Psychopharmacology’s blind spot

July 17, 2012

As I started to discuss in a previous post, psychopharmacology has a blind spot: the accumulation of antidepressants (ADs) in human brains over time. And I don’t just mean the accumulation that occurs during the weeks to months minimally required for a person to respond positively to ADs, but also the long-term accumulation following years of chronic AD use. Unfortunately, the blind spot is exacerbated by reports of trace amounts of ADs in the serum and cerebral spinal fluid (CSF) of responders, which only encourages psychopharmacologists to dismiss AD accumulation as gratuitous. However, serum and CSF drug levels don’t accurately census subcellular and membranous reservoirs of ADs, which old theory predicts and new experiments bear out.


In order to eliminate this blind spot, psychopharmacologists have to stop ignoring the direct effects of ADs on membranes, even if some of these effects are observed in cellular models in the lab or at doses that are ostensibly “too high” to be clinically relevant. (For more on the specious “too high” concentration argument, see my recent post about worms on Prozac). That’s obviously easier said than done.


So let’s start where we have consensus. First, everyone agrees that ADs (tricyclics + SSRIs) are slow acting. Second, everyone agrees that the effects of ADs on serotonin reuptake are fast. The uncertainty arises when atttempts are made to reconcile the rapid inhibition of serotonin reuptake by ADs in all experimental model systems examined with the slow-to-emerge neurogenesis induced by ADs in specific brain regions of chronically dosed rodents, changes which presumably also occur in the brains of people taking ADs.


Is that time lag the result of a slow cascade of physiological adaptations triggered directly by increased synaptic serotonin concentrations? Or are there non-serotonin-based, slow-acting mechanisms involved as well?


One important clue comes from a simple electron microscopy study described in an elegant paper by Renate Lüllman-Rauch’s group from 1979, and published in Biochimica et Biophysica Acta (my favorite journal name to say aloud). The authors used transmission electron microscopy and the freeze-fracture technique to examine various tissues, including retinas and adrenal glands but alas not brains, of rats that were chronically treated with the phenethylamine drug chlorphentermine. Chlorphentermine was once used as an appetite suppressant, not surprising given its structural similarity to amphetamine. (As an aside, sertraline/Zoloft, a major focus of the Perlstein Lab, is actually a chemical descendent of amphetamine).


It was known back in 1979 that lipophilic weak bases like chlorphentermine induce the formation of “myeloid bodies,” or densely packed, often concentric, arrays of phospholipids. The mechanistic interpretation is that acidic phospholipids are bound to amphipathic weak base drugs, and the resulting drug-lipid complexes are no longer recognized by the digestive enzymes (e.g., phospholipases) that normally break down phospholipids in the maintenance of cellular membrane homeostasis.


I’ve taken the liberty of reproducing several stunning electron micrographs from the Lüllman-Rauch paper, which tell better than a thousand words what chlorphentermine accumulation looks like in animal tissues.

Two myeloid bodies are indicated by orange arrows. They are located inside lysosomes, degrading organelles where all cellular components, including phospholipids, are recycled. High-resolution magnifications revealed the fine structure of those inclusions, which are indicative of lamellar phase phospholipids:

The authors also observed non-lamellar (i.e., non-bilayer) architecture, namely hexagonal phase lattices:

My lab has produced very similar images of hexagonal lattices, but from yeast cells treated with the SSRI Zoloft. These images are obviously 2-D slices. So what is the 3-D structure of those chlorphentermine-induced inclusions? Here was an artist’s rendition of the freeze-fracture data, revealing concentric bundles of hexagonal tubules of phospholipids:

Besides providing visually arresting proof of phospholipidosis in response to chronic chlorphentermine accumulation, the authors noted that phospholipid inclusions were different in different cell types, in this case retina vs. adrenal tissue:


The above-mentioned biochemical and physicochemical knowledge supports the interpretation of the present findings, namely that retinal pigment epithelium prefers to accumulate the phospholipids in hexagonal rather than in lamellar phase. In other types of cells, which develop both types of inclusion bodies, the composition of stored lipids as well as the local conditions may be different such that both lipid phases, and additionally transitions between both, may occur.”


The question you’re probably asking yourself right now is, do antidepressants induce cell membrane changes in the human brain over long time scales, and is it required for their therapeutic efficacy? The short answer is no one seems to have looked carefully. In the next post on psychopharmacology’s blind spot, I’ll dig a little deeper..

  • arnieperlstein

    Looking forward to your deeper digging….

  • arnieperlstein

    I have a layman’s question—how many drugs administered to humans for various diseases have recognized side effects which involve accumulation of the drug in cell membranes?

    • Ethan Perlstein

      I don’t know the exact number. A well known case is amiodarone (, which is an antiarrhythmic agent.

      • arnieperlstein

        As we (coincidentally) know from recent personal experience, amiodarone can interact toxically with other medications (like colchicine)–I wonder if the buildup in the membrane is associated with that toxicity? And perhaps in some strange way that relates to the positive effects of membrane buildup of SSRI’s?

        Just a layman having some fun thinking about stuff way over my head….

  • Tetyana Pekar

    Ethan, are there known long-term side-effects of SSRIs (outside of like low libido, I mean things that become apparent after long-term use, that can be linked to ADs?) Have there been studies on patients on 150 or 200mg of sertraline? It was introduced in 1991 (Wikipedia tells me), so I guess 20 years is the max, really, for sertraline, but what about other ADs? Could this be related to the whole notion of “poop-out” effect (and is it actually real)? Maybe this is completely not your field of expertise, but these are just the questions that came to me as I read the post. I have to read more, but, I freaking LOVE LOVE LOVE the design of your site. So sleek and sexy.

    • Ethan Perlstein

      Thanks for your comment and questions, Tetyana!

      I need to bone up on my clinical knowledge, so I don’t have too many answers. I do know that Zoloft (Pfizer’s SSRI) followed Prozac but only 2-3 years later. I’ve heard of the poop-out effect, but nothing deep. There is something called SSRI discontinuation syndrome, basically a fancy name for withdrawal.

      There must be people who have been on SSRIs for a decade or more but I haven’t come across any damning evidence. Then again I haven’t performed an exhaustive search. David Healy, an Irish psychiatrist and self-described psychopharmacologist, is on Twitter and has a blog where he talks about the possible birth defect risk increase associated with pregnant mother taking SSRIs. Most people assume that drug accumulation cause side effects, but my lab’s work with yeast suggests that drug accumulation of Zoloft may actually be relevant to its therapeutic effects.

      My aim with these blog posts is to promote a balanced approach to understanding the full breadth of complex pharmacology of antidepressants.

      • SlimNm has a lot of anecdotal evidence on long-term SSRI-use.

  • SlimNm

    Do SSRIs accumulate in the body and last longer than it’s half-life would suggest?

  • Oleg Yu. Orlov

    Oleg Orlov
    Ethan: Would you hopefully be so kind to provide full reference regarding Renate LuellmannRauch’s publication (if not the .pdf).