Résumé: Measuring individual feed intake of fish in farms is complex and precludes direct selective breeding for feed conversion ratio (FCR). Here, we estimated the individual FCR of 588 sea bass using individual rearing under restricted feeding. These fish were also phenotyped for their weight loss at fasting and muscle fat content as possible indirect indicators of FCR. The 588 fish were from a full factorial mating between parental lines divergently selected for high (F+) or low (F-) weight loss at fasting. The pedigree was known back to the great grandparents. A subset of 400 offspring and their ancestors were genotyped for 1,110 SNPs, which allowed estimating the genomic heritability of traits. Individual FCR and growth rate in aquarium were both heritable (genomic h² = 0.47 and 0.76, respectively) and strongly genetically correlated (-0.98), meaning that under restricted feeding, faster growing fish were more efficient. FCR in aquariums was significantly better for fish with two F- parents (1.38), worse for fish with two F+ parents (1.51) and intermediate (1.46) for crossbred fish (F+/F- or F-/F+). Muscle fat content was positively genetically correlated to growth rate in aquarium and during fasting. Thus, higher growth rate in aquariums, lower weight loss at fasting and fat content are all traits that could improve FCR in aquarium. Improving these traits would also improve FCR of fish in normal group rearing conditions, as we showed that groups composed of fish with good individual FCR were significantly more efficient in groups. The FCR of groups was also better when the fish composing the groups had, on average, lower estimated breeding values for growth rate during fasting (losing less weight). Thus, FCR in aquarium and weight loss at fasting are both promising to improve FCR of fish in groups. Finally, we showed that the reliability of estimated breeding values was higher (from +10% to +125%) with single-step genomic BLUP than with pedigree-based BLUP, showing that genomic data would enhance the accuracy of EBV prediction o in selection candidates from a limited number of sibs individually phenotyped for FCR in aquariums.

Résumé: Feed conversion ratio (FCR), the ratio between feed intake and body weight gain, is of major interest for im-proving aquaculture sustainability through reduced feed costs and environmental impacts. Demonstrating whether FCR measured in juvenile fish is an accurate predictor of their performance during the whole rearing period is critical to developing genetic improvement programs for this trait. This is especially true for estimates obtained in individually reared fish, for which this has high implications regarding the size of the necessary rearing structures. We obtained individual FCR from 30 male Nile tilapia Oreochromis niloticus from the GIFT strain individually reared in a recirculating system, from 36 to 260 g mean weight. They were fed twice a day and uneaten pellets were counted every day to determine the feed intake of each fish. Individual growth was monitored every week. Feed conversion ratio was estimated over two-week periods and over the whole rearing period (210 days). Phenotypic correlations between the two-week FCRs and global FCR estimations were mostly significant (ranged from 0.38 to 0.64). A significant phenotypic correlation between growth and FCR was also found: faster -growing fish had a better (lower) FCR. Individual breeding values for global FCR were estimated using FCR phenotypes from the present study and previously published heritabilities for FCR in Nile tilapia. Potential estimated genetic gain for global FCR was 2.2% per generation with 50% selection intensity. When selecting fish on their FCR from only a two-week period, approximately 50% of the reference genetic gain could be obtained with the same selection intensity. FCR measured during a two-week period at juvenile stage could be a moderately accurate approximation of the whole rearing period FCR, and could be used as a lower cost criterion to select for FCR in future genetic improvement programs using individual rearing of fish.

Résumé: Sea bass is a major species in Mediterranean aquaculture, and is now being subject to selective breeding programs for faster growth. In terrestrial species, it was demonstrated that fast growth may be linked to a correlated degradation of fitness traits. In this experiment, we evaluated 600 young sea bass from a factorial mating of 76 sires and 13 dams. The sires were from four genetic groups, wild (W), domesticated (D), and selected for growth (2 groups, M and P). The 600 offspring were submitted to two acute confinement stress challenges at 6 weeks intervals, and plasma cortisol at one hour post stress was measured. The same fish were also submitted to two swimming challenges at a 5 days interval, where the maximum sustained swimming speed (Umax) of each fish was evaluated. Parentage was assessed by genotyping of 12 microsatellites. 554 fish had both valid parentage and phenotypes. Cortisol had a low repeatability (r = 0.30 between the two successive measurements) while repeatability was moderate for Umax (r = 0.62). However, genetic correlations between successive measurements were very high (> 0.96) for both traits, indicating that successive measurements were related to the same trait. Heritability was moderate for mean post-stress cortisol (h2 = 0.34 ± 0.09) and Umax (h2 = 0.48 ± 0.08). When Umax was expressed in m.s− 1, it was negatively correlated to cortisol (rA = − 0.48 ± 0.08) and weakly correlated to body weight (rA = 0.12 ± 0.16), but figures changed when it was expressed in Body Lengths.s− 1(h2 = 0.55 ± 0.08, rA = − 0.10 ± 0.19 with cortisol and rA = − 0.64 ± 0.07 with body weight, respectively). Cortisol was moderately negatively correlated with body weight (rA = − 0.36 ± 0.18). The four lines did not differ for cortisol or Umax, but when Umax was expressed in BL.s− 1 it tended to be lower in the two selected lines – which were also significantly larger. However, this is likely due to a phenotypic decrease of relative Umax with increasing body size. We conclude that selection for growth and/or domestication should not impact maximum sustained swimming speed in the European sea bass, but may tend to favour animals with low cortisol responsiveness. These traits could be used to orientate functional capabilities other than productivity in sea bass.