Diet Shaped Dog Domestication
by Elizabeth Pennisi on 23 January 2013

Fido may prefer steak, but his digestive system is also geared up for rice and potatoes. That's the conclusion of a new study, which finds that dogs have evolved to eat a more varied diet than their wolf ancestors. The shift parallels genetic changes seen in people and bolsters the idea that dogs and humans share similar evolutionary stories.

Dogs evolved from wolves more than 11,000 years ago, somewhere in Eurasia, though exactly when and how is under debate. The shift from wolf pack member to family pet involved more than just the ability to get along with people, says evolutionary geneticist Erik Axelsson from Uppsala University in Sweden. He and his colleagues compared dog and wolf DNA to learn which genes were important for domestication.

They sequenced DNA from 12 wolves from around the world and from 60 dogs belonging to 14 breeds. They first looked for individual letters in DNA, called bases, that varied from one genome to the next, identifying about 4 million of these so-called single-nucleotide polymorphisms (SNPs). They ignored regions with the most SNPs and instead focused on places where there were very few or no SNPs. That lack of variation signals DNA that was so important for survival during domestication that any variation there was lost, so most dogs have the same SNPs. Those regions were the ones the researchers were most interested in following up on.

The analysis turned up 36 regions, with 122 genes in all, that may have contributed to dog evolution, the team reports online today in Nature. Nineteen of these regions contain genes important for the brain, eight of which are involved with nervous system development, which makes sense given the importance of behavioral changes in the transition to becoming man's best friend, Axelsson notes.

More surprising were genes for digesting starch. Dogs had four to 30 copies of the gene for amylase, a protein that starts the breakdown of starch in the intestine. Wolves have only two copies, one on each chromosome. As a result, that gene was 28-fold more active in dogs, the researchers found. More copies means more protein, and test-tube studies indicate that dogs should be fivefold better than wolves at digesting starch, the chief nutrient in agricultural grains such as wheat and rice. The number of copies of this gene also varies in people: Those eating high carbohydrate diets -- such as the Japanese and European Americans -- have more copies than people with starch-poor diets, such as the Mbuti in Africa. "We have adapted in a very similar way to the dramatic changes that happened when agriculture was developed," Axelsson says.

Dogs and wolves have the same number of copies of another gene, MGAM, which codes for maltase, another enzyme important in starch digestion. But there are four key differences between the sequence in dogs and wolves. One difference causes dogs to produce longer versions of maltase. That longer protein is also seen in herbivores, such as cows and rabbits, and omnivores, such as mouse lemurs and rats, but not in other mammals, suggesting length is important to plant-eaters. These differences make the dog maltase more efficient, the researchers report.

Axelsson thinks these results support the idea that wolves began to associate with humans who were beginning to settle down and farm. Waste dumps provided a ready source of food, albeit not meat, the usual diet. Thus early dogs that evolved more efficient starch digestion had an advantage, he notes.

The finding of these diet-related genes is "very surprising and very exciting," says Elaine Ostrander, a geneticist at the National Human Genome Research Institute in Bethesda, Maryland, who was not connected to the study. "It hints that there are a lot more [genes] to be found" involved in domestication, she adds. As more researchers compare wolf and dog DNA, Ostrander expects more genetic differences between dogs and wolves to emerge. "We are really going to figure [dog evolution] out."

Robert Wayne, an evolutionary biologist who studies dogs at the University of California, Los Angeles, but was not involved with the work, is also pleased with the study. He says he gets contacted often by pet owners wondering if dogs, like wolves, should eat primarily meat. "This [study] suggests no, dogs are different from wolves and don't need a wolflike diet," he says. "They have coevolved with humans and their diet."

What do you guys think?