Aquaculture Europe 2021

October 4 - 7, 2021

Funchal, Madeira

Add To Calendar 06/10/2021 14:50:0006/10/2021 15:10:00Europe/LisbonAquaculture Europe 2021GENETIC BASIS OF RESISTANCE TO VIRAL NERVOUS NECROSIS IN GILTHEAD SEA BREAM Sparus aurata AT THE LARVAL STAGEFunchal-HotelThe European Aquaculture Societywebmaster@aquaeas.orgfalseDD/MM/YYYYaaVZHLXMfzTRLzDrHmAi181982

GENETIC BASIS OF RESISTANCE TO VIRAL NERVOUS NECROSIS IN GILTHEAD SEA BREAM Sparus aurata AT THE LARVAL STAGE

 

S. Faggion1*, R. Franch1, M. Babbucci1, F. Pascoli2, G. Dalla Rovere1, L. Biasini2, S. Iori1, M. Caggiano3, H. Chavanne3, A. Toffan2, P. Carnier1, L. Bargelloni1

 

1 Department of Comparative Biomedicine and Food Science, University of Padova, Viale dell’Università, 16, 35020 Legnaro (PD), Italy

2 Division of Comparative Biomedical Sciences, OIE Reference Centre for viral encephalopathy and retinopathy, Istituto Zooprofilattico Sperimentale delle Venezie (IZSVe), Padova, Italy

3 Panittica Italia Società Agricola S.R.L., Strada del Procaccio 72016 Torre Canne di Fasano, Italy

Email: sara.faggion@unipd.it

 



Introduction

Nervous Necrosis Virus (NNV) is one of the major viral pathogens in aquaculture, affecting a wide range of fish species and causing high mortality rates. The gilthead sea bream (Sparus aurata) has long been considered resistant to NNV, until recently, when significant mortalities caused by a reassortant NNV strain were reported in sea bream hatcheries (Volpe et al. 2020). Since the larval stage is the most susceptible life-stage to NNV, vaccination is not a feasible option due to the immaturity of the immune system. Selective breeding to enhance resistance against the reassortant NNV strain might be a possibility as a disease preventive action. Here, we analysed for the first time the genetic basis of viral nervous necrosis (VNN) mortality in gilthead sea bream larvae and we assessed the accuracy in the genomic prediction of this trait.

Materials and methods

The experimental fish were generated at a commercial hatchery through controlled crosses in three independent full factorial matings (10 sires × 12 dams; 10 sires × 7 dams; 10 sires × 5 dams). At 15 days post-hatching (dph), larvae were transferred to the IZSVe experimental facility. At 19 dph, larvae were infected by immersion adding to the tank the reassortant strain VNNV/S.aurata/Farm1/461-1/Nov2014. The final infectious titre was verified by titration of the water (105.45 TCID50/ml). The challenge trial ended at day 9, when no more symptomatic/dying larvae were detected for at least 24 hours. Larvae showing symptoms of VNN infection and surviving individuals were collected for DNA analysis and recorded as 0 (asymptomatic) or 1 (symptomatic). The experimental infection protocol was evaluated by the IZSVe Animal Welfare Body and Ethics Committee (Opinion CE.IZSVE.3/2016 of 24/10/216) and subsequently approved by the Italian Ministry of Health (Law decree 101/2017-PR of 02/02/2017). All the experimental fish and their parents were genotyped using the Med_Fish SNP array, which contains over 27,000 SNPs for the gilthead sea bream (Peñaloza et al. 2021). Variance components for mortality was estimated using Bayesian procedures with a univariate sire-dam threshold model. Heritability was computed using the sire variance only, to avoid potential non-genetic maternal effects. Genome-wide association analysis (GWAS) was performed to test the association between VNN mortality phenotype and SNPs. Genomic prediction of VNN mortality was performed implementing three Bayesian regression models: Bayes B (BB), Bayes C (BC) and Bayesian Ridge Regression (BRR; GBLUP equivalent). Prediction performance was assessed by means of 5 independently-generated 5-fold cross validations (CV). In each CV, 80% of the data were used to train the model and 20% served as a validation set. Three metrics were used to evaluate model performance in classification: Matthews correlation coefficient (MCC), the area under the ROC curve (AUC) and accuracy (ACC). Pedigree indices were estimated using an animal model through 5-fold CV: in each CV, 80% of the data was used to estimate the indices of the remaining 20% of the animals. Performance of the indices in classification of VNN mortality was assessed using the same metrics used for genomic prediction (MCC, AUC, ACC).

Results

First symptoms of VNN infection were detected at day 6 post-challenge, followed by a peak of two day; then mortality sharply decreased up to day 8. A total of 1184 individual larvae were collected (513 dying and 671 survivors). Genomic DNA was extracted from the tissue of 1044 larvae and 54 parents and used for individual genotyping. A total of 974 larvae, 47% symptomatic and 53% asymptomatic, were successfully genotyped and parentage assignment to a unique parental pair was achieved for all the 974 fish. Individuals were allocated to 160 families, with a number of offspring per family ranging from 1 to 55. After removing one sire and one dam that generated only one offspring, 972 individuals from 28 sires and 22 dams were retained. Overall, 26,591 SNPs with MAF > 0.05 were scored. The estimate of heritability for VNN mortality was moderate (h2 = 0.1921; 95% highest posterior density intervals: 0.0006, 0.5790), with a probability being greater than 0.2 equal to 0.49. Classification of the observed VNN mortality using the genomic prediction of the phenotype of mortality as classifier was significantly more accurate than random guessing of the classes, with consistent results across Bayesian models (Table 1); using the pedigree indices to classify the mortality phenotype resulted in similar performances (AUC = 0.5875, ACC = 0.5798, MCC = 0.2550). The GWAS failed to identify any genome-wide QTL for VNN mortality overcoming the significance threshold.

Discussion

VNN is an emerging threat for gilthead sea bream hatcheries and this is the first study that explored the genetic basis of resistance to VNN in this species. Experimental infections in early developmental stages are scarcely reported, especially with adequate sample size to estimate variance components and genetic parameters. The estimate of heritability for VNN mortality suggests the feasibility of selective breeding programmes for increased resistance to VNN of fish larvae/juveniles, overcoming the problem of vaccination. The practical exploitation of genomic information due to the availability of genome-wide dense marker panels might offer the opportunity of developing prediction tools for the studied trait.

References

Peñaloza C, Manousaki T, Franch R, Tsakogiannis A, Sonesson A, Aslam ML, Allal F, Bargelloni L, Houston RD, Tsigenopoulos CS. (2021). Development and testing of a combined species SNP array for the European seabass (Dicentrarchus labrax) and gilthead seabream (Sparus aurata). Genomics, in press.

Volpe E, Gustinelli A, Caffara M, Errani F, Quaglio F, Fioravanti ML, Ciulli S. (2020). Viral nervous necrosis outbreaks caused by the RGNNV/SJNNV reassortant betanodavirus in gilthead sea bream (Sparus aurata) and European sea bass (Dicentrarchus labrax). Aquaculture, 523: 735155.