Depressed mice, gene therapy, and p11

Reader David sent me this paper the other day, and asked if I could blog about it. I said ok, maybe, and then I read...

..."Gene therapy"...


Sounds very cool, doesn't it? Sounds like the FUTURE! Where's my JETPACK!!!?!?!

But of course "gene therapy" is kind of a buzzword. A lot of people throw it around, but it seems like a lot of people don't know what it really MEANS, and what it can be used for.

But it turns out, it can be used for quite a lot! And it may not be quite so far in the future. After all, they're marketing jetpacks. Alexander et al. "Reversal of Depressed Behaviors in Mice by p11 Gene Therapy in the Nucleus Accumbens" Science Translational Medicine, 2010.

So let's start with gene therapy and what it is, and then we'll go into why they used it in this particular paper. Gene therapy is based on the idea of inserting a gene into someone's genome, either in the whole body or in specific parts, to change the gene expression of that cell or group of cells, and to use this technology to treat disease. In this case, what we're talking about is viral-mediated gene expression. This is where we use a virus (for our own nefarious purposes mwah-ha-ha-ha!!), take out the nasty bits of the viral DNA, and load the virus with the gene you want to express. You then inject the virus into your area of interest (normally this is really site specific), and the virus, using its own virusy ways, will insert your gene of interest into your area of interest. The gene will get incorporated into the genome, and get expressed by your cells!

Like So.

This is a really useful technique that is now being widely used in a lot of fields, including neuroscience. You can insert genes to increase expression of a specific protein, OR use genes that DECREASE expression of that protein. So it allows you to carefully and selectively look at a specific thing in a specific brain region (in neuroscience, anyway).

Now we've got gene therapy, what about p11?

p11 is a relatively new protein on the depression scene (I think the original paper was only as early as 2006), but the effects got everyone excited. p11 is a protein that controls the expression of two serotonin receptors. Serotonin is thought to be one of the major players in mood disorders like depression (while selective serotonin reuptake inhibitors may not work in many patients, this doesn't mean that serotonin isn't playing a role, it's the role that it's playing that's up for debate), and so the consequences of p11 became suddenly interesting. And it turned out that mice with a knockout of the p11 gene displayed what we call a depressive phenotype, where they show more immobility (which used to be called behavioral despair, but now we don't think that's accurate) in tests such as the forced swim test and the tail suspension test, tests which are sensitive to the effects of antidepressants.

The thing is though, that these p11 knockout mice have p11 knocked out EVERYWHERE. That's some big changes, and they are also changes that exist during the whole of the mouse's life, and so could affect a lot of other things. In order to really narrow down what p11 is doing and where it is most important in depression, the authors of this paper wanted to take a normal mouse, and change p11, in very specific brain regions.

The two brain regions they picked were the nucleus accumbens (they abbreviate it NAcc, but I hate that, I prefer NAc, because to me NAcc means the NAc CORE, but whatever) and the anterior cingulate cortex. The nucleus accumbens is a brain area located here:


I blog about it a lot because it's thought to be involved in the rewarding and reinforcing properties of drugs like cocaine. But many people have thought that it could ALSO be involved in some of the issues associated with depression, specifically things like anhedonia, which is a lack of the ability to feel pleasure, and thus very closely related to things like reward.

The anterior cingulate cortex on the other hand, is located here:


They are interested in the anterior cingulate cortex because a HUMAN study found that depressed humans had lower levels of p11 in the anterior cingulate cortex.

So what did they find? Let's start with the NAc.


Pretty, huh?

To start with, they used a virus to insert a gene into the NAc which knocked DOWN p11 in normal mice, only in the NAc. All of the pretty glowy pictures there are of them proving that they can do that. What we are interested in is the three graphs at the bottom (the bar graphs), which all show immobility. These are immobility measures in the tail suspension test and the forced swim test. The first two (left and middle) show that mice with a knockdown of p11 only in the NAc show increases in immobility in these tests, which indicate a pro-depressive phenotype. The graph on the right shows treatment with imipramine, an antidepressant, and shows that the antidepressant still exerts some effects when p11 is knocked out. So it looks like, so far, that these mice are more "depressed", but that the still respond to antidepressants.

So then the question is, can ADDING p11 rescue a "depressed" mouse? To do this, they used p11 knockout mice, that had had p11 knocked out all their lives, and this time used the viral vector to INCREASE p11 in the NAc. When they increased p11, they got a reversal of the "depressive" effects seen in p11 knockout mice. These mice showed more swimming in the forced swim test, more struggling in the tail suspension test, and even showed increased sucrose preference (which is through to be a measure of anhedonia). You can see all that here:


The diagonal bars show the animals that had p11 increased. The sets of bars on the right are the p11 knockouts, and you can see they show longer immobility, until the p11 is increased in the diagonal bars.

But of course, all this stuff was in mice. How does this relate to HUMANS?

They looked at human post-mortem tissue (that's dead) from depressed and control patients, and they tested each NAc for p11.


Looks nice, yeah?! The depressed patients showed a DECREASE in p11 in the NAc (which is comparable in the mice to knocking down p11 in that area, as shown in the first set of figures).

They WERE going to look at the anterior cingulate, but it all got shoved to the supplemental data because there was no difference (and I had to hunt around for that little sentence. Stupid supplemental data!). But that's ok.

So this all lines up, and it all looks pretty nice. Sci likes this paper, it's neatly done, and they nicely knock something down specifically, increase it specifically, and look at behavior as a consequence. Good and thorough. But the question is: now what? p11 controls the expression of two serotonin receptors (the 5-HT1B and the 5-HT4). Which of these receptors is undergoing the changes from lack of p11 that is causing the depressive phenotype? Is it both? Are they expressed in the NAc? Or do changes in p11 changes serotonin receptor expression elsewhere? Other brain areas like the hippocampus have been implicated in depression, and also in the actions of antidepressants, does p11 play a role there as well?

And once we've got this worked out, what are we going to do about it? It's too much to ask to give a viral vector in the brain of every depressed person. In fact, we can't really do that yet at all, it's a very risky technology still (especially in brain). Are there drugs that increase p11? What about the specific receptors, the 5-HT1B and 5-HT4? 5-HT1B and 5-HT4 receptor activating drugs have been shown to be effective in animal antidepressant tests, but have never been developed. Should we develop those? Or go after p11? And of course, although there are studies showing p11 is lower in depressed patients, there are also studies out there showing that p11 isn't really changed in depression, after all.

Ah, science, always with the more questions! But you keep an eye on p11, this little protein looks like it might have some potential.

Alexander B, Warner-Schmidt J, Eriksson T, Tamminga C, Arango-Llievano M, Ghose S, Vernov M, Stavarche M, Musatov S, Flajolet M, Svenningsson P, Greengard P, & Kaplitt MG (2010). Reversal of Depressed Behaviors in Mice by p11 Gene Therapy in the Nucleus Accumbens. Science translational medicine, 2 (54) PMID: 20962330

8 responses so far

  • Another exemplary post. =)
    Neuroscience is quite far from my area of study, but I love reading these posts! You break down the basics well (for someone with a background in molecular biology anyway). Aaaaaand, it's cool to read/understand the whole gene therapy buzz, as it's being thrown around like mad these days.
    It's almost like being in a journal club here.

  • Anton says:

    Very nice article. I hope depression will soon be cured with genetic transformations, but I doubt it will happen in the near future. After all, humans are not mice, and research that is neccessary to avoid side effects can take years if not decades.

    • scicurious says:


      I think you may be a form of spam (, but you're making a useful comment with something I wanted to address so I'm going to go with it.

      The major issue here with using gene therapy in the brain (specifically) of humans is that we have to use viruses to get it in there. There are concerns about this for various reasons. For example, viruses tend to do better in the brains of rats and mice than they do in humans. There are potential long term side effects that we haven't been able to see in the rodent model. All sorts of things, really. We'll get there eventually (I hope), but yes, gene therapy for a brain condition is still a difficult proposition.

  • Tybo says:

    I have to ask if you know why exactly it is that viral vectors work better in mice than humans. I do know there's the general issue of the surface area of a target region to its volume, in that exposing all the cells in a target region to the gene of interest is easier in mice, but is there actually something in the base cell machinery that makes mice more likely to take up a target gene than humans?

  • Eric says:

    I would be curious about that as well Tybo. My son has albinism (specifically the OCA1 variety). Genetic tests have show us exactly where the issue is in his genome. Given that, it SEEMS like fixing the gene should be relatively easy using something similar to what they did in this study, but I know there are a lot of reasons why that isn't true. I would be curious as to why we don't have this tech yet and how far away we are from gene theraphy in humans (5 years? 10? 50?). Is it simply that the body is too good at fighting off viruses to allow for large scale changes on the order I'm talking about? With the brain you obviously have to deal with the whole blood/brain barrier, but what about the rest of the body?

    • scicurious says:

      Short answer: I have no idea why.

      Long answer: In the case of albinism specifically, how could we ensure that the virus gets into and successfully integrates into every basal cell in the skin (in order to get the correct pigmentation)? Most viruses don't infect every cell, just a subset, and even then it's a percentage of the subset. So for a lot of these, you'd have to be able to use the virus for a specific subset, and then be able to get by with part of that subset. So, say, diabetes might be a good candidate, you could infect, say, a specific type of pancreatic cell to produce insulin even if the beta cells which normally do it are lacking.

      That's not to say we couldn't get there eventually, but right now the tools are limited. As well, Tybo, in the human brain, the blood/brain barrier appears to be a lot more discriminatory in humans than in mice. But, they ARE working on it. See here: For example.

  • Alfred says:

    Plz can you give us the full text of this article, a link or something, because I couldn't find the full text unless I subscribe, and I don't have money.

  • r.a. says:

    hey there, great post. Thanks for helping me understand the charts. Your clear, down to earth prose is a pleasure to read. Very well done.

Leave a Reply