Should we use in vivo or in vitro evidence to characterize a drug?

[WSH]

Thank you for your thorough reply.

It's true that I've completely forgotten to check the in vivo evidence.

That reminds me of 8-OH-DPAT. They have proven in vitro that it's a full agonist at inhibiting forskolin-stimulated cAMP production at postsynaptic 5-HT1A receptors in the hippocampus.
But then Montigny came along and proved that it's actually a rather weak partial agonist using in vivo electrophysiology.

So yeah, in vivo > in vitro.

I agree, MK-801 does NOT seem to produce its effects via D2 in vivo.

 

[serotonin2A]

Thanks for the reply!

CSF contains approximately 150 mM Na+. Running a binding experiment using a buffer that contains a lower concentration of sodium would potentially yield results that are not physiologically relevant. Agonist binding to the mu-opioid receptor is known to be sodium-dependent, and something similar could be occuring with D2.

I'll admit that I have an in vivo bias. But that's because ultimately the point of pharmacology is to understand how drugs act in living systems in vivo!

I don't have a bad impression of 8-OH-DPAT (except for the fact that it isn't selective). My understanding is that 5-HT1A efficacy is tissue-dependent. 5-HT1A receptors are coupled to different G proteins in different brain regions (dorsal raphe, hippocampus, and cortex). This leads to regional differences in receptor reserve. Many 5-HT1A agonists show complex effects where they act as full agonists in certain brain regions and partial agonists in others.

 

[WSH]

I'll admit that I have an in vivo bias. But that's because ultimately the point of pharmacology is to understand how drugs act in living systems in vivo!

Yes, it goes like this when it comes to reliability: in vivo human testing > in vivo animal testing > in vitro assays

5-HT1A receptors are coupled to different G proteins in different brain regions (dorsal raphe, hippocampus, and cortex). This leads to regional differences in receptor reserve.

I've heard of it, but hasn't this "receptor reserve theory" already been disproven?

Quoting Montigny and Blier:

The possibility that this apparent partial agonistic property in dorsal hippocampus might have been due to a low receptor reserve in this region can be ruled out on the following grounds: in our previous experiments (using the same paradigm as for R- and S-enantiomers), we have shown that even when a marked suppression of firing activity of the same neurons is obtained with microiontophoretic application of 5-HT itself, following the restoration of their firing with acetylcholine, then raising the application current of 5-HT was still able to produce a potent inhibition (Godbout et al 1991). This finding thus shows that the blockade of the effect of 5-HT by R-(+)-OH-DPAT and S-(-)-OH-DPAT cannot be ascribed to the activation of a maximal number of 5-HT1A receptors by the latter drugs. This finding also suggests that there are spare receptors in the CA3 region with respect to inhibition of firing.

 

[sekio]

the ECMD found MXE did not bind to cloned human mu opioid recptors in vitro... I think that's damning enough evidence.

 

[WSH]

the ECMD found MXE did not bind to cloned human mu opioid recptors in vitro... I think that's damning enough evidence.

The examples of MK-801 and 8-OH-DPAT that we have discussed here show that in vitro assays can't really prove anything.

In vivo is where it's at, and so far there haven't been any MXE in vivo tests to my knowledge.

 

[endotropic]

The examples of MK-801 and 8-OH-DPAT that we have discussed here show that in vitro assays can't really prove anything.

In vivo is where it's at, and so far there haven't been any MXE in vivo tests to my knowledge.

Nicotine has naloxone reversible in vivo effects, but that doesn't tell us anything about its binding profile.

 

[ebola?]

Yes, it goes like this when it comes to reliability: in vivo human testing > in vivo animal testing > in vitro assays

Well, there's some information that you can glean via in vitro assays that you can't get with in vivo testing via current instruments.

ebola

 

[sekio]

in vitro assays can't really prove anything.

Well, no, they prove things, just not the exact same proofs that in vivo tests do.

 

[WSH]

Well, no, they prove things, just not the exact same proofs that in vivo tests do.

OK, I'll take back that they can't prove anything, that was badly phrased.

What I meant was that in vivo testing is more reliable than in vitro assays.

So when in vitro says one thing and in vivo says the opposite (e.g. as it is with MK-801), then the in vivo evidence is obviously more reliable.

 

[ebola?]

What I meant was that in vivo testing is more reliable than in vitro assays.

More precisely, I think you mean that in vivo testing is more valid (specifically, externally generalizable), not reliable. In vivo testing tends to lack reliability in comparison to in vitro testing, as exemplified by the larger n typically required. This is because animal bodies provide varying systemic contexts in which in vivo measurements are taken, so you have greater variability to collapse across.

So when in vitro says one thing and in vivo says the opposite (e.g. as it is with MK-801), then the in vivo evidence is obviously more reliable.

It depends on the evidence you're trying to collect and the specific methodology used. For example, I would take in vitro measurement of binding affinity and efficacy over in vivo evidence using animal-trained stimulus-discrimination and experimental pretreatment with selective antagonists (when specifically examining interaction at receptors, not downstream effects).

ebola

 

[WSH]

It depends on the evidence you're trying to collect and the specific methodology used. For example, I would take in vitro measurement of binding affinity and efficacy over in vivo evidence using animal-trained stimulus-discrimination and experimental pretreatment with selective antagonists (when specifically examining interaction at receptors, not downstream effects).

OK, so it depends.

Let's take MK-801 as an example. What do you trust more? The in vitro evidence that it's a direct D2 agonist or the in vivo evidence that it's not?

 

[Zilpe]

Forgive my ignorance. But how exactly do they do in vivo testing for receptor binding affinity? It makes sense that in vitro testing could eliminate some variables that are important to the overall effects in an organism but at the same time it's hard for me to conceive of an in vivo test that would be accurate. It's not like you can just look inside the brain of a living person to see what's going on at a neurochemical level, can you?

 

[ebola?]

Let's take MK-801 as an example. What do you trust more?

I find the critiques of the 2005 study rather convincing, but this has nothing to do with the general nature of in vivo versus in vitro evidence.

ebola

 

[WSH]

I find the critiques of the 2005 study rather convincing, but this has nothing to do with the general nature of in vivo versus in vitro evidence.

ebola

Thank you, you seem to be very knowledgeable, so may I ask you about the 8-OH-DPAT situation?

The in vitro evidence (inhibition of forskolin-stimulated cAMP production at postsynaptic 5-HT1A receptors in the hippocampus) suggests it's a full agonist, but the in vivo evidence (Montigny's electrophysiology paradigm) suggests it's only a partial agonist.

What's your take on it?

 

[endotropic]

Thank you, you seem to be very knowledgeable, so may I ask you about the 8-OH-DPAT situation?

The in vitro evidence (inhibition of forskolin-stimulated cAMP production at postsynaptic 5-HT1A receptors in the hippocampus) suggests it's a full agonist, but the in vivo evidence (Montigny's electrophysiology paradigm) suggests it's only a partial agonist.

What's your take on it?

Every time you move further away from receptor activation you get more and more response amplification. So in this case you have receptor activation -> g-protein activation -> Adenylyl cyclase inhibition -> changes in cAMP levels. Every receptor activates multiple g-proteins, every g-protein inhibits multiple adenylyl cyclase, etc. Eventually you saturate the response (there's only so much adenylyl cyclase around to inhibit), so if 8-OH-DPAT activates 60% of the receptors it binds to, by the time you get down to cAMP levels the effect is just as strong as you would get with a full agonist like 5-HT.

tl;dr cAMP accumulation is a low efficacy requiring assay, so partial agonists behave the same as full

If you look up 8-OH-DPAT efficacy on an in vitro assay that measures g-protein activation (e.g. [35S]GTPgammaS binding) you'll see it reported as a strong partial agonist (60-80% of 5-HT from what I recall).


This thread hasn't been about MXE/mu for a while now so I'm going to split this into its own discussion (PM me if you'd like a new title).

 

[WSH]

This thread hasn't been about MXE/mu for a while now so I'm going to split this into its own discussion (PM me if you'd like a new title).

I already thought about it, thank you for thinking ahead.

I'm gonna reply later to the rest of it when i have more time.

 

[serotonin2A]

GPCRs (5-HT1A included) can be coupled to multiple downstream effector mechanisms. The electrophysiological effects of 5-HT1A activation may only be partially mediated by effects on adenylyl cyclase/cAMP, or there may be no relationship whatsoever. So it would not necessarily be suprising that the results of a cAMP assay failed to yield the same result as an electrophysiological study. They are really assessing two different things that may or may not be related.

 

[serotonin2A]

You can use ex vivo binding or PET to assess receptor occupation in living animals.

 

[endotropic]

GPCRs (5-HT1A included) can be coupled to multiple downstream effector mechanisms. The electrophysiological effects of 5-HT1A activation may only be partially mediated by effects on adenylyl cyclase/cAMP, or there may be no relationship whatsoever. So it would not necessarily be suprising that the results of a cAMP assay failed to yield the same result as an electrophysiological study. They are really assessing two different things that may or may not be related.

As far as I know Gi/o coupled receptors like 5-HT1A cause changes in membrane excitability primarily by activating GIRK channels via Gbeta/gamma release, with calcium channel regulation playing a smaller role. The cyclase effect is tertiary to those effects if it impacts an e.phys experiment at all.