Non-peptide ligands for neuropeptide receptors?

[Deleted member 170540]

The endogenous compounds neuropeptide Y and the more recently found neuropeptide S seem to have some interesting effects... NPY is involved in regulation of appetite, for instance. NPS regulates food intake, anxiety and wakefulness.

Intracerebroventricular injection of either of these neuropeptides causes an anxiolytic effect in animals, which is not accompanied by sedation. In fact NPS seems to increase wakefulness and exploratory behaviour. Both peptides seem to have something to do with the reward circuits of the brain. Rats learn to self-administer NPS cerebroventricularly ( http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3066039/ ) and direct microinjection of NPY in the nucleus accumbens produces a conditioned place preference ( http://www.ncbi.nlm.nih.gov/pubmed/8278431 ). Also, antagonists of these neuropeptides are anxiogenic and lab animals react to them aversively.

The 'rewarding' properties of these peptides suggest that agonists of the corresponding NPS and NPY receptors could serve as future recreational drugs or medical anxiolytics... Unfortunately the peptides themselves can't cross the blood brain barrier (although some NPS crosses the BBB if administered intranasally), so one would need to develop non-peptide agonists. Unfortunately there are only a few known non-peptide ligands for these receptors, and they are all antagonists...

How does one actually develop a non-peptide ligand for a peptide receptor? Obviously one can't go about synthesising lots of random compounds and testing them all for receptor affinity. There has to be a more rational way of deducing the structural features a molecule must have to bind to a particular receptor. Is there some computer program that tries to guess 'promising' molecular structures?

 

[Sturnam]

How does one actually develop a non-peptide ligand for a peptide receptor? Obviously one can't go about synthesising lots of random compounds and testing them all for receptor affinity. There has to be a more rational way of deducing the structural features a molecule must have to bind to a particular receptor. Is there some computer program that tries to guess 'promising' molecular structures?

Actually, they usually just synthesize lots of compounds.. It's called high-throughput screening. They literally synthesize thousands of compounds and screen them for activity.

Alternatively, they can start from a library of (usually) hundreds of thousands of compounds, find the one with the highest affinity for the desired target, and then use high-throughput screening from there.

 

[Deleted member 170540]

There's lots of known non-peptide agonists that mimic the action of opioid peptides at the mu, delta and kappa receptors, but theres currently only non-peptide antagonists for the NPS, NPY and NPFF receptors. Is there some reason why it's easier to develop antagonists than agonists for the peptide receptors? Or is it just that they don't want to develop NPS agonists because those compounds may become future drugs of abuse?

 

[negrogesic]

Algorithmic modeling techniques seem to be inherently limited by our current-past understanding of these target sites. While the number of 'novel' compounds generated are likely immense, these compounds are ultimately bound within the parameters of our 'current' understanding. To gauge the activity of the compounds as either 'agonist-antagonist' is simplistic; however, in respect to any disparity of antagonist over agonist, I wouldn't think that programs designed to generate 'keys for locks' would care which way they 'turn' (admittedly, a poor metaphor). Point being: Future abuse is unlikely to be major determinate/factor in the modeling of these compounds. Development of 'drugs' from these models could be mediated by such concerns as abuse, profitability etc...

I am reminded of a famous William Borroughs rambling in Naked Lunch:

Followers of obsolete unthinkable trades, doodling in Etruscan,addicts of drugs not yet synthesized, black marketeers of World War III, excisors of telepathic sensitivity, osteopaths of the spirit, investigators of infractions denounced by bland paranoid chess players, servers of fragmentary warrants taken down in hebephrenic shorthand charging unspeakable mutilations of the spirit, officials of unconstituted police states, brokers of exquisite dreams and nostalgias tested on the sensitized cells of junk sickness and bartered for raw materials of the will, drinkers of the Heavy Fluid sealed in translucent amber of dreams

 

[Sturnam]

There's lots of known non-peptide agonists that mimic the action of opioid peptides at the mu, delta and kappa receptors, but theres currently only non-peptide antagonists for the NPS, NPY and NPFF receptors. Is there some reason why it's easier to develop antagonists than agonists for the peptide receptors? Or is it just that they don't want to develop NPS agonists because those compounds may become future drugs of abuse?

Well, for the analogy (lots of non-peptide agonist for opioid receptors), it's that we discovered that series of receptors from agonist to receptor (mu, delta, etc). We already had a non-peptide, and natural compound to base our further SAR from, and actually, the existence of morphine lead us to really characterize the mu opioid receptor. Whereas with the NPS, NPY, and some related, we discovered their existence through genetic mapping/clonic/protein related ways. Basically, that we're starting at the opposite side of the problem as with the opioid receptors. We have a natural receptor, a protein, and we have to find some small molecule that binds into it and activates it. At that point, finding one by methods other than random screening hundreds of compounds seems to be out of our reach, as negrogesic said.

And as far as antagonists vs agonist, I think that's just what ended up happening with drug discovery. Only antagonists. I think it's the case with a few neuropeptide receptors.

But no, I don't think they're not making them because of potential abuse potential.

 

[Nagelfar]

Is there some reason why it's easier to develop antagonists than agonists for the peptide receptors?

As negrogesic alluded to with the words "limited by our current understanding": knowing the method of action first (i.e. there being already existent agonist examples on hand), and having a guess & check to see if that method of action is blocked, is simpler to do. If only because the variables of the activity to be altered are already familiar and have an expected baseline to work from.

In other words, all antagonists have to do is interfere with simultaneous release of agonists; their shape must block/inhibit, but nothing else beyond. If, say, its a receptor agonist instead of an antagonist: then signal processing is required and an action potential being fired is a further step necessitated for its creation. Another step beyond just conforming to the ligand site needed to bind at to block the entry of one which already sets off such a signal. Agonists, at least receptor agonists (I'm not so sure of allosteric agonists anymore), are more innately complex than antagonists.

 

[Deleted member 170540]

Thank you for your replies...

I found some info about the SAR of neuropeptide S agonists and it seems that the peptide chain can be shortened to only 10 amino acids while retaining activity. Probably still too big peptide to cross the bbb, though.

Also found a very recent journal article about the nasal use of NPS: http://www.ncbi.nlm.nih.gov/pubmed/21871467

Our findings provide evidence for memory-enhancing and anxiolytic effects of icv NPS in a non-social context. We could further show that these effects are context-specific, as social memory and social preference behavior remained unchanged after icv NPS. The effects of icv NPS were replicated by nasal application of the neuropeptide. Thus, nasal application of NPS seems to be a useful method in rodents for screening for behavioral or physiological effects before more specific and time-consuming, intracerebral methods are employed, and may represent a viable therapeutic approach for NPS treatment of patients with psychiatric illnesses such as anxiety or panic disorders.