The Frontal Cortex

The Revival of Shock Therapy

A new book, Shock Therapy, has recently been published, which offers a contrarian take on the history of electroconvulsive therapy, or ECT. I haven’t read the book, but Barron Lerner reviews it in Slate:

The authors believe that electroconvulsive therapy is incredibly effective. And yet for decades, a severely depressed patient–even one on the brink of suicide–might not have been offered the therapy, or if her doctors had proposed it, she or her family might well have declined it. In explaining why, the authors demonstrate that though we may assume medical treatments get adopted or rejected based on objective statistics, in fact data are often misinterpreted and manipulated by outside influences that end up overpowering them.

Scientists are finally beginning to understand the precise mechanisms that make ECT such an effective treatment for severe psychiatric disorders. For the most part, it seems that ECT relies on the same cellular mechanisms as typical anti-depressants, like Prozac. All of these treatments work by increasing the production of a class of proteins known as trophic factors. (The serotonin story is vastly oversimplified, if not flat out wrong.) Trophic factors make neurons grow. What water and sun do for trees, trophic factors do for brain cells.

What does an up-regulation of trophic factors have to do with mood disorders? The simple answer is that nobody quite knows. But there’s some suggestive if preliminary evidence that trophic factors ease depression by increasing neurogenesis, or the birth of new cells, in the hippocampus. Here’s how I described the research in a 2006 article:

In December 2000, Ronald Duman’s lab published a paper in the Journal of Neuroscience demonstrating that antidepressants increased neurogenesis in the adult rat brain. In fact, the two most effective treatments they looked at–electroconvulsive therapy and fluoxetine, the chemical name for Prozac–increased neurogenesis in the hippocampus by 75% and 50%, respectively. Subsequent studies did this by increasing the exact same molecules, especially trophic factors, that are suppressed by stress.

Duman was surprised by his own data. Fluoxetine, after all, had been invented by accident. (It was originally studied as an antihistamine.) “The idea that Prozac triggers all these different trophic factors that ultimately lead to increased neurogenesis is just totally serendipitous,” Duman says. “Pure luck.”

But demonstrating a connection between antidepressants and increased neurogenesis was the easy part. It is much more difficult to prove that increased neurogenesis causes the relief provided by antidepressants, and is not just another of the drugs many side-effects. To answer this question, Duman partnered with the lab of René Hen at Columbia.

The research team, led by post-doc Luca Santarelli, effectively erased neurogenesis with low doses of radiation. All other cellular processes remained intact. If the relief from depression was due to changes in serotonin, then halting neurogenesis with radiation should have had no effect.

But it did. Hen and Duman’s data was unambiguous. If there is no increase in neurogenesis, then antidepressants don’t work in rodents. They stay “depressed.”

Duman and Hen’s work was greeted, as expected, by a howl of criticism. Mice aren’t people. The experiment was flawed. The radiation wasn’t specific enough. Robert Sapolsky, whose work on stress paved the way for much of Duman’s own research, is one of the most incisive skeptics. He argues that neurogenesis researchers have no plausible model for how decreased neurogenesis might cause the symptoms of depression. Why would having a handful fewer new cells in the hippocampus have such an effect? “The more expertise someone has about the hippocampus,” Sapolsky wrote in a review in Biological Psychiatry, “the less plausible they find this novel role.”

Since then, Duman has gone on to show that one trophic factor in particular, VEGF, seems to play a large role in increasing neurogenesis. (Although BDNF also seems to be important.)

Update: Vaughan has a typically superb summary of ECT’s history over at Mind Hacks.

In a nutshell, it [ECT] seems to be the most effective treatment for severe depression, seems to impair memory, is disliked and stigmatised, and is difficult to research. Most notably, as a patient, your mileage may vary. Some people have no benefit, some have huge improvement; some have no side-effects, some have ongoing difficulties. Most have some of each.

Comments

  1. #1 Rachael
    January 4, 2008

    This is, randomly enough, exactly my line of research.

    “by increasing the production of a class of proteins known as trophic factors.”

    The neurotrophin story can be quickly complicated. The (four) behavioral studies that have looked at exogenous neurotrophic factor application have conflicting results, and the genetic models are all over the place. A small cohort of researchers have published some rather convincing studies showing that neurotrophin delivery will desensitize neurons in a manner that is dose and time-course dependent. So, boosting neurotrophic factor alone might not guarantee increased plasticity, but this work is not readily cited in the molecular psychiatry field and it is unclear how often behaviorists are aware of this complication. In my current studies I am observing a convincing pro-depressive effect following particular kinds of exogenous neurotrophin delivery to the hippocampus. I couldn’t speak for Ron Duman’s viewpoint, but we have talked about the my experimental results and he seemed to find them consistent. Perhaps (and this is my own speculation here) increased neurotrophin production is merely a side effect that occurs when cells are in a neurogenesis-ready state. It could be more effective to achieve this state in a manner other than artificially boosting neurotrophin levels, which would be back to square one (chancing upon effective drugs rather than discovering them )

    Anyway, I don’t disagree with what you wrote at all. Just entering my own soapbox about what the role of neurotrophic signaling could be. From my perspective, boosting neurotrophic factor production might very well be a dead end from a therapeutic perspective, though neurogenesis still seems to be key.

    Here’s a paper by Carl Diesseroth’s group in Stanford that searches for better metrics to explain pro and anti depressive like behaviors. They observed changes in regional hippocampal activity that could account for behavior regardless of whether the actual behavior was pro- or anti- depressive. This is very, very cool research.

    Sorry this was so long… I can’t help the soapbox!

  2. #2 MattXIV
    January 4, 2008

    I think you’re overstating the importance of neurogenesis in depression and the relative certainty regarding the mechanism. As Duman himself says “[The impact of antidepressants on neurogenesis] is probably not the whole story. You can’t explain the entire action of antidepressants through neurotrophins and neurogenesis.” It fits some of the elements, like the lag between the introduction of the drug and the full response and the correlation between mood and the state of the hippocampus, but leaves others unexplained, such as the effects of SSRIs on “healthy” people and people with bipolar disorder, the fact that ECT doesn’t have a lag time like SSRIs, the effects of drugs like MDMA that acutely alter serotonin levels, the correlation between heightened monoamine oxidases and depression, and the correlation of differences in genes involved in serotonin production with both lower serotonin levels and depression. I’d read the evidence as suggesting that both monamines concentrations and neurogenesis are involved in mood regulation and that there may be a feedback relationship between them.

  3. #3 Jonah
    January 4, 2008

    Thanks for the good comments. Matt, I completely agree that the link between neurogenesis and depression remains very very tenuous. for starters, traumatic brain injuries also increase neurogenesis, but i dont think we are going to start recommending concussions for serious depression. so there are lots of variables to disentangle. nevertheless, it’s still interesting and will certainly be a big topic of future research into depression.

    And Rachael, thank you very much for your informed take. Certainly, the trophic factor story is very complicated, especially when one is trying to get beyond mere correlation. (I certainly wasn’t aware of just how contradictory the studies of exogenous trophic factor delivery were. Thanks for the tip. What do you think accounts for the contradictory results?) But here is a recent summary of a talk given by Duman, in which he reiterates the basic, and perhaps oversimplified, theory I laid out in the post:

    “Dr. Duman noted the substantial number of studies supporting the key observations that comprise the neurotrophic hypothesis�from requirements for the interaction of genes and environment to the pathways underlying neuroplasticity. He focused on neurotrophic growth factors implicated in stress and depression�particularly BDNF, which is a member of the nerve growth factor (NGF) family, along with NGF, neurotrophin-3 (NT-3), and NT-4/5. Other groupings include the VEGF family, which includes such members as VEGF-A and VEGF-B; the fibroblast growth factor (FGF) family, whose members include FGF-2 and FGF1-23; and the insulin-like growth factor (IGF) family, which includes insulin, IGF-1 and IGF-2.

    Dr. Duman noted the substantial number of studies supporting the key observations that comprise the neurotrophic hypothesis�from requirements for the interaction of genes and environment to the pathways underlying neuroplasticity. He focused on neurotrophic growth factors implicated in stress and depression�particularly BDNF, which is a member of the nerve growth factor (NGF) family, along with NGF, neurotrophin-3 (NT-3), and NT-4/5. Other groupings include the VEGF family, which includes such members as VEGF-A and VEGF-B; the fibroblast growth factor (FGF) family, whose members include FGF-2 and FGF1-23; and the insulin-like growth factor (IGF) family, which includes insulin, IGF-1 and IGF-2.

    He then proceeded to summarize how stress influences the expression of BDNF in the hippocampus. �You can see there are many different types of stressors, different lengths of stress, and � almost every case that�s been reported has shown that stress decreases the expression of BDNF� in hippocampus and, in some cases, the prefrontal cortex as well,� he said. �It�s one of the most consistent effects, which really hadn�t been reported before for the action of stress. In contrast to that, antidepressant treatment increases the expression of BDNF in hippocampus.�

    http://www.neuropsychiatryreviews.com/07jul/neurotrophic.html

  4. #4 Dave Briggs
    January 4, 2008

    Scientists are finally beginning to understand the precise mechanisms that make ECT such an effective treatment for severe psychiatric disorders. For the most part, it seems that ECT relies on the same cellular mechanisms as typical anti-depressants, like Prozac. All of these treatments work by increasing the production of a class of proteins known as trophic factors. (The serotonin story is vastly oversimplified, if not flat out wrong.) Trophic factors make neurons grow. What water and sun do for trees, trophic factors do for brain cells.

    I had heard about this lately and even though ECT seems kind of cruel and barbaric if it works it certainly warrants further studies. As you pointed out, if you have a severely depressed patient who is suicidal, to find an effective course of action can be a matter of life and death!
    Dave Briggs :~)

  5. #5 Anon regular reader
    January 4, 2008

    Dave Briggs – its not at all cruel and barbaric.

    I had six ECT treatments five years ago while having a major depressive episode.

    They put me to sleep (i.v. sedation), did the ECT, then woke me up. I felt somewhat confused afterward, and slept all day, but otherwise felt no ill effects.

  6. #6 Rachael
    January 4, 2008

    It was either going to be a short answer or a thorough one, and I decided to aim to be thorough… Note that from the perspective of the neurotrophin hypothesis, application of BDNF should upregulate TrkB, pTrk and pErk, and lead to downstream plasticity events to produce antidepressant-like behavior. Here are the four behavioral BDNF infusion studies that I am aware of:

    (1) *Siuciak…Lindsay (from a company, Regeneron). Chronic (minipump) raphe nuclei infusion at 12-24ug/day in Raphe Nuclei
    = antidepressant
    (2) Shirayama…Duman. Single hippocampal infusion 0.25-1ug
    =antidepressant
    (3) Hoshaw…Lucki. Single intracerebroventricular infusion 0.1-1ug
    = antidepressant
    (4) Eisch…Nestler. Chronic (minipump) Nucleus Accumbens 2.5 ug/day
    = prodepressant

    *Regeneron published in a different journal (in the same year, probably on the same data), strong downregulation of TrkB. Neither paper referenced the other.

    The first three studies are consistent with antidepressant effects. Nestler’s group suggested their study was prodepressive because it was in the NAc (which can sometimes be funky like that). My interpretation is different due to this paper: Xu,Michalson,Racine,Fahnestock Neuroscience 126(2004)521-531. They used a model of seizure kindling and found that continuous doses of BDNF in the hippocampus knocked down TrkB and pTrkB and decreased seizure occurrence, whereas bolus doses had the opposite effect. They also found that they could (behaviorally) overcome the knockdown by dosing at very high levels; presumably this was because TrkB was not completely eliminated, very high levels of BDNF could still trigger downstream cellular events. Following the citation trail from this particular paper leads to you to a number of studies describing arguably negative effects from BDNF (TrkB/Erk knockdown, neuronal necrosis and cell death, p75, etc). One study demonstrated that if you give a neuron a small dose of BDNF, wait a week, and dose again, the neuron will stop responding to BDNF.

    My (not otherwise published!) research observed knockdown of TrkB and Erk following a slightly altered chronic infusion model (I use a polymer implant not a minipump) and a strong prodepressant effect in forced swim.

    When you start piecing together all of these papers, the dose response curve is almost too complicated to interpret, but it is consistent with the idea that exogenous neurotrophin application can be detrimental to plasticity related pathways with differing biochemical and behavioral effects. Although these reports are scattered in different fields, it seems that exogenous BDNF will sometimes trigger positive events and sometimes negative events. And it also seems quite likely to assume that how BDNF functions depends not only on dosing level and timing, but also on individual animal or human experience. This is what leads me to the conclusion that fiddling with BDNF might not be the best approach.

    Regarding BDNF, here is a really excellent review: “Is it time to reassess the BDNF hypothesis of depression?” by JO Groves in Molecular Psychiatry, 2007.

    Here is an “okay” review for genetic models: Urani, et al, “Modeling depression with transgenic mice: The neurotrophin hypothesis revisited ” in Clinical Neuroscience Research 2003.

    With regard to Dr. Duman – I should probably make clear that I’ve only had a few conversations with him about this data, so although he believed the experimental results, I doubt that my meager (engineering) grad student results would influence his own work. I think in the end it boils down to how one interprets the neurotrophin theory. I disagree with the second part of this statement, a lack of neurotrophic signalling factors results in the decreased plasticity/neurogenesis observed in depressed patients, and returning neurotrophin levels to normal will restore a healthy phenotype, but I am inclined to agree with this statement dysfunctional regulation of neurotophin related pathways contributes to the etiology of psychiatric disorders. Unfortunately, the second statement is less helpful in developing treatments.

  7. #7 Rachael
    January 4, 2008

    That’s weird that a link I put in there doesn’t work, but it doesn’t matter because it just says the same thing I typed here (ref to an SFN abstract)… okay, I’ve talked enough for a week! Thanks for posting on this very interesting topic!

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.