Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=tfab20

Download by: [UNAM Ciudad Universitaria] Date: 26 May 2016, At: 10:56

Food Additives & Contaminants: Part B

ISSN: 1939-3210 (Print) 1939-3229 (Online) Journal homepage: http://www.tandfonline.com/loi/tfab20

Survey of aflatoxins in maize tortillas from MexicoCity

Pável Castillo-Urueta , Magda Carvajal , Ignacio Méndez , Florencia Meza &Amanda Gálvez

To cite this article: Pável Castillo-Urueta , Magda Carvajal , Ignacio Méndez , Florencia Meza &Amanda Gálvez (2011) Survey of aflatoxins in maize tortillas from Mexico City, Food Additives &Contaminants: Part B, 4:1, 42-51, DOI: 10.1080/19393210.2010.533390

To link to this article: http://dx.doi.org/10.1080/19393210.2010.533390

Published online: 27 Jan 2011.

Submit your article to this journal

Article views: 86

View related articles

Food Additives and Contaminants: Part BVol. 4, No. 1, March 2011, 42–51

VIEW DATASET

Survey of aflatoxins in maize tortillas from Mexico City

Pavel Castillo-Uruetaab, Magda Carvajalb*, Ignacio Mendezc, Florencia Mezad and Amanda Galveze

aDepartamento de Ciencias Quımicas, Universidad Nacional Autonoma de Mexico, Ciudad Universitaria, 04510 Mexico,D.F., Mexico; bDepartamento de Botanica, Instituto de Biologıa, Universidad Nacional Autonoma de Mexico,Ciudad Universitaria, 04510 Mexico, D.F., Mexico; cDepartamento de Estadıstica, Instituto de Investigaciones en MatematicasAplicadas y en Sistemas, Universidad Nacional Autonoma de Mexico, Ciudad Universitaria, 04510 Mexico, D.F., Mexico;dDepartamento de Ciencias Bioquımicas, Universidad Nacional Autonoma de Mexico, Ciudad Universitaria, 04510 Mexico,D.F. Mexico; eFacultad de Quımica, Universidad Nacional Autonoma de Mexico, Ciudad Universitaria,04510 Mexico, D.F., Mexico

(Received 8 February 2010; final version received 15 August 2010)

In Mexico, maize tortillas are consumed on a daily basis, leading to possible aflatoxin exposure. In a survey of396 2-kg samples, taken over four sampling days in 2006 and 2007 from tortilla shops and supermarkets inMexico City, aflatoxin levels were quantified by HPLC. In Mexico, the regulatory limit is 12mg kg�1 totalaflatoxins for maize tortillas. In this survey, 17% of tortillas contained aflatoxins at levels of 3–385 mg kg�1 orvalues below the limit of quantification (5LOQ) and, of these, 13% were412mg kg�1 and 87% were below theregulatory limit. Average aflatoxin concentrations in 56 contaminated samples were: AFB1 (12.1 mg kg

�1); AFB2

(2.7 mg kg�1); AFG1 (64.1 mg kg�1) and AFG2 (3.7 mg kg

�1), and total AF (20.3 mg kg�1).

Keywords: cereals; cereals and grain; cooked foods; plants; processed foods; aflatoxins; environmentalcontaminants; mutagenic compounds; mutagens; mycotoxins; trace elements (toxic)

Introduction

In Mexico, maize tortillas are a staple food, with

average consumption of 325 g per day (Fernandez-

Munoz et al. 2006). Tortilla consumption has recently

extended to the US, Canada, and some European

countries (Islas-Hernandez et al. 2006). Maize tortilla

production involves a traditional lime-cooking treat-

ment called ‘nixtamalizacion’, which is a thermal

process involving calcium hydroxide, considered as a

possible control in eliminating aflatoxins. This lime

treatment causes anatomic, physical and chemical

changes in maize grains (Gutierrez et al. 2007; Rojas-

Molina et al. 2007), and there are contradictory reports

about its aflatoxin-controlling benefits. Therefore,

several studies have been done to evaluate the degree

of retention of aflatoxins in grain during the process of

tortilla making (Giddey et al. 1977; Price and

Jorgensen 1985; Torreblanca et al. 1986).The major aflatoxins of concern in maize grains are

designated B1, B2, G1 and G2 (Liu et al. 2006).Some reports (De Arriola et al. 1988; Torres et al.

2001) have indicated that, during lime cooking with

CaO (1–3% w/w), levels of aflatoxins in dough

diminished in 97% of the samples and in 98% of

tortillas. This data refers to the loss of aflatoxin

fluorescence but, when the pH changes to acid,

aflatoxins reactivate and recover their lost fluorescence

(Price and Jorgensen 1985). The process of boiling

maize in water with lime reduced aflatoxin levels by

20 to 46% (Price and Jorgensen 1985).The role of pH and temperature in the reduction of

aflatoxins during manufacture, and the effect of frying

tortillas was also analyzed (Torres et al. 2001). The

heat and lime treatment applied to maize grains

produces hydrolysis of the lactone ring of aflatoxins,

forming soluble salts that are lost when washed

(Coomes et al. 1966; Torres et al. 2001). The closure

of the lactone ring reactivates previously degraded

aflatoxins, and this is possible at pH 9.5 (Giddey et al.

1977). Nevertheless, acidification of dough and taking

it later to a pH of 6–8 closes the lactone ring, re-

converting lost aflatoxins into active compounds. The

increase in aflatoxins by acidification was 0–18%. Due

to the contradictory reports on the effect of lime

cooking on aflatoxins, the purpose of the present study

was to identify and quantify aflatoxins in tortillas.

The process of tortilla sampling in Mexico City is

quite complex. Although, Ribeiro (2005) identified

45,000 tortilla shops in the city, according to the

Ministry of Economy, the National Chamber of

Industrialized Maize only lists 2447 tortilla shops and

292 dough mills.

*Corresponding author. Email: magdac@servidor.unam.mx

ISSN 1939–3210 print/ISSN 1939–3229 online

� 2011 Taylor & Francis

DOI: 10.1080/19393210.2010.533390

http://www.informaworld.com

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016

Mexico City is a central receiver node for all types

of maize from the rest of the country; therefore, it canbe considered a representative example of the situation

in the whole country, having the highest populationdensity (5871 inhabitants km�2) and representing18.3% of the total Mexican population (INEGI

2005). Therefore, a study of aflatoxin levels in tortillasin Mexico City is relevant.

Materials and methods

Sampling

Since tortillas are not necessarily produced from grainof the previous harvest period, it was more represen-

tative to establish two tortilla sampling periods, i.e. dryand wet, instead of harvest seasons.

Materials, solvents and reagents

Aflatoxin standards (B1, B2, G1, G2) were acquiredfrom Sigma-Aldrich (St. Louis, MO, USA). Aflatoxin

standard solutions (1 mgml�1) were prepared asdescribed in method 970.44 (A) (AOAC 2005).Standards were dissolved in benzene: acetonitrile

(98:2, v/v) and stored in amber-coloured vials at 4�C.All solvents were HPLC grade and the remaining

substances were AR grade from J.T. Baker (Xalostoc,Mexico).

Buffer phosphate solution was prepared accordingto the method of Harrison et al. (1991). Easi-Extract

immunoaffinity columns for total aflatoxin wereobtained from R-Biopharm-Rhone (Glasgow,Scotland).

The derivatizing solution consisted of 5ml trifluor-

acetic acid in 2.5ml glacial acetic acid and 17.5mldeionized water.

Instrumentation

The chromatography pump (Agilent Series 1100) wasequipped with a Rheodyne 7125-075 injector with a

fixed 20-ml loop (Cotati, CA, USA), fluorescencedetector (Perkin Elmer LCI-100) and a data integrator(Perkin Elmer LC–10) to analyze aflatoxins.

Separation was obtained with a reverse-phase C18

column of 4.6mm I.D.� 250mm length� 5 mm dp

(Prodigy ODS 2, Phenomenex Torrance, USA).Mobile phase was water/acetonitrile/methanol

(65:15:20, v/v/v). The chromatographic column waskept at room temperature (20�C), with a mobile phase

flux of 1.2mlmin�1. Wavelength of the fluorescencedetector was adjusted to �ex max 365 nm and �emmax

450 nm.

Study design

Sampling and sample treatment

To obtain representative data on tortillas in MexicoCity, 198 samples of 2-kg each were taken from 98sampling points over two successive years (2006 and2007), giving a total of 396 samples. In each sampledarea, three supermarkets and three tortilla shops werechosen (Figure 1). The samples were weighed, dried at60�C for 48 h and ground dry in 5-l industrial blendersto a fine powder. Later, several amounts of the entireground sample were chosen randomly to give ahomogenized sub-sample of 50 g for chemical analysis.

For recovery percentage studies, three ‘blank’ sub-samples, with no AF, were enriched with a mixture of10, 20 and 30 ng g�1 of each one of AFB1, AFB2,AFG1 and AFG2. They were left to settle overnight to

evaporate the solvents and then chemically analyzed.

Aflatoxin extraction from maize tortillas

To extract the aflatoxins, 50 g dry weight of tortillasamples were blended at high speed, for 2min with

Figure 1. Mexico City’s districts and sampling points:1) Gustavo A. Madero, 2) Azcapotzalco, 3) MiguelHidalgo, 4) Cuauhtemoc, 5) Venustiano Carranza, 6)Cuajimalpa de Morelos, 7) Alvaro Obregon, 8) BenitoJuarez, 9) Iztacalco, 10) Iztapalapa, 11) Coyoacan, 12)La Magdalena Contreras, 13) Tlalpan, 14) Xochimilco, 15)Tlahuac, 16) Milpa Alta.

Food Additives and Contaminants: Part B 43

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016

100ml of a mixture of methanol/H2O (80:20, v/v),adding 1 g of NaCl to clarify the extract (DOF 2002).The blended solution was filtered (Filtrate 1) and 2mlwere diluted with 14ml of phosphate-buffered saline(PBS), to adjust 1 g of analyzed sample. Then, thediluted Filtrate 1 was filtered through a glass-fiberfilter (Filtrate 2) (Commission Directive 1998).

The immunoaffinity columns (Easi-extractTM) fortotal aflatoxins were previously activated with 20mlPBS, at pH 7.4, after which Filtrate 2 was applied at aflow-rate of 5mlmin�1. The immunoaffinity columnwas then washed with 20ml of distilled water and,finally, dried by passing air through it. Bound aflatoxinwas eluted with 2ml of acetonitrile, HPLC grade, in2min; finally, air was pushed through the column tocollect the last drops of eluate.

One aliquot of the eluate was used for HPLCquantitation and aflatoxin identification; another wasstored as reserve; both aliquots were dried and storedat �4�C. The aliquot for HPLC quantitation wasdissolved in 100 ml acetonitrile added to 400 ml ofderivatizing solution and shaken in a vortex (G-560,Bohemia, NY, US) for 30 s. The vials were subjected tovapor bath at 65�C for 10min (Wilson and Romer1991), taken to room temperature and 20 ml injected tothe chromatographicr pump for HPLC analysis.

HPLC analysis

HPLC is the most common method used to quantifyaflatoxins (Jaimez et al. 2000). All solvents weredegassed for 30min using vacuum filtration equip-ment. Water of chromatographic grade was obtainedby filtering distilled water and deionizing in a micro-filtration system (Millipore, Bedford, MA, US) with0.45 mm� 47mm nylon membranes.

Analytical quality assurance

Validation of the analytical method

The method employed for validation and qualityassurance is as follows:

Precision of the system. Five replicates of the AFB1,AFB2, AFG1 and AFG2 standards were prepared at aconcentration of 10 ngml�1 via individual dilutions,derivatized as previously described and the responseswere measured by HPLC (Horwitz 1995; Thompsonand Wood 1995; Thompson 2000). The standarddeviation (SD) and coefficient of variation (CV%)were recorded.

Adequability of the system. An adequability solution(Garcıa and Alcantara 2002) of a mixture of the fourAFs, in the following proportions AFG1/AFB1/AFG2/

AFB2 (40:40:20:20 ngml�1), was injected in triplicateunder the following elution conditions:

Mobile phase ACN/MeOH/H2O (15:20:65, v/v), flowspeed 1.2mlmin�1, fluorescence detection at 36 nmexcitation and 450 nm emission. The CV% of theanalytical response was measured under the followingparameters: K ’: capacity factor42; R: resolution42;�: selectivity; N: number of theorical dishes.

Linearity of the system (calibration curves). Eachconcentration of AFB1, AFB2, AFG1 and AFG2

were injected in triplicate to construct the calibrationcurves. The average of the results, as well as SD, CV%,limit of quantification (LOQ) and limit of detection(LOD), were determined (Garcıa and Alcantara 2002).

The response of the analytical signal (area underthe curve of each chromatographic peak) was plottedwith the concentration of each AF, giving the curveequation, from which the different statistical parame-ters calculated: slope value (b1), origin ordinate (b0),correlation coefficient (r2), and the confidence intervalfor the slope (IC (�1)).

Exactitude and repeatability of the method. There wasno certified reference material (CRM) for tortillas;therefore, the method of addition/recovery was appliedto replicates of 50 g samples of tortillas spiked with10 mg kg�1 of each AF in triplicate. They were homog-enized and left to rest for 25 h; afterwards, they werechemically extracted and analyzed as previouslydescribed. The chromatographic area values helpedto obtain the mathematical average (Y

�

), SD, CV% andthe confidence interval for the population media(IC (m)) of the recovery percentage (Horwitz 1995).

Linearity of the method. Blank tortilla samples wereprepared and then each tortilla sample was spiked with10 mg kg�1 each of AFB1, AFB2, AFG1 and AFG2;each one again with 20 and 30 mg kg�1, they wereextracted and analyzed by HPLC. AF concentrationversus chromatographic area was plotted and thevarious parameters were calculated: spiked standardversus recovered standard, slope value (b1), ordinate oforigin (b0), correlation coefficient (r2) and confidenceinterval for the slope (IC (�1)), confidence interval forthe ordinate of origin (IC (�0)) and coefficient ofvariation regression (CVy/x).

Recovery percentage. Recovery percentage of eachspiked sample was calculated as recovery¼ (amountfound/amount added)� 100 (Thompson et al. 1999;Garcıa and Alcantara 2002).

44 P. Castillo-Urueta et al.

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016

Detection and quantification limits. The detection andquantification limits were determined taking intoaccount the calibration curves and the standarddeviation of the regression (Garcıa and Alcantara2002). Thus, three concentrations were prepared (10,20 and 40 ng g�1), the analytical response measuredand the following parameters were calculated: slopevalue (b1), correlation coefficient (r2), standard devia-tion of the regression (Sy/x) and the confidence intervalfor the slope IC (�1). To obtain the detection limits(Table 1), the following equation was applied:

LOD ¼3:3xSy=x

b1

and the quantification limits (Table 1) were calculatedusing the following equation:

LOQ ¼10xSy=x

b1

Statistical analysis

To compare the four surveys during the wet and drysampling periods of 2006 and 2007, a Wilcoxonnon-parametric statistical analysis was applied.Results were referred to regulatory tolerance limits.Contingence tables were used, with binary-dependentvariables if total aflatoxin concentrations of thesamples were greater than the tolerance limits. Thetolerance limit for total aflatoxin in maize tortillas is12 mg kg�1, in accordance with the Mexican OfficialRule (DOF 2002).

Results and discussion

The % recovery from a mixture of the four aflatoxinsin triplicate, enriched subsamples were: AFB1 (80%),AFB2 (80%), AFG1 (78%) and AFG2 (83%) from 10,20 and 30 mg kg�1 of each AF concentration. The %recovery percentage used to determine the accuracy ofthe method is shown in Table 2. All the r2 wereapproximately 0.999.

Of the 792 kg (396 samples) of tortilla analyzed,17% (68/396) contained aflatoxins and values belowthe limit of quantification. An average of 1.5% (6/396)had several aflatoxins; 9% (34/396) had AF at levelsbelow the limit of quantification and, in 82.5% of thetortilla samples, AF were not detected. From the 34AF positive samples, 21% (7/34) exceeded the maxi-mum permissible limit (12� 2 mg kg�1) in Mexicanlegislation (NOM–187-SSA1-2001) for maize tortilla,equivalent to 2% of total samples, considering a rangeof �2 mg kg�1 given by the HPLC, and 79% (27/34)had AF values below the legal limit. The average AFconcentration in 56 contaminated samples was: 39/56samples had AFB1 at an average of 12.1 mg kg�1, whichis the most frequent and dangerous AF, although thelevels were not high; 15/56 samples had AFB2

(2.7 mg kg�1); 9/56 samples had AFG1, giving anaverage of 64.1mg kg�1; 8/56 samples had AFG2

(with an average of 3.7mg kg�1), and 20 samples withvalues below the limit of quantification. Finally, fortotal AF, the average amount was 20.3mg kg�1

(Table 3). AFB1 may have been present at low levelsbecause it was biotransformed to AFG1 and whyAFG1 had the highest concentrations. Table 4 showsa summary of the AFs and percentage of non-detectable and detectable samples.

Aspergillus flavus produces AFB1 and AFB2, andAspergillus parasiticus produces the four aflatoxinsAFB1, AFB2, AFG1 and AFG2 (Hesseltine et al. 1970).However, not all strains can do so (Diener and Davis1987). Harrison et al. (1987) analyzed maize used forindustrial purposes and found that 32% of the isolatedfungi were A. flavus or A. parasiticus that produceAFB1 and can sometimes produce AFB1 and AFB2.

Table 2. Percentage recovery (�2 SD) used to determine theaccuracy of the method.

Aflatoxins

*SpikedAF

(mg kg�1)

*RecoveredAF

(mg kg�1)Recovery

(%)RSD(%)

AFB1 9.9 8.6 87 1.8AFB2 9.9 8.5 86 1.2AFG1 9.9 8.7 88 2.4AFG2 9.9 8.7 88 2

AFtotal 39.6 34.5 87 1.4

AFB1 19.9 17.2 86 0.9AFB2 19.9 17.1 86 1.2AFG1 19.9 17.6 88 2.5AFG2 19.9 17.2 86 1.3

AFtotal 79.6 69.1 87 1.4

AFB1 29.7 28.4 96 1.2AFB2 29.7 27.1 94 0.8AFG1 29.7 27.2 92 1.8AFG2 29.7 27.2 92 1.3

AFtotal 118.8 109.9 93 0.5

Note: *Average of three replicates (n¼ 3).

Table 1. Limits of detection and quantification foraflatoxins (�2 SD).

Aflatoxins

Limit ofdetection(LOD)(mg kg�1)

Limit ofquantification

(LOQ)(mg kg�1)

RepeatabilityCV(%)

AFB1 0.5 1.4 1.8AFB2 0.6 1.9 1.2AFG1 0.4 1.2 2.4AFG2 0.4 1.2 2AFtotal 1 3.1 1.5

Food Additives and Contaminants: Part B 45

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016

Table 3. Aflatoxin levels in tortillas from the four surveys (�2 SD). Retention times (min) forAFB1 (8.12–8.80), AFB2 (19.47–20.9), AFG1 (6.31–6.55) and AFG2 (13.91–14.8). Aflatoxin[AFB1, AFB2, AFG1,AFG2 and total AF (AFtotal)] concentrations (mg kg�1) per gram of tortilladry weight.5LOD¼Below limit of detection;5LOQ¼ below limit of quantification.

Borough Sample AFB1 AFB2 AFG1 AFG2 AFtotal

Survey I on April 2006 (�g kg�1)Iztap 1 13 0 0 0 13

2 8 0 0 0 8

3 3 0 0 0 3

4, 7 0 0 0 0 05 13 0 0 0 13

6 5 0 0 0 5

Mad 8 2 0 0 0 2

9 5LOD 0 0 0 –10 2 0 0 0 2

Nov–13 0 0 0 0 0Obr 14, 16, 18, 19 0 0 0 0 0

15 0 3 0 0 317 0 0 0 5LOQ –

Coy 20 0 2 2 0 4

21, 22, 24 0 0 0 0 023 0 0 0 5LOQ –

25 0 5LOQ 0 0 –

Tlal 26 0 3 0 0 3

27 0 2 0 0 228 0 2 0 0 2

29 0 3 0 0 3

30–31 0 0 0 0 0Cuau 32–37 0 0 0 0 0Carr. 38–44 0 0 0 0 0Azcap 45–47, 49, 50 0 0 0 0 0

48 4 2 4 5 15

Iztac 51–56 0 0 0 0 0Xoch 57–62 0 0 0 0 0Jua 63–67 0 0 0 0 0

68 5 5LOD 4 2 11Hgo 69–74 0 0 0 0 0Tlah 75, 77, 80 0 0 0 0 0

76 0 5LOQ 4 0 4

Contr 81 106 5LOQ 0 0 10682–86 0 0 0 0 0

Cuaj 87–89, 91–92 0 0 0 0 090 30 5LOQ 0 0 30

M.Alta 93–98 0 0 0 0 0

AFtotal April 06 191 17 14 7 229

Survey II on November 2006 (�g kg�1)Iztap 1, 3–7 0 0 0 0 0

2 2 0 0 0 2Mad 8,11–13 0 0 0 0 0

9 5 0 0 0 5

10 2 0 0 0 2

Obr 14–19 0 0 0 0 0Coy 20–25 0 0 0 0 0Tlal 26–29, 31 0 0 0 0 0

30 1.4 0 0 0 1.4Cuau 32, 35–37 0 0 0 0 0

33 5LOQ 2 0 0 2

34 2 0 0 0 2

Carr 38–44 0 0 0 0 0Azcap. 45 2 5LOQ 0 0 2

46, 48, 50 0 0 0 0 047 5LOQ 0 0 0 5LOQ

49 0 2 0 0 2

(continued )

46 P. Castillo-Urueta et al.

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016

Table 3. Continued.

Borough Sample AFB1 AFB2 AFG1 AFG2 AFtotal

Iztac. 51–56 0 0 0 0 0Xoch. 57–62 0 0 0 0 0Jua. 63–68 0 0 0 0 0Hgo 69–74 0 0 0 0 0Tlah. 75–80 0 0 0 0 0Contr. 81 8 0 0 0 8

82 5LOQ 4 385 4 39383, 84, 86 0 0 0 0 0

85 5 5LOQ 5 6 16

Cuaj 87–92 0 0 0 0 0M.Alta 93, 94, 0 0 0 0 0

96–9895 10 0 0 0 10

Carr 99 0 0 0 0 0

AFtotal November 06 39 8 390 10 445

Survey III on April 2007 (�g kg�1)Iztap 1–Jul 0 0 0 0 0Mad Aug–13 0 0 0 0 0Obr 14–19 0 0 0 0 0Coy 20–25 0 0 0 0 0Tlal 26–28, 30 0 0 0 0 0

29 7 0 0 0 730 0 0 0 0 031 141 0 150 0 291

Cuau 32 3 0 0 0 3

33–37 0 0 0 0 0Carr 38–41 0 0 0 0 0

42 6 0 0 0 6

43 6 0 0 0 6

44 0 0 0 0 0Azcap 45–48 0 0 0 0 0

49 0 5 0 0 5

50 8 0 0 0 8Iztac 51–56 0 0 0 0 0Xoch 57–60,62 0 0 0 0 0

61 0 0 5LOD 0 5LOQ

Jua 63, 64, 66–68 0 0 0 0 065 6 0 0 0 6

Hgo 69–74 0 0 0 0 0Tlah 75, 77–80 0 0 0 0 0

76 11 0 0 0 11Contr 81 6 0 0 0 6

82–86 0 0 0 0 0Cuaj 87, 90–92 0 0 0 0 0

88 8 0 0 0 8

AFtotal April 07 202 5 150 0 357

Survey IV on November 2007 (�g kg�1)Iztap 1, 2, 4–7 0 0 0 0 0

3 0 0 0 4 4Mad. Aug–13 0 0 0 0 0Obr. 14–16, 19 0 0 0 0 0

17 0 3 0 0 3

18 0 5LOQ 0 0 –Coy 20, 21, 24, 25 0 0 0 0 0

22–23 2 0 0 0 2

5LOQ –Tlal 26–31 0 0 0 0 0Cuau 32–37 0 0 0 0 0Carr 38–42,44 0 0 0 0 0

43 0 2 0 0 2

(continued )

Food Additives and Contaminants: Part B 47

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016

Table 4. AFs and percentage of non-detectable and detectable samples.

Samples

Non-detectable AF Detectable AF

Sampling periods AF Amount of samples % Amount of samples % Total

AFB11 87 88.8 11 11.22 982 89 90.8 9 9.18 983 90 90 10 10 1004 91 91 9 9 100

Total 357 39 396

AFB21 91 92.9 7 7.14 982 95 96.9 3 3.06 983 99 99 1 1 1004 97 97. 0 3 3 100

Total 382 14 396

AFG11 94 95.9 4 4.08 982 96 98 2 2.04 983 99 99 1 1 1004 98 98 2 2 100

Total 387 9 396

AFG21 96 98 2 2.04 982 96 98 2 2.04 983 100 100 0 0 1004 96 96 4 4 100

Total 388 8 396

Table 3. Continued.

Borough Sample AFB1 AFB2 AFG1 AFG2 AFtotal

Azcap 45–50 0 0 0 0 0Iztac 51–56 0 0 0 0 0Xoch 57, 60, 61 0 0 0 0 0

58 0 0 0 2 2

59 0 0 0 2 2

62 3 0 0 0 3

Jua 63, 65, 67 0 0 0 0 064 5 0 0 0 5

66 4 0 0 0 4

68 5 0 0 0 5

Hgo 69, 72–74 0 0 0 0 070 4 0 0 0 4

71 11 3 17 0 31

Tlah 75–80 0 0 0 0 0Contr 81, 82, 85, 86 0 0 0 0 0

83 0 0 6 5LOQ 6

84 5LOQ 5LOQ 0 0 –

Cuaj 87 4 0 0 0 488–92 0 0 0 0 0

M.Alta 93–94,96–97 0 0 0 095 9 0 0 0 9

98 0 5LOQ 0 0 –47 8 23 8 86

Notes: Iztap¼ Iztapalapa; Mad.¼Gustavo A. Madero; Obr.¼ Alvaro Obregon; Cuauh.¼Cuauhtemoc; Azcap.¼Azcapotzalco; Contr.¼La Magdalena Contreras; Carr.¼VenustianoCarranza; Coy¼Coyoacan; Iztac.¼ Iztacalco; Xoch¼Xochimilco; M.Alta¼Milpa Alta;Hgo¼Miguel Hidalgo; Jua¼Benito Juarez; Cuaj¼Cuajimalpa; Tlah¼Tlahuac; Coy¼Coyoacan; Tlal¼Tlalpan.

48 P. Castillo-Urueta et al.

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016

AFB1 is the most mutagenic, teratogenic andcarcinogenic aflatoxin (Richard 2007), and it wasfound in 64.6% of the ‘tortillas’ at levels rangingfrom 3.0 to 140.3 mg kg�1, which are lower from thosefound by Torreblanca et al. (1986). There might be aconnection between the daily amount of AF ingestedby Mexicans via tortillas/other maize products and thefact that Mexico has the highest incidence of liverdiseases on the American continent (OPS 2002).

Maize contamination by A. flavus and A. para-siticus is variable due to environmental favourableconditions, such as dryness and hot weather (McGeeet al. 1996). Sterigmatocystin dimethyl can form AFB1

and AFG1 (Trail et al. 1995), and both represent a highrisk to human health. In the present study, AFB1 wasmore frequent (�1.4mg kg�1) in Mexican tortillas, butAFG1 presented higher concentrations (385.2mg kg�1).

This was probably due to differences in biosyn-thetic activities as the microsome enzyme fractioninvolved in G-group aflatoxin formation is less stablethan the fraction involved in B-group formation (Asaoet al. 1963; Yabe et al. 1999).

Although many samples did not exceed the legallimits for AF, they are accumulated in DNA and RNAas adducts, causing damage over time (Wang andGroopman 1999). Therefore, tortillas must be includedas risk food for AF.

The presence of AFG1, a potent rat kidney carcin-ogen (Butler et al. 1969), indicates that it was producedby either A. parasiticus or A. nomius (Erhlich et al.2007), because these fungi can synthesize it simulta-neously with AFB1. Other studies reported that G and Bgroup aflatoxins are independently produced from acommon branching intermediate O-methyl sterigmato-cystin for AFB1 and AFG1, and dihydro-O-methylster-igmatocystin for AFB2 and AFG2 (Yabe et al. 1999).The ord1 gene from A. flavus (Prieto and Woloshuk1997) and two genes from A. parasiticus – the nadAgene, involved in the formation of G group aflatoxin(Cai et al. 2008) and the ordA (Yu et al. 1998), arerelated to the ability to produce AFB1, AFB2, AFG1

and AFG2. Both the ordA and ord1 gene encodecytochrome P450-type monooxygenases based on theirDNA sequences, and the biosynthesis of aflatoxinsrequires NADPH.

Wilcoxon/Kruskal–Wallis and Wilcoxon testsshowed that the two sampling periods (April andNovember) were not statistically different.Contingency table analysis showed that the frequencyof mutagen level, 410 mg kg�1 AFB1, changed withsampling period (P50.05).

The contamination levels of aflatoxin in maizetortillas for the different districts were not statisticallysignificant for any particular aflatoxins or for theentire range of AFs.

There was a significant difference between aflatoxinconcentrations for the ‘‘type of stores’’ only for AFB1

(p¼ 0.0320) and AFG2 (p¼ 0.0465); for total aflatoxinthere was no significant difference. Wilcoxon/Kruskal–Wallis tests showed that, for AFB1, the supermarketmean was 1.35 and for the ‘tortilla’ shop it was 1.12.For AFG2, the supermarket mean was 0.028 and forthe ‘tortilla’ shop it was 0.1134.

To observe the dispersion of data, a descriptivemethod was applied to the aflatoxin distribution.A non-parametric Wilcoxon test was applied to com-pare the sample medians for the existing differences inconcentration of each aflatoxin between different typesof stores (tortilla shops and supermarket).

Conclusion

Aflatoxins are present in maize tortillas in MexicoCity. Alkaline treatment reduces aflatoxin levels intortillas, but does not eliminate them completely.Of 396 tortilla samples from tortilla shops and super-markets in Mexico City, 21% were contaminated withaflatoxins (3–385 mg total AF kg�1). Of these, 10%were above the maximum permissible limit establishedby Mexican law NOM–187-SSA1-2001, i.e. 12 mg kg�1

in tortillas. Although 90% of the contaminated sam-ples comply with national regulatory levels, there is acancer risk by the accumulation of aflatoxins fromfood in DNA (Essigman et al. 1977; Groopman et al.1988; Wild et al. 1990; Wogan 1992; IARC 2002). Thepresent study is the most complete analysis of AF inmaize tortillas in Mexico, by the number of samplesand with validated analytical methodology.

Acknowledgements

The authors acknowledge the following institutions andindividuals: Direccion General del Personal Academico(DGAPA), Universidad Nacional Autonoma de Mexico forfunding Project IN226807 Papiit, DGAPA, UNAM; ConsejoNacional de Ciencia y Tecnologıa (CONACyT) for the PhDscholarship No. 181805 which has allowed Pavel CastilloUrueta to work on this research topic; Instituto de Biologıaand Posgrado de Ciencias Quımicas, Universidad NacionalAutonoma de Mexico, specifically its PhD programme, andthe administration and facilities provided; Jorge Lopez, JoelVillavicencio and Georgina Ortega Leite from the Institute ofBiology, Universidad Nacional Autonoma de Mexico forcomputing and data base support.

References

AOAC. 2005. Official methods of analysis. 17th ed. Howitz

W, editor. Food Composition; Additives; Natural

Contaminants. Subchapter 1: Mycotoxins. Subchapter 2:

Aflatoxins. Vol. II. Chapter 49:1–42. Washington (DC):Association of Official Analytical Chemists.

Asao T, Buchi G, Abdel-Kader MM, Chang SB, Wick EL,

Wogan GN. 1963. Aflatoxins B and G. J Am Chem Soc.85:1706–1707.

Food Additives and Contaminants: Part B 49

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016

Butler WH, Greenblatt M, Lijinsky W. 1969. Carcinogenesis

in rats by aflatoxins B1,G1 and B2. Cancer Res. 29:2206.Cai J, Zeng H, Shima Y, Hatabayashi H, Nakagawa H,

Ito Y, Adachi Y, Nakajima H, Yabe K. 2008. Involvement

of the nadA gene in formation of G-group aflatoxins in

Aspergillus parasiticus. Fung Genet Biol. 45:1081–1093.

Commission Directive. 1998. Commission Directive 98/53/

EC. Laying down the sampling methods and methods of

analysis for the official control of the levels for certain

contaminants in foodstuffs. Off J Eur Commun. L 201:93.Coomes TJ, Crowther PC, Fewell AJ, Francis BJ. 1966.

Experimental detoxification of groundnut meals contain-

ing aflatoxin. Nature. 209:406–407.

De Arriola MC, De Porres E, De Cabrera S, De Zepeda M,

Rolz C. 1988. Aflatoxins fate during alkaline cooking of

corn for tortilla preparation. J Agric Food Chem.

36:530–533.Diener UL, Davis ND. 1987. Biology of Aspergillus flavus

and A. parasiticus. In: Zuber MS, Lillehoj EB, Renfro BL,

editors. Aflatoxin in maize. Proceedings of the workshop

held in El Batan. Mexico: CYMMYT/UNDP/USAID.

p. 33–40.

DOF 2002. Diario Oficial de la Federacion. Productos y

servicios. Masa, tortillas, tostadas y harinas preparadas

para su elaboracion y establecimientos donde se procesan.

Especificaciones sanitarias. Informacion comercial.

Metodos de prueba. Mexico, D.F. Mexico: Norma

Oficial Mexicana, 18 de agosto NOM–187-SSA1/SCFI-

2002. 1–29.Erhlich KC, Kobbeman K, Montalbano BG, Cotty PJ. 2007.

Aflatoxin producing Aspergillus species from Thailand. Int

J Food Microbiol. 114:153–159.

Essigmann JM, Croy RG, Nadzan AM, Busby WF,

Reinhold VN, Buchi G, Wogan GN. 1977. Structural

identification of the major DNA adduct formed by

aflatoxin B1 in vitro. Proc Natl Acad Sci USA.

74:1870–1874.Fernandez-Munoz JL, San Martın-Martınez E, Dıaz-

Gongora JAI, Calderon A, Alvarado-Escobar A,

Ortiz-Cardenas H, Leal-Perez M. 2006. Steeping time

and cooking temperature dependence of calcium ion

diffusion during microwave nixtamalization of corn.

J Food Eng. 76:568–572.

Garcıa MA, Alcantara A. 2002. Guıa de validacion

de metodos analıticos. Colegio Nacional de Quımicos

Farmaco-Biologos, A.C. Mexico, D.F. Mexico.

p. 1–122.Giddey C, Brandt J, Bunter G. 1977. The detoxification of

oil-seed cakes polluted by aflatoxins: research and devel-

opment of an industrial process. Ann Technol Agric.

27:331�338.

Groopman JD, Cain LG, Kensler TW. 1988. Aflatoxin

exposure in human populations: measurements and rela-

tionship to cancer. Crit Rev Toxicol. 19:113–145.Gutierrez E, Rojas-Molina I, Pons-Hernandez JL,

Guzman H, Aguas-Angel B, Arenas J, Fernandez P,

Palacios–Fonseca A, Herrera G, Rodrıguez ME. 2007.

Study of calcium ion diffusion in nixtamalized quality

protein maize as a function of cooking temperature. Cereal

Chem. 84:186–194.Harrison MA, Silas JC, Carpenter JA. 1987. Incidence of

aflatoxigenic isolates of Aspergillus flavus/parasiticus

obtained from Georgia corn processing plant. J Food

Qual. 10:101–105.Harrison JC, Carvajal M, Garner RC. 1991. Immunological

detection of aflatoxin B1- DNA and aflatoxin B1-serum

albumin adducts in rat and human tissues. In: Human

Carcinogen Exposure: Biomonitoring and Risk

Assessment. Oxford University Press. p. 255–265.Hesseltine CW, Shotwell OL, Smith M, Elis JJ,

Vandergraft E, Shannon G. 1970. Production of various

aflatoxins by strains of Aspergillus flavus. In: Herzberg M.,

editor. Toxic micro-organisms. Washington (DC): UJNR

and US Dept. Interior. p. 490.Horwitz W. 1995. Protocol for the design, conduct and

interpretation of method-performance studies. Pure Appl

Chem. 67:331–343.IARC. 2002. Some traditional herbal medicines, some

mycotoxins, naphthalene and styrene. Monographs on

the evaluation of carcinogenic risks to humans. Vol. 82.

Lyon: International Agency for Research on Cancer.INEGI 2008. Instituto Nacional de Estadıstica, Geografıa e

Informatica [Internet]. II Conteo de Poblacion y Vivienda

2005. 2008. Mexico. INEGI. [Accessed: 2008 August 22].

Gobierno de Mexico. Available from: http://www.inegi.

gob.mx/est/contenidos/espanol/sistemas/conteo2005/iter2005/

consultafiltro.aspx?npag¼1Islas-Hernandez JJ, Rendon-Villalobos R, Agama-

Acevedo E, Gutierrez-Meraz F, Tovar J, Arambula-

Villa G, Bello-Perez LA. 2006. In vitro digestion rate

and resistant starch content of tortillas stored at two

different temperatures. Lebensm.-Wiss. Technol. 39:

947–951.

Jaimez J, Fuente CA, Vazquez BI, Franco CM, Cepeda A,

Mahuizer G. 2000. Application of the assay of aflatoxins

by liquid chromatography with fluorescence detection in

food analysis. J Chromatogr A. 882:1–10.Liu Z, Gao J, Yu J. 2006. Aflatoxins in stored maize and rice

grains in Liaoning Province, China. J Stor Prod Res.

42:468–479.

McGee DC, Olanya OM, Hoyos GM, Tiffany LH. 1996.

Populations of Aspergillus flavus in the Iowa cornfield

ecosystem in years not favorable for aflatoxin contamina-

tion of corn grain. Plant Dis. 80(7):742–746.OPS 2002. Organizacion Panamericana de la Salud. La salud

de las Americas. Vol. 1. Publicacion Tecnica No. 587.

Washington (DC): OMS.Price RL, Jorgensen KV. 1985. Effects of processing on

aflatoxin levels and on mutagenic potential of tortillas

made from naturally contaminated corn. J Food Sci.

50:347–349.Prieto R, Woloshuk CP. 1997. ord1, an oxidoreductase gene

responsible for the conversion of O-methylsterigmatocys-

tin to aflatoxin in Aspergillus flavus. Appl Environ

Microbiol. 63:1661–1666.Ribeiro S. 2005. Las tortillas, Seccion: Numeros. Suplemento

semanal de La Jornada. 17 de enero. Available from:

www.cna.org.mx/dtorio/96.htmlRichard JL. 2007. Some major mycotoxins and their

mycotoxicotoxicoses-An overview. Int J Food Microbiol.

119:3–10.Rojas-Molina I, Gutierrez-Cotez E, Palacios-Fonseca A,

Banos L, Pons-Hernandez JL, Guman- Maldonado SH,

Pineda-Gomez P, Rodrıguez ME. 2007. Study of structural

50 P. Castillo-Urueta et al.

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016

and thermal changes in endosperm of quality proteinmaize during traditional nixtamalization process. Cereal

Chem. 84:304–312.Thompson M. 2000. Towards a unified model of errors inanalytical measurement. Analyst. 125:2020–2025.

Thompson M, Wood R. 1995. Harmonized guidelines for

internal quality control in analytical chemistry labora-tories. Pure Appl Chem. 67:649–666.

Thompson M, Ellison SLR, Fajgelj A, Willetts P, Wood R.

1999. Harmonised guidelines for the use of recoveryinformation in analytical measurement. Pure Appl Chem.71:337–348.

Torreblanca RA, Bourges H, Morales J. 1986. Aflatoxins inmaize and tortillas in Mexico. In: DF. Zuber MS, LillehojEB, Renfro BL, editors. Aflatoxin in Maize: Proceedingsof the Workshop CYMMYT/UNDP/USAID held in El

Batan, Mexico. Chapter 4; p. 310–317.Torres P, Guzman-Ortız M, Ramırez-Wong B. 2001.Revising the role of pH and thermal treatments in

aflatoxin content reduction during the tortilla and deepfrying processes. J Agric Food Chem. 49:2825–2829.

Trail F, Mahanti N, Linz JE. 1995. Molecular biology of

aflatoxins biosynthesis. Microbiology. 141:755–765.

Wild CP, Jiang YZ, Allen SJ, Jansen LAM, Hall AJ,Montesano R. 1990. Aflatoxin–albumin adducts in human

sera from different regions of the world. Carcinogenesis.11:2271–2274.

Wilson Th J, Romer Th R. 1991. Use of MycoSepmultifunctional cleanup column for liquid chromato-

graphic determination of aflatoxin in agricultural pro-ducts. J AOAC. 74:951–956.

Wang J-S, Groopman JD. 1999. DNA damage by mycotox-

ins. Mutat Res. 424:167–181.Wogan GN. 1992. Aflatoxins as risk factors for primaryhepatocellular carcinoma in humans. Cancer Res Suppl.

52:2114s–2118s.Yabe K, Nakamura M, Hamasaki T. 1999. Enzymaticformation of G-group aflatoxins and biosynthetic relation-ships between G and B group aflatoxins. Appl Environ

Microbiol. 65(9):3867–3872.Yu J, Chang PK, Ehrlich KC, Cary JW, Montalbano B,Dyer JM, Bhatnagar D, Cleveland TE. 1998. Characteriza-

tion of the critical amino acids of Aspergillus parasiticuscitochrome P450 monooxygenase encoded by ordA that isinvolved in the biosynthesis of aflatoxins B1, G1, B2 and G2.

Appl Environ Microbiol. 64:4834–4841.

Food Additives and Contaminants: Part B 51

Dow

nloa

ded

by [

UN

AM

Ciu

dad

Uni

vers

itari

a] a

t 10:

56 2

6 M

ay 2

016