06 February 2017

Beware self-reacting fragments

Long-time readers will know that I have a peculiar fascination for artifacts of all kinds, particularly when they provide learning opportunities. A lovely example by Gerhard Klebe (Philipps-Universität Marburg) and collaborators has just appeared in Angew. Chem. Int. Ed.

The researchers have long been using the aspartic protease endothiapepsin (EP) as a model protein: we previously discussed how they compared half a dozen fragment-finding methods against EP, more recently arguing that crystallography is the best of the bunch. The new paper focuses on Compound 1. This molecule was a hit in five out of six fragment screens, each employing a different method, and was among the top ten hits in four of the screens. It produced the highest thermal shift (+3.4 °C), strongly inhibited the enzyme in two different biochemical assays, and even showed a dissociation constant of 115 µM by isothermal titration calorimetry (ITC).

Crystallography, though, told a different story. The researchers obtained high resolution structures (initially 1.25 Å, and ultimately 1.03 Å!) These revealed that the bound ligand was actually compound 2, which is composed of three molecules of compound 1. A variety of experiments, including anomalous scattering, high resolution mass spectrometry (HR-MS), and MS/MS fragmentation supports this assignment.

So what’s going on? The team notes that, although compound 1 does not aggregate and is not a PAINS compound, high-level quantum mechanical modeling suggested that the chlorine is susceptible to nucleophilic displacement, which probably wouldn’t surprise many medicinal chemists. Rearrangement of the resulting dimeric molecule produces compound 4, which could then react with yet another molecule of compound 1 through a radical mechanism to produce compound 2.

Allowing compound 1 to sit in buffer or methanol provided support for this mechanism and allowed the isolation of compound 4 and other degradation products, though compound 2 itself could not be detected. The researchers suggest it is particularly reactive and only stable when surrounded by the protein.

I applaud the investigators for pursuing this fascinating bit of science. This is academic research in the best sense of the phrase.

This is also the kind of investigation that would fall outside the scope of most industrial researchers, where the mandate is to discover promising drug leads as quickly as possible. More somberly, this story could have ended in embarrassment or worse had the researchers been less rigorous. The difficulty (and unlikelihood) of such lengthy investigations is why triaging shortcuts such as PAINS filters have been introduced, and why scientists using these tools must still be cautious: even molecules that aren’t PAINS can act through pathological mechanisms.

This is also why I believe that arguments that PAINS filters are inadequately defined and should thus be discarded are misguided. Sure, some PAINS molecules are drugs, and any rubric can be improved. But the nice thing about fragment screens is that they often produce a plethora of hits to pursue. A flawed triaging scheme will jettison some pearls among the pebbles, but without triage, far more resources will be lost chasing will-o'-the-wisps.


Peter Kenny said...
This comment has been removed by the author.
Peter Kenny said...

Hi Dan, I think that you’re confusing the criticism of PAINS with the criticism of Ro3 (neither hydrogen bond acceptors nor hydrogen bond donors are defined and the method for calculating logP is not given). The criticism of the PAINS filters, both by me and by Alex Tropsha and colleagues, can be summarized as ‘data does not support the rhetoric’. Specifically, we have argued that PAINS filters are typically used outside their applicability domain and are based on analysis of proprietary data. Perhaps my challenge to the use of PAINS filters to define JMC editorial policy will be dismissed as coming from outside the peer-reviewed literature. However, public criticism coming from a JCIM editor of data analysis in a JMC article is likely to be taken seriously. One question that you may wish to consider is how many of the hits from the six AlphaScreen assay panel were actually shown to be interfering with the assay?

Dan Erlanson said...

Hi Pete,

Perhaps I should have written "justified" instead of "defined", though one could argue that data underlying definitions contribute to the definition.

Wordplay aside, I think where we differ is that I believe that too many people spend too much time pursuing artifacts, and that even an incomplete and imperfect tool to avoid this is better than none. Keep in mind that the JMC guidelines, which are not
enforced, do not forbid PAINS from being published, they merely ask for additional experiments. Why would you disagree with this, particularly since you are so earnest in your call for more data to back up the PAINS definitions?

Peter Kenny said...

Hi Dan,

I think where we differ is in our respective perceptions of the value, predictive power and scope of the original PAINS analysis. Screening artifacts are (and have always been) troublesome and, in an ideal world, I would want to see assay results confirmed by direct affinity measurement (SPR is particularly good because you can get a handle on stoichiometry). It’s also worth remembering that assay interference can also lead to false negatives.

The journals do probably need to tighten their requirements for confirming assay results (whether or not the compounds match PAINS filters). One easy option would be to require that concentration responses be given in the supplementary information. I would argue that a public database of observed screening artifacts (both interference of undesirable mode of action) will be necessary to make real progress. If offered the choice between a predictive model and the data used to train it, I would generally opt for the latter.

Anonymous said...

One of the problems about filters such as PAINS and rules such as voldemort is that people think because they have taken them into consideration its plain sailing all the way.

Nearly every presentation I have seen in the last 3 years that describes a library, fragment or otherwise, makes a big play about applying a PAINS filter - downstream they may still have assay interference but will they investigate it properly because they 'filtered all bad compounds out'.

Its the same as the edict from above that all compounds that do not meet chemists eye be removed from screening sets - then lo and behold the first 2 screens read out with chemists running in horror from the hit set while management force them back saying 'you told us all material in the library would be chemically tractable'