- Email: [email protected]
The effects of bakery processing on natural deoxynivalenol contamination
lntemational Joumal ofFeodrvlicrobidogy International Journal of Food Microbiology
37
( 1997) 21-25
The effects of bakery processing on natural deoxynivalenol contamination M.S. Neira”‘b, A.M. Patina”,
Received
2 September
1996; received
E.J. Martinez’,
G. Moltb’, S.L. Resnik”“‘*
in revised form 14 February
1997: accepted
17 March 1997
Abstract The aim of this study was the evaluation of the influence of the breadmaking process on initial deoxynivalenol (DON) contamination. Samples (92) were taken from four batches of eight different types of products in a low-technology bakery. The final products, as well as the corresponding flours, doughs and fermented doughs were analyzed. Extracts were obtained with acetonitrile:water (84: l6), the clean up was made with a multifunctional column and DON was quantified by thin layer chromatography by visual comparison with standards. Confirmation was made by electron capture gas chromatography. The contamination levels in flour samples ranged from 500 pg/kg to 2000 kg/kg on dry weight basis. The results showed a positive correlation between the initial contamination level and the reduction of DON after fermentation. A significant reduction was observed as a consequence of the breadmaking process. 0 I997 Elsevier Science B.V. Ke~~n~d.s: Deoxynivalenol:
Wheat: Bakery products
1. Introduction In Argentina, wheat is one of the most important cereals, not only because of its high local consumption (Gallo et al., 1992) but also because of its significance in international trade. Different sorts of breads and pastries are regularly consumed by the population. Within this context, deoxynivalenol (DON), a naturally occurring metabolite produced by ‘kCorresponding author. UBA Ciudad Universitaria,
Departamento de Industrias, FCEyN, 1428 Buenos Aires, Argentina
0168.16OS/97/Sl7.00 0 1997 Elwvier P/I SOl68-1605~97)00018-X
Science B.V. All right\
the genus Fusarium, is a widespread contaminant of wheat. Toxicological effects of DON (Ueno, 1983; Pacin, 1992) have been shown and immunological disorders have also been indicated (Tryphonas et al., 1984). Many physical and chemical methods have been used to reduce mycotoxin contamination of feedstuffs. such as segregation of contaminated from non-contaminated kernels in water and saturated sodium chloride, milling, cleaning or washing, sieving and dehulling (Trenholm et al., 1991; Scott, 199 I ). Bakery processing has been reported to reserved
reduce DON contamination. On the other hand. El Banna et al. ( 1983) and Scott ( 199 1) suggested that DON is highly stable to bakery processes. Concerns related to DON contamination of wheat have emphasized the need for studies to assess the human intake of this mycotoxin. The aim of this study was to evaluate the influence of breadmaking processes on the level of initial DON contamination of bakery products.
2. Material
and methods
Samples were purchased locally at a traditional bakery. Wheat flours. doughs. fermented doughs and baked products of eight different types were sampled from four consecutive batches. amounting to a total of 92 samples. Samples were taken at three different stages of the process: 31 at the end of the dough preparation (doughs). 31 after the fermentation (fermented doughs) and 30 after baking (baked products). The products considered were bran bread, butter croissant, fat croissant, ‘figazzas’ (french bread dough plus fat), french bread, homemade bread. ‘libritos’ (wheat flour kneaded with water, fat, salt, malt and yeast) and Vienna bread. Fermentation times varied between 5 and 11 h at 25°C and baking between IO and 40 min at 2 10°C. The moisture contents expressed in percentage. were determined at each stage and ranged from 28.9 to 42.7 for doughs. from 28.9 to 39.4 for fermented doughs and from 19.3 to 30.5 for baked products. Approximately 300 grams of representative subsampies were stored at - 20°C until analyses were performed.
Prior to the acetonitrile:water extraction, doughs, fermented doughs and baked products were defatted with three successive extractions with 100, SO and 25 ml of n-hexane. The water component of the extraction acetonitrile:water (84: 16) mixture was adjusted to balance the water content of each sample and then blended for 3 min at high speed with an Osterizer blender.
The extract was filtered through Whatman No4 paper. Approximately 8 ml of the filtrate were placed in a IS X 85 mm culture tube and 4 ml of the filtrate was passed through a Mycosep 225 column (Romcr Labs.. Inc.. MO. USA). The purified extract was then transferred to another tube and evaporated to dryness in a 60°C water bath under vacuum (Wilson and Romer. I99 I ). Detection and quantification of DON were carried out by thin layer chromatography (TLC) and verified by gas chromatography with electron capture detection (Scott and Kanhere, 1986; Scott et al., 1989; Croteau et al., 1994; Pacin et al., 1997). Merck silica gel tic plates (No. 5553) and toluene:acetone ( I:2) developing solvents were used in the TLC analysis. After development, plate\ were air dried, sprayed with 20’% alcl i ethanol:water ( 1: 1) solution and heated at 150”~ for 10 min. The levels of DON were estimated visually (R, = 0.5) by comparison with standard spots under ion g wave UV light (a 25 p,l Hamilton syringe was used for spotting 20 pJ of each sample. along with I, 5. 10 and 20 ~1 of a SO yg/ml of DON standard solution). A Hewlett-Packard gas chromatography model 5890 Series II was used for DON confirmation or to evaluate the TLC DON quantitication, under the following conditions: column, Hewlett-Packard HP-S capillary column (25 m by 0.2 mm (i.d.) and 0.5 km film thickness}: carrier. nitrogen gas at a flow-rate of 1 ml/min and the nitrogen make-up gas at a flowrate of SO ml/min: column temperature. I min at 120°C. then 30”C/min to 200°C. held for 2 min, 5”C/min to 2SO”C, held for I 1 min. The injector and ECD temperatures were 250°C and 300°C respectively. The dried extract residues were derivatired with HT.2 toxin as internal standard with heptaHurobutyric anhydride acid in the presence of catalyst solution (2 mg of l-N.N-dimethyl aminopyridine in I ml of toluene:acetonitrile (95:s)) prior to the in,jection.
To describe the different sets of data boxplots were used. A boxplot is a graphical display invented by Tukey ( 1977) that shows a measure of dispersion (the interquartile range) represented by the width of the box, a measure of location (the median) repre-
MS. Neira et al. I Internchv~nl
Journul oj’ Food Microbiology
sented by a line inside the box, and the presence of possible outliers, represented by asterisks. It also gives an indication of the symmetry or skewness of the distribution. To test differences between central values of two independent groups, the T-test was used in case both groups could be considered normal. To test the hypothesis of normality the Wilk-Shapiro test was applied (Shapiro and Francis, 1972).
3. Results and discussion All flour and bran samples examined were contaminated with deoxynivalenol in concentrations ranging from 500-1000 pg/kg (results not shown). In Fig. 1, box-plots of DON contamination in doughs, fermented doughs and bakery products are presented. Mean and median values for doughs were 1370 kg/kg and 1450 p+g/kg, for fermented doughs 1020 pg/kg and 1060 kg/kg and for bakery products 710 kg/kg and 700 kg/kg, respectively. The Wilk-Shapiro test was applied to the three data sets: in all cases the hypothesis of normality was not rejected. By T-test the mean DON level of baked doughs was less than that of fermented doughs (P <
Fig. I. doughs.
Box and Whisker Plot of deoxynivalenol C = baked product\.
contamination
37 (1997) 21-25
23
O.OOOl), which was less than that of doughs (P < 0.0001). There was no observed relationship between the degree of reduction and the initial level of DON contamination during baking. And, as it was expected, a positive relationship was observed between the increase of baking time and that of the percentage of reduction on DON-contaminated bakery products (data not shown). On the other hand, during fermentation it was found that the percentage of DON reduction increased with the initial level of DON contamination. This was confirmed by the estimated linear trend shown in Fig. 2. This result did not agree with those of Young et al. (1984) and Scott et al. (1992). The percentages of DON reduction between doughs and fermented doughs, fermented doughs and baked products, and doughs and baked products are shown in Fig. 3. The mean reduction between doughs and fermented doughs and between fermented doughs and baked products were 2 1.6% and 28.9%, respectively. Finally, the mean and median reduction between doughs and final products were 44.3% and 38.0%, with a maximun and a minimum reduction rates of 96.6% and 16.8%, respectively. Further studies must be performed under controlled
at different stages of breadmaking
processes.
A = doughs, B = fermented
24
+
60
+
g
+
+
c 2 ;
+
+
40
% L.
%
s: 20
+
+
+
O200
1600
900
initial Fip. 2. Percentage deoxynivalenol
reduction
2300
DON (ug/kg)
vs. lnltial
DON
contanmation
during
fermentation
procrs~~
100
* 80
*
od 3 Y a z
60
4
40
20
0
Fig.
3. Box
reduction.
and Whisker
B =
fermented
plot
of the percentage
doughs-baked
products
of deoxynivalenol reduction,
C 2
reduction
on bakery
doughs-baked
products
proceuing. reduction.
A
=
dough\-femlentrd
doughs
experimental conditions to determine ture of this phenomenon.
the exact na-
Acknowledgments The authors wish to acknowledge Mrs. G. Cano for her technical assistance and Consejo National de Investigaciones Cientificas y Tecnicas, Comisibn de Investigaciones Cientificas de la Provincia de Buenos Aires, Merck Quimica Argentina, Universidad de Buenos Aires for financial support.
References Croteau. M.S., Prelusky, D.B.. Trenholm. H.L.. 1994. Analysis of trichothecene mycotoxins by gas chromatography with electron capture detection. J. Agric. Food Chem. 42, 9288933. El Banns. A.A., Lau, P.Y.. Scott, P.M., 1983. Fate of mycotoxins during processing of foodstuffs II- Deoxynivalenol (vomitoxin) during making of Egyptian bread. J. Food Protect. 46, 4844486. Gallo. A., Bourdignon. F., Pacin, A.. Barbieri, T., Resnik, S., 1992. Evaluation of food intake by means of 24 hour dietary recall in a town of the provmce of Buenos Aires, Argentina. Ecol. Food Nutr. 28, 2999317. Pacin. A., 1992. Las micotoxinas coma factor de riesgo en la salud humana. In: Memorias Seminario International sobre Micotoxinas. Asociacion Colombiana de Postcosecha de Granos, Santafe de Bogota, Colombia, abril 1991. IDEMA, pp. 87-108. Pacin. A.M., Resnik, S.L., Neira, M.S., Molto, G., Martinez, E.. 1997. Natural occurrence of deoxynivalenol in wheat, wheat flour and bakery products in Argentina. Food Add. Contam. (In press).
Scott, P.M., Lombaert, G.A., Pellaers. P., Bacler, S., Kanhere, S.R., Sun. W.F., Lau. P.Y., Weber, D., 1989. Application of capillary gas chromatography to survey of wheat for live trichothecenes. Food Addit. Contam. 6, 4X9-.500. Scott, P.M., Kanhere, S.R., 1986. Determination of nivalenol and deoxymvalenol in cereals by electron-capture gas chromatography. J. Assoc. Off. Anal. Chem. 69. 889-893. Scott, P.M., 1991. Possibilities of reduction or elimination of mycotoxins present in cereal grains. In: Chelkow,aki, J. (Ed.), Cereal grains. Mycotoxins. fungi and quality in drying and storage. Elsevier, Amsterdam, pp. 529-572. Scott, P.M., Kanhere, S.R., Daley, E.F., Farber, J.M., 1992. Fermentation of wart containing deoxynivalenol and searalenone. Mycotoxin Res. 8, 58866. Shapiro, S.S., Francia, R.S., 1972. An approximate analysis of variance test for normality. J. Am. Statist. Assoc. 67. 2lS216. Trenholm. H.L., Charmley, L.L., Prelusky. D.B., Warner. R.M., 1991. Two physical methods for decontamination of four cereals contaminated with deoxymvalenol and zearalenone. J. Agric. Food Chem. 39, 356-360. Tryphonas. H., O’Grady, L., Arnold, D.L., McGuire. P.F.. Karpinski, K., Vesonder, R.F.. 1984. Effects of deoxynivalenol (vomitoxin) on the humoral immunity of mice. Toxicol. Lett. 23. 17-24. Tukey. J.W.. 1977. Exploratory Data Analysis. Massachusetts, Addison-Wesley Publishing Company. Ueno Y. (Ed.). 1983. Trichothecenes. Chemical, Btological and Toxicological Aspects. Tokyo, Kodansha. Wilson. T.J.. Romer, T.R., 1991. Use of the Mycosep multifunctional column for liqutd chromatographic determination of allatoxins in agricultural products. J. Assoc. Off. Anal. Chem. 74. 95 I-956. Young, J.C., Fulcher, R.G., Hayhoe, J.H., Scott. P.M., Dexter, J.E., 1984. Effect of mtlling and bakin g on deoxynivalenol (vomitoxin) content of Eastern Canadian wheat. J. Agric. Food Chem. 32, 6599664.
37
( 1997) 21-25
The effects of bakery processing on natural deoxynivalenol contamination M.S. Neira”‘b, A.M. Patina”,
Received
2 September
1996; received
E.J. Martinez’,
G. Moltb’, S.L. Resnik”“‘*
in revised form 14 February
1997: accepted
17 March 1997
Abstract The aim of this study was the evaluation of the influence of the breadmaking process on initial deoxynivalenol (DON) contamination. Samples (92) were taken from four batches of eight different types of products in a low-technology bakery. The final products, as well as the corresponding flours, doughs and fermented doughs were analyzed. Extracts were obtained with acetonitrile:water (84: l6), the clean up was made with a multifunctional column and DON was quantified by thin layer chromatography by visual comparison with standards. Confirmation was made by electron capture gas chromatography. The contamination levels in flour samples ranged from 500 pg/kg to 2000 kg/kg on dry weight basis. The results showed a positive correlation between the initial contamination level and the reduction of DON after fermentation. A significant reduction was observed as a consequence of the breadmaking process. 0 I997 Elsevier Science B.V. Ke~~n~d.s: Deoxynivalenol:
Wheat: Bakery products
1. Introduction In Argentina, wheat is one of the most important cereals, not only because of its high local consumption (Gallo et al., 1992) but also because of its significance in international trade. Different sorts of breads and pastries are regularly consumed by the population. Within this context, deoxynivalenol (DON), a naturally occurring metabolite produced by ‘kCorresponding author. UBA Ciudad Universitaria,
Departamento de Industrias, FCEyN, 1428 Buenos Aires, Argentina
0168.16OS/97/Sl7.00 0 1997 Elwvier P/I SOl68-1605~97)00018-X
Science B.V. All right\
the genus Fusarium, is a widespread contaminant of wheat. Toxicological effects of DON (Ueno, 1983; Pacin, 1992) have been shown and immunological disorders have also been indicated (Tryphonas et al., 1984). Many physical and chemical methods have been used to reduce mycotoxin contamination of feedstuffs. such as segregation of contaminated from non-contaminated kernels in water and saturated sodium chloride, milling, cleaning or washing, sieving and dehulling (Trenholm et al., 1991; Scott, 199 I ). Bakery processing has been reported to reserved
reduce DON contamination. On the other hand. El Banna et al. ( 1983) and Scott ( 199 1) suggested that DON is highly stable to bakery processes. Concerns related to DON contamination of wheat have emphasized the need for studies to assess the human intake of this mycotoxin. The aim of this study was to evaluate the influence of breadmaking processes on the level of initial DON contamination of bakery products.
2. Material
and methods
Samples were purchased locally at a traditional bakery. Wheat flours. doughs. fermented doughs and baked products of eight different types were sampled from four consecutive batches. amounting to a total of 92 samples. Samples were taken at three different stages of the process: 31 at the end of the dough preparation (doughs). 31 after the fermentation (fermented doughs) and 30 after baking (baked products). The products considered were bran bread, butter croissant, fat croissant, ‘figazzas’ (french bread dough plus fat), french bread, homemade bread. ‘libritos’ (wheat flour kneaded with water, fat, salt, malt and yeast) and Vienna bread. Fermentation times varied between 5 and 11 h at 25°C and baking between IO and 40 min at 2 10°C. The moisture contents expressed in percentage. were determined at each stage and ranged from 28.9 to 42.7 for doughs. from 28.9 to 39.4 for fermented doughs and from 19.3 to 30.5 for baked products. Approximately 300 grams of representative subsampies were stored at - 20°C until analyses were performed.
Prior to the acetonitrile:water extraction, doughs, fermented doughs and baked products were defatted with three successive extractions with 100, SO and 25 ml of n-hexane. The water component of the extraction acetonitrile:water (84: 16) mixture was adjusted to balance the water content of each sample and then blended for 3 min at high speed with an Osterizer blender.
The extract was filtered through Whatman No4 paper. Approximately 8 ml of the filtrate were placed in a IS X 85 mm culture tube and 4 ml of the filtrate was passed through a Mycosep 225 column (Romcr Labs.. Inc.. MO. USA). The purified extract was then transferred to another tube and evaporated to dryness in a 60°C water bath under vacuum (Wilson and Romer. I99 I ). Detection and quantification of DON were carried out by thin layer chromatography (TLC) and verified by gas chromatography with electron capture detection (Scott and Kanhere, 1986; Scott et al., 1989; Croteau et al., 1994; Pacin et al., 1997). Merck silica gel tic plates (No. 5553) and toluene:acetone ( I:2) developing solvents were used in the TLC analysis. After development, plate\ were air dried, sprayed with 20’% alcl i ethanol:water ( 1: 1) solution and heated at 150”~ for 10 min. The levels of DON were estimated visually (R, = 0.5) by comparison with standard spots under ion g wave UV light (a 25 p,l Hamilton syringe was used for spotting 20 pJ of each sample. along with I, 5. 10 and 20 ~1 of a SO yg/ml of DON standard solution). A Hewlett-Packard gas chromatography model 5890 Series II was used for DON confirmation or to evaluate the TLC DON quantitication, under the following conditions: column, Hewlett-Packard HP-S capillary column (25 m by 0.2 mm (i.d.) and 0.5 km film thickness}: carrier. nitrogen gas at a flow-rate of 1 ml/min and the nitrogen make-up gas at a flowrate of SO ml/min: column temperature. I min at 120°C. then 30”C/min to 200°C. held for 2 min, 5”C/min to 2SO”C, held for I 1 min. The injector and ECD temperatures were 250°C and 300°C respectively. The dried extract residues were derivatired with HT.2 toxin as internal standard with heptaHurobutyric anhydride acid in the presence of catalyst solution (2 mg of l-N.N-dimethyl aminopyridine in I ml of toluene:acetonitrile (95:s)) prior to the in,jection.
To describe the different sets of data boxplots were used. A boxplot is a graphical display invented by Tukey ( 1977) that shows a measure of dispersion (the interquartile range) represented by the width of the box, a measure of location (the median) repre-
MS. Neira et al. I Internchv~nl
Journul oj’ Food Microbiology
sented by a line inside the box, and the presence of possible outliers, represented by asterisks. It also gives an indication of the symmetry or skewness of the distribution. To test differences between central values of two independent groups, the T-test was used in case both groups could be considered normal. To test the hypothesis of normality the Wilk-Shapiro test was applied (Shapiro and Francis, 1972).
3. Results and discussion All flour and bran samples examined were contaminated with deoxynivalenol in concentrations ranging from 500-1000 pg/kg (results not shown). In Fig. 1, box-plots of DON contamination in doughs, fermented doughs and bakery products are presented. Mean and median values for doughs were 1370 kg/kg and 1450 p+g/kg, for fermented doughs 1020 pg/kg and 1060 kg/kg and for bakery products 710 kg/kg and 700 kg/kg, respectively. The Wilk-Shapiro test was applied to the three data sets: in all cases the hypothesis of normality was not rejected. By T-test the mean DON level of baked doughs was less than that of fermented doughs (P <
Fig. I. doughs.
Box and Whisker Plot of deoxynivalenol C = baked product\.
contamination
37 (1997) 21-25
23
O.OOOl), which was less than that of doughs (P < 0.0001). There was no observed relationship between the degree of reduction and the initial level of DON contamination during baking. And, as it was expected, a positive relationship was observed between the increase of baking time and that of the percentage of reduction on DON-contaminated bakery products (data not shown). On the other hand, during fermentation it was found that the percentage of DON reduction increased with the initial level of DON contamination. This was confirmed by the estimated linear trend shown in Fig. 2. This result did not agree with those of Young et al. (1984) and Scott et al. (1992). The percentages of DON reduction between doughs and fermented doughs, fermented doughs and baked products, and doughs and baked products are shown in Fig. 3. The mean reduction between doughs and fermented doughs and between fermented doughs and baked products were 2 1.6% and 28.9%, respectively. Finally, the mean and median reduction between doughs and final products were 44.3% and 38.0%, with a maximun and a minimum reduction rates of 96.6% and 16.8%, respectively. Further studies must be performed under controlled
at different stages of breadmaking
processes.
A = doughs, B = fermented
24
+
60
+
g
+
+
c 2 ;
+
+
40
% L.
%
s: 20
+
+
+
O200
1600
900
initial Fip. 2. Percentage deoxynivalenol
reduction
2300
DON (ug/kg)
vs. lnltial
DON
contanmation
during
fermentation
procrs~~
100
* 80
*
od 3 Y a z
60
4
40
20
0
Fig.
3. Box
reduction.
and Whisker
B =
fermented
plot
of the percentage
doughs-baked
products
of deoxynivalenol reduction,
C 2
reduction
on bakery
doughs-baked
products
proceuing. reduction.
A
=
dough\-femlentrd
doughs
experimental conditions to determine ture of this phenomenon.
the exact na-
Acknowledgments The authors wish to acknowledge Mrs. G. Cano for her technical assistance and Consejo National de Investigaciones Cientificas y Tecnicas, Comisibn de Investigaciones Cientificas de la Provincia de Buenos Aires, Merck Quimica Argentina, Universidad de Buenos Aires for financial support.
References Croteau. M.S., Prelusky, D.B.. Trenholm. H.L.. 1994. Analysis of trichothecene mycotoxins by gas chromatography with electron capture detection. J. Agric. Food Chem. 42, 9288933. El Banns. A.A., Lau, P.Y.. Scott, P.M., 1983. Fate of mycotoxins during processing of foodstuffs II- Deoxynivalenol (vomitoxin) during making of Egyptian bread. J. Food Protect. 46, 4844486. Gallo. A., Bourdignon. F., Pacin, A.. Barbieri, T., Resnik, S., 1992. Evaluation of food intake by means of 24 hour dietary recall in a town of the provmce of Buenos Aires, Argentina. Ecol. Food Nutr. 28, 2999317. Pacin. A., 1992. Las micotoxinas coma factor de riesgo en la salud humana. In: Memorias Seminario International sobre Micotoxinas. Asociacion Colombiana de Postcosecha de Granos, Santafe de Bogota, Colombia, abril 1991. IDEMA, pp. 87-108. Pacin. A.M., Resnik, S.L., Neira, M.S., Molto, G., Martinez, E.. 1997. Natural occurrence of deoxynivalenol in wheat, wheat flour and bakery products in Argentina. Food Add. Contam. (In press).
Scott, P.M., Lombaert, G.A., Pellaers. P., Bacler, S., Kanhere, S.R., Sun. W.F., Lau. P.Y., Weber, D., 1989. Application of capillary gas chromatography to survey of wheat for live trichothecenes. Food Addit. Contam. 6, 4X9-.500. Scott, P.M., Kanhere, S.R., 1986. Determination of nivalenol and deoxymvalenol in cereals by electron-capture gas chromatography. J. Assoc. Off. Anal. Chem. 69. 889-893. Scott, P.M., 1991. Possibilities of reduction or elimination of mycotoxins present in cereal grains. In: Chelkow,aki, J. (Ed.), Cereal grains. Mycotoxins. fungi and quality in drying and storage. Elsevier, Amsterdam, pp. 529-572. Scott, P.M., Kanhere, S.R., Daley, E.F., Farber, J.M., 1992. Fermentation of wart containing deoxynivalenol and searalenone. Mycotoxin Res. 8, 58866. Shapiro, S.S., Francia, R.S., 1972. An approximate analysis of variance test for normality. J. Am. Statist. Assoc. 67. 2lS216. Trenholm. H.L., Charmley, L.L., Prelusky. D.B., Warner. R.M., 1991. Two physical methods for decontamination of four cereals contaminated with deoxymvalenol and zearalenone. J. Agric. Food Chem. 39, 356-360. Tryphonas. H., O’Grady, L., Arnold, D.L., McGuire. P.F.. Karpinski, K., Vesonder, R.F.. 1984. Effects of deoxynivalenol (vomitoxin) on the humoral immunity of mice. Toxicol. Lett. 23. 17-24. Tukey. J.W.. 1977. Exploratory Data Analysis. Massachusetts, Addison-Wesley Publishing Company. Ueno Y. (Ed.). 1983. Trichothecenes. Chemical, Btological and Toxicological Aspects. Tokyo, Kodansha. Wilson. T.J.. Romer, T.R., 1991. Use of the Mycosep multifunctional column for liqutd chromatographic determination of allatoxins in agricultural products. J. Assoc. Off. Anal. Chem. 74. 95 I-956. Young, J.C., Fulcher, R.G., Hayhoe, J.H., Scott. P.M., Dexter, J.E., 1984. Effect of mtlling and bakin g on deoxynivalenol (vomitoxin) content of Eastern Canadian wheat. J. Agric. Food Chem. 32, 6599664.