It’s high times in the land of neuroscience. In the last two weeks, we’ve had three high-profile papers from the wizards of optogenetics, all related to depression, all linked to dopamine.Now, by this point, everyone in the blogosphere has covered the gee whiz aspect of this story, so I thought I’d delve into a couple of problematic issues these studies raise for those of us who think about neuroscience for a living.
First of all, lest I be mistaken: these are really excellent studies. As far as I’m concerned, the hype over optogenetics is pretty well-deserved. Having recently slogged through just a tiny corner of the vast literature on electrical microstimulation, it’s a welcome relief to read a paper in which the authors know exactly what cell type they’re stimulating, as well as where it’s coming from and where it’s going to. This is rapidly, rapidly becoming the gold standard for systems neuroscience, and it’s exciting to see real progress being made on questions where we’ve often had to make do with correlational studies. What it feels like, to those of us outside the inner circle, is that while we plebes make do with circumstantial evidence; the optogenetics crowd proves things.
Of course, that’s not entirely fair. Sometimes, even the kids with the fanciest toys come up hard against biological reality.
And that’s what this post is about: the really tough questions that stare back at us from the depths of the brain.
Two neuromodulators and the prefrontal cortex walk into a bar…
So here’s the breakdown: three experiments; all in rodents. One in mice (Chaudhury et al.), one in rats (Warden et al.), one with both (Tye et al.). The question is what dopamine and serotonin have to do with depression.
Now dopamine, if you’ll recall, is the brain’s pleasure chemical. Except that it’s not, as reviewed by Bromberg-Martin et al. (2010), Salamone and Correa (2012), and Berridge (1996) (from way back). In fact, we don’t quite know what dopamine does, except that it seems to be important in motivation and reinforcement. (I’m deliberately avoiding the term “reward,” since, as people point out, the more we use it, the slipperier and less useful its definition gets.)
Then there’s serotonin. Which, when it’s not fulfilling its duties as another misunderstood feel-good drug, is involved in such unglamorous processes as regulating eating and certain social processes (certain being my weasel-word for “I’m not an expert on this”).
So dopamine. And serotonin. And sure, you say, it makes sense that the brain’s motivation chemical would be involved in depression. But that’s where things get tricky. Because the antidepressant drugs we have are all based on modulating serotonin and/or norepinephrine (=noradrenaline), not dopamine. Throw in the fact that acetylcholine, one of the other big brain neuromodulators, seems to play a key role in alertness and sleep/wake cycles, and you might start thinking that it’s a great big neuromodulator party up in there, where everybody’s out of whack and wrecking the place up. Which is probably correct, or at least much closer to the truth. The brain is a complicated place, friends.
Okay, you say. Fine, you say. But at least we know that depression results from these chemical imbalances. But do we? Turns out that when you look at real, depressed humans in the MRI or PET machine, there’s a whole network of cortical areas implicated as well. So take that, brainstem.
Now bear in mind, it’s not as if we know nothing about the roles of all these players. It’s just that a lot of them seem to be onstage at once, and the drama in question is a gothic Faulkner novel of full of morbid and incestuous relationships. Which means, in practice, that symptoms only give us very indirect information about root causes, and even very effective treatments may only work because of indirect effects.
Now back to the good part
First off, for those of you who haven’t read the originals or any of the other coverage, here’s the rundown: In the Tye paper, the focus is on dopamine neurons in the ventral tegmental area (VTA), particularly on those that project to the nucleus accumbens (NAc; a key node in the reinforcement and motivation network and a big target for dopamine). In the Chaudhury paper, we keep these two actors but add a third, the medial prefrontal cortex (mPFC), another target of dopamine release. In the Warden paper, mPFC sticks around, but the focus is on one of its projections back to the brainstem, to the dorsal raphe nucleus (DRN), the site of serotonin manufacture.
And what do they find? Tye: stimulation of dopamine neurons reduces depressive signs; inhibition of dopamine neurons increases them. Warden: excitation of mPFC projections to serotonin neurons produced antidepressive behavior; mPFC excitation of the lateral habenula, another brainstem nucleus, had the opposite effect. Chaudhury: phasic (burst-like) but not tonic (regular) firing of dopamine neurons increases susceptibility to depression; inhibition induced resilience. Susceptibility seems to be related to dopamine projections to the NAc, while resilience seems to depend on the connections to mPFC.
So what do we make of all this? Well, more dopamine and serotonin release (in the right pattern) seems to be antidepressive. Nucleus accumbens and prefrontal cortex connections seem to be key. And in some measure, we knew these things indirectly, but it’s tremendous to see them demonstrated with this much experimental control.
But science is complicated, y’all
Now, for the stuff that doesn’t make it into the science news.
On the face of it, the Chaudhury and Tye results seem to contradict each other. In the Tye paper, firing of dopamine cells projecting to NAc reduces depressive symptoms. But in the Chaudhury paper, inhibiting this pathway makes mice resilient. Now, the Chaudhury paper contrasts phasic/bursting stimulation with tonic/regular stimulation, finding that it’s the former that disposes animals toward depression. But the Tye paper uses stimulation parameters very similar to Chaudhury’s phasic stimulation.
So why did Tye find antidepressant effects and Chaudhury increased susceptibility to depression when stimulating the same pathway in (nearly) the same way? Several pieces reporting on the findings point out that the paradigms are different (acute stress for the Tye vs chronic stress for the Chaudhury; more on this below), but I think that more honest answer is that this really is a puzzle. My assumption is that both groups were very careful, but it’s entirely possible that results can vary from one lab to another due to changes in animal housing, experimental procedure, and five hundred other small things that no one would think to ask about.
In the case of the Warden paper, there’s less of a puzzle than a bunch of extra questions: was the result due primarily to serotonin, as the study suggests, or was the antidepressant effect mediated through areas interacting with DRN? After all, it’s well known that because dopamine and serotonin are related chemically, the systems responsible for admitting them can interact with one another. For instance, in Parkinson’s disease, when most of the dopamine-producing cells have been destroyed, serotonin-releasing cells can uptake and subsequently release dopamine manufactured from precursor chemicals in medications used to treat the disease. Personally, I have no intuition about whether to consider dopamine or serotonin as more fundamental in this case, but the fact that the systems interact strongly raises questions about what’s causing the behavioral effect when you stimulate one or the other.
Along the same lines: it’s known from very careful work (I know the papers from Hikosaka lab, but they cite others) that the lateral habenula directly inhibits VTA. This may explain why stimulating LHB in the Warden paper produces a depressive effect, but it’s not open-and-shut.
Times is tryin’ on rats and mice. Perils in animal models of depression.