Implementation of the Bacillus cereus microbiological plate used for the screening of tetracyclines in raw milk samples with STAR protocol – the problem with false-negative results solved

Abstract

In antibiotic residue analyses the first step of screening is just as important as the following steps. Screening methods need to be quick and inexpensive, but above all sensitive enough to detect the antibiotic residue at or below the maximum residue limit (MRL). In the case of a positive result, the next step is conducted and further methods are used for confirmation. MRLs stated in European Union Regulation 37/2010 for tetracyclines in raw milk are: 100 µg kg–1 for tetracycline, 100 µg kg–1 for oxytetracycline, 100 µg kg–1 for chlortetracycline and no limit for doxycycline because it is prohibited for use in animals from which milk is produced for human consumption. The current five-plate microbiological screening method for the detection of antibiotic residues in raw milk was found to be simple and inexpensive, but not specific, sensitive and reliable enough to detect tetracycline at MRL in routine raw milk screening procedures. Spiking samples with tetracycline at the MRL level and applying them on Bacillus cereus ATCC 11778 microbiological plates often gave false-negative results, indicating that tetracyclines may have to be inactivated or masked. Tetracyclines seem to bind to a certain component in milk. Consequently, when applying samples to the B. cereus microbiological plate the antibiotic cannot inhibit the growth of B. cereus which disables the formation of inhibition zones on the test plate. After adding the appropriate amount of citric acid into the milk samples, we solved the problem of false-negative results. During the validation 79 samples of milk were spiked with tetracyclines at different concentrations: 100 µg kg–1 for tetracycline, 100 µg kg–1 for oxytetracycline, 80 µg kg–1 for chlortetracycline and 30 µg kg–1 for doxycycline. Concentrations used in the validation matched the requirements for MRLs (they were either at or below the MRLs) stated in European Union Regulation 37/2010. The sensitivity of the validation was 100%.