About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« The Complicated Causes of Cancer | Main | Pfizer / Bayer? »

September 9, 2008

Antipsychotics: Do They Work For A Completely Different Reason?

Email This Entry

Posted by Derek

As I’ve noted here, and many others have elsewhere, we have very little idea how many important central nervous system drugs actually work. Antidepressants, antipsychotics, antiseizure medications for epilepsy – the real workings of these drugs are quite obscure. The standard explanation for this state of things is that the human brain is extremely complicated and difficult to study, and that’s absolutely right.

But there’s an interesting paper on antipsychotics that’s just come out from a group at Duke, suggesting that there’s an important common mechanism that has been missed up until now. One thing that everyone can agree on is that dopamine receptors are important in this area. Which ones, and how they should be affected (agonist, antagonist, inverse partial what-have-you) – now that’s a subject for argument, but I don’t think you’ll find anyone who says that the dopaminergic system isn’t a big factor. Helping to keep the argument going is the fact that the existing drugs have a rather wide spectrum of activity against the main dopamine receptors.

But for some years now, the D2 subtype has been considered first among equals in this area. Binding affinity to D2 correlates as well as anything does to clinical efficacy, but when you look closer, the various drugs have different profiles as inverse agonists and antagonists of the receptor. What this latest study shows, though, is that a completely different signaling pathway – other than the classic GPCR signaling one – might well be involved. A protein called beta-arrestin has long been known to be important in receptor trafficking – movement of the receptor protein to and from the cell surface. A few years ago, it was shown that beta-arrestin isn’t just some sort of cellular tugboat in these systems, but can participate in another signaling pathway entirely.

Dopamine receptors were already complicated when I worked on them, but they’ve gotten a lot hairier since then. The beta-arrestin work makes things even trickier: who would have thought that these GPCRs, with all of their well-established and subtle signaling modes, also participated in a totally different signaling network at the same time? It’s like finding out that all your hammers can also drive screws, using some gizmo hidden in their handles that you didn’t even know was there.

When this latest team looked at the various clinical antipsychotics, what they found was that no matter what their profile in the traditional D2 signaling assays, they all are very good at disrupting the D2/beta-arrestin pathway. Since some of the downstream targets in that pathway (a protein called Akt and a kinase, GSK-3) have already been associated with schizophrenia, this may well be a big factor behind antipsychotic efficacy, and one that no one in the drug discovery business has paid much attention to. As soon as someone gets this formatted for a high-throughput assay, though, that will change – and it could lead to entirely new compound classes in this area.

Of course, there’s still a lot that we don’t know. What, for example, does beta-arrestin signaling actually do in schizophrenia? Akt and GSK-3 are powerful signaling players, involved in all sorts of pathways. Untangling their roles, or the roles of other yet-unknown beta-arrestin driven processes, will keep the biologists busy for a good long while. And the existing antipsychotics hit quite a few other receptors as well – what’s the role of the beta-arrestin system in those interactions? The brain will keep us busy for a good long while, and so will the signaling receptors.

Comments (6) + TrackBacks (0) | Category: Biological News | The Central Nervous System


1. Retread on September 9, 2008 8:17 AM writes...

Well, if beta arrestin is important in antipsychotic effects, studying just what it does is going to be incredibly complex. For starters look at

[ Proc. Natl. Acad. Sci. vol. 104 pp. 12011 - 12016 '07 ] New functions for beta-arrestins have been found in addition to turning off G protein coupled receptors (GPCRs). These include (1) clathrin mediated endocytosis of receptors (2) signal transducers for MAP kinases, AKT and PI3K. Both the endocytic and signaling roles of beta-arrestins rely on their ability to serve as adaptors and scaffolds. Some 24 binding partners of the arrestins have been found.

The present work greatly expands the number by finding proteins interacting with beta-arrestin using LC tandem mass spectroscopy.

71 proteins interacted with beta arrestin1, 164 interacted with beta-arrestin2 and 102 interacted with both. Some proteins bound only after agonist stimulation (of an associated GPCR) while others dissociated. Interestingly S-arrestin (visual arrestin) and X-arrestin (cone arrestin) were also found in heteromeric complex with the beta-arrestins. Some of the interacting proteins can also be found in the nucleus.

If drug effects were simple, we'd have figured them out long ago. Making life even more difficult (and more interesting) is the fact that we are far from knowing all the players in the game, and even when we do, their interactions with other players are likely to be crucial. Consider the dynamical effects of introducing a very obscure governor into the presidential race. Does anyone care to predict how it will play out?

Permalink to Comment

2. Wavefunction on September 9, 2008 8:46 AM writes...

Would be really interesting to see how this pathway might be related, if at all, to extrapyramidal symptoms. But first let's find out how it's related to the usual effects.

Permalink to Comment

3. Anonymous on September 9, 2008 7:54 PM writes...

I think this is a perfect example of what was discussed in 'PNAS:Read it or not'.
Personally I wouldn't read too much into this study.

It shows:
(1) D2 agonists activate B-arrestin
OK. This has been shown before.

(2) Antipsychotics that are known to bind to the D2 receptor can antagonize D2-mediated arrestin activation. However these molecules don't show any agonist activity towards arrestin.

OK. This makes sense. This is the nature of competitive binding. It also makes sense that they don't activate arrestin as these molecules are typically inverse agonists.

(3) The antipsychotics in (2) are either partial agonists or inverse agonists for the G-protein pathway to cyclase.

OK. This data has been published by 100's of researchers in dozens of labs around the world.

(4) Bringing it all together in a conclusion:
Because all the antipsychotics could inhibit Arrestin activation yet had different profiles in the cyclase assay- this must mean that the target/pathway selective effect of these molecules is mediated by B-arrestin. (the authors didn't specifically state this- but it was certainly very strongly implied - and I think the naive reader would draw this conclusion).

Well sorry to break it to you conclusion #4 could be wrong in so many ways.

Ignoring all the well trodden data about the multi-target actions of antipsychotics these studies neglect some long known caveats about cellular pharmacology- as well as some published data.

The crux of the paper is that apiprazole is a partial agonist at cAMP but is a neutral antagonist at arrestin. This molecule 'makes or breaks' the paper as far as I am concerned as it is the molecule that dissociates cannonical G-protein pathways from antipsychotic efficacy.

The main problem with this paper is overinterpreting the comparison of data from a quantative and highly sensitive assay (cAMP) with a less quantitive and less sensitive assay (BRET/imaging) in an overexpressed system.

The simple observation for this data are that apiprazole is a partial agonist for arrestin and cAMP; and that (1) there is more 'receptor reserve' in the cAMP pathway compared to the arrestin pathway (especially as the cell lines are overexpressing arrestin ) or (2) the arrestin assay is not sensitive enough to detect the weak partial agonism of apiprazole (the 'true' efficacy of apiprazole is reported to be about 5% of dopamine).

Furthermore, a great paper was already published this year in Neuropharmacology (2008 Jun;54(8):1215-22). This paper shows that apiprazole is a partial agonist at the arrestin pathway. Thus essentially demonstrating that G-protein activation and arrestin activation are directly correlated (when assayed with the requisite sensitivity).

Sorry for the lenght of the post- but its a complicated topic.

Permalink to Comment

4. Derek Lowe on September 10, 2008 7:38 AM writes...

Good points indeed. I find the receptor reserve argument especially worth pursuing, since I assume that nothing is really known about the tone of the beta-arrestin signaling yet.

I'll take a look at the paper you cite, and try to blog about the topic again.

Permalink to Comment

5. Retread on September 10, 2008 8:10 AM writes...


Shades or reading the medical literature (as I did for 30+ years) where my attitude to every paper soon became -- how are they lying to me? Was the design of the work adequate to support the conclusion (usually were there enough patients)? Was the conclusion supported by the data (almost never in studies using historical control, or non placebo controlled double blinded studies)?

I had always thought that the august pages of PNAS would be better, and was rather shocked by the recent post in this blog about PNAS. Most of the commenters agreed that the papers were substandard.

I was a classmate and friend of Nick Cozzarelli, the late editor of PNAS for 10 years. His work on the topoisomerases has certainly held up and was of excellent quality.

Along the lines of unexpected mechanisms of drug action, you might be interested in the 15 May post of Chemiotics on the Skeptical Chymist blog -- Do you know where your drug is (and what it is doing)? in which the totally surprising effect of tricyclic antidepressants in lowering intracellular ceramide levels (and possibly treating cystic fibrosis) is described.

Permalink to Comment

6. shockwave on November 15, 2009 4:19 PM writes...

I never had symptoms till starting antipsychotic. Now I have migraines nausea vomiting, I never I repeat never have appetite. I also developed muscle problems like I have had a cramp in my elbow for almost a year. The doctors did a biopsy and genetic testing they found the medication to have caused a cellular dysfunction in my muscles and most likely throughout the rest of my body and organs that is responsible for my symptoms and their is no cure. The genetic testing ruled out all known muscle diseases caused by genetics or genetic mutations, or inheritance, the medication is probably the cause. Basically this medicine is given to people who don't have good health care it is billed as a cure. All patients are labeled for life with this diagnosis (schizophrenia) and there is no way to prove you do not have the illness. In that respect is not a very scientific diagnosis. In science for a theory to be accepted and it still doesn't mean it is 100 percent guaranteed to be true, there has to be the possibility to disprove it, this is not the case with schizophrenia.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

Messed-Up Clinical Studies: A First-Hand Report
Pharma and Ebola
Lilly Steps In for AstraZeneca's Secretase Inhibitor
Update on Alnylam (And the Direction of Things to Come)
There Must Have Been Multiple Chances to Catch This
Weirdly, Tramadol Is Not a Natural Product After All
Thiola, Retrophin, Martin Shkrell, Reddit, and More
The Most Unconscionable Drug Price Hike I Have Yet Seen