Jessica Hekman: Is there a ‘good dog’ gene?

Darcey. Image Credit: Emily Nordvang, All rights reserved

Click here to listen to the full podcast episode

Kat: Jessica Hekman, the self-styled Dog Zombie, works with Elinor Karlsson and her team studying the brains and genes of dogs to understand canine behaviour, including trying to figure out why certain dog breeds behave in certain ways. She’s also the owner of three dogs - Fitz the Border Collie, Dashiel the English Shepherd and Jenny, a genetic pick’n’mix - some of whom you’ll hear making cameo appearances in this interview. As far as she’s concerned, based on her own personal experience, breed certainly does influence behaviour. 

Jessica: The Border Collie is very Border Collie. He likes to herd things. We're having an issue right now where he doesn't want the English Shepherd to be allowed to leave the back porch, so there's just this narrow exit from the porch, and so the Border Collie wants to control the exit and not let the other dog's off. Right? Because that's a nice narrow opening where he can control what they do. 

The English Shepherd is also very, very smart, but a lot less intense and a lot less focused on herding because they're a more general purpose farm dog, so I do see that. 

And then Jenny, it's interesting, as I said, she's a mixy mix and she's got a lot of stuff in her, and so who knows. I have actually had her ancestry tested and so I know what's in there. And so it's an interesting question of whether I'm seeing what I want to see as the results of the test or whether it's really because of her ancestry. She is 25% Labrador Retriever and she is very interested in food and in toys, which are both Labrador Retriever traits, and she's 12% samoyed and I can see that, that's more of a physical thing, like I can see it in her soft coat, but she also has this distinctive woo woo woo way that she barks, which is not how most Labs bark, but is more typical of how samoyed would bark. So I see it, and then my boss, Elinor Karlsson, would say "Oh, you're just telling yourself stories".

Georgia: Amazing. So how do you go about studying dog behaviour and its relationship with genetics?

Jessica: So the behaviour, we have a website with a lot of behaviour surveys on it. I think we have about 220 questions that people can answer right now. And we have people who have pet dogs and also people who have working dogs actually, who can come to that website and answer those questions about their dogs. And that in itself is a really useful dataset because they also tell us things about how old their dog is and what the dog's breed is. 

And so we can start doing analysis, and we're in fact about to publish a paper with our first analysis of, do we see different behaviours by breed? Do we see different behaviours by age? By size? Things like that. And then we also, when we have the funding, we can pay for people to have their dogs sequenced. So that is we would send them a kit and they would use the kit to send us back some saliva, actually that would go directly to a sequencing laboratory, which sequences it. And then we can look to see a particular behavioural trait that we're interested in, then we can look and see, are there any particular places in the genome that correlate with that trait?

Georgia: I mean, I know a personality in genetics is quite a fraught topic in general, is this easier or more difficult to study in dogs?

Jessica: The theory is that it's easier, that’s why we're doing it, and part of the theory has to do with working dogs, is that dogs have been under stronger selection pressure for behaviour than humans, right? Humans are free breeding, we hope, and dogs we are choosing to put them together based on the traits that we want. And a lot of those traits are about morphology right? So a lot of them, particularly in our pet dogs, are about what colour the dog is, how big the dog is and how long the dog's fur is but, both in the pet dogs but even more so in the working dogs, there's selection pressure for how the dog behaves. And so we're hopeful that we'll be able to take a population of dogs that behave in one way and then compare them to a population of dogs that have been bred to behave in a different way and that we'll be able to see differences that way, but it still requires a lot of dogs.

So with humans, some of the recent behavioural studies, they're hitting sample sizes in the hundreds of thousands now and we're hoping that we're only going to need maybe 10,000 dogs to get some good answers, but that doesn't mean that we're there yet, right? We're still at around 2000 dogs. So right now what we're doing is well with the 2000 dogs we can start looking for the easier stuff while we're collecting the information to look for the behavioural stuff. And so right now we're looking at things like; genes related to size and ear carriage and white coat colour and things like that, which are amenable at these sample sizes, and then we're collecting the information for later and doing the breed behaviour studies and that kind of stuff.

Georgia: So you're aiming for a much bigger sample size, but are there any sort of personality gene stories that are starting to emerge at this stage?

Jessica: Yeah, we are seeing a couple of things like a spike for something like woo woo barking, we call it woo woo barking, and I mentioned earlier that my dog Jenny does that specific woo woo bark. So we're finding some genes related to that. We are mostly starting to see some behaviours that do really depend on breed. 

So for example, Border Collies do come out as being much, much more biddable than other breeds, meaning that they are more likely to pay attention to their handler and want to do what the handler wants them to do. So that's something that we are hoping that we're going to be able to use the overlap of breed with the genomics of where stuff is in the DNA to try to find some answers, but we're not getting any really strong hits for behaviour yet.

Georgia: I love the woo woo bark. I know exactly what you mean.

Jessica: Yeah, I say that Jenny says woo, and my Border Collie Fitz actually says wow, and then it's only the English Shepherd Dash who actually says woof.

Georgia: You mentioned earlier, working dogs are a big part of this question, they have this incredibly specialised behaviour. So do you think this will be easier to pick apart? The sort of the unique code for like, you mentioned, herding behaviour in sheep dogs?

Jessica: I really hope so. So as I mentioned, we're starting with the guide dog schools and that's because they have really gotten their act together to organise and standardise on recording the behaviour for many generations of their dogs, again, in a standardised way. So that they're all using the same set of questions to say, how does the dog behave in this situation? How does the dog behave in that situation? So it's this lovely synergy of both that the dogs are under selective pressure for particular behaviour and that they have this standardised assessment so that we can trust that the answers for this school and that school are somewhat similar. 

So I would love for us to start expanding into other types of working dogs, I mean, we're going to have to, and we fully intend to, but to take your example of herding dogs now there's definitely a strong genetic component to the ability to herd, but the people who are working with herding dogs do not have the same standardised way of assessing dogs. 

So we're going to have to figure that out. What are we going to do? Are we going to go to herding trials and take the dogs who succeed the most at a herding trial and compare them to dogs who do less well, but a lot of that again is due to how they've been trained, right? So we're going to have to try to figure out how to do those phenotype assessments, the behavioural phenotype assessments. That's the really big challenge there.

Georgia: And so what are the goals for this study? What are you hoping to achieve once you have your samples? And once you start to get a clearer picture of what's going on.

Jessica: So two things; one thing is that we really want to be able to give back to the people who are breeding these working dogs. So we really want to be able to help them set up what is known as genomic estimated breeding values. And what that means is that you would be able, when you have a litter of hopeful guide dog puppies just born, you'd be able to genetically test each of them. And we'd give back and assessment of, this one is probably going to do better on the confidence trait that you've been studying, and this one's probably going to do better at what they call body sensitivities, so whether they mind if there's a harness on them, and this one's going to do better at biddability, so paying attention to what the person wants them to do.

Jessica: And then hopefully they'd be able to take those scores and say, okay, we have an idea of at least genetically where these puppies are starting off from. And we know that this isn't going to be fully predictive and that there's going to be environmental influence, but then they'd be able to make some decisions right there. And so they'd be able to say, well, based on this test, this puppy looks like they're going to do really well and so we definitely want to hang onto this one and go ahead and train them for the full year for the guide dog program, this puppy hasn't been doing so well on our assessments, and also the genetic test came back saying maybe he's lower confidence. So putting those pieces of information together, this is one that we're going to redirect out to be adopted out as a pet. 

That's really important to the guide dog schools who put a lot of money and energy into raising up a lot of dogs that not all of which succeed either as guide dogs or as breeders. Right? And so they'd really like to know earlier, is this a dog that we want to hang on to? So we're hoping to give back in that way.

Georgia: To save the dog the humiliation of failing guide dog school?

Jessica: Yeah! I mean, that could be anthropomorphism, but I think it's certainly fair to say that if a dog is going to be a pet, shouldn't it have the luxury of being able to be placed in that pet home at eight weeks of age rather than at one year of age? So I think it is better for the dog's welfare for them to be able to be put into a job that they're going to be successful at.

Georgia: And you mentioned there was a second application on top of this one.

Jessica: Yeah, So our hope is that we can start to understand what is different in the brain of a dog, for example, a dog who's very confident or a dog who's very not confident, say very anxious all the time, with genetics. The hope is that if you were studying confidence, you would find some versions of particular genes that make a dog more or less confident, and then the next step would be to try to figure out what those genes actually do in the brain, what their function is. So I think the best illustration of this is to talk about some work that Karlsson lab did recently but it's this past work looking at compulsive disorder in dogs.

So dogs do have a compulsive disorder very similar to what we call obsessive compulsive disorder in humans. So what that looks like in a dog, one thing is tail chasing. So a little bit of tail chasing can be normal and not to be something to be concerned about, but these compulsive dogs are doing it so long that it really interferes with their quality of life. So they'll spin around chasing their tail until they fall down until they're too dizzy to stand and then they'll get up and do it all over again. So that's sort of what a compulsive behaviour would look like.

Other ones are things like flank sucking so that the dog sucks on its own flank, and they can do this for so long that they actually injure themselves or in Border Collies, a typical thing that you would see is that they are obsessive about lights or shadows, and just spend their day really focusing on chasing lights or shadows around the house.

So Karlsson lab, as I said, did this work looking at obsessive disorder in dogs and found a couple of genes coded for proteins that are involved in a particular type of synapse in the brain. So a synapse is where two neurons in the brain come together and they need to be together to send messages back and forth. And the proteins that were involved in canine compulsive disorder were involved in holding the two neurons together, binding them together so that that synapse stayed strong. So then damaging one of those would affect compulsive disorder in dogs.

And what was interesting, we found that, but then a couple of years later published a paper which has this beautiful figure in it, where there's a drawing of the synapse and illustrations of where these proteins that are coded for by these genes are part of the synapse and those are labeled dog, and then there's also other proteins involved in the holding the synopsis together, and those are labeled human. And those are from other papers where they were associated with obsessive compulsive disorder in humans, but still part of the same system. And then there's others still part of the same system labeled mouse, and they're from yet other studies that found an association between compulsive disorder in mice and still other genes that code for proteins that are involved in this same system.

And so basically then you can start to see that the system itself is very important in compulsive disorder. So it may be that the cause of compulsive disorder in these three different species is totally different genes, but the genes are all involved in the same system. And when you put it all together like that, you start to see the importance of this particular system. And again, that will then hopefully help other laboratories start working towards finding specifically targeted therapies, either medications or behavioural approaches.

Kat: That’s Jessica Hekman, and you can check out her work at dogzombie.com.

Image Credit: Pug, Tabitha Dale, all rights reserved