[Spontaneous models of human diseases in dogs: ichthyoses as an example]. / Modèles spontanés de maladies humaines chez le chien: exemple des ichtyoses.

Ichthyoses encompass a heterogeneous group of genodermatoses characterized by abnormal desquamation over the entire body due to defects of the terminal differentiation of keratinocytes and desquamation, which occur in the upper layer of the epidermis. Even though in humans more than 40 genes have already been identified, the genetic causes of several forms remain unknown and are difficult to identify in Humans. Strikingly, several purebred dogs are also affected by specific forms of ichthyoses. In the Golden retriever dog breed, an autosomal recessive form of ichthyosis, resembling human autosomal recessive congenital ichthyoses, has recently been diagnosed with a high incidence. We first characterized the disease occurring in the golden retriever breed and collected cases and controls. A genome-wide association study on 40 unrelated Golden retriever dogs, using the canine 49.000 SNPs (single nucleotide polymorphisms) array (Affymetrix v2), followed by statistical analyses and candidate gene sequencing, allowed to identify the causal mutation in the lipase coding PNPLA1 gene (patatin-like phospholipase domain-containing protein). Screening for alterations in the human ortholog gene in 10 autosomal recessive congenital ichthyoses families, for which no genetic cause has been identified thus far, allowed to identify two recessive mutations in the PNPLA1 protein in two families. This collaborative work between "human" and "canine" geneticists, practicians, histopathologists, biochemists and electron microscopy experts not only allowed to identify, in humans, an eighth gene for autosomal recessive congenital ichthyoses, but also allowed to highlight the function of this as-yet-unknown skin specific lipase in the lipid metabolism of the skin barrier. For veterinary medicine and breeding practices, a genetic test has been developed. These findings illustrate the importance of the discovery of relevant human orthologous canine genetic diseases, whose causes can be tracked in dog breeds more easily than in humans. Indeed, due to the selection and breeding practices applied to purebred dogs, the dog constitutes a unique species for unravelling phenotype/genotype relationships and providing new insights into human genetic diseases. This work paves the way for the identification of rare gene variants in humans that may be responsible for other keratinisation and epidermal barrier defects.

PNPLA1 mutations cause autosomal recessive congenital ichthyosis in golden retriever dogs and humans.

Ichthyoses comprise a heterogeneous group of genodermatoses characterized by abnormal desquamation over the whole body, for which the genetic causes of several human forms remain unknown. We used a spontaneous dog model in the golden retriever breed, which is affected by a lamellar ichthyosis resembling human autosomal recessive congenital ichthyoses (ARCI), to carry out a genome-wide association study. We identified a homozygous insertion-deletion (indel) mutation in PNPLA1 that leads to a premature stop codon in all affected golden retriever dogs. We subsequently found one missense and one nonsense mutation in the catalytic domain of human PNPLA1 in six individuals with ARCI from two families. Further experiments highlighted the importance of PNPLA1 in the formation of the epidermal lipid barrier. This study identifies a new gene involved in human ichthyoses and provides insights into the localization and function of this yet uncharacterized member of the PNPLA protein family.

Ancestral T-box mutation is present in many, but not all, short-tailed dog breeds.

Dogs differ greatly in their morphological characteristics including various tail phenotypes. Congenitally short-tailed dogs are present in many breeds; however, the causative mutation located in the T-box transcription factor T gene (C189G) had only been described in the bobtailed Pembroke Welsh Corgis. We investigated here the presence of the T gene mutation in 23 other breeds (360 dogs, including 156 natural short tailed) in which natural bobtailed dogs exist. In the 17 breeds in which the C189G mutation was observed, there was a perfect correlation between this mutation and the short-tail phenotype. However, 6 breeds did not carry the known substitution or any other mutations in the T gene coding regions. No dogs were found to be homozygous for the C189G mutation, suggesting that the homozygous condition is lethal. In order to study the effect of the T gene mutation on litter size, we compared the number of puppies born from short-tailed parents to that born from long-tailed parents. In the Swedish Vallhund breed, we observed a 29% decrease in the litter size when both parents were short tailed. Given that the T gene mutation is not present in all breeds of short-tailed dog, there must be yet other genetic factors affecting tail phenotypes to be discovered.