Research article

Genetic diversity of canine olfactory receptors

Stéphanie Robin¹ , Sandrine Tacher² , Maud Rimbault¹ , Amaury Vaysse¹ , Stéphane Dréano¹ , Catherine André¹ , Christophe Hitte¹ and Francis Galibert¹

¹Institut de Génétique et Développement de Rennes, CNRS UMR 6061, Université de Rennes 1, 2 Avenue du Professeur Léon Bernard, CS 34317, 35043 Rennes, France

²Systèmes d'Elevage Nutrition Animale et Humaine, INRA UMR 1079, Domaine de la Prise, 35590 Saint Gilles, France

author email corresponding author email

BMC Genomics 2009, 10:21doi:10.1186/1471-2164-10-21

Published:	14 January 2009

Abstract

Background

Evolution has resulted in large repertoires of olfactory receptor (OR) genes, forming the largest gene families in mammalian genomes. Knowledge of the genetic diversity of olfactory receptors is essential if we are to understand the differences in olfactory sensory capability between individuals. Canine breeds constitute an attractive model system for such investigations.

Results

We sequenced 109 OR genes considered representative of the whole OR canine repertoire, which consists of more than 800 genes, in a cohort of 48 dogs of six different breeds. SNP frequency showed the overall level of polymorphism to be high. However, the distribution of SNP was highly heterogeneous among OR genes. More than 50% of OR genes were found to harbour a large number of SNP, whereas the rest were devoid of SNP or only slightly polymorphic. Heterogeneity was also observed across breeds, with 25% of the SNP breed-specific. Linkage disequilibrium within OR genes and OR clusters suggested a gene conversion process, consistent with a mean level of polymorphism higher than that observed for introns and intergenic sequences. A large proportion (47%) of SNP induced amino-acid changes and the Ka/Ks ratio calculated for all alleles with a complete ORF indicated a low selective constraint with respect to the high level of redundancy of the olfactory combinatory code and an ongoing pseudogenisation process, which affects dog breeds differently.

Conclusion

Our demonstration of a high overall level of polymorphism, likely to modify the ligand-binding capacity of receptors distributed differently within the six breeds tested, is the first step towards understanding why Labrador Retrievers and German Shepherd Dogs have a much greater potential for use as sniffer dogs than Pekingese dogs or Greyhounds. Furthermore, the heterogeneity in OR polymorphism observed raises questions as to why, in a context in which most OR genes are highly polymorphic, a subset of these genes is not? This phenomenon may be related to the nature of their ligands and their importance in everyday life.