The Vetrospective
Host Dr. Michael Kent explains + breaks down the Science behind the research about our Domestic Companion Animals.
The Vetrospective
Dog Genetics
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Host Dr. Michael Kent sits with Dr. Danika Bannasch and explores the world and pedigree of dog genetics.
So we understand the genes and the alleles that give different coat colors in dogs, and in fact blue versus brown eyes in dogs and other traits. We also understand many of the alleles that give and cause diseases in dogs, and that's what I've been working on for 25 years and a lot of other dog geneticists around the world.
SPEAKER_01Hello and welcome to Vetrospective. This is your host, Dr. Michael Kent. I'm a professor in radiation oncology at the UC Davis School of Veterinary Medicine, and your host. Today we'll be talking about dogs and genes. A T C G. These are the letters that make up the code of life, the instruction manual that is the guide our cells use to build each of us and pretty much most life that we know. The bases put into different combinations are the building blocks for genes. Genes, in turn, are the blueprint of life, and our understanding of them is helping shape the future of diseases and medicine. As dogs evolved from their ancestor, the wolf, becoming domesticated and really co-evolving with us, we created breeds. This was done by selectively breeding dogs for the traits that allowed us to coexist in the same environment and take on the many roles dogs play in our lives. Whether it's working dogs, sporting and hunting dogs, or dogs that have been bred to be our companions, genetics is at the heart of it. It gave us breeds diverse as chihuahuas, pugs, labs, golden retrievers, and even the really large Great Dane. Unfortunately, in creating these breeds, we inadvertently also bred in characteristics we did not want, those that can lead to certain diseases. Today's guest is Dr. Danica Banish, a veterinary geneticist here at UC Davis, who will help us better, really get a better understanding of the ins and outs of this complex but really cool and interesting topic. Dr. Banish received her undergraduate degree in genetics from UC Davis and then went on to do her PhD in molecular biology from Princeton University before returning to UC Davis for her veterinary degree, a postdoc position, and a residency in genetics. Then she became faculty. She has run an active genetics laboratory for over 25 years here and has published over 140 scientific papers and is now associate dean for research here at the Vet School. Welcome to Vetrospective, Dr. Banish. Thank you for joining me here today.
SPEAKER_00Oh, thank you for having me.
SPEAKER_01No, it's great. I appreciate it. So, first, I want to ask you a question I asked most of my guests. How did you decide to pursue a career in veterinary medicine and why genetics in particular?
SPEAKER_00So I I actually always wanted to be a dog geneticist. I just didn't know how to do that because there weren't really any dog geneticists.
SPEAKER_01Was this at five years old? You knew the word genetics or geneticists? So tell me a little more.
SPEAKER_00We wrote in my high school yearbook that I wanted to study animal genetics. So I had figured it out by then that it was kind of the science behind dog breeding. And I think that's where it came from. Um I, when I was uh 12 years old, um, I did some dog walking for some um neighbors and friends of ours, and they took me to the Philadelphia dog show. So that was the first dog show I went to, and they gave me the AKC dog breed book. And I went through and just marveled at all of the different sizes and shapes and colors of dog breeds and got really fascinated by that and wanted to study why that happens. So yeah, I mean, it it I don't like to tell the students about my history because people don't usually know what they want to do when they're 16 years old. I do not use it, it is it is highly unusual, but I guess I'm pretty um pretty focused. No, that's good.
SPEAKER_01Um, so each of our cells have genes in them, right? They're in the nucleus and they're organized into chromosomes. I think most people have heard of that. But how does that translate into actually making a dog?
SPEAKER_00So the the genes are sort of a blueprint for proteins, and it's actually the proteins in the body that and the proteins that are made that come together to give different cell types. So your bone cells are different than your skin cells, for example. And then all of those different cell types go and turn into organs, and the organs work together to make a dog or a human.
SPEAKER_01Pretty much any species. So now you get one set of genes from each parent, right? And they mix up a bit.
SPEAKER_00They do. There's there's some shuffling um in both of the parents so that the offspring actually have a sort of slightly different complement of it's not it's not genes, but versions of genes. So we all have the same genes. In fact, humans and dogs have pretty much the same genes. There's just the sequence of the genes is a little bit different.
SPEAKER_01Okay, so the gene kind of codes for an idea of what a liver should do, but maybe a dog has a variation on it compared to a human. Right. And within dogs, there could be variation as well.
SPEAKER_00Right. Less variation than there is between dogs and humans, but yeah, there's variation between dogs, or same as there's variation between humans.
SPEAKER_01So, in a sense, the genes are a code book for us, right? And to understand how we're built and each cell is regulated.
SPEAKER_00Yep, yep. I mean, people use the term colloquially now and say, like, oh, my DNA or my blueprint, right? And they're they're sort of synonymous.
SPEAKER_01So if we want to get down to seeing my blueprint, we're talking about sequencing. You know, and this is the process of getting the genetic information out of a cell. Do we just take a cell or a group of cells and put it in a machine? Is and just kind of in that we could talk about if what kind of sample do we use? Because I know you're gonna have um either spitting or cheek swabs or a blood sample. Is there advantage over one over the other?
SPEAKER_00So we can sequence um DNA from lots of different sources. So you can get DNA from a cheek swab, you can get DNA from blood, you can get DNA from a tissue sample. Um and the difference is really in the quality and the amount of DNA that you can extract out. So for modern sequencing techniques, cheek swabs will work, but blood is probably a little bit better because you get enough extra in case something happens.
SPEAKER_01So you get more DNA than you do off a cheek swab.
SPEAKER_00Yeah. And you asked about the sequencing process. So um things have changed in the last 20 years or so. Um, sequencing is very efficient and not very expensive. So we can sequence the whole genome of a dog for about$500. Um what did that cost before?
SPEAKER_01Like to sequence the first dog, or or do we even know that?
SPEAKER_00Yeah, I mean, it's definitely gone down drastically, and the human genome project costs billions of dollars.
SPEAKER_01And that was for the first one person.
SPEAKER_00Yeah. Yeah. Yeah. So those that sequencing is a little bit different. So we talk, we'll go back to dogs for a second. There's something called a genome assembly. And that is the sequence of a dog put together into chromosomes, and we use that as sort of a benchmark. So currently, we use the genome assembly of a German shepherd named Mishka, and that's the one that most dog geneticists would align to. So if I wanted to sequence the DNA of a patient in the hospital because it has a disease that I think is inherited, I sequence that DNA, and we do that by chopping the DNA up into little tiny bits and sequencing the little tiny bits. And then you have to align those little tiny bits to a genome assembly. So in this case, we align to Mishka, the German Shepherd, and then we use bioinformatics tools to identify the differences between my patient and Mishkada.
SPEAKER_01Okay, I'm gonna stop you just here for a second. And bioinformatics tools is kind of this black box. So you you wind up with all these little sequences that you have and you know, a big Excel sheet, let's say, that's just got in on an Excel sheet. Yeah, so it's huge, right? So what are what is bioinformatics and how do you kind of use that?
SPEAKER_00So bioinformatics is basically the application of computer science to biology. So in this case, you know, you you said at the beginning that's just four base pairs, right? And it's just the order of those four base pairs on each chromosome that's what matters. Or in the case of sequencing a patient, it's just the differences between that dog and all of the rest of the dogs that matter. So bioinformatics is the ability to be able to pull out those differences and maybe get them to a form where you can look on an Excel file, but technically each patient that we sequence has maybe six million variants compared to Mishka, the genome assembly. So you can't even open that in Excel file.
SPEAKER_01Yeah, that's a that's a big file. So obviously, we need bioinformatics and software to help us sort all this out. So what is normal? You said we're using Mishka, one dog, one German Shepherd. Why don't we go back to the wolf as the cold standard? Since this is where dogs came from. Or how do we know a gene has changed and not just a variation on normal? Because I know if you sequence me and you, we're gonna have some pretty big differences there.
SPEAKER_00Uh yeah, yeah, we we will. Um, there were a lot of questions there. So I know, sorry. So there's there's a number of genome assemblies available now. For a long time, we we had one, which was Tasha the Boxer. Um, and then Mishka came about, and that was an improvement in that there was more sequence put together than in Tasha. So we switched to Mishka. I will say there is actually really good Greenland wolf assembly. So sometimes we use the wolf. Um it's not just about how good the assembly is, it's also how good the annotation is. So what do you mean by annotation? So most of the genome doesn't code for genes, the protein coding genes, and those are the ones that we know the most about. So the the genome assembly that we use is the one that has the most information about the protein coding genes because those are the ones we look at first. So I told you, I gave you an example that I sequence a patient's DNA. The very first thing I do is look for changes in the protein coding sequence to see if there's something that might be causing their disease there. And that's called exons, right?
SPEAKER_01Yeah. Now, when I was in school forever ago, they called the rest of it junk DNA. And I know it's not junk DNA. So, what are those other portions of the DNA doing that aren't just making our proteins?
SPEAKER_00So there's a large amount of the genome that's repetitive DNA, about 40% of the dog genome is repetitive DNA. About just a couple percentage codes for proteins. And the rest, we're not quite sure what it does yet, but I think we're starting to realize that it probably does something, um, and that we shouldn't just uh ignore it.
SPEAKER_01Interesting. So, like I said in the intro, we bred dogs initially from wolves. How closely related are dogs back to the prototypic wolf?
SPEAKER_00But we estimate that domestication occurred somewhere between 10,000 and 25,000 years ago. And um you can certainly tell by DNA that an animal is a wolf versus a dog. Um, it there are enough differences there that we can tell that. But between dogs, there aren't that many differences. So you you can look at a compilation of things of DNA sequences to try and tell breeds apart. Um, but they're not as different as you might think.
SPEAKER_01So how different are dogs breeds?
SPEAKER_00You know, are they are they related to each other very closely or or you know, so it it doesn't take many um variants to sort of change uh French bulldog from a wolf, right? So interesting. Um there's variants that would make it smaller, there's variants that would make it brachiocephalic. So in other words, not having a muzzle, there's variants that would make it tail kinked. Um, but it and there's probably variants that are affecting the coat color that change it, but it's really, you know, a dozen or so. Um it's not a huge number. And so there's not that many variants that differentiate the different dog breeds from each other.
SPEAKER_01Aaron Powell So the blueprints are for each dog breed are pretty similar in just a small number of changes in certain genes.
SPEAKER_00Right. And then there's also some changes in frequency of alleles so that you can identify by the DNA that it belongs to a certain breed, but those aren't changes that are changing proteins.
SPEAKER_01Aaron Powell So what do you mean by frequency of alleles?
SPEAKER_00I know I'm asking a lot of basic questions that uh someone who knows genetics should know already, but please we talked about genes because that's a term that's familiar to people, but really the the genes are all the same. So you and I have the same genes, all the dogs have the same genes. The differences are sequences that we call alleles, which are basically the different versions of the gene that exist.
SPEAKER_01Aaron Powell What gives me dark hair and you brown hair?
SPEAKER_00Right, exactly. Although hair color is a little bit complicated in people, it's a little bit simpler in dogs. But yeah, there's just a couple of okay.
SPEAKER_01Eye color, is that more simple? No, of course not. I couldn't come up with a good uh a good uh analogy for you.
SPEAKER_00But so But it is it is straightforward or more straightforward in dogs. So we understand the genes and the alleles that give different coat colors in dogs, and in fact, blue versus brown eyes in dogs and and other traits. We also understand many of the alleles that give and cause diseases in dogs, and that's what I've been working on for 25 years and and a lot of other dog geneticists around the world.
SPEAKER_01And why I have you sitting here today. So basically, these alleles are different to create different individuals, and it's this combination of all the different alleles of your genes that make me, me, and you, you.
SPEAKER_00Yep.
SPEAKER_01Now, I know there's this term kind of called genetic diversity, and in dogs, it's you know, I've seen some controversies about whether we bottleneck dogs and you know, and what I kind of want to get an idea is are dogs less genetically diverse than let's say cats or people?
SPEAKER_00Um, well, they're definitely less genetically diverse than people. Um within dog breeds, the um inbreeding is actually quite high. Um, part of that is due to how dog breeds were initially created. So someone had a dog that they liked, and they started breeding more dogs that were similar. And the quickest way to get dogs that were similar would be to breed relatives to each other. Line breeding, it's called, right? Well, initially it might have even been inbreeding to get a dog. So I see on your computer you have a sticker of a golden retriever. I think you might be sort of fond of golden retrievers. Golden retrievers started from one particular male stud dog who was bred to a number of different females and the entire breed was created from them. From one dog? Well, from a handful of dogs. Yeah. And then once um, once there were sort of typical looking dogs of breeds, the kennel clubs were organized and started and started limiting the individuals that could be registered as that breed. And so once you sort of close off the gene pool and don't allow any new dogs to enter, then you're sort of forcing continued inbreeding. And what we end up with today are most dog breeds that have uh really, really high inbreeding and no sort of avenue to add additional genetic material. So the purebred dogs, you can't go out and bring in another unrelated dog.
SPEAKER_01So I know there's tests that you can order now that you can cheek swab or get saliva from your dog, and you take a mixed breed dog, and it'll tell you where they came from. Is it golden retriever? Is it a um golden doodle? Is it a 10% this breed and 20% that breed? Um are these tests getting better? Are they good now? Do you think um it provides information?
SPEAKER_00Yeah, they're actually pretty good. Um, and uh I think people are really interested in what their mix might be and what sort of combination of breeds it might be. Um, I think it can give people some information about expectations, about you know, how big that dog might be or what it will be like. Um and in addition to telling what Nix as a dog is, um, a lot of these Pamel tests will also test for potentially inherited diseases. Um, so you know, each particular breed might have four or five tests for inherited diseases for that breed. And if your dog is a mix of a couple of different breeds, then you can find out about those inherited diseases that those breeds may have. So they would run specific tests based on what breeds they have found your dog to be and get like a golden retriever panel and a Labrador panel, or no, the companies that run tests that will tell you what breeds your dogs are run a panel of um health tests and they always run the same panel. So there are other companies that have put together specific panels for breeds, um, and they don't all necessarily run the same tests.
SPEAKER_01But yeah. And when you get, let's say, a positive result. Now, does that mean that dog's going to develop the disease in its lifetime? Is this something maybe you don't want to know?
SPEAKER_00Um, well, hope I would assume that the information would be helpful for their health care. So it depends on what the mode of inheritance is. If it's a recessive trait and your dog just has one copy of it, then it it will normally be fine. Um, if it's a recessive trait and they have two copies of it, then you need to worry about that disease and talk to your veterinarian about what the results are so that the veterinarian can guide the health care of the animal.
SPEAKER_01So, Danica or Dr. Banish, you know I'm basically a cancer guy, right? And I'm mostly clinician. I like to do some other things. So I'm going to ask you some basic question there. So, what's the difference between the genes that are passed on and let's say the genetic changes we see in a cancer cell?
SPEAKER_00Yeah, so it's one of the things I teach the first-year BET students is the difference between those. So the genes that you pass on are exactly that. The allele, the bad allele, is passed on through the eggs and the sperm and can go to the next generation. That's really different than the mutations that occur that cause cancer. Those mutations accumulate in a cell, let's say a bone cell. Um, and when enough mutations accumulate, and mutations and cancer-causing genes, then that cell gets transformed into a cancerous cell which grows and makes a tumor and makes cancer. So those aren't necessarily passed on, although there are inherited cancers where a predisposition to having cancer can't be passed on.
SPEAKER_01Yeah, and that's obviously a lot of more work needs to be done in that area. And I wanted to go back to coat color a second. So you had said that we've kind of figured out coat color in individual breeds a lot in the dog. And has coat selecting for coat color led to disease? You know, if we are selecting for one thing, often you pull unrelated genes along that might lead to disease.
SPEAKER_00So there are definitely some coat colors that um have some bad things associated with them. Um one in dogs and a couple in other species and horses and cats. But um the one that comes to mind right away in dogs is myral coat colors. So myral is that variegated gray. Um, if it's on a black background, uh gray spots that you can see. And if um dogs have particular elections. Alleles at that gene. And the dog has two copies of them. Sometimes they can be born deaf or blind. And so this is a case where the variance associated with the change in pigment also affects hearing and development of the eye.
SPEAKER_01So this is just kind of one disease. So while coat color is pretty interesting, it's just a small piece of the overall puzzle of looking at disease and looking at breeds and if there's breed predilection for disease.
SPEAKER_00Yeah. And most of the time the coat color isn't associated with problems. It's just uh something that you know people have preferences for.
SPEAKER_01So about how many genetic diseases in the dog have we identified and can we test for now? Do you know do you have any idea of that? I know that's a big broad question.
SPEAKER_00Yeah, we're uh quite a few. Um I think we're up to about four or five hundred. Um four or five hundred.
SPEAKER_01Wow, that's a lot. That is a lot.
SPEAKER_00Yeah.
SPEAKER_01And how is that compared to, let's say, people? Do you know? Are there multiple diseases also that have been identified in humans? Or I know you're not a human geneticist.
SPEAKER_00I'm not a human geneticist. Um I think that you know there's a lot more that's been done in people. Um and and we do things a little bit differently in people, so we routinely would sequence people to try and identify what variants they might have. Um the interesting thing is that people aren't inbred. So they don't all have the same variant in a gene that causes the disease. They will have different variants. But in the purebred dogs, we always are looking for just, I mean, rarely is there more than one variant in a gene that causes the same disease.
SPEAKER_01So the alleles, which are gonna vary across individuals, sometimes they're good, or you could have multiple good alleles, and occasionally one of those alleles might actually lead to a disease. Yeah. Yeah. Okay. So over the years I've worked here, I've heard some really great work uh has come into your lab. And I think the first one I heard was the first one I heard of was about dalmatian dogs, particularly male Dmatian dogs. So I know dalmatian dogs carry a genetic variant that leads to hyperurecemia or hyperurikemia, right? And where they excrete high levels of uric acid into their urine, which makes stones that accumulate in the bladder. And if they accumulate in the bladder, they can try to urinate them out through their urethra or the tube that goes from your bladder to the outside, and they get stuck. And this would make male domations in particular, you know, if they get stuck, then your bladder gets bigger and bigger and eventually ruptures and it's fatal. So can you talk to me a little bit about this story and kind of talk to our listeners about it? Because I it's a really cool story.
SPEAKER_00Yeah, yeah, it's a it's a really neat story. I um I used to have Dalmatians. Um, they they were my favorite breed. Um and uh we had a urologist here who was a professor of mine, Dr. Ling, who I remember him. Yep, who uh encouraged me to try and figure out, you know, why Dalmatians have this hyperurcoseria. Um and this was a really long time ago. So before any whole genome sequencing could be done in dogs, and before the genome assemblies were made, and we were really fortunate that a um dog slash mouse geneticist um did a cross. So he crossed uh Dalmatian to a pointer, and then he back crossed to Dalmatians. And each generation he selected for low uric acid. Um and so this cross was actually done in the 1970s, and I had always known about it. Um, and uh for a long time it was really just one person who was doing this cross. Um, and some Dalmatian breeders in California decided that they also wanted to participate in the cross and try and get rid of the bad allele of this gene.
SPEAKER_01And so they were still maintaining a Dalmatian in a sense, but they were breeding in another breed to fix this problem.
SPEAKER_00They just did one cross to a pointer and they were taking the normal allele, and each generation they were selecting for the normal allele as they crossed to Dalmatians.
SPEAKER_01But they couldn't read the gene yet, so they were just finding the dog who didn't make the increased uric acid in the urine and then breeding them back in.
SPEAKER_00And anyone who has dalmatians, purebred dalmatians, that don't have the normal gene, um, it's really hung. You can take Dalmatian urine and put it in the fridge and it'll turn to sand. So it's pretty easy to phenotype them. And he he also did some special testing of the urine, but nonetheless, he did this for 14 generations. And then Denise Powell, who was a breeder in California, decided that this was something she was really passionate about, and she started breeding these low uric acid dogs and allowing us to collect samples from the dogs, including urine and a little DNA sample. And we use that information to identify the region of the chromosome where the gene was located, and then eventually the gene and eventually the mutation. So nowadays I could do that project in oh, I don't know, a couple of hours. But um so a different approach. But the project at the time uh took my graduate student a couple of years, um, but eventually we did find the gene and the mutation. Once we had the mutation, we confirmed that all Dalmatians had the mutation, unless they were crossed with pointers. And eventually the um Dalmatian Club of America and the American Kennel Club allowed those dogs to be registered. So now you can actually get a Dalmatian that doesn't make bladder stones, um, which is pretty exciting.
SPEAKER_01That's huge, right? So you you breed out the problem then.
SPEAKER_00Right. So you can breed out the problem. The other thing that we found out about this mutation is that while it was known to be um in Dalmatians, um, we also had reports of it being in bulldogs and jackrossel terriers and weimarners and black Russian terriers. But as these testing companies have tested more and more dogs, it turns out that it's the fourth most common mutation in all dogs that are tested. Um, and so knowing what the gene is and the mutation is actually helpful for all dog breeds because there are other dog breeds that have this. They just don't have it as commonly as the Dalmatian did.
SPEAKER_01So a breeder could go get these tests done of their dogs that they're gonna breed and hopefully try to avoid that and breed out disease. Yep. That's that's really cool. So is it unusual that one gene causes a problem or some diseases caused by multiple genes of their interactions? So are the risks if we fix one problem that we create another, a kind of genetic whack-a-mole almost?
SPEAKER_00It shouldn't be. There were there was definitely concern when we started offering genetic tests to dog breeders that um this could cause rises in allele frequencies of other bad alleles. But um, as long as breeders, even though the dogs are inbred and there isn't much you can do about that now, you can continue to try and um breed relatively unrelated animals to each other and do testing each generation. So while you're eliminating one allele, you shouldn't necessarily have this sort of sweep.
SPEAKER_01The knockoff effect. So these tests can be really useful in saving a breed. I was just wanted to kind of wrap things up and ask you where do you see things going? What is the what do you look for in the future of genetics in dog breeds and genetics in in veterinary medicine?
SPEAKER_00I think we've gotten really good at solving the simple Mendelian traits, so traits where it's just one gene involved. Um, and I think the future is going to be tackling some of the more challenging questions in um dog health. And so that's cancer, autoimmune, and inflammatory diseases, and the one that we really haven't solved, which is epilepsy. So all three of those um diseases are um definitely not one gene. There might be some examples of some seizure disorders that are caused by one gene, but a number of dogs. Yeah, small number of dogs. But idiopathic epilepsy is um not understood at the genetic level, and yet dogs like to have seizures, for example. And the same thing with cancer. There have been some alleles identified that seem to um carry a higher risk of cancer, but they don't explain the risks of breeds that have high cancer rates. Um, so I think setting those more complex diseases is what we're going to see next. And I think that's really exciting.
SPEAKER_01This is really exciting, and I really enjoyed speaking with you about this today. I I'd really love to have you back so we could talk about more examples of genetic diseases. There's there's just so much that we've learned and so much more that we need to learn. So thank you again, uh, Dr. Banis, for joining us today.
SPEAKER_00Thanks for having me.
SPEAKER_01The vetrospective, as with life, takes a village. I want to thank those who suggested I start this project and everyone who has encouraged and supported me along the way. Particularly, I want to thank our producer and director, Dene Blythe Unti, Nancy Bay, who is our program coordinator, our sound mixer, Andy Cowett, and theme music was composed and produced by Tim Gehagen. Thank you all, and we'll see you next time.