Quantitative assessment of transferable antibiotic resistance in zebrafish (Danio rerio) fed Hermetia illucens-based feed

Antibiotic resistance (AR) is an issue in aquaculture since the excessive use of antibiotics administered both for prophylaxis and metaphylaxis may lead to selection for resistant bacteria, that can be found in edible products (e.g., seafood) marketed for human consumption. There is a paucity of data in the scientific literature on the dynamics of AR genes in fish feeds, including those with insect meal as an ingredient. This study represents the first attempt to assess the dynamics of genes conferring resistance to antibiotics conventionally used in clinical practice such as macrolide-lincosamide-streptogramin B (MLSB) [erm(A), erm(B), erm(C)], vancomycin (vanA, vanB), tetracyclines [tet(M), tet(O),tet(S), tet(K)], -lactams (mecA, blaZ) and aminoglycosides [aac(6’)-Ie aph(2”)-Ia]. For this purpose, zebrafish (Danio rerio) fed Hermetia illucens-based diets was used as an experimental model. AR genes were assessed using optimized quantitative PCR (qPCR) assays. The dynamics of the selected AR genes were studied during a 6-month feeding trial with zebrafish fed diets including increasing levels of H. illucens (Hi) larvae (25%, 50%, 75%, and 100%) that were previously reared on coffee silverskin supplemented with 10% Schizochytrium sp. microalgae. A diet with no insect additions was used as the control. qPCR detected erm(B), tet(K), tet(M), tet(O) and tet(S) genes in all the analyzed diet samples, whereas the same samples were negative for mecA, vanA, vanB and aac-aph. In zebrafish at their larval, juvenile and reproductive stages, erm(A), erm(C), vanB and aac-aph genes were never detected, whereas erm(B), tet(M) and tet(S) genes were widespread irrespective of the stage of development or feed. According to principal component analysis, a clear separation among fish fed diets containing the highest or the lowest insect inclusion levels was generally observed. These results showed a significant contribution of insect-based feed to the spread of the studied AR genes compared with control diets. This finding suggests the occurrence of complex interactions between the microbiota and the resistome occurring in feed and in the fish gut. Notably, when present, AR genes could be transferred to commensal or pathogenic microorganisms that may reach the food chain. These results represent a first baseline for future AR risk assessments in the edible insect feed chain. Until then, prudent use of antibiotics during rearing of edible insects is recommended.