Genetic diversity of French rainbow trout lines

MOTS-CLES : genetic variability; selection ; aquaculture ; trout

Selection in aquaculture is a recent activity in comparison with terrestrial species. Rainbow trout lines have been selected in France from closed lines since the 1990s. Initial constitution then selection may have led to a decrease in genetic variability. This variability is a guarantee of long-term genetic progress and the capacity of lines to adapt to a changing environment.

In order to characterize genetic variabiltiy between different lines, principal component analysis and a measure of genetic differentiation (FST) may be used to assess whether populations are genetically close or not.

To characterize genetic variability within a line, two indicators are traditionnally used: efficient population size (Ne) and inbreeding. In a real population, all the individuals do not participate in the same way to the reproductive process and it is therefore not the total number of animals in a population that determine genetic variability.

Thus the efficient size of a population, Nen corresponds to the theoretical number of reproducing animals that would contribute equally to a population from a genetic point of view. The inbreeding coefficient represents the probability at a locus that the two alleles be identical since they come from a common ancestor. A rapid increase of the level of inbreeding that accompanies a loss in genetic diversity may lead to inbreeding depression, that is to negative effects on reproduction, survival or production performances of individuals.

The recent possibilities of high-throughput genotyping make it possible to revisit the analysis of genetic diversity, in particular by having a much more exhaustive vision of the entire genome and its evolution in comparison with classical pedigree analyses.

The objectives of this study are to evaluate genetic diversity of four commercial lines and two French experimental lines. To do this, between 32 and 49 lines were genotyped using the AxionTM trout genotyping array containing 57501 SNP (single nucleotide polymorphisms) including 38350 exploitable SNP after quality controls.

Our results show a moderate differentiation between commercial lines (FST de 0.08 à 0.15), equivalent to those estimated between bovine breeds for example.The efficient numbers are low in all lineages. A stabilization or even a slight increase in the effective numbers was, however, observed on the last generations of three of the commercial lines, probably in relation with the development of a mating optimization strategy in their selection programs. The average level of inbreeding determined on molecular information varies between 10% for an unselected line and 19.5% for a selected line. Recent inbreeding, accumulated over the last 3 generations, varied between 3 and 8% according to the line. The use of molecular information provides a different and more detailed vision of the use of pedigrees, notably revealing events prior to the rational management of lines such as founding effects during the creation of the populations that the lines originate from. This study also shows a strong heterogeneity of estimated inbreeding levels between one chromosome and another or between one region and another on the same chromosome. This could be associated to indirect effects of selection on a localized reduction of genetic variation in certain areas of the genome.

The objective is now threefold. The first step is to find out whether certain stages of the life cycle (fertilization, first feeding, etc.) are responsible for the loss of genetic variability in a non-selected population. Secondly, we will study the effect of inbreeding on fish performance, considering the variability in the level of inbreeding observed along the genome. Finally, we will propose an evolution of the management of selection programs to enhance and preserve genetic variability along the genome as much as possible.

Funding and Partnerships
This study was performed as part of a CIFRE PhD study supported by the SYSAAF and ANRT (n° ?) and the projects « 57K-Truite » and « SG-Truite » co-funded by FranceAgrimer (N°2015-0638), FEAMP (RFEA47 0016 FA 1000016) and the ANR (n°2017/0239).
The study was performed in partnership with four French rainbow trout breeders (Viviers de Sarrance, Les sources de l’Avance, Les fils de Charles Murgat et Bretagne Truite) with technical support from the SYSAAF (Syndicat des Sélectionneurs Avicoles et Aquacoles Français),