Depression Post 3: Studying Depression in the Lab

Welcome to depression post three! Previously I covered the symptoms and etiology of depression, and some of the most common antidepressant medications and how they work. Now I’d like to go into some of the research behind it. After all, it’s not like we just grab a depressed person and say “here, take this and call me in the morning”. Every new drug that comes out on the market has to go through rigorous clinical testing to determine whether or not it works, and just as important, whether or not it is safe. And even then, drug companies and the FDA make mistakes.
This post (to save me writing another 2,000 words at a time) is going to focus on the way we study depression (along with other psychiatric disorders) using animal behavioral models, particularly rodent models. But of course, you can’t put a mouse on a couch and ask it how it feels about its mother. What researchers have found is that there are ways of studying the efficacy of traditional antidepressants. So rodent tests for depression are not so much tests for depression so much as they are screening for possible antidepressant therapies.


Tests of Antidepressant Efficacy
1) Forced Swim Test: Take a mouse (or rat), put it in a bucket of water where it has to swim. Wait, and ponder how there is noting quite so cute as a mouse swimming. Pretty soon the mouse will give up when it realizes it can’t get out, and start floating in the water. The time it takes to give up is estimated as the time it takes for the mouse to despair. Researchers have found that antidepressants like Prozac or Celexa increase the amount of time that the mice spend swimming and trying to get out of the bucket. So you can screen for an antidepressant and whether or not it will work by giving to a mouse in the forced swim test.

2) Tail Suspension Test: This time, string the mouse up by its tail. Wait for the mouse to stop wriggling around. It will give up pretty soon (though it’s best to use fat mice for this, skinny mice can climb their own tails and so are no good). This is just like the Forced Swim Test, and works pretty much the same way.
3) Sucrose Drinking: This test measures anhedonia, or whether or not animals are sensitive to pleasurable stimuli. And the mice love it! Take a mouse and give it a choice between water, and a sweet solution like sugar water, Kool Aid, or Tang (you would be AMAZED at the number of scientific studies that use things like Tang). Obviously the mice like the sweet stuff better, and how much MORE they like it (how much they drink) is a measure of anhedonia. So you can take mouse models of depression and see if they drink sugar water, then give them antidepressants and see if they drink more.
So how do you get depressed mice in the first place? First of all, some strains of mice or rats are more sensitive to depression than others, or more sensitive to antidepressants. So once you’ve picked your strain, you then have several options:
1) Parental Separation: taking a mouse (or a rat, or a monkey) away from its mother for a while (either raising it on a cloth surrogate or for rodents, just taking it away for an hour a day), can produce mice that are more sensitive to stress and give up swimming more easily.
2) Social Defeat: Put a mouse in a cage with a bigger mouse. The small mouse (or the non-resident mouse, mice are very territorial) will lose the fight. If you have a mouse lose a fight several times, they start to get depressed. In monkeys, if you put monkeys in groups, there is a hierarchy that is soon established. The monkeys at the bottom of the heap show signs of depression and are sensitive to antidepressants.
3) Chronic Stress: Put a mouse in a cage, and put that cage in a rat cage. In the wild, rats eat mice, so even just the smell of them can stress out a mouse. A lot of times you can just put a piece of bedding that’s been in a rat cage in a mouse case, or something that smells like cats, etc. If you keep it in there for several days, you get a chronic model of stress, and the mice that have been stressed have very different reactions in the depression tests.

Those are some of the most well known models of depression and behavioral tests for antidepressant efficacy. Of course there are also genetic models (with different versions of the serotonin transporter), and there are ways to look at animals with depressed phenotypes using PET imaging and MRI. Next up, the serotonin hypothesis of depression, and why it’s probably a lot more complicated than that.

18 Responses

  1. How do tests (1) and (2) adjust for the fact that a smarter mouse will realize the hopelessness of the situation more quickly?

  2. I’m such a wuss.
    I’m no PETA-fan, but I love animals. Reading this made me realize how glad I am that my field does not require me to make mice depressed or stressed out for the purposes of my research.
    I’d much rather fuck with people. (Within ethical guidelines, of course)😉

  3. Agreed. These studies make me wince too. I really don’t like the idea of causing a conscious being to suffer via my intention.
    Although I understand the value of such research, it’s no doubt undesirable and serves as a screaming reminder for us to find better ways to research these things.
    Why can we not just use the mouse to make sure the drugs are physiologically safe instead of psychologically harming them?
    Once the drugs are shown to be biologically safe, the clinical trials could go ahead. Perhaps this is avoided for financial reasons, but I think the cost is worth it to uphold morality.
    If there are indeed other, significant, unavoidable, scientific reasons though, please educate me!
    Great summary, Sci.
    Thanks,
    Flym.

  4. JLK and Flymises: don’t worry too much about the Forced Swim Test and Tail Suspension. The tests are only run for a maximum of six minutes at a time (for the FST) and two minutes at a time (for the TST). And animals are very rarely put in the test more than twice. In social defeat, the animals are watched the entire time so they can’t be really hurt.
    It’s true that we could just test the drugs for safety, but there are VERY few humans I know willing to take a drug that sure, might be safe, but with no knowledge of what it might DO. Not only that, how would we know what dose was even effective? All we would know would be the dose that wouldn’t kill you. We’d have no idea how effective it was, or sometimes, even what disease it would be effective for, unless we can test it in animal models. I can’t really think of an effective way to test for that efficacy in humans unless we already had an idea of what we were looking for. Until these tests were invented, most of the psychiatric drugs used today were discovered purely by chance, and with the amount that people suffer from psychiatric disease, we need to be inventing and testing new drugs at a much faster rate to find the most effective treatment.
    Using these tests, we can take chemical structures that we know are similar, test them for safety AND the most effective dose, and use that to calculate similar human doses. I admit that it’s not the best system, but it’s what we have. And animal protocols are designed to ensure that animals never suffer more than they absolutely have to for a given test.
    I hope that helps. Please let me know if you have any more questions!
    llewelly: “smarter” mice would be compensated for because you have to run the test in more than one animal, usually about 6-12 per group. So any effects of an animal being smarter or not would probably come out in the wash.

  5. It wasn’t the tests that made me wince, personally. It was the idea of taking a baby mouse away from its mother in order to make it depressed and putting a mouse in a cage with a bigger mouse that beats it up several times also to make it depressed.
    I know, I’m a sap. I cried when my gerbil died and buried it in a celestial seasonings sleepytime tea box in the backyard. I can’t help it.

  6. Sci- I still think that if you’re studying knockout mice, and a particular mutation we *think* will make the mice more depressed also makes them cleverer, you’d have problems with this introducing artefacts in the FST or TST.
    I didn’t know about the Tang taste test, but it makes me a little happier to know that there is a not-mean-sounding depression test for mice.
    At the same time, as a depressed grad student with an unhealthy creme brule habit, I can’t help but think that a drug that increases resistance to despair in helpless situations might be a more important thing to identify than something that would somehow make creme brule even tastier. Then again, maybe the existance of grad students argues that intelligence has nothing to do with avoidance of hopeless situations.

  7. […] this study was done in mice.  And Sci has blogged a little before on how to look at depression in mice.  What she hasn’t really blogged about much is what MAKES for a stressed or […]

  8. […] take it, it’ll just huddle there and look unhappy.  This is a way of generating a model of depression in mice.  Sci doesn’t generally like it, because there are other ways to generate depressive like […]

  9. […] sucrose and took longer to approach a tasty bit of food. They even didn’t swim as much in the forced swim test. The depressive like behaviors correlated with a decrease in […]

  10. […] this study was done in mice. And Sci has blogged a little before on how to look at depression in mice. What she hasn’t really blogged about much is what MAKES for a stressed or depressed […]

  11. […] won’t take it, it’ll just huddle there and look unhappy. This is a way of generating a model of depression in mice. Sci doesn’t generally like it, because there are other ways to generate depressive like […]

  12. […] sucrose and took longer to approach a tasty bit of food. They even didn’t swim as much in the forced swim test. The depressive like behaviors correlated with a decrease in […]

  13. […] Why is this important? Recently scientists have found that antidepressant drugs (like Prozac and Zoloft) increase the birth of new neurons in the hippocampus, and that these increases in neurogenesis are associated with behaviors that we have come to associate with antidepressant properties. Basically if an antidepressant is working in a rat, you will see neurogenesis and you will also see increases in sucrose consumption, increases in tasty food consumption, and increases in struggle in tests like the forced swim test or tail suspension test. […]

  14. […] Depression Post 3: Studying Depression in the Lab | Are you … […]

  15. […] Depression Post 3: Studying Depression in the Lab | Are you … […]

  16. […] Depression Post 3: Studying Depression in the Lab | Are you … […]

  17. […] Depression Post 3: Studying Depression in the Lab | Are you … […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: