Your dog is basically a super social wolf, and scientists may have found the gene that makes him want to cuddle with you.
A new study shows that friendliness in dogs is associated with the same genes that make some people hyper-social.
The study, published this week in the journal Science Advances, found that structural variations in three genes on chromosome 6 are correlated with how much canines socialize with humans. An analysis of DNA from two dozen animals revealed that these genes look very different in dogs than they do in wolves.
Mutations in the same genes are also linked with a rare developmental disorder in humans called Williams-Beuren Syndrome, or WBS. People with WBS are typically hyper-social, meaning they form bonds quickly and show great interest in other people, including strangers. Other symptoms include developmental and learning disabilities as well as cardiovascular problems.
To Bridgett vonHoldt, who studies canine genetics at Princeton University, some of these traits sounded a lot like the behaviors of domesticated dogs, especially compared with wolves.
For example, dogs like to stay close to humans and gaze at them for longer periods of time than wolves do. Dogs also tend to be less independent in problem-solving when they're around people, and they retain their affinity for humans throughout their lives.
"Many dogs maintain their puppy-like enthusiasm for social interactions throughout their life, whereas wolves grow out of this behavior and engage in more mature, abbreviated greetings as they age," said Monique Udell, who studies animal behavior at Oregon State University and co-authored the new study. "One might think of how a young child greets you versus a teenager or adult relative."
These behaviors are typical of what scientists call domestication syndrome, and researchers have noticed them in other kinds of domesticated animals as well. But they don't fully understand how the underlying genetic changes develop.
"Everyone wants to find the genes that make dogs different from wolves, and try to understand how domestication changed the genome," vonHoldt said.
She already had a head start. In 2010, as part of her doctoral research, vonHoldt had mapped the entire genome of 225 gray wolves and 912 dogs from 85 breeds. There were a few genes that stood out as consistently different between dogs and wolves, especially the WBS gene WBSCR17. But vonHoldt still didn't have a handle on how those genetic differences were related to behavior.
VonHoldt met Udell three years ago, and they started chatting about canines. Together they realized that if they combined vonHoldt's genetic expertise with Udell's canine behavior data, they might be able to find the missing link.
They decided to examine the social behavior of a group of dogs and a group of wolves and then analyze their DNA in the region that included the WBSCR17 gene.
This idea was pretty new. While scientists have analyzed the genes responsible for specific clinical disorders in dogs, something as complex and varied as social behavior is a lot trickier to study.
After selecting 18 dogs and 10 gray wolves who had been socialized with humans, they began their work with a series of behavioral tests.
When the dogs were given a puzzle box with a sausage hidden inside, only two of the 18 were able to open the box whether or not a human was present. Wolves performed much better: eight of the 10 opened the box when a human was present, and nine of them opened the box when they were left alone.
When dogs were in the presence of a human, they spent a median of 20% of their time looking at the person and only 10% of the time looking at the box. The wolves, on the other hand, spent nearly 100% of the time looking at the box whether or not a human was present.
Udell said that these results agreed with previous studies that have shown that dogs are not as good at independent problem-solving as wolves, and that they get more distracted by social stimulation.
Next came the sociability test, which took place in four phases. In each, a human sat in a chair near a dog or a wolf for two minutes, and researchers recorded how much time the canine in question spent within 1 meter of the person.
In the first round, the person was a stranger who sat passively in the chair, not making eye contact or speaking to the animal. In the second round, the stranger actively engaged with the canine. Then the two phases were repeated with an owner or caretaker instead of a stranger.
The researchers determined that when familiar humans were present, dogs spent a median of 93% of their time near people while wolves spent only 36% of theirs. However, when the humans were strangers, the dogs stuck around 53% of the time and wolves 28% — a difference too small to be statistically significant considering the small number of animals involved.
While there was some variation between individual dogs and individual wolves, what was really important to the researchers was to determine whether there was a link between the sociability of each canine and what his or her DNA looked like.
So they took blood samples from 16 of the dogs and eight of the wolves and analyzed a large chunk of DNA on chromosome 6, including the region associated with WBS. (Two of the dogs and two of the wolves were not included in this part of the study because researchers were missing some behavioral data or because they couldn't get enough blood for the DNA analysis.)
They found mutations in three genes that were much more common in the hyper-social canines, most of which were dogs. These three genes are called GTF2I, GTF2IRD1 and WBSCR17, and have also been shown to cause an increase in social behavior in mice and are believed to do the same in humans.
Interestingly, two of the wolves were very social and dog-like in their behavior, while one of the dogs acted quite wolf-like. The team found that the two social wolves had more mutations in these three genes, while the wolf-like dog had fewer mutations.
Adam Miklósi, an ethologist at Eötvös Loránd University in Hungary, said that the results would have been more convincing with a larger sample size.
The study authors acknowledged that their sample size was small, but they pointed out that the link between DNA and behavior was quite distinct. "I think it's astounding that we have significance," vonHoldt said.
Despite the difficulty in finding human-socialized wolves to use as subjects, the authors said they would like to study larger samples in the future. They also want to understand how this trend varies by dog breed.
VonHoldt is especially curious about exactly how the genetic mutations result in increased social behavior, and plans to research that question in more detail.
Eventually, she may figure out how wolves evolved into man's best friend.