I blogged on this article at "In the Pipeline" (http://pipeline.corante.c...), and Prof. Perlstein asked if I would come over and add the comments that have accumulated there to the paper itself, where the readers here could see them. Here's what the readership of my site has had to say so far:

2. Kare on May 16, 2012 8:00 AM writes...

This result is also interesting given the unexplained influence of omega-3 fatty acids on cognition, behavior and mood. Presumably, membrane biophysics is playing a role there, as well.
Permalink to Comment

3. Innovorich on May 16, 2012 8:04 AM writes...

There were very similar results published several years ago for the effects of anesthetics acting via changes to membrane fluidity rather than (at least directly) through any particularly protein target.
Permalink to Comment

4. road on May 16, 2012 8:06 AM writes...

The other enantiomers of sertraline are apparently inactive, which suggests that the psychoactive effects of sertraline are not due to pysichochemical interactions with membranes. It's a pretty perfect control, IMHO.
Permalink to Comment

5. luysii on May 16, 2012 8:20 AM writes...

Ah, the antidepressants, we hardly know ye. For another example, this time with amitriptyline, a classic tricyclic antidepressant, and its possible use in cystic fibrosis, because it inhibits an enzyme involved in ceramide metabolism of all things see https://luysii.wordpress.....

It's post #13 in a series concerning how little we understand all the things our drugs are actually doing.
Permalink to Comment

6. PPedroso on May 16, 2012 8:21 AM writes...

I look forward to read Dr. Kell feedback on this.
Not from a negative we-were-right perspective but just to check how his team will incorporate this new data into their research since I see some direct implications with their line of thought.
Permalink to Comment

7. Curious Wavefunction on May 16, 2012 8:52 AM writes...

#4 road: Just a thought: Aren't phospholipids chiral? Could this result in the preferential interaction of only one sertraline stereoisomers?
Permalink to Comment

8. Anonymous on May 16, 2012 8:58 AM writes...

Did the mood of the yeast cells improve? Did they happen to make beer taste better afterwards?
Permalink to Comment

9. NoDrugsNoJobs on May 16, 2012 9:32 AM writes...

I think #4 makes a pretty good point. Correct me if I'm wrong but not only does the correct enantiomer have most or all of the in vivo effects but that in turn correlates with the binding affinity for the serotonin transporter. Having said that, clearly we don't know the downstream effects close to well at all. There is another antidepressant known as tianeptine which putatively promotes serotonin uptake - the opposite of an SSRI of sorts. What both tianeptine and SSRI's share is a tendency to grow the neurites in the hippocampus and prevent stress induced hippocampal shrinkage. Are there alternative mechanisms for the broader response - appears likely.
Permalink to Comment

10. Ethan Perlstein on May 16, 2012 10:37 AM writes...

Hi, I'm actually the corresponding author of the paper in question. In the spirit of Open Science, I encourage comments on my paper on the PLoS ONE website. As for this comment thread, if I may I'd like to chime in on the question of stereochemistry.

First, it's true that the enantiomer of sertraline has lower affinity (about 10-fold, but still submicromolar) for the serotonin transporter and is much less selective for the serotonin transporter, at least in synaptosomal assays. Please see the original paper on sertraline by the Pfizer scientists who developed it here: http://www.ncbi.nlm.nih.g.... (Unfortunately it's behind a paywall). However, I'm not aware of data showing that sertraline's diastereomers are without behavioral effects in mice. But even if the diastereomers were without acute behavioral effects in mice, it would be interesting to see if the well-known physiological changes associated with chronic AD treatment in mice, e.g., BDNF expression, are also observed for the diastereomers.

Second, even though the enantiomer of sertraline is less potent and less selective for the serotonin transporter, it would still be predicted to accumulate in brain cells because it's lipophilicity is unchanged. The $64,000 question is whether sertraline is an antidepressant because of dual activities at hSERT and on cellular membrane homeostasis. That question could be teased apart synthetically by creating a less lipophilic sertraline analog that still possessed high potency and selectivity at hSERT, but I'm not aware of that experiment having been done either. If anyone wants to do the med chem, I'd be happy to test them in yeast and outsource the behavioral assays in mice!
Permalink to Comment

11. ZINC on May 16, 2012 10:59 AM writes...

@10 Ethan, there's usually no need to have some make a fresh compound. Especially in the case of known drugs, many people have made the compounds already and they are available for purchase, usually at prices academics can live with. For instance, here are the top 50 matching purchasable compounds of sertraline from ZINC:

http://zinc.docking.org/r...

I'm sure with some clever searching you can narrow down the list to less lipophilic ones.
Permalink to Comment

12. ZINC on May 16, 2012 11:01 AM writes...

P.S. you may get lucky and find a less lipophilic one that also has potency for hSERT, since ChEMBL activity data is also imported into ZINC.
Permalink to Comment

13. John Irwin on May 16, 2012 12:22 PM writes...

You can specify maximum calculated logP of 3 as follows, for instance, http://zinc.docking.org/r...
Permalink to Comment

14. lynn on May 16, 2012 12:47 PM writes...

Have they tested effects on bacteria, especially S. aureus and B. subtilis? It's amazing how many human health drug candidates [or at least compounds in industrial chemical libraries] kill bacteria via membrane interactions. However, these effects are often abrogated in the presence of human serum, usually due to the compound's binding to albumin.
Permalink to Comment

15. MoMo on May 16, 2012 5:23 PM writes...

Antidepressants do more than just anti-depress!

For fun I used to read all of CombinatoRx's patents, just to see how many times tricyclic antidepressants showed up as synergists in just about every disease known to man.

But Lynn knows the score here. Big lipophiles can change transport dynamics in all cells everywhere, and special old-school ones like Trifluoroperazine do all kinds of biological tricks. It causes orgasms during sneezes! Some side effect!

Somewhere someone is thinking-Where can I get some? ACCCHOOO!
Permalink to Comment

16. Fred on May 16, 2012 5:27 PM writes...

im with luysii. we just dont know what we are doing. I think membrane phase transitions are an overlooked and potentially important mechanism in drug action. I have read anesthetics work by altering the Tm of membranes and that you can revive an anesthetized person by placing them in a hyperbaric chamber and crankin it up. does anyone know if this is true?
Permalink to Comment

17. Ethan Perlstein on May 16, 2012 5:58 PM writes...

Thank you all for a lively and stimulating discussion thread. I wish this kind of exchange was happening on the comments thread of my paper on the PLoS ONE website. I reiterate my invitation bring some of this discussion directly to the paper here: http://www.plosone.org/ar...

Thanks!

-Ethan
Permalink to Comment

18. DJN on May 16, 2012 6:36 PM writes...

Not only antidepressants but beta-blockers have very unusual effects on red cell membranes. Crenellated cells arise and unusual pH changes occur within the cells. Found this out the hard way back in 1971 when looking for compounds that would shift the oxygenation curve of hemoglobin in vivo. DJN
Permalink to Comment

19. newnickname on May 17, 2012 7:05 AM writes...

@16: "anesthetics work by altering the Tm of membranes and that you can revive an anesthetized person by placing them in a hyperbaric chamber and crankin it up. does anyone know if this is true?"

If you replace "person" with "newt", "mouse" or other small animal, it is definitely true. See the work of KW Miller (Harvard), 1970s - 1980s. Put a mouse in a hyperbaric chamber on a slowly rotating (rolling) platform (drum) and the mouse has a reflex action to right itself. Apply the anesthetic and the mouse takes a snooze and rolls over and over as the platform (drum) rotates. Crank up the pressure with an inert gas (e.g., He) and the mouse wakes up and the righting reflex returns. Bring the pressure back down to one atmosphere and the mouse goes back to sleep. I don't remember for sure, but I don't think the pressures are too bad: 1000 psi? 1500 psi? ("tank pressure").

The closest human "experiments" I can think of would be the rescue of divers from Raptures of the Deep (nitrogen narcosis; the bends) by placing them in hyperbaric chambers. It works.

Small pressure cells ("bombs" to us chemists - I hope I didn't set off a Homeland Security alert by typing that) are easy to come by and can accommodate yeast more easily than mice.