San Nicolas island foxes (Urocyon littoralis) are thought to be least genetically diverse of all vertebrates. However, they have been able to maintain diversity and heterozygosity of their MHC haplotypes.
On my previous post entitled “Why is inbreeding bad?” I noted that one of the real issues with inbreeding is an impaired immune system as the result of losing diversity and heterozygosity in MHC/DLA haplotypes.
The MHC stands for the Major Histocompatibility Complex. The MHC is cell surface molecule that is encoded by a relatively large gene family. The MHC molecule controls some of the actions of the leukocytes, “the white blood cells.” As you might remember from your high school health class, leukocytes are the blood cells that act as police, taking out infectious diseases and foreign materials, including cancerous cells.
Possessing a greater diversity in MHC haplotypes allows the organism a heightened ability to control what leukocytes do when exposed to a variety of challenges to the immune system. Reduced diversity winds up restricting the immune system’s ability to fight disease. Possessing a heterozygous MHC haplotypes also provides greater immune resistance than possessing homozygous MHC haplotypes.
In humans, the MHC is called the human leukocyte antigen system, usually abbreviated as HLA. In dogs, it is called the dog leukocyte antigen system, which is abbreviated at DLA.
One cannot look at an organism and tell what its MHC haplotypes are. It’s not like looking at fur or hair color.
And as a result, these haplotypes are very easily lost within a closed registry system. Often, the haplotypes not lost entirely within a breed or strain, but because of the popular sire effect and continuous linebreeding from “the best,” a huge percentage of the population in any one breed can be quite homozyous in its MHC haplotypes.
This has real health and welfare effects. Autoimmune diseases are becoming more and more common in purebred dogs. Some breeds are highly susceptible to cancer, which might be partially the result of the decrease in MHC diversity of dogs within the closed registry system. Allergies are also likely implicated to this reduction in MHC diversity and heterozygosity. The MHC also has some effect upon reproductive cells, which is one reason why certain breeds and strains have issues with fertility.
So what can be done?
Well, the easy answer is to do away with rigid closed registry systems and allow the different dog breeds to exist as very genetically diverse populations.
However, it is possible to breed for great MHC diversity in domestic dogs, even within the closed registry system.
To understand how this is possible, we will have to change species again.
Take the San Nicolas population of the island fox (Urocyon littoralis). This population is considered the least genetically variable of all sexually reproducing animals.
However, despite its lack of genetic diversity, this population of island fox has relatively diverse MHC haplotypes.
How does it do this?
Well, a study led by Andres Aguilar (2004) found that the San Nicolas foxes likely were able to maintain their diverse MHC haplotypes through what is known as balancing selection. Balancing selection occurs when population keep multiple alleles actively maintained at frequencies that are higher than the mutation rate. It almost always happens within populations in better for an organism to be heterzygous for certain alleles than homozygous for them.
As we have seen with MHC haplotypes, it is better to be a heterzygote, so there is a strong selection pressure for the foxes to keep their MHC haplotypes diverse.
One way in which they likely do this is that they choose mates that have MHC haplotypes that are different from their own.
This might sound a little crazy, but even humans have innate mechanisms that are designed to keep our MHC haplotypes diverse.
There is a famous study in Switzerland that asked women which men were most attractive by smelling the shirts the men had slept in for two nights. Almost without exception, the women selected the men who had different MHC haplotypes from their own. The pheromones that are connected with sexual attraction in humans are also giving out genetic information such as what MHC haplotypes a potential mate might have. And it’s so innate that women are able to obtain this information without understanding exactly what they are responding to.
It’s likely the same way in island foxes, which, like humans, typically reproduce within a bonded pair system. Foxes, like all wild dogs, are pretty particular about their mates. After all, the survival of their offspring is dependent upon having a partner that can provide for them.
Domestic dogs, for whatever reason, don’t seem to be as particular about their mates, and thus, it has been easier to selectively breed them. Wolves and other wild dogs very rarely inbreed, but it is relatively easy to get dogs to inbreed and to do so for multiple generations.
And as a result, it has been easier for different breeds to lose their MHC haplotypes.
Further, it is not just the mere number of haplotypes a breed possesses. Heterozygosity has proven to be more important in determining immune health than just possessing a certain number of haplotypes.
That means that within an inbred population, there has to be some conscious effort to maintain both diversity and heterozygosity.
That is pretty hard to do when one is breeding for physical or behavioral traits– and the immune system genes cannot be readily ascertained.
Well, we do have a solution. Genoscoper currently offers an MHC II test, which would allow breeders to see what haplotypes their dogs posess and what to do with them. Mars Veterinary also offers a test that examines the MHC (among other things). It is called Optimal Selection from Wisdom Panel, and it was initially tested with breeders affiliated with the Dandie Dinmont Terrier Club of America (A NEW GENETIC BREEDING TOOL ON THE HORIZON (1)). Over a two year period, this selection tool has allowed Dandie Dinmont breeders to increase their litter sized from 2.75 to 4.0 puppies per litter. Dandie Dinmonts are a rare terrier breed that has issues with genetic diversity, and these tests could be a boon for breeds that have very small founding and effective populations.
So science now allows for us to do with domestic dogs what the San Nicolas island foxes were able to do on their own. We can maintain relatively inbred populations and still keep the MHC haplotypes diverse and heterozygous.
However, here are some caveats:
First of all, virtually no dogs are bred without some eye to competition. Competition, whether in show or working trials, has a selection pressure of its own, and a breeder has to considered a wide variety of issues before choosing which dog to breed with which other dog. As I’ve noted before, MHC haplotypes are hard to see, so how many breeders would choose diversity and hetrozygosity of MHC haplotypes over the ability of the dog to win in the show ring or in the trial? I would like to think that very few breeders would choose immune health over ribbons and titles, but that’s probably not the case. Keeping the MHC diverse and heterozygous is a long-term project, and it requires a bit more dedication and discipline to breed for than physical or behavioral traits.
Further, breeders have to balance a wide variety of issues when selecting which dogs to breed, and MHC haplotype diversity alone should not be the sole criteria. Other issues with temperament and hereditary diseases have to be considered in light of the MHC.
Finally, many of these breeds have either entirely lost MHC haplotypes or certain haplotypes are very rare within the breed.
You cannot do DIY balancing selection when you don’t have the haplotypes in the first place.
Even if we might have these tools, one should keep in mind that it is much more cost effective and easier to try to maintain genetic diversity within breeds and strains in the first place. It makes much more sense to keep breed registries more open and then operate selective breeding principles within much more diverse populations. Breeding within much more diverse populations requires some understanding of how different traits are inherited, but one can readily learn these skills. It is not rocket science, but one can figure it out. If people living thousands of years ago0–before there was even a Mendelian theory of inheritance– were able to produce so many different types of dogs, then we surely can.
DIY balancing selection can be a tool within conservation genetics, but I think it has to be paired with a fundamental understanding of how population genetics works.
Otherwise, you’re going to get into silly arguments with people who think that you can just inbreed and inbreed without consequence. No rare dog breed can be saved in the long term if its immune system doesn’t work properly.
It’s really that simple.
In the end, selecting for diverse and heterozygous MHC haplotypes within a limited gene pool is less preferable than trying to maintain genetically diverse strain. We have intentionally reduced genetic diversity within dog breeds and strains, yet as a population, dogs remain one of the most genetically diverse of all domestic animals. We have the ability to correct some of the errors without resorting to sophisticated genomic technology, but because the dog culture at large values purity and homozygosity at all costs, we are left looking at the only other option available.
Inbreeding requires so much scientific knowledge and acumen to pull off correctly that it really shouldn’t be done. And the validity registry systems that ultimately reduce dog breeds to inbred populations needs to be questioned.
Reblogged this on A Bitter CynoAnarchist Rages On and commented:
Scottie talks about new tools to allow breeders to preserve diversity, on the Retriever, Dog, & Wildlife blog.
[...] DIY Balancing Selection (retrieverman.wordpress.com) [...]
It would be interesting to know more about the population ecology of the St Nicholas foxes. If they’re like their mainland sister species, they start breeding at a young age and have large litters. Meaning that a vixen who has five litters produces a couple dozen kits that die without reproducing for each one that survives. That gives natural selection a lot to select from. But how does this act on MHC/DLA? Is the immune system that important in surviving to reproductive age and finding a mate? What are the main mortality factors for juveniles?
p.s. just checked some facts and litter size for the grey fox is a bit under four . . . question remains, what is the major agent of natural selection? Even at 4/litter, one litter per year, females that make it to whelp average 3 or 4 litters, the fraction of pups/kits making it to breeding age will be small. Is this a crash-through-disease island population (no natural enemies)? How does this relate to the MHC/DLA picture?
The interesting thing about this “species” is that it is the largest predatory mammal on the islands.
For most of its time on the islands, it was the apex predator, and it had virtually no predation at all. That means that more island fox kits survived to adulthood than gray foxes on the mainland.
The real problem with some island fox populations is that bald eagles no longer nest on the islands. Bald eagles don’t normally mess with relatively large mammals– they do sometimes, but it’s not their main prey source. However, right now, the islands have attracted golden eagles, which do regularly attack large mammalian prey. It is believed that golden eagles came to the islands in search of feral piglets to eat. Yes, these islands have feral pigs, goats, and an introduced herd of bison.
So think of historical population genetics island foxes as being like what gray foxes would be like if gray foxes lived without any competition or major predators.
“question remains, what is the major agent of natural selection? ”
This is the wrong question. There is no MAJOR agent of natural selection.
Balancing selection presupposes many factors acting together. The MHC would have a factor in parasite and disease resistance. That is only a small part of the whole picture. Obviously, in the case of these foxes, variance at the MHC is an important factor in their survival, or they wouldn’t have it after such a severe bottleneck. There are other species that have experienced the same type of bottleneck that do not show MHC variation. This shows that diversity at the MHC was not an important factor of survival *for that species in that environment under those particular selective pressures.* If the pressures change, however, all bets may be off.
MHC issues affect different species and populations in different ways.
you said: Domestic dogs, for whatever reason, don’t seem to be as particular about their mates, and thus, it has been easier to selectively breed them.
== I’m not so sure that’s true. The street or village dogs seem pretty diverse, so it may well be that left to their own devices, dogs are just as selective as any other. Plenty of breeders will note that a given bitch will “have nothing to do” with the desired stud dog. The problem is that most of the time, the bitches aren’t given a choice. Either it’s the only male available (recreating the “Isle Royale” scenario) or the breeder interfers . It may well be one of the negative aspects of using AI breedings. I don’t think it’s likely that breeders will start lining up a bunch of prospective studs and letting a bitch choose for herself, but I do think that perhaps more attention should be taken when a bitch doesn’t want to accept a particular male.
Peggy Richter.
Studies of street dogs show that bitches often mate with multiple partners, and pair bonds between male and female dogs in those populations are virtually nonexistent.
The high genetic diversity in African and Asian street dogs is that they have never undergone selection to form actual breeds, and they have been that way for thousands of years.
The link you gave to support that “Allergies are also implicated to this reduction in MHC diversity and heterozygosity” describes MHC involvement in allergic reactions but does not say anything about diversity or heterozygosity (or loss of diversity and/or presence of homozygosity). Or did I miss it?
It does not. The one on cancer doesn’t say that either.
The point was just to show that the MHC is implicated with allergic reactions, too.
Optimal immune response is associated with diverse and heterozygous DLA/MHC genes. That’s the general point of this piece.
http://www.lhasaapso.org/articles/mhc.html
Here’s one that actually says it.
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0022402
Worst case scenario.
And don’t think it couldn’t happen in dogs.
Something similar exists in golden hamsters, which are very, very inbred.
http://nsdl.niscair.res.in/bitstream/123456789/536/1/MHC_HLA.pdf
I do not disagree with your underlying theme that loss of genetic diversity is a bad bad thing. I do not disagree that loss of MHC haplotypes and increased MHC homozygosity are bad. I only question whether everything can be laid upon loss of MHC haplotypes and increased MHC homozygosity.
<>
Something similar DOES happen in dogs, i.e. CTVT, a transmissible tumor. It is mentioned in the article, in fact. I get a different take-home message from this article than you do. Look at the last paragraph. They could be wrong–it happens–but the researchers concluded that the devils are “immunocompetent” despite their MHC situation and they appear to attribute the lack of immune recognition of the tumors to the fact that “DFTD tumour cells are likely to have evolved adaptive mechanisms to survive unhindered in the tissues of the new host.”
Thanks for the article. There’s a good run-down of different mechanisms by which tumor cells can avoid immunosurveillance or avoid destruction.
I’m not saying that everything can be laid upon it, but this factor is a huge issue that has largely been ignored by the dog fancy. It still is.
Nowhere in this post do I say it’s the cause of everything.
We don’t know everything about it, but we know that in virtually all cases, diverse and heterozygous MHC haplotypes are necessary for optimal immune response.
This is an exception, but it’s not that big of one when one considers how inbred lab mice strains are:
http://www.genetics.org/content/176/4/2501.full.pdf
Dog breeds aren’t that inbred.
And what we’ve seen in dogs is that this is important, and it’s important enough that even in inbred forms of wild dog, they select for it on their own.
Two things that aren’t causing all of this are vaccines and dog food.
http://www.nsf.gov/news/news_summ.jsp?cntn_id=123040&WT.mc_id=USNSF_51&WT.mc_ev=click
This article explains why even though MHC diversity is important, it is not an end-all solution.
A dog breed who is very diverse is more tolerant of being wiped out by a plague. It is better for 20% of the population with rare haplotypes to survive than it is for the entire population to be demolished.
East Siberian Laikas are very well-noted for tolerating diseases native to Asia, where imported dogs from Europe still have a hard time coping with outbreaks in Siberia.
The article also explains why many indigenous breeds in North America went extinct upon contact with the Europeans.
Not to mention that in a landrace population you have natural outbreeding. You are not starting with a limited number of genes, then breeding in such a way that you are losing them with each generation.
If your native Laikas have a disease outbreak in one area, other dogs are brought in from another area. These dogs will have had gene flow from yet another area, due to their landrace status. This virtually guarantees a certain amount of diversity.
Contrast that to dogs in a closed registry. You are starting with dogs that are from a very limited gene pool. Distemper runs through your dogs and kills most of them. You get dogs from another kennel, that shares the same foundation as your kennel. Your new dogs are not going to be any less susceptible than your old dogs.
If you go back and look at the old turn of the century dog books (hit Google books) you will find repeated mentions of inbred dogs being more susceptible to disease.
Awesome article, Dave.
No one is saying that everything can be laid the feet of MHC homozygosity. There are background genes involved. There is a current study on IGs that implicates MHC in one strain and just plain tight breeding in another, in regards to autoimmune disease. I haven’t gotten my hands on the full study yet.
Heterozygosity is, however, the default in regards to wild canids. And we do have a test that will help us to maintain diversity at the MHC in dogs.
Eventually technology will catch up, we use what we have available for now.
[...] DYI Balancing Selection [...]