Skip to main content

In Silico

Unlovely Polyphenols

Here’s a new paper from the folks at the Burnham Institute and UCSD on a new target for vaccinia virus. They’re going after a virulence factor (N1L) through computational screening, which is a challenge, since this is a protein-protein interaction.
They pulled out a number of structures, which have some modest activity in cell infection assays. In addition, they showed through calorimetry that the compounds do appear to be affecting the target protein, specifically its equilibrium between monomeric and oligomeric forms. But the structures of their best hits. . .well, here’s the table. You can ignore compounds 6 and 8; they show up as cytotoxic. But the whole list is pretty ghastly, at least to my eyes.
These sorts of highly aromatic polyphenol structures have two long traditions in medicinal chemistry: showing activity in assays, for the first part, and not being realizable as actual drugs, for the second. There’s no doubt that they can do a lot of things; it’s just that getting them to do them in a real-world situation is not trivial. Part of the problem is specificity (and associated toxicity) and part of it is pharmacokinetics. As you’d imagine, these compounds can have rather funky clearance behavior, what with all those phenols.
So I’d regard these as proof-of-concept compounds that validate N1L as a target. I think that we’ll need to wait for someone to format up an assay for high-throughput (non-virtual) screening to see if something more tractable comes up. Either that, or rework the virtual screens on the basis that we’ve seen enough polyphenols come up on this target already. . .
Note: readers of the paper will note that our old friend resveratrol turns up as an active compound as well. It’s very much in the polyphenol tradition; make of that what you will.

25 comments on “Unlovely Polyphenols”

  1. anchor says:

    Derek: It seems to me that these compounds will be devoured by the liver. Indeed these are medicinal chemist’s nightmare! Lot of things can go wrong here and I see lot of red flags!

  2. Sili says:

    Completely Off Topic:
    Via World of Weird Things I just found this great documentary of the history of ‘cold’, i.e. thermodynamics made sexy. I’m only half an hour in – just got to Carnot – but so far I like it a lot.
    It’s a Nova programme, so go enjoy your tax dollars at work (I’ll just sponge off you like a filthy furrinner).

  3. partial agonist says:

    I wonder if these types of compounds turn up in protein-protein interaction screens since they are a type of compound that may tends to self-aggregat, giving the type of extended surface often needed for protein-protein interactions.
    I just reviewed an HTS for a protein-protein interaction target, and the hit list was just about as depressing.

  4. JC says:

    There’s an empty building in the next parking lot where they used to work on drugs for protein-protein interactions.

  5. screener says:

    Polyphenols are known promiscuous inhibitors (i.e. quercitin). I would also hesitate to call them “proof-of-concept compounds that validate N1L as a target”. Promiscuous inhibitors are not good tools for target validation since you have no idea what they are doing in cells, and they are likely hitting a number of different targets.

  6. Evorich says:

    Could they not have tested a bunch of known non-poly-phenol bcl-2 inhibitors if all they wanted was target validation and tool compounds??

  7. John says:

    100,000 experienced medicinal chemists out of work and the NIH is throwing money at academicians to publish papers like this one. Go figure.

  8. partial agonist says:

    One might also say
    100,000 experienced medicinal chemists out of work and big pharma mainstay Glaxo Smith Kline is throwing tons of money at worthless promisculous compounds like these and laying off even more chemists. Go figure.

  9. Les Lane says:

    Phenols binds nonspecifically to the amide backbone of proteins as well as to anything else that’s a hydrogen bond acceptor (as evidenced by the solubility of proteins in phenol).

  10. Curt F. says:

    1. Is there some new trend to omit explicit representation of the H atoms bonded to heteroatoms? I’m not used to seeing formulas with phenols written as -O instead of -OH. Can I write methylamine at “-N” and be done?
    2. This paper is still a paper, and although I am not an expert in medicinal chemistry, it still looks like it reports novel data in the field that deserved to be recorded in the annals of science. If people have problems that this research was funded in the first place, I guess you should have said something when the PIs’ grants got funded. Did anyone protest in 2006 when the NIH’s award of this grant was announced?

  11. processchemist says:

    Docking programs in the hands of some academical scientists can give funny results. I remember, years ago, an university spin off looking for the synthesis of an almost impossible derivative of a terpene… The logic underlying the funding of academic research remains a mistery, to me.

  12. dddd says:

    Yuk. Looks like cytochrome food to me.

  13. cliffintokyo says:

    Isosteres of Phenyl-OH anyone? -NHSO2R?

  14. Blah says:

    It’s really tiring hearing from all of these out of work people being total trolls on these boards… enough already. NIH-funded projects is how all of you got your training in the first place. So why turn around now and keep nagging that the NIH should not be funding academic projects anymore? What do you suggest? That they invest in privately held companies?!?

  15. Hap says:

    NIH is obtaining knowledge and training more scientists. As knowledge, this paper is sort of problematic because it suggests really promiscuous binders (which are likely to have bad pharmacokinetics, to boot) to inhibit protein-protein binding events – you don’t know what they’re doing, and the hits are members of a class notoriously hard to refine into lead compounds. So you haven’t really gotten much – even if you have a hit, you can’t refine it easily and can get little selectivity, so as a probe it’s probably unhelpful.
    As training, it’s more helpful, because the people working are learning how to look for drugs, but it still has the problem that it ignores a lot of knowledge on the characteristics of the hits. If you’re training people to make drugs, or look for them, then finding ligands isn’t going to be sufficient, and since even people that don’t do drug discovery know that, it seems woefully insufficient as training, unless you’re training people to be unemployed, which you could do for far cheaper. NIH is supposed to pay to train people, but if the training ignores many of the complexities of developing drugs (like “a ligand is not necessarily a drug”), then it’s not very helpful. You want the people NIH funds to be aware of the limitations of their methods, so that at least their students are so aware.
    Doing drug discovery without an awareness of what it entails is not a good use of anyone’s money, either as training or as information. You can either not do drug discovery under NIH funding or do it with an awareness of the issues that make it problematic (and perhaps even with an eye towards solving them).

  16. Blah says:

    Hap — I get it. It’s a poor lead compound or inhibitor. It’s a bad article in general. But do you know how many industry projects are just as unlikely to lead to an FDA approved drug?
    Then, who would you say should get NIH funding? This is a serious question. Who should get it?

  17. barry says:

    re: Cliffintokyo#13
    The isosteres for phenols depend on what the phenol is doing. If it’s an H-bond donor, the sulfonamides sometimes work. If it’s an H-bond acceptor, anisoles or even Ar-F sometimes work. If it’s playing both roles simultaneously, it may be irreplaceable. If its binding is mediated by waters, you may want to replace both the Ar-OH and the bound water.

  18. David Formerly Known as a Chemist says:

    It’s no surprise these compounds show up as actives. They show up as actives in MANY screens. A sizable portion of my career was spent with a chemistry CRO that did contract medchem work for many small companies. I can’t count the number of times I’ve been presented with similarly awful structures as so-called “lead compounds”. Polyhydroxylated anthraquinones, imines, anilines, all kinds of garbage. Most of these “leads” were presented by newly-started companies that screened early combichem libraries, or other similarly poorly-fashioned libraries. A number of them came from university spin-outs, where such libraries were screened. Problem is, academic labs usually don’t have access to high-quality screening collections, and thus end up using trash like this. Fortunately, this seems to be getting better.

  19. Hap says:

    If you want to play in the real world, you play by real world rules. If you’re going to train people to do drug discovery, don’t show up with half-assed target assessments and expect cash – in that case, why bother, because there are lots of for-profits who can do it better (and the “training” isn’t going to help your trainees, either). Either do drug discovery or focus on more general biochemical or pharmacokinetic concepts, particularly since one of the major gaps in knowledge is the assessment of pharmacokinetics and drug metabolism.
    Don’t fund drug discovery if the people you’re funding have no idea what the hell a drug looks like. I’m not really seeing what’s so hard here.

  20. partial agonist says:

    To be precise here, I am sure that the NIH did not fund the optimization of polyphenols as great leads for this target. They funded the development and implementation of a HIGH THROUGHPUT SCREEN, with apparently decent screening methodology, against a pretty reasonable target.
    The disconcerting thing is that the authors did not feel the need to go further down their hit list a bit and tell us what compounds emerged that are tractable (if any), even if they had a 15 uM IC50 instead of 1 uM.
    I don’t think that the NIH is at fault for the follow-through on the assay results, and I don’t think that the follow-through is any more reflective of what is bound to come out of an academic lab than is is reflective of what will come out of a lab in the USA, or a lab in the state of California, or a lab in a city that begin with the letter “L”, or in studies reported in papers whose first author has a last name beginning with “C”.
    Moral: you just can’t fall in love with IC50 data without factoring in the pros and cons of the structure that it is associated with.
    I say, blame the authors without assuming that their mindset reflects a widely accepted way to go about doing drug discovery.

  21. Ronathan richardson says:

    As someone who’s done a bit of academic screening, I think the fault lies with the screening centers and “medicinal chemists” they employ. There seem to be a LOT of people that work on screening as their primary job but don’t know how to go through a list of hits and pick the real ones, or even moreso, eliminate the crap compounds from a library. It’s really not hard. Alternatively, the NIH should require that these screening centers get real pharma-vet chemists (not hard to find, especially if you can promise a stable job!) to oversee the chemistry side of things.

  22. cliffintokyo says:

    re: #17
    Assumed that weak acid (H-bond donor) props of ArOH usually predominate.
    Most isosteres of this interaction might be *too big* (I mean steric hindrance of course), depending on the *receptor space* (3D geometry)

  23. cliffintokyo says:

    Way to go! Those sterically hindered sulfimides look nicely set up to do some receptor-selective tricks….or somesuch. (Don’t *forget* to tell the pharmacologists though)

  24. Glenn says:

    This reminds me of a project in which flavones showed activity in HTS assay. Some inexperienced scientist got all excited about them even after I told them they won’t find any SAR other than activity seemed to correlate with the number of OHs present.

Comments are closed.