A report published this January in the journal Science illustrates the kinds of genetic adaptations which co-occurred with the domestication of modern dogs. Most striking were changes in numerous genes responsible for carbohydrate metabolism and brain function, suggesting what may have been key adaptations to the post-domestication environment.
The researchers sequenced DNA from 60 dogs representing 14 different breeds and from 12 wolves found at locations around the world. They first identified millions of molecular markers or polymorphisms in the populations of both wolves and dogs. Polymorphisms are regions of DNA which differ either at a single nucleotide/base pair (single-nucleotide polymorphism, SNP) or which differ in the number of times a sequence of nucleotides is repeated (copy number variant, CNV).
Next, they constructed a statistical model to distinguish genetic markers present in dogs from those found only in wolves. Because the study included many millions of genetic markers, the authors restricted further analysis to those genetic sites which were >5 standard deviations from what would be predicted by the model if a genetic marker only differed between dogs and wolves by chance. Even accounting for these statistical issues, it is equally important to address whether any detected genetic differences represent adaptations, that is changes which are selected for by a new environment (i.e. which provide better fitness), or whether such changes occur randomly as populations diverge (genetic drift).
Based upon their analysis, the authors identified 122 novel genes potentially associated with canine domestication. Most notable among these genes were large clusters thought to be associated with the development of the nervous system as well as starch and fat metabolism, although the functions of many of these genes has only been poorly studied.
In order to perform preliminary confirmation that some of these genes were selected for, the authors studied in more detail the difference in function of some of these genes between dogs and wolves.
For example, they found that amylase, a gene which breaks down carbohydrate and is expressed by humans in the mouth, had 7-fold more copies in the average dog genome than in that of the wolf. This genetic change was associated with nearly 30 times higher gene expression in the pancreas of the dog and higher activity of the amylase enzyme in serum (an artificial environment). Because the pancreas is involved in injecting enzymes involved in metabolism into the digestive tract, this change is highly suggestive of an increased ability in dogs to efficiently digest carbohydrates. Similar changes in gene expression and enzyme activity were found in an enzyme responsible for the conversion of maltose to glucose, whereas a third gene polymorphism, found primarily in dogs rather than wolves, appears to harbor a mutation that has been previously shown to increase glucose uptake from the small intestine into the blood.
This study pretty nicely shows just how quickly genetic selection can take place given the strong (albeit, artificial) selection pressures placed on domesticated dogs by their human handlers. Although exact data at which wolves were originally captured and domesticated is indeterminate (and possibly may have happened numerous times), any changes found between dogs and wolves which resulted from selection likely happened more recently than 50,000 years ago or so.
I am reminded about certain human fad diets which propose that humans haven’t adapted properly to eating grain-heavy diets. Clearly, however, our consumption of grains with the advent of technology overlaps fairly well with grain consumption by domesticated dogs. This “technological” selection may have had equally or potentially more rapid effects on human carbohydrate metabolism.
Clearly, this is an empirical question, and the above study is a product of a growing trend in human genomics which may soon provide more definitive answers to ancient questions about the human diet.