Lead Generation Strategies

Many biomolecular interactions form very long-lived complexes. These can be extremely challenging to measure leading to unfavorably long incubation times in biochemical/biophysical assays and inaccurate measurements. SPR has the advantage of measuring kinetics with high accuracy, the “chaser” method extends the range of measurable half-lives with high accuracy but are low throughput. We present adjustments to the chaser assay to improve throughput and its application to measuring protein interactions.

Direct-to-biology (D2B) platforms enable efficient hit identification and optimisation through the screening of crude reaction mixtures. They are especially well-poised for covalent drug discovery, where MS-based screening helps to untangle on-target binding from assay interference, reducing the risk of false positives. The power of a D2B approach will be illustrated for the identification and optimisation of covalent inhibitors of WRN helicase, a challenging therapeutic target implicated in microsatellite-unstable cancers.

PARP-mediated poly(ADP-ribose) formation recruits and organises DNA repair proteins, while PARG-mediated PAR degradation terminates repair and preserves genomic integrity. Guided by x-ray structures of PDD00017272 bound to PARG, we pursued covalent inhibition of nearby Cys872 to address the challenging catalytic pocket and long PARG half-life. Structure-guided hydrazide acrylamide inhibitors showed biochemical activity but limited cellular potency and plasma stability. Subsequent scaffold hopping, new exit-vector design, and incorporation of a bicyclo[1.1.0]butane carboxamide warhead yielded FORX-425, a potent covalent PARG inhibitor with strong cellular activity, high plasma stability, and promising mouse pharmacokinetics, highlighting BCB warheads for challenging pockets.

The kinase MPS1 is a pivotal component in the cell division process and reversible inhibitors have been evaluated in clinical trials for the treatment of aggressive types of solid tumours. This talk will cover the development and comprehensive structural, biochemical, and cellular characterisation of the first MPS1 covalent inhibitor. This proof of covalent ligandability alongside promising biological results will guide future drug discovery efforts towards MPS1-targeted cancer therapies.

The basic helix-loop-helix leucine zipper (bHLH-LZ) transcription factor MITF is a melanoma oncogene, but considered undruggable as direct targeting is unprecedented. We have identified fragments binding to the DNA binding domain of MITF and optimised them to sub-micromolar ligands by fragment merging. Detailed structural and biophysical validation will be presented. NMR studies and MD simulations indicate that the interconversion of kinked and straight helices is slowed down by compound binding.

Based on our experience with fragment-based drug discovery, we have established Frag2Drug—a new open-access, state-of-the-art drug discovery facility that integrates biophysical screening and protein crystallisation. Using surface plasmon resonance and high-throughput x-ray crystallography, Frag2Drug will accelerate the discovery of drug leads and pharmacological tools for unmet medical needs, serving both academic and industrial users. In this talk, we will present past work and the capabilities of our platform.

Quantifying ternary complex formation, essential for targeted protein degradation by PROTACs and molecular glues, remains challenging. We here present a proximity binding assay that simultaneously measures binary and ternary interaction kinetics using a biosensor. A Y-shaped DNA scaffold brings target proteins and E3 ligase receptors into proximity, enabling detection of ternary complexes by FRET and binary binding by fluorescence quenching, supporting real-time analysis with minimal sample consumption.

Molecular glues that form inhibitory ternary complexes between FKBP12 or CYPA and target proteins (rapamycin/FK506 and daraxonrasib/cyclosporine) have been used clinically. However, the systematic discovery of molecular glues against proteins of interest has been limited. We constructed and screened a multi-million compound DEL, biased to bind and resurface FKBP12, against a set of 25 purified protein targets. Hits were characterised in biophysical and cellular assays with the targets BRD9 and QDPR.