Pattern Recognition - Selective Serotonin Rewait what
March 23rd, 2014
09:10 am

[Link]

Previous Entry Share Next Entry
Selective Serotonin Rewait what
The SSRIs--selective serotonin reuptake inhibitors--are a class of drugs commonly used as antidepressants. They 'work' by keeping neurons from absorbing serotonin from the synaptic cleft back into the cell, increasing the amount of available serotonin. This is good because serotonin is made of tiny pixies who want to be free, so keeping it out of the cells makes them happy, and that makes the patient happy by the magickal principle of contagion.

Well, that's about as close to the truth as any other version, and closer than the popular 'chemical imbalance' non-explanation, which is just updated humoralism. You need more serotonin, or black bile, one of the two.

If you don't believe me, consider tianeptine.  Tianeptine is a selective serotonin reuptake enhancer--it encourages the neurons to suck more serotonin out of the synaptic cleft, reducing the amount of available serotonin.

Yep, it's an antidepressant. Apparently a pretty effective one.

What the hell is going on?  Well, either:

  • Nobody actually knows how the hell these things work, or...

  • Tweaking serotonin in either direction can treat depression.

Actually, the first option is pretty much true regardless. Do SSRIs actually work directly via serotonin, or by indirectly promoting neurogenesis, or what? Last I checked (which I admit was a few years ago) it seemed very much up in the air.  But the second option appeals to my interest in dynamical systems, and my own 'punching the TV set' hypothesis of psychopharmacology.  You youngsters may not remember such things, but when an old CRT TV set was on the blink, hitting/shaking it would often knock its components back into their sockets and get it to work again.  By analogy, some psychiatric disorders may amount to getting stuck in a pathological stable orbit of a dynamical system, and perturbing the parameters of the system may knock it back into a healthy orbit*.


*: Yes, this is super-waffly, things are vastly more complicated than that, and it's unlikely that this is a fruitful metaphor for treating most illnesses. On the other hand, look at bipolar disorder.  What is that but an eigenvalue wandering into the right half-plane**?

**: This is also a super-waffly metaphor but perhaps slightly less so***.

***: Note also that there's a separate medical application for this metaphor where it works quite well: cardiac arrhythmia. Perturbing the system to knock it back into a healthy orbit is precisely what a defibrillator does.

Tags:

(17 comments | Leave a comment)

Comments
 
[User Picture]
From:luagha
Date:March 23rd, 2014 06:48 pm (UTC)
(Link)

Also related to use of Prozac for obsessive-compulsive disorders.

As I have had it explained to me:

A woman has to wash her hands three times before she takes the stairs. She is unable to walk on stairs if she doesn't - she will collapse and would have to be dragged. Her brain is in a rut to speak - a local minima from which she can't climb out.

Take prozac for three months. Now all peaks and valleys are lessened.

Return to the stairs. With someone standing there counseling her and saying exciting things like, 'You don't need to wash your hands, you can do it, stop walking to the bathroom' she can, as a supreme effort of will, walk up the stairs. Where before it was impossible.

After successfully walking up and down the stairs with help three or four times, the compulsion is gone. The rut/local minima is no more.

Then the patient can stop taking prozac.
From:(Anonymous)
Date:March 23rd, 2014 11:55 pm (UTC)
(Link)
is this from a study somewhere or anecdotal experience or what?
[User Picture]
From:luagha
Date:March 24th, 2014 04:15 am (UTC)
(Link)
It is how it was described to me on a radio show; the process of prozac therapy for compulsions.
[User Picture]
From:mizkit
Date:March 23rd, 2014 07:32 pm (UTC)
(Link)
I really like the idea that essentially we're all CRT TVs.
[User Picture]
From:twoeleven
Date:March 23rd, 2014 09:21 pm (UTC)
(Link)
Nobody actually knows how the hell these things work, or...

It appears so:
Fluoxetine (Prozac) was the first major breakthrough for the treatment of depression since the introduction of tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs) nearly 30 years earlier. It was the first selective serotonin reuptake inhibitor (SSRI) approved by the United States Food and Drug Administration, offering superior efficacy and reduced side effects relative to TCAs and MAOIs. Though a debate remains regarding the exact mechanism by which the clinical efficacy of fluoxetine is manifested, the importance of fluoxetine and related SSRIs to the field is unquestionable. [...]
there are some classes of drugs for which not knowing the mode of action (MoA) is pretty typical; psychopharm is one of them. i'm too lazy to look up the paper, but i read an article within the past year that claimed that the SSRI activity of the class is a secondary MoA, and their main effect is due to another interaction with the CNS.

i don't think it's so bad we don't know quite what we're doing, since we do that a lot in medicine, but i'm not sanguine about giving a drug known to fiddle with brain development to kids. ISTM that's it's a huge uncontrolled experiment, and we'll figure out the results the hard way.
[User Picture]
From:luagha
Date:March 24th, 2014 01:07 am (UTC)
(Link)
We are already.

(This was a sly reference to the prevalence of school shootings et al; but actually if measured over time school disasters have been fairly steady since they were first really measured with some school bombings in the 1930s. It's a naturally statistically bursty phenomenon because it's exciting and rare and it bleeds and it leads, but measured in decades the numbers haven't changed. It's just the media coverage that has amplified the attention.)
From:lhn
Date:March 24th, 2014 02:51 am (UTC)
(Link)
Do you have a cite for that? Not that I disbelieve it-- in general, my default assumption of any "X is worse than ever before!" story is that X is either roughly steady, has gotten substantially better to the point that exceptions are major news, or at worst is cyclical. (Rebuttable, of course-- it's just the way to bet.) But in the case of school attacks I'd be interested in having something to point to.
[User Picture]
From:luagha
Date:March 24th, 2014 04:15 am (UTC)
(Link)
It's googleable.
[User Picture]
From:st_rev
Date:March 24th, 2014 02:00 am (UTC)
(Link)
Well, one thing to consider is that the SSRIs are the first "success story" for rational drug discovery: pick a target and a mechanism, and screen compounds by assay against it. Whether they work by serotonin reuptake inhibition or not, that's why they were investigated in the first place.

Which suggests in turn that rational drug discovery has been a trillion-dollar voodoo ritual.
[User Picture]
From:twoeleven
Date:March 24th, 2014 02:36 am (UTC)
(Link)
i'm not sure why you say that.

1) my objection is only their use in children; they're a great success otherwise.

2) even if their nominal MoA is secondary to their main effect -- i haven't seen anything which says that it's irrelevant, but obviously, i can't read the whole literature -- the method works as desired.¹

3) rational drug design has worked quite well for the later NSAIDs, anti-viral protease inhibitors and NNRTIs, and other recent drug classes. it also makes french-named company quite rich via their newer rationally-designed ag chemicals.

it's not a perfect method by any measure -- i can rattle off a bunch of limitations¹ -- but it's much better than the lamp-post logic of analoging around known-good compounds or making a bunch of stuff and hoping to get lucky. current methods usually use a mixture of everything we've got (screening random compounds, in silico design of molecules based on estimates of what the target is or how it works, combinatorial synthesis of lots of somewhat-likely candidate molecules, and rational design starting from target structure). if it was an easy problem, we'd have solved it already. :)

1 (yup, used twice): the most obvious limitation is that rational design often can't say what else a drug candidate will interact with. we can make educated guesses from similar molecules, but that's usually about it. so: tissue studies, animal studies, and phase 1 clinical studies.
[User Picture]
From:st_rev
Date:March 24th, 2014 02:48 am (UTC)
(Link)
Sorry, forgot who I was talking to. You're right, of course. I'm overstating my case recklessly.

Edited at 2014-03-24 02:49 am (UTC)
[User Picture]
From:twoeleven
Date:March 24th, 2014 03:15 am (UTC)
(Link)
Oh... heh. :) Since you defined rational drug design, I thought you were making the notional play to the lurkers, and so I was wondering if I was missing your real point.
[User Picture]
From:st_rev
Date:March 24th, 2014 03:21 am (UTC)
(Link)
I pay a lot more attention to psychopharmacology than other areas, which is a recipe for despair. But I also read In the Pipeline; I should know better.

Edited at 2014-03-24 03:21 am (UTC)
[User Picture]
From:st_rev
Date:March 24th, 2014 03:25 am (UTC)
(Link)
I slipped in a tiny bit of a troll about tianeptine; the linked article notes that current thinking is it actually works by a different mechanism. But that raises a lot of the same issues, and more importantly the comparison is just hilarious to me.
[User Picture]
From:twoeleven
Date:March 24th, 2014 03:36 am (UTC)
(Link)
I saw; I'm not sure what that says, beyond that we're not perfect. Psychopharm is a hard area to get right, since the blood-brain barrier keeps out a lot of things we'd like to try, and most of the reasonable experiments can only be run on humans.
From:(Anonymous)
Date:May 8th, 2014 01:14 pm (UTC)
(Link)
A couple suggestive data points in favor of the "punch it and it'll get better" theory:

all-nighters are sometimes recommended as a treatment for depression
one-time use of MDMA can end a depressive episode
electroshock therapy, when it works, *really* fucking works

Actually, I wonder what would happen if you tried tryptophan and 5-HTP in the evening, *then* stayed up all night (without the aid of stimulants)? and then had sex?

That's the closest thing I can think of to provoking a release of serotonin without any drugs you can't buy at a drugstore.
[User Picture]
From:st_rev
Date:May 9th, 2014 06:14 pm (UTC)
(Link)
Sleep deprivation definitely works as a short-duration antidepressant, but it's not sustainable, and the side effects are severe.

I've had more success with choline and dopamine precursors, generally.
Powered by LiveJournal.com