The capacity of silage inoculant bacteria to bind aflatoxin B1 in vitro and in artificially contaminated corn silage

The objectives were to examine the aflatoxin B-1 (AFB1)-binding capacity of silage bacteria and factors affecting the responses. Experiments 1 and 2 examined the effects of bacterial strain and population on the AFB(1)-binding capacity of 10 bacteria. When applied at 10(6) cfu/mL to an in vitro medium, only Lactobacillus plantarum PT5B bound the AFB(1) and the binding capacity was low (4%). When applied at 10(9) cfu/mL, all 10 bacteria bound AFB(1), but L. plantarum R2014 (Lp) and EQ12, Lactobacillus buchneri R1102 (Lb), and Pediococcus acidilactici R2142 and EQ01 (Pa) had the greatest capacity (23.9 to 33%). Experiment 3 examined the AFB(1)-binding capacity of viable and nonviable (HCl-treated) forms of Lp, Lb, and Pa at different pH. Nonviable Lb and Lp, but not Pa, increased AFB(1) binding. Binding of AFB(1) was greatest at pH 2.5 and least at pH 8. As the nonviable Lb and Lp that bound AFB(1) in experiment 3 would not be effective silage inoculants, experiment 4 examined effects of benign versus severe treatments (85 vs. 100 degrees C; pH 2.5 vs. <1) on the viability of Lp, Lb, and Pa. The population of bacteria was reduced from 9 to 4 log cfu/mL by treatment with HCl at pH 2.5 and to 2 log cfu/mL by 85 or 100 degrees C, whereas acidification at pH <1 eliminated the bacteria. Experiment 5 determined the effect of the ensiling duration and benign treatment methods [37 (viable cells) or 85 degrees C (heated cells) or acidification with HCl at pH 2.5 (acid-treated cells)] on binding of AFB(1) and silage quality during the fermentation of corn forage. Corn forage was ensiled after treatment with only deionized water (control), AFB(1) (30 mu g/kg of fresh forage), or a mixture of AFB(1) and 109 cfu/g of each of the treated bacteria. Adding AFB(1) alone to corn forage reduced the pH decline during the first 3 d of ensiling and increased or tended to increase butyric acid concentration and final pH after ensiling for 21 d. Bacterial inoculation inhibited these negative effects. The fermentation profile of silage treated with Lb and Pa did not differ from those of the control silage. In all silages treated with the toxin, the AFB(1) concentration decreased linearly (from 30 to <= 0.35 mu g/kg) within 3 d of ensiling. Certain silage bacteria can bind AFB(1) but the efficacy depends on several factors.