*2.1. Instrumental Analysis of Mycotoxins*

Analysis of 54 biomasses as well as the corresponding 6 MRS broth samples was carried out by high-performance liquid chromatography (HPLC) coupled with UV detection upon solvent extraction on the basis of a method optimization. Recent methods use mainly acetonitrile for the extraction of mycotoxins from foodstuffs, followed by cleanup with different modes of solid phase extraction (e.g., immunoaffinity, dispersive, etc.) to eliminate matrix effects. For cultivated bacteria or fungal strains, methanol [33] or chloroform [34,35] are the most frequently applied solvents, then the extracts are either subjected to a cleanup [33] or only filtered [35] prior to the analysis by a liquid chromatographic method. For complex matrices, gradient elution is applied [33], but for the bacteria investigated in the present study, a simple isocratic method gave sufficient separation. Typical chromatograms of target compounds are shown in Figure 1. The retention times were 6.07 and 7.13 min for AFB1 and ST, respectively. Although mycotoxins could be determined directly from the MRS broth, removal of the most polar matrix components resulted in better baseline and less interference (see Figure 2). For the extraction of MRS broth, the more volatile solvent, dichloromethane, allowed good recoveries (93.4 ± 3.1 and 97.6 ± 4.8% for AFB1 and ST, respectively), while for biomass, addition of 10% methanol to dichloromethane significantly increased the extraction efficiency by enhancing the solvent penetration to the cells. Ultrasound agitation seemed to be less effective than shaking of samples. Both analytes were determined from HPLC peak areas at the corresponding retention times with excellent linear calibration characteristics. For quantification of target compounds, peak areas determined for AFB1 at 365 nm and for ST at 240 nm were used. Peak purity for ST was checked by the ratios of signal intensities (peak areas) recorded at 240 and 325 nm, which was found 2.03 for standard solutions. The linear regression values of external calibration curves were 0.9992 and 0.9997, and the slopes were 110.7 and 145.3 for AFB1 and ST, respectively. The limits of detection, determined with standard solutions, were 0.010 µg/mL for both mycotoxins, and they were the same in spiked liquid matrices extracted from blank.

**Figure 1.** *Cont*.

**Figure 1.** Chromatograms of samples extracted from biomass containing Aflatoxin B1 (AFB1) (**a**) or sterigmatocystin (ST) (**b**) at levels of 0.070 and 0.236 µg/mL, respectively.

**Figure 2.** Chromatograms of a sample measured directly without any extraction (upper line) and that of the same sample extracted from MRS broth (lower line) containing AFB1 at the level of 0.50 µg/mL.
