For most of history, people bred dogs to produce a particular type suitable for work, companionship, or some other purpose. Knowledge about genes and the principles of inheritance were used to improve the predictability of many traits breeders cared about- color, size, coat type, instincts, and so on. Before the advent of kennel clubs, breeders did what worked for them and avoided breeding in ways that caused problems.
When dog breeding became a hobby and kennel clubs were formed, breeders placed value on the purebred dog and created closed registries- dogs would only be considered "pure" bred if their parents were already registered in the studbook.
Over the ensuing years, generation after generation of purebred dogs have been produced from animals already registered in the studbook. Breeders have used selective breeding to produce dogs to suit their needs, carefully choosing breeding pairs and evaluating the offspring to determine which were of high enough quality to be suitable for breeding themselves.
At the time kennel clubs were created about the turn of the 20th century, very little was known about genetics. Scientists were arguing about the mechanisms of evolution, and Mendel the monk was puttering in his garden doing experiments on peas. Nobody had any notion that creating closed populations of purebred dogs might not be a good idea. But as genetics began to develop as a discipline alongside the theory of evolution, biologists began to think about how new species evolved and the role that genetics might play. They realized that changes in the traits of animals over time could lead to the evolution of new species, and that this was a reflection of changes in the particular mix of genes present in the animals that form a species. If the animals in one population of a species became different enough genetically from the animals in another population, they could cease to interbreed and become separate, distinct species. The process of evolution was really just a change in the frequencies of particular alleles that determined the traits that make the new species unique.
Apart from the means of selection, the genetics of dogs and other domestic animals (and humans) are the same as those of every other vertebrate. When somebody decides to create a new breed of dog, they are breeding to produce a group of animals that will share a particular mix of genes that code for the features desired. A breeder starts with a group of dogs and their particular mix of genes, and combine, add, remove, and shuffle genes by selective breeding until they get what they want. In the resulting population of animals, the genes that distinguish the breed will be common to all of the dogs. The process of maintaining type in a breed (or a species, for that matter) is about achieving stability of allele frequencies in the population. Refining or improving a breed, or developing a different type, or breeding for better herding skills or smaller size, or whatever, is all about changing the frequencies in the population of the alleles you want (to become more frequent) or the ones you don’t (to become less frequent).
When you breed a litter, you bring together the mix of genes (in the parents) that you think will produce the traits you want in the puppies. But in a larger sense you are pushing the genetic composition of the breed a bit in one direction or another. You can’t increase the frequency of some genes without decreasing the frequency of others, so selecting for something necessarily selects against something else. While you might only be considering the genes in the litters you produce, your choices and all the choices being made independently by other breeders are changing the allele frequencies in the population as a whole. That might be good, but oftentimes it’s not. Perhaps you’ve never had a dog with PRA, but when it starts showing up in your breed it’s a problem for everybody. Managing the problem then requires cooperation among breeders- people need to test, they need to be honest about information, and they need to adopt breeding strategies to manage the problem in the entire breed. With genetic diseases now popping up faster than researchers can study them, breeders are facing a landscape littered with unseen genetic landlines, and they are forced to pick their way through with every litter, hoping with fingers crossed that they will avoid all the potential disasters.
It doesn’t have to be this way. The increasing burden of inherited disorders in dogs is the completely predictable outcome of the way purebred dogs have been bred. For the many disorders caused by recessive mutations, the problem suddenly appears because the mutation has become common in the breed. The mutation has probably always been there and has been passed down harmlessly from generation to generation for hundreds or even thousands of years, and as long as a dog only got one copy, the normal copy it is paired with does the job and the dog is not affected. The mutation only becomes a problem if it becomes common enough in the population for a puppy to end up with two copies. It’s all about gene frequencies, and this is no different than the genetics of any other trait. If you want your dog to have prick ears, you select for prick ears and against soft ones, increasing the frequency of the prick ear genes and reducing the frequency of the soft ear genes in the population. If PRA or some other problem suddenly pops up in a breed, it’s because the frequency of the PRA gene in the breed has increased because breeders have selected for the PRA gene. Not deliberately, of course, but inadvertently, in the process of selecting for some other trait. Maintaining type and managing genetic diseases in a breed are done exactly the same way- by controlling the frequencies of the alleles you care about, good or bad, in the entire population. Genetic diseases in dogs are the manifestation of failure to manage the frequencies of the genes that cause them.
Not all genetic diseases are caused by a single gene. Many problems are polygenic, and of course the immune system is extremely complex, with the highest genetic diversity in the genome. Polygenic problems are extremely difficult to understand; identifying the genes involved is very hard - there might be tens or even hundreds of them, each having only a small effect, and breaking just one can wreck an entire system. And how do you fix something like this? How do you restore the immune system to health when scientists don’t fully understand how it works? Complicated things are very easy to break and really, really hard to fix. For most things, geneticists are not going to have the answers.
Controlling genetic diseases in dogs is going to require some changes in the way dogs are bred. The problems are not coming out of nowhere- dogs have been bred in a way that makes them inevitable. So we need to change a few things, and more than anything we need to provide breeders with the education, information, expertise, and tools that they need in order to breed healthy dogs in a sustainable way. This is going to take some effort on everybody’s part, but if the breeders of every other domestic animal and commercial crop can learn how to do this, certainly dog breeders can do it too.
It’s time to begin. The education is available, the information is there, there are experts to do the dirty work, and it can be done. Certainly we can all agree that the best way forward is to first do no harm.