyse a second set of samples. An assessment was made of recently published studies on meat alternative products to inform which additional contaminants should be included in the study. Using information from a variety of sources including Mihalache et al., 2022, 2023 and 2024; Lin et al., 2023; and FAO, 2020 (6-10 and 16) a table of suggested analyses was shared with FSS. In addition to the analytes in Table 3 a number of other mycotoxins and plant toxins were reported to occur in PBPP in the literature. Mycotoxins reported were Alternaria toxins, enniatins, beauvericin, citrinin, sterigmatocystin. Other alkaloids found were β-carboline alkaloids and pyrrolizidine alkaloids. The exposure assessment by Mihalache (9) found, based on the model used, consumption of plant-based meat alternatives led to non-tolerable exposure to alternariol in pea and soy + wheat based products. They also noted health concerns related to liver and renal cancer for aflatoxins and OTA respectively. 

The list of requested analytes for Phase 2 was more extensive than Phase 1. Following discussion with FSS it was agreed the analyses listed in Table 5 would be carried out on the samples. All samples would be analysed for all analytes except erucic acid, where only samples that contained rapeseed oil as an ingredient would be analysed. Some methods used in Phase 1 were also used for Phase 2 (11+ and ergot alkaloids), however a number of additional methods were also required to allow tests to be carried out for all the requested analytes.