Three dimensional fragments revisited

A long-running debate in the fragment world centers on the utility of “three dimensional” fragments. Proponents argue that these (often aliphatic) fragments may be more novel, have better physicochemical properties, and have more vectors for elaboration than “flatter” (mostly aromatic) molecules. Skeptics retort that hit rates are likely to be lower for these more complex molecules, and good luck making analogs. Two papers published late last year add more data to the debate.

The first paper, published in J. Med. Chem. by William Pomerantz and collaborators at the University of Minnesota and Eli Lilly, describes the results of a fragment screen against the bromodomain BRD4(D1), a popular member of the BET family. The 467 fragment library was enriched for shapely fragments as assessed by plane of best fit (PBF), which is the “average distance of a non-hydrogen atom from a plane drawn through the compound such as to minimize the average.” For example, "flat" benzene has a PBF of 0 while the cofactor NADPH has a PBF of 1.53.

The library was screened using ligand-observed (CPMG) NMR, and 34 hits were confirmed using protein-observed fluorine (PrOF) NMR. All of these were competitive with the known ligand (+)-JQ1, consistent with binding at the acetylated lysine recognition site. The average PBF of the hits was 0.44, essentially the same as the library itself (0.46). This is higher than the average PBF (0.36) of all fragments crystallized with BRD4 in the protein data bank.

Structures of all the hits are provided, and some of them are indeed quite unusual. The researchers characterized a substituted thiazepane crystallographically and were able to optimize this to a 32 µM binder with good ligand efficiency. This fragment was also selective against a handful of other bromodomains.

The researchers had previously screened BRD4(D1) under identical conditions with a more traditional, “flatter” library with an average PBF of 0.26. Interestingly, in that case the hits were less shapely than the library as a whole, with an average PBF of 0.17. The confirmed hit rate was also higher: 20% vs 7%. That said, the fragments in the traditional library tended to be smaller (averaging 180 Da vs 241 Da), so the molecular complexity of this library was likely to be lower, which could account for the higher hit rate.

The second paper, published in Bioorg. Med. Chem. Lett. by Ulrich Grädler and collaborators at Merck KGaA, EMD Serono, Edelris, and Proteros, focuses on cyclophilin D (CypD), which has been implicated in cardiovascular disease and multiple sclerosis. Unlike BRD4, this is a tough target: an HTS screen of 650,000 compounds in a biochemical assay yielded just 178 hits, none of which confirmed. Undeterred, the researchers screened 2688 fragments by SPR at 2 mM, resulting in 58 confirmed hits, all quite weak (millimolar). Crystallography was attempted on most of them, yielding six structures, including such shapely specimens as compounds 3 and 7.

Compound 3 binds in the lipophilic S2 pocket of CypD, overlapping with the aniline moiety of previously reported compound 2. Fragment merging led to compound 14, with nearly 40-fold improved affinity over compound 2. A similar strategy merging compound 3 with fragment 8 led to low micromolar compound 27, two orders of magnitude more potent than the starting fragments. Perhaps most impressively, fragment linking compound 3 with compound 7, a shapely fragment which binds in the S1’ pocket, led to submicromolar compound 39, with affinity more than 10,000-fold higher than either fragment.

So in the end, fanciers of shapely fragments and detractors alike can feel vindicated by these papers. Hit rates might be lower for three dimensional fragments, but the resulting hits are likely to be less precedented. In the case of CypD, a shapely fragment led to three different series for a target that had resisted HTS. Of course, there is still some way to go: no cell, permeability, or stability data are provided for any of the molecules, and medicinal chemists may blanch at the seven stereocenters in compound 39. But these are interesting starting points, and it will be fun to see where they end up.

Fragments in the clinic: AMG 510

Few cancer targets are as prominent as KRAS, a molecular switch that has been called “the beating heart of cancer.” Mutations that cause the switch to be stuck in the “on” state occur in roughly a quarter of human tumors. The protein’s role in driving cancer has been known for nearly forty years, but for most of that time it has been considered undruggable.

Until now.

When bound to GTP, KRAS is turned on, promoting cell proliferation until the GTP is hydrolyzed to GDP. Activating mutants impair this hydrolysis. An obvious approach to targeting KRAS would be to develop small molecules that bind in the nucleotide-binding pocket, as has been done successfully for kinases. Unfortunately, the extremely high affinity of KRAS for GTP, along with the high intracellular concentration of GTP, makes this impossible.

One activating mutant replaces a glycine with cysteine (G12C), providing a convenient handle for covalent inhibitors. In 2013, Kevan Shokat and colleagues at University of California San Francisco described in Nature how they used Tethering to identify fragments such as 6H05 that bound to this cysteine through a disulfide bond. Because disulfides are not stable in cells, they replaced this “warhead” with an acrylamide moiety and made a few other tweaks to arrive at compound 12. Extensive optimization described by Yi Liu and collaborators at Kura, Wellspring, and Janssen in a 2018 Cell paper led to molecules such as ARS-1620, which showed potent biochemical and cell-based activity and promising results in mouse xenograft models.

In 2014, Carmot Therapeutics began a collaboration with Amgen to discover covalent inhibitors of the G12C mutant of KRAS. Carmot’s technology, Chemotype Evolution, entails rapid synthesis and testing of large libraries around an existing molecule such as a fragment (see here for a nice animation). In this case, fragments chosen included simple acrylamides, the thought being that – unlike the disulfides in Tethering – these could be carried through into the final molecule.

Initial hit 2 was truncated and cyclized to compound 4, which had comparable activity but with fewer atoms and rotatable bonds. Additional iterations of Chemotype Evolution and medicinal chemistry ultimately led to compound 1, with nanomolar activity in cells containing KRAS^G12C. Crystallography revealed that the tetrahydroisoquinoline moiety bound in a previously cryptic pocket formed by movement of a histidine side chain. These interactions contributed to the high affinity of the molecule; more details are provided in a paper published last year in ACS Med. Chem. Lett. by Victor Cee, myself, and our collaborators at Amgen and Carmot.

Unfortunately, although compound 1 was more potent in cell assays than ARS-1620, it had low oral bioavailability and rapid clearance in mice and rats. However, as described in J. Med. Chem., Brian Lanman and colleagues at Amgen superimposed ARS-1620 with compound 1 and realized that it would be possible to access the cryptic pocket from the former compound. This strategy proved successful, ultimately leading to AMG510, which entered the clinic in August 2018. And although it is still early, a paper in Nature by Jude Canon, J. Russell Lipford, and collaborators at Amgen and elsewhere describes promising responses to the drug by a handful of patients with non-small-cell lung carcinoma.

Amgen is not the only company to have capitalized on the Tethering results: Mirati used the information to develop their clinical-phase MRTX849. And Janssen has also entered the clinic with JNJ-74699157.

There are multiple lessons here. First, as we’ve seen previously, a single fragment can lead to multiple clinical compounds. Second, progress often requires considerable changes to the initial fragment. This story is clearly a case of fragment-assisted drug discovery, and in the interest of space I’ve had to omit most of the lovely medicinal chemistry, not to mention biology and biophysics, detailed in these five papers.

Another lesson is that covalent fragments can enable lead discovery for targets not accessible through other means. But this enablement may require non-conventional molecules; the initial fragments violate the rule of three, and the disclosed clinical compounds have molecular weights in excess of 500 Da. Despite these challenges, remarkably rapid progress is possible: less than five years elapsed between the first Nature publication and the entry of AMG 510 into the clinic.

Most important, this research has led to possibly life-extending molecules – one of the responders described in the second Nature paper had been on drug for 42 weeks. Practical Fragments wishes her or him, and everyone else involved with the trials, the best of luck.

FBLD by Cryo-EM

X-ray crystallography is the most popular fragment-finding method according to our latest poll. This is in part due to the previously unmatched level of detail a crystal structure can provide, and in part due to the increasing speed and automation for data collection and processing. But many proteins can never be coaxed into crystals, and while it is possible to advance fragments in the absence of structure, it is rarely easy.

In the past few years, cryogenic electron microscopy (cryo-EM) has come to rival crystallography in terms of resolution (see here for details). The first mention of the technique on Practical Fragments was in 2013, when Teddy wrote that he “could not figure out how you would use [it] in screening/FBHG. However, the point of emerging technology is to emerge…” And emerge it has. At FBLD 2018 researchers from Astex presented the first cryo-EM structure of a bound fragment, and in a new paper in Drug Disc. Today Harren Jhoti and collaborators at Astex and Isohelio provide details for two proteins.

The first target, β-galactosidase (Bgal), is a model protein that has previously been characterized by cryo-EM. The researchers solved the bound structures of three small molecules, two of them fragment-sized, to resolutions of 2.2-2.3 Å. The quality of the maps is such that they could easily be mistaken for crystallographic data: the density is clear and includes ordered water molecules. Induced conformational changes are evident, stereochemistry is unambiguous for all the ligands, and one piperidine ring even shows a hole in the middle. Even more impressive, the structures were solved using automated software.

But three structures do not a screen make. For this, the researchers chose the oncology target pyruvate kinase 2 (PKM2). They designed a small library of 68 highly soluble fragments and screened these at 5 mM. The structures of two complexes are shown, and while the resolution is lower than for Bgal, clear contacts with the protein are evident. Next, the researchers screened cocktails of four fragments, each at 25 mM. Fragments in mixtures were chosen to have diverse shapes, and two structures are shown demonstrating that the technique can distinguish the binders.

Throughput is still an issue: each round of data collection and processing – whether individual compound or cocktail – took about a day. However, the researchers expect that improvements in software and hardware should enable 400 fragments to be screened in less than a month. Indeed, given that much of the focus of cryo-EM has been on generating novel structures, one can imagine various shortcuts for screening, such as collecting smaller numbers of images to look for any evidence of bound ligands. I wonder too if something like PanDDA could be developed for cryo-EM.

Both proteins described have been previously characterized by crystallography, but these are still early days, so it is only a matter of time before we see fragment structures in proteins that haven’t been crystallized. That will open thousands of proteins heretofore inaccessible for structure-based design. The researchers conclude by predicting that “cryo-EM will have a transformative impact on the pharmaceutical industry in the near future.” I would not bet against this.

Fragment events in 2020

Updated as of 24 May - we'll provide another update in early July.

June 2-4 CANCELLED: CHI is launching an inaugural Expanding Chemical Space program as part of their World Pharma Week in Boston, and it looks like there will be several talks on FBDD. Update: although this event has been canceled, some of the talks will be moved to Discovery on Target, which is also schedule to be held in Boston Sept 16-18. You can read impressions of last year's event here and 2018 here. 

June 15-17 Postponed to March 10-12, 2021:  Although not exclusively fragment-focused, the Eighth NovAliX Conference on Biophysics in Drug Discovery will have lots of relevant talks, and returns to Boston this year. You can read my impressions of the 2018 event here, the 2017 Strasbourg event here, and Teddy's impressions of the 2013 event here, here, and here.

April 13-17 August 24-28: CHI’s Fifteenth Annual Fragment-Based Drug Discovery, the longest-running fragment event, will be held in San Diego VIRTUALLY April 14-15 August 25-26. This is part of the larger Drug Discovery Chemistry meeting. You can read impressions of the 2019 meeting here, the 2018 meeting here, the 2017 meeting here, the 2016 meeting here; the 2015 meeting here, here, and here; the 2014 meeting here and here; the 2013 meeting here and here; the 2012 meeting here; the 2011 meeting here; and 2010 here.

September 20-23 CANCELLED: FBLD 2020 will be held for the first time in the original Cambridge (UK). This will mark the eighth in an illustrious series of conferences organized by scientists for scientists. You can read impressions of FBLD 2018, FBLD 2016, FBLD 2014,  FBLD 2012, FBLD 2010, and FBLD 2009. Please watch this space; the conference may be rescheduled next year.

December 17-18: What better way to close the year than in Hawaii? The second Pacifichem Symposium devoted to fragments will be held in Honolulu. Pacifichem conferences are held every 5 years and are designed to bring together scientists from Pacific Rim countries including Australia, Canada, China, Japan, Korea, New Zealand, and the US. Here are my impressions of the 2015 event. Abstract submission is open but closes June 1.

Know of anything else? Please leave a comment or let us know!