University Place

A Dog Model of Degenerative Polyneuropathy

[Tom Zinnen, Outreach Specialist, Biotechnology Center, University of Wisconsin-Madison]

Welcome everyone to Wednesday Nite @ The Lab. I’m Tom Zinnen, I work here at the UW Madison Biotechnology Center. I also work for UW-Extension Cooperative Extension, and on behalf of those folks, and our other co-organizers, Wisconsin Public Television, the Wisconsin Alumni Association, and the UW Madison Science Alliance, thanks again for coming to Wednesday Nite at The Lab. We do this every Wednesday night, fifty times a year.

Tonight’s rather special, because this is a Wednesday night about the lab, and this is about the Labrador Retriever, and a particular type of degenerative polyneuropathy that that breed, and other types of dogs get. Our speaker tonight is Susannah Sample. She’s with the School of Veterinary Medicine.

She was born in Hinsdale, Illinois, and grew up in La Grange. She went to Lyons Township High School, in the suburbs there. Went to Northwestern University and got her undergraduate degree in biomedical engineering. Then she came up here to Wisconsin, where she got her Masters, her D.V.M., and her PhD, and then, just last year, she became a board-certified veterinary surgeon. So, I’m looking forward to hearing about One Health Medicine: A Naturally Occurring Dog Model of Degenerative Polyneuropathy. Please join me in welcoming Susannah Sample to Wednesday Nite at The Lab.

[applause]

[Susannah Sample, Assistant Scientist, School of Veterinary Medicine, University of Wisconsin-Madison]

Well, thanks for that introduction. To go through a quick outline of what I’m going to talk about tonight, I was going to start by discussing a little bit about canine genetics, and genome wide association studies, followed by a discussion about this condition that I’m studying in dogs which I call Acquired Laryngeal Paralysis Polyneuropathy, some people would just refer to it as Lar Par, more commonly. And I’m going to go through a few basic definitions about genetics and genomics to get everyone on the same page to explain what we’re doing and why, and then discuss our genetic study, and a little bit about future directions.

So, to start, when you talk about canine genomics and genome wide association studies, dogs have really become more prevalent in this area, and so I’m going to talk a bit about why dogs are really great models.

In 2010, there was this paper that came out with the title Man’s Best Friend is Biology’s Best in Show, and I just think that that’s a really great encompassing title to talk about the idea that dogs are really important when it comes to comparative genomic research. And the reason for this is that when you talk about pure bred dogs, they are a very highly inbred group of animals. Selective breeding has created over 400 breeds, and most of this has happened within even just the last few hundred years, and as a consequence of this, we have a very high prevalence of breed specific diseases.

Ultimately, this really, when you think about it, is humanity’s greatest genetic experiment. It started, you know, thousands of years ago, when we first took gray wolves and domesticated them into what’s referred to now as village dogs, and over the last few hundred years, taken those dogs and created dog breeds to the extent that you end up with, you know, situation where you’ve got this little guy and a Great Dane, that are actually the same species, and when you really sit back and think about that, it’s a pretty amazing phenotypic or – or visual diversity within – within dogs.

So, importantly, when you think about canine genomics or genetics, the canine and the human genome are actually very similar. So, at this point, about 92 percent of genes that we know of in people have been mapped in dogs, and that’s obviously increasing over time.

Another important feature is that people and their dogs share a common environment, so environmental influences are obviously quite similar, and most diseases that we see in dogs do have a corresponding disease in humans.

So, talking about one of these diseases is acquired laryngeal paralysis polyneuropathy. So, like I mentioned, in the world a lot of people call this disease Lar Par, and it’s most commonly what we call an idiopathic degenerative polyneuropathy. So, what that means, idiopathic means we have no idea what causes it. Degenerative means, obviously, it’s something that degenerates over time, and polyneuropathy means a disease of many nerves.

About 70 percent of these cases occur in Labrador Retrievers. The next most common breed, at least in the United States, are Golden Retrievers. Rarely, dogs can get laryngeal paralysis from other causes. Most commonly, that’s going to be from a tumor, maybe in their neck, or from some sort of surgical trauma. Technically speaking, endocrine diseases like diabetes or hypothyroidism could cause a disease like this, but that’s not really been well-documented.

In Labrador Retrievers, the age of onset or presentation is very typical. So, they tend to be about 11 maybe 12 years of age, and owners usually report that they’ve noticed a change over the past year, and it strongly resembles certain neuropathic conditions in people, so the example I’m going to use tonight is a disease condition called Charcot-Marie-Tooth disease.

So, talking a little about Charcot-Marie-Tooth, it’s the most common inherited polyneuropathy in people, or peripheral neuropathy in people, and what that means is that it’s a disease of nerves that aren’t necessarily coming in your spinal cord or your brain, but they – they come from those places. So, they’re what’s innervating your body from your cord or from your brain.

And it’s actually, relatively common. It affects about one in 25 hundred people in the United States.

It ends up having a progressive degeneration of motor and sensory nerves, so these people tend to experience muscle atrophy. They have chronic pain. They have hand and foot deformities, like you see in that picture there. And interestingly, and importantly, it’s the longest nerves in the body that tend to be the most affected. So, in people, those nerves are usually what innervates your distal limbs, so your hands or your feet, obviously progressing over time.

There are a lot of different subtypes, so when you think about Charcot-Marie-Tooth disease, you have to realize it’s an umbrella term. So, its – it’s a disease term for a typical presentation that is caused by a lot of different genetic mutations. So, I may have C.M.T., and someone else may have C.M.T. in the audience, but what’s causing that would be two different genes.

So, at this point, there’s over 80 genes that are associated with it. Now, there are some that are a lot more common than others. So, there’s four genes responsible for most of the genetically diagnosed versions of this disease, but not all cases have been found – we haven’t found a genetic cause for every type of case.

The other important part of this is that most forms are inherited in an autosomal dominant fashion. That’s not always true, but the vast majority of them are a dominant disease.

There’s no disease modifying therapy currently available for people with this condition, and so the treatment ends up being symptomatic, so we support – help to support people with disabilities, or you treat muscle cramps, or spasms, or pain.

One of the big concerns you have with any time you are dealing with a – a neuropathic disease is that when you talk about drug trials there’s obviously a lot of ethical concerns that come up about the idea of trying drugs that affect your nervous system, particularly when you’re dealing with children. Now, Charcot-Marie-Tooth can affect children, but generally, it’s thought of as being a middle-age to older disease because it’s progressive over time, so people tend to start having signs that really affect their life when they’re a bit older.

And like I said, not all genetic underpinnings are identified, so one question we have is: Is this disease we talk about in dogs, a potential model for a condition in people?

So, let’s talk about laryngeal paralysis. Like I said, when you look at this clinically you can see signs of a peripheral polyneuropathy, so again, nerves not from, not your cord or your brain, but nerves coming from those places, that are affecting those nerves that are longest in the body. Now, in the dog, the two longest nerves you have are the recurrent laryngeal nerve, which, I have no idea why, but it – it’s true in people too, this exits your – your brain, it comes down, it wraps around your heart, and then comes all the way back up to innervate your larynx. So, in a giraffe, this thing’s like eight feet long.

[laughter]

It’s the most redundant nerve ever.

And the other long nerve you have is the nerves in dogs that innervate their hind limbs, those are another very long nerve. So, when you look at these clinical findings, dogs end up with laryngeal paralysis, which is their larynx, or voice box is paralyzed. They can end up with respiratory distress. Their esophagus isn’t going to have correct mobility anymore. And you end up with problems in your hind limbs too. So, clinically we see hind limb weakness and decreased reflexes in the back leg. So, I’m going to talk a little bit more about all of those in a moment.

But clinically, when people come in with their dogs having this disease, the first thing to realize is that dogs don’t ever exhibit symptoms, they exhibit clinical signs. So, what that means is a dog can’t tell you if when they’re eight, if they’re feeling tingly feelings in their paws, right? They can’t tell you if their stomach hurts, or if they’re nauseous. Instead, they vomit, right? So, they show you the clinical sign that they’re not feeling good, but they can’t tell you, you know, three hours beforehand like: I’m not really feeling so good. So, consequently, we don’t really have a way of identifying this potentially earlier on until there’s really a major problem, where these nerves aren’t working at all.

The typical progression when owners report in Labradors, is usually relatively slow from months to years, they start noticing a few changes. And usually it’s the changes associated with the voice box, or the larynx, are the most noticeable. So, oftentimes dogs have a voice change, or instead of barking normally, they have a very hoarse bark. Maybe they won’t want to exercise quite as much. They end up with very loud breathing, and in very severe situations, they can end up in extreme respiratory distress. They can turn blue, which we call cyanosis, and potentially even collapse.

Hind limb weakness is also often noted when you start talking to owners about it, but unfortunately, it’s often misinterpreted or misdiagnosed. These are older dogs, so the hips get blamed a lot. That they have hip dysplasia, or old-dog arthritis, or they’re just slowing down, when really, they – they have a neuropathy that’s affecting their mobility.

On examination findings, you see severe upper respiratory noise, and like I said, some neurologic exam findings, you can find. So, I’m going to go through all of these. They have absent paw replacement tests. So, I’ll show you what that means. Decreased withdrawal reflexes, and I’ll discuss that and then also show you some images of a dog with hind limb weakness.

So, the severe upper respiratory distress, or noise and potential respiratory distress, is something that you can generally hear these dogs coming down the block. They sound something around Darth Vader slash train.

[laughter]

And so, I’m going to play a video of this, and you have to realize that there’s a dog – it’s not like this dog’s just been exercising. He was just sitting down and stood up. So, you get a sense of how noisy he really is.

He’s very excited.

Shake it off, shake it off. Shake it off.

So, you know, later on in the disease, these dogs are actually relatively obvious when they have the disease, ’cause you can hear it.

Other things that you see are, laryngeal paralysis on exam. So, these are images of a larynx and they’re taken through a small camera that we actually use to look inside abdomens of dogs, so the resolution’s not great, and I apologize, but to get you oriented, we’re looking down the throat of a dog. So, up here, is the esophagus. That little crease right here. This is the trachea inside that dark hole, and these are cartilage that is associated with the voice box. This is a tongue depressor.

So, to give you a sense of a normal dog, when they breathe in, you’ll see that –

[plays wrong video]

whoops, that’s not it.

When they breathe in, the larynx opens, right? So, normally when you breathe in, these, sort of, double doors open up, and then close. So, I’m breathing in, I’m breathing out. I’m coughing, I’m breathing in, I’m breathing out.

So, in a dog that’s paralyzed, I don’t know how well you can see this, but, there’s a little bit of mucus right here, and so you can see that move when the dog breathes, and you’ll notice that the larynx is not moving at all. So, he breathes in and nothing’s really happening. And that’s all to do with this recurrent laryngeal nerve.

Another sign we see, are the paw replacement tests. So, normally, if you take the hind limb of a dog, and their paws, and you flip them over, they should flip right back. So, dogs that have this condition, oftentimes, don’t do that.

[playing video of paw replacement test with sick dog]

So, you can see that this dog, if you flip his paws over, it’s just not really connecting that that’s happened, and they don’t flip them back.

Another sign, and this is a little hard to see, but it’s decreased withdrawal reflexes. So, I’m going to show you a normal dog, and then I’m going to show you a dog that has decreased reflexes. So, this is a dog that’s -a withdrawal reflex.

[playing video of normal dog]

So, what I’m going to do first with this dog is show you what a normal range of motion – range of motion is, so this is what he should do. You pinch his toe, he should pull back, and all of his joints should flex. So, you can see that his ankle flexes, his knee flexes, and his hock – and his hip flexes. It’s the same – its the same reflex that if touch a hot stove, you pull your hand back. That’s not something that goes all the way up to your brain and back, that’s a local reflex, and if you ever have the opportunity to do this,

[laughter]

you’ll notice that your – your – your wrist, your elbow, and your shoulder all flex back, so that’s normal.

So, in a dog that has this condition, what ends up happening, oh

[mistakenly advances slide]

is they go away.

[returns to correct slide and plays video of sick dog]

There. So, I’m going to again, I’ll show normal, so that’s how he should flex his leg, and what you’ll notice is he kicks back when you pinch his toe. So, he doesn’t flex this joint correctly. He’s really not flexing it.

The computer’s touchy.

And he kicks back. And that’s because his hock’s not innervated, or his ankle isn’t having normal innervation.

Lastly, you start seeing, hind limb weakness. So, when you watch a dog walk normally, you can notice that, he’s trotting in a very, sort of, graceful way.

[playing video of a normal dog walking]

He’s flexing his joints correctly. When he puts his feet down, they’re not totally under his body, but they’re – theyre, you know, in an appropriate place. That’s really a normal trotting dog. Now, when you see a dog that’s weak, what you’ll notice, is that he walks in a very stiff-gaited manner.

[playing video of a sick dog walking]

He’s not really flexing his joints correctly. He’s maybe a little bit more of a wide based stance, and that all just has to do with him being relatively weak.

So, these are very classic signs that we can see on neurologic exam.

Unfortunately, this can often become an emergency. Particularly when you have owners who just aren’t aware that their dog has this condition. They don’t appreciate that maybe these things aren’t just old dog changes. This particularly happens when dogs are excited, and it’s high temperatures. So, their respiratory rate goes up, they start panting. That airway flow, and flow of oxygen going by actually causes some trauma almost to the laryngeal mucosa, which is just like the cheek, like inside your cheek, so you can imagine it swells relatively quickly. It becomes inflamed, starts to swell, the dogs aren’t breathing so well. Now, they’re really upset, and you end up in this sort of vicious cycle, which can, in very severe cases, end up with what we call severe dyspnea, in other words, a very difficult time breathing. They can end up with cyanosis, they turn blue because they’re not getting enough oxygen. They can collapse. And because dogs cool themselves so much panting, they can end up with hyperthermia, and actually die of heatstroke.

So, unfortunately, this really is a life limiting condition. The most common problem we have with dogs that die of this disease is that they get aspiration pneumonia, and what that means is essentially food goes down the wrong tube, and whether it’s food going in, or actually most likely, what’s more likely is their actually getting regurge, or esophageal reflux, because their esophagus isn’t working quite right, and now, their larynx, which protects their airway isn’t working, and you can imagine, that they can end up with what we call silent aspiration. So, they can end up with pneumonia. Sometimes, that can be catastrophic to the point where there’s really not much we can do, and sometimes it’s something that just happens over, and over again. And you have these older dogs, that are in the hospital a lot, and they’re not able to live their dream.

On the flipside, you can have severe laryngeal swelling, like I talked about, where you will have dogs that die of heatstroke, or you have dogs that die literally of suffocation. And sometimes, what ends up happening, is these dogs just degenerate to the point where they don’t have the quality of life that they want to have. They can’t chase their squirrel, you know, they can’t go do the things that makes them wake up every morning, and owners just say, he’s not doing what he needs to do to be happy.

So, in terms of treatment options, there’s, like Charcot-Marie-Tooth, there is no disease modifying treatment available. We have two options. We have conservative management, and then we do have one surgical option. So, conservative management usually involves weight loss to keep these dogs from getting so hot. Stress management reduction so they don’t get too excited. Exercise restrictions, obviously, so they don’t end up panting so much, and avoiding high temperatures. Surgically, we can go and open up their airway to a degree, which essentially gives them sort of a rescue tube to breathe through, so hopefully they’ll never asphyxiate, or suffocate, or they’ll never have a heatstroke event.

And there’s a number of different ways to do this, and if you put ten surgeons in a room, you’ll probably get eight opinions about which is the best. But generally, what’s done, in 2016, is a cricoarytenoid lateralization, which is commonly referred to as a tie-back surgery.

There’s other options as well, but what it essentially does is if you look at these images here. So, we have the larynx again, up here is the esophagus. Trachea again is here. So, if you look at this opening in the trachea, what these procedures essentially do, is open that up to a degree, so you can see that this area, which we call the glottic opening, is a little bit bigger. And it just gives them a rescue, so, they don’t end up in those really dangerous situations.

We talk about prognosis with these surgeries. Aspiration pneumonia is a complication. And if you read through the literature of different lateralization procedures, or opening procedures, on average, somewhere around 15-20 percent of dogs probably get aspiration pneumonia after the procedure in the immediate, sort of, few days. Now, not very many dogs die of that, they just need to be in the hospital a little bit longer. But you can have catastrophic aspiration events, as a result of this surgery. They’re very rare, but as soon as you see one happen, it makes you nervous.

Unfortunately, we have no pre-operative means of determining which dogs are going to get aspiration pneumonia post-operatively, so there’s not really a good way to recommend to owners whether or not this surgery is something that’s going to be more dangerous for their dog versus another. But you also need to keep in mind, that aspiration pneumonia is also common in dogs that don’t have this surgery, so it – it’s kind of a complication of the disease in general.

And these are two images of a thorax of a dog that shows you aspiration pneumonia. So, this is a normal dog. So, when you’re looking at this, the head’s this way. His little butt’s that way. This is his heart. So, you’re looking at his lungs, and this is his trachea coming down. And that’s – thats really a normal thoracic radiograph. When you look at this dog, what you can see is that there is this area that is white, and what looks almost like a tree coming out of that, and those are what we call air bronchograms.

And essentially what causes that is when there’s fluid in the lungs, the bronchus, which still has air in it, contrasts very well on these radiographs, and you can see sort of this tree-like pattern that occurs and it’s a very classic radiograph of what we call aspiration pneumonia, or any pneumonia.

So, the key points here is that laryngeal paralysis in dogs really is a devastating disease. It can be life-threatening. The clinical features indicate that it’s a peripheral neuropathy. It’s similar to a number of human degenerative neurologic diseases, one of which I discussed with you today, and it really does have a potential to be a naturally occurring animal model of human peripheral neuropathies.

I’m going to take a quick drink. This is one of the dogs in our study. He’s very handsome. And apparently a winner.

[laughter]

Alright, so I’m going to talk a little bit about genetics. So, stick with me. I – I’ll try to make this as painless as possible.

So, we’re going to go back to like fifth grade, and D.N.A. So, D.N.A. is the inheritance information of your body, and each strand is essentially made up of four bases. So, A, G, T, and C. And each base, essentially creates what’s called a nucleotide. So, a given nucleotide would be like, A, and when you match A with its pair, which is T, you get a base pair. So, A always pairs with T, C always pairs with G.

That’s the basic you need to know about the nucleotide, the base pair, and D.N.A.

Secondly, the genome. So, the genome is the entire set of D.N.A. for any animal. You have to realize that about 99. 9 percent of it is exactly the same in a given species. The caveat of that is point one percent is actually a lot of D.N.A., because the D.N.A. is – is huge, right? [coughs]

So, when you talk about D.N.A., oftentimes, we’ll discuss alleles, and alleles are alternative forms of D.N.A. at a D.N.A. segment at a given location in the genome. The way I always like to think of this conceptually is that, from your mom and your dad you get two alleles, right? So, there – you’ve got two, one from each parent. When I refer to the word locus, which I might slip into every once in a while, that essentially just refers to a place in the D.N.A. Alright, it’s like a house, a location, a locus.

Okay. I promise this is the last one. Genotypes versus phenotypes.

So, a genotype is a pair of alleles at a locus, and so you can be either a homozygote or a heterozygote, and a homozygote is when you inherit two identical version of the allele from your parents, or heterozygote is when you inherit two different alleles, and the phenotype is the observable characteristic as a result of the genotype. So, just to bring back freshman year in high school, a great example of this is eye color. So, when you look at brown eyes versus blue eyes, you probably all, like, drew this chart back in the day. You can either be a homozygote major, where you’re big B big B, and you get brown eyes. Brown eyes being a dominant trait. You can be a heterozygote, where you’re big B little b. You end up with brown eyes, because brown is dominant. Or you can be a homozygote minor, which is the minor allele, which is little b little b, and you end up with blue eyes, because blue eyes is a recessive trait.

So, now that we’ve bought that flashback, let’s get into something more fun. So, genome wide association studies – the entire point of this is to come up with a way to scan the genome and find a biologically significant variations that contribute to a trait or a disease, and when you think about this, it’s really a very nascent field. So, it wasn’t even really thought of until the mid 90’s. The human genome wasn’t sequenced until 2001. It wasn’t until 2005 that the first human G.W.A.S. came out. That same year the canine genome was sequenced, and in dogs, our first G.W.A.S. came out six years ago. So, this is really um, a relatively new field, but it’s very rapidly evolving because it’s been, I think realized, that it’s a very powerful technology.

So, how this works. I’m doing to talk about SNiPs and L.D. in more – in more detail in a minute. But, essentially, there’s commercial arrays out, we call them SNiP chips, that have a whole bunch of these SNiPs scanned across the genome, and you send in D.N.A. for your SNiP chips, and you get a whole bunch of data back, and you can analyze them, and there’s two real ways to approach G.W.A.S. You can either do a case-control association, like I’m doing, so the dogs either have laryngeal paralysis or they don’t, or you can do quantitative trait – trait association, so for instance height. So, there are some genes that are more correlated with being tall, and other genes that are more correlated with being short.

And the power of a SNiP to detect an association is dependent on something called linkage disequilibrium. So, SNiPs in a simplified fashion.

They’re variations in a nucleotide base at a single locus in the genome. So, in one spot in the genome, there are variation in this A, C, T, and G, that’s recognized within a species.

So, if you look at this here. Essentially, this is, say, a strand of D.N..A, and here is three subjects, and here’s a bunch of SNiPs. So, essentially these arrays take a bunch of SNiPs and they put a scan throughout the genome, and I’ll get back, say three subjects, so at SNiP one, this – this subject is an A-T, and this subject’s an A-A, and this subject’s an A-T. And then obviously you have that, except for a few hundred thousand, or a million SNiPs.

Some of these SNiPs are biologically significant, and some of them make absolutely no difference. So, these studies, essentially, use these SNiP markers, and like I said, they’re just spread out across the genome.

So, what about linkage disequilibrium, like, what – what – what does that mean? So, when you think about linkage disequilibrium, the easiest way to conceptualize it is to realize that there are chunks of D.N.A. that are inherited together. It’s easiest to think about this in a linear fashion, like D.N.A’s one big strand, and there’s big chunks that are inherited. In reality, sometimes these aren’t even on the same chromosome. Sometimes, they’re pretty far away from each other on a given strand of D.N..A, but when you’re thinking about it for now, just think about ’em as a chunk of D.N.A. that’s all inherited together.

And you have very long chunks of D.N.A. in new populations, and shorter chunks in old populations, and we’ll talk about that in a moment.

The other thing to keep in mind, is that pure bred dogs are not bred randomly, right? There’s nothing natural about a stud dog being bred 400 times in its life. So, you know, the lifetime matings for given dogs can be very high, and it just really highlights the idea that pure-bred dogs, their genetics, and – and breeding are very controlled.

Alright, so this is the tough bit.

These are called triangle plots, and they’re really the easiest way to describe linkage disequilibrium, and the way to think about this, is it’s like a heat map. So, along this top area here are a bunch of SNiPs. So, R.S. is a common, and then a bunch of numbers behind it, is a common way of identifying a given SNiP.

And so, within each one of these, what you can do is connect them, so, here’s SNiP number three, and we’re going to connect it to SNiP number nine, and you can see, because it’s sort of a reddish color, those two are in some degree of linkage. In other words, they are often inherited together. Alright, so the more red you have within these plots, the more linkage disequilibrium exists.

So, to put that into context, when you look at human populations, you can see newer populations, like Caucasians, which in the history of humanity, Caucasians are actually a very young population, versus an ancient population like the Yorubans, and when you look at the difference in these maps, you can tell that the Caucasians have a whole lot more red than the Yorubans. So, newer populations have higher linkage disequilibrium – in these heat maps, they have more red, because more of these chunks of D.N.A. are inherited together, and associated with each other.

So, that’s people. So, what about in dogs?

Well, when you look at a Caucasian, you’re like: Yeah, there’s a lot of red there. And then you look at a pure-bred dog, and it’s like red-out, right? So, you know, pure-bred dogs have extremely high linkage disequilibrium, and that becomes important when you’re doing genetic studies, because within one of these blocks or chunks of D.N.A., when you’re looking at SNiPs and you’re trying to find a disease, you only need, theoretically, one SNiP within that block to determine – to – to tag a disease locus.

So, within here, if you’re saying, here’s my disease, so in humans the disease is here, and in dogs it’s here. In order to find that disease in people, because they have smaller linkage disequilibrium, because their D.N.A. isn’t inherited in as big of chunks, you need all of these markers to try to find that – that particular mutation, versus a dog, because you have bigger chunks of D.N.A., and you only need one of these markers in each chunk, you end up with needing a whole lot less markers.

And this is really reflected in these BeadChips, or SNiP chips that I was telling you about, that you can get commercially. So, in people, their SNiP chips have over 4 million SNiPs across the human genome versus dogs – we’re like really excited because we just got bumped up to 220 thousand. So, you know, when you’re looking at that, it really just gives you a sense of how fewer, you know, markers you really need to look at this in dogs.

So, the key points of this is that genome wide association studies are a means to scan the genome for variants that associate with a given disease or a given trait. And the power of G.W.A.S. is really influenced by linkage disequilibrium, so dogs have a very high linkage disequilibrium compared to people, and what this means is that when you’re studying diseases in dogs, you need fewer sample sizes, you need fewer dogs enrolled, and you need a lot less markers across the genome to try to find a problem.

This is Botchy. He’s fabulous. Okay.

So – I’m now going to talk about Mendelian versus complex traits. I promise this gets exciting in a minute.

[laughter]

Wait till you see the videos.

So, when you talk about Mendelian disease, these are diseases that you either have the gene or you don’t, so on this sort of map here, and you’re saying: Well what gives me this disease? Well, Mendelian diseases are pretty much, your genetics gives you the disease, versus an infectious disease, say Ebola, where like, you – it’s in the environment, you kinda got it regardless of what your genes are. And sitting in the middle of this, are what’s called complex traits. And complex traits are gene or diseases that you may have genetic susceptibility to it, but you also probably have environmental factors, and other factors, that have, sort of, tipped you over the edge to getting the disease. So, an example of this is Alzheimer’s. Right?

So, there’s no genetic test for Alzheimer’s. There’s some genes that may predispose you to it more than others, but it’s complex, there’s a lot more going on.

So, this is my video.

So, wait for it.

So, this is our dog, and right now he’s on the scale of being disease negative, alright? So, he has no disease. So, when you’re talking about Mendelian diseases, right, you have a gene, and it gives you the disease.

[in video,ball (gene) is dropped into empty bowl causing bowl with toy dog to lift higher]

So, that mutation just pulled him over, and now our dog is diseased.

When you talk about complex diseases, which are a little bit more difficult, essentially you have a whole lot of genes that are kinda contributing together, in sort of an additive effect

[In video, a cup of Skittles (various genes) is added to the empty bowl causing bowl with toy dog to lift higher]

or Skittles, depending upon your perspective, and they – they contribute together to create a situation where you have a disease, but that’s really not totally accurate.

Because, if you’re a dog, you may have all those Skittles, and all those gene mutations that are putting you in a position where you are susceptible but there’s going to be something in the environment

[in video milk bones are added to the Skittles side making it outweigh the toy dog side]

e. g. milk bones, that might tip you over to getting the disease, right? So, maybe obesity, or other environmental factors that really contribute to you getting that. So, it’s really the difference between monogenic or Mendelian traits, and complex diseases.

When we think about Mendelian disease, there’s a number of examples of these in people, like cystic fibrosis, or Huntingdon’s disease. There’s a paper that came out in 2015, looking at canine idiopathic epilepsy, which is a seizure disorder and idiopathic again, meaning we don’t know what causes it, where they used

[advances one slide too far]

whoop, getting ahead of myself

they used 157 cases and they had 179 controls. That may seem like a lot, but you have to realize that if this study was done in people, you’d have thousands, and we’ll get to that.

And essentially, they ended up doing a study, and – and what I want to talk about actually here is what – what this whole plot is. And so, it’s called a Manhattan plot, and I think it’s because it’s supposed to look like the Manhattan skyline, but each one of these little dots here is a SNiP. And along here, you’ve got your chromosomes. So, all these SNiPs have been tested, and you’re associating them versus dogs that have the disease and don’t have the disease, and is there a difference? And then on the Y-axis here is the P value.

And so what you can see is that, you know, you’re not really doing anything, there’s not too much of a difference, and then suddenly on chromosome 37, it’s like whoop, whoa, so there’s is something going on at chromosome 37, and in this area formed what we call a volcano. So, essentially, there were SNiPs that sort of came up and then finally one that’s like Hello, and that’s tagging the – the disease.

So, another example of this was squamous cell carcinoma of the digit, so skin cancer on the toes, and it happens in a number of – a number of dogs, and this study, they ended up actually finding mutation that causes this, and again, if you look at this, there’s 31 cases and 34 controls. It’s – its very few, few dogs, but in this case, they used pure-bred standard poodles, scanned the genome using G.W.A.S. and ended up finding an area on chromosome 15 of interest, and went further on, and actually did find the mutation.

Now what about complex diseases. So, this gets a little bit trickier. So, one of the, to me, most outstanding papers on complex disease that’s come out came out in 2014 on schizophrenia. And what they found was 108 associated loci, so when we talk about there’s a lot of different genes that contribute to this, this is what we’re, sort of, discussing.

And so, this line here is their – their line of significance. What’s particularly outstanding about this is that in order to do this study, they had, you know, however many million SNiP markers and over 150 thousand people in this study. When you think about how much data, energy, time and money it takes to do that, it is absolutely outstanding. So, if you have a dog model, you might be in a little bit better situation.

So, the key point here is that dogs really have a unique genomic architecture, and linkage disequilibrium in dogs, like we discussed, is a lot greater than it is in people, so again, you get smaller sample sizes, you have few markers. And we know that G.W.A.S. in dogs can be used successfully to identify regions of the genome that are associated with disease traits. And obviously it’s a big bonus, if you ever create – find a disease marker in a dog, that relates to a person, you now have a one health situation. You really benefit a lot of species.

Okay. This is my take a drink time.

So, getting back to laryngeal paralysis. It has a very strong breed predisposition, which really suggests a genetic basis. In veterinary medicine, it’s very common that there are certain breeds that get certain diseases, to the point where, if you know the signalment, in other words, if you know the age, the gender, the neuter status, and the breed of dog, and someone tells you what the – the clinical presentation problem is, you’re going to have a pretty good idea of what’s probably going on. I mean, there’s always unicorns, but there’s a lot of horses.

[laughter]

So, when you’ve got a breed that is very commonly gets a disease, and no other dog breed really gets it, you have to be thinking to yourself: You know it seems like it’s probably something genetic. And we look at pedigrees of dogs in this study, it really does indicate that laryngeal paralysis is heritable in – in the Labrador.

So, we’ve done a G.W.A.S. and right now, we’ve got 81 Labrador Retrievers. So, we have 60 cases, and 21 controls. And you’ll understand why we have more cases than controls in a moment, but the cases were all seen at U.W. Veterinary Care. They received either surgical treatment for laryngeal paralysis, so they had that tie-back surgery, or they had clinical signs that were very classic of the disease. Then we had 21 controls. So, these dogs are all – I think they’re actually 12, but 11 and a half to 12 years of age. They’ve been checked with us at least twice. They have no signs of – of respiratory obstruction. They have no signs of progression of neurologic disease. So, we’re pretty confident that they – they won’t be getting the disease. And all these dogs are screened -screened for relatedness. So, we get pedigrees from them, and if there’s any direct siblings we exclude them.

So, just to throw a Q-Q plot at you. Essentially, this is – this is one of the ways of looking at our results. So, what this is is an observed versus expected P values. The basics of this is, if you look at this line right here. I’m going to use my mouse. This is what – and each one of these little dots are SNiPs. Alright, so these are all what we kind of expect the SNiPs to be, and then, suddenly at the top, there’s a view that break away because what we’re observing is not what we expect, because they’re associated with the disease, because they’re different. And these SNiPs all lining up along this line like this tells us that we have corrected, statistically, for relatedness of the dogs within the study.

So, I’m just giving you a little bit of our Manhattan plot. This is just one chromosome. But what you can see here is that we’ve created this sort of, there’s actually, almost two volcanoes, but this is the one of interest. So, there’s sort of these volcanoes of SNiPs coming up, and there’s one that’s really significant and this lies in an area of the genome that there’s a number of genes that are pretty interesting, that could be associated with neurologic disease.

Perhaps more interestingly, is when you look at our phenotype-genotype table. So, along this side, these are cases of dogs that have the disease, and these are controls. So, like I said, we have 60 cases, we have 21 controls, we have a total of 81 dogs, and now we get back to our brown-eyed, blue-eyed. So, you can have homozygous major, in other words your big B big B. All of our cases, or all dogs that have this a major allele have the disease. And then you have heterozygous dogs, so big B little b, you know, you brown-eyed heterozygote, and in these, you got 28 cases, and you have 4 – 4 controls. And then little b little b we’ll get to in a minute. But what this map essentially, really strongly indicates, is that this is a dominant disease. So, just like with brown eyes, if you have one of these big A alleles, there’s a pretty good chance, you’re going to get the disease.

Now these homozygous minor alleles don’t totally match up, because there are a number of dogs that get the disease that you wouldn’t expect to based on their genotype. And there’s a lot of reasons this can happen with G.W.A.S. So, for instance, this SNiP is probably tagging an area of the genome, but it’s probably not the actual mutation that’s causing the problem itself. This could be more than one disease. And so just like in C.M.T., in Charcot-Marie-Tooth, where there’s a lot of diseases with a common clinical picture, it’s probably true in dogs too, that there’s a lot of diseases that could have a common clinical picture. My hope is that it’s not true within a given breed, but it’s certainly a possibility. You could also have, modifier locus, so in other words, a dog maybe needs to have this, or this contributes to part of a disease but there’s actually something else happening nearby that is – that is either supporting it or preventing the disease from happening. So, there’s a lot of reasons why you could end up with dogs in this situation.

The other issue is that our phenotyping might not be perfect. It can be difficult, especially earlier on in this disease, to be absolutely sure a dog has a condition, which is why we recheck these dogs fairly frequently.

So, what about the whole Labrador population. So, fortunately, because we do a number of genetic studies on Labradors, we actually have a whole set of dogs’ D.N.A. that has gone through these SNiP chips, that are not associated with this study. And a lot of these dogs admittedly are from the Madison and Wisconsin area, but they are actually nationwide. So, we have about 210 additional dogs that aren’t enrolled in the study, that were not recruited for anything to do with laryngeal paralysis, that are pure-bred Labradors, and if you look at their risk allele in this situation, you have the very unfortunate realization that this disease could be affecting even up to 70 percent of older Labradors. Now, clinically, when you go into a hospital and you see a lot of old Labs, that’s not surprising. There are a lot of dogs that get this disease.

So, the key point is that we’ve found a region in G.W.A.S. It has number of candidate genes, so we’re really excited about that. It does appear that this is an autosomal dominant trait, although we do need to do more statistical analysis on our pedigrees to really confirm that. And, at least in the dogs that we have in our lab – in our laboratory, you know, up to 70 percent of the population may be at risk of developing this disease in old age.

So, what’s next?

Well, we need to look at this region a little bit more, and so, we’ve just started an experiment looking at whole genome sequencing, which is really exciting, with the hope of eventually, not only identifying mutation but potentially even being able to create some sort of a genetic test for this.

We also need to evaluate this in other breeds, because like I mentioned, could this be a clinical phenotype where you have laryngeal paralysis that’s actually a result of a number of different genotypes, where there’s more than one disease causing this problem.

So overall, I hope that I’ve at least introduced you the idea of the comparative value between dogs and people with One Health Medicine, where dogs are really very valuable models for us. The huge benefit from a veterinary perspective is obviously, if you find a disease in dogs, and it applies to people, you know you’ve got – youve got a – a – a strong funding source to continue to look at this in dogs and – and you have a mutual benefit for both species.

I do need to disclose my funding.

And, you know, it takes a village to raise me. So, I’d like to thank all of my collaborators and mentors, but particularly all the owners and participants which come from throughout the United States and Canada, and mail – mail in their D.N.A. and help us out beyond what I can describe to you.

[slide of dog in veterinary scrubs appears on screen]

[laughter]

I’d like to thank you, and I can take any questions you might have.

[applause]