*4.3. Mycotoxin Extraction and Analytical Determination*

The mycotoxin content of the samples was determined by UV detection after high performance liquid chromatographic separation (HPLC-UV) on the basis of literature methods for AFB1 [19]) and ST [36,37]. First, the mycotoxin was extracted as follows. The cultures in the Falcon tubes were centrifuged for 40 min at room temperature at 4000 rpm. The supernatant contains the remaining unbound mycotoxin and the residue, referred to as the biomass hereinafter, contains the mycotoxin bound by the bacteria. One millilitre of the supernatant transferred to an empty Falcon tube was shaken with 1 mL of dichloromethane for 20 min in a horizontal shaker in the dark. From the dichloromethane

phase, 0.5 mL was taken out and concentrated in a clean Eppendorf tube at 45 ◦C under a fume hood. For the extraction of the mycotoxin from the biomass, 1.8 mL of dichloromethane and 0.2 mL of methanol were added to the Falcon tube containing the biomass. The mixture was pipetted into Eppendorf tubes. The tubes were vortexed (Vortex Genie 2, MO BIO Laboratories, Carlsbad, CA, USA) for 20 min in the dark and then centrifuged at 3000 rpm for 10 min. One ml of the supernatant was evaporated as before. The residues of the extracts were resolved in 1.0 mL eluent, and determined by HPLC on a Younglin YL9100 HPLC system equipped with a YL9150 autosampler (YL Instruments Co., Anyang, Korea). For the analysis, 30 µL of the extracts were applied onto a Brisa (Technochrome) C18 column (5 µm, 15 cm × 0.46 cm) at 30 ◦C. The separation was carried out at a flow rate of 1 mL/min using isocratic elution, containing 60:20:20 or 40:30:30 (*v*/*v*%) of water, methanol and acetonitrile for AFB1 and ST, respectively. The detector wavelengths were 365 nm or 325 and 240 nm for AFB1 and ST, respectively. All determinations were performed in triplicates from three parallel samples. Relative standard deviations established for binding capacities for three parallel samples ranged between 0.53% and 1.35%.
