I admit I got a little excited when I saw a Twitt RT’d earlier today from Noah Gray.

Soon to be in press: Strong evidence supporting the neurogenesis-depression hypothesis, from @jsnsndr: http://j.mp/qn6fyD

I’m sort of vaguely aware and following a literature that is trying to establish a causal link between depression and alterations in hippocampal neurogenesis, proliferation and general plasticity based on creating new, functioning cells.

This is generally a non-human literature, typically in mouse models and….highly correlative.

With respect to this latter, the state of the art for a long time has been to treat markers of neurogenesis (there are many stages and concepts here which I’ll glomp under one heading. Follow the link in the tweet to the Functional Neurogenesis blog for all your background reading) as dependent variables to be reported. Not manipulated. This is, perhaps obviously, the case for any post-mortem human brain analysis but also for many animal models to date. In essence, you do some thing to the animal and then look at the markers afterward. Then you report whether neurogenesis is up, down or unchanged. So far so good. But this approach doesn’t quite get at the question of causality which is important if we think that altering the effects on neurogenesis (say by a drug or behavioral therapy*) would have a beneficial effect on the affective disorder itself. It could, after all, be a result of depression with no causal role.

There is also a question of whether a given animal model is a valid representation of a human affective disorder. Here we can think about issues of predictability- does it matter, for example, what the mouse model looks like if the ability to predict what anti-depression, anti-anxiety or anti-mania drugs will work (ultimately) in the clinic is high? Of course not…if your goal is drug development.

If you goal is understanding the neurobiological underpinnings of the human disease, well, you want to be careful.

My take on the current state of the nonhuman models of depression is that we are not yet at the point where we have high confidence in calling any model “depression”. They are models, they have various points of high correspondence to human illness but they also have limitations. They are, in cases, highly predictive of drugs which will ultimately prove useful in the clinical setting.

So I confess that when I see a scientist (Noah Gray is, after all, a scientist by training even if he currently inhabits an Editor job) tweet “strong evidence”, well, I’m looking for some coolio stuff.

Following the link to the post on Functional Neurogenesis, I found the post title to include the word “confirmed”! yay, let’s read!!!!

So I’m excited to say that we will soon be publishing what (I think) is the best evidence that impaired adult neurogenesis actually causes depressive symptoms (in mice).

“in mice”. Fuck.

Okay, so let’s scratch the “strong evidence” and the “neurogenesis-depression hypothesis confirmed stuff” for now.

Pretty cool paper, by the sounds of it. Again, from my distinctly nonspecialist position, this is the next step. Manipulations of the neurogenesis processes as independent variables to provide stronger evidence that there is a causal relationship. Between these processes and a behavioral or physiological phenotype. I can’t really say where this all fits into the “first demonstration” or “best demonstration” or “critical demonstration” picture so as to give you a valance for exactly how cool it is. But I do recognize that these approaches are the next place the field needs to go to better establish that the neurogenesis-depression hypothesis could be “confirmed”.

Me, until the state of animal models are better and more convincingly established, I want to see data in human subjects before I am willing to concede either “confirmation” or “strong evidence”.

ps. Do take a read over the Functional Neurogenesis blog. It is really quite good and this area of neuroscience is fun. Many of you may still labor under the old belief that the adult brain doesn’t grow new neurons and can’t repair itself. It’s no liver, but the brain does have some capacity to generate new neurons.