Aflatoxin M1 Concentration in Various Dairy Products: Evidence for Biologically Reduced Amount of AFM1 in Yoghurt.

BACKGROUND: Aflatoxin M1 (AFM1), a carcinogenic substance is found in milk and dairy products. The effect of season and type of dairy products on AFMi level in northern Iran was investigated in this study.
METHODS: Three hundred samples (each season 75 samples) including raw and pasteurized milk, yoghurt, cheese, and cream samples were collected from three distinct milk producing farms. The samples were subjected to chemical and solid phase extractions and were analyzed by using HPLC technique. Recovery percentages, limit of detection and limit of quantification values were determined.
RESULTS: Seventy percent and 98% were the minimum and maximum recoveries for cheese and raw milk, respectively and 0.021 and 0.063 ppb were the limit of detection and limit of quantification values for AFM1. We found that in autumn and winter the highest level (0.121 ppb) of AFM1 in cheese and cream samples and failed to detect any AFM1 in spring samples. Interestingly, our data showed that the yoghurt samples had the lowest level of AFM1 in all seasons.
CONCLUSION: There are significant differences between the AFM1 levels in dairy products in various seasons and also various types of products, suggesting spring and summer yoghurt samples as the safest products from AFM1 level point of view.

Introduction

Aflatoxins (AFs) are known mycotoxins that threatening humans and animals health with causing disorders such as hepatocellular carcinoma (HCC). In addition to causing the certain cancers, AFs suppress the immune system and potentiate the other hepatotoxic compounds toxicity. The carcinogenicity and mutagenicity of AFs are mainly related to the conversion of AFB1 into AFB1 8, 9-epoxide, which is able to bind the proteins and nucleic acids (i). Aflatoxins-induced reproductive disorders including disruption of testicular structure, secretion of androgens and generation of meiotic micronucleate giant spermatocytes have been previously reported (2).

Aflatoxin M1 (AFM1) is the hydroxylated metabolite of aflatoxin B1, which is found in milk and dairy products. Its weaker carcinogenicity than AFB1 has been reported in laboratory animals. Besides, previous studies showed that there are potential hazardous human exposure to AFM1 through the consumption of milk and milk products (3). Therefore, several countries have legislated action levels of AFM1 in milk and dairy products (4). As AFM1 is heat stable compound, there-fore its level in contaminated foods remains unaffected by heating methods including pasteurization and sterilization (5). There are reports indicating that adding sequestrating agents to the AFB1-contaminated feeds, resulted in a significant reduction of AFM1 level in milk (6). However, an effective sequester which be able to lower the AFM1 level in milk and dairy products without negative impact on nutritional status has not been introduced yet. In addition of the diet level of AFB1, ecological conditions, animal species and farm management strategy also affect the AFM1 concentration in milk and dairy products (7). Tajkarami and his co-workers (2007) reported significant differences between 5 different regions of Iran, regarding the AFM1 concentration in milk (8).

There are controversial and limited internationally published data about the effect of season and type of diary products on the occurrence of AFM1; therefore we aimed to investigate the effect of various seasons and different type of dairy products (raw and pasteurized milk, yoghurt, cheese, and cream) on AFM1 level in Urmia, located in northern Iran.

Materials & Methods

Chemicals and standard solutions

The reference standard of AFM1 (from Aspergillus flavus, 10μg, A 6428) was purchased from Sigma— Aldrich (Germany). Acetonitrile, methanol, and water were of HPLC grade and were purchased from Baker (Deventer, the Netherlands).

Samples

Milk samples in each season (15 raw milk samples from three distinct milk producing farms) were collected in 2012 according to the Iranian national standard milk sampling method (INS No.419) and after taking raw milk (500 ml) the rest of samples were subjected to be pasteurized (30 min in 63 °C) and /or converted to the other dairy products including yoghurt, cheese, and cream. The samples were stored at 4 °C and analyzed within the 5 days.

Chemical and solid phase extraction

A 50 g milk sample or dairy products homogenate after addition of 20 ml deionized water was centrifuged at 10 °C in a polypropylene tube at 1500 × g for 15 min. Thereafter, once for milk samples and twice for cheese, yoghurt and cream samples defattation processes by using n-butanol was conducted. After defattation, 50 ml pure methanol was added to samples and was shaken vigorously for 10 min, followed by 15 min centrifugation at 1100 × g. The upper layers of the samples after addition of the same volume deionized water were subjected to solid phase extraction by using preconditioned immunoaffinity column (Aflastar M1, Romer labs, Diagnostic GmbH, 3430, Tulln, Asturia). The cartridges were washed with 10 mL water, dried by vacuum for 15 s and then eluted with 3 mL of methanol. The eluate was evaporated under a gentle stream of nitrogen. The residue was re-dissolved in 200 mL of mobile phase.

Method evaluation

To evaluate recoveries, the uncontaminated samples were spiked with 0.5, 0.8 and 1 ppb of standard AFM1. The recovery test was performed five times at the concentration of 0.8 ppb. The limit of quantification level was determined by the addition of decreasing concentrations of the AFM1 standard to uncontaminated samples. Standard curve for AFM1, were constructed with 5 points (from 0.5— 10 ppb).

HPLC analyses

To analyze the prepared samples, we used a Waters Breeze 1525 HPLC system equipped with Waters 1525 Binary pump, on-line degasser, Waters 717 Plus Auto sampler and Waters 2475 fluorescence detector. To detect the AFM1 by fluorescent detector, excitation was set at 360 nm and emission at 440 nm. The analytical column was an ODS 5 μm, 4.6 mm × 250 mm C18 column (Nucleodur) and the mobile phase consisted of Acetonitrile: Methanol: H2O (20:20:60, V/V/V), which was pumped at a flow rate of 1 ml/min. Waters Breeze software also was used to analyze the data.

Statistical analyses

The statistical methods used in this study were based on normal confidence intervals and analysis of variance (ANOVA).

Results

The constructed standard curve demonstrated linearity within the concentration range and the calculated coefficients of correlation was found 0.9998. We found the retention time of 9.6 min for AFM1 in standard and test samples.

To validate the method, which was applied to analyze the dairy products in this study, we determined the recovery percentages for AFM1 at three concentrations in each single product. The obtained data indicated that the highest (98.5 %) and lowest (69 %) recoveries belong to raw milk and cheese samples, respectively. The recovery percentages for various dairy products are presented in Table 1. Moreover, the limit of detection (LOD) and limit of quantification (LOQ) levels of 0.021 and 0.063 ppb were estimated.

Table 1

Percentage of recovery for AFM1 in various dairy products

AFM1	Raw Milk	Pasteurized Milk	Yoghurt	Cream	Primary cheese
0.5 (ppb)	98.5 ± 3.6	97 ± 14	88 ± 7	80 ± 2.3	74 ± 3.4
0.8	95 ± 7.2	91 ± 3.9	81 ± 4.1	72 ± 4.5	75 ± 7.1
1.0	88.1 ± 4.4	85.9 ± 3.2	85 ± 3.3	87 ± 7.2	69 ± 1.8

We found that there are significant differences (P<0.05) between 3 distinct milk producing farms in terms of AFM1 level in raw milk at summer time.

Figure 1 shows that in all three seasons the samples which were collected from farm II were contaminated with high level of AFM1 when compared to two other farms. Additionally a season-dependent increase of AFM1 level in raw milk and in all three farms was obtained.

Fig. 1

Aflatoxin M1 level in raw milks from three various farms at different seasons

We found the highest concentration of AFM1 in the samples, which collected from farm B at autumn season and followed by samples, which analyzed during the winter season. None of the collected samples at spring time was found to be contaminated with AFM1.

Comparing the three farms samples revealed that 2 farms from 3 at autumn and all three farms at wintertime showed AFM1 level exceeded from standard concentration of 0.05 ppb. There are no significant differences (P>0.05) between raw and pasteurized milk samples. We found that AFM1 concentration in the corresponding cheese and cream samples were slightly but significantly higher than that in raw and pasteurized milk samples (Table 2). Interestingly, analyses of yoghurt samples revealed that the AFM1 level significantly (P<0.05) was reduced when compared to the original raw milk and also to the other dairy products which made from the corresponding samples.

Table 2

AFM1 Levels (ppb) in milk and dairy products in Urmia City

Season	Farm	Raw Milk	Pasteurized Milk	Yogurt	Cream	Primary Cheese
	I	N.D.	N.D.	N.D.	N.D.	N.D.
Spring	II	N.D.	N.D.	N.D.	N.D.	N.D.
	III	N.D.	N.D.	N.D.	N.D.	N.D.
	I	0.035 ± 0.007	0.035 ± 0.003	0.029 ± 0.002	0.032 ± 0.007	0.031 ± 0.006
Summer	II	0.082 ± 0.005	0.13 ± 0.04	0.029 ± 0.003	0.09 ± 0.008	0.098 ± 0.004
	III	0.029 ± 0.004	0.021 ± 0.005	0.021 ± 0.006	0.031 ± 0.005	0.031 ± 0.005
	I	0.108 ± 0.03	0.085 ± 0.002	0.032 ± 0.008	0.116 ± 0.007	0.121 ± 0.09
Autumn	II	0.111 ± 0.009	0.105 ± 0.04	0.033 ± 0.007	0.121 ± 0.06	0.112 ± 0.05
	III	0.035 ± 0.004	0.036 ± 0.007	0.032 ± 0.008	0.042 ± 0.005	0.047 ± 0.006
	I	0.099 ± 0.008	0.097 ± 0.02	0.033 ± 0.003	0.110 ± 0.07	0.113 ± 0.02
Winter	II	0.106 ± 0.06	0.102 ± 0.03	0.033 ± 0.007	0.113 ± 0.03	0.114 ± 0.04
	III	0.100 ± 0.04	0.098 ± 0.01	0.033 ± 0.005	0.113 ± 0.04	0.119 ± 0.07

Discussion

This study showed that AFM1 level in dairy products is affected by season and product type. The data indicated that the highest level of AFM1 was detected in the cheese and cream samples which prepared at autumn and winter seasons. Moreover, the lowest concentration of AFM1 was found in yoghurt samples in all seasons.

To evaluate the accuracy of the analytical method, we estimated the LOD, LOQ and recovery percentages for each single product. The values for LOD and LOQ were not different for dairy products, however, our results showed various recovery percentages for different products, suggesting the important role of food matrix in recovery of certain compound. The effect of various matrixes in the recovery of certain compounds has been well documented (9).

AFM1 is the hydroxylated metabolite of AFB1, which is produced by cytochrome P450 system in the liver and released mainly into milk and urine and in less extent into the feces. It has been reported that AFM1 level in the milk has a direct relationship with the concentration of AFB1 in consumed contaminated feed (10). Previous studies reported that consuming the contaminated feedstuffs with more than 70 mg AFB1 will resulted in the release of AFM1 at the higher concentration than standard level (0.05 ppb) in milk (11). There are different regulations in various countries and in Iran. According to decision (5925 NSO) of the national standard organization, the AFM1 level should not exceeded from 0.05 ppb in milk samples (12). Therefore, the first message of the current study might be warning about the out of range presence of AFM1 in milk and dairy products in northwest of Iran at autumn and winter seasons. There are many factors that could affect the AFM1 level in the milk and dairy products including: the diet source of lactating animals, ecological and economical factors and management policy.

Our findings in the current study indicate that there is no detectable AFM1 in the milk and dairy products during the springtime. This finding may be explained with two facts: either due to using fresh forage during the springtime, there is no AFB1 contamination, or the level of contamination is under the sensitivity of used method (13). It is well documented that the feedstuffs such as concentrate and silage due to having the favorite condition of fungi growth including high humidity and temperature and improper storage condition may result in the mycotoxins production (14–16). Aflatoxin B1 has been found in various feed including cottonseed, barley, soy bran, pellet wheat, corn silage and sorghum silage in different countries (17). Exposure of lactating cattle to mycotoxins occurs mainly through the consumption of contaminated feed and may cause mycotoxicosis. In case of exposure to AFB1, dairy cattle not only may cause aflatoxicosis, but it may lead to the production of its metabolic product in milk and dairy products that consequently affect the public health (18).

We found no significant differences between the raw and pasteurized milk samples in terms of AFM1 concentration and this finding is supported by previous reports indicating that the pasteurization processes do not affect the AFM1 level in milk (19). One of the remarkable findings of the current study is that we detected the lowest range of the AFM1 concentrations in yoghurt samples, corresponding to those raw milk samples, which the AFM1 level at autumn and winter seasons exceeded from standard level. There could be several explanations for this finding including: the presence of various lactic acid bacteria in yoghurt samples which are able to destroy the AFM1 (20). Up to 34% reduction of AFM1 concentration in yoghurt samples compared to original raw milk samples has been previously reported. Moreover, it has been also demonstrated that the different strains of lactic acid bacteria do have various capability in the degrading of AFM1, where Lactobacillus acidophilus and L. rhamnosus were able to reduce AFM1 level by 18.5 and 49.6 %, respectively. Another reason for the AFM1 reduction in yoghurt samples may be related to affinity of the present bacteria in binding to AFM1 molecules, which by reduction of pH level would be able to degrade the bound molecules (21). Previous studies which have been performed in four Iranian large cities of Tehran, Esfahan, Shiraz and Yazd during winter and summer 2009 indicated that the AFM1 concentration in yoghurt samples was significantly lower than in pasteurized milk samples (22). We also found that the AFM1 concentration in cream and cheese samples were slightly higher than that in the corresponding milk samples. The high fat content of cream and lipophilic property of AFM1 and high affinity of AFM1 molecules to casein may explain the higher concentration of AFM1 in cream and cheese samples, respectively (23).

Conclusion

Our data indicate that both season and type of dairy products could affect the AFM1 level as the highest concentrations were found in autumn and winter seasons in cheese and cream samples, while the lowest level of AFM1 was detected in yoghurt samples. These findings may help the dairy products producers to use the properly prepared/stored feedstuff during the cold seasons for milk producing animals and also might lead the food authorities to consider the knowledge-based use of probiotics to produce the safe foods.

Ethical considerations

Ethical issues (Including plagiarism, Informed Consent, misconduct, data fabrication and/or falsification, double publication and/or submission, redundancy, etc.) have been completely observed by the authors.