The uprising of diseases is one of the main challenges in the current Nile tilapia production requiring alternatives to improve the resistance of animals. Genome-wide association studies (GWAS) are used to map QTLs (quantitative trait loci), allowing the identification of associations between desired traits and available genetic markers such as SNPs (single-nucleotide polymorphism). In this sense, the aim of this study was to perform GWAS using a 60K SNP array for animals challenged for Francisella orientalis.
About 1,000 animals from 112 families were challenged for F. orientalis. Approximately 334 animals were randomly distributed among three 2000-L experimental tanks for the challenge, evenly distributing the number of animals per family. Before the challenge, the animals were weighed and inoculated with a pre-defined lethal dose of 0.1ml of inoculum per 10g of body weight. After inoculation, time to death (TD) was recorded during 15 days. Approximately 5 animals per family (~500 animals) had genomic DNA extracted and were genotyped using the Axiom™ TilShrv1 array with 60K SNPs. We used the single-step genomic best linear unbiased predictor (ssGBLUP) method that uses both information from genotyped and non-genotyped animals to obtain the solution of SNPs and obtain a proportion of variance explained by each marker windows. We also exploited the genomic regions associated to resistance (TD) looking for genes close to this significant SNPs. Results showed significant genetic variance for F. orientalis resistance with a heritability of 0.282 ± 0.08 confirming the possibility for genetic selection for resistance in this population. The GWAS results showed genomic regions possibly associated to resistance for Francisella orientalis in Nile tilapia in chromosomes 1, 6, 9 and 15 (Fig. 1). These regions showed SNPs with proportion of genetic variance explaining higher than 1%. In addition, it was possible to observe a polygenic architecture for this trait as positive associations of small magnitude were found in different chromosomes. In this sense, genomic selection is more recommended to improve this trait. We found 82 genes close (100,000 bp) to the significant SNPs. Some of these genes are related to important biological processes, for example: FAT atypical cadherin 1a (a protein receptor for Gram-positive bacteria required for entry into host epithelial cells) and Kruppel-like factor 1 (participates in the development and homeostasis maintenance of immune systems). In conclusion, the resistance for F. orientalis presents a polygenic architecture requiring genomic selection to be improved. We also found relevant genes possibly associated to this resistance in Nile tilapia.