Whole genome re-sequencing reveals recent signatures of selection in three strains of farmed Nile tilapia (Oreochromis niloticus)

Abstract

Nile tilapia belongs to the second most cultivated group of fish in the world, mainly because of its favorable characteristics for production. Genetic improvement programs and domestication process of Nile tilapia may have modified the genome through selective pressure, leaving signals that can be detected at the molecular level. In this work, signatures of selection were identified using genomewide SNP data, by two haplotype-based (iHS and Rsb) and one F ST based method. Whole-genome re-sequencing of 326 individuals from three strains (A, B and C) of farmed tilapia maintained in Brazil and Costa Rica was carried out using Illumina HiSeq 2500 technology. After applying conventional SNP-calling and quality-control filters, ~ 1.3 M high-quality SNPs were inferred and used as input for the iHS, Rsb and F ST based methods. We detected several candidate genes putatively subjected to selection in each strain. A considerable number of these genes are associated with growth (e.g. NCAPG, KLF3, TBC1D1, TTN), early development (e.g. FGFR3, PFKFB3), and immunity traits (e.g. NLRC3, PIGR, MAP1S). These candidate genes represent putative genomic landmarks that could be associated to traits of biological and commercial interest in farmed nile tilapia. Nile tilapia (Oreochromis niloticus) is a teleost fish of the Cichlidae family native to Africa and the Middle East. The geographic range of the species extends from 8°N to 32°N 1. The first record of domestication is dated around 3,500 years ago as evidenced in paintings at the Theban tombs in Egypt 2. Nowadays, this species is the second most cultivated group of fish in the world 3. Favorable characteristics for production include rapid growth, adaptability to different culture conditions, tolerance to high densities, disease resistance, easy reproduction, and tolerance to low concentrations of oxygen 4. Genetic improvement programs (GIPs) for Nile tilapia began in 1988 as an approach to counteract the production decrease generated by introgressions with Mozambique tilapia (Oreochromis mossambicus) 5,6. Since then, nearly twenty GIPs have been established for Nile tilapia around the world 7,8. GIPs aim to improve traits of commercial interest, such as growth rate, disease resistance, cold and salinity tolerance 7. The GIFT (Genetic Improvement of Farmed Tilapia) 9 Nile tilapia strain was developed by the ICLARM (International Centre for Living Aquatic Resources Management, now the WorldFish Center), in collaboration with the Norwegian Institute of Aquaculture Research (AKVAFORSK, now NOFIMA Marin) 1. The implementation of GIPs for the GIFT population has been successful, because growth rate in Nile tilapia has doubled in five generations, showing that this species had a positive response to selection 1 .

Genomics to accelerate genetic improvement in tilapia

“…Lin et al (2016) detected that the phenotypic variation explained by QTL associated with growth-related traits under saline conditions was lower than 3%, with some important genomic regions found in LG12, LG20 and LG22. Evidence of the polygenic nature underlying harvest weight and fillet yield was provided by a GWAS performed using genotypes from a 50K SNP panel in a commercial Nile tilapia population from Costa Rica (Yoshida et al, 2019b), which is in agreement with previous studies for growthrelated traits in other fish species (Tsai et al, 2015;Yoshida et al, 2017b;Reis Neto et al, 2019). In addition, the same authors tested the use of genotype imputation from low-to high-density SNP panels, suggesting a reduction of genotyping cost up to 69%, depending on the breeding population size, without considerable losses in accuracy of genomic predictions when compared against the use of true dense SNP genotypes (Yoshida et al, 2019b).…”

“…Evidence of the polygenic nature underlying harvest weight and fillet yield was provided by a GWAS performed using genotypes from a 50K SNP panel in a commercial Nile tilapia population from Costa Rica (Yoshida et al, 2019b), which is in agreement with previous studies for growthrelated traits in other fish species (Tsai et al, 2015;Yoshida et al, 2017b;Reis Neto et al, 2019). In addition, the same authors tested the use of genotype imputation from low-to high-density SNP panels, suggesting a reduction of genotyping cost up to 69%, depending on the breeding population size, without considerable losses in accuracy of genomic predictions when compared against the use of true dense SNP genotypes (Yoshida et al, 2019b). Recent studies have also demonstrated the absence of major QTL for resistance against S. agalactiae, and also that genomicbased approaches outperform conventional pedigree-based methods for the genetic evaluation of this trait in Nile tilapia (Lu et al, 2020) and hybrid red tilapia (Sukhavachana et al, 2020).…”

“…In aquaculture, fillet yield is a key trait, but difficult to select for (Vandeputte et al, 2020). However, a large phenotypic variation, ranging from 32 to 45%, and low (0.06 AE 0.04) to medium (0.25 AE 0.07) values of heritability have been reported for fillet yield in Nile tilapia, suggesting that this trait can be improved through selection (Rutten et al, 2005;Gjerde et al, 2012;Joshi et al, 2018bJoshi et al, , 2020aThodesen (Da-Yong Ma) et al, 2011;Yoshida et al, 2019b). However, fillet-related traits are always laborious and costly to measure; furthermore, these traits cannot be recorded on alive breeding candidates (Gjerde et al, 2012a).…”

“…spp., including regions in LG1, LG3, LG20 and LG23 (Lee et al, 2004;Shirak et al, 2006;Cnaani et al, 2008;Eshel et al, 2014;Palaiokostas et al, 2015;Conte et al, 2017;C aceres et al, 2019a, 2019b). In addition, several genes have been suggested to have a role in sex determination in the tilapia genus, including Wilms tumor suppressor protein 1b (wt1b), cytochrome P450 of family 19 subfamily A member 1 (cyp19a), the anti-M€ ullerian hormone (Amh) and Doublesex/Mab-3 Related Transcription Factor 2 (dmrt2) (Lee et al, 2004;Shirak et al, 2006;Cnaani et al, 2008;Li et al, 2015aLi et al, , 2015bSun et al, 2017;C aceres et al, 2019a, 2019b. The multiple regions identified to have a role in sex determination in tilapia indicate that this is most likely a complex trait, which can also be modulated from environmental stimulus.…”

“…Identifying causative mutations will lead to a better understanding of biological mechanisms behind QTL (Meuwissen et al, 2013). Previous studies used WGS information to understand the evolutionary diversification of African cichlid fish (Brawand et al, 2015), identify signatures of selection (Hong Xia et al, 2015;C adiz et al, 2020) and sex determination regions through GWAS in tilapia (C aceres et al, 2019a, 2019b). In contrast, genomic prediction studies using WGS in tilapia are still not avaliable.…”