NMR poll!

Among fragment-finding techniques, nuclear magnetic resonance (NMR) ranks near the top. Protein-detected methods, like HSQC/HMQC-based SAR by NMR, helped usher in fragment-based drug discovery as a practical endeavor. More recently, ligand-detected methods such as line broadening (or CPMG), STD, and WaterLOGSY appear to have gained the edge. There are also more boutique methods, such as ILOE and spin labeling. And of course, some people proudly embrace their fluorine fetishism.

So what’s your favorite flavor? Now's your chance to weigh in on our latest poll (on the right). The first question asks whether you use NMR, and the second asks which methods you use. PLEASE ANSWER BOTH QUESTIONS - the free version of Polldaddy doesn't track individuals, so we need the answer to the first question to know the total number of respondents.

And, as always, your comments are welcome.

Crystallography as a primary screen: the case of HisRS

X-ray crystallography plays a starring role in fragment-based lead discovery. But, with a few exceptions, it is rarely the primary screen. One of these exceptions was reported recently in Acta Crystallogr. D by Wim Hol and coworkers at the University of Washington, Seattle.

The researchers were interested in the histidyl-tRNA synthetase (HisRS) from Trypanosoma cruzi, the parasite that causes Chagas disease. They had previously constructed a library of 680 fragments in pools of 10, designed such that the fragments within each pool had different shapes to facilitate crystallographic screening. Crystals of HisRS•His (ie, the enzyme in complex with its histidine substrate) were soaked with the pools, such that the final concentration of each fragment was 1.5 mM. Fifteen of these pools showed new electron density, and these in turn yielded 15 different fragment hits when the pools were deconvoluted. Two additional fragments from the deconvolution process showed weak density, though these were not further pursued.

Strikingly, all 15 fragments bound to the same site, a site not observed in the absence of the fragments. This is a narrow groove described by the researchers as a “document sleeve,” and in fact all the hits are single six-membered aromatics or double aromatics with few or no aliphatic substituents. Most of the fragments are also quite small, with the majority having just 8 or 9 non-hydrogen atoms. Although all the fragments bind in roughly the same plane, there is considerable variation in the positions of substituents, and some of the fragments appear to bind in multiple orientations.

Next, the researchers tested their hits in orthogonal assays. Only one fragment showed thermal stabilization of the enzyme, and only three showed any inhibition in a functional assay (at most 39% at 2 mM of fragment). Thus, these are very weak binders.

The fragment-binding pocket is a few Ångstroms away from the histidine substrate, making linking the two ligands feasible. In preparation for this step, the researchers acetylated the amino group of the most potent fragment. This caused little change in the functional activity, but crystallography revealed that the fragment’s binding orientation had flipped around, such that the acetamide group pointed away from the histidine and towards a cysteine residue. Attempting to turn lemons into lemonade, the researchers added electrophiles to try to interact with the cysteine residue. Some of these molecules had measurable IC₅₀ values (on the order of 1 mM), and crystallography of one of these showed covalent bond formation between the fragment and the targeted cysteine. This cysteine residue is found in the T. cruzi enzyme but not the human version, and indeed these molecules appear to be more active against the parasitic HisRS.

This is a nice example of fragment screening by crystallography that illustrates one of its main challenges: crystallography is capable of detecting extremely weak binders that may prove difficult to advance. Still, the researchers have taken some promising initial steps, and it will be fun to see what they come up with.

Monday Morning Non-Blogging

I know you all come here looking for pithy breakdowns of recent publications in the fragment world.  Today Dan and I are both in Boston at CHI's Discovery on Target conference.  We are teaching our fragment course today, which is really focused on PPIs and Biophysics for this crowd. So, blogging will be limited this week (maybe). 

Fragments vs ERK2

Extracellular-regulated kinase 2 (ERK2) is one of just two known substrates of the kinases MEK1 and MEK2, themselves the subjects of considerable clinical efforts to treat cancer. In a paper just published online in Bioorg. Med. Chem. Lett., Daniel Burdick and colleagues at Genentech describe how they have used FBLD to tackle ERK2.

A library of just 635 fragments was screened against the protein using STD NMR, yielding 54 hits, and SPR, yielding 78 hits. Thirteen of these came up in both assays, and compound 1 had the second-highest ligand efficiency. Not surprisingly, X-ray crystallography revealed that this purine binds to the hinge region of the kinase. The electron density also showed something else binding nearby, which the researchers interpreted as an imidazole molecule left over from the protein purification. Thus, they set out to grow their fragment in this direction.

The purine moiety of compound 1 was not well-suited for growing towards the imidazole, and purines have also been picked over extensively by numerous groups, so the researchers used scaffold-hopping to develop compound 3. This turned out to have acceptable affinity and dramatically improved ligand efficiency. Growing led to compound 14, and structural characterization of a related molecule confirmed that the added heterocycle bound in the same region as the originally observed imidazole.

Next, the researchers grew in a different direction, ultimately leading to compound 39, with low nanomolar potency. Although no cell activity or selectivity data are reported, the authors note that the series underwent further optimization that will be reported in future.

This is a nice, concise description of fragment-based lead discovery and optimization that incorporates multiple biophysical methods, structure-based drug design and modeling, and creative medicinal chemistry. It is not clear whether targeting ERK2 has advantages over MEK or RAF, but work like this is precisely what is needed to generate chemical probes to answer this question.

Is this still a thing? And why?

As regular readers here know, we often discuss metrics because everyone uses them.  Last year, agent provcateur Pete Kenny unleashed a broadside against those who defended metrics.  This seems to be  like the corpse flower that blooms once a year, stinks the place up, yet everyone runs to go see it.  Well, recently Pete posted in the LI group: "Ligand efficiency validated fragment-based design?"and asked whether or not people agreed with the statement.  This of course has inspired a wave of comments.  I disagreed with the statement, but not for the "metrics suck" argument.  I strongly urge people to go read the thread, unless of course you have something better to do.  

To me, this is not about the validity of metrics.  [Let me add here, that I prefer the "LEAN" metric (pIC50/HAC) because it can be done in your head on the fly.]  I think people have a good understanding of what they do, their limitations, and their strengths.  I disagreed with the statement because of the use of the word "validated".  In the development world, we talk about our assays very specifically: they are qualified or validated.  A validated assay is one that has been shown to be accurate, specific, reproducible, and rugged for the analyte in the concentration range to be measured.  Put plainly, this means that if you expect to measure analyte X at 5 uM, you have to show that for all samples it will be measured in you can identify it, measure 5 uM accurately, and do it every time.  That's a validated measurement.  When you are qualifying an assay, the bar is much lower.  An assay is considered qualified if it has been demonstrated to be "fit for purpose".  Fit for purpose means that it will do the job, but you haven't beat the sugar out of it to make sure it is "valid".  To me, ligand efficiency is fit for purpose of driving medchem decisions; it is qualified for that purpose, but not validated (N.B. I am not saying "not valid".) 

Dry solutions for crystallography

To some, X-ray crystallography may be a rather dry topic. However, the process generally entails lots of liquids. In particular, the commonly used practice of crystal soaking entails transferring protein crystals to a new solution containing dissolved ligands, which is both tedious and can cause crystals to shatter or dissolve. A new paper by Jean-Francois Guiçhou at Université de Montpellier and collaborators in Acta Cryst. D aims to streamline the process, and so lower barriers for obtaining structural information that could guide drug design.

Rather than manually transferring crystals to new solutions, the researchers pre-coated crystallization plates with ligands and then grew protein crystals in them. They first dissolved the ligands, transferred these to the wells, and allowed the solvent to evaporate. Although they tested a variety of solvents, including acetone, tetrahydrofuran, ethanol, acetonitrile, 2-propanol, water, and DMSO, only the last two proved suitable; most of the rest wicked up the well, spreading over too large of a surface (though methanol has been used by Beryllium, née Emerald). DMSO is, of course, the most commonly used solvent for storing small molecules, and so should work for most ligands. DMSO is not very volatile, but only 1 µl was used per well, and putting the plates in a fume hood for a week left behind dry ligand.

To make things easier still, the researchers used special crystallization plates that could be put directly into an X-ray beam (in situ crystallography), further diminishing the amount of manipulation required. The technique was tested against four different proteins: the old standard hen egg-white lysozyme and the drug targets cyclophilin D, PPARγ, and Erk-2.

For lysozyme, the water-soluble fragment benzamidine was used, and the resulting structures showed the fragment binding in a similar manner as previously described. So did structures of PPARγ bound with the high affinity ligand rosiglitazone. Cyclophilin, though, was not as successful: of nine fragments attempted, only one produced a structure. In contrast, three fragments produced structures using conventional approaches. ‘Dry’ crystallization was more successful with two more potent (micromolar or better) cyclophilin ligands. Interestingly, dry crystallization succeeded with one ligand that had previously been characterized only by co-crystallization; even week-long soaking experiments had not worked.

Finally, Erk-2 was screened against 14 ligands designed as hinge-binders with low solubility in water. Crystals were obtained with five of the ligands, and four were large enough to generate good-quality structures.

Overall this seems like a convenient approach, though it does seem prone to false negatives. What do the crystallographers out there think – is this a practical solution?

ATAD2 Again...Now with a good tool.

Epigenetics is big.  We keep on beating that drum.  Just to prove it, today's paper is on a target we have talked about before: ATAD2.  That previous paper was unsatisfying: leading to my summary: "if you throw enough fragments at a target you can find a few that bind."  Today's entry  from GSK has produced the first micromolar inhibitors of ATAD2.  

As noted previously, ATAD2 is "undruggable" or at least VERY difficult to find chemical matter against.  To add to the difficulty,  the BET activity needs to be minimized.  With that in mind, they set a high threshold of activity (pIC50 greater than7) and 100 fold selectivity against BRD4 (a representative BET domain).  The ATAD2 site is more polar and flexible than BET.  The authors felt that this would be exploitable to create selective molecules.  To address ATAD2 they started with Ac-K mimics from previous BET work.  They supplemented this with diverse cores not represented.  One such array (which I read as libraries, somebody correct me if I am wrong) was based on the cpd 1,

Cpd 1. 

which is similar to the chemotypes discussed last year.   A crystal structure of 1 was solved, confirming that it bound as expected.  

Additional arrays were made around this core and tested in a TR-FRET assay.  30,000 compounds gave a 0.25% hit rate.  Confirmation was performed by HSQC NMR.  A subset of compounds interacted at the Ac-K site based upon comparison to compounds with known binding modes.  In this case, the peak that shifted upon binding were the same.  I would like to know if this was by visual inspection of spectra or if it was accomplished using PCA, or similar method.  It probably doesn't matter, but intrigues the NMR jock in me.

In rounds of medchem and X-ray confirmation, they were able to drive the potency against ATAD2 to the single digit micromolar.  The ligand efficiencies were maintained right around 0.30. Compound 57 (R=4-Me) and 60 (R=4-OMe) had the "best balance of ATAD2 and BET activity".  These compounds were also active in a cell-based assay known to be sensitive to BET inhibitors.  However, there is no selectivity.  ATAD2/BET pIC50 for 57 was 1.1 and 60 was 1.0. So, despite the selectivity threshold they developed, these compounds are not selective.  Despite that, I think this paper shows that the aphorism Undruggable =Undone is true.

Polypharmacology for Kinases

We've been a roll with epigenetics and PPIs lately.  So, its a nice break when a kinase paper comes out.  But, in keeping with the theme of hard targets, today's paper is about a tyrosine kinase.  We've started to see more and more FBDD on TKs.  One problem is that TKs can acquire resistance to drugs, quickly eliminating their therapeutic usefulness.  One way around this is to use polypharmacology: "optimized inhibitory profiles for critical disease-promoting kinases, including crucial mutant targets."  In this work, they are targeting RET and VEGFR2 dual inhibitor using a in silico/fragment approach.  

Compound design was largely based upon homology modeling the "DFG-out" RET structure utilizing the VEGFR2 structure as a template.  Their Kinase Directed Fragments (KDF) are shown in Figure 1.

Figure 1.

Their fragment design rationale makes some interesting comments.  They state that a "hinge binding" fragment alone can aggregate at high concentrations needed to achieve activity in a biochemical screen.  So, their fragments have an additional moiety that interact with the lipophilic or ribose pocket.  

Accordingly, KDFs have larger molecular weights and are generally more active than the fragments contained in traditional libraries, permitting screening in the micromolar range.

I would say the first statement is conjecture and the second untrue.  17 heavy atoms is squarely in the regime of what people consider "fragment" sized.  I think instead the authors are using the wrong tool for the job.  Using a biochemical screen to find fragment actives is akin to hammering a nail with a screwdriver.  Sure, you can do it, but why would you?  

Rather expectedly, they identified compound 1 as a promising starting platform.  Of course, the criteria for selecting this compound are kept highly secret.  It did "effectively" inhibit RET at 100 (63%) and 20 uM (28%) in the presence of 190 uM ATP [Km for RET 12uM].  It had VEGFR2 activity of 59% at 100 uM. 

Add caption

Modeling allowed them to generate the compounds showed in Figure 2.

Figure 2. 

Pz-1 had activity less than 1 nM against RET, RET(V804M/L)[a gatekeeper mutant],  and VEGFR2.  This equipotency was also demonstarted in cell-based assays.  Against a panel of 91 other kinases at 50 nM, Pz-1 had significant activity against 7 others (TRKB, TRKC, GKA, FYN, SRC, TAK1, MUSK).  So, in the end using primarily modeling and a biochemical assay they were able to generate a polypharmacological TK inhibitor.  I leave it to those more well versed in the biology whether or not those 7 other kinases pose a potential problem.  I however would argue that they generated an agent with polypharmacology against 9 kinases not 2.