Probiotic Strategies for Detoxification of AFM1 in Skim Milk Using Bifidobacterium Lactis and Streptococcus Thermophiles

1. Introduction

Aflatoxins (AFs), as some of the most important mycotoxins, are natural by-products that cause serious food quality and safety problems worldwide. AFs are produced by the fungal species Aspergillus, particularly Aspergillus flavus, Aspergillus parasiticus and Aspergillus nomius ( 1 , 2 ). They are commonly found in foods and feeds such as cereals, oilseeds, spices, and nuts, especially in tropical regions. Among the 20 known AFs, AFB1, AFB2, AFG1, and AFG2 are the primary ones. The highest toxicity is associated with AFB1, produced by A. flavus, which is often found in the feed of dairy ruminants. Aflatoxin M1 (AFM1) is a hydroxylated metabolite of AFB1 formed in the liver and excreted in the milk of animals or humans who have consumed an aflatoxin-contaminated diet. Various factors, including the species type, diet, and individual factors such as lactation period and milk production yield, influence the conversion of AFB1 to AFM1 ( 3 , 4 ). Milk and dairy products with high consumption rate across all ages, especially children, serve as a vehicle for contaminants that pose serious risk to human health. Aflatoxins are known for their heat-resistant properties, which means that even the pasteurization process of milk doesnot effectively inactivate AFM1 contamination. ( 5 ).

Despite affecting numerous important agricultural products and increasing economic costs, AFs are carcinogens and hepatotoxic agents. Due to the high incidence of AFM1 in milk, several countries have implemented strict control policies to reduce the risk of aflatoxin exposure ( 6 ). Efforts to detoxify contaminated products have been ongoing for decades ( 7 ). Several strategies have been developed to prevent the growth of mycotoxigenic fungi and to eliminate or inactivate AFM1 in milk. However, these strategies have limitations, such as reduced nutritional value, low organoleptic qualities, low efficiency and safety concerns, and high costs. Recently, biological methods have gained attention as alternative strategies to chemical and physical treatments ( 8 , 9 ). Certain lactic acid-producing microorganisms have recently attracted interest due to their ability to detoxify AFM1 in contaminated milk ( 10 , 11 ).

Aflatoxin detection and quantification of aflatoxins are critical for safety concerns. Various methods are available, while enzyme-linked immune-sorbent assay (ELISA) and High-performance liquid chromatography (HPLC) are the most widely used. Given food safety, public health risks, and economic factors related to the presence of aflatoxin in food and animal diets such as silage, combining beneficial microorganisms is probably the most effective strategy for achieving the optimal effect. Bifidobacteria are abundant in the normal gut flora, used in dairy products as probiotics, and have shown potential in aflatoxin detoxification. Some species of Bifidobacteria, such as Bifidobacterium bifidum and Bifidobacterium lactis, have been reported to possess aflatoxin detoxification properties. Additionally, S. thermophilus, as a major dairy starter, exhibits antimicrobial, antioxidant, and antitoxin effects ( 12 ).

Even though there are some reports on AFB1 detoxification by different microbes, the effects of Streptococcus thermophilus and Bifidobacterium lactis have not been compared. The objective of this study was to identify the most effective method for detecting aflatoxin M1 contamination in skim milk using HPLC. Given thatphysicochemical parameters, such as temperature and the concentrations of AFM1 and probiotics, could affect the detoxification of AFM1 ( 13 , 14 ), we initially investigated the detoxification effects of bacteria at two levels of bacteria concentration (8 and 10 logs CFU/mL) and incubation temperature (4 and 42ºC) ), along with three levels of AFM1 concentrations (0.1, 0.25 and 0.5 μg /mL) during storage at the different time point (30, 60, 120 min and 24h), using HPLC. In the second step, we chose the best strategies for each bacterium and compared the individual probiotics to determine the most effective strategy for detoxifying AFM1.

2. Materials and Methods

2.1.Preparation of Bacteria

The bacterial strains used in this study, Bifidobacterium animalis subsp. lactis Bb12 and Streptococcus thermophilus PTCC7788, were purchased from Chr. Hansen (Denmark). To prepare the cell suspension, 1g of lyophilized bacteria was cultivated in 100 mL of De Man, Rogosa, and Sharpe (MRS) broth and M17 broth and incubated at 37°C for 24 hours. The culture media was then centrifuged at 3500 ×g at 4°C for 15 minutes to harvest the cells. The turbidity of suspension was standardized to match that of a 10 McFarland standard, which corresponds to approximately 3×10¹⁰ CFU/ mL. The cell suspension was counted using a hemocytometer (Neubauer counting chamber) to obtain two final concentrations of 10 and 8 log CFU/ mL ( 15 ).

2.2.Preparation of AFM1

AFM1 powder (6795-23-9, Aokin, Germany) was diluted in acetonitrile to achieve a concentration of 10 µg/mL. The AFM1 standard solution was further diluted in acetonitrile to obtain a concentration of 1 µg/mL and stored at 4°C until use ( 16 ).

2.3.Contamination and Inoculation of Skim Milk

Skimmed milk was prepared by mixing skim milk powder (115363, Merck, Germany) with distilled water in a ratio of 1:10 (w/v). The skim milk samples were agitated for 5 minutes and then centrifuged at 3500 ×g at 4ºC for 10 minutes to separate the cream. After centrifugation, the upper cream layer was completely removed from the skim milk. Subsequently, samples were spiked with three different concentrations of AFM1 working solutions (0.1, 0.25, and 0.5 μg /mL) at 42ºC . After milk contamination, 9 mL samples,both separately and in combination, were inoculated with bacteria at two concentrations (10 and 8 log CFU /mL) and incubated at two temperatures (4 and 42ºC) for different time points (30, 60,120 min, and 24 h). The skim milk samples with AFM1 and bacteria were the positive control, while samples without bacteria acted as the negative control. After due time, the samples were centrifuged at 2750 ×g for 5 minutes to harvest the supernatant, which was then to evaluate the residual aflatoxin ( 17 ). Each treatment sample was named according to Table 1.

2.4.AFM1 Determination By HPLC Method

The detection of AFM1 residues in skim milk was evaluated using the HPLC method, as described by Sarlak et al.( 18 ), with minor modifications. An HPLC system (Waters Alliance 2695 Separations Module) equipped with a column (Grom Sil C18 ODS-5ST, 5μm×250×4.6mm) and a fluorescence detector 2475 were used in this study. The excitation and emission wavelengths were set at 365 and 465 nm, respectively. The mobile phase consisted of water, methanol, and acetonitrile in a ratio of 60:20:20. The flow rate was set at 1 mL/min and the injection volume was 150µl.

The percentage of AFM removed by bacteria was calculated as follows.

%AFM1 removal=100 * [1 - (peak area of sample)/ peak area of positive control)].

2.5.Statistical Analysis

The data in the study were described using frequency (percent) for qualitative variables and Mean±SD for quantitative variables. The normality of the distribution of quantitative variables was assessed using the Shapiro–Wilk test. To compare the means of normal and non-normal quantitative variables between the two treatment groups (B. lactis and S. thermophilus), the independent T-test and Mann-Whitney U test were applied, respectively. The Kruskal-Wallis test was used to compare the mean of quantitative variables between the three toxin concentrations (0.1, 0.25, and 0.5 μg /mL). If there was a significant difference between the three toxin concentrations, the Dunn-Bonferroni test was utilized to identify which toxin concentration pairs caused these differences. In addition, the effects of the treatment type, temperature, toxin concentration, and bacterial concentration on AFM1 removal percent were evaluated using the generalized estimating equations (GEE) model. All statistical analyses were performed using SPSS version 20 , with a significant level of 0.05.

3. Results

We examined the percentage of AFM1 removal by two bacteria (B. lactis and S. thermophilus) at two levels of bacteria concentration (8 and 10 logs CFU/mL), two incubation temperatures (4 and 42ºC) ,and three levels of AFM1 concentrations (0.1, 0.25 and 0.5 μg /mL) during storage at the different time points (30, 60, 120 and 1440 min) using HPLC.

3.1. Effect of Bacterial Concentration

Our findings showed a significant difference between the two treatment types at both bacteria concentrations  after 30 minutes (P <0.05). At 60, 120, and 1440 minutes, the mean of AFM1 removal was significantly higher in the B. lactis treatment compared to S. thermophiles at the 10 log CFU/mL concentration (P <0.05). No significant difference was observed in the mean of AFM1 removal percentage between the two levels of bacteria concentrations within treatment types (P>0.05) (Table 2). Figures 1-A and 1-B show the trends of AFM1 removal percentage over time in both treatment types by bacterial concentration (BC). Results from the Friedman test revealed that the mean AFM1 removal rate at each treatment type increased significantly over time in both bacterial concentrations (P <0.05) (Table-5).

Figure 1. Change trends of AFM1 removal percent in both treatment types in terms of bacterial concentration (BC).

3.2. Effect of incubation temperature

Table 3 shows the results of evaluating the percentage of AFM1 removal between treatment types at different time points and temperatures. We found that the mean percentage of AFM1 removal showed a significant difference between the two treatment types at 30, 60, and 120 minutes (P <0.05) (Table-5).

At 1440 minutes, the mean percentage of AFM1 removal in the S. thermophiles group was significantly lower compared to the B. lactis group,but only at 4ºC (P <0.05). In the S. thermophilus group, a significant difference was observed in the mean percentage of AFM1 removal between the two temperatures at 30, 60, and 120 minutes (P<0.05). However, in the B. lactis group, no significant difference was observed in the mean percentage of AFM1 removal between the two temperatures (P>0.05). As shown in Figures 2-A and 2-B, the mean percentage of AFM1 removal in both treatment types increased over time at both temperatures. Results from the Friedman test indicated that the mean AFM1 removal rate at each treatment type increased significantly over time at both temperatures (P <0.05). 

Figure 2. Change trends of AFM1 removal percent in both treatment types in terms of temperature (T).

3.3. Effect of treatments on AFM1 concentration

Table 4 presents a comparison of the AFM1 removal percentage between two treatment types at different time points across three toxin concentrations (0.1, 0.25, and 0.5 μg/mL). Our findings revealed a significant difference between the two treatment types at 0.25 and 0.50 μg/mL toxin concentrations at 30 and 60 minutes (P <0.05). At 120 and 1440 minutes, we observed that the mean AFM1 removal rate in the B. lactis group was significantly higher than that in the S. thermophiles group at 0.50 μg/mL toxin concentration (P<0.05). In the B. lactis group, a significant difference was observed in the mean AFM1 removal percentage among the three toxin concentrations (0.1, 0.25, and 0.5 μg/mL) at 30, 60, and 120 minutes, based on the results of the Kruskal–Wallis test (P <0.05). To understand which mean differences between the two toxin concentrations had caused significant differences among the three toxin concentrations, we used the Dunn-Bonferroni post-hoc test. According to Dunn-Bonferroni post-hoc test results in the B. lactis group, there was a statistically significant difference in mean AFM1 removal percentage between 0.1 and 0.5 μg/mL toxin concentrations at times 30, 60, and 120 minutes (P <0.05). In contrast, no significant difference was observed in the mean AFM1 removal percentage between the three toxin concentrations (0.1, 0.25, and 0.5 μg/mL) in the S. thermophiles group (P>0.05). Figures 1-A, 1-B, and 1-C show trends of AFM1 removal percentage over time in both treatment types by toxin concentrations (TC). The Friedman test results showed that the mean AFM1 removal rate at each treatment type increased significantly over time across all three toxin concentrations (P <0.05).

The generalized estimating equations (GEE) model was used to investigate the effects of treatment type, temperature, toxin concentration, and bacterial concentration on AFM1 removal percentage. Results from the GEE model showed that there was a statistically significant difference in the mean AFM1 removal percentage between two treatment types, two temperatures, two bacterial concentrations, and three toxin concentrations at baseline or first measurement (30 minutes) (P<0.05) (Figure 3).

Figure 3. Change trends of AFM1 removal percent in both treatment types in terms of Toxin concentration (TC).

After adjusting for the effects of other variables in the model, we found that the mean AFM1 removal percentage in the B. lactis group was 14.90 units higher than in the S. thermophiles group at baseline or the first measurement (30 minutes). Additionally, the mean AFM1 removal percentage at 4ºC was 9.84 units lower than at 42 ºC at the first measurement (30 minutes). When comparing bacterial concentrations, the mean AFM1 removal percentage at 8 log CFU/mL was 3.04 units lower than at 10 log CFU/mL at the first measurement. Regarding toxin concentration, the mean AFM1 removal percentage at baseline (30 minutes) was 13.31 and 9.32 units lower in 0.10 μg/mL and 0.25 μg/mL, respectively compared to 0.5 μg/mL. However, other variables, including time and interaction of time with treatment type, temperature, toxin concentration, and bacterial concentration, had no significant effect on the rate of AFM1 removal (P>0.05) (Table 4).

4. Discussion

Milk and dairy products contaminated with AFM1 have become major food safety concerns. Thus, it is important to implement strategies for reduction and to monitor the presence of AFB1 in feedstuffs. The present study investigated the ability of S. thermophilus and Bifidobacterium animalis (subspecies lactis) as probiotic bacteria, to detoxify AFM1 in contaminated milk, considering factors such as bacterial population, incubation temperature, and toxin concentration. We found that AFM1 detoxification from milk was time-dependent, with significant AFM1 removal occurring at an earlier time of exposure. Other studies confirm that the removal of AFM1 is a rapid process that depends on the bacterial strain ( 19 ). We found that B. lactis and S. thermophilus had significant ability to remove AFM1 at 120 minutes and 60 minutes, respectively.

Bacterial concentration was one of the factors influencing AFM1 reduction in skim milk for both B. lactis and S. thermophiles.  Similar results were reported by Sarlak et al. ( 18 ), who investigated the removal of AFM1 from fermented milk drinks (doogh) by probiotic strains. They showed the percentage of AFM1 removal was higher at 10 log CFU/mL of Lactobacillus. acidophilus compared to 7 log CFU/ mL (99 vs 95%) over 28 days. Additionally, 7 log CFU/mL of L. acidophilus exhibited more AFM1 binding capacity than 7 log CFU /mL of B. lactis (75%).

We found that the high concentration of B.lactis (10 logs CFU/ mL) led to 67.65 % AFM1 removal after 24 hours in milk. We also found that the lower concentration level of S.thermophiles (8 logs CFU/ mL) could remove more AFMI, suggesting that the cell wall structure is more related to the type of microorganisms involved in removing toxins.

Two enzymatic and absorption mechanisms have been proposed to reduce aflatoxin by microorganism strains. Since it has been reported that viable and non-viable bacteria can bind AF, the surface of the cell wall is the dominant mechanism of toxin elimination. The removal of AFM1 in contaminated skim milk with 0.5 ng/mL of AFM1 inoculated with 1010 cells /mL of heat-killed strains, including Bifidobacterium lactis FLORA-FITBI07 and a pool of LAB, was approximately 12% at 60 minutes at 42ºC ( 10 ). The stability of bacterial-AFM1 binding was evaluated using repeated washing by Panwar et al. ( 20 ). They highlighted of the bacterial cell walls due to the release of AFM1 after washing and suggested mechanisms of action in aflatoxin detoxification likely involving noncovalent binding rather than metabolic inactivation.

Our result indicated that the highest percentage of toxin removal for both bacterial types related to an incubation temperature of 42ºC compared to 4ºC. It may be attributed to to the heat treatment affecting components of the cell wall, such as polysaccharides and peptidoglycans, which can disrupt the the cell membrane and enhance aflatoxin binding to cell wall components and plasmatic membrane.

We also found that the highest affinity for B. lactis binding to AFM1 was observed when the toxin concentration was high (0.5 μg /mL). Our findings agree with previous studies showing that toxin binding increases with higher toxin concentration ( 13 , 21 , 22 ). For example, Karazhiyan et al. showed a similar upward trend in toxin removal by yeasts with increasing toxin concentrations from 100 to 750 pg /mL ( 21 ).

The level of AFM1 binding by S. thermophilus in PBS and yogurt spiked with 50 μg /L and incubated at 42 ºC increased over time and was approximately 35% and 38% after 6 hours, respectively. The higher removal rate in yogurt may be associated with the better binding ability of AFM1 to casein molecules ( 23 ). Such data were in good correlation with our finding, which indicated that the highest removal AFM1 for S. thermophiles in milk was related to 0.1 and 0.25 μg /mL (24 and 22.8% respectively) after 60 minutes at 42°C and reaching 45% at 24 hours with 0.5 μg/mL. The beneficial effect of lactic acid fermentation on the reduction of AFM1 level using starter cultures of L. bulgaricus and S. thermophiles in milk fermentation showed a significant reduction in AFM1 concentration from 0.075 and 0.207 to 0.068 and 0.198 ppb, respectively. Barukcic et al. ( 24 ) investigated the potential of the probiotics (Lactobacillus acidophilus La-2, Bifidobacterium animalis subsp. lactis BB-12 and Streptococcus thermophiles) to reduce AFM1 in milk contaminated with 54 ng /L AFM1 over a period of 21 days. Their results demonstrated approximately a 50% reduction in AFM1 concentration. These findings align with our results, showing the ability of probiotics in detoxification of AFM1. The present study confirmed the detoxification capability of probiotic bacteria. They indicated that the amount of AFM1 removal by tested bacteria depends on the strain, bacterial population, incubation temperature, and toxin concentration, while storage time had a significant effect. Our findings showed that the significant removal of AFM1 in skim milk contaminated with 0.5 μg/mL and treated with 10 log CFU/mL B. lactis was 57.7% at 120 minutes at 42ºC. Similarly, the significant removal of AFM1 in skim milk spiked with 0.1 and 0.5 μg/mL of AFM1 and inoculated with 8 log CFU/mL S. thermophiles was 24% and 45% at 60 minutes and 24 hours both at 42 ºC. Additionally, the best strains showed the highest AFM1 removal (87%) at 0.5 μg/mL at 24 hours. These findings suggest potential future application of these future applications of these bacteria to control AFM1 in the dairy industry. However, more studies are needed to investigate the mechanisms involved in toxin removal by B. lactis and S. thermophiles, especially considering changes in physicochemical factors.

Acknowledgment

Not applicable

Authors' Contribution

Study concept and design: M. A. SAA. A

Acquisition of data: AE. K.

Analysis and interpretation of data: M. T.

Drafting of the manuscript: N. SH.

Critical revision of the manuscript for important intellectual content: SAA. A, M. A.

Statistical analysis: M. T, N. SH.

Administrative, technical, and material support: SAA. A, AE. K.

Ethics

We hereby declare all ethical standards have been respected in preparation of the submitted article.

Conflict of Interest

The authors declare no relevant financial or non-financial interests to disclose

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data Availability

All data generated or analyzed during this study are included in this published article and its supplementary information files.

References

1. Kumar P, Mahato DK, Kamle M, Mohanta TK, Kang SG. Aflatoxins: A global concern for food safety, human health and their management. Frontiers in microbiology. 2017; 7:2170.
2. Jallow A, Xie H, Tang X, Qi Z, Li P. Worldwide aflatoxin contamination of agricultural products and foods: From occurrence to control. Comprehensive reviews in food science and food safety. 2021; 20(3):2332-81.
3. Dohnal V, Wu Q, Kuca K. Metabolism of aflatoxins: key enzymes and interindividual as well as interspecies differences. Archives of toxicology. 2014; 88(9):1635-44.
4. Lima CMG, Costa HRD, Pagnossa JP, Rollemberg NdC, Silva JFd, Dalla Nora FM, et al. Influence of grains postharvest conditions on mycotoxins occurrence in milk and dairy products. Food Science and Technology. 2021; 42
5. Mollayusefian I, Ranaei V, Pilevar Z, Cabral-Pinto MMS, Rostami A, Nematolahi A, et al. The concentration of aflatoxin M1 in raw and pasteurized milk: A worldwide systematic review and meta-analysis. Trends in Food Science & Technology. 2021; 115:22-30.
6. Vagef R, Mahmoudi R. Occurrence of Aflatoxin M1 in raw and pasteurized milk produced in west region of Iran (during summer and winter). International Food Research Journal. 2013; 20(3):1421.
7. Ismail A, Goncalves BL, de Neeff DV, Ponzilacqua B, Coppa CFSC, Hintzsche H, et al. Aflatoxin in foodstuffs: Occurrence and recent advances in decontamination. Food Research International. 2018; 113:74-85.
8. Goncalves BL, Muaz K, Coppa CFSC, Rosim RE, Kamimura ES, Oliveira CAF, et al. Aflatoxin M1 absorption by non-viable cells of lactic acid bacteria and Saccharomyces cerevisiae strains in Frescal cheese. Food research international. 2020; 136:109604.
9. Pop OL, Suharoschi R, Gabbianelli R. Biodetoxification and Protective Properties of Probiotics. Microorganisms. 2022; 10(7):1278.
10. Corassin CH, Bovo F, Rosim RE, Oliveira CAFd. Efficiency of Saccharomyces cerevisiae and lactic acid bacteria strains to bind aflatoxin M1 in UHT skim milk. Food control. 2013; 31(1):80-3.
11. Mahmood Fashandi H, Abbasi R, Mousavi Khaneghah A. The detoxification of aflatoxin M1 by Lactobacillus acidophilus and Bifidobacterium spp.: A review. Journal of food processing and preservation. 2018; 42(9):e13704.
12. Allam NG, Ali EMM, Shabanna S, Abd-Elrahman E. Protective efficacy of Streptococcus thermophilus against acute cadmium toxicity in mice. Iranian journal of pharmaceutical research: IJPR. 2018; 17(2):695.
13. Ismail A, Levin RE, Riaz M, Akhtar S, Gong YY, de Oliveira CAF. Effect of different microbial concentrations on binding of aflatoxin M1 and stability testing. Food control. 2017; 73:492-6.
14. Sokoutifar R, Razavilar V, Anvar AA, Shoeiby S. Degraded aflatoxin M1 in artificially contaminated fermented milk using Lactobacillus acidophilus and Lactobacillus plantarum affected by some bio-physical factors. Journal of Food Safety. 2018; 38(6):e12544.
15. Bovo F, Corassin CH, Rosim RE, de Oliveira CAF. Efficiency of lactic acid bacteria strains for decontamination of aflatoxin M1 in phosphate buffer saline solution and in skimmed milk. Food and Bioprocess Technology. 2013; 6(8):2230-4.
16. Elsanhoty RM, Salam SA, Ramadan MF, Badr FH. Detoxification of aflatoxin M1 in yoghurt using probiotics and lactic acid bacteria. Food control. 2014; 43:129-34.
17. Pierides M, El-Nezami H, Peltonen K, Salminen S, Ahokas J. Ability of dairy strains of lactic acid bacteria to bind aflatoxin M1 in a food model. Journal of food protection. 2000; 63(5):645-50.
18. Sarlak Z, Rouhi M, Mohammadi R, Khaksar R, Mortazavian AM, Sohrabvandi S, et al. Probiotic biological strategies to decontaminate aflatoxin M1 in a traditional Iranian fermented milk drink (Doogh). Food control. 2017; 71:152-9.
19. Adibpour N, Soleimanian-Zad S, Sarabi-Jamab M, Tajalli F. Effect of storage time and concentration of aflatoxin m1 on toxin binding capacity of L. acidophilus in fermented milk product. Journal of Agricultural Science and Technology. 2016; 18(5):1209-20.
20. Panwar R, Kumar N, Kashyap V, Ram C, Kapila R. Aflatoxin M1 detoxification ability of probiotic lactobacilli of Indian origin in in vitro digestion model. Probiotics and antimicrobial proteins. 2019; 11(2):460-9.
21. Karazhiyan H, Mehraban SM, Karazhyan R, Mehrzad A, Haghighi E. Ability of different treatments of Saccharomyces cerevisiae to surface bind aflatoxin M1 in yoghurt. Journal of Agricultural Science and Technology. 2016; 18:1489-98.
22. Namvar Rad M, Razavilar V, Anvar SAA, Akbari-Adergani B. Selected bio-physical factors affecting the efficiency of Bifidobacterium animalis lactis and Lactobacillus delbrueckii bulgaricus to degrade aflatoxin M1 in artificially contaminated milk. Journal of Food Safety. 2018; 38(4):e12463.
23. El Khoury A, Atoui A, Yaghi J. Analysis of aflatoxin M1 in milk and yogurt and AFM1 reduction by lactic acid bacteria used in Lebanese industry. Food control. 2011; 22(10):1695-9.
24. Barukcic I, Bilandzic N, Markov K, Jakopovic KL, Bozanic R. Reduction in aflatoxin M1 concentration during production and storage of selected fermented milks. International journal of dairy technology. 2018; 71(3):734-40.