Introduction
Cardiomyopathy syndrome (CMS) is a severe viral cardiac disease of Atlantic salmon caused by piscine myocarditis virus (PMCV). The disease typically manifests several months after sea transfer, during the second year of production when fish reach approximately 3.6 kg. Clinical signs are often absent until mortality begins, resulting in compromised fish health and welfare as well as substantial economic losses for producers. CMS is characterized by a slow progression of pathology, and even minor stressors, such as treatment for sea lice infestation, may trigger mass mortality events [1]. Histopathological examination reveals severe inflammation and necrosis primarily affecting the endocardium and spongy myocardium of the atrium and ventricle [2], accompanied by extensive infiltration of mononuclear immune cells, including lymphocytes and macrophages [3]. Estimates of heritability for resistance to CMS vary widely depending on phenotypic definitions (e.g., survival, viral load, or histology), environmental conditions (field versus controlled challenge), and population structure. Previous studies have identified genomic regions on chromosomes 12, 23 and 27 associated with genetic variation in CMS resistance, with the major quantitative trait locus (QTL) on chromosome 27 explaining approximately 30% of the genetic variance [4]. In the present study, we aimed to fine-map these genomic regions, with primary focus on the QTL located on chromosome 27, and to identify causative genes and/or mutations underlying host resistance to CMS.
Materials and Methods
The study population originated from the MOWI ASA breeding nucleus and included four year classes (YC2016, YC2017, YC2019, YC2021). Individuals were selected for whole-genome resequencing (WGR) to fine-map previously identified QTL regions and for functional validation using transcriptomic (atrium) and proteomic (plasma) analyses. WGR was conducted on YC2016–YC2019, while YC2021 was used for expression analyses. All individuals had pedigree data, CMS phenotypes (challenge or field), and genotypes from the NOFSAL03 (~55K SNPs) array. Sampling was stratified by genomic breeding values (gEBV) and haplotype status, limiting relatedness (≤4 per sire, ≤2 per dam). Individuals were classified into three groups resistant (FUL; gEBV >1.25 SD, two favorable haplotypes), intermediate (HET; >0.625 SD, one copy), or susceptible (DEL; <1.25 SD, no copies). A total of 157 individuals were selected for WGR and 30 (n=10 per group) from YC2021 for transcriptomic and proteomic analyses. Sequencing (150 bp paired-end) was performed on the BGISEQ platform. Reads were quality-filtered (using Trimmomatic), aligned to ICSASG_v2 (using BWA-MEM), and variants called using SAMtools/bcftools, and variant annotations via Ensembl VEP. Transcriptomic data were generated on Illumina NovaSeq 6000, processed with STAR, and analysed with edgeR. Plasma proteomics was conducted using Q-Exactive Orbitrap, with analysis in MaxQuant and statistical evaluation in Perseus and Unscrambler. Genome-wide association analysis using sequence-derived variants was performed using linear mixed model in GCTA [5] with "--mlma-loco" function, including year class as a fixed effect and additive genetic effects based on a genomic relationship matrix: , where is the vector of phenotypic records, is the overall mean, and represents fixed effects with corresponding design matrix . The vector denotes additive genetic effects , with as the genomic relationship matrix and design matrix , while e represents residual effects .
Results and Discussion
Alignment of sequence reads within the target QTL region on chromosome 27 (10–12 Mbp) identified 8,813 variants, including five high-impact SNPs spanning approximately 13 genes. Among these, two immune-related genes, MR1 and HA1F, emerged as strong candidates. Genome-wide association analysis reinforced a robust signal in this region; however, inclusion of sequence-derived variants did not further increase statistical significance beyond that captured by the SNP array. Moreover, high-depth (~100×) resequencing comparing FUL and DEL groups revealed reduced or absent read alignments in specific segments of the susceptible DEL group relative to resistant FUL individuals, indicating the presence of a structural variant, most likely a deletion spanning exonic regions of the HA1F gene. This putative deletion was experimentally validated at both genomic and transcriptomic levels using a targeted PCR assay on 30 individuals (10/group), which showed complete absence of amplification in DEL individuals. Additionally, differential expression analysis identified MR1 and HA1F as the only significantly regulated genes between FUL and DEL groups. Specifically, MR1 was strongly upregulated in susceptible fish, whereas HA1F exhibited markedly reduced expression. Proteomic analyses corroborated these findings, demonstrating significantly elevated levels of the MR1 protein product (A0A1S3PYN4) in the DEL group. The concordant evidence from fine-mapping, structural variant detection, transcriptomics, and proteomics strongly supports a central role for MR1 and HA1F in resistance to CMS. These results suggest that dysregulation of key immune pathways, potentially driven by structural variation, may impair host defence mechanisms, with reduced HA1F expression likely contributing to compromised immunity in susceptible individuals.
Figure 1: A) Copy number variation analysis identified a putative deletion within the QTL region on chromosome 27 in the DEL group. B) Integrative Genomics Viewer visualization shows near-complete loss of read coverage across the region in DEL compared to FUL. C) PCR validation confirms the deletion, with no amplification in DEL group D) Transcriptome analysis revealed concordant differentially expressed genes with fine-mapping results, supporting the findings. />
Acknowledgment
The results produced in current study are a part of "CMS-Edit" and the "Genomics4Robust" projects. CMS-Edit project was funded by The Research Council of Norway under grant agreement number 294504 while Genomics4Robust project is an internally funded project from Nofima AS under project number 13281.
References
1. Garseth Å H, et al.: Cardiomyopathy syndrome in Atlantic salmon Salmo salar L.: A review of the current state of knowledge. J Fish Dis 2018, 41(1):11–26. 2. Rodger HD, et al.: Clinical cardiomyopathy syndrome in Atlantic salmon, Salmo salar L. J Fish Dis 2014, 37(10):935–939. 3. Timmerhaus G, et al.: Comparison of Atlantic salmon individuals with different outcomes of cardiomyopathy syndrome (CMS). BMC Genomics 2012, 13(1):205. 4. Boison S, et al.: QTLs Associated with Resistance to Cardiomyopathy Syndrome in Atlantic Salmon. Journal of Heredity 2019, 110(6):727–737. 5. Yang J, et al.: GCTA: a tool for genome-wide complex trait analysis. Am J Hum Genet 2011, 88(1):76–82.