Severity of Mitral Valve Degeneration Is Associated with Chromosome 15 Loci in Whippet Dogs

Abstract

Mitral valve degeneration (MVD) is the most common form of heart disease in dogs, frequently leading to left-sided congestive heart failure and cardiac mortality. Although breed-specific disease characteristics and overrepresentation point towards a genetic origin for MVD, a causative mutation and complete molecular pathogenesis are unknown. Whippet dogs are overrepresented in incidence of MVD, suggesting an inherited component in this breed. Expressivity of this condition is variable with some dogs showing evidence of more severe disease at earlier ages than other dogs. This phenomenon makes a traditional case versus control genetic study prone to phenotyping error. This study sought to avoid these common pitfalls by identifying genetic loci associated with severity of MVD in Whippets through a genome-wide association study (GWAS). 138 Whippet dogs were characterized for MVD by echocardiographic examination and a novel disease severity score was developed and adjusted for age in each subject. Single nucleotide polymorphism (SNP) genotype data (170k Illumina CanineHD SnpChip) was obtained for DNA isolated from blood of each study subject. Continuous variable genome wide association was performed after correction for population stratification by efficient mixed model association expedited (EMMAX) in 130 dogs. A genome wide significant association was identified on chromosome 15 (peak locus 57,770,326; Padj = 0.049) and secondary loci of suggestive association were identified on chromosome 2 (peak locus 37,628,875; Padj = 0.079). Positional candidate genes were identified within the primary and secondary loci including follistatin-related protein 5 precursor (FSTL5) and Rho GTPase-activating protein 26 (ARHGAP26). These results support the hypothesis that severity of MVD in whippets has a genetic basis and warrants further study by either candidate gene sequencing or next-generation techniques.

Fig 1. A two-dimensional (A) and two-dimensional with color Doppler (B) echocardiogram from a dog are displayed.

The view is left apical four-chamber showing the left ventricle (LV), left atrium (LA), right ventricle (RV) and right atrium (RA). The mitral valve is denoted by the white arrow in image A. This image represent a dog with severe mitral valve prolapse as it is observed to buckle back into the LA during systole. Severe mitral valve regurgitation and chamber enlargement are also apparent, yielding a disease severity score of 6 in this dog.

Fig 2. A box and whiskers plot displays the median and range of ages across each severity score group that was present within this study.

Fig 3

A) Q-Q plot from the continuous variable association with λ = 1.07 demonstrating mild population stratification as seen by the continual deviation from the line (consistent obtained P values deviate from those which are expected). B) Q-Q plot from the efficient mixed model association expedited (EMMAX) analysis in which the deviation from expected P values is corrected yielding a single, pronounced deviation from expect P values, consistent with the significant association results obtained.

Fig 4

A) Manhattan plot demonstrating the genotype association to MVD severity obtained by the mixed linear model, additive, continuous variable analysis (EMMAX). The most significant regions of interest are noted on chromosome 15 and 2. Raw (P_raw) and Bonferroni corrected P values (P_adj) are displayed for the top 2 SNPs. The red lines denote the level of genome wide significance (Bonferroni). B) The same manhattan plot is focused in on a region of chromosome 15 to demonstrate the flanking region of the most significant SNP. C) The same manhattan plot is focused in on a region of chromosome 2 to demonstrate the flanking region of the secondary peak of significance.