Development of complementary methods for the detection of animal by-products in feedingstuffs

Since the mid-1980s, several transmissible spongiform encephalopathies have been reported in humans and animals. Bovine spongiform encephalopathy was first identified in cattle in 1986 and, in subsequent years in other animal species. Consumption by cattle of meat and bone meal containing carcasses of infected animals was incriminated. The use of such animal by-products was first banned in feed for ruminants and then, with few exceptions such as the use of fish meal for non-ruminants, extended to all feeds for farmed animal. Thanks to prevention and control efforts, a gradual lifting of the feed ban is now possible. The reintroduction of non-ruminant processed animal proteins into the feed of aquaculture animals was adopted in 2013. In this context of dynamic relaxation of the ban, the challenge remains the development of complementary methods or the adaptation of official methods aimed at refining the identification of feed materials. Indeed, the combination of light microscopy and PCR does not differentiate among certain animal by-products and this can make the results uninterpretable.
This PhD thesis focused on (1) the evaluation of current legislation and available methods of analysis for the detection of animal proteins; (2) the development of a fluorescence in situ hybridisation method combining the advantages of microscopy (localisation and identification of type of tissues) and molecular biology techniques (identification of the species by DNA); (3) the identification of specific peptides from blood products by high resolution mass spectrometry and (4) the development of a sensitive and routinely applicable tandem mass spectrometry method paving a new way to the detection of processed animal proteins.
The results obtained allowed to (1) highlight current and foreseen analytical gaps as part of the feed ban relaxation dynamics; (2) propose a method for a specific detection of processed bone particles of ruminant origin independently of contamination by other by-products such as milk powder; (3) identify 9 biomarker peptides of plasma powder of bovine origin and 7 biomarker peptides of bovine haemoglobin powder, 6 of which also allow the detection of blood meal of bovine origin and (4) develop a UHPLC MS/MS multi-target method allowing the simultaneous detection of several by-products of ruminant origin (blood meal, blood products and dairy products) at a limit of detection (LOD) of 0.1% (w/w). Finally, new analytical schemes incorporating the results of this thesis as well as potential sources of improvement for future analytical developments have been proposed.