GENETIC VALIDATION OF THE XRAY METHOD FOR PHENOTYPING FEED INTAKE IN ATLANTIC SALMON

Introduction

Feed plays a crucial role in the economic and environmental performance of Atlantic salmon production with the cost of feed accounting for over 50% of the total production cost (http://www.fiskeridir.no) and 73-80% of the carbon footprint (Winther et al., 2020). Improving the feed efficiency of Atlantic salmon has potential to be a win-win situation for improving profitability and resource efficiency. Direct selection for improved feed efficiency requires records of feed intake on replicated family groups in tanks through daily recording of the given and the wasted feed in each tank. Due to the large number of tanks required this is both costly and labour demanding and accesses half of the genetic variation. An alternative method is to feed with pellets containing radio opaque markers and use Xray radiography to count the total markers consumed by individual fish. The Xray method for daily feed intake has been shown to be heritable in rainbow trout (Kause et al., 2006) and chinook salmon (Walker et al., 2012), however estimates for Atlantic salmon are lacking. The advent of genomic selection and advances in digital radiography and image analysis offer the possibility to improve the feasibility of obtaining individual genomic breeding values for feed intake and feed efficiency of the breeding candidates. The objective of the present studies was to 1) estimate genomic parameters for Xray feed intake records and 2) validate individual Xray feed intake records with the “gold standard” tank feeding method at family level in replicate tanks.

Materials and methods

A total of 2450 Atlantic salmon parr from 35 nucleus families of the MOWI Genetics AS (MOWI ASA, Øyerhamn, Norway) were PIT tagged and transported to Nofima’s Research Station for Sustainable Aquaculture, Sunndalsøra. All fish were genotyped with a customized 65K SNP chip. The PIT-tagged fish were sorted by family and divided into two parallel trials; the first with a total of 700 fish equally split over two replicated tanks with an equal number of fish from all families in each of the two tanks (Xray trial) and the second trial with 1750 fish split into 70 tanks with two tanks per family and 25 fish per family (tank feeding trial). The tanks were divided in two different rooms in the same building.

In the Xray trial, fish were acclimated to the two tanks for three weeks. During this period fish were transitioned to a feeding regime where they received the entire daily feed ration over a six hour period. A day prior to Xray, the feed was switched to an identical dietary formulation including glass beadlets. After consuming the daily ration fish were anesthetised and Xrayed. Beadlets were counted using a customised image analysis software developed by Nofima. Number of recorded beadlets per fish were converted to the amount of feed intake (g) per fish (FIXray) using a calibration standard based on feed samples of known mass. FIXray was analysed by a mixed linear animal model using DMU (Madsen and Jensen, 2014) including the fixed effects of tank (with two levels), a fixed regression on time within the Xray recording day, a random animal additive genetic effects captured by the genomic relationship matrix. The heritability was estimated from variance components (h2 = Va/Vp). The average feed intake per fish, per family and tank (FIXrayF) was analysed by a linear model including the fixed effect of tank (with two levels) and a random effect of family. The variance components for the family effect was expressed relative to the total variance (H2 = Vf/Vp). 

In the tank feeding trial, daily feed rations and refusals were collected over 50 days. At day 50 the trial, fish were euthanized and recorded for body length and weight. The average family biomass per tank was modelled as a function of time throughout the trial and daily feed intake records (FITF) were matched to the Xray feed intake records of the Xray trial (FIXray) on an equal average family body weight basis. The accumulated feed intake per tank (FITF) was analysed by a linear model including the fixed effect of room (with two levels) and the random effect of family (H2 = Vf/Vp). A bivariate analysis was run FIXrayF and FITF with the fixed effects of tank (with two levels) for FIXray and the fixed effect of room for FITF, and the random effect of family for both traits. The covariance between random effects of family used to compute the correlation between for families between methods. 

Results

Results of the feed intake at individual and family level are presented in Table 1 along with their respective intraclass correlation coefficients. The heritability (h2) of FIXray was 0.22 ± 0.05. The family means of FIXray explained 38% of the total variation which was very similar to the family effect (37%) of the total variation in the accumulated FITF. Crucially, the rank correlation between the families FIXrayF and FITF was high and positive (0.85 ± 0.25).

Conclusion

For the first time, significant heritability of feed intake using the Xray method has been reported in Atlantic salmon. Validation at the family shows that a substantial proportion of the variation in feed intake among the families are co-explained by both methods. This demonstrates the feasibility of the Xray method for recording feed intake in Atlantic salmon and lays down the foundation for research into feed efficiency in Atlantic salmon. 

Funding

The Xray trial was developed by the Nofima internal project “PrecisionVision”, the gold standard trial was part of the EU project “NewTechAqua” (H2020 No 862658) and the EU project AquaImpact  (H2020 No 818367).

References

Kause, A., Tobin, D., Dobly, A., Houlihan, D., Martin, S., Mäntysaari, E.A., Ritola, O., Ruohonen, K., 2006. Recording strategies and selection potential of feed intake measured using the X-ray method in rainbow trout. Genet. Sel. Evol. 38, 389–409. https://doi.org/10.1051/gse:2006011

Madsen, P., Jensen, J., 2014. A user’s guide to DMU, version 6, release 5.0. Cent. Quant. Genet. Genomics Dept. Mol. Biol. Genet. Univ. Aarhus Res. Cent. Foulum Box 50, 8830 Tjele Denmark.

Walker, S.P., Ingram, M., Bailey, J., Dodds, K.G., Fisher, P.J., Amer, P.R., Symonds, J.E., 2012. Chinook salmon (Oncorhynchus tshawytscha) feed conversion efficiency : evaluation and potential for selection. Proc. New Zeal. Soc. Anim. Prod. 72, 227–230.

Winther, U., Hognes, E.S., Jafarzadeh, S., Ziegler, F., 2020. Greenhouse gas emissions of Norwegian seafood products in 2017.