DCM is fatal. The heart fails, sometimes in the absence of any symptoms of a problem, and the dog simply drops dead - in the middle of a game of fetch, during a run on an agility course, or while the family is away during the day at work and school. Most dogs die in their prime and even younger. 

​(In the first graph below, the data are divided into age cohorts and no animal appears in more than one age group.)
​

Where did this horrible problem come from?

We have good data about the history of DCM in the breed. In 1990, the incidence was already quite high, with more than 25% of dogs affected. Since then, DCM has increased essentially linearly by about 1.5% per year. At this rate, by 2040, 100% of the Doberman breed will be afflicted with DCM.

I wrote last summer about the tragedy of watching this noble breed go extinct before our very eyes (see Are We Watching the Extinction of a Breed?).

What are we doing to deal with this problem? Breeders are routinely monitoring their dogs for the electrical abnormalities that are signs of DCM. They are trying to select from lines that appear to be less afflicted with the problem.

However, the question nobody seems to be asking is whether it is even possible to rid the breed of this problem through selective breeding. Can better monitoring and ever more selective breeding reduce the incidence of this horrible problem in Dobermans? Is there enough genetic diversity in the breed to "breed away" from DCM?

We now have new data from a separate study (Dreger et al 2017) on a different cohort of dogs (see Inbreeding of Purebred Dogs Determined from DNA). They used a different method for estimating the proportion of the genome that is homozygous called "runs of homozygosity" (ROH). This method identifies blocks of consecutive homozygous loci, then adds the total length of these blocks and divides by the total length of chromosomes covered by markers to produce the inbreeding coefficient. Lower values of inbreeding are better.

The hand thing about this estimate of inbreeding is that the value indicates two things: 1) the probability that an animal will inherit two copies of the same allele from an ancestor (i.e., homozygous for that allele), and 2) the fraction of all loci that are homozygous. 

From this study, the average inbreeding coefficient of the Doberman is 43% (red arrow on the enlarged graph on the left, blue on the graph for all breeds on the left). This means that 

• on average, nearly half the genome of a dog is homozogyous, with two copies of the same allele;
• the risk of any particular locus being homozygous for the same allele - whether good or bad - is 43%;
  
• on average, 43% of the genomes of any two dogs are the same.

In addition to the data on inbreeding and heterozygosity of the whole genome in Dobermans, we now also have information specifically about the genes of the immune system.

The immune system protects an animal from all manner of outside invaders, from bacteria and viruses to fungi and parasites. It must be able to recognize a bewildering diversity of pathogens as foreign, then marshall the specific cellular defense mechanisms necessary to destroy them. At the same time, it must be able to distinguish "self" from "non-self"; failure to do this is the cause of autoimmune disorders, in which the immune system attacks one of the body's own tissues.

The genes for the immune system in the dog are called the "dog leukocyte antigens" (DLA). They tend to be inherited as blocks of genes called haplotypes. There are two types of DLA haplotypes called DLA Class I and DLA Class II. Like single alleles, an animal can inherit two copies of the same haplotype or they can be different (i.e., homozygous or heterozygous).

In most animals, the genes of the immune system are the most diverse in the entire genome, and in wild animals there is strong selection to keep them that way. In purebred dogs, however, inbreeding, strong selection, bottlenecks, and genetic drift have reduced the genetic diversity across the genome, including the Class I and Class II DLA.

Take for example these data for DLA diversity from UC Davis in nine breeds of dogs (see Inbreeding and the Immune System: Unintended Consequences). The graph shows the number of haplotypes found in a survey of each breed, separated into Class I and Class II DLA.

This is the assessment of Dr Niels Pedersen, who is conducting a study of genetic diversity in Dobermans at UC Davis.

It is highly unlikely that the desperate genetic situation of the Doberman can be improved by selective breeding within the closed gene pool of the breed. It should be made very clear to breeders that they will not restore health to this breed by selecting against health problems. Furthermore, trying to select less related parents in an effort to improve diversity in the offspring is a bit like pushing your peas around on your plate; you might look like you're accomplishing something, but with little effect.

The only hope for this breed is the initiation of a sound, comprehensive cross-breeding program, under the guidance of population geneticists, that will introduce new genetic diversity into the breed. The longer it takes to begin genetic rescue, the more difficult it will be and the less likely it is to be successful.

REFERENCES

Pedersen NC, L Brucker, NG Tessier, H Liu, M Cecilia, T Penedo, S Hughes, A Oberbauer, & B Sacks. The effect of genetic bottlenecks and inbreeding on the incidence of two major autoimmune diseases in Standard Poodles, sebaceous adenitis and Addison's disease. Canine Genetics and Epidemiology 2:13. DOI 10.1186/s40575-015-0026-5.
 

See also these posts about the Doberman:

Beuchat C. 2016. Are we watching the extinction of a breed? (or, Why are we focused on consequence instead of cause?)

Beuchat C. 2016. Are we watching the extinction of a breed? (part 2)

Beuchat C. 2017 Are preservation breeders preserving the Doberman? (No.)

By Carol Beuchat PhD

For decades, scientists have been following a population of wolves on Isle Royale that has served as a model system for the study of the evolution and genetics of small, isolated populations. Isle Royale is a remote island in Lake Superior, USA, and wolves first arrived via an ice bridge from Canada in the 1940's. They found a healthy population of moose and no other predators, and their numbers grew. Over the years, the populations of the wolves and moose cycled in the dynamics of predator and prey. Good years for the moose population would be followed by an explosion in the number of wolves, which would drive the moose population back down, and the cycle would begin again. Scientists were able to observe the effects of the introduction of canine parvovirus, which nearly wiped out the wolf population in the 1980s. When the wolf population failed to recover, despite an abundance of moose to support them, scientists became concerned that the population was suffering from the negative effects of inbreeding. The wolf population was descended from only one male and two females.

A hard winter in 1996 took a heavy toll on the moose population, which collapsed to only 500 animals. This reduced the available food for the wolf population that was still trying to recover from low numbers.

Then in 1997, a wolf called "Old Grey Guy" (designated "M93") crossed the ice bridge that occasionally forms between the island and the mainland and joined the population on Isle Royale. With the injection of fresh genetic diversity from an unrelated animal into the population, the wolves began to thrive and increased from three packs to four. But the superior health and fertility of M93's descendants allowed them to proliferate at the expense of the older lineages until nearly 60% of the genetic variation in the population came from M93. As the level of relatedness of the animals in the population increased, so too did inbreeding. Pack after pack failed until by 2011 there was only a single pack of nine wolves remaining on Isle Royale. 

Today, only two closely-related wolves are left. They are half-sibs and also father-daughter. If they produced a pup, the expected level of inbreeding would be 44%. It is possible that, just by chance, they are less related than would be expected from their pedigree relationships, and if they produced a pup it might, just by chance, be less inbred than predicted. But in the absence of new blood added to the population, inbreeding can only increase. The last pup seen with the pair in 2016 was deformed and small in size, and it disappeared in 2016. This is probably the end of the road for the Isle Royal wolves.

The graphs below are from a paper just published that documents the genetic and population deterioration of the Isle Royale wolves from 1995 to the present (Hedrick et al 2017). They tell a simple story. A small, isolated population of animals initially thrived, but a series of unfortunate environmental events and unavoidable inbreeding have resulted in collapse and probably extinction. 

​Nothing about the Isle Royale wolf story is a surprise. Inbreeding and loss of genetic diversity reduce the ability of a population to tolerate changes in environmental conditions, disease, fluctuating food supply, and the many other random events that can affect survival in the wild. Eventually, animals in closed, isolated populations go extinct.

Our purebred dogs don't have to survive extreme weather and occasional food shortages. We assist reproduction, nurse the weak pup, avoid allergies with special diets, and otherwise keep the dogs as healthy as possible with the best veterinary care. But it's sobering to compare the levels of inbreeding that did in this population of wolves to those typical of purebred dogs. Before the arrival of M93, the average inbreeding in the wolves was a bit more than 20%. After a few generations, inbreeding began to increase again and was most recently greater than 30%. In the vast majority of purebred dog breeds, the average level of inbreeding exceeds 25% and in too many breeds exceeds 40%. For breeds with closed stud books, inbreeding can only continue to increase. The Isle Royale wolves provide a real-life experiment confirming what we know about the consequences of small population size and genetic isolation.

We are inadvertently repeating this experiment, now with a captive population of purebred dogs. How high can the level of inbreeding go before a breed simply goes extinct? Do we really want to find out?

REFERENCES

Hedrick PW, M Kardos, R Peterson, & JA Vucetich. 2017. Genomic variation of inbreeding and ancestry in the remaining two Isle Royale wolves. J Heredity (in press). 

The UK Kennel Club has updated their table of "vulnerable" and "at watch" breeds using the registration numbers for 2016. 

Why are these breeds considered to be at risk of extinction? I'll show you using a bit of population genetics.

The graph below shows how the size of a population affects the rate of inbreeding. Let's say we have a population of 50 dogs, 50:50 male and female. If we start with 50 unrelated dogs at generation zero on the graph and let them breed randomly, the average level of inbreeding in the population will increase over the generations. By 10 generations, the average inbreeding coefficient is about 10%, and by 30 generations it is about 25%. The effects of inbreeding depression (low fertility, high puppy mortality, low disease resistance, shorter lifespan, etc) begin to be significant above about 10%. For perspective, inbreeding of 25% is what you get from crossing full siblings from unrelated parents. 
​

The thing to remember here is that this dependence of inbreeding on population size is not some obscure biological principle; it is simply a matter of math. Breeders can increase the rate of inbreeding by deliberately breeding closely related animals, or they can slow it down a bit by avoiding close breedings, but it will nevertheless increase over time. 

The reason certain breeds are on the UK KC "vulnerable" list is because their numbers are so low that inbreeding will increase rapidly. With this will come increased incidence of genetic disorders caused by recessive mutations, decreased disease resistance, smaller litters, higher puppy mortality, and the many other deficits that are typical of inbreeding depression. Eventually, small populations become more and more inbred until they go extinct.

Have another look at the table at the top. The numbers are for total registrations, but you can estimate the size of the breeding population by diving registrations by some estimate of litter size (e.g., 5) and using that to estimate the number of breeding females.

For the Bloodhound, for example, 53 registrations translates to about 10 litters from 10 bitches. If we have an equal number of males, the total population size will be 20. You can see that the average level of inbreeding will increase at a rate of about 1% per generation (orange symbols and line). So, by 25 generations, the average mating will be the equivalent of a full sibling cross (25%). 

Notice that the breeds on this list have had small numbers for years, so they are already highly inbred. For instance, the average inbreeding coefficient of the bloodhound determined from DNA is already stratospheric at 43% (Dreger et al 2017). No amount of clever breeding will prevent this from continuing to increase. 

Most populations of wild animals don't survive with inbreeding levels above 25-30%. They are overtaken by small litter size, high puppy mortality, genetic disease, and physical defects. Modern veterinary care and our willingness to pay for it is sustaining many breeds with inbreeding levels even higher than this, but how much longer can they last?

If these breeds are going to survive, they need much larger population sizes, and the already breathtaking levels of inbreeding must be reduced to improve health. Otherwise, they will eventually go extinct.

Will this happen? ​Time will tell.

Check out
ICB's online courses

​*** BASIC POPULATION GENETICS FOR DOG BREEDERS ***
Starts 6 March 2017
​
*******************

Join our Facebook Group
ICB Breeding for the Future
...the science of dog breeding
*******************

Visit our Facebook Page
ICB Institute of Canine Biology
...the latest canine news and research

By Carol Beuchat PhD

Without the immune system, we would be vulnerable to infection by bacteria, viruses, fungi, and parasites. Your body needs to recognize the foreign invader and mount an appropriate response to eliminate it before it does damage. There are zillions of kinds of potentially dangerous beasties out there, so the immune system is necessarily very complex. The immune system must also be able to tell the difference between invaders and your own tissue; failure to do this results in autoimmune disorders. Even cancer cells are targeted by the immune system.

​We're going to talk about the relationship between the inbreeding and the immune system in dogs. Just to make sure you appreciate the complexity of your body's defense system, take a few minutes to watch this little video. You don't need to remember any of this; just notice the number of moving parts and how intricately they are connected.

No doubt about it, your body's defense system is extremely complicated, and all of its magic depends on exceptionally high diversity in the genes that run it. 

One of the consequences of inbreeding and loss of genetic diversity in purebred dogs is decreased diversity in the genes of the immune system. This has lead to weaker immune systems and higher rates of autoimmune disorders.  

To get an idea of how much variation there is in the diversity of immune system genes among breeds of dogs, I've compiled the data for a number of breeds. The genes tend to occur in sets called "haplotypes", and there are two main classes of these, DLA Class I and DLA Class II. (DLA stands for "dog leukocyte antigen", just a fancy name for these particular immune system genes.)

You can see that the breeds with the highest number of haplotypes are the Standard Poodle and the Havanese, with about 40 versions of DLA Class I haplotypes and roughly 30 of DLA Class II. At the other end of the graph are several breeds that have only about 10 of each class of haplotypes (Doberman, Flatcoated Retriever, Black Russian Terrier, and Alaskan Klee Kai). In terms of total diversity (Class I + Class II), the number of haplotypes for these breeds ranges from about 70 (40 + 30) to as few as about 20 (10 + 10). 

(Remember, all of these variations are in the population but not in a single animal.)

 
When you're making breeding decisions, you probably don't think much about the genes of the immune system. But there is good evidence that inbreeding and loss of genetic diversity take a toll on DLA diversity. Breed populations with higher levels of inbreeding like the Doberman and Flatcoat have fewer DLA haplotypes, which should compromise the function of the immune system.
 

We know that inbreeding produces homozygosity, and this in turn increases the expression of genetic disorders caused by recessive mutations. We know also that homozygosity causes something called inbreeding depression, which is manifested as lower fertility, smaller litter sizes and higher puppy mortality, shorter lifespan, and many other detrimental effects. But inbreeding and loss of genetic diversity also affects how well the immune system does its job. 

Geneticists talk about the importance of genetic diversity for maintaining a healthy gene pool and reducing the incidence of genetic disorders in dogs. We could eliminate every mutation in a breed but this would not produce "healthy" dogs if the immune system has been compromised. Even the breeds with the "best" DLA diversity (Standard Poodles and Havanese) have only a fraction of the diversity present in the ancestral dog, and they have their share of allergies, skin disorders, and cancer. 

The bottom line is that breeders need to protect the DLA diversity that remains in their breed, and as they contemplate breeding strategies to improve overall genetic diversity, they should specifically consider DLA diversity as well.

REFERENCES

Dreger DL, M Rimbault, BW Davis, A Bhatnagar, HG Parker, & EA Ostrander. 2016. Whole genome sequence, SNP chips and pedigree structure: building demographic profiles in domestic dog breeds to optimize genetic trait mapping. Disease Models & Mechanisms 9: 1445-1460. doi:  10.1242/dmm.027037

Check out
ICB's online courses

​*** BASIC POPULATION GENETICS FOR DOG BREEDERS ***
Starts 6 March 2017
​
*******************

Join our Facebook Group
ICB Breeding for the Future
...the science of dog breeding
*******************

Visit our Facebook Page
ICB Institute of Canine Biology
...the latest canine news and research

Yesterday's post was about a new venture in the UK called "PUPscan", which endeavors to identify puppies that will develop hip dysplasia as adults on the basis of a sonographic examination of the hips in puppies 6-8 weeks old. The gist of my discussion is that I had basic questions about what they were measuring and the scientific evidence to support assertion that their scans would be predictive of hip dysplasia in the adult. They did not provide answers to these questions despite repeated attempts to extract basic information from them. 

Last night, I was alerted to a paper (Fischer et al, 2010) that examined the efficacy of using ultrasound of puppies to predict hip dysplasia in adult dogs. ​This is the very thing they appear to be doing, so the results of that study are worth a closer look. 

They examined 566 puppies of 17 different breeds, from age 16 to 49 days (mean = 32.5 days). They clearly describe the measurements they made, which included α-angle (as illustrated), as well as a measure of joint laxity (distraction value, DV). They also assessed the repeatability of their measurements and subjected the data to rigorous statistical analysis. They note that they did not record β-angle as had been done in other studies because of low repeatability. Of the animals they examined as puppies, 203 (35.9%) were radiographed as adults for classification of hip status. 

Their results:

"...No statistical significant correlation could be found between sonographical determinations of α-angle, joint laxity, DV and conventional radiographic hip joint classification at the age of 12 to 24 months...None of the
parameters assessed by ultrasonography in puppies could ultimately be linked to adult CHD. Results of our study suggest that static and dynamic ultrasonography of hip joints in puppies between 16 and 49 days of age is technically feasible but cannot be recommended for detecting puppies that will develop CHD between the ages of 12 and 24 months."

Their conclusion is very clear. They were unable to predict the development of hip dysplasia based on ultrasound examination of young puppies. Further, they note that similar sonographic assessments have been done on newborn infants to detect dysplastic or immature hips for many years. However, they note that the hips of  puppies always appear normal at birth and dysplasia develops postnatally, so the structural anomalies in newborn infants are not evident until later development in puppies.

There is one question I asked that is not answered by this study: 

• What is the basis for your claim to be assessing "congenital genetic dysplasia"?​​

PUPscan claims to be able to distinguish "genetic" dysplasia from dysplasia resulting from trauma or damage. In fact, as Riser et al (1985) explain, the hips of all puppies are anatomically normal at birth, and hip dysplasia is the consequence of abnormal stresses on the structures of the hip that result from laxity in the joint. That is, there is no "genetic" hip dysplasia in which genes are responsible for misshaping of the hip during post-natal development. Rather, hip dysplasia is in fact the result of trauma and damage because of abnormal biomechanics. He notes that genetics does play a role in the body conformation of a dog (e.g., size, weight, muscle development), and these features are related to breed-specific differences in the prevalence of hip dysplasia, with large, heavy dogs being at highest risk. From this, 

To be fair, assessing joint laxity in young puppies seems like a reasonable idea. In fact, I've given it some thought myself, and I haven't ruled out the possibility that it might be a useful thing to do. While this study failed to find that any of the things they measured were predictive, it's possible that focusing on a single breed (instead of combining the results from 17 breeds) might reveal something missed in this study. 

The bottom line

A published, peer-reviewed study failed to find any evidence that ultrasound examination of young puppies was predictive of the development of hip dysplasia as adults.

Finally, this post appeared on the PUPscan Facebook page today:

REFERENCES

Fischer A, A Flock, B Tellhelm, K Failing, M Kramer, & C Thiel. 2010. Static and dynamic ultrasonography for the early diagnosis of canine hip dysplasia. Journal of Small Animal Practice 51: 582-588. DOI: 10.1111/j.1748-5827.2010.00995.x.

Riser WH, WH Rhodes, & CD Newton. 1985. Hip dysplasia. Ch 83 in Textbook of Small Animal Orthopedics, CD Newton, Ed. Lippencott Williams & Wilkins. 

Check out
ICB's online courses
​
*******************

Join our Facebook Group
ICB Breeding for the Future
...the science of dog breeding
*******************

Visit our Facebook Page
ICB Institute of Canine Biology
...the latest canine news and research
​