FBLD 2024

The FBLD meetings have always been calendar highlights. Starting in 2008, before Practical Fragments even existed, they have graced cities around the world in 2009, 2010, 2012, 2014, 2016, and 2018. The plan was for 2020 to be held in Cambridge, UK, but for obvious reasons that didn’t happen. Last week, Boston hosted a triumphant return of the event. With more than 30 talks and dozens of posters I’ll just touch on a few major themes.

 

Crystallography

High-throughput crystallography was prevalent, as befits its growing role in fragment finding. (If you haven’t yet voted in our methods poll on the right side of the page please do so!) Debanu Das (XPose Therapeutics) described how crystallographic screens of just a few hundred fragments identified hits against DNA-damage response proteins such as APE1; these have been advanced to high-nanomolar inhibitors with cell activity. And Andreas Pica described the ALPX platform that enabled screening >4000 hits from an HTS screen against PDEδ resulting in >500 structures.

 

The Diamond Light Source was a pioneer in developing high-throughput crystallography methods, and several speakers described continued progress. Blake Balcomb noted that since 2015 they have collected >240,000 datasets and identified >30,000 ligands. Of these, some 3750 have been deposited into the Protein Data Bank.

 

A crystallographic fragment hit is just the start, and Frank von Delft emphasized that “fragment progression is neither fast nor cheap.” His goal is to take a 100 µM binder to a 10 nM lead in less than a week for less than £1000. Toward this end he and his team are using rapid chemical synthesis and crude reaction screening along with various computational approaches and crowd-sourced science. The COVID Moonshot, which we wrote about here, is one model, and Diamond is trying to create a “Moonshot factory” to pursue other viral targets.

 

Computational Approaches

Computational methods are potentially the least expensive fragment-to-lead method, and these were well represented. One challenge is screening the massive chemical space represented by make-on-demand libraries, and Pat Walters (Relay) described how this can be done using Thompson Sampling, an active-learning method that traces its origins to 1933. Applied to lead discovery, the method involves breaking larger molecules into component fragments and iteratively searching for better binders. Pat showed that searching just 0.1% of a library of 335 million molecules consistently found 90% of the best hits.

 

Most computational methods rely on experimental data, and over the past 25 years Astex has generated >100 crystal structures on each of more than 40 targets, with >6600 bound fragments in total. Paul Mortenson described how these are being used to develop generative models, with chemists providing feedback on suggested molecules.

 

Artificial intelligence is the centerpiece of Isomorphic Labs, which has unfettered access to AlphaFold 3. Rebecca Paul described an example starting from a literature fragment in which the predicted affinities matched well with experiment – and the molecules were considerably more potent than those suggested by an experienced medicinal chemist.

 

Recognizing the need for experimental affinity data for fragments, Isomorphic worked with Arctoris to screen 5420 fragments against 65 kinases covering the diversity of the kinome. After carefully curating the data, including rescreening the actives at a different CRO, they found 485 fragments with an IC₅₀ of 300 µM or better. Interestingly, only about half of these fragments are known kinase binders.

 

Sandor Vajda (Boston University) suggested there may be limitations to machine learning models. He found that using AlphaFold 2 to find cryptic pockets was dependent on their representation in the PDB, with rare experimental states not being predicted. Sandor also proposed an interesting hypothesis that cryptic pockets created only by the movement of side chains are not very ligandable because the side chains move on such a rapid time scale that they effectively act as competitive inhibitors to ligands.

 

Success Stories

No FBLD meeting would be complete without success stories, and FBLD 2024 was no exception. Chaohong Sun noted that nearly 80% of the targets at AbbVie taken into fragment-based screening are novel. Of these, more than 80% yield actionable hits, though 44% are not pursued for a variety of reasons, including finding hits from other sources, hits at novel sites with no obvious function, and changes to the portfolio. Chaohong described a series of STING agonists that was taken forward to low nanomolar leads with in vivo activity.

 

Michelle Arkin (UCSF) described progress on creating molecular glues to link 14-3-3 proteins to the estrogen receptor, which we last wrote about here. Covalent binders to the 14-3-3 protein stabilize the interaction with ERα by more than 100-fold and show activity in cancer cell models.

 

Multiple talks focused on SARS-CoV-2 targets. Ashley Taylor (Vanderbilt) described fragment screens against the papain-like protease PL^Pro that led to both covalent and non-covalent inhibitors. James Fraser (UCSF) described how a massive crystallographic screen against the Nsp3 macrodomain Mac1 led to high nanomolar compounds, which we wrote about here. And Adam Renslo (UCSF) discussed the further optimization of Mac1 inhibitors to yield molecules that could protect mice from a fatal challenge of the virus.

 

A drawback of pursuing novel targets is that sometimes the biology proves uncooperative. Andrew Woodhead described a successful fragment screen at Astex against the oncology target elF4E that led to mid-nanomolar binders that could disrupt the protein-protein interaction with eIF4G in cells. Surprisingly, these molecules had no effect on cell viability, and a series of mutational and targeted-protein degradation experiments suggested that blocking a larger region of the protein-protein binding site might be necessary.

 

Drugs are the ultimate success stories, as David Rees reminded participants in “25 years of thinking small.” In addition to providing an overview of FBLD at Astex, David added up the sales of all seven FDA-approved fragment-derived drugs, which totals more than $3 billion. Harder to quantify—though infinitely more valuable—are the added years of life for patients with once-untreatable cancers. These numbers will only grow as the dozens of fragment-derived molecules in the clinic continue to advance.

 

I’ll close on that note. If you missed FBLD 2024, you’ll have another chance next year: FBLD 2025 is planned for Cambridge (UK) September 21-24 next year. Barring global pandemics.

New poll: structural information needed for F2L and fragment-finding methods

With elections taking place around the world, Practical Fragments is getting into the action. Our new poll revisits two questions from past years to see how things have changed.

 

Our first question asks, “how much structural information do you need to begin optimizing a fragment?” When we ran this poll back in 2017 a third of respondents needed crystallography to begin a fragment-to-lead campaign, while only a quarter would move forward with SAR only. But when Wolfgang Jahnke, Ben Davis, and I published a review in 2018 about advancing fragments in the absence of crystal structures, we found an abundance of approaches. It will be interesting to see whether these numbers have shifted.

 

Our second question asks what method(s) you use to find and validate fragments. For consistency with previous polls please click every method you use, whether as a primary screening technique or for validation. Please note too that we’ve added cryo-electron microscopy. You can read about these methods below, and if you select “other” please describe in the comments.

 

• Affinity chromatography, AS-MS, capillary electrophoresis, or ultrafiltration
• BLI (biolayer interferometry)
• Computational screening
• Cryo-EM (cryogenic electron microscopy)
  
• Functional screening (high concentration biochemical, FRET, cell-based, etc.)
• ITC (isothermal titration calorimetry)
• Literature or known fragments
• MS (mass spectrometry, native or covalent)
• MST (microscale thermophoresis)
• NMR – ligand detected
• NMR – protein detected
• SPR (surface plasmon resonance)
• Thermal shift (or DSF)
• X-ray crystallography
• Other

 

Please vote on the right hand side of the page; click the vote button for each question. (If you don’t see the poll you may need to (1) turn off private browsing, since the free Crowdsignal version we use for the blog cannot support surveys in this mode or (2) view web version on your phone.)

Casting light on target-guided synthesis

Target-guided synthesis, in which a protein templates the formation of its own inhibitor, is a concept first proposed decades ago. There are roughly two flavors. Dynamic combinatorial chemistry (DCC) involves reversible formation of the product, and we wrote in 2017 about some of the challenges. Kinetic target-guided synthesis (KTGS) involves irreversible chemistry, for which the options are limited. The classical click chemistry azide-alkyne cycloaddition is so slow that reactions usually take days, which can be a problem for delicate proteins. A recent (open-access) paper in Angew. Chem. Int. Ed. by Cyrille Sabot et al. describes a bright way to accelerate things.

 

The researchers turned to photochemistry, specifically diazirine chemistry. Illuminating 3-trifluoromethyl-3-phenyldiazirines leads to loss of nitrogen and formation of highly reactive carbenes. The carbenes are so hot that they can react indiscriminately with proteins, as we described here. However, the reaction with thiols is faster than the reaction with other functional groups on proteins, so the researchers reasoned that a library of thiols could out-compete the protein.

 

The carbonic anhydrase bCA-II was chosen as a model protein. Sulfonamide-containing molecules such as compound 5 are known to be good inhibitors. This “anchor” molecule was incubated at 60 µM with seven different diazirines, each at 400 µM, in the presence or absence of 30 µM bCA-II and then irradiated with 365 nM light for a few minutes. Most of the reactions produced similar amounts of product in the presence or absence of bCA-II, but compound 1b yielded about threefold more of compound 2d in the presence of bCA-II, suggesting the reaction was being templated by the protein. 

 

Control experiments lend credence to this hypothesis. First, adding a known competitive bCA-II inhibitor reduced the formation of compound 2d to background levels. Second, other proteins did not cause a similar enhancement in the formation of compound 2d. Finally, conducting the experiment with phenylmethanethiol (ie, a variant of compound 5 lacking the sulfonamide moiety essential for interaction with bCA-II) did not cause an enrichment of the photochemical product in the presence of the enzyme.

 

Chiral HPLC was used to show that compound 2d was slightly enriched for the (R)-enantiomer, with an enantiomeric excess of around 10%, when the reaction was conducted in the presence of bCA-II but not in the absence. The two enantiomers were synthesized and tested, and the (R) form did indeed have slightly better activity (300 nM vs 330 nM).

 

This is a thoughtful, well-conducted investigation. But it makes me even less sanguine about the practicality of KTGS for finding new chemical matter, for several reasons. First, the efficiency of the reaction is poor: the researchers calculate the yield of compound 2d at around 1% of the enzyme concentration, so low that they used single-ion monitoring (SIM) mass spectrometry to detect it. Because of this low efficiency, the concentration of enzyme used needs to be quite high.

 

The most serious strike against KTGS is the fact that all of the diazirines generated potent (sub-micromolar) inhibitors. One of them was even slightly better than compound 2d but did not show enrichment in the presence of bCA-II. False negatives seem to be a major problem, as we’ve written previously.

 

One caveat to my caveats is that compound 2d is only marginally more potent than the starting compound 5. NMR experiments conducted with diazirine 1b suggest binding to the protein, though the affinity was not quantified. Perhaps a different fragment linking system, in which both fragments have measurable affinity for the target, would be better suited to demonstrate the utility of KTGS. For now, this paper does a nice job highlighting its drawbacks.

Fragments vs herpesviridae

The name herpes makes most people think of painful ulcers in the mouth, or worse. But herpesviruses are actually a family of viruses that can also cause chicken pox, mononucleosis, and other diseases. Some 95% of adults are infected by at least one type of herpesvirus, and these can become deadly if people become immunocompromised, such as during an organ transplant. A drug that would inhibit all forms of herpesviruses would be useful, and the first steps are described in a recent ACS Med. Chem. Lett. paper by Michael Plotkin and colleagues at Merck.

 

The details of the primary screen are sparse, though the researchers did say they physically screened more than 100,000 compounds to identify molecules such as compound 5, a modest inhibitor of the DNA polymerases from both cytomegalovirus (CMV) and varicella zoster virus (VZV). (For most compounds the paper reports biochemical activity towards both of these polymerases as well as antiviral activity for CMV, VZV, herpes simplex virus 1 (HSV-1), and HSV-2, but for simplicity I’ll only show data for CMV here. The compounds generally have comparable activity towards different viruses.)

 

Hydrogen bond acceptors such as the ketone in compound 5 were found to be essential for activity, and exploring a variety of analogs led to compound 12, which in addition to submicromolar biochemical activity against the DNA polymerases also showed antiviral activity against CMV and other herpesviruses.

 

The paper goes into considerable detail on the lead optimization. The (S) enantiomer of compound 12 was an order of magnitude more potent than the (R) enantiomer. Modifications made to both of the phenyl rings ultimately led to compound 44, with low nanomolar biochemical activity against the polymerases and sub-micromolar antiviral activity against CMV, VZV, HSV-1, and HSV-2. Importantly, the researchers note that they did not have crystal structures during optimization, a useful reminder that structural information is not always necessary.

 

Compound 44 had modest oral bioavailability in rodents, but closely related compound 42 containing a trifluoromethyl group in place of the bromine was better, albeit with slightly lower biochemical potency. This molecule led to high survival rates in mice when dosed either before or after being exposed to HSV-1. In separate studies, the compound reduced CMV viral load. For both HSV-1 and CMV compound 42 compared favorably to acyclovir and ganciclovir, two commonly used drugs.

 

Although there is still some way to go to a drug, the researchers end by promising to describe “further progress of this series.” I look forward to reading about this.

Fragments in Brazil

Most of the fragment events we’ve highlighted are in the US, Europe, and Australia, but that does not fully reflect where all the good science is happening. In a recent ACS Med. Chem. Lett. paper, Carolina Horta Andrade, Maria Cristina Nonato, and Flavio da Silva Emery introduce CRAFT: the Center for Research and Advancement in Fragments and molecular Targets.

 

Established in 2021, CRAFT is a collaboration between the University of Saõ Paulo and the Federal University of Goiás. The center is focused on endemic diseases of Brazil. As the researchers note, only one of the 60 or so fragment-derived drugs that have entered the clinic is an anti-infective, so there is clearly significant need. CRAFT also has an educational and training component reminiscent of the European FragNet and the Australian Centre for Fragment-Based Design.

 

One focus of CRAFT is fragment library design, including underexplored heterocyclic systems. Importantly, the researchers are investigating new synthetic methodologies to be able to functionalize different regions of the fragments. They are also exploring fragments similar to or derived from natural products.

 

Targets are of course essential, and CRAFT is investing in protein production and characterization, such as the enzyme DHODH from Leishmania; we’ve written recently about a fragment approach to the mammalian counterpart.

 

Finally, CRAFT is investing in structure-based design, ligand-based design, and phenotypic screening. And in 2024 no venture would be complete without use of machine learning.

 

Academic laboratories often struggle with downstream drug discovery efforts such as drug metabolism and pharmacokinetics. CRAFT recognizes this and has partnered with the Welcome Centre for Anti-Infectives Research to train participants in DMPK.

 

The researchers “invite the global scientific community to collaborate with us in addressing neglected diseases.” I hope they succeed. Five years ago we highlighted the consortium Open Source Antibiotics, but that site seems to be updated infrequently. The COVID Moonshot has been more successful but is arguably less urgent given the billions of dollars of industry money that poured into research on SARS-CoV-2. From an ethical perspective society should invest more on combating tropical diseases. And as the planet warms, these diseases will increasingly move out of the tropics.