Plant-based diet in rainbow trout (Oncorhynchus mykiss Walbaum): Are there genotype-diet interactions for main production traits when fish are fed marine vs. plant-based diets from the first meal?
Résumé
In the last years, the increase of aquaculture production has led to the evolution of feed composition with an increasing substitution of fish meal and fish oil with terrestrial plant products. In the meantime, selective breeding of fish has been widely developed. The ability to grow on plant-based diets has recently been proven to be genetically variable, pointing out the interest to increase knowledge on the potential consequences of substitution of fish meal and fish oil on current breeding programs. Moreover, heritabilities of major production traits other than growth also need to be estimated in this new environment. Experimental rainbow trouts (about 3000) were derived from a factorial mating design (25 sires and 10 dams), separated in two batches and fed two experimental diets since the first meal. The marine diet (M) was based on fish meal (FM) and fish oil (FO), while the substitution rate by plant products was total in the plant-based (PB) diet (free of FM and FO). We estimated heritabilities under both diets and genotype by diet interactions for body weight, fork length, condition factor, TGC (thermal-unit growth coefficient) and processing yields: viscera yield, carcass yield, filet yield and head yield. Due to the reduced growth of fish fed the plant-based diet, the estimation of genetic parameters was done on fish from both dietary treatments measured at the same age and also on fish measured at the same weight assuming that family rankings could evolve with further growth. Survival rates were high and did not depend on the diet. At 343 days, the body weight of the M batch (440.2 g ± 11.3) was twice that of the PB batch (216.0 g ± 12.6). After 465 days, the PB batch reached 407.4 g ± 17.1. Whatever the comparison (at same age or same weight), there were significant diet-induced differences for processing traits except for head yield (P = 0.80, same age) and filet yield (P = 0.82, same weight). The main findings are the high heritabilities observed in fish fed the plant-based diet for body weight (0.69 ± 0.13), TGC (0.65 ± 0.12) and the generally low, though significant, genotype by diet interactions for final body weight, fork length, carcass yield, viscera yield, head yield and filet yield between fish fed source-contrasted diets. Estimates of genetic correlation between diets were the lowest for early weight (0.67 ± 0.12) and indicate differential abilities for initial growth between M and PB diets.