Introduction
Ever since mycotoxins have been described as emerging feed contaminants for European aquaculture, it has become evident that deoxynivalenol (DON) is the most prevalent mycotoxin in aquafeeds (Koletsi et al., 2021). Different fish species display different sensitivities to DON. Compared to other fish species, rainbow trout (Oncorhynchus mykiss ) is very sensitive to DON (Hooft et al., 2011; Koletsi et al., 2021). Experimentally, studies on the effects of DON have generally targeted the concentration of the mycotoxin instead of focusing on the composition of the diet. Studies with DON in carnivorous fish have hitherto been performed against a background of optimal quality marine-based diets. Yet, as argued above, ingredient composition of aquafeeds has diversified rapidly to include more plant-based ingredients. This not only introduces the risk of feeding mycotoxin contaminated ingredients but also introduces a risk that the negative effects of DON can be amplified by sub-optimal diets. Information on the interaction between DON and diet composition is generally lacking . Therefore, this study investigated if dietary composition influences the effects of DON on the health and performance of rainbow trout (Oncorhynchus mykiss ).
Material and methods
Four experimental diets (2x2 factorial design) were formulated which differed in 1) the type of protein source; fishmeal (FM) versus soybean meal-based (SBM) and 2) the DON content of wheat; clean versus naturally contaminated wheat. Triplicate groups of n=30 fish were assigned to each diet: (1) CON-FM; DON= 0 µg/kg feed; (2) DON-FM; DON=1200 µg/kg feed; (3) CON-SBM; DON= 46 µg/kg feed; (4) DON-SBM; DON= 1300 µg/kg feed. The 8 week experiment was divided into two feeding periods: after 6 weeks of restrictive feeding, fish were fed ad libitum for 2 weeks. Influences on performance were evaluated by determining growth, protein and energy gain metrics, and on health parameters through the determination of histopathological changes in the liver and gastrointestinal tract (GIT).
Results
Restrictive feeding showed negative effects of DON and dietary composition on performance but did not show an interaction between DON and diet composition. Similarly, subsequent ad libitum feeding showed effects of DON and/or diet composition on growth, feed efficiency and body biometrics, but no interaction effects. These data confirmed the challenging nature of the SBM-based diet and confirm previously noted negative effects of DON on performance. Neither DON, nor diet composition indicated significant effects on liver health, nor was there an interaction effect. We discuss that the combination of DON and a sub-optimal diet based on SBM could accentuate the effects of DON, in particular concerning GIT functioning. Indeed, the histopathological assessment of mucosal fold width, enterocyte width and goblet cell density indicated significant interaction effects between DON and diet composition in the midgut. Yet, the differences were generally small and interaction effects were restricted to the midgut. Overall, this study suggests that DON is harmful to rainbow trout regardless of diet quality.