NEW YORK (GenomeWeb) – Researchers have tied more than 200 regions of the dog genome to their domestication.
Dogs were domesticated from gray wolves between 10,000 years and 40,000 years ago, likely in Eurasia, though there is still debate on when and where it took place, and modern breeds were developed more recently, about 300 years ago.
A number of traits shared among domesticated animals — including increased tameness, smaller brain size, and floppy ears — are absent among their wild relatives, a pattern that has been dubbed the domestication syndrome.
A University of Michigan-led team of researchers conducted a whole-genome SNP analysis of 43 village dogs and 10 wolves. By focusing on village dogs, which never underwent breed selection, the researchers hoped to be better poised to sniff out signals of domestication rather than breed selection.
As they reported last night in BMC Biology, the researchers unearthed nearly 250 candidate domestication regions. These regions encompass genes like RAI1 that are associated with neural crest development, which has been hypothesized to play a role in domestication.
"The neural crest hypothesis posits that the phenotypes we see in domesticated animals over and over again … can be explained by genetic changes that act in a certain type of cell during development called neural crest cells, which are incredibly important and contribute to all kinds of adult tissues," first author Amanda Pendleton, a postdoc at UMich, said in a statement.
Previous studies have homed in on dozens of potentially domestication-linked sites in the dog genome. But when Pendleton and her colleagues used their set of village dogs to similarly search for candidate domestication regions using Fst statistics, only about half the regions they found overlapped with previously reported ones. The researchers traced this discrepancy to the previous studies' reliance on modern breed dogs and suggested that some of the loci had to do instead with later breed formation.
Instead, she and her colleagues turned to the XP-CLR statistic to search for domestication-linked loci. This approach is designed to identify loci that are under selection based on patterns of correlated multilocus allele frequency differences between two populations. In this study, the researchers used wolves as their reference population and, after filtering based on what haplotypes ancient dogs harbored, homed in on 246 candidate loci within 50 kilobases of 429 unique genes. Most of these regions were not reported in previous studies, they noted.
Many of the genes within these candidate domestication regions support the neural crest hypothesis, the researchers said. For instance, they noted that the highest scoring locus they unearthed via XP-CLR centers on RAI1, which is involved in the retinoic-signaling pathway that is important in development. In humans, lacking the RAI1 gene causes Smith-Magenis syndrome, while its duplication leads to Potocki-Lupski syndrome — both developmental disorders. Some of the symptoms seen in those conditions such as craniofacial changes and sleep disturbances are reminiscent of the shorter muzzles and switch from nocturnal to diurnal life among domesticated dogs.
Similarly, the researchers noted that alterations to SERPINH1 in one of their candidate domestication regions that affects collagen, including in neural crest cell-derived tissue, could contribute to dogs' floppier ears.
The researchers also identified 67 regions of extreme copy number differences between their village dogs and wolves. In particular, they homed in on two large duplications affecting AMY2B, which, when present in greater numbers, increases starch metabolism efficiency.
"Altogether, we conclude that while primary selection during domestication likely targeted tameness, genes that contribute to determination of this behavioral change are also involved in critical, far-reaching pathways that conferred drastic phenotypic changes in dogs relative to their wild counterparts," the authors wrote in their paper.