Analytica Chimica Acta 829 (2014) 68–74

Aflatoxin B1 and M1 in milk

Scaglioni P.Ta,*, Becker-Algeri T. b, Drunkler D. b, Badiale-Furlong E. a

a Laboratório de Ciência de Alimentos, Escola de Química e Alimentos, Universidade Federal do Rio Grande-FURG, Avenida Itália, km 8, Bairro Carreiros, RioGrande do Sul 96203-900 Brazilb Programa de Pós Graduação em Tecnologia de Alimentos, Universidade Tecnológica Federal do Paraná-UTFPR, Avenida Brasil, Câmpus Medianeira, 4232 -Parque Independência, Medianeira, Parana 85884-000, Brazil

H I G H L I G H T S G R A P H I C A L A B S T R A C T

� The mobile phase chosen consistsacetic acid:acetonitrile:methanol(55:10:35).

� We obtained a separation factor of2.3 for AFLAB1 and AFLAM1.

� The milk samples were contaminat-ed by aflatoxins above the legislatedlimits.

� Aflatoxins tend to get attached to clotafter acid precipitation of milk.

A R T I C L E I N F O

Article history:Received 28 February 2014Received in revised form 13 April 2014Accepted 18 April 2014Available online 26 April 2014

Keywords:Aflatoxin B1

Aflatoxin M1

Analytical methodMilkLiquid chromatography

A B S T R A C T

The aflatoxin M1 (AFLAM1) is a mycotoxin that results from the hydroxylation of the aflatoxin B1 (AFLAB1).It contaminates the milk of animals fed with a diet containing its precursor. In this work, we determinedthe occurrence of AFLAB1 and AFLAM1 in milk, as well as the chromatographic conditions to quantifythese mycotoxins. The extraction and quantification of AFLAB1 and AFLAM1 in naturally contaminatedand artificially spiked milk samples which are produced and marketed in the state of RS were performedusing the AOAC official method and UHPLC with fluorescence detection. We obtained a separation factorof 2.3 for AFLAB1 and AFLAM1 using a mobile phase consisting of 1% acetic acid:acetonitrile:methanol(55:10:35). The analytical curves had a wide linearity range and the limit of quantification (LOQm)concentrations of AFLAB1 and AFLAM1 were equal to 0.5 and 0.25 mg L�1, respectively. Samples ofpasteurized and ultra-high-temperature processed (UHT) milk showed natural contamination, and thelevels for both aflatoxins ranged from 0.7 to 1.5 mg L�1. Raw and concentrated milk samples onlycontained AFLAM1, with a maximum average concentration of 1.7 mg L�1. These concentrations, higherthan permitted by legislation, confirm the existence of a health risk, as well as highlight the relevance ofsearching for alternatives to reduce this contamination.

ã 2014 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Analytica Chimica Acta

journal homepa ge: www.elsev ier .com/locate /aca

* Corresponding author. Tel.: +55 53 32935374/81011107; fax: +55 53 32338645.E-mail addresses: priscilascaglioni@gmail.com, ptscaglioni@gmail.com

(P.T. Scaglioni).

http://dx.doi.org/10.1016/j.aca.2014.04.0360003-2670/ã 2014 Elsevier B.V. All rights reserved.

1. Introduction

Milk is a food that provides macro and micronutrients for thegrowth, development and maintenance of human health. However,it may also be a vehicle of environmental and food contaminants,causing various physiological alterations in individuals whoconsume it. Microorganisms present in milk, as well as theirmetabolites, are able to migrate in the fluids and tissues of

P.T. Scaglioni et al. / Analytica Chimica Acta 829 (2014) 68–74 69

breeding mammals subsequently harming their health. There is anespecial concern for children and infants, who are more suscepti-ble than adults to the toxic effects of contaminants such asaflatoxins present in milk [1].

Aflatoxins are secondary metabolites produced by fungal speciessuch as Aspergillus flavus, Aspergillus parasiticus and Aspergillusnomius. These organisms contaminate agricultural products andproduce especially aflatoxins B1 (AFLAB1), B2 (AFLAB2), G1 (AFLAG1)and G2 (AFLAG2). Animals that ingest food contaminated withAFLAB1 and AFLAB2 can hydroxylate these aflatoxins and yield dairyproducts contaminated with Aflatoxins M1 (AFLAM1) and M2

(AFLAM2) [2]. Aflatoxins are factors involved in the etiology ofhuman liver cancer. In animals, the ingestion of contaminated foodwith a certain concentration of aflatoxins was followed byteratogenesis and, especially during the first embryonic phase,occurs the malformation of fetuses and the reabsorption of embryos[3,4]. The AFLAB1 is classified as Group 1 (carcinogenic to humans),while AFLAM1 is classified as Group 2B (possibly carcinogenic tohumans) [5]. The World Health Organization recommended thereduction of the AFLAM1 levels in milk and dairy products to aminimum, in order to minimize the potential risk it poses. The toxicand carcinogenic effects have been extensively demonstrated inseveral species, especially inyounganimals [6,7]. There is, therefore,a great concern regarding the health of children, considering theirhigh consumption of milk and dairy products, their low body weightand their great susceptibility to aflatoxins [5].

The first methods for determining AFLAM1 in samples of milkand dairy products were developed in the mid 60s, based on thefluorescent property of the toxin when exposed to ultraviolet light.Thus, the first technique to identify and quantify was the thin layerchromatography (TLC) [8]. Later, with the development of methodsfor ultra high performance liquid chromatography (UHPLC), therewas a notable increase in the degree of precision of thedeterminations, however with high costs [9]. The use of methodsfor multitoxins is recommended to better assess the possibility of asynergistic effect, as well as to circumvent the cost impact ofanalytical determinations. These methods are especially importantwhen there is a metabolic relation between two analyzed toxins,like in the case of AFLAM1 and AFLAB1. Both toxins can occur inmilk if the precursor AFLAB1 was abundant enough in the food tosurpass the metabolic capacity of the animal [10].

Reviews and validations of methods for the determination ofAFLAM1 are constantly made in order to refine the techniques ofextraction, purification and quantification, in order to improve thereliability of the results concerning the impact of this compound inmilk and meet the current demand for minimum generation ofwaste. However, there are few studies on adapting these methodsto determine the precursor toxin (AFLAB1) that may be circulatingin the fluids of animals without being metabolized too.

In this study we sought to standardize a method tosimultaneously quantify the concentrations of AFLAB1 and AFLAM1

in milk destined to human consumption.

2. Materials and methods

2.1. AFLAB1 and AFLAM1 standards

The standards for AFLAB1 and AFLAM1 were purchased fromSigma–Aldrich Chemical Company. They were resuspended inbenzene:acetonitrile (98:2), resulting in the desired concentra-tions for each analysis.

2.2. Milk sampling

The milk samples were collected from July to October 2012,amounting to 40 samples. They were acquired from producers in

the southern Brazil (raw milk), local shops (pasteurized milk andultra-high temperature-UHT) and a beneficiation industry of theregion (raw, pasteurized, concentrated, powder and UHT).

Samples of raw milk were stored in polyethylene terephthalatepacks provided by the producer and kept under cooling tempera-ture (4 �C) until the time of analysis, which did not exceed 24 hafter the acquisition thereof.

The samples obtained in the local commerce were kept underrefrigeration from the time of acquisition (pasteurized milk) orafter opening the package (UHT). The sampling unit consisted ofsealed containers without apparent deformation, within thevalidity period, containing 1 L of sample and from differentmanufacturing batches. They were chosen considering theavailability of lots in major supermarkets in the city of Rio Grande,RS, Brazil. In view of the geographical distance between the factoryand the laboratory where the analysis were performed, thesamples derived from the processing industry were inoculatedwith an specific preservative to maintain the characteristics of thematerials unchanged, namely 2-bromo-2-nitropropane-1,3-diol(Bronopol1).

2.3. Extraction of AFLAB1 and AFLAM1 from milk

Using a method described by AOAC [11], the volumes of both thesample and the solvents were reduced three times in order tominimize the generation of waste. After that, the extraction wascarried out using methanol and celite. A subsequent partition wascarried out using 4% sodium chloride and hexane. The extractionand the removal of any remaining water in the extracts wereperformed, respectively, with chloroform and anhydrous sodiumsulfate. The final extracts were dried using nitrogen current andstored at freezing temperature until the time of quantification.

2.4. Establishment of the chromatographic conditions for quantifyingAFLAB1 and AFLAM1 by UHPLC- FL

Tests were carried out to choose the mobile phase that wouldresult in the best chromatographic separation of aflatoxins withthe lowest interference of the matrix components. These testswere based on the solvent mixture established by AOAC [11],namely 1% acetic acid:acetonitrile:methanol (40:35:25).

The run was performed on a ultra high-efficiency liquidchromatograph (UHPLC) consisting of a LC-10 AT pump, a DGUdegasser, a CBM-20A controller, a SPD-20A fluorescence detector, a20 mL 7725i manual injector, a LC Solution-Shimadzu software anda Kromasil1 C18 5 mm 250 � 4.6 mm column. The standards andextracts of the samples were eluted at 35 �C with a flow rate of1.0 mL min�1. The detection of mycotoxins was performed inwavelengths of 360 nm to 450 nm for absorption and emission,with a loop of 20 mL.

The retention factor (k) was calculated dividing the correctedretention time of each aflatoxin by the dead time. The separationfactor (a) was determined by the kA/kB ratio, knowing that thesolute B is more retained in the column than the solute A [12,13].

Two standard curves were built, one for each aflatoxin, using thestandard solutions of AFLAB1 and AFLAM1, which were diluted in thesolvent mixture that makes up the mobile phase, at increasingconcentrations (4.5 mg L�1; 9 mg L�1; 18 mg L�1; 27 mg L�1; 54 mgL�1; 81mg L�1; 108 mg L�1; 162mg L�1 and 270 mg L�1). Two othercurves, one for each aflatoxin, were made from the fortification (intriplicate) of milk samples with eight increasing concentrations ofaflatoxins (0.25mg L�1; 0.5 mg L�1; 4.5 mg L�1; 18 mg L�1; 54mg L�1;81mg L�1; 108 mg L�1 and 162mg L�1). These samples went throughthe same extraction process as the others, according to the AOACmethod, and were subsequently quantified by UHPLC-FL. Then, the

70 P.T. Scaglioni et al. / Analytica Chimica Acta 829 (2014) 68–74

range of linearity, the coefficients of determination and correlation,as well as the limits of detection (LOD) and quantification (LOQ) foreach mycotoxin were determined.

The determination of the detection limit of the instrument(LODi) was performed by injecting the standard mixture solutionof AFLAB1 or AFLAM1 diluted in the solvent mixture that makes upthe mobile phase. These injections were made at decreasingconcentrations, until the acquiring of a 3:1 ratio between theanalyte peak and the base line noise near the retention time of thepeak of aflatoxins. To determine the quantification limit of theinstrument (LOQi), the analytical solutions prepared were injectedin the solvent until a 10:1 ratio between the analyte peak and thebaseline noise was obtained. To determine the detection limit(LODm) and the quantitation limit of the method (LOQm), extractsof the spiked samples were injected until the same relation to theinstrumental limits described was achieved [14].

2.5. Recovery test of the AOAC extraction method

The determination of the recovery percentage of the methodwas performed by fortifying milk samples with the standards ofAFLAB1 and AFLAM1. The aliquots of each standard weretransferred to flasks and, after evaporation of the solvent, aliquotsof milk were added. Six levels of fortification were used, based onthe legislated limits for milk AFLAM1 in Brazil (0.5 mg kg�1) andconsidering the LOQm estimated for each of the aflatoxins. Thefollowing concentrations were tested: 0.25 (only for AFLAM1);0.5 mg L�1; 3 mg L�1; 6 mg L�1; 8 mg L�1 and 10 mg L�1. The extrac-tion and the quantification of the dried extract in 1 mL of themixture that consisted in the mobile phase in the UHPLC-FL wereperformed by the AOAC method [11]. The percentage recovery wasestimated by relating the concentration of aflatoxins found withthe expected concentration, as shown in Eq. (1).

%R ¼ C1 � C2

C3

� �� 100 (1)

where %R-recovery percentage; C1-concentration determined inthe spiked sample; C2-concentration determined in the non-spiked sample; and C3-concentration of the standard used for thefortification.

2.6. Occurrence of the AFLAB1 and AFLAM1 in milk samples and theirdistribution in fractions of the raw material

A natural contamination by aflatoxins was observed in milksamples that underwent different processes such as, raw milk,pasteurized, UHT, concentrated and powdered.

To investigate the distribution of aflatoxins in raw materialduring the processing of milk for cheese production, an acidcoagulation of milk was performed, by the method determined bythe Ministry of Agriculture, Livestock and Food Supply [15]. To

Table 1Separation efficiency of the mycotoxins eluted by different mobile phases.

Exp. Acetic acid 1%:acetonitrile:methanol Retention time (tr) (m

AFLAM1

1 40:35:25 3.4

2 50:15:35 5.2

3 55:10:35 6.6

4 60:05:35 11.8

5 65:10:25 13.5

Exp.-experiment, AFLAB1-aflatoxin B1, AFLAM1-aflatoxin M1.

separate the milk serum from its clot, a solution of 10% glacialacetic acid was added to a sample of pasteurized milk preheated to35 �C until the sample reached a pH of 4.6. The levels of AFLAB1 andAFLAM1 and the protein content in both fractions were deter-mined. This procedure was performed in samples of control milk(without contamination by addition of the standard) and onsamples spiked with both aflatoxins standards in the finalconcentration of 5 mg L�1. For these determinations, we usedthe extraction method of AOAC [11], as well as a subsequentquantification by UHPLC-FL.

3. Results and discussion

To obtain an optimal chromatographic separation betweenAFLAB1 and AFLAM1 and a reasonable run duration, different ratiosof mobile phase were tested (Table 1). Tests were made for theretention factor, which indicates the relation between the timeeach mycotoxin was retained in the stationary phase and the timethat it was being carried by the mobile phase. The separation factorwas also tested, indicating the selectivity of the chromatographicsystem of mycotoxins eluting at adjacent peaks, both estimated forthe different elution gradients of their stationary phase (C18).

The proportion of the mobile phase used in Experiment 1 wasindicated by the official method [11] using a mixture of patterns ofboth aflatoxins, at the same concentration (0.01 mg mL�1), andresulted in peaks eluted in less than 5 min of run, which could beaffected by other components of the sample eluted in thebeginning of the run. To increase the retention time, the plots of1% acetic acid and methanol were increased, considering that thepolar character of these solvents can help a better retention ofaflatoxins in the column. In Experiment 2, standards withconcentrations of 0.5 mg mL�1 were used, resulting in a smallretention time. In the Experiments 4 and 5, the retention time ofAFLAB1 was too long, which promoted the generation of waste andthe reduction of the analytical efficiency, resulting in a 25 min run.Experiment 3 had the most suitable retention times for bothaflatoxins, with a decreased risk of the initial interference in a totalrunning time of 15 min. To confirm the efficiency of the separationunder these conditions, we tested it with the AFLAB1 standard(0.5 mg mL�1), the AFLAM1 standard (0.1 mg mL�1) and a mixture ofboth aflatoxins (0.05 mg mL�1), obtaining the same retention timein all cases. In this gradient also occurred the largest separationfactor, indicating that the peaks are presented as far as possiblefrom each other, ensuring the best selectivity among theperformed tests and a sufficiently small interval not to result inpeak broadening (Fig. 1).

The analytical curves for both aflatoxins, as well as theirparameters (Table 2), show that the procedure was suitable for thequantification of both aflatoxins. Their indicators meet therecommendations of analysis reliability, considering the recom-mendations of ANVISA [16] (a correlation coefficient of 0.99) and ofINMETRO [17] (values above 0.9). Furthermore, the linearity of thecurves showed a wide range of application, and the matrix curves

in) Retention factor(k)

Separation factor (a)

AFLAB1 AFLAM1 AFLAB1 AFLAM1� AFLAB1

4.4 0.36 0.76 2.118.0 1.08 2.20 2.0411.8 1.64 3.72 2.2720.0 3.72 7.00 1.8825.0 4.40 9.00 2.05

P.T. Scaglioni et al. / Analytica Chimica Acta 829 (2014) 68–74 71

have the advantage of carrying the interferents inherent to thesample extract.

The limits of detection and quantification for both theinstrument and the method were estimated by the signal-noiseratio, which establishes the minimum concentration at which theanalyte can be easily detected. Botura [18] determined the LOD andLOQ values for a modified method to determine AFLAM1 in goatmilk using UHPLC-FL as equal to 0.2 and 0.5 mg L�1, respectively.However, in order to validate a method to quantify AFLAM1 incow’s milk, Oliveira [19] determined LODm and LOQm values asequal to 0.003 and 0.007 mg L�1, respectively, using HPLC–MS/MS.The values of LODm and LOQm found in this study were amongthose cited by the authors and were considered satisfactory,

Fig. 1. Elution chromatograms of the AFLAB1 and AFLAM1 in UHPLC-FL using differe40:35:25; (2) 50:15:35; (3) 55:10:35; (4) 60:05:35; (5) 65:10:25.

considering that the LOQm presented a value below the legislatedfor milk AFLAM1 in Brazil (0.5 mg kg�1) [20].

According to the guideline of the European Commission [21],which makes specific performance recommendations for essayson mycotoxin recovery, when AFLAM1 is used at concentrationlevels greater than 0.05 mg L�1, the recommended recovery value is70–110%. Our results (Table 3) show recoveries that weremaintained within acceptable levels. Furthermore, when it isnecessary to recover mycotoxins with different polarities, theefficiency can be compromised due to a greater or lesser affinity ofeach mycotoxin with the eluent. Such is the case in therecuperation results, as the contamination levels increased. Inthis study, AFLAM1 showed higher recoveries in comparison to

nt mobile phase gradients composed of 1% acetic acid:acetonitrile:methanol. (1)

Table 2Analytical parameters evaluated in UHPLC-FL.

Analytical parameters AFLAB1 AFLAM1

Solvent curves y = 2,947,476x – 8261.30 y = 2,246,144 x – 6973.83Linearity (mg L�1) 4.5 a 108 4.5 a 270Coefficient of correlation 0.9987 0.9991Coefficient of determination 0.9974 0.9983LODi (mg L�1) 1.5 1.5LOQi (mg L�1) 4.5 4.5

Work curve y = 1,506,689x – 14,579.84 y = 693,877.2x – 5680.73Linearity (mg L�1) 0.5 a 162 0.25 a 162Coefficient of correlation 0.9918 0.9933Coefficient of determination 0.9836 0.9866LODm (mg L�1) 0.2 0.09LOQm (mg L�1) 0.5 0.25

AFLAB1-aflatoxin B1, AFLAM1-aflatoxin M1, LOD1-instrument detection limit, LOQ1-instrument quantitation limit, LODm-method detection limit, LOQm-method quantitationlimit.

Table 3Recovery of the AFLAB1 and AFLAM1 extracted from milk by the AOAC method.

Fortification (mg L�1) Recovery (%) (CV%)

AFLAB1 AFLAM1

0.25 – 75 (29)0.5 83 (3) 91 (15)3.0 78 (16) 114 (15)6.0 73 (14) 111 (33)8.0 76 (21) 94 (11)10.0 98 (12) 77 (19)Average 82 94

AFLAB1-alfatoxin B1, AFLAM1-aflatoxin M1, CV%-coefficient of variation.

72 P.T. Scaglioni et al. / Analytica Chimica Acta 829 (2014) 68–74

AFLAB1, a fact that was already expected, since the conditions ofthe official method were designed to quantify AFLAM1. However, itwas verified that AFLAB1 can also be quantified within reliableanalytical conditions.

Atanda et al. [22] evaluated the AFLAM1 contamination in milkand ice cream consumed in Nigeria using the same extractionmethod of this study [11], obtaining recovery values of 84.6, 85 and88% for contamination levels equal to 20, 25 and 30 mg L�1. Thus,according to the findings of the authors cited above, even withcontamination levels much larger than the legislated limits(0.5 mg kg�1) the method still presents a satisfactory recoveryfor AFLAM1.

Samples of raw, pasteurized, UHT, concentrated and powderedmilk were analyzed in order to determine their natural contami-nation by AFLAB1 and AFLAM1 (Table 4). Fig. 2 shows chromato-grams of pasteurized milk with and without natural contaminationsamples. We note that the retention time of both aflatoxins has notchanged in relation to the standard.

Other studies on the incidence of AFLAB1 in milk are scarce,considering that the legislation of most countries only imposes

Table 4Incidence of AFLAB1 and AFLAM1 in different samples of milk.

Milk sample AFLAB1

% incidence Average contamination (mg L�1)(CV%)

% above thelegislationa

Raw 0 (0/7) ND –

Pasteurized 41.7 (5/12) 1.476 (51) 100

UHT 13.3 (2/15) 0.690 (20) 100

Concentrated 0 (0/3) ND –

Powdered 0 (0/3) ND –

ND = not detected, CV% = variation coefficient.a Number of contaminated samples/number of analyzed samples.b Brazilian Legislation for AFLAM1 in fluid milk (0.5 mg kg�1).

maximum concentrations to AFLAM1 in milk and its derivates.However, more attention should be given to the study of AFLAB1,considering that in Brazil it has been found in limits higher thanthe ones legislated for its hydroxylated derivate (AFLAM1) inpasteurized and UHT milk. Gurbay et al. [10] verified thecontamination by both aflatoxins in 100% of the analyzed samplesof human milk originated from Turkey, finding levels from 0.095 to4.1 mg L�1 for AFLAB1 and 0.061 to 0.3 mg L�1 for AFLAM1. Resultssuch as the ones presented in this work show that the aflatoxin canbe excreted in its non hydroxilated form through the human andanimal organism. This fact suggests that the animal had a diet withhigh mycotoxin levels, which surpassed the metabolic capacity ofthe individual.

Another fact to be highlighted is that all of the samples with theincidence of AFLAM1 had this mycotoxin in concentrations abovethe legislated limit in Brazil. This enforces the necessity of a moreadequate control both in milk processing and in the handling ofanimal producers. Only the powdered milk samples did not showcontamination by the aflatoxins. This fact shows the need to studythe processing conditions that brought about this adequacy forconsumption. However, the collection period of the samples mayhave influenced this elevated incidence since, during winter,granaries are the predominant source of the cattle due to pasturescarcity. Therefore, due to its preparation conditions, this type offood presented higher levels of contamination by mycotoxins.

Other recent studies show that the contamination by AFLAM1

occurs in different samples from many parts of the world. Duarteet al. [23] verified the incidence of AFLAM1 in 40 samples of semi-skimmed pasteurized and UHT milk commercialized in Portugaland reported that 27.5% of the samples were contaminated with anaverage concentration of 0.0234 mg L�1. In addition, 5% of thesesamples had values above the limits of the European legislation(0.05 mg L�1). Rahimi et al. [24] verified the presence of AFLAM1 in

AFLAM1

%incidence

Average contamination (mg L�1)(CV%)

% above the legislation b

28.6 (2/7) 0.835 (9) 10058.3 (7/12) 0.884 (42) 10066.7 (10/15) 1.168 (27) 10066.7 (2/3) 1.718 (8) 100

0 (0/3) ND –

Fig. 2. Chromatograms of pasteurized milk without (1) and with (2) natural contamination by aflatoxins.

P.T. Scaglioni et al. / Analytica Chimica Acta 829 (2014) 68–74 73

79% of samples of raw cow milk, with an average concentration of0.06 mg L�1 (�57.4). Among these samples, 36% had contaminationlevels above the legislated limit of the European Community.

An acidic clotting of a pasteurized milk sample was performedto verify the distribution of aflatoxins and of the proteic contentbetween the serum and the clot (Table 5) of control and spiked(5 mg L�1) samples.

Both aflatoxins adhered to the milk clot, with a recovery of 102%for AFLAM1 in this fraction, when the control sample was coagulated.A recovery of 100 and 111% was obtained for AFLAB1 and AFLAM1,respectively, in the spiked sample clot. These results reinforce theimportance of establishing maximum permitted levels for dairyproducts obtained by milk coagulation. They also highlight thenecessity of a rigid control of raw materials, considering thatcurrently the Brazilian legislation establishes only a maximumtolerable limit for AFLAM1 in cheese (2.5 mg kg�1).

Deveci [25] quantified AFLAM1 during the production andstorage of canned white cheese from milk samples spiked with themycotoxin in two concentrations (1.5 and 3.5 mg kg�1). For both

Table 5Distribution of AFLAB1 and AFLAM1 between coagulated milk fractions.

Sample Humidity(%)(CV%)

Proteins (%(CV%)a

Control Milk 88.9 (0.5) 2.7 (7.4)

Serum 94.8 (0.1) 0.6 (17.0)

Clot 27.3 (1.9) 11.7 (15.0)

Spiked Milk – –

Serum – –

Clot – –

AFLAB1-aflatoxin B1, AFLAM1-aflatoxin M1, CV%-variation coefficient, n = 3.a Estimated in humid basis.

concentrations, after the cheese was produced, about 60% of theAFLAM1 remained in the clot cheese, while close to 30% wastransferred to the milk serum.

In a small cheese production from cheese artificially contami-nated with AFLAM1, López et al. [26] verified, that 60% of theAFLAM1 was found in the serum and 40% in the cheese. Hassanin[27] investigated the stability of the AFLAM1 during the productionand storage of yogurt, cheese and acidified milk. The authorconcludes that milk AFLAM1 is transmitted to its products, sincethe presence of AFLAM1 in cheese can be due to the fact that thetoxin binds to casein and, on the other hand, that part of the milkserum remains in the strands of the clot. Contrarily to thesestudies, in the present study none of the aflatoxins was detected inthe serum, indicating the total permanence of the mycotoxins inthe proteic fraction, when the acid condition was used forcoagulation. Thus, it was demonstrated that the contaminationbetween the serum and the clot depends on the coagulationmethod used in the fractioning. This is an important subject to bestudied in order to develop recommendable strategies of

) AFLAB1 (mg L�1) (CV%) AFLAM1 (mg L�1) (CV%)

ND 1.20 (11.1)ND NDND 1.22 (6.4)5.11 (9.5) 5.86 (5.7)ND ND5.42 (8.2) 6.48 (11.5)

74 P.T. Scaglioni et al. / Analytica Chimica Acta 829 (2014) 68–74

processing to minimize the risk of contamination by milk productconsumption.

4. Conclusion

The chromatographic method validated for mycotoxins resultedin a better proportion of the contents of the mobile phase, with aseparation factor of 2.3. In addition, the analytical parameters of themethod fulfilled the requirements of reliability, showing correlationcoefficients above 0.99, with LOQm values of 0.5 and 0.25 mg L�1 forAFLAB1 and AFLAM1, respectively. AFLAB1was detected in 42% of thepasteurized milk samples and 13% of the UHT milk samples, in levelsbetween 0.7 and 1.5 mg L�1. AFLAM1 was present in 29% of the rawmilk samples, 58% of pasteurized, 67% of UHT and 67% ofconcentrated, and all the samples that showed a natural contamina-tion by both aflatoxins were above the legislated limit in Brazil(0.5 mg L�1). Only the powdered milk samples did not showa naturalincidence of both aflatoxins, considering the quantification limits ofthe method. Acid milk coagulation promoted the accumulation ofcontaminants in the clot

References

[1] A. Kwiatkowski, A.P.F. Alves, Importância da detecção e do controle deaflatoxinas em alimentos, Saúde e Biologia 2 (2007) 44–53.

[2] E.E. Creppy, Update of survery, regulation and toxic effects of mycotoxins inEurope, Toxicology Letters 127 (2002) 19–28.

[3] V.M. Scussel, Micotoxinas em Alimentos, Editora Insular, Florianópolis, 1998.[4] J. Zhao, R.B. Shirley, J.D. Dibner, F. Uraizee, M. Officer, M. Kitchell, M. Vazquez-

Anon, C.D. Knight, Comparison of hydrated sodium calcium aluminosilicateand yeast cell wall on counteracting aflatoxicosis in broiler chicks, PoultryScience 89 (2010) 2147–2156.

[5] International Agency of Research on Cancer- IARC, Monographs on theEvaluation of Carcinogenic Risks to Humans, WHO, Lyon, 2002 v. 82.

[6] H.A. Partanen, H.S. El-Nezami, J.M. Leppänen, P.K. Myllynen, H.J. Woodhouse,K.H. Vähäkangas, Aflatoxin B1 transfer and metabolism in human placenta,Toxicological Sciences: an Official Journal of the Society of Toxicology 113(2010) 216–225.

[7] F. Caloni, A. Stammati, G. Friggè, I.D. Angelus, Aflatoxin, M1 absorption andcytotoxicity on human intestinal in vitro model, Toxicon 47 (2006) 409–415.

[8] OPS, Organización Panamericana de la Salud, Micotoxinas Critérios de SaludAmbiental, OPS, Washington, 1983.

[9] R.D. Stubblefield, H.P. Van-Egmond, Chromotographic methods of analysis foraflatoxin M1, in: VAN-EGMONDHP (Ed.), Mycotoxins in Dairy Products,Elsevier Applied Science, London, 1989.

[10] A. Gurbay, A. Sabuncuoglu, G. Girgin, G. Sahin, S. Yigit, M. Yurdakok, G.Tekinalp, Exposure of newborns to aflatoxina M1 and B1 from mothers’ breastmilk in Ankara, Turkey, Food and Chemical Toxicology 48 (2010) 314–319.

[11] W. Horwitz, Official Methods of Analysis International, 17th ed., Association ofOfficial Analytical Chemists AOAC, USA, 2000.

[12] C.H. Collins, G.L. Braga, P.S. Bonato, Fundamentos de Cromatografia, Editora daUnicamp, Campinas, 2006.

[13] L.V. Soares, Curso Básico de Instrumentação Para Analistas de Alimentos eFármacos, Editora Manole Ltda, Baueri-SP, 2006.

[14] M. Ribani, C.B.G. Bottoli, C.H. Collins, C.S.F. Jardim, L.F.C. Melo, Validação emmétodos cromatográficos e eletroforéticos, Química Nova 27 (2004) 771–780.

[15] A.C.R. Krolow, M.E.R. Ribeiro, BRASIL, MAPA–Ministério da Agricultura,Pecuária e Abastecimento. Obtenção de leite com qualidade e elaboraçãode derivados, Embrapa Clima Temperado 154 (2006) Documentos, ISSN 1806-9193.

[16] BRASIL, ANVISA. Agência Nacional de Vigilância Sanitária, Resolução RE n� 899,de 29/05/2003. Disponível em: www.anvisa.gov.brwww.anvisa.gov.br. Acessoem: dez; 2013.

[17] Brasil, INMETRO. Instituto nacional de metrologia, normalização e qualidadeindustrial, orientações sobre validação de métodos de ensaios químicos. DOQ-CGCRE -008; 2003.

[18] M.B. Botura, Otimização de Métodos Analíticos Para Determinação deAflatoxinas em Rações e Leite de Cabra, e sua Ocorrência no Estado da Bahia.Dissertação (Mestrado em Medicina Veterinária Tropical), Escola de MedicinaVeterinária da Universidade Federal da Bahia, Bahia, 2005, pp. 2005.

[19] M.S. Oliveira, Validação de Metodologia Analítica Para Análise de AflatoxinaM1 e sua Ocorrência em leite Bovino Comercializado no sul do Brasil.Dissertação (Mestrado em Ciência e Tecnologia de Alimentos), UniversidadeFederal de Santa Maria, Santa Maria, 2010.

[20] BRASIL. ANVISA-Agência Nacional de Vigilância Sanitária, Resolução RDC n� 7,de 18/02/2011. Disponível em: www.anvisa.gov.brwww.anvisa.gov.br. Acessoem: dez; 2013.

[21] COMISSÃO EUROPÉIA - Regulamento (CE) n� 401/2006 da Comissão, de 23 deFevereiro de 2006. Métodos de amostragem e de análise para o controle oficialdos teores de micotoxinas nos gêneros alimentícios. Disponível em http://eur-lex.europa.eu/pt/index.htm Acesso: dez; 2013.

[22] O. Atanda, A. Oguntubo, O. Adejumo, J. Ikeorah, I. Akpan, Aflatoxin M1

contamination of milk and ice cream in Abeokuta and Odeda local govern-ments of Ogun state, Nigeria, Chemosphere 68 (2007) 1455–1458.

[23] S.C. Duarte, A.M. Almeida, A.S. Teixeira, A.L. Pereira, A.C. Falcão, A. Pena, C.M.Lino, Aflatoxin M1 in marketed milk in Portugal: assessment of human andanimal exposure, Food Control 30 (2013) 411–417.

[24] E. Rahimi, M. Bonyadian, M. Rafei, H.R. Kazemeini, Occurrence of aflatoxina M1

in raw milk of five dairy species in Ahvaz, Iran, Food and Chemical Toxicology48 (2010) 129–131.

[25] O. Deveci, Changes in the concentration of aflatoxin M1 during manufactureand storage of white pickled cheese, Food Control 18 (2007) 1103–1107.

[26] C. López, L. Ramos, S. Ramadán, L. Bulacio, J. Perez, Distribution of aflatoxin M1

in cheese obtained from milk artificially contaminated, International Journalof Food Microbiology 64 (2001) 211–215.

[27] N.I. Hassanin, Stability of aflatoxin M1 during manufacture and storage ofyoghurt, yoghurt-cheese and acidified milk, Journal of the Science of Food andAgriculture 65 (1994) 31–34.