Description and chemistry of deoxynivalenol

4-Deoxynivalenol (DON), also known as vomitoxin belongs to trichothecenes chemical family: tetracyclic sesquiterpenes with 12,13-epoxy group, therefore the deoxynivalenol is a 12,13-epoxy-3,4,15-trihydroxytrichotec-9-en-8-one. It is produced by several Fusarium genus, especially Fusarium culmorum and Fusarium graminearum (Gibberella zeae). Fusarium is a soil fungi which grow on the crop in the field. grows optimally at a temperature of 21ºC and 0,87 of water activity. F. graminerum grows optimally at 25ºC and about 0,88 of water activity. Deoxynivalenol can be found as a natural contaminant in various cereals crops (maize, barley, oats, rye and wheat) and also in processed grains (malt, beer, bread, cookies, cakes, pastas, breakfast cereals biscuit and croissants).

DON is a white crystalline substance with a melting point of 151-153ºC. It is optically active and is soluble in ethanol, methanol, ethyl acetate, water and chloroform. DON has a great stability during storage/milling and in the processing cooking of food; it does not degrade at high temperatures (1,2,3).


Biochemical mode of action and immunotoxicity

DON has a potent immunosuppressive activity; it inhibits the synthesis of DNA and RNA and the protein synthesis at the ribosomal level. The IgA is notably affected. The humoral and cellular immunity is suppressed, and the susceptibility for infectious diseases is increased. The mycotoxin has a haemolytic effect on erythrocytes. It causes the gastroenteric syndrome affecting the digestive system, at high doses it can induce vomiting (emesis) in pigs and at lower concentrations DON reduces growth and feed consumption (1,2,4).

In 1993 the IARC (5) classified the DON in Category 3 (not classifiable as to its carcinogenicity to humans).There are no indications for carcinogenesis and/or mutagenic properties of DON, therefore, the immunotoxicity and the general toxicity of DON are the critical effects to be considered (5,6).


Toxicity, occurrence, legislation and effects in humans

The evaluation for human risks can be based by extrapolating the relevant toxicological endpoints such as the NOAEL (no observed adverse effect levels) or LOAEL (lowest observed adverse effect levels) from the toxicity studies in mouse and swine, using appropriate safety and / or uncertainty factors, generally ranging from 50 to 50000 with a risk level of 1/100000 which is considered as more appropriate for unavoidable natural contaminants (7).

The NOAELs values reported for DON are 40, 60, 100, 250, 500, and 375 micrograms/kg body weight (b.w.)/day, from the toxicity studies and the relevance of the toxicological endpoints in swine (subchronic, 85-100 days), swine (subchronic, 94-96 days), mouse (chronic, 2 years), mouse (immunotoxicity), mouse (teratogenicity), and mouse (reproduction toxicity), respectively, and according to the critical effects, such as reduced growth (8), reduced growth and effects on liver and serum albumin (9), reduced growth (6), increased susceptibility to infections (10), foetal skeleton abnormalities (11), and postnatal mortality (12), respectively.

The studies with pigs were not used to calculate the TDI (tolerable daily intake) because the feed used was naturally contaminated with DON and the toxicological influence due to the presence of other toxins was not excluded (8,9).

For the estimation of a provisional TDI (NOAEL/ uncertainty factor,UF) was decided to use the NOAEL (100 micrograms/kg b.w./day) of the chronic diet study with mouse (2 years) according to the critical effect of reduced growth. From the NOAEL divided by an uncertainty factor of 100, a provisional TDI (PTDI) of 1 microgram/Kg b.w./day was derived (1,3). This TDI is provisional because the studies of the chronic toxicity were done in one animal especies.

In several countries, the concentrations (micrograms/Kg) of DON which were found in cereals and some cereals-based products, varied between 1 to 5,700 in wheat; 3 to 3,370 in maize; 4 to 9,000 in barley; 4 to 760 in oat; 6 to 5,100 in rice and 13 to 240 in rye. In wheat-based breakfast cereals were found contamination levels of 103 to 6,040 micrograms/Kg (average of 754 micrograms/Kg) (13-16)

The European Communities (EC) has no legislation for DON neither in human foods nor in animal feeds. In 1999, in the Netherlands were proposed the following general DON concentrations (micrograms/Kg) limits of 120 for cleaned wheat, 60 for bread and 120 for food products with a wheat content higher than 33%. For food products with a wheat content lower than 33% it was suggested to monitor only the cleaned wheat used as an ingredient (17).

In Asia (during 1961 to 1985) there were 35 outbreaks of acute human disease attributed to consumption of grains (scabby wheat and mouldy corn) contaminated with DON at concentrations of 3,000 to 93,000 micrograms/Kg. The symptoms described were gastrointestinal upset, dizziness, vomiting, nausea, diarrhoea, headache, fever and blood in the stool. All of them were reversible effects because when the contaminated food was replaced with uncontaminated food, the signs and symptoms disappeared. At least 7,818 victims were affected and became ill from 5 to 30 min. after consumption of contaminated grains. No deaths were reported.
In 1987, an outbreak of disease with the same symptoms previously mentioned affected about 50,000 people (150 families) in India. The problem was attributed to the consumption of bread made from mouldy wheat and contaminated with DON at concentrations of 340 to 8,400 micrograms/Kg. The wheat was reported to contain other trichothecenes, such as acetyldeoxinivalenol, nivalenol, and T-2 toxin at concentrations of 600 to 2,400 micrograms/kg, 30 to 100 micrograms/Kg and 550 to 4,000 micrograms/Kg, respectively. Probably the other mycotoxins increased the DON toxicity (14,18).

The total intake of DON in the African and Middle Eastern diets was estimated in 0.77 and 2.4 micrograms/Kg b.w./day. We have no data about the total intake of DON in the European, Latin American and Far East diets, however the major source of intake in Latin American, European and Middle Eastern was wheat (64 to 88% of total intake), whereas the sources in Africa and Far East diets were more varied, wheat, rice, and maize; wheat and rice, respectively (14,19).


The risk for children

In the Netherlands is considered that the children 1-4 years old have the highest wheat intake, which is of 4.5 to 8.5 g/Kg b.w./day. In boys (1-4 years) the consumption of bread contributes for approximately 61% to the total wheat intake and the cookies, breakfast cereals and pasta contribute for approximately 5.8, 6.8 and 4.9% to the total wheat intake, respectively. In girls (1-4 years) the consumption of bread contributes for approximately 64% to the total wheat intake and the cookies, breakfast cereals and pasta contribute for approximately 4.8, 5.9 and 5.4% to the total wheat intake, respectively. Fast growing children can be vulnerable to growth reduction effect caused by DON and which one was previously mentioned in experimental animals (17).
Assuming a body weight of 10 kg, the contribution of the DON intake can be calculated and compared with the TDI. The DON concentrations (micrograms/kg) limits proposed in 1999 in the Netherlands for bread (60 micrograms/kg), cookies (120 micrograms/kg), breakfast cereals (120 micrograms/kg) and pasta uncooked (120 micrograms/kg), will be taking in account for calculating (17).

Boys and girls (1-4 years) with a body weight of 10 kg may intake daily 51g of bread + 72 g of wheat-based foods (cookies, breakfast cereals, pasta) and 46 g of bread + 46 g of wheat-based foods (cookies, breakfast cereals, pasta), respectively. Considering the bread and the wheat-based foods contaminated with DON at concentrations of 60 and 120 micrograms/kg, respectively (the limits previously referred), the DON intake daily would be of 11.7 micrograms for boys and 8.3 micrograms for girls. Dividing these data by 10 (average of body weight) the DON intake daily per kg body weight of boys and girls would be 1.17 and 0.83, respectively. Comparing these values with the TDI (PTDI) of 1 microgram/kg b.w./day, the daily DON intake for boys and girls would be 117 and 83% of the TDI (PTDI).
At these concentrations limits, adverse health effects are not expected for children and the general population. The previously DON concentrations limits were calculated considering a child with a high wheat consumption of 8.5 g/kg b.w./day, if we assume a child with a median wheat consumption of 4.5 g/kg b.w./day, the general concentrations limits are two-fold higher and considering growth retardation as a reversible toxic effect two-fold higher concentration limits may be considered temporarily acceptable (17,20). However these values are significantly lower than the established limits for DON of 2,000 and 1,000 micrograms/kg in United States and Canada, respectively, in wheat and its byproducts for human consumption (26).



Prevention and control

The formation of deoxynivalenol can be reduced applying preharvest measures to control infection with Fusarium spp by the reduction of the inoculum in host debris and other reservoirs in the field. For reducing the inoculum it is recommended the crop rotation and rotation of wheat and maize with non-host crops. Climate, sources of fungal inoculum and potential insect vectors can interact to produce specific mycotoxin occurrence, such as DON. The damage caused by insects to kernels may predispose them to infections during storage. The fungal infections in growing crops can be reduced by rapid drying and correct storage of the harvest crops and the appropriate use of antifungal agents at inhibitory concentrations. When mycotoxin contaminated grain is suspected or identified it can, in part, be reduced the contamination by the removal of contaminated grain by mechanical separation techniques. Physical, chemical and biological methods have been used for decontamination DON in cereals. Some of the treatments reduced the concentrations of DON, others were ineffective and others cannot yet applied on a commercial scale. The gravity separation and washing procedures, can reduce DON concentrations in wheat and maize. Thermal processing is usually ineffective. At the moment, there are not effective techniques for direct DON inactivation and/ or detoxification (20).


Analytical methods

According to our experience we recommend two methods for determination of deoxynivalenol in wheat, maize and wheat-based products. One of them by Thin Layer Chromatography (TLC)(21, 22), the other one by Immunoaffinity Column and Liquid Chromatography (HPLC) (24).


Thin Layer Chromatography

The mycotoxin DON is extracted from sample with 200 mL mixture of acetonitrile + water (84+16). The extract is filtered and 20 mL filtrate are collected in 25 mL graduated cylinder. For wheat, proceed with column chromatography cle anup. For corn, the extract is deffated with 50 mL of hexane and proceed with column chromatography cleanup.

The extract is passed through a column of mixed alumina + charcoal + Celite (0.5g + 0.7 g + 0.3 g). As solution reaches top of packed bed, rinse cylinder with 10 mL acetonitrile + water (84+16) and then add to column. The solvent is evaporated to dryness on a steam bath. Ethyl acetate is added to the residue and heated to dissolve DON. After cooling, the residue is transferred to a vial with additional ethyl acetate and the extract is evaporated to dryness on steam bath under stream nitrogen and is dissolved in chloroform + acetonitrile (4 + 1) for TLC on an AlCl3-impregnated silica gel plate developed with chloroform + acetone + isopropanol (8 + 1 + 1).


The plate is heated in a 120ºC oven for 7 min; a blue fluorescent spot is produced under longwave ultraviolet light. DON is quantitated visually and/or fluorodensitometrically by comparison with reference standards. The minimum detectable amount of DON is ca 20 ng/spot. The limit of DON determination is ca 40 micrograms/kg for wheat and 100 micrograms/kg for corn. Recoveries of DON added to wheat and corn at 100, 500, and 1000 micrograms/kg levels were 85, 93, and 88% and 77, 80, and 80%, respectively (21,22).

We recommended: I.- Deffating not only in corn but also in wheat and corn and wheat- based products. II.- Confirm by developing two plates more with chloroform + ethyl acetate + ethanol (140 +20 +40) and chloroform + methanol (93+7). III.- Perform the quantitative analysis fluorodensitometrically or by the limit of detection method (23).


Immunoaffinity Column and Liquid Chromatography (HPLC). Deoxynivalenol Levels in Wheat-Based Breakfast Cereals Marketed in Portugal.

The author refers to a part of the article "Determination of Deoxynivalenol in Wheat-Based Breakfast Cereals Marketed in Portugal" published by Maria Ligia Martins and H.Marina Martins in Journal of Food Protection, Vol. 64. No. 10, 2001. Pages 1848-1850. The partial reproduction of the original article was done with the permission of the authors previously mentioned.


ABSTRACT

Deoxynivalenol (DON), also known as vomitoxin, is one of a group of closely related secondary fungal metabolites - the trichothecenes, and is produced predominantly by several species of the genus Fusarium, especially F. graminearum. The present study was carried out to evaluate the natural occurrence of DON in different kinds of wheat-based breakfast cereals widely consumed by the population. A total of 88 commercially available samples of wheat - based breakfast were randomly collected from different supermarkets, in Lisbon, Portugal. The samples were analyzed using immunoaffinity column and DON was quantified by liquid chromatography. Detection limit was 100 micrograms/kg. Average recovery of DON was 80 %. Of 88 analyzed samples, 72.8% contained levels of DON between 103 and 6,040 micrograms/kg, with mean level of 754 mg/kg, and 24 samples (27.2%) were not contaminated (<100 micrograms/kg ). These results indicate an incidence of this mycotoxin in these products and the authors suggest a monitoring for the prevention of moulds and mycotoxins. This is the first report in Portugal on natural contamination with DON in wheat-based breakfast cereals.


MATERIALS AND METHODS

Samples.

A total of 88 packaged samples of commercial wheat cereals for breakfast (24 samples of bran, 20 wheat flakes, and 44 of wheat and fruits) were purchased in different supermarkets in Lisbon-Portugal. Each 250 g sample was ground in a blender, and subsamples of 50 g were analyzed.


DON determination and quantification by liquid chromatography.

DON analysis was carried out following the method described by Cahill et al.(24). A sample of 50 g was extracted in distilled water by blending for 3 min at high speed, filtered through both a fluted and glass microfiber filter paper and applied to an immunoaffinity column, (DONtest HPLC; VICAM, Watertown, Mass.). Subsequently the column was washed with distilled water and the toxin was eluted from the column with methanol, evaporated to dryness in a rotary evaporator and redissolved in 300 microliters of acetonitrile-water. Determination of DON was carried out by isocratic reverse-phase liquid chromatography using a LiChrospher 100 RP-18, 5 mm column 25 x 4.6 mm EcoPack (Merck, Portugal). The mobile phase was acetonitrile-water, filtered through a 0.22-mm filter membrane, degassed, and used at a flow rate of 0.6 ml/min. DON was detected using a Merck - Hitachi L7420 UV detector set to 218 nm. Data were analyzed with a computing integrator (Compaq Deskpro, Merck-Hitachi).
DON was obtained from Sigma-Aldrich (Spain), D-0156. Working standards solutions, limit of detection, and percentage recovery were determined according to the method previously referred to herein. The limit of detection was 100 micrograms/kg. Recovery was determined by spiking DON standards at levels of 100.0, 250.0, and 500.0 micrograms/kg to wheat samples. Recovery averages were 98.0, 85.0, and 88.0 % respectively.


RESULTS AND DISCUSSION

The analysis of the 88 samples of wheat and bran showed that 72.8% (64 samples) were contaminated with DON. The detected levels ranged between 103 and 6,040 micrograms/kg. Twenty-four samples ( 27.2 %) did not reveal the presence of toxin (<100 micrograms/kg).
A summary of the results for bran, wheat flakes and wheat and fruits is shown in Table 1. Of 24 samples of bran, eight (33.3%) were not contaminated with DON. Eight samples (33.3%) contained levels between 101 and 1,000 micrograms/kg, four samples (16.7%) were contaminated with levels of 1,001 to 5,000 micrograms/kg and four others (16.7%) contained DON at >5,001 micrograms/kg. The incidence of DON on 20 samples of wheat flakes analyzed showed that four samples (20%) were negatives (< 100 micrograms/kg). Fourteen samples (70.0%) were contaminated with levels ranging from 101 to 1,000 micrograms/kg, and two samples (10.0%) had levels between 1,001 and 5,000 micrograms/kg. Concerning to the 44 wheat and fruits samples, 12 (27.3%) were not contaminated, 26 samples (59.1%) had levels of 101 up to 1,000 micrograms/kg, and 6 ( 13.6%) had contamination between 1,001 to 5,000 micrograms/kg (Table 1). Dalcero et al.(25) studied DON contamination in wheat from Cordoba, Argentina during 1993 to 1994 harvest season. In 40 samples analyzed, they found levels ranged between 300 and 4,500 micrograms/kg. These results agree with those obtained by our screening.
The levels of contamination are important, considering that United States and Canada have established limits for DON of 2,000 and 1,000 micrograms/kg, respectively, in wheat and its byproducts for human consumption (26).
Reports from the United States (27) have shown that about 40% of the 483 wheat samples from the 1993 crop year contained DON levels that were greater than advisory levels of 2,000 micrograms/kg (26). The results suggest a risk for consumers of wheat products and the need to monitor final products before consumption.


Table 1

		DON Levels in micrograms/kg
Commodity	Incidence	ND <100 (%)	101 to 1,000 (%)	1,001 to 5,000 (%)	> 5,001(%)
Bran	16/24	8 (33.3)	8 (33.3)	4 (16.7)	4 (16.7)
Wheat flakes	16/20	4 (20.0)	14 (70.0)	2 (10.0)	-
Wheat and fruits	32/44	12 (27.3)	26 (59.1)	6 (13.6)	-
Total	64/88	24 (27.3)	48 (54.5)	12 (13.6)	4 (4.6)

ND: not detected, < 100 micrograms/kg

Bibliography

(1) Eriksend, G.S and Alexander, J (eds). (1998). Fusarium toxins in cereals - a risk assessment. Nordic Council of Ministers; TemaNord 1998: 502, pp.7-27 and 45-48: Copenhagen.

(2) Rotter, B.A., Prelusky, D.B., Pestka, J.J. (1996). Toxicology of deoxynivalenol (vomitoxin). J. Toxicol Environ Health, 48:1-34.

(3) Ehling, G., Cockburn, A., Snowdon, P., and Buchhaus,H. (1997) The significance of the Fusarium toxin deoxynivalenol (DON) for human and animal health. Cereal Research Commun, 25:443-447.

(4) Deijns, A.J., Egmond, H.P. van., Speijers, G.A.J and Loveren, H. van. (1997). Immunotoxiciteit van natuurlijke toxinen. Een literatuur overzicht. RIVM-rapport 388802007, pp. 16-17. Rijks Instituut voor Volksgezondheid en Milieu, Bilthoven.

(5) IARC. (1993). Monographs on the evaluation of carcinogenic risks to humans; Vol. 56: Some naturally occurring substances, food items and constituents, heterocyclic aromatic amines and mycotoxins. International Agency for Research on Cancer, World Health Organization, pp. 397-333: Lyon.

(6) Iverson,F., Amstrong, C., Nea, E., Truelove, J., Fernie, S., Scott, P.M., Stapley, R., Hayward, S and Gunner, S. (1995). Chronic feeding study of deoxynivalenol in B6C3F1 male and female mice. Teratogenesis Carcinogenesis Mutagenesis, 15: 283-306.

(7) Kuiper-Goodman, T. (1990). Uncertainties in the risk assessment of three mycotoxins: aflatoxin, ochratoxin, and zearalenone. Can. J. Physiol. Pharmacol., 68: 1017-1024

(8) Bersjo, B., Matre, T., and Nafstad, I. (1992). Effects of diets with graded levels of deoxynivalenol on performance in growing pigs. J. Vet. Med., A39: 752-758.

(9) Bersjo, B., Langseth, W., Nafstad, I., Hogset Jansen, J., and Larsen, H.J.S. (1993). The effects of naturally deoxynivalenol-contaminated oats on the clinical condition, blood parameters, performance and carcass composition of growing pigs. Vet. Res. Commun., 17: 283-294.

(10) Tryphonas, H., Iverson, F., Ying So, E.A., MgGuire, P.F., O´Grady, L., Clayson, D.B., and Scott, P.M. (1986). Effects of deoxynivalenol (vomitoxin) on the humoral and cellular immunity of mice. Toxicol. Lett., 30: 137-150.

(11) Khera, K.S., Whalen, C., Angers, G., Vesonder, R.F., and Kuiper-Goodman, T. (1982). Embryotoxicity of 4-deoxynivalenol (vomitoxin) in mice. Bull. Environm. Contam. Toxicol., 29 : 487-491.

(12) Khera, K.S., Arnold, D.L., Whalen, C., Angers, G., and Scott, P.M. (1984). Vomitoxin (4-deoxynivalenol): effects on reproduction of mice and rats. Toxicol. Appl. Pharmacol., 74: 345-356.

(13) COST (European cooperation in the field of scientific and technical research). (2001). Occurrence of toxicogenic fungi and mycotoxins in plants, food and feed in Europe. In: A.Logrieco (Ed.), Agriculture and biotechnology. European Commision (COST Action 835), Luxembourg, pp. 1-207.

(14) JECFA (Joint FAO/WHO Expert Committee on Food Additives). (2001). Fifty-sixth meeting, Geneva, 6-15 February 2001, pp.1-33.

(15) Martins, M.L., and Martins,H.M. (2001). Determination of deoxinivalenol in wheat-based breakfast cereals marketed in Portugal. J. Food Prot. 64: 1848-1850.

(16) WHO (World Health Organization) (2002). Evaluation of Certain Mycotoxins in Food. Fifty-sixth report of the Joint FAO/WHO Expert Committee on Food Additives. WHO Technical Report Series 906. Geneva, pp. 1-62

(17) Pieters, M.N., Fiolet, D.C.M., and Baars,A.J. (1999). Deoxynivalenol: Derivation of concentration limits in wheat and wheat containing food products. RIVM (Rijks Instituut voor Volksgezondheid en Milieu), RIVM report 388802 018. Bilthoven. October, 1999. pp.1-32.

(18) Kuiper-Goodman, T., 1994. Prevention of Human Mycotoxicoses Trough Risk Assesment and Risk Management. In: J.D.Miller and H.L.Trenholm (eds.), Mycotoxins In Grain, Compouns Other Than Aflatoxin. Eagan Press, St.Paul, Minnesota, pp. 439-469.

(19) WHO (World Health Organization), 2002. Evaluation of Certain Mycotoxins in Food. Fifty-sixth report of the Joint FAO/WHO Expert Committee on Food Additives. WHO Technical Report Series 906. Geneva, pp. 1-62

(20) Gimeno,A., and Ligia Martins, M. (2003) Risk Analysis of the most Relevant Human Mycotoxicosis. I Panamerican Symposium on Mycotoxins for Industry. April 1-4, 2003. Mexico City. Abstract in Proceedings book, p. 34. The complete article " Analisis de Riesgo de las mas Relevantes Micotoxicosis en Humanos " had been submmited for publication in the Revista Mejicana de Micologia.

(21) Trucksess, M.W., Nesheim, S., and Eppley, R.M. (1984). Thin Layer Chromatographic Determination of Deoxynivalenol in Wheat and Corn. J.Assoc.Off.Anal. Chem. 67, 40?43.

(22) Official Methods of Analysis. (1990). Deoxynivalenol in Wheat, Thin Layer Chromatography Method, First Action 1986. 15 th Ed., AOAC, Sec. 49. Natural Poisons (Peter M.Scott, Associate Chapter Editor), Sec. 986.17, pp. 1205-1206.

(23) Gimeno, A., and Martins, M.L. (2002). Metodologias para el Control de Micotoxinas en Alimentos Compuestos y Materia primas. In www.engormix.com (Ir a: micotoxinas). Consulted in 16-05-2003.

(24) Cahill, L.M., Kruger, S.C., McAlice, B.T., Ramsey,C.S., Prioli, R., and Kohn, B. (1999). Quantification of deoxynivalenol in wheat using immunoaffinity column and liquid chromatography. Journal of Chromatography, 859, 23-28.

(25) Dalcero, A., Torres, A., Etcheverry, M., Chulze, S., and Varsavsky, E. (1997). Occurrence of deoxynivalenol and Fusarium graminearum in Argentinian wheat. Food Add. Contam. 14 : 11-14.

(26) FAO (Food and Agriculture Organisation), (1997). Worldwide Regulations for Mycotoxin 1995. A compendium. FAO Food and Nutrition: Paper 64. (Rome: Food and Agriculture Organization of the United Nations).

(27) Trucksess, M. W., Thomas, F., Young, K., Stack, M.E., Fulgueras, W.J., and Page, S.W. (1995). Survey of deoxynivalenol in U.S.1993. wheat and barley crops by enzyme- linked immunosorbent assay. J. Assoc. Off. Anal. Chem. Intern. 78: 631-636.