Survey of milk for mycotoxins (Number 17/01)

Friday 14 September 2001

Food Survey Information Sheet

Key Facts

• Mycotoxins are naturally occurring toxins produced by moulds which may grow on cereals and other materials used as animal feeds. In dairy cattle, some of these toxins can be metabolised and their metabolites transferred into milk
• One hundred milk samples (50 retail and 50 farm gate), both conventionally produced and organic, were analysed for aflatoxin and ochratoxin A
• Ninety-seven percent of conventionally produced, and all the organically produced, samples contained no measurable amount of aflatoxin M₁
• Of the 3 percent of samples which did contain aflatoxin M₁, none were above the current EU legal limit of 0.05 microgram/kg.
• Ochratoxin A was not detected in any of the samples
• These survey results do not raise any food safety concerns

Summary
This survey analysed 50 retail and 50 farm-gate milk samples, produced by both conventional and organic methods, for the mycotoxins: aflatoxin M₁ (AFM₁) and ochratoxin A (OTA). Milk samples that contain residues of mycotoxins indicate that they originated from a dairy herd which had consumed contaminated feed.

Aflatoxin M₁ was found in three percent of conventional farm-gate milk samples at levels ranging from 0.01 - 0.021 microgram/l. None of these samples are above the limits set in EC Regulation (EC) No. 1525/98 of 0.05 microgram/l. AFM1 was not detected in any retail milk samples or organically produced farm-gate samples.

Ochratoxin A was not detected in any of the samples.

Background
The Agency has a well-established monitoring programme for mycotoxins in food. Mycotoxins are naturally occurring toxins produced by moulds, growing on a wide variety of foods and animal feeding stuffs. Mycotoxins that pose a risk to human health include aflatoxins (AFL) and ochratoxin A (OTA), both of which are potential contaminants of animal feed.

Aflatoxins are produced primarily by two species of mould, Aspergillus flavus and A. parasiticus and to a lesser extent by A. nomius. Each of these moulds can grow on a wide range of cereal crops especially in tropical and sub-tropical climates. Different moulds which generally favour climates with a lower temperature and humidity produce OTA. The growth of all these moulds is dependent on numerous environmental factors (e.g. temperature and humidity) and on crop storage and so the occurrence of the toxin cannot always be predicted. The moulds' presence is not necessarily indicative of the presence of the toxin and conversely toxins can be formed without visual evidence of mould signifying the need for diligence in this area.

Earlier studies have identified specific contamination problems and these have been addressed through improved monitoring practices supported by regulations. Table 1 summarises the major fungal species responsible for the contamination of foods by each of these mycotoxins, the foods that are known to be susceptible and the most recent expert committee opinion on possible human health effects.

Aflatoxin B1 (AFB1) is metabolised in ruminants to produce aflatoxin M₁ (AFM₁) and this metabolite may be transferred into cows' milk. AFM₁ is about 90 percent less toxic than AFB₁, however, it is still considered to be a human carcinogen. Amounts of AFB₁ in complete feed for diary cattle are controlled by the Feeding Stuffs Regulations 2000. These specify a maximum concentration of AFB1 (5.0 microgram/kg), hence reducing the potential AFM₁ carry over in to milk.

OTA has not been monitored in UK milk before but it has been found previously in animal feed.

Aflatoxins are both acutely and chronically toxic in animal studies and are recognised as animal and human carcinogens, with AFB1 being the most potent.1 The Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT) recommended that the levels of AFB₁ in food should be reduced to the lowest technologically achievable. This was subsequently endorsed by the Food Advisory Committee (FAC).

Previous work has shown that very low levels of AFM₁ may occur in retail milk. Surveys performed in 1995/6 determined the concentration of AFM₁ in retail milk and milk products and farm gate milk. Most samples analysed contained very low or unquantifiable levels of AFM₁.

European Commission Regulation (EC) No. 1525/98 came into force on the 1st of January 1999 which was implemented in the UK by a Statutory Instrument (The Contaminants in Food (Amendment) Regulations 1999 (S.I. [1999] No. 1603)). The Regulation sets limits for AFB₁ and total aflatoxins in groundnuts, nuts, dried fruit and cereals and a limit for AFM₁ in milk (0.05 microgram/kg). Recently, the Codex Alimentarius Commission has approved a higher limit for AFM1 of 0.5 microgram/kg. The Joint FAO/WHO Expert Committee on Food Additives and Contaminants (JECFA) concluded that the higher limit is sufficient to protect human health.

EC Legislation controlling OTA is due to come into force shortly. This will set limits in cereals (5 microgram/kg) and dried vine fruits (10 microgram/kg). The limits result from the evaluation undertaken by several expert committees of the toxicity of OTA. In 1995, JECFA set a provisional tolerable weekly intake (PTWI) for OTA of 0.1 microgram/kg b.w./week, which was reaffirmed this year. In 1998, the EC's Scientific Committee for Food (SCF) issued its most recent opinion on OTA advising that the tolerable daily intake (TDI) should be less than 0.005 microgram/kg b.w./day. The COT concluded that OTA is a genotoxic carcinogen and has been linked to kidney damage in pigs and humans.

Sampling Strategy
A total of 100 samples were taken for this survey. Mintel purchased 50 samples from retail shops and 50 farm gate samples were supplied by National Milk Records. The samples were then transferred to the analytical laboratory, RHM Technology Ltd. A minimum sample size of 1 litre was taken. In a departure from normal practice, a proportion significantly greater than the market share was allocated to organic samples. This was necessary to obtain sufficient data to carry out a meaningful statistical analysis of the results. Of the 100 samples analysed, ten were farm-gate organic and ten were retail organic samples.

Methodology
Details of the analytical procedures used and the limits of detection (LOD) are summarised in Table 2. The samples were analysed in batches. Each batch included a reagent blank and at least one spiked sample fortified at a specific level for each of the toxins analysed. All results were corrected for recovery. Measurement uncertainty details can be found at Annex 1. Recovery values are detailed in Annex 2.

Quality Assurance
The methods described are UKAS accredited and have been validated i.e. evidence has been, and continues to be, generated to demonstrate that the methods do what is intended and are reliable and reproducible. Control of mycotoxin analyses is monitored using naturally contaminated in-house reference materials and by participation in UK and EU proficiency schemes, collaborative and intercomparison trials. All samples in this survey were stored at ?20oC and thoroughly mixed prior to analysis.

Results
All sample data can be found in Table 3, in the "see also" box.

Aflatoxin M₁
Aflatoxin M₁ was found in three percent of samples, all of which were conventional farm-gate milk. Levels ranged from 0.01 ? 0.021 microgram/kg. None of these samples were above the current maximum EC level of 0.05 microgram/l. AFM₁ was not detected in any retail milk samples or organic farm-gate samples (Table 4).

Ochratoxin A
OTA was not detected in any of the milk samples (Table 4).

Milk samples were also analysed for zearalenone, alpha and beta zearalenol and their conjugates. These analyses produced anomalies which the Agency is still investigating. However, in order to report promptly the results of AFL and OTA analysis, these data will be reported in due course.

Interpretation
Aflatoxin M1 Only three of the 100 samples contained quantifiable levels of AFM₁ (Table 4), all of which were conventional farm-gate milk (0.01 ? 0.021 microgram/kg). These results compare favourably with the 1995 survey where 1 percent of retail milk was greater than 0.05 microgram/kg. Fifty four percent of samples were between 0.01- 0.04 microgram/kg, and only 45 percent were below the detection limit of 0.01 microgram/kg.9 In the 1996 survey, four percent of farm-gate milk samples were over the EU limit, and 83 percent were below the detection limit.

These results demonstrate the continuing effectiveness of feed industry controls as a result of European Directive (EC) No. 1999/29/EEC Undesirable Substances in Animal Nutrition implemented in the UK by The Feeding Stuffs Regulations 2000 (S.I. No. [2000] 2481) which sets maximum limits for AFB₁ in feed

Conclusion
Only three of the 100 samples contained quantifiable levels of AFM₁ (Table 4) which were all below the EU limit of 0.05 microgram/kg. These results demonstrate the continuing effectiveness of the Feeding Stuffs Regulations limiting the amount of AFB1 contamination of feed.

The findings of this survey do not raise any food safety concerns.

The Food Standards Agency will continue to monitor for mycotoxins in milk.

Units of Measurement
Microgram : one millionth of a gram
Kilogram (kg): one thousand grams
Microgram per kilogram (microgram/kg)

References

1. Scudamore, K.A. Mycotoxins; An Independent Assessment of MAFF-Funded Applied Research and Surveillance, 1993-1996. London: MAFF, 1999.
2. Skaug, M.A. Analysis of Norwegian milk and infant formulas for ochratoxin A. Food Additives and Contaminants. 1999, 16 (2), 75-78.
3. Prelusky, D.B., Scott, P.M., Trenholm, H.L., Lawrence, G.A. Minimal transmission of zearalenone to milk of dairy cows. Journal of Environmental Science and Health. 1990, B25(1), 87-103.
4. Mirocha, C.J., Pathre, S.V., Robison, T.S. Comparative metabolism of zearalenone and transmission into bovine milk. Food Cosmet. Toxicology, 1981, 19, 25-30.
5. Hagler, W.M., Danko, G.Y., Horvath, L., Palyusik, M., Mirocha, C.J., Transmission of zearalenone and its metabolite into ruminant milk. Acta Veterinaria Academiae Scientiarum Hungaricae, 1980, 28 (2), 209-216.
6. Kim, E.K., Shon, D.H., Ryu, D., Park, J.W., Hwang, H.J., Kim, Y.B., Occurrence of aflatoxin M1 in Korean dairy products determined by Elisa and HPLC. Food Additives and Contaminants, 2000, 17 (1), 59-64.
7. The Feeding Stuffs Regulations 2000 (SI No [2000]2481).
8. Ministry of Agriculture, Fisheries and Food. Survey of aflatoxin M₁ in farm-gate milk. Food Surveillance Information Sheet, 78, 1996.
9. Miller, J.D., Trenholm, H.L. Mycotoxins in Grain Compounds Other than Aflatoxin. United States of America, Eagan Press, 1994.
10. MAFF. Surveillance of mycotoxins in the UK. 4th Report of the Steering Group on Surveillance, The Working Party on Mycotoxins, Food Surveillance Paper, No 4. London: HMSO, 1980.
11. Ministry of Agriculture, Fisheries and Food. Survey of aflatoxin M₁ in retail milk and milk products. Food Surveillance Information Sheet, 64, 1995.
12. Official Journal of the European Communities, L 201, (17/07/98), Commission Regulation (EC) No. 125/98, 43-46.
13. Anon. Aflatoxin M₁ standard approved by Codex. Food Chemical News, 18/07/2001.
14. Anon. EU Food Law. July 2001, Agra Europe (London),16.
15. World Health Organisation. Evaluation of Certain Food Additives and Contaminants. Forty-forth Report of the Joint FAO/WHO Expert Committee on Food Additives, 1996, WHO Technical report series, No.859, Geneva: WHO
16. European Commission Scientific Committee on Food (28 September 1998). Opinion of OA CS/CNTM/MYC/14
17. MAFF 1993. Mycotoxins: Third Report. The Thirty-sixth Report of the Steering Group on Chemical Aspects of Food Surveillance Sub Group on Mycotoxins. Food Surveillance Information Paper, No 36. London:HMSO