Where did dogs come from?

What we know and what we don’t know about dog domestication

Want to learn more about dog domestication? It's not too late to sign up for my online class, From Domestication to Inbreeding!

Dogs evolved from wolves. We’ve been certain of that for several decades by now. But there remain a lot of questions: exactly when did dogs first appear? Did they join their fate with humans when we were hunter-gatherers, or were they attracted to us after the Agricultural Revolution, because we had begun to farm? Which group of ancient wolves did they come from? Knowing more about where dogs began will help us understand modern dogs and their behavior better. Academics are currently conducting a very polite debate about these questions in journals, waged over the course of years.

Grey Wolf

Why is the problem such a hard one? Until recently, the tools that we were using to get information about ancient canids were very limited. Our first tool was archaeology: digging up the remains of ancient canids, trying to figure out if the animal was more dog-like or more wolf-like, and then estimating the age of the find. It’s not entirely straightforward to tell an ancient dog from an ancient wolf using only bones, especially when many archaeological finds are incomplete. The important parts of the skeleton for this work are the teeth and skull: dog muzzles are shorter than wolf muzzles, so that their teeth are more crowded into the available space, and the last premolar and first molar are smaller in dogs than in wolves. Some interesting finds have suggested that dog-like canids first appeared between 15,000-30,0000 years ago — that’s just before agriculture was first developed.

A well publicized 1997 paper from Vilà et al. popularized a new tool for dating dog domestication: analysis of mitochondrial DNA, or mtDNA. Mitochondrial DNA is the DNA inside the mitochochondria in our cells. Mitochondria used to be free-living organisms; they began to live symbiotically in the cells of multi-cellular organisms billions of years ago, but still have their own separate DNA. Mitochondrial DNA gains new mutations at a regular rate, and this can be used as a molecular clock: compare the mtDNA of two different species, and by counting the differences, you can estimate how many years ago the ancestral species split into the two new species.

The problem is that this molecular clock isn’t very reliable or very precise: we don’t really know exactly how fast mtDNA mutates, which makes the clock hard to calibrate. The 1997 findings suggested that dogs and wolves separated about 130,000 years ago — an order of magnitude more than the archaeological estimates suggested! Other mtDNA studies have been conducted since then, with a variety of results, none of them conclusive. It turns out that dog and wolf mtDNA divergence is particularly difficult to analyze because dogs and wolves can and do still interbreed. My golden retriever may not have a wolf in his immediate ancestry, but I suspect you don’t have to go back all that many thousands of years to find one — certainly not all the way back to the domestication split. And there are quite a few populations of dogs in the world with much more recent wolf ancestry than that. This interbreeding really screws up the molecular clock.

In the last few years, though, the revolution in genomic tools — cheap and efficient sequencing of complete genomes — has gotten to the point where it’s affordable to completely sequence the genomes of a number of dogs and wolves for a study. This is significantly changing the kinds of things we can learn about how dogs and wolves genetically differ. Instead of guessing at changes in mtDNA, we can look at the actual genes that differ between the two species. These new studies have set the date of dog domestication at 11,000-32,000 years ago, a date which is similar enough to the archaeological findings to make a lot of sense.

We’ve learned a lot of interesting things from these new sequencing studies beyond just a more precise date of domestication. A little more than a year ago, Axelsson et al. found that dogs make more of an enzyme for digesting starch than wolves do. The enzyme is called amylase, and dogs have multiple copies of the gene, whereas wolves have only one. These researchers wondered if this improved ability to digest starch meant that dogs were domesticated after the appearance of agriculture — if starch digestion was part of the domestication process. However, a study published in January 2014 by Freedman et al. dug deeper into the amylase question and discovered that in fact, not all dogs have extra amylase genes. Some ancient breeds, like the husky, do not. Neither does the dingo. These very recent findings suggest that dogs were in fact domesticated before the Agricultural Revolution, and that some breeds later developed an improved ability to eat what we eat, adapting to their new post-domestication diet. You might imagine that such a change would have been less important to the husky, living in the cold north as it did, where meat was on offer much more often than plants.

Freedman et al. also suggested that dogs didn’t actually evolve from wolves. Wait, what? It's possible that both dogs and wolves evolved from a different ancient canid which doesn’t exist any more. Freedman came to this conclusion using a somewhat complicated genomic analysis which doesn’t tell us anything about what such a canid would have been like, but it’s an idea which resonates with reservations I’ve always had about the “dogs came from wolves” theory. Wolves are so shy, so hesitant to come near humans, and so focused on making their living by hunting. The ancestors of dogs seem more likely to have been scavengers, willing to live close to humans. Maybe some ancient canid did give rise to both species — the one moving closer to human civilization and becoming dogs, the other farther away and becoming wolves. With several studies coming out every year about dog domestication, we may learn more very soon.

For more information, check out “How Much Is That in Dog Years? The Adventof Canine Population Genomics,”a recent open-access review article that provided much of the information in this story.

Note: this story was originally published in the summer 2014 issue of  The APDT Chronicle of the Dog.

Image by Isster17 (Own work) [CC BY-SA 4.0 (http://creativecommons.org/licenses/by-sa/4.0)], via Wikimedia Commons

References

Axelsson, Erik, et al. “The genomic signature of dog domestication reveals adaptation to a starch-rich diet.” Nature 495.7441 (2013): 360-364.

Freedman, Adam H., et al. “Genome sequencing highlights the dynamic early history of dogs.” PLoS genetics 10.1 (2014): e1004016.

Larson, Greger, and Daniel G. Bradley. “How Much Is That in Dog Years? The Advent of Canine Population Genomics.” PLoS genetics 10.1 (2014): e1004093.

Vilà, Carles, et al. “Multiple and ancient origins of the domestic dog.” Science 276.5319 (1997): 1687-1689.