Detection and characterization of genomic regions associated to Dilated Cardiomyopathy in Great Danes

Abstract: Canine dilated cardiomyopathy (DCM) is a lethal myocardial disease prevalent in large and giant-sized dog breeds, including the Great Dane (GD). The goal of this study was to identify genetic risk factors for DCM and to contribute to a future development of DCM genetic markers in order to facilitate disease diagnosis and risk allele detection. Furthermore, in the long term, the identification of novel genes and pathways that may contribute for development of future therapies. Two different approaches of genome wide association study (GWAS) using data from ~170,000 Single Nucleotide Polymorphism (SNP) markers were conducted in order to identify disease-associated regions of the genome. A previous GWAS to this study used a different cohort of individuals and analytical software. This methodology detected six candidate regions on chromosomes 3, 16, 19, 29 and 30 (two associated regions on chr19). The current GWAS tested nine different models in a dataset containing 127 Great Danes (62 cases and 65 controls). The GenABEL package for R was used for the analysis. The elected model 3 included population stratification and had an inflation factor lambda (λ) equal to 1.04. Two regions on chromosome 19 overlapped with the first GWAS. Region 1: between 23-23.3 Mb containing two rRNAs; Region 2: between 44-46Mb containing three genes: SPOPL, NXPH2 and a LRP1B orthologue. No marker reached Bonferroni statistical significance, and the low power could be due to insufficient number of cases and controls. It is suggested that SNP typing of extra Great Danes and a better use of the available phenotype data would help to improve the statistical power. The correction for complex gene-gene interactions such as epistasis may be an interesting approach to detect novel candidates. The regions detected in the precedent GWAS were re-sequenced using Illumina Next generation sequencing technology in five animals (three controls and two cases), and the data was available for analysis in the current study. Six candidate non-synonymous mutations were detected in the DMXL2, TMC3, ZMAT4, MESDC2 genes and in a paralogue to RPL10A. DMXL2 encodes for a protein important for the Notch pathway in mammals. In humans, a SNP in this gene has been associated with ischemic stroke. The polymorphism in DMXL2 was confirmed through the genotyping of 319 individuals and the non-reference allele is slightly more frequent on verified cases as opposed to verified controls, however, the difference is not statistically significant (t-test p-value= 0.366; 2X3 Fisher test p-value= 0.258). Further analysis on both regions on chr19 should be carried out in order to detect possible novel mutations, copy number variants, genes and regulatory elements that could influence the DCM phenotype.