Jeff Schoenebeck - Dog genetics

Kat - You're listening to the Naked Genetics podcast with me, Kat Arney. Still to come, we'll be learning about a legless gene of the month. But first, humans aren't the only animals to be getting their genomes analysed - our four-legged friends have been getting in on the act too, as I found out when I spoke to Jeff Schoenebeck from the Roslin Institute in Edinburgh.

Jeff - The differences between pugs and great Danes and these dogs with very disparate shapes, what's remarkable is that it seems like in dogs, these things that are driving shape differences can be held in a couple of hands. So, this is quite a striking difference to other species and I think about human stature and the genetics of something like that in humans which we know from studies, there's hundreds of variants that seem to be important that are playing into this determining stature. In dogs, what we have are, at the moment, it depends on the study and how it's designed but somewhere in the 10s of variants that can be playing into this. That's a remarkable difference, but bear in mind, a lot of these variants had to be recognised by breeders in the litter. So, I think the difference between dogs and humans or other species is that the variants that are particularly affecting morphology probably have larger effects than what would typically be found in a population of humans.

Kat - Different dog breeds have been very highly selected for. They want the long ears of a basset. They want the kind of the luxurious hair of a poodle. How do you start pinning down some of these genetic differences that lead to these really quite disparate traits?

Jeff - So, we begin by comparing the skulls - the shapes and sizes of skulls across animals. Our study design is based on patients that are coming through the referral hospital, many of them coming through The Royal Dick School of Veterinary Studies. These patients are coming because there's a diagnostic purpose for them to have a CT scan. But we can take that information when a dog gets a head scan and we can use it to reconstruct the head and we can derive size and shape from those animals. And so, we can feed this information into our genome wide association studies which for all intents and purposes tell us what area of the dog genome and what piece of DNA is carrying a particular genetic change and is driving a pug to have a shorter snout. That's the first part. So, we want to find a signal that's driving a particular shape-related event and we want to identify what chromosome and what region in the chromosome that change is located on. 

And then we want to look across our data and by comparing dogs of similar shapes to those that lack that shape and identify what is the minimal region where this mutation or this genetic change could fall in. Is it a hundred thousand positions, is it 90,000? Once we have an interval defined then we leverage these whole genome sequencing technologies that have come on board in recent years. We sift through genotypes and genetic variants and we try and infer what is the best candidate there that can be causing this morphological change. Ultimately, we want to understand what is the change at the DNA level that's driving the trait. In future, we'd like to characterise the biology behind this, but it's still a bit of early days for us. We're still trying to define what are the places in the dog genome that drive size and head shape.

Kat - More broadly, looking at other studies that have been done, what sort of genes have been identified that gives some of the varieties to dogs that we see?

Jeff - I think among the morphology studies, size is where the most data exists. We see genes that make sense, things that affect insulin signalling, insulin growth factor 1 which we know in a myriad of species can affect overall animal size. Growth hormone receptor - again, this is a gene who captures signalling molecules and signals downstream to modulate growth. So, these are fantastic. They make sense in terms of why they've been selected by breeders to modulate animal size. From the skull side of things, it's still pretty wide open and this is again the interest of my lab is to put this issue to rest so to speak. We can infer that BMP signalling is probably involved, bmp standing for bone morphogenetic protein. So that makes sense. There's still a lot to be done.

Kat - One of my favourite traits in dogs is the little stumpy ones, things like corgis and dachshunds, and bassets. Is there a similarity between these kind of stumpy dogs in terms of their genetics?

Jeff - Absolutely and this is a fantastic work Heidi Parker in Elaine Ostrander's lab, wondered the same exact thing - what is driving dogs to have smaller legs? The technical term for this is asymmetric chondrodysplasia.

Kat - Cute stumpiness.

Jeff - Exactly. So she compared all the belly scraping dogs to those that stand high off the ground and looked for genetic differences between the dogs with the short legs and the dogs with the long legs. She found this beautiful signal on chromosome 18, used other tools to see, well, was there breeder mediated selection in this area on chromosome 18 and indeed, there was. And then through using sequencing and mapping technologies, she was able to find a region and it turns out that it's an additional copy of Fgf4.

Kat - What does Fgf4 do and how did an extra copy of it get into these short dogs?

Jeff - Fgf4, fibroblast growth factor and there's a large family of FGFs and they're involved in a host of different biological processes. But they have a very prominent role in development. FGF4 in particular is a gene that's required for normal limb growth. In many species, this has been demonstrated. And so, what we think occurred in dogs, and this is due to the structure, this additional FGF gene that was identified, and what we infer is that it's in fact, a copy of RNA - the message of the gene had been reverse transcribed back to DNA and that DNA got inserted back into the genome. So, the dogs wound up with an extra copy of Fgf4. Somehow, this extra copy, the overexpression of this gene is probably perturbing limb development, so it's restricting limb growth.

Kat - And then presumably, some owner of that dog went, "Wow! That's cute. Let's make more of those."

Jeff - Absolutely. Getting back to this idea of form and function, here, it's men saying, "These dogs have a form that can suit very well in the function of going into holes of badgers or rabbits and chasing out prey or unwanted pests in the farmland." And so, many of these dogs that have this FGF retrogene are dogs that are used for hunting in burrows and then the other large bodied dogs that have this are cattle herders. So, the idea being that they're so low to the ground, if a cow kicks, it'll swoop up above the head and they'll miss it. It won't connect with their head.

Kat - That was Jeff Schoenebeck from the Roslin Institute