ARTICLE IN PRESS

Food Control xxx (2009) xxxxxx

Contents lists available at ScienceDirect

Food Control
journal homepage: www.elsevier.com/locate/foodcont

Occurrence of AFM1 in urine samples of a Brazilian population and association

with food consumption
Alessandra de Cssia Romero a, Tnia Raquel Baroni Ferreira a, Carlos Tadeu dos Santos Dias b,
Maria Antonia Calori-Domingues a, Eduardo Micotti da Gloria a,*
a
b

Department of Agroindustry, Food and Nutrition, ESALQ, Universidade de So Paulo USP, Av. Pdua Dias, 11, CEP 13418-900, Piracicaba, So Paulo, Brazil
Department of Exact Sciences, ESALQ, Universidade de So Paulo USP, Av. Pdua Dias, 11, CEP 13418-900, Piracicaba, So Paulo, Brazil

a r t i c l e

i n f o

Article history:
Received 6 April 2009
Received in revised form 10 August 2009
Accepted 17 August 2009
Available online xxxx
Keywords:
Aatoxin M1
Biomarker
Urine
Food consumption

a b s t r a c t
This survey evaluated the presence of AFM1 in human urine samples from a specic Brazilian population,
as well as corn, peanut, and milk consumption measured by two types of food inquiry. Urine samples
from donors who live in the city of Piracicaba, State of So Paulo, Brazil were analyzed to detect the
presence of aatoxin M1 (AFM1), an aatoxin B1 metabolite, which may be used as aatoxin B1 exposure
biomarker. The AFM1 analysis was performed using immunoafnity clean-up and detection by high-performance-liquid chromatography with uorescence detector. A total of 69 samples were analyzed and 45
of them (65%) presented contaminations P1.8 pg ml 1, which was the limit of quantication (LOQ). Seventy eight percent (n = 54) of the samples presented detectable concentrations of AFM1 (>0.6 pg ml 1).
The AFM1 concentration among samples above LOQ ranged from 1.8 to 39.9 pg ml 1. There were differences in food consumption prole among donors, although no association was found between food consumption and AFM1 concentration in urine. The high frequency of positive samples suggests exposure of
the populations studied to aatoxins.
2009 Elsevier Ltd. All rights reserved.

1. Introduction
Human exposure to aatoxins is a concern worldwide because
aatoxins are potent cancer-promoting agents, especially liver cancer (International Agency for Research Cancer (IARC) 1993). One
way of evaluating aatoxin exposure through diet is to estimate
the aatoxin concentration in foods and the level at which these
foods are consumed. However, several other factors may impair
this estimation such as heterogeneity in relation to food contamination, difculties in measuring food consumption, seasonal variations associated with food contamination and effect of food
preparation on contamination level, among others (Hall & Wild,
1994; Polychronaki, 2007).
An alternative for a more precise evaluation of total aatoxin
exposure is to measure the presence of aatoxin metabolites in
body uids such as milk, urine and blood (Hall & Wild, 1994).
The detection of several aatoxin metabolites in body uids has
already been reported in the literature (Essigmann et al., 1977;
Gorelick, 1990; Mykkanen et al., 2005; Sabbioni, Skipper, Buchi,
& Tannenbaum, 1987). However, these metabolites can only be
used as evidence of aatoxin exposure after their validation as

* Corresponding author. Tel./fax: +55 19 34225678.

E-mail address: emgloria@esalq.usp.br (E.M. da Gloria).

biomarkers. For aatoxins, the presence of adducts, aatoxin N7

guanine in urine, aatoxin lysine in blood and aatoxin M1 in urine
is considered a valid alternative for aatoxin exposure verication
(Groopman & Kensler, 1999).
Aatoxin M1 (AFM1) is a hydroxylated metabolite of aatoxin B1
(AFB1) originated from animal and human metabolism after the
ingestion of aatoxin B1, which may be present in body uids such
as urine, milk, and blood (Groopman, 1994). Only from 1.2% to 2.2%
of the aatoxin present in the diet is excreted as AFM1 in human
urine (Zhu et al., 1987).
The use of AFM1 present in urine as biomarker of human exposure to aatoxins has already been reported in several world populations. In Taiwan, the analysis of urinary excretion performed by
Hatch et al. (1993) with 250 people resulted in concentrations
ranging from 0.003 to 0.108 ng ml 1. Data obtained by Cheng, Root,
Pan, Chen, and Campbell (1997) with people from China and Taiwan revealed 6466% of positivity, with concentrations ranging
from 3.2 to 108 ng for 12 h and from 2.7 to 17 ng for 12 h among
people who live in China and Taiwan, respectively. In Sierra Leone,
Africa, concentrations ranged from 0.1 to 374 ng ml 1, according to
a study by Jonsyn-Ellis (2000). In Ghana, concentrations reported
by Jolly et al. (2006) reached maximum levels close to 12 ng mg 1
of creatinine. In the Czech Republic, Malir et al. (2006) reported the
occurrence of AFM1 in 57.6% of urine samples analyzed, with concentrations ranging from 0.019 to 19.219 ng g 1 of creatinine.

0956-7135/$ - see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodcont.2009.08.004

Please cite this article in press as: Romero, A. d. C., et al. Occurrence of AFM1 in urine samples of a Brazilian population and association with food consumption. Food Control (2009), doi:10.1016/j.foodcont.2009.08.004

ARTICLE IN PRESS
2

A.de.C. Romero et al. / Food Control xxx (2009) xxxxxx

In Brazil, there are several reports on the occurrence of aatoxin

in foods aimed at human and animal consumption. However, corn
and peanut, as well as corn and peanut-based products are the
main foods related to aatoxin contamination (Caldas, Silva, &
Oliveira, 2002; Gloria, Fonseca, & Souza, 1997; Gloria, Romero,
Carvalho, Calori-Domingues, & Gonalves, 2006; Kawashima & Valente-Soares, 2006; Rodriguez-Amaya & Sabino, 2002; Sabino et al.,
2000). Milk and dairy products are also frequently reported as contaminated in Brazil with low aatoxin levels (Gonalez et al., 2005;
Oliveira & Ferraz, 2007; Oliveira, Germano, Bird, & Pinto, 1997);
Oliveira, Rosmaninho, & Rosim, 2006; Shundo & Sabino, 2006.
However, there are few investigations on the presence of aatoxin exposure biomarkers in the Brazilian population. Scussel,
Haas, Gong, Turner, and Wild (2004) analyzed blood samples from
50 blood donors in a hospital in the city of So Paulo, Brazil, and
observed the presence of AFB1-lysine in 62% of the samples analyzed, with concentrations ranging from <3 to 57.3 pg mg 1. Navas
(2003) reported the analysis for AFM1 of samples of human breast
milk collected in a hospital in the city of So Paulo as well. Only
one sample presented a contamination level of 0.024 ng ml 1.
The detection/quantication limit for the methodology proposed
by that author was 0.01 ng ml 1.
This study aimed to conduct a survey on the occurrence of aatoxin M1 in the urine of individuals that live in Piracicaba, So Paulo, Brazil. Donor feeding habits were evaluated by means of two
types of food inquiries and consumption data were tested for correlation with contamination levels observed in urine samples.
This study has been approved by Comit de tica em Pesquisa
(Research Ethics Committee), Faculdade de Odontologia de Piracicaba - Universidade Estadual de Campinas (Protocol No. 095/2005).

paper (Whatman Schleider & Schuell, Maidstone, England, product

number 934-AH). Later, 20 ml of ltered extract was transferred to
a 50 ml capacity vial and 20 ml of sodium acetate buffer (pH 5.0)
was added. The pH of the mixture was measured and corrected
to 5.0 using an appropriate volume of a 0.1 M glacial acetic acid
solution whenever necessary. The mixture was directly passed
through an immunoafnity column (Neocolumn, Neogem Europe,
UK) at a ow rate of approximately 1.01.5 ml min 1. After adding
the mixture the column was washed with 40 ml of ultrapure water
(Milli Q, Millipore, Bedford, MA, USA). The column was dried by
applying positive pressure with a syringe and bound AFM1 was
eluted with 2.0 ml of HPLC-methanol which was recovered in a
4 ml vial previously treated with acid. The eluate was evaporated
under nitrogen gas and reconstituted with 500 ll of the mobile
phase before liquid chromatograph analysis.
Detection and quantication of sample extracts were performed
by high-performance-liquid chromatography (HPLC) with a liquid
chromatography system equipped with a LC-10AT Shimadzu pump
(Kyoto, Japan), a Shimadzu RF-10AXL uorescence detector (excitation 365 nm and emission 460 nm), an injection volume of
100 lL, and a reverse phase column (250  4.6 mm, particle size
of 3 lm) and precolumn (Synergi Fusion, Phenomenex Inc., Torrance, CA, USA) kept at room temperature. The mobile phase consisted of an isocratic mixture of water and acetonitrile at a volume
ratio of 75:25 and a ow rate of 1.0 ml min 1. A calibration curve
was prepared using standard AFM1 solutions in mobile phase at
concentrations of 0.005, 0.01, 0.02, and 0.03 ng ml 1. The standard
obtained (Sigma, St. Louis, MO, USA, product code 6428, 10 lg) as
puried crystalline AFM1 was dissolved in HPLC-grade acetonitrile
and its concentration was determined by spectrophotometry
according to Trucksess (2005, chap. 49).

2. Material and methods

2.5. Aatoxin M1 identity conrmation

2.1. Donor selection

The AFM1 identity was conrmed by the formation of AFM1

hemiacetal derivative (AFM2a). The sample extract was evaporated
under nitrogen gas and re-dissolved with 200 ll of HPLC-n-hexane
and 200 ll of triuoroacetic acid (TFA). The mixture was vortexed
during 1 min and left to rest for 10 min. The mixture was evaporated under nitrogen gas and re-suspended in 500 ll of mobile
phase before injection into the liquid chromatography system.
The disappearance or reduction of the original AFM1 peak and
the appearance of a new peak at an earlier retention time conrmed the presence of AFM1 in the sample.

Sixty-nine individuals were randomly selected among residents

from different regions of the city of Piracicaba, So Paulo, Brazil as
urine donors.
2.2. Food Inquiries
The individuals selected as donors answered two types of food
inquiries, namely a feeding frequency inquiry (FFI) on the monthly
consumption by individuals, and a 24-h recollection (IR24), in
which the consumption of all foods ingested on the day before urine collection were recorded. Among the FFI component items and
items reported by individuals in the IR24, foods were selected and
grouped as corn-, peanut-, and milk-based foods, whose contamination by aatoxins has been more frequently reported in the scientic literature.
2.3. Sampling
Each donor collected an early-morning urine sample using a
300 ml polyethylene ask. The sample volume ranged from 50 to
300 ml among donors. After collection, samples were identied
and immediately sent to the laboratory, and were stored at
18 C until analysis.

2.6. Method evaluation

The limit of detection (LOD) was considered as the lowest AFM1
concentration that could be reliably detected (>3:1 signal to noise
ratio) in blank urine sample chromatograms. The quantication
limit (LOQ) was considered as three times the LOD.
The method was studied for its recovery rate using human urine
sample as blank and for this a non detect AFM1 urine sample was
used. Seven replicates were made. AFM1 standard solution was
added to samples immediately before analyze to give the concentration of 4.0 pg ml 1 of AFM1. The recovery rate was considered as
the concentration of AFM1 quantied over the concentration
added. The repeatability measured by relative standard deviation
was checked over replicates of urine samples used in the recovery
study.

2.4. Aatoxin M1 analysis in urine

2.7. Statistical analysis
The extraction and purication of urine samples for AFM1 determination were performed as recommended by Kussak, Andersson,
and Andersson (1995), with some modications. A 30 ml volume
from the urine sample was ltered through a glass microbre lter

The results obtained were analyzed via the LSD Means Comparison Test (p < 0.05) between the individuals total and the 90 percentile of AFM1 concentrations in urine. The correlation between

Please cite this article in press as: Romero, A. d. C., et al. Occurrence of AFM1 in urine samples of a Brazilian population and association with food consumption. Food Control (2009), doi:10.1016/j.foodcont.2009.08.004

ARTICLE IN PRESS
3

A.de.C. Romero et al. / Food Control xxx (2009) xxxxxx

consumption of food groups and AFM1 concentrations in urine was

tested via Spearmans correlation Test. For statistical analysis purposes, AFM1 contamination results lower than the limit of detection of the methodology employed were considered numerically
equal to zero, while values between the limits of quantication
(1.8 pg ml 1) and the detection (0.6 pg ml 1) were considered as
the mean between those extremes, that is, 1.2 pg ml 1. Statistical
analyses were performed using Statistical Analysis System software (SAS, 2004).

Table 2
Comparison between consumption means reported in the FFI and IR24 inquiries for
food groups considered as risk groups for contamination by aatoxins.

3. Results and discussion

3.1. Method evaluation

The AFM1 retention time considering the liquid chromatographic conditions used was 14.0 min. The linearity of the calibration curve was 0.9907. The limit of detection (LOD) considering a
nal sample extract of 500 ll and an injection volume of 100 ll
was 0.6 pg ml 1 and the limit of quantication (LOQ) was
1.8 pg ml 1. The recovery rates ranged from 73% to 97% with a
mean value of 83% and relative standard deviations for seven
spiked samples (4 pg ml 1) of 20.7%. The AFM1 identity conrmation method by TFA described showed that the original AFM1 peak
in the positive sample disappeared and a new peak appeared at a
shorter retention time of 5.1 min.
3.2. Occurrence of aatoxin M1
The AFM1 levels detected in urine samples are presented in Table 1. From 69 samples analyzed, 54 (78%) presented detectable
AFM1 concentrations. Forty-ve samples (65%) had concentrations
ranging from 1.8 to 39.9 pg ml 1.
3.3. Food intake
The FFI was used to estimate food intake habits among the individuals sampled and conrmed that the foods generally reported
as contaminated with aatoxins in Brazil were also part of the regular diet of donors.
Milk and milk-based products were the foods consumed at the
highest frequencies and quantities, and were consumed by all individuals (69), followed by corn and corn-based products (67) and
peanut and peanut-based products (39).
IR24 was used to obtain the ingestion of foods on the day that
preceded collection of the urine sample. Consumption of milk
and milk-based products were the most frequent among the 69
individuals, and was reported by 61 individuals. Corn and peanut
consumption had lower frequencies, and were reported by 12
and 14 individuals, respectively (Table 2).
The consumption of corn and corn-based product food group
signicantly differed from peanut and peanut-based products, as
well as from milk and milk-based product in the FFI I and II IR24

Table 1
Distribution of AFM1 contamination among urine samples from donors, State of So
Paulo, Brazil.
AFM1 (pg ml
<0.6
0.61.7
1.89.7
9.839.9

15
9
37
8

21.7
13.1
53.6
11.6

Food groups

Mean intake in inquiries (g/day)*

FFI

Corn and corn-based products

Peanut and peanut-based
products
Milk and milk-based products

II

II

23.2a,**,A,**
5.1b,A

23.9a,A
9.1b,A

13.3a,A
7.1a,A

76.7a,B
35.0b,B

286.1b,B

286.1b,B

307.7b,A

348.0b,A

In both inquiries, I corresponds to the mean of all individuals and II corresponds

to the mean of individuals that reported some level of intake only.
**
The same lower case letters in the same column indicate that the means are not
different from each other. In each inquiry, equal upper case letters on the same row
indicate that the means are not different from each other (LSD Test, p < 0.05).

II inquiries. In IR24 I, corn and corn-based products and peanut

and peanut-based products did not differ from each other, but
were both different on average from milk-based products.
The comparison between means for all individuals and only
those that reported consumption (Table 2) indicates that for FFI
the mean of all individuals could be used as an instrument representative of the group. For IR24, the corn and peanut groups had
signicantly higher means when the analysis was made only between individuals that consumed those foods, demonstrating that
for the evaluation of exposure to aatoxins via those foods, it
would also be important to consider consumption means only between individuals that reported some level of consumption.

3.4. Association between consumption of risk foods and urine AFM1

The correlation analysis for food intake and detected AFM1 levels, obtained by Spearmans correlation Test, showed statistically
non-signicant r values (Table 3), indicating that no correlation
trend exists between those factors. Consequently, in spite of the
differentiated consumption of donors in relation to food groups,
no association was found between the specic consumption of
any of those food groups and AFM1 concentrations in the urine of
individuals, as well as between the total consumption of those food
groups and AFM1.
A correlation analysis between consumption AFM1 levels was
also performed; comparisons were made for the group of donors
with the highest AFM1 levels observed and all other donors. To
accomplish that, the 90 percentile for AFM1 concentration in urine
showed that 10% of the highest AFM1 concentrations in urine ranged from 9.8 to 39.9 pg ml 1 while the other 90% showed values
lower than 9.7 pg ml 1. The means comparison between both
groups of individuals considering the 90 percentile of contamination did not detect signicant differences in food intake between
individuals, although the mean AFM1 contamination values were
signicantly different between groups.

Table 3
Spearmans correlation Test between AFM1 and consumption reported in the FFI and
IR24 inquiries for food groups.
Food groups

Mean (standard deviation): 5.96 (6.39)

Median*: 3.96
Coefcient of variation*: 107%
Values for the interval between 1.8 and 40.0 pg ml

Spearmans correlation Test

FFI

IR24

Corn and corn-based products

Peanut and peanut-based products
Milk and milk-based products

0.0174
0.1774
0.0876

IR24
p

0.8869
0.1448
0.4741

0.1809
0.1076
0.1352

0.1368
0.3788
0.2681

Please cite this article in press as: Romero, A. d. C., et al. Occurrence of AFM1 in urine samples of a Brazilian population and association with food consumption. Food Control (2009), doi:10.1016/j.foodcont.2009.08.004

ARTICLE IN PRESS
4

A.de.C. Romero et al. / Food Control xxx (2009) xxxxxx

Table 4
Comparison between mean AFM1 values and consumption obtained by FFI, considering the 90 percentile of contamination values in urine.
90 percentile

Consumption of groups of foods and food-based products FFI (g/day)

AFM1 concentration (pg ml

AFM1 range

(n)

Corn

Peanut

Milk

Mean

09.7
9.739.9

90
10

(61)
(8)

22.9a,*
24.9a

4.9a
6.4a

282.9a
310.3a

2.6a
15.4b

Identical lower case letters on the same column indicate that means do not differ between each other (LSD Test, p < 0.05).

The lack of association between AFM1 levels present in urine

and consumption of certain types of food investigated in this study
shows that those foods do not make an important or expressive
contribution toward exposure to aatoxins through the diet. Perhaps in association with other foods they may contribute partially
and fortuitously toward total exposure. However, the possibility
that the source of exposure to aatoxins here detected comes from
a particular type of food or food group not addressed in the food
inquiries formulated in this study cannot be ruled out (see Table 4).
Studies conducted in regions of Asia and Africa were successful
in obtaining correlations between food ingestion and excretion of
biomarkers for aatoxins or with aatoxicosis signals (AzzizBaumgartner et al., 2005; Gan et al., 1988; Lewis et al., 2005; Lye
et al., 1995; Zhu et al., 1987). However, it is observed that those regions are greatly susceptible to the occurrence of aatoxins in
foods, providing greater food exposure. In regions of China, it is
estimated that the daily ingestion of AFB1 from corn consumption
is 184.1 lg; however, AFB1 ingestion from total diet is even higher,
estimated between 6.5 and 2027 ng kg 1 d 1. In regions of Africa,
the ingestion of aatoxins by peanut consumption ranges from
9.9 to 99.2 ng kg 1 d 1, but exposure via total consumption may
reach from 3.5 to 183.7 ng kg 1 d 1 (Williams et al., 2004). There
are few studies in Brazil dealing with human exposure estimation
resulting from the presence of aatoxins in foods. Oliveira, Germano, Bird, and Pinto (1997) estimated exposure to AFM1 in 4month-old children and obtained values of 3.7 ng kg 1 d 1 for
those individuals after analyzing milk samples for infant consumption in day-care centers in So Paulo, Brazil. Similarly, in a study on
the occurrence of aatoxins in foods, Amaral, Nascimento, Sekiyama, Janeiro, and Machinski (2006) estimated exposure via aatoxin concentrations obtained in corn our and consumption
data available in the scientic literature, and obtained an exposure
value of 0.14 ng kg 1 d 1 provided by the consumption of that
corn-based food.
The AFM1 values in urine observed in this study were lower
than those previously reported for other countries however, the
number of positive samples observed was within the same range
(Cheng et al., 1997; Hatch et al., 1993; Jolly et al. 2006; Jonsyn-Ellis, 2000; Malir et al., 2006). It can be seen that although AFM1 in
urine is a validated biomarker to evaluate exposure to aatoxins,
in this study the collection of a punctual urine sample that was
not representative of a 12 or a 24 h period, prevented the
estimation of ingestion of aatoxins and, consequently, a correlation between total AFM1 excretion per time period with food consumption in the previous day. The possibility of AFM1 glucoronide
and sulphate (Eaton, Ramsdell, & Neal, 1994) conjugated excretion
must be considered with an additional factor which can affect the
correlation (Mykkanen et al., 2005) despite free AFM1 be a validated biomarker. Nevertheless, the high rate of positive samples
observed constitutes important information because it shows that
those populations are exposed to aatoxin.
Since the presence of AFM1 in urine is considered a short-term
AFB1 exposure biomarker (Groopman, 1994) and aatoxin contamination level is heterogeneous and could be seasonal (Council for
Agriculture Science & Technology (CAST), 2003) perhaps a 24 h urine sample and a different sampling frequency could more ade-

quately show AFM1 contamination in the urine of those

populations.
4. Conclusions
This study is the rst to report AFM1 in urine samples from Brazilian populations. In the Brazilian populations mentioned here,
concentration levels were not as high as those reported for other
countries but their frequencies, resulting from the low AFM1 detection limit achieved in this study, were as high as or even higher
than those, showing that most donors were exposed to aatoxin.
The lack of association between consumption of the foods investigated in this study, considered as normally contaminated with
aatoxins in Brazil, and AFM1 levels in urine suggest that exposure
of this population via their diet did not have those foods individually as the main sources of exposure. Further AFB1 biomarker studies are required to better evaluate the true levels of aatoxin
exposure over a long period, with urine collections that would
quantitatively represent AFM1 excretion among individuals.
Since the city of Piracicaba is considered a region with high social and economic indicators (Instituto Brasileiro de Geograa e
Estatstica IBGE (Internet)., 2007), the observed high frequencies
of aatoxin exposure suggests that maybe other less developed regions of the country present disturbing levels of aatoxin exposure,
corroborating the need for new studies to evaluate individual
exposure to aatoxins in Brazil.
Acknowledgment
The authors would like to thank Fundao de Amparo Pesquisa do Estado de So Paulo (FAPESP), Brazil, Grant No. 2005/533086, for nancial support.
References
Amaral, K. A. S., Nascimento, G. B., Sekiyama, B. L., Janeiro, V., & Machinski, M. Jr.,
(2006). Aatoxinas em produtos de milho comercializados no Brasil e riscos
para a sade humana. Cincia e Tecnologia de Alimentos, 26(2), 336342.
Azziz-Baumgartner, E., Lindblade, K., Gieseker, K., Rogers, H. S., Kieszak, S., Njapau,
H., et al. (2005). Aatoxin investigation group. Case-control study of an acute
aatoxicosis outbreak, Kenya, 2004. Environmental Health Perspectives, 113(12),
17791784.
Caldas, E. D., Silva, S. C., & Oliveira, N. O. (2002). Aatoxinas e ocratoxinas a em
alimentos e riscos para a sade humana. Revista de Sade Pblica, 99, 195197.
Cheng, Z., Root, M., Pan, W., Chen, J., & Campbell, C. T. (1997). Use of improved
method for analysis of urinary aatoxin M1 in a survey of Mailand China e
Taiwan. Cancer Epidemiology, Biomarkers and Prevention, 6, 523529.
Council for Agriculture Science, Technology (CAST) (2003). Mycotoxins: Risks in
plant, animal, and human systems. IA: Ames.
Eaton, D. L., Ramsdell, H. S., & Neal, D. L. (1994). Biotransformation of aatoxins. In
D. L. Eaton & J. D. Groopman (Eds.), The toxicology of aatoxins: Human health,
veterinary and agricultural signicance (pp. 4572). London: Academic Press.
Essigmann, J. M., Croy, R. G., Nadzan, A. M., Busby, W. F., Jr., Reinhold, V. N., Buchi, G.,
et al. (1977). Structural identication of the major DNA adduct formed by
aatoxin B1 in vitro. Proceedings National Academy Science USA, 74, 18701874.
Gan, L. S., Skipper, P. L., Peng, X. C., Groopman, J. D., Chen, J. S., Wogan, G. N., et al.
(1988). Serum albumin adducts in the molecular epidemiology of aatoxin
carcinogenesis: Correlation with aatoxin B1 intake and urinary excretion of
aatoxin M1. Carcinogenesis, 9(7), 13231325.
Gloria, E. M., Fonseca, H., & Souza, I. M. (1997). Occurrence of mycotoxins in maize
delivered to the food industry in Brazil. Tropical Science, 37, 107110.

Please cite this article in press as: Romero, A. d. C., et al. Occurrence of AFM1 in urine samples of a Brazilian population and association with food consumption. Food Control (2009), doi:10.1016/j.foodcont.2009.08.004

ARTICLE IN PRESS
A.de.C. Romero et al. / Food Control xxx (2009) xxxxxx
Gloria, E. M., Romero, A. C., Carvalho, A. P. P., Calori-Domingues, M. A., & Gonalves,
P. V. M. (2006). Perl da contaminao com aatoxinas entre embalagens de
produtos de amendoim. Cincia e Tecnologia de Alimentos, 26(2), 660665.
Gonalez, E., Felicio, J. D., Pinto, M. M., Rossi, M. H., Nogueira, J. H. C., & Manginelli, S.
(2005). Ocorrncia de aatoxina M1 em leite comercializado em alguns
municpios do estado de So Paulo. Arquivos do Instituto Biolgico, 72(4),
435438.
Gorelick, N. J. (1990). Risk assessment for aatoxin: I. Metabolism of aatoxin B1 by
different species. Risk Analysis, 10, 539559.
Groopman, J. D. (1994). Molecular dosimetry methods for assessing human
aatoxin exposures. In D. L. Eaton & J. D. Groopman (Eds.), The toxicology of
aatoxins: Human health, veterinary, and agricultural signicance (pp. 259280).
San Diego, CA: Academic Press.
Groopman, J. D., & Kensler, T. W. (1999). The light at the end of the tunnel for
chemical-specic biomarkers: Daylight or headlight? Carcinogenesis, 20, 111.
Hall, A. J., & Wild, C. P. (1994). Epidemiology of aatoxin related disease. In D. L.
Eaton & J. D. Groopman (Eds.), The toxicology of aatoxins: Human health,
veterinary, and agricultural signicance (pp. 223258). San Diego, CA: Academic
Press.
Hatch, M. C., Chen, C.-J., Levin, B., Ji, B.-T., Yang Hsu, S.-W., Wang, L.-W., et al. (1993).
Urinary aatoxin levels, hepatitis-B virus infection an hepatocellular carcinoma
in Taiwan. International Journal of Cancer, 54, 931934.
Instituto Brasileiro de Geograa e Estatstica IBGE (Internet). (2007) <http://
www.ibge.gov.br/home/statistica/calendario2007.shtm> Accessed 21.07.07.
International Angency for Research Cancer (IARC) (1993). Some naturally occurring
substances Food items and constituents, heterocyclic aromatic amines and
mycotoxins. Monographs on the evaluation of carcinogenic risks to humans No.
56 (pp. 489521). Lyon: IARC.
Jolly, P., Jiang, Y., Ellis, W., Awuah, R., Nnedu, O., Phillips, T., et al. (2006).
Determinants of aatoxin levels in Ghanaians: Sociodemographic factors,
knowledge of aatoxin and food handling and consumption practices.
International Journal of Hygiene and Environmental Health, 209, 345358.
Jonsyn-Ellis, F. (2000). Seasonal variation in exposure frequency and concentration
levels of aatoxins and ochratoxins in urine samples of boys and girls.
Mycopathologia, 152, 3540.
Kawashima, L. M., & Valente-Soares, L. M. (2006). Incidncia de fumonisinas B1,
aatoxinas B1, B2, G1 e G2 e ocratoxina A e zearalenona em produtos de milho.
Cincia e Tecnologia de Alimentos, 26(3), 516521.
Kussak, A., Andersson, B., & Andersson, K. (1995). Immunoafnity column clean-up
for the high-performance liquid chromatographic determination of aatoxins
B1, B2, G1, G2, M1 and Q1 in urine. Journal of Chromatography B, 672, 253259.
Lewis, L., Onsongo, M., Njapau, H., Schurz-Rogers, H., Luber, G., Kieszak, S., et al.
(2005). Aatoxin contamination of commercial maize products during on
outbreak of acute aatoxicosis in eastern and central Kenya. Environmental
Health Perspectives, 113(12), 17631767.
Lye, M. S., Ghazali, A. A., Mohan, J., Alwin, N., & Nair, R. C. (1995). An outbreak of
acute hepatic encephalopathy due to severe aatoxicosis in Malaysia. American
Journal Tropical of Medicine and Hygiene, 53(1), 6872.
Malir, F., Ostry, V., Grosse, Y., Roubal, T., Skarkova, J., & Ruprich, J. (2006). Monitoring
the mycotoxins in food and their biomarkers in the Czech Republic. Molecular
Nutrition Food Research, 50, 513518.

Mykkanen, H., Zhu, H., Salminen, E., Juvonen, R. O., Ling, W., Ma, J., et al. (2005).
Fecal and urinary excretion of aatoxin B1 metabolites (AFQ1, AFM1 and AFBN7-guanine) in young Chinese males. International Journal of Cancer, 115,
879884.
Navas, S. A. (2003). Possvel exposio de crianas as aatoxinas M1 e ocratoxina A,
atravs do leite materno, na cidade de So Paulo. Dissertation, Faculdade de
Engenharia de Alimentos da UNICAMP.
Oliveira, C. A. F., & Ferraz, J. C. O. (2007). Occurence of aatoxin M1 in pasteurized,
UHT milk and milk powder from goat origin. Food Control, 18, 375378.
Oliveira, C. A. F., Germano, P. M. L., Bird, C., & Pinto, C. A. (1997). Immunochemical
assessment of aatoxin M1 in milk powder consumed by infants in So Paulo,
Brazil. Food Additives and Contaminants, 14(1), 710.
Oliveira, C. A., Rosmaninho, J., & Rosim, R. (2006). Aatoxin M1 and cyclopiazonic
acid in uid milk traded in So Paulo, Brazil. Food Additives and Contaminants,
23(2), 196201.
Polychronaki, N. (2007). Biomarkers of aatoxin exposure and a dietary
intervention: Studies in infants and children from Egypt and Guinea and
young adults from China. Dissertation (doctoral in public health and clinic
nutrition) Kuopio, University of Kuopio, 114p.
Rodriguez-Amaya, D. B., & Sabino, M. (2002). Mycotoxin research in Brazil: The last
decade in review. Brazilian Journal of Microbiology, 33(1), 111.
Sabbioni, G., Skipper, P. L., Buchi, G., & Tannenbaum, S. R. (1987). Isolation and
characterization of the major serum albumin adduct formed by aatoxin B1
in vivo in rats. Carcinogenesis, 8, 819824.
Sabino, M., Lamardo, L. C. A., Milanez, T. V., Inomata, E. I., Zorzetto, A. P., Navas, A.S.,
et al. (2000). Twenty years of aatoxin contamination in groundnuts and
groundnut products in So Paulo State, Brazil. In Ofcial program and
abstractbook of the 10th international IUPAC symposium on mycotoxin and
phycotoxin, 2000 (pp. 151). So Paulo, Brasil.
SAS, (2004). Online Doc 9, 1.3 SAS. Statistical analysis system. SAS Institute Inc., Cary.
Scussel, V. M., Haas, P., Gong, Y. Y., Turner, C. P., & Wild, C. P. (2004). Study of
aatoxin exposure in a Brazilian population using an aatoxinalbumin
biomarker. In H. Njapau, S. Trujillo, H. P. Van Egmond, & D. Park (Eds.).
Mycotoxins and phycotoxins: Advances in determination, toxicology and exposure
management. Proceedings of international IUPAC symposium on mycotoxins and
phycotoxins, 2004, Bethesda (Vol. 11, pp. 197202). Netherlands: Wageningen
Academic.
Shundo, L., & Sabino, M. (2006). Aatoxin M1 in milk by immunoafnity column
cleanup with TLC/HPLC determination. Brazilian Journal of Microbiology, 37,
164167.
Trucksess, M. W. (2005). Natural Toxins. In W. Horwitz (Ed.), Ofcial methods of
analysis of the association of ofcial analytical chemists international (18th ed.,
pp. 199). Gaithersburg, MD: Association of Ofcial Analytical Chemists
International.
Williams, J. H., Phillips, T. D., Jolly, P. E., Stiles, J. K., Jolly, C. M., & Aggarwal, D. (2004).
Human aatoxicosis in developing countries: A review of toxicology, exposure,
potential health consequences, and interventions. American Journal Clinical
Nutrition, 80, 11061122.
Zhu, J. Q., Zhang, L. S., Hu, X., Xiao, Y., Chen, J. S., Xu, Y. C., et al. (1987). Correlation of
dietary aatoxin B1 levels with excretion of aatoxin M1 in human urine. Cancer
Research, 47, 18481852.

Please cite this article in press as: Romero, A. d. C., et al. Occurrence of AFM1 in urine samples of a Brazilian population and association with food consumption. Food Control (2009), doi:10.1016/j.foodcont.2009.08.004