How Canine Genetics Can Lead to Cures for Diseases of Humans – and Dogs

An interview with a geneticist who compares the genomes of dogs and people.


“One of the conditions we’ve had the best success with so far is OCD — obsessive compulsive disorder. We conducted a study with Doberman pinschers, a breed that tends to get it, working with Tufts Animal Behavior Clinic Director Nicholas Dodman and his colleague Alice Moon-Fanelli. The first important step was phenotyping — identifying Dobermans who had the condition and also those who didn’t. It’s a very distinctive behavior in Dobermans. They chew on themselves. They put their own flanks in their mouths and suck.

“After Drs. Dodman and Moon-Fanelli distinguished dogs with OCD from those who didn’t have it, we were able look at their DNA and find a place on chromosome 7 where the cases and controls were really different, in a gene called CDH2. It’s already known where that gene is in people, but we didn’t know until now that it had anything to do with OCD.”

Elinor Karlsson, PhD, is talking about how the study of genetics in specific dog breeds sheds light on disease processes in humans, and how that might lead to treatments (and even preventatives) for both species. A geneticist, she has conducted much of her research at the Broad Institute, affiliated with both Harvard University and the Massachusetts Institute of Technology, and has recently moved to the University of Massachusetts Medical School. To find out the exact nature of how studying the dog genome could help lead to cures for human conditions (interest in such research intensified in 1999 with the identification of a gene for canine narcolepsy), she graciously sat down with us recently for an extended interview at her office.

Your Dog: If both people and dogs have the same genetic cause for OCD, why did it take a study of dogs to learn that the gene could be an OCD trigger? Why couldn’t you just find it out by looking at people with and without OCD instead of looking at Doberman pinschers?

Elinor Karlsson, PhD: It has to do with the fact that dogs of a particular breed are much more similar to each other than people are to each other. If you think back to high school biology, you’ll recall that all of our DNA, our genes, is made up of four nucleotide bases: adenine [A], guanine [G], cytosine [C], and thymine [T] — AGCT.

In any two people, we differ about once every 900 bases. It doesn’t sound like much, but it adds up to enough differences to create a lot of static when you’re looking at the genomic map. On the other hand, two dogs of a specific breed differ only once in every 1,600 bases — almost two times as infrequently. Today’s breeds were created only within the last couple hundred years from just a very few dogs, so the genetic variation between dogs of any one breed remains quite small. And because two dogs within a breed are so much more similar, when one has a condition and the other doesn’t, it’s much easier to find the gene or genes responsible for the difference. You need to look at a much shorter amount of genetic material to see it. You also need fewer dogs than you would people to confirm the difference, in part because, as you know, different breeds tend to be more prone to certain medical conditions. Without knowing it, breeders have done a lot of the groundwork for geneticists looking for clues to disease.

Consider that we were able to locate the gene in question with Drs. Dodman and Moon-Fanelli identifying only 92 Doberman pinschers with OCD and 68 Doberman pinschers without it. The same experiment in people would take thousands of subjects to be able to pinpoint the gene with the same confidence. Yet once we know the trigger gene in dogs, it’s easy to locate it in people because the entire human genome has already been mapped out. It’s just a matter of finding out that that particular gene has a certain kind of significance.

Your Dog: But is the gene for something like OCD in people really similar to the one in dogs? How close can the comparison really be?

Dr. Karlsson: It’s extraordinarily close. It seems like dogs and humans are really different until you compare humans to flies, or yeast. Humans and dogs are actually pretty similar — they’ve both got limbs, heads, digestive systems; we eat the same food. The things that are different between dogs and humans are much fewer than the things that are the same.

And with certain diseases, the two species are more similar than you might imagine. Osteosarcoma [bone cancer] comes to mind. If you compare the tumors from dogs and people, they’re almost indistinguishable. The progression of the diseases is nearly identical in the two species, too. The tumor originates in a particular spot in the long bones and tends to spread to the lungs, with death usually resulting from the cancer spreading there. But it’s rare in humans, so it has been hard to get an understanding of the genetics. Yet 25 percent of greyhounds seem to be dying of bone cancer. In dogs, through greyhounds, we’ve found one gene that we’re very confident is the source of the actual mutation that causes the bone cancer. Bone cancer is a terribly painful type of malignancy, even worse for dogs than for people. So we want to be able to get at the genetic underpinnings as the start point for finding a way to prevent it, or at least keep it from progressing.

Your Dog: How can identifying a gene help researchers come up with a way to cure — or prevent — a disease?

Dr. Karlsson: First, it’s important to understand that a single gene mutation is rarely responsible for the development of a disease. Disease processes are complex and usually involve a number of genes. For instance, the CDH2 gene explains some of obsessive compulsive disorder, but not all of it.

Another thing to keep in mind is that when we find a region of the genome that is different between sick and healthy dogs, it is just the very first step. We then look at all the DNA in and around the region to see if we can find the precise genetic difference that makes one dog — or person — get sick and not another. Often what we see is that the difference is not in the gene itself but nearby, and that it is regulatory — it changes when that gene is turned on or off. When genes are turned on, they make proteins, the molecules that get our cells to function. If you can figure out how these regulatory changes affect the protein and increase disease risk, you can begin to research drug therapies that might help regulate the action of those proteins so they don’t get a process going when they’re not supposed to, or, conversely, set a series of chemical events in motion when they otherwise might not have done their job. Then maybe you’re keeping a process from occurring that could generate a cancer, or you’re keeping a cancer from progressing. That’s what it’s about — finding, understanding, and targeting the pathway that needs to be shut off or turned on to stop a disease from developing or getting worse.

Your Dog: Have any drugs been developed to treat or stave off a disease as a result of genetic research in dogs that then gets applied to people?

Dr. Karlsson: Not a drug but gene therapy. Dogs are capable of getting an eye problem called retinitis pigmentosa. A dog will gradually become blind because a photoreceptor in the eye — something that allows the eye to make use of light — stops working. People get it, too. It’s an inherited trait. They’ve been testing gene therapy in dogs to keep the problem from causing blindness, and even to reverse blindness, and they’re getting pretty close to being able to do it in humans. They literally go in and fix the faulty gene. Normally that’s extremely difficult to do because a gene is in every single cell of your body. But the eye is kind of a little ball of fluid that’s self-contained. That means you can inject the gene therapy in just the eye itself and effect a change. It’s one of the easiest places in the body to be doing gene therapy because you’re not trying to change the gene in your blood, your skin, and every other tissue. They’re now up to testing the gene therapy in humans that they’ve been using in dogs. If it works, we will have successfully conquered a cause of blindness in both people and dogs. At some point researchers will also be able to extrapolate from canine genetics to human genetics to develop drugs that will keep ‘domino’ genes from turning on or off and thereby stop diseases from forming or progressing.

Your Dog: Can you give any other examples of diseases of both dogs and humans that geneticists are working to learn the causes of — and then possible treatments?

Dr. Karlsson: Well, the ridge in Rhodesian ridgebacks — one of the first traits we worked on in dogs — is actually a neural tube defect of sorts that comes from a gene mutation. It’s not like spina bifida, where the spinal column doesn’t close completely; it’s not harmful to the dog. But because it has to do with how the neural tube closes around the spinal column during fetal development, it can help us understand how neural tube defects occur in people.

We focus in general on diseases that are close to human diseases — cancer, diabetes, heart disease, epilepsy, deafness, and even psychiatric diseases like the OCD we were talking about earlier. And we benefit from the fact that in humans and dogs alike, family history is one of the strongest risk factors for all diseases — there’s a heritable component. That’s one of the things that is often a distinguishing characteristic of a breed; it has a higher incidence of a particular disease than other breeds (which is what makes it so easy to find on the breed’s genome).

Of even greater benefit to research that can help humans is that the canine lifespan is approximately seven times shorter than that of people, so diseases of old age, like cancer, take hold earlier and typically run their course in just a few years. So a cancer that takes a decade or longer to play out in a person will take only a year or two in dog. That means clinical trials that test the effectiveness of a drug to combat a disease might take only a year or two in dogs, whereas it might take anywhere from five to 15 years in people. In that way, dogs can provide a useful testing ground for drug therapies once they are developed.

Your Dog: Does all of this research — tracing a disease to a particular gene or genes and then working to come up with medical therapies to treat the disease — hurt the dogs in any way?

Dr. Karlsson: I’m happy to be able to say no. To do genetic testing on a dog, all you need is a blood sample. You can do cheek swabs for saliva, which also contains DNA, but dogs don’t have the cleanest mouths, so that makes it not the easiest. As technology improves, we’ll be able to get around that problem. But at this time, blood samples provide much better quality DNA than saliva, and more of it.

As far as drug testing, the aim will be to try drugs on sick or vulnerable dogs that we want to also be able to try on humans; we’re not looking to give them substances that we wouldn’t want to take ourselves. But starting with dogs, whose lives and therefore whose diseases run a faster course, allows you to see more quickly the effect the drug is having, or whether it’s having any effect at all.

All of this is a great step forward. Since the beginning of the twentieth century more than 100 years ago, the default model for genetic studies in mammals has been the mouse. And we’ve learned a lot using mice as models. But mice are in many significant ways quite different from people — and dogs. For instance, many diseases that occur spontaneously in the human and canine species have to be induced in laboratory mice. So the mouse has been great for studying genes once they have been manipulated to cause a problem, but it’s not so useful for studying mutations that occur seemingly out of nowhere and have to be dealt with in the real world in that way. Also, human and dog diseases are complex, involving many genes, as I said before. So while it’s relatively easy to test the effect of a single gene mutation in a mouse that was manipulated by researchers, that doesn’t help scientists get at the complex interplay between genes that cause disease in the human and canine populations. With the use of breeds to get at the causes, and then the cures, for those shared diseases, dogs are going to become man’s best friend in new and ever more important ways.


Please enter your comment!
Please enter your name here