The Replacement Dog: how a veterinarian / dog rescuer / geneticist searched for the right puppy

The death of my fifteen year old golden retriever Jack wasn’t just about my loss. It was about finding someone else to do his job, which for the last five years had been serving as a security blanket for Jenny, my shy collie mix. Jenny depended on him to tell her when people did not intend to eat her, to run interference with well-meaning strangers, and to demonstrate calm at the veterinary clinic. We could not remain a single dog household for long.

Jack and Jenny

I adopted Jenny at age 13 months. She had never been off the farm where she was born until her surrender to a shelter, and the world proved much larger than she expected. Jenny has excellent dog skills but a crippling anxiety upon encountering new people or environments. I adopted her knowing about her idiosyncrasies because I wanted to study anxiety in dogs, and wanted to experience it first hand. And so I have; living with Jenny has informed my understanding of anxiety in a way that reading about it never could have.

In my research, studying the way genetics and environment interact to affect the risk of anxiety has brought me back time and again to the importance of early environment: in utero environment, early maternal care, and puppy socialization. How the brain changes during and after the socialization period turns out to be a huge part of my research interests, and to learn from the source as I had done with Jenny, I would need a puppy, and a very young one at that. I’ve only adopted adult dogs in the past, but now is an excellent time for me to raise a puppy, as I work from home many days.

A puppy who would grow up to fit in well with Jenny had to fit a specific mold: confident around people and other dogs, but not so pushy as to annoy her. Someone she could play with. Someone male, because I didn’t want to deal with girl dog politics for the next ten years.

Jenny

Now, I have counseled others that adopting a very young mixed breed puppy from a shelter or rescue group means you really have no idea at all who you have just brought home, and that there is no shame in purchasing a dog from a responsible breeder. However, in practice, I balked at purchasing a dog. I completed an intense shelter medicine internship at the University of Florida several years ago, and I still feel part of that community. As the distance between the present day and that experience increases, I find myself holding tighter to those connections and looking for new ways to remain a part of sheltering. Purchasing a dog who was not going to be in want of a home felt a bit like eating humanely raised meat: I tell myself it’s okay for others to do it, but when I actually try to do it myself, some part of me rebels.

Yet as I looked at puppies from local rescue groups, in short order I found myself in a panic: could I really adopt a puppy whose genetics were completely unknown, whose parents I most likely couldn’t meet, and who had almost certainly had some early life trauma before ending up in foster care? Genetics and early experience are both critical in shaping the adult personality, and while I hope I could handle dealing with another shy dog, Jenny needed someone dependable, not another neurotic failing to keep it together when the mailman drove past.

When I started to seriously consider purchasing a dog, I had to decide on a breed. I love retriever-collie mixes: ideally the best of both worlds, retriever-social and collie-smart. But finding a responsible breeder of retriever-collie mixes seemed a tall order. Border collies are too intense for me. Australian shepherds have their tails docked so short. And I wanted to find a breed that is not recognized by the AKC, that is absolutely not bred for looks, that possibly even has open stud books to keep the genetics pool large and diverse.

I found the Scotch Collie and the English Shepherd. The Scotch Collie club had an open stud book policy going for it (good for them!). The English Shepherd club had a closed stud book policy (open it up, guys!) but it had been open relatively recently, the breed isn’t recognized by the AKC, and the dogs can’t be shown in conformation classes. The breed is a versatile working breed. Both breeds have lovely breed standards that accept a wide range of phenotypes (for example, 30-80 lbs in adult weight - a wide range!), which in itself tells the story of breeding for temperament and not looks.

In the end, I chose the English Shepherd based on the fact that there are more of them around, so it was easier to find a litter promptly. Waiting a few months would mean potty training a puppy in January in the Midwest, an experience I’ll leave to others.

The English Shepherd club maintained a list of breeders who had available puppies, with lots of information about the parents. It’s a well designed resource, and I link to it not to encourage others to run out and get an ES puppy (they are smart and high energy and not for everyone) but to provide an example of what kinds of information should be provided about available litters.

I screened the descriptions of parents: I discarded those who weighed more than 70 lbs, as managing Jack in his dotage had been hard on my back. I discarded those who were described as “protective” or “taking some time to warm up to people.” I checked that the parents had passed the relevant genetic tests (for this breed, tests for several eye diseases and hip dysplasia). Then I looked at the remaining breeders’ websites.

The breeders I liked talked about how they raised the puppies: giving them lots of positive experiences. They talked about what they did with the puppies’ parents - agility, nosework, herding. They often had long applications for potential owners to fill out, which asked all the questions they should: How will you exercise this dog? Do you have a fenced yard? What will you do if he is destructive?

I found a litter in Virginia with a male who sounded perfect: confident, social, and by the way athletic. My husband and I stuffed Jenny into the car and drove 9.5 hours to pick up our boy. He cost, by the way, probably more than twice what a rescue puppy would have cost, but I have paid for knowing that he is clear of some genetic diseases for which he might have been at risk, and for knowing that he was in the uterus of a calm, happy mother; raised with a litter who had plenty of high quality food and safe places; and had extensive early socialization (including the Early Neurological Stimulation and Early Scent Stimulation programs). He has proven, in his first week and a half with us, to be social and sweet, willing to settle down when asked so long as he is given plenty of exercise and mental stimulation, and terrifyingly smart. He and Jenny are already wrestling for hours daily, laying the foundation for what I trust will be a long friendship.

That is the story of how we found Dashiell.

Dashiell

Geek version of the fat mutant labs FAQ

In the face of overwhelming demand (three people thought it sounded like a good idea), here is the geek version of the fat mutant labs FAQ, the nitty gritty about the study findings.

What gene is mutated and what does it do?

The gene itself is one of these weird ones that actually codes for multiple proteins. (Basic genetics usually assumes that one gene codes for one protein, of which there may be a few different but similar versions.) The gene in this study is POMC, or proopiomelanocortin. The “pro” means that it codes for a protein which doesn't itself do anything until it gets cut up some more. The rest of the long name describes the things it gets cut into:

• opio: short for opioid. Opiods are potent pain relievers; the classic opioid is morphine. If you or your pet has had surgery, you’ve probably used an opioid for pain relief during or after it. (There is currently a big scandal about drug companies and opioids in the news.) In this case, two of the products snipped out of POMC are endogenous (made by the body) opiods, β-endorphin and enkephalin. They are feel-good substances.
• melano: refers to melanocyte stimulating hormone (MSH). MSH has a bunch of different effects in different tissues, but most importantly has been associated in humans with effects in “controlling appetite,” our authors tell us.
• cortin: the coolest thing this gene makes is ACTH, which is released into the bloodstream to tell the adrenals (down by the kidneys) to release the “stress hormone,” cortisol. This is the system I study! But it is, it turns out, not really relevant to this particular study.

What was the mutation?

It was a 14 base pair deletion. The gene itself is thousands of base pairs long; in the dogs with the obesity-associated allele, they were missing just 14 base pairs. But remember, DNA bases (nucleotides) are translated into proteins in sets of 3. (Three nucleotides codes for one amino acid; a string of amino acids makes up a protein.) If you remove a chunk of bases that is not a number divisible by 3, suddenly the translation machinery that reads the DNA and produces proteins gets completely off track. It is just reading in sets of three. Now it’s off by one or two and suddenly it's creating entirely different amino acids, so the resulting protein is completely different after this point. This is called a “missense” mutation, because a protein is still generated, it's just made with different amino acids.

For example, the sentence “mad man sat” becomes gibberish if you mess up the spaces and make  “adm ans at” after removing just one letter.

So that’s what happened here: the first chunk of the POMC protein in these dogs with this allele is fine, but the second chunk is gibberish from the body’s point of view. Since the POMC protein gets cut up into smaller proteins with actual functions, this means that some of its products were fine, and some were not. The article has a lovely illustration of the situation, highlighting in red the products that are affected by the mutation.

Image from Raffan et al., 2016

The products that are broken in dogs with this allele are β-MSH and β-endorphin. The first, you will remember, has been associated with control of appetite in humans. (And it’s apparently somewhat different in rodents, so it’s hard to test its function in laboratory mice, so it was exciting to find it in dogs so we can learn more about obesity!) The second is one of the endogenous opioids, a feel good substance.

How do these broken proteins cause obesity?

We don’t know. The authors write:

The mechanism by which reduced β-MSH and β-endorphin due to the mutation causes behavioral and weight phenotypes remains to be precisely elucidated...  However, studies of humans with POMC mutations resulting in aberrant forms of β-MSH ... have suggested that β-MSH is important in controlling appetite and obesity development in man, with hyperphagia notable in patients with both mutations... The role of β-endorphin in regulating appetite, satiety, and energy balance is less well understood, but it has been proposed to underlie oro-sensory reward in high-need states or when the stimulus is especially valuable. However, mice selectively lacking β-endorphin are hyperphagic and obese, suggesting that the loss of both neuropeptides could contribute, in combination, to the phenotype seen in dogs carrying this frameshift POMC mutation.

 So, we don’t know, but we know MSH problems make people more likely to eat a lot, and β-endorphin may have to do with the feeling of reward after eating.

How did they find the gene?

They guessed! They had a small number of genes that they thought might have to do with obesity, and they checked them out in some labs. They looked for different versions of the genes in different dogs, and then looked for alleles (versions) which were more common in fat labs than not fat labs. They found one hit — this particular allele of POMC.

This is known as a “candidate gene” approach: when you pick a gene that you think might have an effect in a particular phenotype and test it out specifically. It has historically been less productive than “hypothesis free” approaches in which you basically ask about all the genes you possibly can if they have something to do with the phenotype. This is because our guesses about which genes affect which phenotypes turn out to be wrong so often. So these authors got lucky to get a result!

Was their sample size big enough?

I hadn’t considered getting into the statistics (I hate statistics) but some people actually asked about them. Yeah, I like their numbers for a candidate gene approach. To get reliable results using some other methods they would have needed more dogs, but when you’re just asking questions about a few genes, it’s fine to have a smaller number (in this case, 310).

I’d be more concerned about where their dogs came from. They tell us:

Labrador retriever samples were collected from dogs from a large assistance dog breeding colony (n = 81) or that were pet dogs from the UK (n = 310). Pet dogs were recruited either after their owners volunteered in response to an email from the UK Kennel Club sent to over 15,000 Labrador retriever owners, or via participating veterinary practices.

This sounds reasonable enough. But if I wanted to play devil’s advocate, I’d suggest that they were biasing themselves to a particular kind of owner, the kind of owner who responds to UK Kennel Club email. These owners may be more likely to breed and/or show labs, and therefore these labs may have a slightly different genetic background than some other group of labs. For example, perhaps some famous show lab sire who sired thousands of puppies happened to have this mutation. And then perhaps labs that are shown are more likely to be fat than labs who are not (because UK judges actually reward obesity in show labs — don’t get me started on that). If that happened, it would throw off the statistics and you might see a spurious correlation between the mutation and obesity.

That’s just me spinning tales. I think their methods are pretty standard; it’s hard to recruit pet dogs to these kinds of studies and they did it the usual way. It's quite interesting that they found a mutation which is so clear in its loss of function of the protein. If the correlation is indeed spurious, subsequent studies using different populations of labs should show us.

Any more questions?

What else do you guys want to know? I tend to focus on stuff in studies that I find interesting. What do you find interesting?

Learn more genetics

As before, I will shamelessly take the opportunity to plug my upcoming genetics class. It is not too late to sign up; it starts Monday, May 9, but you can sign up several days late.

Raffan, Eleanor, et al. "A Deletion in the Canine POMC Gene Is Associated with Weight and Appetite in Obesity-Prone Labrador Retriever Dogs." Cell Metabolism (2016).

Fat mutant labradors FAQ

The study about a mutation associated with obesity in labrador retrievers has received massive news coverage for a dog genetics article (and you can see how happy the researchers are about it at @GODogsProject). So what does it mean for your dog?

Image: Raffan et al., 2016

• Do all labs have this mutation? No. They tested 383 UK labs and found that of 383 Labrador retrievers from the UK, 78% of them didn't have this mutation at all; 20% had one copy of the mutation (were heterozygous); and 2% had two copies (the maximum number you can have; they were homozygous). They tested some US labs as well and found similar frequencies.
• Do any other breeds have this mutation? Yes, it was also found in flat-coated retrievers, a breed closely related to the lab. The researchers tested 38 other breeds, testing 8-55 dogs per breed. (The list of breeds they tested is provided.) They tested 55 golden retrievers, another closely related breed to the lab, without finding this mutation. That doesn’t mean it’s not out there, but it does suggest that if it is present in other breeds, it’s much less common in them than in labs.
• If my dog has this mutation, does it mean my dog is doomed to be fat? No. They did show an association between the mutation and weight: dogs who have one copy of the mutation are, on average, 1.90 kg (4.18 lbs) heavier than dogs who have no copies. Dogs who have two copies are on average 3.8 kg heavier than dogs who have no copies. (This is in labs, but the numbers in flatties are very similar.) But that’s an average. It’s not all about genetics. Some dogs who have this mutation won’t put on that much extra weight, and some dogs who have it will put on more. The gene they studied will interact with other genes to affect your dog’s eating habits and metabolism, and of course in weight gain as in behavior, the environment (food type, food amount, exercise) is a huge factor.
• If my dog is fat, does it mean my dog probably has this mutation? Not necessarily. There are lots of reasons to get fat.
• Is this “the gene” for weight gain? In dogs as in humans, multiple genes control weight. This is just one, albeit one with a pretty impressive effect in this breed. And again, remember the importance of environmental factors!
• How does this mutation cause weight gain? It may have to do with causing dogs to want to eat more (certainly a trait we’ve all seen in labs!). It may also change their metabolism directly, affecting how they turn calories into energy.
• Where did the mutation come from? Both labs and flat-coats are descended from the St. John’s water dog, a breed which is no longer around. The researchers have reason to believe that the mutation dates back to that breed.
• Does this mutation make labs easier to train? Possibly. The researchers tested a population of labs who are used as breeding stock for assistance dogs, and found that many more of them carried the mutation than in the general population: 23% had zero copies, 64% had one copy, and 12% had two copies. Additional genetic analysis suggested that these dogs are being actively selected for this mutation (unbeknownst to the people who are selecting them!). This suggests that something about this mutation makes dogs better at assistance work — perhaps making them more food motivated and easier to train.
• Does this mutation make flat-coats fat, too? It does, and yet flatties aren’t known for obesity the way labs are. It’s a bit of a head scratcher.
• What's the big deal? Didn’t we already know that labs are food-obsessed mutants? I know, right?

Raffan, Eleanor, et al. "A Deletion in the Canine POMC Gene Is Associated with Weight and Appetite in Obesity-Prone Labrador Retriever Dogs." Cell Metabolism (2016). It’s open access and, as modern genetics papers go, not that hard a read. Check it out!

Want to know more about dogs and genetics? I have a class on it starting Monday, May 9! We will learn concepts like homozygosity and heterozygosity, and I will be happy to discuss this study in more depth.

Dog genome ruminations

The other day I was re-reading the original dog genome paper, as you do. This is the paper published in 2005 to accompany the release of the first full dog genome sequence (of a boxer named Tasha) and accompanying annotation (a mapbook of what genes are located where in the very long sequence of bases that is the genome).

You might think that a genome paper wouldn’t be very interesting, because basically the point of it is to say “here is this genome. We published it. It was a lot of work, and it’s done, and now you can use it.” But most groups try to have something interesting to say in their descriptions of a new genome, and this one actually had a lot of interesting stuff about dog genomics in it.

Don’t just take my word for it. It’s open access, so you can read it for yourself.

Lindblad-Toh, Kerstin, et al. “Genome sequence, comparative analysis and haplotype structure of the domestic dog.” Nature 438.7069 (2005): 803-819.

The dog was one of the earlier mammals to be sequenced, so a lot of this paper consists of comparisons between dog and the other sequences we had at the time, human and mouse. We already knew that humans and mice were more closely related than humans and dogs in one sense: they share a most recent common ancestor. This means that as you follow the branches (and tangles) of the tree of life, first you get a branch that divides the most recent common ancestors of human, mouse, dog, and relatives from species like opossum and chicken; then you get a branch that divides the most recent common ancestors of human and mouse and relatives from dog and relatives; and only then do you get a branch that divides the most recent common ancestors of human and mouse. It looks like this:

Tirindelli, Roberto, et al. "From pheromones to behavior." Physiological reviews 89.3 (2009): 921-956. Fig 5

So we’d expect that human and mouse would share more genomic sequence than dog and human, right? Each of those branches in the tree of life represent a point at which one species becomes two, with resulting divergence in genomic sequence. So if the species divergence between humans and mice happened more recently than the species divergence between humans and dogs, then the genomes of humans and mice should look more similar than the genomes of humans and dogs. But it turns out, as this dog genome introductory paper reports, that dog and human share more genomic sequence, more base pairs, than human and mouse do. So how can that be, if humans and mice are closer together on the branches of the tree?

There are several forces contributing to this result, but the one that made me smile was the different rates at which each species reproduces. In the time since humans, mice, and dogs branched off from their shared common ancestor (before humans and mice branched off from their shared common ancestor), mice have had many more generations than humans and dogs. They reproduce so quickly compared to us and dogs that they have more chances to change their genetics from generation to generation. And as a result, while the number of divisions (places where the tree branches) are greater between human and dog than human and mouse, the number of generations of mice between today’s mouse and that last common ancestor of mice and humans and dogs is greater in mice than in dogs or humans. As the paper’s authors put it:

The lineage-specific divergence rates (human < dog < mouse) are probably explained by differences in metabolic rates or generation times, but the relative contributions of these factors remain unclear.

The other way of looking at it is saying that species age at different rates. So while behaviorally modern humans appeared around 50,000 years ago, and dogs appeared arguably 10,000-32,000 years ago, nevertheless the human population is about 4,000 generations old while the dog population is around 9,000 generations old. Because dog generations are shorter.

We created them, but they’re now older than us. Just like how my dog was younger than me when I got him, but aged right past me. Science!

Why genetics and dog training?

I won’t lie to you. When I first started thinking about teaching genetics courses for the Association of Professional Dog Trainers, I was mostly excited about the second class, which covers behavioral genetics of dogs. The first class was just something we had to do in order to get everyone up to speed on the basics of genetics, to have the information they needed to understand the second course.

But, of course, basic genetics is relevant to every day life with dogs and is interesting on its own. I don’t blog about genetics much, because I’m shoulders deep in highly technical stuff in my PhD program which is hard to communicate to people who aren’t equally immersed in the field. But when I stop to think, it’s not hard to come up with questions about dogs that you can’t answer without basic genetics.

• In the past decade, new advances in technology have enabled the discoveries of more and more genes in both humans and dogs. These discoveries get reported in the popular press, such as the gene for small size in dogs (discovered in 2007). What exactly is a gene? What does it do? What does it mean to have different “versions” of a gene? It’s hard to understand these news tidbits if you don’t really get some of these basic concepts.
• When you breed a lab and a poodle, you get a labradoodle with very predictable appearance. But if you breed two labradoodles, you can't predict what the puppies will look like. Some will look more like labs, others more like poodles. They're all the same genes, so why is one generation so different from another?
• Why is blue merle color associated with deafness in dogs, so that if you breed two blue merles to each other, you're almost certainly going to have some deaf puppies? 

It’s easy to get caught up in the details of a field and forget that that’s not all there is. I’m trying to remember to get my nose out of the books (or PDFs of articles) once in a while and look around me.

(Genetics is beautiful and fascinating and I’m extremely lucky to have the chance to talk about it, through the lens of a shared love of dogs, in my upcoming classes with the APDT.)

What has cheap genome sequencing done for dog science lately?

So you have the full sequence of a couple few dog genomes, and a wolf genome to boot. (Yes, these days sequencing is cheap enough that genomics researchers can do this. There are more errors in these less expensive, “shallowly sequenced” genomes than the one that we use for the standard canine reference, but even with errors, you can still get the whole genome to play with.) So you have these genomes. And you are curious about domestication. What makes a dog different from a wolf? These genomes each are made up of millions of nucelotides, so when you do a straightforward comparison between dog and wolf, you get hundreds of thousands of differences in nucleotides, ranging from single nucleotides that are different to long stretches where chunks of the genome are repeated in one species but not the other. And what to make of the differences between pairs of dogs — are those important too? It can seem an overwhelming problem.

Luckily, in addition to having fast and cheap access to full genome sequences, we also have powerful computers for analyzing these sequences (and one of my favorite parts of my PhD program is that I get to use my programming skills in addition to my biomedical skills). What people do is think of patterns that suggest that certain areas are the interesting ones, and tell computers to look for those areas. It turns out that if you have a couple of genomes of animals of the same species, you can compare them to find regions where there is very little variation between animals. This suggests that this area is important — everyone has to have exactly the same sequence here, because anyone who has any differences is less fit and less likely to survive to pass on their genes. This is called a selective sweep, because at some point in the past, this change swept through the genome and everyone eventually got the identical copy of this region.

For an added bonus, if you have the ancestral species — in this case I am obviously talking about wolves, which are ancestral to dogs — you can compare this region in that species. If you find that this region is the same in all the dogs but different in the wolves, you have an area which is highly suspicious for being involved in domestication. So you can ask a computer to go find some of these low variation regions for you,

There are a lot of statistical tests that you can do to convince yourself that this area has sufficiently low variation to be interesting, but that area doesn’t, and it has been my pleasure this week to be reading about those in great detail. (Being a grad student rocks, but then sometimes there is statistics.) But the most recent papers I have been reading have pretty much done away with statistical tests to convince themselves that certain areas are involved in domestication. What they have done is to use stats to find areas that are just potentially interesting, and then they actually go look at the areas and see what they see. What known genes are in that area? Anything that could have to do with domestication? Yes? So let’s see how that gene differs in a whole bunch of dogs and wolves. The same in all the dogs, and different from that in all the wolves? Awesome. So what does this gene actually do? Can we understand how the genomic change between dogs and wolves — the mutation — changed the protein? Did it change the protein’s function? Or maybe dogs make more, or less, of that protein. Labs have just been selecting specific genes from these areas and investigating them intensely and seeing what they find out.

The best example of this approach (and the one I find the most interesting, because it was done in dogs, not pigs or chickens like the other papers I have been reading) was published early this year. You have probably heard it if you are interested in dog domestication, because it made a big stir by declaring that dogs had evolved to be better at digesting starch than wolves.

Axelsson E., Ratnakumar A., Arendt M.L., Maqbool K., Webster M.T., Perloski M., Liberg O., Arnemo J.M., Hedhammar Å. & Lindblad-Toh K. (2013). The genomic signature of dog domestication reveals adaptation to a starch-rich diet, Nature, 495 (7441) 360-364. DOI: 10.1038/nature11837

But when you read about this paper, did you know how they figured out that dogs are better at digesting starch? They did one of these low-variation genomic scans. They found some interesting regions. They looked at what genes were in these regions. They found a lot of genes involved in digestion, so they decided to chase that first. (They also found some interesting genes that work in the brain, and hopefully we will see a followup paper on that soon.) They focused on a few genes and tried to figure out what they did and how they had changed between dog and wolf. In at least one case they found that dogs just expressed a lot more of a particular protein than wolves do, and that protein is involved in digesting starch.

There are a lot more regions to look at in dogs, and there are some interesting things to hunt down in tame foxes, too, of course. We are in a fascinating time for genomics. The technology is becoming so inexpensive that we can actually look at the code of genomes belonging to individual animals more more readily than we could just a few years ago, and this is a game-changer. There should be many more discoveries to come soon about the mechanics of canid domestication!

2013 Canine and Feline Genetics Conference

I was privileged to attend the 7th International Conference on Advances in Canine and Feline Genomics and Inherited Diseases this past week at the Broad Institute in Cambridge, Massachusetts. It wasn’t a large conference: about 150 dog and cat genetics researchers who get together every year and a half or so to catch each other up on what they’ve discovered recently, give each other suggestions about how to proceed or where to get good samples, and give their graduate students a chance to give some talks. I took notes on Twitter (#canfelgen) on my favorite talks (er, those of them that were not too technical; there were quite a few talks that I enjoyed hugely but that did not lend themselves to 140 character summaries). My apologies in advance if I got anything wrong; I was typing with my thumbs as fast as I could and may have made some mistakes.

Robert Wayne, Analysis of recent and ancient canine genomes suggest a new hypothesis for dog origins
Robert Wayne of UCLA talked about ancient canid genomes and “a new hypothesis for dog origins.” We are still not sure which gray wolf population was directly ancestral to modern dogs, and his work has shown that in fact no single population appears to fit the bill. Wayne believes that in addition to all the gene mixing between dog and wolves since domestication (which obviously muddies the picture), there was an ancient population of canids that gave rise to both dogs and wolves which we have not yet found samples from.

Wayne explained that ancient populations of wolves were much more diverse genetically than modern day populations, and we will really need to look to those ancient populations to solve the mystery of dog origins. About 10,000 years ago, populations of both wolves and dogs shrank dramatically in size, which explains why wolf populations are less diverse today than before that bottleneck. This was right around the time that domestication may have been happening, but we don’t know if the events are related.

So, based on this information, Wayne Lab embarked on a study of ancient canid DNA, comparing samples from dogs and wolves both from about 15,000-30,000 years ago. They found that the oldest dog populations were in Europe, not Asia. One interesting finding was in the black coat gene, which was relatively recently introduced from dogs into wolves and has swept through wolves. Apparently, being homozygous for black coat reduces fitness in wolves, but being heterozygous increases fitness. They don’t know why yet.

Wesley Warren, Genetic signatures of selection in the domestic cat lineage
Wesley Warren of the Washington University School of Medicine talked about using cats as models to study domestication. Comparing what we know about the behavior of cats to work in other species (dogs, horses, and chickens), he hypothesizes that cats aren’t actually undergoing significant domestication at all, because they are still very competent at living independently from humans and hunting their own prey. He talked about his work developing a SNP chip for cats, to aid in future genomic research in that species. A SNP chip is basically a library of known polymorphisms in the cat genome — single nucleotides that are known to differ between different individuals. Having all of these SNPs cataloged and available for use on a chip makes looking for correlations between these differences and things like behaviors or diseases much easier. At one point, his chip was used to discover the gene for curly coats in cats.

Warren talked about his recent work comparing the genomes of domesticated cats with their nearest wild relative. He found differences in RALY, a coat color gene. He found that cats have fewer receptors for smell than dogs do, but more for pheromones, and he wants to compare both olfactory and pheromone receptor genes in domesticated cats to big cats. He also talked about the 99Lives project, a project to get more cats sequenced (a theme which was returned to later in the conference).

Anna Kukekova, Simple behavioral pattern: is it simple?
Anna Kukekova of the University of Illinois talked about her work with tame foxes (if you don’t know about the Russian farm fox project, check out the excellent summary at the Thoughtful Animal). Kukekova opened by demonstrating the difference between the lines of foxes selected for tameness and the foxes selected for aggression with video in which a researcher performed a behavioral test on one fox from each line: first standing by a fox’s cage, then opening the door and reaching for the fox, then trying to pet the fox. The videos, shown side by side, were dramatically different: on the left, a clearly wild animal, both cowering from and menacing the human who stood near it. On the right, an animal reacting to human presence just as a dog in a shelter might, almost in a spasm of enthusiasm, wagging its tail, soliciting affection, rolling over to let the human pet its belly. Kukekova talked about analyzing the differences in behavior statistically, and how some of the most important behaviors they found for consistent differentiation between the populations were pricking ears forward, wagging tails, and approaching humans.

Kukekova investigated foxes which were second-generation crosses between tame and aggressive lines. The tame behavior was highly heritable, which was already known. Animals with heritage from both lines usually showed intermediate behavior, on the spectrum between tameness and aggression. What was interesting was that some of the crossed animals showed what she called “switching” behavior: the animals showed tame behavior at some points in the test, and aggressive behavior at others. For example, some of these animals were aggressive to humans who stood at their cage doors, but friendly when the door was opened. Others were friendly when the door was closed, but aggressive once it was opened.

Tara Baxter, Genomic approaches to identify putative canine behavior-associated genes
Tara Baxter of Cornell talked about her method of trying to track down some genes that are associated with different behaviors in dogs. This is a tough problem, as behaviors are usually influenced by multiple genes as well as by the environment, so tracking down a gene that influences aggression (for example) is a lot harder than tracking down a gene that is all by itself responsible for a disease. Baxter reviewed test results from owners who filled out a CBARQ (behavioral survey) about their dogs; she had access to a database of 19,000 surveys, so an impressive sample size. Using these tests, she averaged behaviors for each breed, getting a score of how likely animals of a particular breed were to display a particular behavior (for example, “begging”). Then she did an association study, using a canine SNP chip similar to the feline one discussed above. She used the chip to compare the SNPs found in individual dogs from various breeds, and looked for correlations between the average breed behaviors and the SNPs that she found in individuals of those breeds.

She had some interesting results which will benefit from more study. For example, for the behavior of urinating while left alone, she found an association in an area which is related to behavioral disorders in humans. Finding this association in an area which seems to affect behavior suggested to her that she might be on the right track, though of course a lot more work will need to be done. She mentioned some other interesting associations that she found as well. She also, of course, found associations that appear spurious, such as the association between a gene for long hair and chasing behavior. One amusing association she found was a relationship between the gene for short legs, such as you might see in a corgi, and a fear of stairs! She commented that sometimes physical traits explain behavior.

...And that is my smattering of summaries from the conference. Here’s hoping that I will manage to attend the next one, in eighteen months, in the other Cambridge — the one in the UK!

Guessing at the mechanisms of dog aggression

I've been thinking a lot lately about how dog aggression works, since the recent dog fighting bust (second largest in history). Fighting dogs are bred for willingness to attack other dogs, but for docility with humans. You don’t want your fighting dog to turn on you in the training yard or in the ring! Willingness to attack another dog, and to continue to attack when the other dog retaliates, is called “gameness.” Despite intense selection on the part of the dog fighters, the dogs show a lot of variation in levels of gameness: some dogs are very game and some are less so, even with training. But it does seem to be true that gameness is heritable, something you can breed for.

So how do you get aggression which is so specific? And what are the fighting dog breeders actually selecting for? What’s different in the DNA of a game dog and a not-game dog? We don’t have any real idea. Recently I came up with one possibility (too new even to be called a theory). It opens more questions than answers, but here’s the story.

There is a well-studied phenomenon in rats and mice related to the position of the fetuses in the uterus. (I know, uterine position is probably not related to genetics, but bear with me for a minute.)  If a female fetus is surrounded by two males, one on each side, she gets more than her usual dose of testosterone in the uterus. Because testosterone helps the developing fetus know what sex to develop into, this extra testosterone makes her develop some masculine characteristics which will stay with her throughout life: she will be what is referred to as a masculinized female. Among other things, her behavior will be affected. Her play style will change to a more rough and tumble style. And she will be more aggressive towards others of her species.

This phenomenon has been demonstrated in multiple species, including guinea pigs, rabbits, and marmots. It is suspected to be in effect in dogs as well: although there are no published papers reporting on it in dogs (at least none that I could find — please let me know if I’m wrong!) I have heard it discussed at dog training seminars as a possibility. And given the range of species it affects and the similarity of effects of reproductive hormones on development across species, it seems really likely to affect dogs.

The big question is: how could this be a genetic phenomenon? The genders of your neighbors in the uterus are random, right? Well, not completely: one of the differences between masculinized and non-masculinized females is that masculinized females have more male offspring. Really. We don't know how that works, though there are some theories about why it may be a useful adaptation to some environments.

Moreover, testosterone doesn't just come from other fetuses. It comes from the mother as well. Some amount of testosterone is normal in development. What if what dog fighters are breeding for, without knowing it, is mothers who produce more testosterone when they are pregnant? Or maybe fetuses which are worse at transforming testosterone into estrogen (as fetuses like to do)? Or fetuses which are more sensitive to testosterone (maybe have more numerous or more sensitive testosterone receptors)?

These questions lead to even more questions, of course, which is why I haven’t even called these ideas a theory yet. Do the more aggressive masculinized female rodents show more aggression to their own species than to humans (which is my initial question about the fighting dogs)? Do male rodents with more males beside them in the uterus show increased levels of aggression? Do we know anything at all about different levels of testosterone released by the dam, not just by uterine neighbors?

There is a lot known about intrauterine position. It is really well studied, partly because it might help us understand the effects of reproductive hormones on fetuses in general, such as possible effects of artificial hormones which are unintentionally introduced into our diets, like BPA. So as I continue to read about it, I hope I’ll start to figure out if this is an idea with legs or just a passing fancy. In the interests of keeping this post readable, I haven’t written about all the interesting facets that I’ve encountered in this phenomenon, so feel free to ask questions. And there are certainly holes in the idea beyond the ones I mentioned, so feel free to point those out, too!

Edited to add: I messed up in suggesting that intra-uterine position might affect dogs the way it has been shown to affect rats, humans, and cattle. Dog placentas are fundamentally different from rat and human placentas, and also different from cow placentas (which form a third category). In short, it would be pretty unlikely for two fetuses to share hormones in-utero in a dog the way they can in rats, humans, and cows. So while I still think it's an interesting idea that a dog fetus could be exposed to different amounts of testosterone in-utero (probably due to processing of hormones by the placenta) and that this could affect its adult behavior, I want to emphasize that it is actually not likely that these hormones could be from other fetuses in a dog. The hormones would be from some difference in the mother, not from a chance alignment of the offspring. So in summary: if your bitch gives birth to one female and two males, that's not a reason to worry about masculinization and temperament in the female.

References

• Ryan B.C. (2002). Intrauterine position effects, Neuroscience and Behavioral Reviews, 26 (6) 665-678. PMID: 12479841
• Monclus R., Cook T. & Blumstein D.T. (2012). Masculinized female yellow-bellied marmots initiate more social interactions, Biology Letters, 8 (2) 208-210. DOI: 10.1098/rsbl.2011.0754
• Hotchkiss A.K., Lambright C.S., Ostby J.S., Parks-Saldutti L., Vandenbergh J.G. & Gray L.E. (2006). Prenatal Testosterone Exposure Permanently Masculinizes Anogenital Distance, Nipple Development, and Reproductive Tract Morphology in Female Sprague-Dawley Rats, Toxicological Sciences, 96 (2) 335-345. DOI: 10.1093/toxsci/kfm002
• Bánszegi O., Altbäcker V. & Bilkó Á. (2009). Intrauterine position influences anatomy and behavior in domestic rabbits, Physiology & Behavior, 98 (3) 258-262. DOI: 10.1016/j.physbeh.2009.05.016
• Correa L.A., Frugone M.J. & Soto-Gamboa M. (2013). Social dominance and behavioral consequences of intrauterine position in female groups of the social rodent Octodon degus., Physiology & behavior, PMID: 23769692