Efficiency of Selected Lactic Acid Bacteria Isolated from some Dairy Products on Aflatoxin B1 and Ochratoxin A

Aflatoxin B1 (AFB1) and ochratoxin A (OTA) are two of the most important of mycotoxins associated with tropical and subtropical climate as in Egypt. Therefore, this study aimed to isolated lactic acid bacteria (LAB) and Bifidobacteria from some local dairy products from Egypt and using some strains to reduce/or remove AFB1 and OTA toxins. Fifty samples, of milk and dairy products, including ten samples each, of Kareish, Damietta cheese (soft type), buffalo’s milk, yoghurt (zabady), and naturally acidified milk (Rayeb)were screened for their load of Lactobacilli, Lactococci and Bifidobacteria, using MRS agar, M17 agar and (MRS-Cys) agar, respectively. The obtained data indicated that 38, 31 and 22 isolates belonged to Lactobacillus, Lactococcus and genus Bifidobacterium, respectively. Fourteen strains isolated from LAB and Bifidobacterium were tested on AFB1 and OTA in PBs contaminated by10 ppb with three times incubation periods (6, 12 and 36 hours) at 37°C. The results indicated that all tested strains were able to reduce AFB1 at different rates ranging between 12.1 to 65.4% after incubation for 6 h. These rates increased to 78.8% and 89.9% after incubation at 37°C for 12 and 36h, respectively. The same results were with OTA, where percentages of reduction ranging from 81.4 to 80.4% were obtained when incubation with LAB cells was done at 37°C and prolonged to 36 h. According to the results, the percentage of reduction increased with the increasing time for each strain.


INTRODUCTION
Mycotoxins can be defined as toxic secondary metabolites that are produced by many species of fungi in the field and /or during storage. Aflatoxin B 1 (AFB 1 ) and ochratoxin A (OTA) are two of the most important mycotoxins regarding their occurrence and toxicity 1 . AFB 1 and OTA were classified by the International Agency for Research on Cancer (IARC) as human carcinogens that belong to Group 1 and Group 2B, respectively 2 . Contamination by AFB 1 and OTA harms human and animal health and causes big economic losses that reach billions of dollars every year. Thus, getting rid of this toxin in food and feed by using certain strategies is needed urgently. Current methods to reduce mycotoxin contamination in food and feed can be classified into physical, chemical and biological methods 3,4 . Many studies used the bacteria, yeasts and molds to reduce or degrade AFB 1 and OTA in foods and feeds 5,6 . Lactic acid bacteria (LAB) are a good choice for control or reduce AFB 1 and OTA from a contaminated medium, generally considered to be safe according to USFDA; some of them also have many benefits on health, which are called probiotics bacteria 7 . Viable and non-viable cells of LAB have an adsorbent ability to bind toxins. Where it overlaps between the toxins and the functional groups of the cell surface happens, adsorption occurs. A lot of factors like temperature, pH, time of incubation, bacterial concentration and type of strain may affect the ability of LAB to bind toxins 8,9 . Therefore, this study aimed to evaluate the ability of some strains LAB, isolated from dairy products, to reduce or degrade AFB 1 and OTA from liquid media at (6h, 12h and 36 h) period incubation.

MATERIALS AND METHODS Samples from dairy and dairy products
Fifty samples were collected from milk and dairy products. The collected samples included ten samples from each of buffalo's milk, yoghurt, naturally acidified milk (Rayeb), Kareish and Domiati cheese (soft type). All samples were analyzed using three isolation media. De Man, Rogosa, Sharp agar (MRS) agar (Oxoid, UK) was used for isolation of lactobacilli, MRS medium supplemented with L-cysteine (0.5 g/liter) (MRS-Cys) agar was used for isolation of Bifidobacteria, however, M17 medium (Oxoid) agar was used for isolation of Lactococci spp.

Microbiological assessment of products
Twenty-five g/ml of each sample was aseptically transferred to conical flask 500 ml and homogenized in 225 ml of sterile tri-sodium citrate 10 . Ten-flods dilutions of homogenates were prepared and inoculated onto plates of MRS, MRS-Cys agar, which incubated in anaerobic condition using Co 2 generated kits (Anaero-Gen. Oxoid) and incubated at 37°C for 48 h, while M17 plates were incubated in aerobic conditions at 32°C. All plates were incubated for a period of three to five days 11,12 .

Identification of LAB and Bifidobacterium
All the selected strains (120) isolates were activated twice in MRS-Cys, MRS and M17 broth at 37°C for 24 h. The selected isolates were observed by optical microscopy to determine their morphology and Gram staining results 13 . Additionally, they were tested for catalase 14 , oxidase, nitrate reductase, gelatinase activities, production of indol, and production of gas from glucose 15 . The API-50CHL identification system (BioMerieux, Lyon, France) was used for Lactobacilii and Bifidobacterium, whereas the API-20 STREP identification system (Bio-Merieux, Lyon, France) was used for Lactococci.

Preparation of culture cells
The strains of lactic acid bacteria were activated overnight (16 h) in 100 ml MRS, MRS-Cys and M17 broth. A cell free solution was obtained by centrifugation of culture at 5000 rpm at 4°C for 20 min. Preparation of AFB 1 and OTA spiked solution AFB 1 and OTA standards were purchased from Sigma, chemical Co. (St. Louis, MO, U.S.A). Stock solutions and standards were prepared and assayed according to AOAC Method 971. 22,2005) 16 . Standards separately were added by 10 ppb to 100 mL phosphate buffered saline (PBS). 17 All solvents were of HPLC grade. The water was double distilled with Millipore water purification system (Bedford, M A, USA).

AFB 1 and OTA binding assay in vitro
One hundred milligrams of bacterial cells were added to100 ml of PBs contaminated by 10 ppb of AFB 1 and OTA both are separate from each other. Samples were incubated at 37°C for 6, 12 and 36h for testability of strains for binding of toxins. Then in each incubation period; filtration for discarding bacterial cells was done through filtration then extraction and determination of AFB 1 or OTA in PBs (100 mL) with three replicates for each incubation time. Extraction and determination of AFB 1 using HPLC AFB 1 was extracted and cleanup using immunoaffinity column (IAC) Aflatest ®-p affinity to obtained the dry film, then determination AFB 1 in samples after derivatives process by adding 100μl of trifluoroacetic acid (TFA) to samples and mixed well for 30 s and the mixture stands for15 min, finally measurement by HPLC) 17 Extraction and determination of OTA using HPLC OTA was extracted and cleaned up using the method described in AOAC, (2007) 18 using the IAC (Ochra Test ®-p affinity column), then concentrations of OTA were determined by High-Performance Liquid Chromatography (HPLC) system 19 .

Statistical analyses
All data were statistically analyzed using the General Linear Model procedure of the SPSS ver. 18 (IBM Corp, NY). The significance of the differences among treatment groups was determined by Waller-Duncan k-ratio. All statements of significance were based on the probability of Pvalue ≤0.05 was considered to be statistically significant.

RESULTS AND DISCUSSION Isolation of lactic acid bacteria and Bifidobacterium
A total of fifty samples of milk and dairy products were investigated for their load of LAB and Bifidobacterium spp. This included 10 samples, each of buffalo's and Rayeb milk, yoghurt, Kareish and Domiati cheese. The results showed in (Fig. 1) as an average log of a colony-forming unit (cfu) of LAB and Bifidobacterium spp.
Yoghurt was the highest product in log count among all other products, for example the The results indicated that using L. rhamnosus gave the highest reduction for AFB 1 after incubation at 37°C for 6, 12 and 36 h, where 65.4%, 72.6% and 89.0%, respectively. When using L. plantarum for removing or degradation of AFB 1 , the increase in removal with increased time of incubation was 61.3%, 72.4 and 85.9% after 6, 12 and 36h, respectively. While Bifidobacterium bifidum and Streptococcus thermophilus reduced AFB 1 88.8 and 88.3% after 36h of incubation time, respectively. Generally, the minimum reduction percentages of AFB 1 were 12.1, 22.6 and 40.4%after incubation for 6h, 12h and 36h, respectively with Lactococcus lactis. ANOVA analysis of the effect of incubation time and type of LAB strains on the content of AFB 1 in liquid media is shown in (Table 2), which shows that there is a significant effect due to the incubation period as well as LAB strain. The obtained data shown in (Fig. 2) reflected that bind of AFB 1  24 , noticed that the ability of AFB 1 to bind in vitro and in vivo depended on LAB strain. Though the complex formed between AFB 1 and bacteria was proved to be stable, the mechanism of binding between AFB 1 and LAB was not illustrated yet. It was thought that AFB 1 binds to cell wall components, including both polysaccharides and peptidoglycans. It was identified before that both hydrogen bonds and Van der Waals interactions were involved in AFB 1 binding [25][26][27] . AFB 1 binds by the LAB cell wall during processes namely adsorption and desorption, primarily, the AFB 1 molecule is trapped inside the single helix of the (1 ➔ 3)-β-D-glucan chain. Then, the AFB 1 molecule is covered by the branched (1➔6)-β-D-glucan chain, where it is preserved inside the helix. It also showed some involvement with both electrostatic interactions and hydrogen bonds between AFB 1 molecules and the bacterial cell wall. Furthermore, AFB 1 can be bound to the bacterial cell wall through weak non-covalent interactions such as associating with hydrophobic pockets on the bacterial surface. The differences in AFB 1 binding by the strains are probably due to the different bacterial cell wall and cell envelope structure 28,29 .

Impact of LAB on the content of OTA in contaminated PBs medium
The obtained results indicated that the highest reduction percentage of OTA after incubation for 6h was 59.3, 52.0 and 50.2% after incubation with cells from L. rhamnosus, L. plantarum and L. reuteri, respectively. Besides, the lowest reduction was 10.5, 12.1 and 15.5% in samples incubated with Lactococcuslactis, Lactococcus cremoris and Leuconostoc mesenteroides, respectively. Incubation with L. rhamnosus for 12 and 36h removed 68.1 and 81.4% from OTA, respectively. On the other hand Str. thermophilus and Bifidobacterium bifidum cells removed 80.3 and 80.4% from OTA after incubation for 36h at 37°C. The ANOVA analysis in (Table 3) showed significant differences between reduction percentages of OTA after incubation with different LAB cells as well as significant differences in incubation time. These results indicated that the incubation time was one of the important factors for the efficiency of LAB cells to bind OTA in PBs. Increasing the incubation time to 12 and 36h led to increasing the reduction of OTA in contaminated PBs as shown in Fig. 3. These results are consistent with many studies conducted in this field 29 31 investigated the fifteen strains belonging to five relevant oenological LAB species that were grown in broth synthetic culture medium spiked with OTA. The portion of OTA removed during the bacterial growth was 8 to 28%. The OTA was removed from the supernatants between 31 to 57% by the bacterial pellet. The most effective mechanism to remove OTA is binding with the bacterial cell wall, where it contained peptidoglycan matrices, polysaccharides, as well as teichoic and lipoteichoic acid, and a protein S-layer. However, binding is based on the adsorption capacity of toxins to the cells and not on enzyme activity. This is where peptidoglycan and exopolysaccharides play an important role. LAB which is thermally inactivated gives higher removal capacity may due to changes on the cell surface. Another proposed hypothesis, namely binding of toxin by bacterial cells is also influenced by electron donor-acceptor and Lewis acid-base interactions, this may due to that the cells of LAB consider as strong electron donors and weak electron acceptors, which is supported by the hydrophilic nature of their surface 32,33 .

CONCLUSION
Incubation of LAB cells with PBs contaminated with AFB 1 and OTA might reduce the toxin contamination. The obtained results indicated that all strains were able to reduce AFB 1 and OTA but at different rates which ranged between 10 to 50% with incubation for 6h and increased to 75 % after incubation for 12h. Moreover, increasing the incubation time with LAB cells to 36h led to removing 80 and more than 88% from AFB 1 and OTA in contaminated PBs.