Ruminations of a dog scientist on a 96-well plate

I've been doing a lot of bench work in the laboratory lately. This involves filling the tiny little wells on a plate with my ingredients (sample, reagents, primers) and then inserting the plate into a reader. The machine takes the plate up with whirring sounds that always fascinate me. I know there are little robot arms in there moving the plate into place, and I wish I could watch the process. But as I listen to the robot work, I sometimes think: is this the closest I get to living, moving animals now? How did I get here, so separated from fur and behaviors and emotions?

96 well PCR plate

My long term research goal is to understand the differences in how brains work in dogs who suffer from fear issues compared to resilient dogs who take life's arrows a bit more in stride. I'm doing this by studying gene expression in the brains of foxes who have been bred to be fearless (“tame”) or fearful (and aggressive — those who study them just refer to this line as “aggressive,” though).

My approach is, at the moment at least, deeply reductionist: what are the differences in gene expression in a few brain regions in these two lines of foxes? In other words, does one group make more of a certain kind of gene than the other? My hope is that I’ll be able to make some conclusions about the differences in function in these brain regions between the two lines of foxes, and that what I find will be relevant to fearful dogs. But I find myself burrowing deeper and deeper into learning about very small parts of the brain, and then very specific functions of those parts to the exclusion of other parts. Currently I’m learning about the pituitary gland — no, wait, just a particular cell type in the pituitary gland, the corticotroph — no, wait, just a particular set of processes of the corticotroph, how it releases one particular hormone into the bloodstream.

So in my daily work, I do things like take some tissue and extract all the RNA from it (throwing out DNA, proteins, cell structure, all sorts of interesting information — that's not what I'm working on or able to assess at the moment). I use PCR to extract a tiny piece of RNA from the complete transcriptome (all the RNA from that tissue), throwing out even more information. And then assess the expression level of that RNA, resulting in just one number. One number out of all that information after a day’s work.

Behavior can’t really be fully understood using this reductionist approach. If I do find a few important gene expression differences in a few small brain regions, they won’t explain the whole story of why an animal has a fearful personality. They’ll be a tiny, tiny piece of a complicated network of interactions involving genetics and life experience. But in order to get at that tapestry we have to first be able to visualize the threads that make it up. So here I am, in the trenches, doing that.

A recovering shy dog.

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.

The future of behavior medication?

We don’t actually know how behavior medications work. We know how they change the operations of cells — for example, we know facts like “this medication makes cells slower to recycle this particular chemical.” But we don’t have a good idea of how those cellular-level changes result in behavior-level changes. We don’t know how these medications make individuals feel better.

And that’s a problem, because not every individual responds to a particular behavior med in the same way. A pathologically fearful dog might have nasty side effects on one med, no response to a second, and then respond beautifully to a third. It’s hard on owners to have to try a variety of medications before finding the right one, especially as it takes a month or two to be sure that a particular medication is or isn’t working. (Oh, yeah, and the same is true for humans who use these drugs.)

If we knew how these medications worked, we might be able to figure out who they would work on without so much cumbersome trial-and-error. Imagine taking your shy dog to a veterinary behaviorist, who would do a genetic test and prescribe the right drug based on the results, to go along with behavior modification exercises.

My current work focuses on gene networks that differ in the brain between animals who are shy and aggressive and animals who are confident and friendly. I've always felt that my requests for funding have been a little hand-wavy as I have argued that surely my findings may help us understand behavioral medications better... You know, someday. Someday maybe my findings will help us design better medications, in fact. But that day seemed a really long way off.

Until I read Ed Yong's story “CRISPR’s most exciting uses have nothing to do with gene editing”. CRISPR is a fancy new gene editing technology that has everyone talking about science fiction coming to pass: being able to edit human (and animal) genes to make designer babies (and animals). Edit out the gene variants for genetic diseases before a baby is born! (But hopefully don't slide down the slippery slope to editing height, skin color, eye color, personality...)

But it turns out that CRISPR may have a more subtle use: gene regulation. Soon, scientists may be able use it to tell individual genes to turn on and off (to make more or less of their product). Rather than permanently editing genes in embryos, we could temporarily modify the output of genes in adults. Suddenly my quest to find the sets of genes affecting shyness seems less quixotic. Maybe my discoveries (do you like how I assume I’ll have discoveries? Let’s just pretend it’s a sure thing) won't have to wait for drug discovery work to be useful. Maybe we’ll be able to directly turn the volume up or down on those particular genes, directly affecting pathological shyness.

Scary? Yeah, it's not something I foresee being used therapeutically in the next few years, not until we understand the brain well enough to be able to predict side effects. But it’s really cool to imagine that some day we may have this sort of fine-tuned control over psychological diseases.

Science with a sense of humor

I’m writing a peer-reviewed article right now. I can almost guarantee it’s something pretty much none of you will be interested in (it is not about dogs or foxes, but about genomics technology), but when it’s out I’ll do my best to blog about it in a way that makes it seem exciting. We’re at the review stage: reviewers give us a bunch of comments, we make the changes to the article, then we write a letter back to the reviewers. The letter is supposed to say things like “Thank you so much for your insightful comments. We made all the changes you suggested!”

One reviewer comment pointed out that at one point in the article, I had referred to humans as a model species. Now, model species are normally species that we use as models for humans. The best examples are laboratory rodents: we study rats in the hopes that what is true for rats is true for humans. The rats are a model species.

The reviewer commented “Are humans really a model species?” At which point my boss basically put her head in her hands and was embarrassed that we hadn’t noticed this stupid gaffe we’d made.

In my first draft of the reply letter to the reviewers, I replied to the question about whether humans are a model species: “They are to this veterinarian!” I, of course, love to read human research in the hope that what is true for humans is true for dogs. (But I made the change in the manuscript.)

I pointed this out to my boss and said, “Did you like my veterinarian joke?”

She: “Yes.”

Me: “Is it OK to have jokes in letters to reviewers?”

She: “No.”

Sigh.

Who funds dog research?

As I move through my training and think ahead to my future career, I wonder: who will pay for all this research I want to do on dogs? I have so many questions to ask!

• What changes happen in the canine brain as it enters, and then leaves, the socialization period?
• How is the brain of a fearful dog different from that of a confident dog?
• What are the genetic differences behind these variations?
• How do environmental differences (prenatal stress, early learning, adult life) change the brain?

In other words, what are the mechanisms in the brain that differ in fearful dogs — receptors, neurotransmitters, synaptic wiring? And how can I learn about them without using invasive (painful and/or terminal) techniques?

Who are the caretakers of Dog, the species, who care about fearfulness? We as dog owners and lovers care, but dog owners and lovers aren’t the ones who are trained to heal unhealthy dogs, to perform research aimed at understanding them, and we (mostly) aren’t the ones who breed them. So who are the groups who are the caretakers of Dog, and what subsets of Dog do they care for?

Veterinarians

We (I am a veterinarian) are trained to heal sick dogs. Relatively few veterinarians perform research compared to those who engage solely in clinical practice. But some do perform research: most commonly as faculty at veterinary schools alongside a clinical practice, or less commonly as researchers without a clinical practice at research instititutions.

Veterinary research, as a result of this strong emphasis on healing the unhealthy, is focused on clinical results. Veterinarians most commonly perform research which asks questions about the effectiveness of particular techniques — medications, surgical approaches, new equipment. Veterinary research very rarely addresses root questions about mechanisms, particularly in the area of behavior. Rather than asking “How are the brains of fearful dogs different?”, veterinary research is more likely to ask how we could fix a fearful dog: “Does this medication make a fearful dog less fearful?”

In fact, as I pursue my mechanism-based questions, I am asked if I miss being a veterinarian. The perception is that because I am engaged in basic, rather than clinical, research, I am no longer working as a veterinarian.

Basic science researchers

If veterinarians do clinical research studies, then who does basic research biomedical studies, studies that look not at how to fix problems but at how the body works? Ph.D. researchers are more likely to do this sort of research, which is why I am currently engaged in obtaining a Ph.D.

Traditionally, Ph.D. researchers have not been interested in dogs. In fact, way back in 2004 when I was originally deciding between a Ph.D. and a D.V.M., I was told by a Ph.D. animal behaviorist, “Ph.D.s don’t study domesticated animals. Veterinarians study those.” (Actually, veterinarians mostly just try to fix unhealthy domesticated animals, not study the healthy ones.)

That perception has changed in a big way in the intervening eleven years. There are now multiple laboratories studying dogs. But where does their funding come from — who cares enough about dogs as dogs, not as models for human problems, to provide the impressive funding needed for a genomics study? (The work I am doing for my Ph.D., sequencing messenger RNA, costs around $45,000.)

The U.S. federal government

The traditional source of funding for basic research is the federal government: the National Institutes of Health for health-based research and the National Science Foundation for more basic research. But these two massive institutions are very much focused on human health — as they should be, as they are funded by the tax dollars of American citizens. The economy can’t support all the research American researchers would like to do, and getting an NIH or NSF grant is becoming more and more difficult as grant funding is cut. Funding to study dogs as models of human disease? Maybe, but isn’t it easier to study laboratory rodents (on which you can perform invasive studies) or work on humans directly? Funding to study dogs as dogs? Go lie down until it passes.

In my experience, the small number of laboratories directly studying dogs are either studying them as models for questions about human health or evolution, operate on a shoestring budget, or have great trouble obtaining funding for what they want to do.

Animal welfare organizations

So who cares about dogs? Animal welfare organizations, some of which are national in scope and do perform research. Some major players in this field are the American Society for the Prevention of Cruelty to Animals (ASPCA), the Center for Shelter Dogs (CSD), and the Humane Society of the United States (HSUS). I am most familiar with the research coming out of the ASPCA and the CSD, and it is exciting stuff. But it is again mostly focused on applied questions: how can we help the shelter dogs in our care?

I reviewed some of the research these two organizations have performed on how to identify and treat food aggression in shelter dogs in my story for the Bark on shelter behavioral assessments. This was ground-breaking research and I am really glad to see it published. But it doesn’t ask the basic (i.e., non-applied) research questions I am interested in: what is it about the brains of these dogs that differs from the brains of dogs without food aggression? That kind of research doesn’t have immediate applied benefit. You can’t take it to a shelter worker with a recommendation about whether or not to put a food aggressive dog on the adoption floor. It is incredibly impressive that these shelter-focused organizations perform any research at all, and it is absolutely appropriate that the research they perform should have a highly applied focus, with clear questions that, when answered, will provide guidance on how to improve the lives of shelter dogs, immediately. They do not have the resources to pursue these sort of mechanism questions that I want to ask, which do not have immediate applicability.

So who cares about understanding how dog brains work, with the hope that that information will provide a base for future applied research? Who cares about the whole species, not just the subset in shelters or the subset in hospitals?

Breed organizations

Breed organizations care very much about the health and welfare of dogs, and in fact have provided funding into the mechanisms behind health issues specific to their breed. A recent paper about associations between spay/neuter status and health issues in Golden Retrievers was partially funded by the American Kennel Club’s Canine Health Foundation (AKC/CHF), and a similar study on Vizslas was funded by the Vizsla Club of America Welfare Foundation. (I blogged about these studies elsewhere.)

These organizations can fund basic research on how and why particular diseases occur in their breeds, and may even be willing to fund expensive genetic studies, such as a recent one on the genetics of cancer in Golden Retrievers, supported in part by both the AKC/CHF and the Golden Retriever Foundation. However, their focus is very much on the problems of a particular breed. My questions are broader: why do dogs of all breeds have different personalities, some more or less fearful? These organizations are really the caretakers of breed subsets of Dog, not of Dog itself.

Who, then?

Who does that leave as a group willing to fund studies on Dog? On problems common to all breeds? On problems which may or may not provide good models for humans? If I hope to one day run a laboratory which studies these problems, who can I hope to help pay for the research?

I would be remiss if I did not mention Morris Animal Foundation here. While their important Golden Retriever Lifetime Study happens to focus on the health issues of a single breed, their mission is to fund research into studies of small animals (dogs and cats), livestock, and wild animals, with no breed limitations. This group is doing important work, and I applaud them.

But one organization is not enough for a laboratory to depend on for survival, especially in these times with research funding so hard to come by. And so I wonder: are we, the dog lovers of the world, the ones to start supporting research into what it is to be a dog? We, who own dogs of all breeds and mixes, with all sorts of problems, who know what problems most plague us as owners — not just medical problems, but behavioral ones?

And so I leave you with my dreams of crowdfunding, in which a researcher proposes a study and asks the public to support it through donations. Such an approach allows the dog community to take the task of answering basic questions about Dogness into their own hands. This direct connection between a researcher and the community affected by their research is a new benefit of this age of social media. Is this approach right for this particular problem? Time will tell.

Do spayed and neutered dogs get cancer more often?

[Note: this post was originally published at the lovely Julie Hecht's Dog Spies blog at Scientific American.]

Where I live, in America, it’s taken for granted that responsible owners spay or neuter their dogs. The population of homeless animals is still large enough that risking an unwanted litter is, to many owners, unthinkable. And spay/neuter is just what people do. But two papers were published, in 2013 and 2014, suggesting that these widely accepted surgical procedures may lead to increased long-term risk of certain kinds of cancers. These studies ignited a debate which had been smouldering for years: are there unwanted health consequences associated with altering a dog’s levels of estrogen or testosterone?

The 2013 paper looked at Golden Retrievers. The authors reviewed data from veterinary hospitals, comparing Goldens who were diagnosed with various diseases, those who were not, and the spay/neuter status of each group; they found a correlation between spaying or neutering and cancers such as osteosarcoma, hemangiosarcoma, and mast cell cancer. The 2014 paper used a voluntary Internet-based survey to perform a similar investigation in the Vizsla breed. They also found correlations between spay/neuter status and mast cell cancer, hemangiosarcoma, and lymphoma.

These are scary results, but I caution that studying the causes of multi-factorial diseases like cancer is incredibly challenging. Take the Golden Retriever study, a retrospective study using data from a veterinary referral hospital. This study was limited to dogs whose owners chose to bring them to a relatively expensive referral hospital. This is the kind of place where you take your pet when he has cancer and you are willing to spend a fair amount of money to help him. As a result, this hospital’s records probably provide a great source of data on companion animals living with concerned owners, particularly owners who have provided excellent medical care for much or all of the animal’s life. However, this hospital’s records are less likely to provide data on animals whose owners have provided sub-optimal care. This kind of bias in sample selection can have a significant effect on the findings drawn from the data.

The Vizsla study used an Internet-based survey instead of hospital records. Like the Golden Retriever study, this study could have found itself with a biased sample of very committed dog owners, in this case owners who engaged in dog-focused communities online and who had enough concern about the health of the breed to fill out a survey. This study additionally suffered from a lack of verified data; owners were asked to give medical details about their dogs and may have misremembered or misinterpreted a past diagnosis.

Don’t get me wrong – these were both important studies, and they did their best with the available resources. I applaud both sets of authors for putting this information out there. But the studies both have their limitations, which makes their findings difficult to trust or generalize to other populations of dogs.

Meanwhile, another 2013 study presented some other interesting results. This study drew data from multiple referral hospitals to determine the causes of death in spayed or neutered versus intact dogs – and they found that spayed and neutered dogs, on average, lived longer than intact dogs. Intact dogs were more likely to die of infectious disease or trauma, while spayed or neutered dogs were more likely to die of immune-mediated diseases or (again) cancer. In other words, while spayed or neutered dogs did get cancer, it didn’t seem to shorten their lifespans.

This study shed a new light on the cancer question. It suggested that perhaps spayed or neutered animals might be more likely to get cancer simply because they were living long enough to get it. Intact animals were more likely to die younger, perhaps simply not aging into the time of life when the risk of cancer rises.

So where does that leave us? Is there a causal link between spaying/neutering and cancer? I think the question is still wide open. What we really need is a study that follows animals forward throughout their lifetimes instead of using retrospective records or surveys to get the data – and, thanks to Morris Animal Foundation’s groundbreaking Golden Retriever Lifetime Study, we are getting just that. This study is enrolling Goldens as puppies and following their health over the course of their lives. It will be years before the study gives us answers, but it provides hope for more solid data. (Of course, it still can’t address the issue of bias, in that owners who enroll their puppies in this study could be highly responsible dog owners who provide excellent medical care!)

We can, however, do something about cancer in dogs without waiting for the results of that study. It is no coincidence that two of the studies discussed here investigated Golden Retrievers. Sixty percent of Golden Retrievers will die of cancer. That is indisputably a problem with the genetics of the breed, and other breeds suffer from similar problems. We should be attacking cancer on all fronts, and this is a front we don’t have to study first. Golden Retriever breeders are between a rock and a hard place, trying to breed for health in a gene pool which doesn’t have enough genetic diversity to support it. The solution is to bring in new blood from gene pools with much lower risk of cancer, breeding dogs who don’t look like purebred Goldens for a few generations to revitalize the breed as a whole. Genetics contribute far more to risk of cancer than whether an animal is spayed or neutered. We clearly have a strong desire as a society to reduce the incidence of cancer in Golden Retrievers and other breeds. While we’re studying risk from spaying and neutering, let’s address the genetics question that we know we can fix.


Image: Rob Kleine, Golden Retriever, Flickr Creative Commons License.

References

Torres de la Riva G, Hart BL, Farver TB, et al. Neutering Dogs: Effects on Joint Disorders and Cancers in Golden Retrievers. PLoS ONE 2013. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0055937

Zink MC, Farhoody P, Elser SE, et al. Evaluation of the risk and age of onset of cancer and behavioral disorders in gonadectomized Vizslas. Journal of the American Veterinary Medical Association 2014;244:309–319. http://avmajournals.avma.org/doi/full/10.2460/javma.244.3.309 [Paywalled]

Hoffman JM, Creevy KE, Promislow DEL. Reproductive Capability Is Associated with Lifespan and Cause of Death in Companion Dogs. PLoS ONE 2013. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0061082
 

Can a father's stressful experiences affect his offspring's stress system?

I am again indebted to my APDT students, who asked a very interesting question which turned into a blog post. This time it was: “Can a father’s stress be passed on epigenetically to his offspring through sperm?” Warning: epigenetic geekery ahead. If you are not in the mood for some technical terminology, this may not be the post for you.

Source: Wikipedia

I’ve blogged about epigenetics before (on the epigenetics of fear and of stress) and there are summaries of what epigenetics is in those two posts, but basically it’s changes in the DNA that don’t involve the sequence of bases. We’ve been so focused on the importance of DNA sequence as we’ve learned more and more about genetics, but in recent years the importance of other factors has started to become obvious. It’s like saying that the content of a book isn’t the only thing controlling whether the book gets read or not — it also matters whether the cover is appealing, how much it costs, and whether it’s shelved where people can find it.

One of my students tried to answer her own question and found this fascinating article:

Rodgers A.B., S. L. Bronson, S. Revello & T. L. Bale (2013). Paternal Stress Exposure Alters Sperm MicroRNA Content and Reprograms Offspring HPA Stress Axis Regulation, Journal of Neuroscience, 33 (21) 9003-9012. DOI: http://dx.doi.org/10.1523/jneurosci.0914-13.2013

Essentially, Rodgers et al stressed out some male mice and then tested their offspring six ways from Sunday to see if the effects had been passed on. And they had, but in some surprising ways.

The mice

Male mice were chosen because we have evidence that environmental effects can be passed on epigenetically through the sperm. Because sperm are made throughout the male’s lifetime, they can easily serve as messengers to pass information about the environment on from fathers to their offspring. Mothers can’t pass this information on through their eggs (so far as we know), because eggs are made before a female is born, and can’t easily be changed later. Of course, a mother has plenty of other chances to pass on information about the environment to her offspring: while they’re in utero and while they’re dependent on her. But for dad, sometimes he just has that one chance; he may never interact with his offspring in any other way than through the information in his sperm.

(Why do parents want to give their offspring information about the  environment? To let them know, at formative times in their lives, how to develop. If the world is a safe one, you don’t need a highly reactive stress system. But if it’s a dangerous place, you need your store of cortisol ready to go. It’s easiest for these sorts of developmental decisions about how to tune the stress system to be made early in development — in utero or post-natally — so that’s why parents have systems to pass the information on early, early, early.)

Male mice, then, were chosen for this study. In addition to the control group of unstressed mice, there were two groups of stressed mice: mice who were stressed in adolescence, and mice who were stressed as adults. Epidemiological research in humans suggests that adolescence is an important time for epigenetic changes in sperm.

The stressors

The males were subjected to a variety of stressors. In reading the list, I was torn between sympathy for the mice, and bemusement at the entries:

Stressors, selected because they are nonhabituating, do not induce pain, and do not affect food or water intake, included the following: 36 h constant light, 15 min exposure to fox odor, novel object (marbles) overnight, 15 min restraint in a 50 ml conical tube, multiple cage changes, novel 100 dB white noise (Sleep Machine; Brookstone) overnight, and saturated bedding overnight.

Wet beds! Scary white noise! Scary marbles! And yet yes, probably very stressful to a mouse, and I should not make fun.

Breeding
The males were given time to recover from the stress and then bred. They were removed from the cage as soon as they had mated with the female, which took 1-3 days, to minimize their interactions with her, so that their stress levels could not affect her. (However, the smart reviewers at F1000 [warning: not open content] note that a stressed male might have been more aggressive in mating, which could cause the female to alter her care of her offspring.)

Offspring stress response
The offspring of stressed males, it turned out, had a less responsive stress response than the offspring of unstressed males. In other words, when these mice were stressed by being restrained in a conical tube for 15 minutes, the ones whose fathers had undergone the variety of stressors had a smaller cortisol response compared to the ones whose fathers had not been stressed. The result was almost exactly the same whether the fathers had been stressed as adolescents or as adults, which surprised the researchers.

Now, if a mouse receives information from his father (or his father’s sperm, but you know what I mean) that the world he’s going to live in is a stressful place, I would have expected that that mouse would develop a more reactive stress system, not less. Worried that terrifying marbles or a wet bed are going to attack you at any moment? Then you had better have your stress response at the ready, right?

The stress system is, of course, much more complicated than that. We don’t understand yet why some models of stress system dysregulation show less reactive responses and some show more reactive responses. For example, humans with depression or PTSD can both show either more or less reactive stress responses than mentally healthy humans. So what exactly does this mean for these particular mice? The next thing I would do is to look at their behavior. Do they act more stressed?

Offspring behavior
The offspring were subjected to quite a few and quite varied tests to see if their stress behaviors were different. The researchers tested things like how much the offspring startled in response to a loud sound; how fearful they were of being in a brightly lit box versus a dark box (mice feel safer in darkness); how long they struggled when suspended by their tails; and more. Really surprisingly (to me, at least), there were no behavioral differences between the offspring of the stressed fathers and the offspring of unstressed fathers, despite this significant difference in stress system responsiveness. So what does that mean?

The researchers tested a bunch of other stuff that left them empty handed as well, like gene expression differences in the brains and adrenals of the different sets of mice. All nothing. But what did they find that was different? They found an epigenetic difference in the fathers’ sperm.

microRNA changes in sperm
Epigenetics is all about gene expression: determining which genes are used frequently to make their associated proteins, and which are left to lie dormant. The two best understood epigenetic mechanisms, acetylation and methylation, affect how much messenger RNA (mRNA) is transcribed from a particular gene. If there is more messenger RNA for a particular gene, then it’s easier to go the next step and make more of the protein that that gene codes for, and that gene’s expression increases. The mechanism that these researchers looked at is different. Instead of methylation and acetylation, they looked at microRNAs (miRNA) in the sperm of these mice. Where methylation and acetylation affect how much messenger RNA is generated, microRNAs attach to the messenger RNA itself after it has been created, and silence it.

The way it works is this: since RNA is basically half of a double strand of DNA, it’s really sticky. It wants to find something that looks like its complement and stick to it. So microRNAs are little bits of RNA that stick to particular messenger RNAs. Then when the cell takes those messenger RNAs and tries to use them to make a protein — it can’t. Because there’s this microRNA stuck to it, blocking the sequence of the message. So microRNAs reduce the expression of a gene, but they do it one step later on in the gene to protein pathway than methylation or acetylation does.

Back to our book example, it’s like if you have a cookbook (the DNA). You copy out a recipe on a piece of paper for later use (the RNA). Then you use the recipe to make cookies (the protein). Methylation puts big rocks in front of the bookshelf so you can't get to it and get at the cookbook. Acetylation glues the pages of the book together so you can’t read it. But microRNAs are your obnoxious husband who draws in marker all over your copied recipe, so you have to go back and copy it out again. (Disclaimer: while my husband is quite capable of being obnoxious, he has never defaced any of my recipes. He has scribbled notes on the medication list for my dogs in the face of my express requests to the contrary, however. Rosie hasn’t been on ciprofloxacin for six months but it still says “cipro” on her meds list. It’s like he’s incapable of thinking ahead.)

There is a lot we don’t know about microRNAs. The whole epigenetics field is like this: we are getting to the point where we can detect these changes, but we still don’t really know what they mean. So in this study, they found that 9 microRNAs were expressed at different levels in sperm of the stressed mice versus the unstressed mice. We can make some predictions, using computer algorithms, about which messenger RNAs these microRNAs were going to stick to and silence, but we don’t know for sure that that’s what they were actually going to do.

Still, the predicted list is pretty interesting, because it contains the messenger RNA for the enzyme which controls methylation. Methylation! Another epigenetic mechanism! So is there some epigenetic chain going on here? The dad passes on microRNAs which will result in the DNA of the offspring being more or less methylated. It’s so hard to know what that means, because methylation has very different effects depending on which gene is affected, and this change is a more global change. But it’s a really intriguing finding, isn’t it?

Conclusions
This study is exciting, but I still felt a bit of disappointment as I read it. No behavior changes? Really? Is it really significant without the behavior changes? I mean, do we really care about stress system changes if there are no behavior changes? Of course we do, and I wonder if future studies will investigate different behaviors, or behaviors at different points in the mouse’s life, and then we’ll understand this system a little better.

What does it mean for dogs? Of course it is immediately applicable to the question: if a male dog is stressed, will this stress affect his offspring? The answer is a nice solid maybe. In some way that we can’t really predict or define.

But at another level, this is another step in our progress towards understanding how genes and the environment interact. Stressful situations change gene expression in the stressed individual and possibly their offspring. How, why? How can we measure it? How can we use our knowledge to help an animal who has been traumatized, or undersocialized? Watching the field of epigenetics unfold is so much fun: everything is new, we understand so little, but the new technologies are coming so fast that we’re learning more and more.

On nature and nurture and their interactions to make a personality

My mom called me yesterday because she had experienced some Science and was excited about it. She was watching a TV episode about aggression and how it appears in nearly every species. She called me to say that she thought my lab should look for the gene for aggression. “It should be easy,” she said, “because it should be the same gene in every animal.”

Aggressive silver fox

Yeah, you’d think that there would be single genes controlling bits of our personalities (human and dog — I think dogs are much more interesting, but in this case it’s much the same problem). Only ten or twenty years ago we thought we were in the endgame to find these genes: once the human genome was sequenced, we expected to be able to do a series of big studies to find these answers. Take a few hundred humans and sort them into “violent“ and “not violent.” Then look at markers in their genomes and use computers to find associations: all the violent people should share one marker, which will tell you where the gene for violence is. Done.

But we did those studies and we found, again and again, that these sorts of personality traits don’t give up their answers this way. In fact, in the case of zero personality traits have we found one (or even two or three) genes that control that trait. Sometimes we find genes that we think control a solid chunk of a trait, only to find that it was a statistical error — if you ask enough questions, you’ll find an interesting set of data just by chance. But if you ask the same question of another set of data (in other words, do another study), you’ll see that the first one was wrong. And this is what we have seen, for trait after trait.

Now, occasionally we’ll find a personality trait for which a little bit of it can be explained by one gene. When I say a little bit, I mean that if there is a normal amount of variety in this trait — say, in how violent a person is, ranging all the way from a pacifist to a psychopath — then the genes we find will explain about 0.1 percent of that variety. The rest is — what? Chance? Environment?

It’s a bunch of things, probably. For one thing, it’s surprisingly hard to define a personality trait. What’s violence? In dogs, we diagnose different kinds of aggression: territorial aggression, owner-directed aggression, dog-dog aggression, fear aggression. Are these all the same thing? Probably not. So instead of looking for one trait, “aggression,” should we look for four traits? Maybe. But do we actually know that those are the right four? Maybe there are six. Maybe there are ten. Maybe there are a hundred. We need to understand the traits we study better, and ask more detailed questions about them.

For another thing, yes, environment is important! Genes are important, but they are nowhere near the whole story. And environment is complicated. Certainly the difference between a pet store puppyhood and early life with a responsible breeder is huge. But can you lump early life experience into two bins, “good” versus “bad”? There are all kinds of variables. In the pet store, what kind of crate was the puppy kept in, how much interaction did it get, how young was it when it arrived? At the breeder’s, were there other adult dogs besides the mother to interact with, were there any small children, were there any bad interactions with other dogs or people? And a hundred, a thousand more questions.

I read recently about a pair of conjoined twins with very different personalities. These two had the same genes, because they came from the same embryo originally. And they had the same environment, because due to being conjoined they had to spend their lives in each other’s company. So how could their personalities differ? The article theorized that they reacted to each other, with one taking a bold, outgoing role and the other becoming shy and retiring in compensation.

And finally, the most interesting idea, in my opinion as a genomics researcher: what if we aren’t going to find the answer by looking at the sequences of DNA that make up genes? What if we are going to find the answer by looking at how the genes are regulated? If it isn’t that my dog is more fearful because some gene is a little broken, but she is more fearful because some gene is getting turned on much more or much less often than it should? It’s hard to investigate gene regulation when you have questions about the brain, because to do it you kind of have to get inside the brain, and it’s hard to do that without killing the person you’re studying. But I think looking at regulation is where things are going to have to go, and researchers are working on finding non-lethal ways of doing it.

So, nature and nurture: both important. Personality: super, super complicated. But also wicked interesting.

[If you’re a dog trainer or just interested in dog genetics, you can learn about the genetics of dog behavior with me this summer in an online course with the APDT!]

Open access dog salivary cortisol data

I finally got around to sharing the data from my study of dog salivary cortisol levels on figshare. I have meant to do this for months. Particularly, I wanted to do it so that I could wear the cool “I’m a figsharer!” t-shirt that Mark Hahnel gave me at scio13. How embarrassing would it be to wear that shirt and have someone ask what you shared and have to admit that you still haven't actually shared anything? But I am a figsharer now. So if you want numbers, go check it out.

Oh, and in case you’re interested in the associated paper, that’s here (but, sadly, not open access):

Hekman, Jessica P., Alicia Z. Karas, and Nancy A. Dreschel. “Salivary cortisol concentrations and behavior in a population of healthy dogs hospitalized for elective procedures.” Applied Animal Behaviour Science (2012). http://dx.doi.org/10.1016/j.applanim.2012.08.007

Looking at dog brains

Today I was privileged to visit Dr. Greg Berns' laboratory to see awake dogs in an fMRI. In vet school, of course I saw dogs getting MRIs of their brains as part of medical diagnostics, in hunts for cancer, stroke, inflammation, etc. But because an MRI requires that the subject hold perfectly still for several minutes at a time, these dogs were under general anesthesia, which is both expensive for the owner and physically difficult on the dog.

In humans, we can use the related technology, functional MRI (fMRI), to see changes in brain activity in response to different stimuli, such as music, smells, or looking at pictures. This is a useful tool in research, for example as we try to figure out which brain areas perform which tasks. In dogs, we haven't been able to do such studies, because the only way to keep dogs still enough for an fMRI has been to anesthetize them, and obviously a sleeping dog isn't going to have a meaningful reaction to external stimuli.

At Dr. Berns' lab, they have trained dogs to hold still in an fMRI machine while resting their chins on a chin rest. Can your dog hold its head perfectly still for minutes at a time? What about in a strange room, with loud machine noises all around, with ear muffs on to protect their hearing? It's an impressive feat, and done using entirely positive methods. (The training protocol was developed by Mark Spivak of Comprehensive Pet Therapy, Inc.)

I was most impressed by the dogs' relaxed body language. They entered the machine willingly, when their owners asked them to. They lay down with their chins on the rest and waited. As I watched from behind, I could see that many of the dogs were lying on one hip or even frog-legged, in very relaxed postures, suggesting that they were comfortable being in the machine. (Have you ever had an MRI? It is a claustrophobic experience. Humans getting MRIs would benefit from the extensive conditioning preparation that these dogs had, as well as having a loved one present to feed them treats periodically!) Some dogs would balk at some points and exit the machine, at which point their handler would ask them to return and they would. Dogs always had the opportunity to leave. At the end of the test, they came out happy and wriggly.

Highlights of the day for me:

• The Boston terrier who hurled himself into the fMRI at full speed and then became rock-still for as long as his owner asked him to. That dog was committed to his fMRI experience! (Who would expect the Boston to be the calmest dog in the magnet?)
• The dogs with their ear protectors wrapped onto their heads with an elastic material normally used to attach catheters and the like. They looked hilarious.
• The treats fed to dogs on the end of long sticks so that they're easier to deliver inside the magnet. Ingenious.
• Personally getting to participate in experiments by giving hand signals to dogs who were in the magnet, watching me intently as they waited for their treats.

The joke around the lab is that these tests will tell us why our dogs really love us: are we best friends or just food dispensers? It is a joke because of course fMRI is not a test for love; science has some trouble testing for squishy concepts like that. But fMRI does give us a new  tool for guessing at what goes on in doggy heads, in addition to having to muck around with hormones like cortisol (as I have done) or strange little cognition tests like separation experiments or pointing experiments, as others have done. We have never been able to use this tool on awake animals before, so this is a huge step forward.

It was a fascinating day. I am deeply happy to see non-invasive research going on which takes the welfare of its canine participants into account, and waiting with bated breath to find out the results of the experiments I saw.


Further reading
 

• Greg Berns' home page at Emory

• Functional fMRI in awake unrestrained dogs (open access)
•  

Why cortisol sucks as a measurement of stress

Standing in the ward of a veterinary hospital, you see a dog jumping up and down in his run, barking. Is he distressed at being here? Or is he just barking to get attention? Obviously, you decide to perform a research study on dogs in the hospital, to measure their stress levels. How are you going to go about getting some sort of numeric measurement of stress, so that you can perform statistics on your data and publish it in a journal?

People have approached similar problems in a lot of ways. One of the most common answers is to measure the dog’s cortisol levels. (Or corticosterone, if it’s actually a rat, mouse, rabbit, or bird.) This is the approach I’m using; cortisol is in saliva, which is why I spent so much time over the last year trying to get dogs to drool more.

What is cortisol, actually? There was an excellent post on mindhacks.com recently about cortisol and how science journalists sometimes misrepresent it. I’m going to use dogs as my examples, but what I have to say is just as relevant to studies on humans, if that is your cup of tea. It might help you to understand some of the news stories floating around about various things which “raise cortisol levels.”

Cortisol is a hormone made by your adrenal glands. Your adrenals sit next to your kidneys, but they produce cortisol in response to hormones released from your brain in stressful situations. So we like to measure cortisol levels because they tend to increase when the brain is sending out “I’m stressed” messages.

OK, but what is stress? What I care about, and what many people who measure cortisol care about, is psychological distress — being yelled at, being scared you’re going to be eaten by a predator, being left in a loud veterinary hospital with no familiar faces around you. Stress is a lot of other things as well, however. It is hunger, illness, feeling too cold, having exercised recently. In fact, cortisol has a normal rise and fall over the course of the day to help your body know that it is time to be awake or to go to sleep. Your adrenals also produce it to help you deal with anything which requires some extra energy. You may need that extra energy for a good reason, such as competing in an athletic event. “Good” stressors like that are known as eustressors. So if you’re going to use cortisol to measure stress, you are going to be measuring both eustress and normal daily stress like hunger, in addition to whatever source of distress you may be interested in.

Knowing this, you’ll try to design your experiment to work around the problem. You’re interested in whether dogs find their time in a veterinary hospital to be distressing. So you will try to remove eustressors from the equation — you will make sure that none of our dogs have exciting things like getting fed or taken for walks happen while we’re studying them. You will also make sure that all of your study dogs are healthy, since illness can raise cortisol levels. And you will measure the dogs’ cortisol levels at exactly the same time of day, because of cortisol’s diurnal cycle. (There is some very interesting debate about whether dogs, unlike all other mammals which have been studied, actually don’t have a daily cycle of cortisol. One theory is that they don’t because they sleep most of the day.) Now you believe you're just measuring distress.

Cortisol is still an awfully bad way to measure distress! Males and females react to stressors in different ways. (This has mainly been reported in humans, but it’s been said that studies of stress in rats which are limited to males miss an important segment of the population.) Age has something to do with cortisol levels as well, though mainly just in the very young and very old. In dogs, it is an open question of whether breed matters, but I’m guessing it does, since personality affects cortisol responses to stress in humans.

So you control for that, too. You get a bunch of dogs of the exact same age, gender, and breed. They are all laboratory animals, so you can be reasonably sure their histories are the same, and you aren’t going to find out at the end that half of them have spent more time in a veterinary hospital than others. (This wasn’t the direction I chose, but some studies do give it a go, using laboratory beagles of similar ages and only one gender. There are obvious ethical implications here, but that’s a post for another time.) You put these dogs in a veterinary hospital and measure their cortisol levels. Now are you measuring their reaction to the hospital setting?

Maybe. The next problem is that all these animals have their baseline “unstressed” cortisol level set at a different point. We don't understand all the genetics having to do with how this system works, but we are learning. We do know that a cortisol level that indicates stress (good or bad) in one individual might indicate total relaxation in another. Many studies deal with this problem by looking only at changes in cortisol levels. They measure cortisol before and after the stressor, and look at the difference, rather than at absolute levels. So let’s assume you can do this in your hospitalized dogs. You keep them in one environment for a few weeks or months, until they have time to settle in and relax, and you keep track of their average cortisol levels there. Then you put them in the hospital and look for a difference. Now are you measuring their reaction to the hospital?

You probably are, but what exactly are they reacting to? Something which is a stressor for one individual isn’t necessarily a big deal for another. For example, the Trier Social Stress Test (TSST) is a test specifically designed to raise cortisol levels in humans. It’s used to study things like how gender affects responses to stress. You stress the person out by making them do some public speaking and public arithmetic. But only 70% of people who take the TSST actually have increased cortisol levels compared to just before they took the test. Doing arithmetic in front of a hostile audience just isn't alarming for some people.

In the case of your hospitalized dogs, some don’t like the noise, and some don’t like being in a cage, and some don’t like having other dogs around, and some don't like all of the above. But some think it’s awesome to be in such an exciting environment with so much going on. The hospital isn’t just one big stressor, it is a lot of different little ones.

If cortisol is such a bad way to measure distress, why do we use it? Unfortunately, it is still the best understood method we have. There are lots of other methods, but they all have their own problems. It’s a good idea to use at least two methods together, actually.

So what do you do? Give up? My approach has been to cross my fingers (maybe close my eyes) and just proceed. I think a lot of research involves just circling around a problem, picking away at it until it starts to give in. Studies of stress may not be able to give precise answers to questions about stressors. But if enough of them are done, our picture of how the stress response works will continue to get clearer and clearer. It’s really hard to know what is going on in the mind of a member of a different species; it can even be hard to know what’s going on inside the mind of a member of your own species. We just have to keep trying.

[ETA: See the follow-up post, “Why do other measurements of stress suck worse than cortisol?”]

Links post

• Pluto's demotion is a great opportunity for science. Timmer concludes that scientists need to do a better job of communicating with the public.
• Westminster dog show. Interesting rant about the issues with dogs bred for the show ring today.
• Sciencefeed: Twitter for scientists
• How will I know when it's time? Musings on knowing when it’s time to euthanize a pet.
• How to avoid “he-said-she-said” science journalism
• American Humane and Cesar Millan have met to discuss the differences between their two camps of dog training methods. A board member of American Humane provides us with an urbane report in which both sides of the debate are recognized. I really appreciate this kind of reporting! I agree with many of the points made in this opinion post, but am not sure its tone is as constructive.
• We're so good at medical studies that most of them are wrong

Spring semester springs

This semester I have (almost) no classes. My one class is Journal Club, which mostly involves listening to other people present journal articles. One week in March I will have to present an article myself, which will involve a frenzied burst of activity, and then again a return to just research.

Just research! I am looking forward to it, although to be honest, last semester’s classes (biostatistics and bioethics) were surprisingly rewarding. The didactic part of this Master’s degree feels much more relevant and interesting to me than the majority of the didactic classes in the DVM coursework. That revelation is part of my increasing interest in pursuing a PhD in the future.

The distraction of classwork last semester caused a lot of dog videos to stack up, waiting to be analyzed. I was pulling in 2-3 videos of 20 minutes each per week; in theory, I would also analyze them every week, breaking each into 5 second bins in which I use a small web application which I wrote to note the dog’s location (front, middle, or back of the run), position (lying lateral, lying sternal, sitting, walking, etc), whether the dog is panting or not, whether the dog has moved at all in the last five seconds, etc. It is a somewhat tedious process, but on the up side, the good DVD player is in my bedroom, so I get to do the analysis in bed with my animals. I’m trying to analyze one or two videos a day for a while until the stack gets thinner. They take 1-2 hours to do, depending on how often the dog changes what it’s doing. The ideal video is of a dog lying still and not moving for 20 minutes. Two dogs have done that so far. I am fond of both of them.

Yesterday I scored two videos in the morning. The mid part of the day I spent reading about cortisol, for use in a summary paper about markers of stress. The summary papers provide useful background immediately, and should become part of my thesis paper this summer.

Then I went in to try to enroll dogs. I continue to work out the best timing strategy for this. There is a sweet spot somewhere around 5:30 pm. Too early, and I miss some dogs which haven’t yet filtered in unannounced. (It’s nice when dogs have appointments two days in advance, but sometimes people call to make surgery appointments and get scheduled for the next day, in which case the only way I have of discovering that dog is finding it in B Ward in the evening.) Too late, and the surgeons have all gone home, so I can’t find them to ask permission to use their patients. (I embarrassingly once said “awesome!” when told “Dr. Depardeau is still here even though it’s eight o’clock at night because his patient is bleeding out and he’s stuck in surgery.”)

Last night, I went in around 4 pm, and scouted. One dog looked like a great candidate. He was in the hospital for, I kid you not, a migrating foreign body in his foot. A tech and I discussed this as we stared at the anesthesia schedule for the next day: a migrating foreign body in his foot? His hospital record didn’t completely explain why they came up with this theory. I looked at the dog in B Ward, and indeed he had an open lesion on his foot, but he was mobile and alert and looked like a good candidiate. However, he had had imaging (radiographs and ultrasound) earlier in the afternoon, and had been sedated for that. It takes six hours for sedation to wear off to the point where I feel comfortable saying that it no longer affects their behavior, so I wouldn’t have been able to enroll him until 9 pm. That’s outside my window; enroll dogs at extremely different times of day and you risk seeing cortisol levels that differ because of cortisol’s diurnal cycle, not because of stress level differences. Oh, well.

I went to the gym for an hour, and came back around 5:30 in hopes that some new dog had shown up. No. I think the hospital is not quite back in full gear after the holidays, and also Mondays are typically slow. At this point I started looking more closely at the two smaller dogs that I had discarded before. I don’t like to use small dogs; they are put in cages, not runs, so I have to ask someone to move them into a run for my use. I also have a feeling (unscientifically) that small dogs react to stressful situations differently than large dogs do. But at this point I just wanted someone to enroll. One of the two turned out to be on meds which disqualified her. The other probably would have worked out — but his doctor had just gone home, so I couldn’t get permission to use him. (Surprising! People on a medicine rotation, as this resident was, are generally around much later than 5:45.) The lesson: if you think you might possibly want to use a dog, get the permission proactively. Don’t wait to see if someone better shows up later.

Then I went home and read more. I can’t quite get over the fact that I get to spend a year reading and writing. It’s great.

So that seems to be my routine for the early part of this semester. So far, so good. Soon I have to start doing things like doing the actual cortisol assay for the dogs I’ve enrolled, and build the results into a stress scale, but for now things are in a pretty solid rhythm.