KRT5 in-frame deletion in a family of German Shepherd dogs with split paw pad disease resembling localized epidermolysis bullosa simplex in human patients.

Split paw pad disease is a scarcely defined phenotype characterized by skin lesions on the paw pads of dogs. We studied a family of German Shepherd dogs, in which four dogs developed intermittent paw pad lesions and lameness. The paw pads of two of the affected dogs were biopsied and demonstrated cleft formation in the stratum spinosum and stratum corneum, the outermost layers of the epidermis. Whole genome sequencing data from an affected dog revealed a private heterozygous 18 bp in frame deletion in the KRT5 gene. The deletion NM_001346035.1:c.988_1005del or NP_001332964.1:p.(Asn330_Asp335del) is predicted to lead to a loss of six amino acids in the L12 linker domain of the encoded keratin 5. KRT5 variants in human patients lead to various subtypes of epidermolysis bullosa simplex (EBS). Localized EBS is the mildest of the KRT5-related human diseases and may be caused by variants affecting the L12 linker domain of keratin 5. We therefore think that the detected KRT5 deletion in dogs represents a candidate causal variant for the observed skin lesions in dogs. However, while the clinical phenotype of KRT5-mutant dogs of this study closely resembles human patients with localized EBS, there are differences in the histopathology. EBS is defined by cleft formation within the basal layer of the epidermis while the cleft formation in the dogs described herein occurred in the outermost layers, a hallmark of split paw pad disease. Our study provides a basis for further studies into the exact relation of split paw pad disease and EBS.

Prospective evaluation of a patented DNA test for canine hip dysplasia (CHD).

Genetic testing has been propagated as a suitable means to specify individual risks for canine hip dysplasia (CHD). However, the current lack of validation of most genetic CHD tests has left dog owners and breeders in the dark about their practical utility. Therefore, the Society for German Shepherd Dogs (Verein für Deutsche Schäferhunde, SV) initiated a prospective study of 935 animals to assess independently the value of a genetic CHD test (European Patent Specification EP 2 123 777 B1) that was developed by Distl et al. (2009) on the basis of the SV animal stock. Dogs were followed-up for 3 years after birth, classified regarding their CHD phenotype using the scheme of the Fédération Cynologique Internationale, and genotyped for the 17 single nucleotide polymorphisms (SNPs) constituting the CHD test in question. Individual SNP genotypes were combined into animal-specific genomic breeding values (GBVs), calculated as the weighted sum of SNP-wise scores as laid down in the patent specification. Logistic regression analysis revealed that, unexpectedly, the odds ratio for CHD decreased, rather than increased, by a factor of 0.98 per unit increase of the GBV. Nevertheless, since this effect was not statistically significant (95% CI: 0.93-1.03), and the area-under-curve of the test was only 0.523, it must be concluded that the genetic test patented by Distl et al. (2009) is unsuitable for individual CHD risk assessment.

A deletion in the VLDLR gene in Eurasier dogs with cerebellar hypoplasia resembling a Dandy-Walker-like malformation (DWLM).

Dandy-Walker-like malformation (DWLM) is the result of aberrant brain development and mainly characterized by cerebellar hypoplasia. DWLM affected dogs display a non-progressive cerebellar ataxia. Several DWLM cases were recently observed in the Eurasier dog breed, which strongly suggested a monogenic autosomal recessive inheritance in this breed. We performed a genome-wide association study (GWAS) with 9 cases and 11 controls and found the best association of DWLM with markers on chromosome 1. Subsequent homozygosity mapping confirmed that all 9 cases were homozygous for a shared haplotype in this region, which delineated a critical interval of 3.35 Mb. We sequenced the genome of an affected Eurasier and compared it with the Boxer reference genome and 47 control genomes of dogs from other breeds. This analysis revealed 4 private non-synonymous variants in the critical interval of the affected Eurasier. We genotyped these variants in additional dogs and found perfect association for only one of these variants, a single base deletion in the VLDLR gene encoding the very low density lipoprotein receptor. This variant, VLDLR:c.1713delC is predicted to cause a frameshift and premature stop codon (p.W572Gfs*10). Variants in the VLDLR gene have been shown to cause congenital cerebellar ataxia and mental retardation in human patients and Vldlr knockout mice also display an ataxia phenotype. Our combined genetic data together with the functional knowledge on the VLDLR gene from other species thus strongly suggest that VLDLR:c.1713delC is indeed causing DWLM in Eurasier dogs.