Plosone study

[markosheehan]

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0009019
does anyone know how reliable this study is. it seems flawed and many of the affinity values dont match up with the PDSP ki database. does anyone know what hot ligands are used for psilocin and LSD in it?

 

[serotonin2A]

The assays were performed using a mixture of agonist and antagonist radioligands. I think each binding assay was only conducted once (i.e., no independent repeats). A few of the Ki values appear to be a outliers -- which is not necessarily unusual (that is why assays are normally repeated several times).

If memory serves, hot ligands for 5-HT1A, 5-HT2A, and 5-HT2C were 8-OH-DPAT, ketanserin, and mesulergine, respectively.

 

[markosheehan]

Do you know if they switched hot ligands between psilocin and LSD for the same receptors. And you do you know what hot ligands they used.? It does not say it on the study

 

[serotonin2A]

Do you know if they switched hot ligands between psilocin and LSD for the same receptors. And you do you know what hot ligands they used.? It does not say it on the study

It would be highly unusual for a competitive binding study to switch hot ligands. As I noted before, the hot ligands for 5-HT1A, 5-HT2A, and 5-HT2C were 8-OH-DPAT, ketanserin, and mesulergine, respectively. I would have to look up the other radioligands.

 

[markosheehan]

Yes it would be unusual but i think it happend based on his unusual results. Like the affinity for psilocin for 2a is very low on the study compared to other serotonin receptors. It doesn't make sense.

 

[serotonin2A]

Yes it would be unusual but i think it happend based on his unusual results. Like the affinity for psilocin for 2a is very low on the study compared to other serotonin receptors. It doesn't make sense.

They definitely used the same hot ligands for psilocin and LSD. I am 100% certain.

As I said, some of the values are probably outliers -- there are many apparent outliers in the data set. PDSP generated data is generally reliable but it is the responsibility of the investigator ordering the assays to carefully screen the data for outliers. The PDSP always includes a control ligand on the screening plates to make sure that the assays are functioning normally, but that is usually the extent of their evaluation.

 

[markosheehan]

Thomas ray will claim his data is the most reliable.
His data was from the PDSP.
Which would you say is more reliable for psychedelic affinitys. His study or other ones

 

[serotonin2A]

Thomas ray will claim his data is the most reliable.
His data was from the PDSP.
Which would you say is more reliable for psychedelic affinitys. His study or other ones

Why would the data in the PLOS One paper be more reliable than other assessments?

Was my critique not clear? Ray's data is filled with outliers, which means it is not reliable. Some of the data in the PLOS One paper does not even match the Ki values reported from other PDSP screenings. Take a look at other PDSP screenings for psilocin, 5-MeO-DMT, LSD, DMT, etc. in the literature and compare the data reported by Ray.

As an example, take a look at the mescaline data for 5-HT2A. Numerous other studies performed by multiple groups have consistantly shown affinity in the ~500 nM range (vs agonist radioligands) or the ~5000 nM range (vs antagonist radioligands).

It should be pretty obvious that Thomas Ray does not really have a good understanding of pharmacology because he tried to measure ligand selectivity using a mixture of agonist and antagonist radioligands. There is nothing inherently wrong with using agonist and antagonist hot ligands to measure affinity, but it makes his selectivity calculations meaningless.

 

[Cotcha Yankinov]

Numerous other studies performed by multiple groups have consistantly shown affinity in the ~500 nM range (vs agonist radioligands) or the ~5000 nM range (vs antagonist radioligands).

Is the difference in measured affinity with agonist vs. antagonist radioligands due to differences in agonist vs. antagonist affinity when the receptor complex is coupled to a G-protein?

Is the receptor homogenate prepared by doing something like engineering a cell to express only 5-HT2A then homogenizing it, and hence some of those receptors may still be coupled when doing the radioligand assay?

 

[serotonin2A]

Is the difference in measured affinity with agonist vs. antagonist radioligands due to differences in agonist vs. antagonist affinity when the receptor complex is coupled to a G-protein?

Antagonists bind to active and inactive conformations with equivalent affinity. By contrast, agonists have higher affinity for the active conformation. So when a receptor is labeled with an antagonist, an agonist displacing drug has to occupy both low-affinity and high-affinity sites. This ends up increasing the concentration required for the displacing ligand to produce 50% occupation (ie, the Ki) because it has to occupy some of the low-affinity sites in order to displace the antagonist.

I hope I explained it in a way that makes sense?

Is the receptor homogenate prepared by doing something like engineering a cell to express only 5-HT2A then homogenizing it, and hence some of those receptors may still be coupled when doing the radioligand assay?

The homogenates could be from a cell line or from native tissue. A proportion of the sites are probably G protein coupled (it depends on the tissue or the cell line used).

 

[Cotcha Yankinov]

Thanks for that explanation, very interesting...

Picture an agonist displacing ligand and antagonist radioligand mixture reaching equilibrium in a homogenate containing some high-affinity state receptors.

If the agonist binding leads to a change in the conformation of the receptors and hence a decrease in the agonist's affinity for them, would the displacing agonist's affinity actually appear to decrease as time went on and more receptors went inactive due to agonist binding? Or is there a perfect equilibrium reached, and the receptors are good about rotating from inactive to active even in a homogenate?

It seems like an inverse agonist radioligand would especially appear to decrease an agonist displacing ligands affinity as the mixture reaches equilibrium and more inactive receptors form, while this may not be seen with a silent antagonist, or when using only low-affinity state receptor homogenates.

Inversely, maybe an antagonist displacing ligand's affinity appears to increase as an agonist radioligand de-activates receptors.

Is the low-affinity state receptor still a site of interest because it may spontaneously couple frequently enough, and a ligand that is already bound there when the receptor becomes active may immediately activate the receptor without needing to re-bind?

 

[serotonin2A]

Picture an agonist displacing ligand and antagonist radioligand mixture reaching equilibrium in a homogenate containing some high-affinity state receptors.

If the agonist binding leads to a change in the conformation of the receptors and hence a decrease in the agonist's affinity for them, would the displacing agonist's affinity actually appear to decrease as time went on and more receptors went inactive due to agonist binding? Or is there a perfect equilibrium reached, and the receptors are good about rotating from inactive to active even in a homogenate?

Two points: (1) Competitive binding studies are usually performed after the binding of the radioligand reaches equilibrium. (2) Antagonists do not change the conformation of the receptor. Rather, they bind non-selectively to the active and inactive conformations.

 

[Cotcha Yankinov]

Hmmmm I may not have been very clear - if an agonist can change the conformation of the receptor to inactive then are more and more receptors going to become inactive as time rolls on in any homogenate containing an agonist, or even in a homogenate will the receptors be cycling through inactive - active states by utilizing loose G protein subunits floating around and hence an equilibrium will be found?

So when a receptor is labeled with an antagonist, an agonist displacing drug has to occupy both low-affinity and high-affinity sites. This ends up increasing the concentration required for the displacing ligand to produce 50% occupation (ie, the Ki) because it has to occupy some of the low-affinity sites in order to displace the antagonist.

The picture I get from this is that the agonist displacing ligand does a decent job at binding to active receptors through the antagonist radioligand's blockade but then the agonist displacing ligand has to fight relatively harder to bind to the remaining inactive receptors - if the sites of interest are the active receptors, do you think using antagonist radioligands isn't preferable?

So let's say that after the agonist displacing ligand reaches a certain concentration and manages to bind to the majority of active receptors (which is in vitro mediating most of the physiological response) it then enters a state of diminishing returns in terms of concentration vs. physiological response.


Using an antagonist radioligand with mescaline for example, the Ki could appear elevated to 5000 nM even though long before then it may have already bound to many active receptors (from 500 nM to 5000 nM it was just diminishing returns in terms of binding to active receptors/physiological response because the majority of active receptors were bound and mescaline was just struggling to bind to inactive receptors)

I hope this makes some shred of sense.

 

[Coolwhip]

The first hole I see in that scenario is association/disassociation rates, ligands don't stay bound to a receptor, they bounce around like in pinball machine(horrible analogy)

 

[Charles_Voynich]

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0009019
does anyone know how reliable this study is. it seems flawed and many of the affinity values dont match up with the PDSP ki database. does anyone know what hot ligands are used for psilocin and LSD in it?

For what it's worth, we've already found a number of issues with PDSP data. It's a good starting point, but it seems inconsistencies are fairly common.

 

[serotonin2A]

Hmmmm I may not have been very clear - if an agonist can change the conformation of the receptor to inactive then are more and more receptors going to become inactive as time rolls on in any homogenate containing an agonist, or even in a homogenate will the receptors be cycling through inactive - active states by utilizing loose G protein subunits floating around and hence an equilibrium will be found?



The picture I get from this is that the agonist displacing ligand does a decent job at binding to active receptors through the antagonist radioligand's blockade but then the agonist displacing ligand has to fight relatively harder to bind to the remaining inactive receptors - if the sites of interest are the active receptors, do you think using antagonist radioligands isn't preferable?

So let's say that after the agonist displacing ligand reaches a certain concentration and manages to bind to the majority of active receptors (which is in vitro mediating most of the physiological response) it then enters a state of diminishing returns in terms of concentration vs. physiological response.


Using an antagonist radioligand with mescaline for example, the Ki could appear elevated to 5000 nM even though long before then it may have already bound to many active receptors (from 500 nM to 5000 nM it was just diminishing returns in terms of binding to active receptors/physiological response because the majority of active receptors were bound and mescaline was just struggling to bind to inactive receptors)

I hope this makes some shred of sense.

Agonists shift the receptors to the "active" conformation, not the inactive conformation. Antagonists don't alter the equilibrium -- they bind to both the active and inactive conformations and occupy them so other ligands cannot bind, but they don't shift the equilibrium one way or the other (hence why they do not act as agonists or inverse agonists).

It sort of makes intuitive sense to use agonist radioligands. However, in practice, they can cause complications. They tend to have relatively low levels of specific binding (most receptors are not G protein coupled at any given time), which necessitates using radioisotopes with very high specific activity (which makes the assays more difficult, more dangerous, and more expensive). For some receptors, it may not be possible to use a radioisotope with high specific activity (adding iodine or bromine to some agonists may disrupt binding or intrinsic activity). I've also seen cases where functional measures showed a better correlation with binding data obtained with antagonist radioligands vs. agonist radioligands.

The Ki values obtained with agonist and antagonist radioligands tend to be correlated, so for certain purposes there is no advantage gained by using an agonist radioligand. Studies geared toward determining selectivity across multiple sites obviously should use agonist radioligands, but if you just want to show there is a ligand-receptor interaction then it may not matter how the receptor is labeled.

 

[Cotcha Yankinov]

Agonists shift the receptors to the "active" conformation, not the inactive conformation.

Sorry, I've gotten confused here - please see the bottom and left depictions here in the 4 stages
https://upload.wikimedia.org/wikipedia/commons/c/c9/GPCR_cycle.jpg

Are those two stages what is meant by "high-affinity state" and "active"?

https://upload.wikimedia.org/wikipedia/commons/e/e6/GPCR_activation.jpg - this depicts the "active state" as one where Ga is in the receptor (as opposed to the above depiction, where the Ga is away from the receptor when depicted as "active"). Leave it to Wikipedia to confuse the shit out of me I suppose. Are the depictions just not super visually accurate?

Or is low-affinity state used to describe a receptor that isn't actively going through those 4 receptor stages at all, and they are not really connected to a viable pool of G-proteins to associate with?

What I was thinking was that if agonists have higher affinity for active states, and agonists shift the conformation to active, then an agonist could cause the subunits to dissociate and then the next agonist would have higher affinity for that receptor that is now in the active state. This could mean that the longer a displacing agonist sat in an assay and shifted the receptors to high-affinity (active) states, the higher its affinity would appear to be if measured with an antagonist radioligand (because the antagonist hasn't seen an equivalent rise in affinity with the generation of more active states).

I suppose this wouldn't be appreciable depending on the difference in agonist affinity for active vs. inactive conformations and depending on the ability of the homogenate to form low-affinity state receptors as the homogenate ages.

The first hole I see in that scenario is association/disassociation rates, ligands don't stay bound to a receptor, they bounce around like in pinball machine(horrible analogy)

Can an agonist stay bound to a receptor and continually set off G-protein dissociation and allow re-association without needing to unbind?

 

[serotonin2A]

Sorry, I've gotten confused here - please see the bottom and left depictions here in the 4 stages
https://upload.wikimedia.org/wikipedia/commons/c/c9/GPCR_cycle.jpg

Are those two stages what is meant by "high-affinity state" and "active"?

https://upload.wikimedia.org/wikipedia/commons/e/e6/GPCR_activation.jpg - this depicts the "active state" as one where Ga is in the receptor (as opposed to the above depiction, where the Ga is away from the receptor when depicted as "active"). Leave it to Wikipedia to confuse the shit out of me I suppose. Are the depictions just not super visually accurate?

Or is low-affinity state used to describe a receptor that isn't actively going through those 4 receptor stages at all, and they are not really connected to a viable pool of G-proteins to associate with?

The first illustration omits much of the story. The second illustration shows the receptor in a low affinity conformation (left) and a high affinity conformation (right), but is oversimplified. Look up the "cubic ternary complex model" to see some more of the complexity. However, even that model is oversimplified.

What I was thinking was that if agonists have higher affinity for active states, and agonists shift the conformation to active, then an agonist could cause the subunits to dissociate and then the next agonist would have higher affinity for that receptor that is now in the active state. This could mean that the longer a displacing agonist sat in an assay and shifted the receptors to high-affinity (active) states, the higher its affinity would appear to be if measured with an antagonist radioligand (because the antagonist hasn't seen an equivalent rise in affinity with the generation of more active states).

The competitive binding assays measure the proportion of receptors that are occupied by the agonist at different concentrations.

I suppose this wouldn't be appreciable depending on the difference in agonist affinity for active vs. inactive conformations and depending on the ability of the homogenate to form low-affinity state receptors as the homogenate ages.

Can an agonist stay bound to a receptor and continually set off G-protein dissociation and allow re-association without needing to unbind?

Yes, that is definitely possible. That is exactly what happens with an irreversible agonist.

 

[Cotcha Yankinov]

The competitive binding assays are measuring how much of the antagonist radioligand is displaced -- it doesn't matter what affinity state the receptor occupies once the radioligand unbinds.

Sorry I'm being clear as mud, but won't the displacing agonist be better at displacing the antagonist radioligand if the agonist has been sitting in the homogenate and shifting the ratio of active vs. inactive receptors towards more active?

As an example, if you measured an agonist displacing ligand against an antagonist radioligand in a homogenate containing mostly active receptors and compared that with a reading from a homogenate of mostly inactive receptors, wouldn't the two apparent affinities differ even though the ligands fundamentally remain the same?

I was thinking one confounder from assay to assay would be that maybe some homogenates are better or worse at cycling from active to inactive. I hope I'm not giving anyone a stroke lol.

 

[serotonin2A]

Sorry I'm being clear as mud, but won't the displacing agonist be better at displacing the antagonist radioligand if the agonist has been sitting in the homogenate and shifting the ratio of active vs. inactive receptors towards more active?

As an example, if you measured an agonist displacing ligand against an antagonist radioligand in a homogenate containing mostly active receptors and compared that with a reading from a homogenate of mostly inactive receptors, wouldn't the two apparent affinities differ even though the ligands fundamentally remain the same?

I was thinking one confounder from assay to assay would be that maybe some homogenates are better or worse at cycling from active to inactive. I hope I'm not giving anyone a stroke lol.

Sorry, when I went back and read my answer, I realized I should have written that differently -- I started to edit my answer, but you responded before I could finish. Competitive binding assays measure the concentration-dependence of agonist binding to the receptor. With an antagonist radioligand, it doesn't matter what state the receptors are in, because antagonists bind to all states nonselectively. So usually, you incubate the test ligand, radioligand, and receptor, and then you measure how many sites are occupied by the test ligand. If the test ligand is an agonist and activates the receptor, that wouldn't have any effect on the subsequent binding of an antagonist radioligand because antagonists are equally happy to bind to the active or the inactive conformation.