A recent study suggesting a novel mechanism of psychedelic action was published just a few days ago. This study by Vargas et al, out of the Olson Lab, aimed to understand why tryptamine psychedelics but not serotonin (5-HT) causes hallucinations and subjective effects.
Where are receptors typically located?
5-HT2A receptors are g-protein coupled receptors (GPCRs), a large family of receptors that are expressed in the brain
GPCRs are embedded into the cell membrane and have an extra and intracellular side. The part of the receptor that is activated by an agonist, like a psychedelic or 5-HT, are found on the extracellular side. The other part of the GPCR that is responsible for downstream effects is on the other side of the cell membrane.
After activation by a drug or ligand, GPCRs can be internalized making them unavailable if more of the drug comes around. This internalizing means that the whole receptor (not just the one part) is now inside of the cell, in a little bubble called the endosome.
It’s thought that these receptors that are inside of the cell will not be active or may be further tagged to be degraded.
There has been research into the intracellular receptors as well, which suggests receptor pools are inside of the membrane, ready to be used.
There has also been some evidence that 5-HT2A receptors are found intracellularly associated with post-synaptic domains.
Fig. 1 GPCR example. The blue part is the membrane, which the receptor is embedded. The ligand typically binds on the top side (extracellular) and the signaling comes from the bottom (intracellular).
What is biased agonism and location bias?
Biased agonism, which is the idea that different ligands at the same receptor can induce different types of changes inside the cells after they bind. This has been shown with psychedelic signaling and the recruitment of different messengers, as well as with other classes of drugs like opioids and their respective receptors.
Biased agonism can be due to a variety of different factors, but boils down to how different ligands can alter the conformation - or the way the receptor sits - on the membrane.
Location bias, in the context of this recent paper, refers to the ability of certain ligands or drugs to preferentially drive GPCR activity from intracellular compartments and/or restrict activity to the plasma membrane, therefore altering signaling.
The rationale of the location bias of the ligands in this paper is their differences in lipid solubility or lipophilicity. This lipid solubility determines how well ligands can passively diffuse across the membrane. For example, serotonin cannot pass through the membrane whereas tryptamines like DMT and psilocin can.
Figure taken from - Biased agonism: An emerging paradigm in GPCR drug discovery by Rankovic et al 2016.
Major Findings:
Replication of results that 5-Meo-DMT induced plasticity both structurally and functionally.
To see structure-activity relationships, they use different levels of N-methylation, which is the addition of the methyl group on the nitrogen of a compound. This makes the compounds more lipophilic, meaning more likely to diffuse across a lipid membrane. They found that 5-HT did not produce increase in growth in rat cortical neurons, but increasing this N-methylation resulted in increased growth of neurons.
To make sure that these receptors are indeed intracellular, they use multiple cell types inlcuding human embryonic kidney cells expressing the 5-HT2A receptor, and cortical neurons expressing a fluorescent protein tied to the 5-HT2A receptor. They find that there is intracellular 5-HT2A receptors in neurons under specific conditions. And that there are differential localization patterns in other cell types.
Using multiple techniques to determine Gq and Beta-Arrestin (2nd messengers signaling), they find that efficacy of tryptamines to produce signaling in these assays was not correlated with dendrite growth.
They cite a negative correlation between the 5-HT2A receptor efficacy and dendritic growth, but note this should be interpreted with caution. A significant correlation between lipophilicity and structural plasticity was reported.
By modifying DMT, psilocybin and ketanserin (a non-hallucinogenic 5-HT2A antagonist) to make them impermeable - unable to cross the membrane - they find that these compounds were only able to produce dendritic growth when also applied with electroporation (putting holes in the membrane to allow things to get into it). Interestingly, membrane impermeable ketanserin was still able to block the effects of 5-HT and 5-Meo-DMT.
To see whether serotonin would produce effects in the cell with a little help from the Serotonin Transporter Protein, they introduce a construct containing SERT into the cells. They then treated the cells with permeable and impermeable DMT and found that both SERT+ or SERT-- cells resulted in increased structural plasticity with the permeable DMT but only resulted in increased plasticity with serotonin in the SERT+ cells.
Now they use mice to see how this would work in a living system. They injected a SERT construct into the PFC of these mice followed by administration of a serotonin releasing agent. They imaged spines 24 h later and found that the SERT+ animals resulted in increased structural plasticity. In behavioral tests, they found a significant reduction in immobility time in the forced swim task.
What does it all mean?
Overall, this research is suggesting three major things:
Intracellular 5-HT2A receptors exist in rat cortical neurons and can be activated
These receptors are only able to be reached by lipophilic compounds that pass through the cell membrane
Psychedelics, since they are more lipophilic, can pass through the membrane and activate these receptors along with membrane bound receptors. 5-HT can only do so in the presence of a transporter protein.
Considerations
Every research paper leaves more questions than answers - especially one of this caliber. This paper challenges the current dogma associated with GPCR signaling, not just with psychedelics.
Some questions and considerations:
Future studies looking at this in vivo should consider something other than the FST with unstressed animals. This test without an added stress model is only assessing unconditioned behaviors and testing the same mice can produce confounded results as they learn to cope.
The control experiments throughout the paper were extremely well done and thought-out. The use of HEK and cortical neurons was important to see what is happening from different levels of the system.
Using ex vivo, or in tissue slices, it would be interesting to see more work done associating the 5-HT2A receptor and its location bias, or the location bias of ligands.
There was little explanation about some of the correlations and assays done - psychlight1 and 2. Have these methods been used in other groups and validated by others outside the Olson lab?
The use of DIV6 cortical neurons was of concern for some as well. These are not really neurons at this day which could confound the “plasticity” because these cells are still maturing and growing new spines. These are day 18 embryonic rat cells (consistent with other literature), that should be allowed to grow further before testing - typically DIV15-21.
Interestingly, the paper used only tryptamine psychedelics and 5-HT. There was no mention of LSD or Lisuride in the paper, and considering they have different lipophilicity than tryptamines, future research is warranted.
This paper focuses heavily on structural plasticity, but doesn't show much functional data. Interested to see more on this in the future.