Dwarfism in Labradors: A Look at Genetic Disease and Inbreeding

Labradors used in a study published last year. Upper left is a dwarf female. Upper right is a dwarf mother and her unaffected daughter. Bottom left are three litter-mates, two unaffected and one dwarf.

One heritable condition many people are not aware of in the ever-popular Labrador retriever breed is dwarfism. As can be seen above, some of the dwarfs have only slightly shortened legs, which can explain why the trait can be easily glossed over. Interestingly enough, there are at least two different forms of dwarfism in the breed: one (osteochondrodysplasia) with a more obvious bent-legged phenotype and the other (skeletal dysplasia 2 aka SD2) with more normal appearing, only subtly shortened legs. I will be concentrating on the latter.

So, why should we even be concerned about dwarfism? Whereas some forms of dwarfism can lead to severe limb malformations and subsequent joint issues, as well as eye problems, SD2 does not. Instead, the gene may be linked to deafness. There hasn't been an examination of SD2 dogs to see if they do in fact have any hearing impairments, but similar genes in other animals lead to at least some deafness. In truth, any sort of abnormality should be concerning, at least to some degree. In the case of these dwarf Labradors, all from working- or field-type stock in a recent study, the decrease in leg length would lessen effectiveness as a working dog, as would any impairment to hearing.

Variation in the Labrador retriever. There is some more extreme type on both ends of the spectrum.

This form of dwarfism also brings up some interesting questions about breed type. Labrador retrievers vary significantly in size and shape, most notably between the comparatively lanky, lightly built working lines and the far more stocky, low-slung show lines. Some less-than scrupulous breeders could use SD2 to their advantage, taking a dog like the third one in the above image and getting it to resemble something more like the first two dogs. I honestly wouldn't be overtly surprised if someone used such a tactic as there are people that are that desperate to win a ribbon. Low-slung and stocky is what's winning right now, after all.

The average height difference between affected and unaffected dogs is only about 6 cm (2.4 in). Scientists were able to identify the likely source of SD2: a single allele change in a gene that is involved in collagen development. The trait is a recessive, which explains the generation jumping seen in the following pedigree.

Pedigree of the dogs in the study. Squares are male, circles are female. The dark individuals are all affected with dwarfism, with the red showing a form not caused by the same gene as the others. The arrow indicates the likely source of the gene, a popular sire who is the common ancestor of every dwarf in the study.

What else is remarkable about this pedigree? Do you see it? There is quite a bit of inbreeding going on. Though it's very difficult to trace all of the relationships (this chart has serious organizational issues), this pedigree is rife with shared ancestors, though most of the duplicates are a few generations back. For example, LA101, who is the black square closest to the top of the pedigree, has a grandfather and a great-great-grandmother who are full siblings. In fact, every affected individual can indeed be traced back to a common ancestor on both their sire's side and their dam's side.

And that's the problem.

This is a classic example of inbreeding leading to genetic disease. Inbreeding is risky business as it leads to an increase in homozygosity. Related individuals are more likely to have identical versions of a certain gene, and if you breed them together, it's quite likely that you will end up with offspring getting two copies of this gene, one from each parent. A lot of dog breeders think this is advantageous, allowing them to more easily select for a desirably trait. However, in all likelihood there will be bad that comes with this perceived good. There are numerous genetic diseases found in purebred dogs, and almost all of them appeared through inbreeding. This is why so many conditions can be traced to a single, common ancestor.

Often, the founding populations of a given dog breed are disturbingly low when viewed from with an eye toward population genetics. When you have breeds numbering in the thousands that are descended from maybe seven individuals, it's no wonder there is so much disease. If there are any detrimental recessive alleles in that small of a population, homozygous individuals affected by that detrimental gene are bound to occur. Labradors, to some extent, are lucky. Their popularity has persisted long enough that there have been fewer genetic bottlenecks, at least not to the extent of, say, the long list of breeds whose populations plummeted around WWII. The fact that retrievers were once interbred is also an advantage, giving a fairly large gene pool for all of the retriever breeds before they were separated. While some breeds are severely lacking in genetic diversity, such as the collie's paltry effective population size of 33, Labradors in Great Britain have an effective population size of 114.

This, however, is still far from spectacular.

For those not in the know, conservation biologists use something called the 50/500 Rule when assessing endangered species for extinction risk. At an effective population size of 500, there is concern that the species will not be able to maintain genetic diversity over a long period of time. At an effective population size of 50, the species is at immediate risk for extinction. They're circling the drain. Domestic animals have the advantage of veterinary care, but no owner in their right mind would prefer to have a sick pet. While the 50/500 Rule definitely doesn't bode well for the health of the aforementioned collie, the Labrador isn't in much better shape. When nearly one hundred thousand dogs have genetic variation equal to little more than one hundred individuals, there is a serious problem. For one thing, it makes it that much harder to avoid mating a certain dog to another that doesn't share a significant percentage of its genotype.

Since purebred dogs are closed populations, lack of genetic diversity is a serious problem. Comparing these populations to endangered species is very appropriate given the tiny effective populations sizes. Inbreeding serves to eliminate heterozygosity, and the more it is done, the more diversity is lost forever. If registries continue to insist on keeping studbooks closed, inbreeding has to stop or the only savior is going to be systematic outcrossing.

Sources are images from Wikimedia Commons for the type comparison (1, 2, 3, 4: all being copyright free or under Creative Commons licenses), Princeton University Press, Population Structure and Inbreeding From Pedigree Analysis of Purebred Dogs, and most importantly, A COL11A2 Mutation in Labrador Retrievers with Mild Disproportionate Dwarfism published in PLOS One, a peer-reviewed, free to access, Creative Commons licensed publication. Authors are Mirjam Frischknecht, Helena Niehof-Oellers, Vidhya Jagannathan, Marta Owczarek-Lipska, Cord Drögemüller, Elisabeth Dietschi, Gaudenz Dolf, Bernd Tellhelm, Johann Lang, Katriina Tiira, Hannes Lohi, and Tosso Leeb. 

What is an Inbred Strain?

These mice are from the inbred strain BALB/c, one of the most commonly used inbred strains in experimentation

Inbred strains are lines of animals bred to be genetically identical. This is done for experimental purposes, as experiments require the elimination of as many unknown variables as possible. This is to eliminate the possibility that the experimental result was caused by a variable that is not being tested for. For example, genetic variation may cause a particular mouse (or rat or guinea pig) to be more resistant to toxin A than another mouse, so using genetically identical mice will eliminate that possibility. This is also why other variables, such as type and amount of food, size of cage, bedding, and other traits not pertaining to the variable being tested should be strictly controlled.

So, how do you go about creating an inbred strain? By inbreeding, of course! However, it takes some time. The process involves brother-sister matings for at least twenty generations. At that point, the inbreeding coefficient is 99% or higher, and as such the resulting individuals are basically genetically identical. Each inbred strain has been given a designation of numbers and/or letters (such as the mice strains BALB/c, C17, MOM, T739, and perhaps the most often used C57BL), and different strains are used for different experiments. This is often due to a particular trait that is known to occur in the strain under certain conditions. There are many strains that have been selected for certain traits for this exact reason.

Though many people are steadfastly against the use of live animal experimentation, you cannot deny how much has been learned. If it weren't for live animal experimentation, we wouldn't know very much about cancer, behavior modification, or numerous other aspects of modern science. Inbred strains continue to be used so that more can be found out about these and other topics.

My first knowledge of inbred strains came from my Psychology course, which was taught by an animal behavior specialist. He often worked with inbred strains, as did most (if not all) of the Psychology majors on the campus. However, they usually worked with rats (such as strain F344).

Source is Rutgers. Image is from Wikimedia Commons under a Creative Commons license.

Recessives and Inbreeding

Continuing on my musings on the silliness that is the "mismark," I would like to present several examples of situations under which "undesirable" markings appear. I plan to do some case studies on certain breeds and why I think some breed standards are completely impractical when it comes to color. To begin, here's some comments on various forms of breeding that involve recessive genes.

One of the most common reasons, perhaps the most common reason why mismarks will appear in breeds is that countless color genes in the canine world are recessive in some way. This makes it quite easy to forget about a gene that was present in the early days of a breed, but disappeared through selective breeding. It only takes one breeding for a recessive to disappear, and if chance favors the production of homozygous dominant individuals and heterozygous carriers, that gene may soon be forgotten. Also, only breeding from the dominant phenotype will lead to production of some homozygous dominant individuals, and eventually a population may end up being mostly made up of homozgous dominant and heterozygous individuals. Of course, since heteozygotes are still in the population, the production of the recessive "mismark" is still very possible. Depending on the percentage of heterozygotes, it may even be almost guaranteed. Remember that if you breed heterozygotes together, you have a 50% chance of producing more heterozygotes (as well as 25% homozygous dominant and 25% homozygous recessive). Even if you take the "undesirable" recessive phenotypes out of the breeding pool, two-thirds of the remaining individuals will still carry the recessive. The likelihood that the recessive will disappear completely is basically zero.

In addition, if there is some sort of recessive being carried by members of a breed, inbreeding will often lead to it appearing at a higher frequency than would be expected if matings occur between more distantly related individuals (see Why Inbreeding Leads to Increased Homozygosity). Remember: all members of a purebred population are relatives due to the small number of individuals that went into starting basically every breed. Often, purebreds have very high inbreeding coefficients, sometimes to the point that, no matter how little relationship two dogs have in their recent pedigrees, they may in fact be as genetically similar as cousins or even closer relatives. So, the more heavy the inbreeding (i.e. the higher the inbreeding coefficient), the more likely that a recessively inherited mismark will appear. If breeders wish to avoid the colors that are seen as "bad," then this is one more reason to add onto the giant pile of reasons to not inbreed.

This silver Labrador is expressing the recessive blue dilution gene (possibly in combination with the liver gene). It is believed by most Labrador breeders out there that the presence of the gene in the breed is the result of a fairly recent out-cross to a Weimaraner. If this is true, the silver Labs will likely be no healthier thanks to the out-cross when compared to regular Labs due to the heavy inbreeding involved in creating continuous lines of silvers. In fact, their health may even be worse than the rest of the breed thanks to that same inbreeding.

One advantage, I will say, that goes along with breeding for a dominant phenotype is there is little to no inbreeding required to get the gene to be fairly consistent. In contrast, the easiest way to get a recessive gene to be consistently produced is very heavy inbreeding. For example, if there is a single dog that appears with a recessively inherited color that a breeder is interested in, and they are the only one of their kind, inbreeding is the only way to guarantee the reappearance of the gene. Let's say a female puppy is born who is an unusual color. As an experiment, she's bred to a male and all of the puppies don't look like her. This means the trait is recessive, and all of the puppies are guaranteed to be carriers. Breeding two of them together would lead to a litter that would likely contain 25% homozygous recessive puppies. If two of those recessive phenotype puppies happen to be opposite genders, then breeding them together would produce an entire litter of recessives! Or, if only a male puppy is produced, he could be bred back to the original bitch (grandmother-grandson) to create the desired litter of recessives. The original bitch could also be bred to her father, since it is quite likely that he was a carrier of the color, and they could produce more like her. However, this is even worse than a brother-sister mating (father-daughter and mother-son matings are the most severe forms of inbreeding possible in mammals). Furthermore, if the desire is to keep the color coming, then only breeding to very close relatives would be possible. Even if out-crosses were done, they would still be relatives and they would still have to be bred back to the heavily inbred original population for the color to reappear. It's creating the breed-within-a-breed I mentioned in a previous post.

These are two white (albino) Doberman pinschers, which are caused by an incomplete form of albinism (making them cream and white with green-yellow eyes). The sort of heavy inbreeding situation I've mentioned happened when the white Doberman appeared on the scene, which is why white Dobermans are far more prone to health and behavioral issues than the rest of the Doberman breed. Sensitive skin is basically guaranteed due to the lack of pigment.

Needless to say, I think anyone who is interested in any "unusual" color in a breed should be very wary. I would say that virtually every breeder that breeds specifically for a color that is not accepted by a breed standard should be approached with extreme care. If the color is recessive, then it is likely that they are involved in very questionable breeding practices. Part of this is that the lack of acceptance from other breeders toward the "unacceptable" color has likely resulted in a very narrow pool of dogs going into creating the lines that they are offering. I reiterate: the narrower the gene pool, the worse off a population will be. If any breeder brags about having a particular color of puppy available, this should be a major warning sign. If that puppy is being sold for a higher price than the "normal" puppies, this is major warning sign number two.

One issue I see with modern genetic research is the possibility that all members of a population may be tested for a particular color gene in an attempt to completely eliminate it. To me, this idea is completely absurd, but I wouldn't put it past some breeders. It should be much more important to deal with inherited health issues rather than genes controlling color, coat type, and other traits in a breed.