Proton-NMR metabolomics of rainbow trout fed a plant-based diet supplemented with graded levels of a protein-rich yeast fraction reveal several metabolic processes involved in growth

Abstract

Background: Plant raw materials are commonly used in aquafeeds, as marine resources are unsustainable. However, full plant-based diets lead to poorer fish growth performance. Objective: We aimed to understand the metabolic effects of a yeast fraction as a protein supplement in a plant-based diet and to integrate such effects with phenotypic traits as a new approach to assess the interest of this raw material. Methods: Juvenile (49 g) rainbow trout (Oncorhynchus mykiss) were fed graded levels of a yeast protein-rich fraction (5% YST05, 10% YST10, 15% YST15) in a plant-based diet (PB) for 84 d. Final body weight, feed conversion ratio, and hepatosomatic and viscerosomatic indexes were measured. Plasma, liver, and muscle 1H-NMR fingerprints were analyzed with principal component analyses, and their metabolite patterns were clustered according to the yeast level to identify concomitant metabolic effects. A regression modeling approach was used to predict tissue metabolite changes from plasma fingerprints. Results: In tissues, the patterns of metabolite changes followed either linear trends with the gradual inclusion of a yeast fraction (2 patterns out of 6 inmuscle, 1 in liver) or quadratic trends (4 patterns in muscle, 5 in liver). Muscle aspartate and glucose (395 and 138% maximum increase in relative content compared with PB, respectively) revealing modification in energy metabolism, as well as modification of liver betaine (163% maximum increase) and muscle histidine (57% maximum decrease) related functions, indicates that the yeast fraction could improve growth in several ways. The highest correlation between measured and predicted metabolite intensities in a tissue based on plasma fingerprints was observed for betaine in liver (r = 0.80). Conclusions: These findings herald a new approach to assess the plurality of metabolic effects induced by diets and establish the optimal level of raw materials. They open the way for using plasma as a noninvasive matrix in trout nutrition studies.