Watch the slideshow

Jacob de Jong1, Christoph von Holst2, Ursula Vincent2, Gisele Gizzi2, Christophe Genouel3, Jacques Michard3, Harry van Egmond1, Tina Zuidema1, Irene Pol-Hofstad1 and Wim Beek1
1   RIKILT- Institute of Food Safety  (RIKILT, Wageningen, The Netherlands)
2   European Commission, Directorate General Joint Research Centre, Institute for Reference Materials and
Measurements (IRMM, Geel, Belgium)
3   Laboratoire de la Direction Générale de la Concurrence, de la Consommation et de la Répression des Fraudes  (DGCCRF, Rennes, France)

In order to protect human health the authorisation of several antibiotics (avoparcin, zinc-bacitracin, spiramycin, tylosin and virginiamycin) and growth promoters (carbadox and olaquindox) as feed additives has been withdrawn in the European Union in 1999. The effective and efficient control of the possible illegal use of these substances requires the availability of multi-screening and confirmation methods.

The main technical and scientific objective of the EU-project SIMBAG-FEED was to develop, improve and validate adequate methods which can be implemented in an overall control strategy applying the following techniques: microbiological inhibition, high voltage electrophoresis (HVE), Thin Layer Chromatography (TLC), High Performance Liquid Chromatography (HPLC) and Liquid Chromatography- Mass Spectrometry (LC-MS/MS).

The methods should be: a) sufficiently sensitive (antibiotics 1 mg/kg; olaquindox 3 mg/kg; carbadox 4 mg/kg); b) multi-analyte; c) able to distinguish between the banned substances and registered feed additives and antibiotics used as veterinary drugs in medicated feed; d) robust;  e) the techniques should be widely available in control laboratories; f) reliable, viz. for the confirmation techniques unequivocal identification is more important than quantification.

Methods have been developed and in-house validated. Between-lab validation of the methods has been performed in two laboratories that were not involved in method development. Validation has been performed according to Commission Decision 2002/657/EC.

Based on the results of the validation studies, the candidate methods for the proposed control strategy were selected and subjected to international collaborative studies.

For the antibiotics a two- or three-step control strategy has been established, comprising of several multi-methods. The microbiological plate test is the method of choice for initial screening of feedingstuffs while LC-MS/MS is the method of choice for confirmation. In between, TLC or HVE can be used as a post-screening step in order to minimise the risk of false-positive results caused by the presence of antimicrobial veterinary drugs. For tylosin, spiramycin and virginiamycin the target limits of 1 mg/kg were reached for all techniques, except for spiramycin with TLC (2 mg/kg). The control strategy is supported by fully validated methods, incl. collaborative studies. For zinc-bacitracin, the limits of detection for the screening and post-screening step are somewhat higher (3 – 5 mg/kg).

For avoparcin and zinc-bacitracin, the post-screening step is only partly validated by means of collaborative studies (viz. a limited collaborative study of HVE) and the LC-MS/MS confirmation could only be implemented by part of the laboratories participating in the collaborative study at the target limit of identification of 1 mg/kg. Next to the abovementioned techniques, an LC-UV multi-method for tylosin and spiramycin and an LC-fluorescence method for virginiamycin were developed and validated by three laboratories.

These HPLC-methods, that were not collaboratively studied, are suitable for post-screening of feeds with contents starting from approx. 10 mg/kg and 3 mg/kg respectively.

For the growth promoters, the control strategy consists of a two- or one step approach, starting with an LC-UV multi-method for screening and quantification, followed by confirmation with LC with diode-array detection (LC-DAD) or LC-MS-MS multi-methods. If LC-DAD is applied directly, the control strategy can be reduced to a single-step approach. The target limits in feeds were reached and the control strategy is supported by fully validated methods, incl. collaborative studies.

Website
http://www.rikilt.wur.nl/UK/research/Projects/EU+projects/Output+SIMBAGFEED/

Acknowledgements
The SIMBAG-FEED project was funded by the European Commission, DG RTD, 5th Framework Programme, Competitive and Sustainable Growth  (GROWTH) Programme, Contract no. G6RD-CT-2000-00413.