Dog Disease Mapping Project (DogDNA)
Image design: Lauren Solomon, Broad Communications
In 2005, Broad researchers gave the world its first complete look at the genetic sequence, or genome, of Canis familiaris, the domesticated dog. They deciphered the DNA sequence of Tasha, a female purebred boxer. Though an interesting revelation, the real story has been and will continue to be in delving deeper into canine genetics of various purebred breeds to find genes that cause disease. “The larger goal, of course, is to use this information to aid in understanding corresponding human diseases,” says Kerstin Lindblad-Toh, scientific director of vertebrate genome biology at the Broad Institute of MIT and Harvard, and professor in comparative genomics at Uppsala University in Sweden. Dr. Lindblad-Toh leads the Broad and Uppsala teams carrying out the dog disease mapping project, or DogDNA project.
Domesticated dogs split from their wolf ancestors approximately 30,000 years ago. Contrast this to the scant 200-year history of most breed creation. In the process of breeding dogs for certain desired traits over successive generations, many genetic mutations have inadvertently come along for the ride with undesirable side effects. This includes a genetic predisposition to many diseases ranging from epilepsy, cancer, diabetes, and hip problems to many more breed-specific conditions.
Since Broad researchers and their colleagues first published the dog genome in 2005 they have found nearly 20 genes known to cause several conditions or traits in various dog breeds. This is cool science, but why all the attention on the domesticated dog?
The answer is that it is much easier to find the gene(s) of disease interest in dogs than in humans. “In human studies, you have to enroll thousands of people and look at millions of SNPs, or markers in the genome, to find areas of disease mutation,” says Elinor Karlsson, a Broad associate researcher who works closely with the DogDNA project. “In some cases in dogs, we can find what we’re looking for in just 20K SNPs for a really simple disease. In a more complex disease like cancer, you only need about 200,000 SNPs and about 200 dogs. It is a much simpler, more powerful platform for finding these genes.”
The plan is that by understanding the genes that cause that disease in dogs, we will learn a lot about the disease in humans as well.
The team has arrived at the point where they have found many genes in dogs previously unknown to be linked with disease and are translating that into a similar search in humans. “If we can understand the functionality of the DNA mutations of the dog and find similar pathways in humans, that opens the door to developing new treatments for the future,” says Lindblad-Toh.
From Simple to Complex Mutations
Researchers at the Broad are in the process of investigating nearly 40 diseases in a variety of breeds, some of which are discussed in more detail here. The team started by mapping a series of disorders that were fairly straightforward in terms of trait characteristics – like coat color in the boxer, the presence of the ridge in Rhodesian ridgeback dogs, and hairlessness in the Chinese crested dog.
The new gene tally continues to mount with the discovery of mutations resulting in cardiomyopathy disorder in boxers, ALS-like disease in the Pembroke Welsh corgi, brittle bone (osteoimperfecta) and day blindness in dachshunds, and a neurodegenerative disorder called neuronal ceroid-lipofuscinosis (NCL) in Tibetan terriers. For many of these disorders, this has led directly to a quest to find similar gene mutations in people.
Over time, the DogDNA effort led to finding the genetic mutations of more complex traits. The first example includes systemic lupus erythematosus (SLE) in Nova Scotia duck tolling retrievers. “We identified a cascade of overactive reactions in the immune response,” explains Lindblad-Toh. “We found about five different genes involved in regulating an immune pathway.” This breed was devastated by a canine distemper virus outbreak in which the only dogs to survive were those with the strongest immune response. The suggestion is that the dogs’ strong immune response may have been protective during the distemper epidemic but may be the root cause of overactive reactions and SLE in subsequent generations. This serves as an example of how natural selection also plays a role in continuing certain mutations found widely across a breed that originally came from a small group of surviving dogs.
A very recent example of finding a gene mutation for complex disease involves the Shar-Pei dog from Asia, bred for its characteristic thickly wrinkled skin. Unfortunately this breed is also predisposed to Familial Shar-Pei Fever (FSF), a periodic fever disease linked with chronic inflammation and kidney failure. The DogDNA effort recently published findings showing that the same genetic mutation causes both wrinkled skin and FSF. The team is now investigating a similar link with human periodic fever disorders, like Familial Mediterranean Fever.
Clues to Cancer
A large area of focus of the team’s work is on understanding cancer genetics in purebred dogs. To date, this has involved studying breast cancer in the English springer spaniel — a breed in which more than 35% of the dogs have mammary tumors — and osteosarcoma in racing greyhounds, rottweilers, great Dane, and Irish wolfhounds.
One of the biggest DogDNA efforts is the golden retriever cancer study. Nearly 60% of these purebred dogs develop bone marrow cancers, including mast cell tumors, hemangioma, and lymphoma. Researchers at the Broad have been collecting dog samples from golden retrievers, both dogs with these cancers and old, healthy golden retrievers to find the gene(s) responsible for these cancers.
"We have found several gene regions that look exciting and are now resequencing the DNA from these dogs to find the specific mutations involved,” says Noriko Tonomura. DVM, PhD, a research associate in the Broad’s vertebrate genome sequencing and analysis group, and research assistant professor at Cummings School of Veterinary Medicine at Tufts University. Going forward, the team hopes to learn how common these mutations are in various breeds. “This will hopefully help breeders choose what dogs to breed to avoid passing on harmful cancer genes,” says Tonomura. It may even help identify if there are any genetic mutations that interact with each other or with specific environmental factors that lead to the development of cancers.
In concert with the colleagues at Mayo Clinic and Dana-Farber Cancer Institute, the DogDNA researchers are now starting to analyze blood and urine samples from human lymphoma patients to find a possible link with human genetic mutations in the same genome regions.
Along with studying the dog genome for causes of physiological conditions, the DogDNA team is searching for clues underlying psychiatric disease. After studying just 140 Doberman pinschers, they recently discovered a gene for canine compulsive disorder (CCD), the dog version of obsessive-compulsive disorder, or OCD. “It is similar to genes for things like autism in humans,” explains Karlsson. They are now expanding this study to include other breeds known to develop CCD, like bull terriers. These dogs are treated with the same drugs used to treat OCD in humans and have almost the same response rate of about 50-60%. The Broad team is now sequencing genes linked to that gene — about 500-600 genes that are in similar pathways — in hundreds of human patients.
The Broad DogDNA team is very much in need of blood samples from purebred dogs, including older, healthy dogs. “We have found the tip of the genetic iceberg and we know there will be more genetic risk factors,” explains Lindblad-Toh. “We can guess that if we had twice as many samples we would find twice as many risk factors and that would help us build a more complete picture of how these genes interact with each other.” Plus, the team is now moving more into a phase of research where it wants to study specific tumors and see what the consequences of mutations are. “For this work, we need tumor and blood samples paired together from the same dog, along with samples from healthy control dogs, so we can look at follow-up effects,” she adds.
To learn more about donating samples, visit the team’s donation website to find specific directions and contact information.
“These discoveries will identify risk factors for certain diseases, which will help us to understand overall health risks for your dog,” adds Lindblad-Toh. The Broad team gets additional support from other academic research centers, including Uppsala University in Sweden, the European LUPA Project, and major animal health and research organizations including the Canine Health Foundation, the Morris Animal Foundation, the American Kennel Club, and the Golden Retriever Club of America.