Myeloid cell leukemia 1 (Mcl-1) is a member of the Bcl-2 family of proteins that is overexpressed and amplified in a wide variety of cancers and causes resistance to many chemotherapies. Although a very promising cancer target, it exerts its activity through protein-protein interactions and is very challenging to drugs with small molecules. We have discovered picomolar Mcl-1 inhibitors for the treatment of a variety of cancers.

In the last few years, drug discovery was revolutionized by the abundance of target 3D structures and the development of ultra-large libraries of drug-like compounds.  Inspired by the fragment screening concept, we have developed V-SYNTHES, a new iterative synthon-based approach for fast structure-based virtual screening of billions of readily available (REAL) compounds (Sadybekov et al. Nature 2022). I discuss the latest developments and applications of V-SYNTHES technology to lead discovery.

We present a novel approach built upon Absolute Free Energy Perturbation, to screen > 25 Million fragments virtually and identify a small number of 25-50 fragments for experimental testing. We present a first proof of concept in an active drug discovery project for an allosteric kinase inhibitor, where it yielded hit rates > 50% with multiple hits at Ligand Efficiency of ~0.4 or better and single digit µM potency. We also present retrospective analysis across different target classes with known fragment hits in the literature.

The advantages of WAC for FBS are the detection of weak binders by screening fragments at low concentrations (<5 μM) and immediate ranking of hits. Here we present the results of a WAC screen towards SMARCA4 as well as subsequent hit validation and expansion efforts.

A multifaceted screening approach, including FBLD, was undertaken to identify inhibitors of the progranulin-sortilin interaction as a possible treatment for neurodegeneration. In this presentation, we will describe our efforts to evolve two hit series using SBDD focusing on the maintenance of ligand efficiency while improving potency. We will also highlight the implementation of high throughput experimentation in iterative cycles to produce a potent series of inhibitors with good CNS properties.

We have developed a proteomics platform to guide the construction of a screening library containing small molecules armed with covalent reactive groups. From these efforts, we have discovered inhibitors of Janus kinase 1 (JAK1) that potently and selectively inhibit JAK1 mediated cytokine signaling through a novel allosteric pocket on the pseudokinase domain, distinct from the pocket bound by recent clinical-stage inhibitors of the related kinase Tyk2.

We present Chemical Space Docking, a novel structure-based technique that efficiently facilitates exploration of libraries with billions of molecules and beyond. This virtual screening method combines two advances: (1) avoids full library enumeration; (2) leverages protein structural information by evaluating products with molecular docking. With chemical space docking we identified inhibitors of ROCK1 kinase from almost one billion commercially available, synthesis-on-demand, compounds. Our virtual screening hit rate was 39%.

The Methylthioadenosine phosphorylase (MTAP)-encoding gene is co-deleted with p16/CDKN2a in ~10% of human cancers, leading to elevated levels of the MTAP substrate methylthioadenosine (MTA) in these cancers. MTA binds Protein Arginine N-Methyl Transferase (PRMT5) to form a PRMT5/MTA complex. We describe biophysics-aided discovery of a 4-(aminomethyl)phthalazin-1(2H)-one fragment and its evolution into MRTX1719, a clinical-stage, selective inhibitor of the PRMT5/MTA complex as a potential precision medicine for treating MTAP-deleted tumors.

The discovery and development of Asciminib (ABL001) will be described. Fragment screening hits were optimized, and an NMR-based conformational assay was set up to understand the conformational changes that are required for allosteric ligands to inhibit the kinase. Using X-ray crystallography, structure-based design, and medicinal chemistry, allosteric ABL inhibitors were optimized to a clinical candidate, ABL001, which recently received FDA approval for the treatment of chronic myeloid leukemia (CML).

In this presentation, a previously undisclosed and functionally-relevant binding site near the C-terminus of IL17A will be described that was identified via a 2D NMR screen.   X-ray feedback informed structure-based design to enhance affinity through linking and fragment merging.  Leads that were identified display ~40 nM binding affinity in biophysical assays (SPR, ITC) and functional inhibition of IL17A-induced activation of cellular signaling.

Biophysical methods are instrumental for characterization of ligand-target interactions to maximize the success of structure determination by EM and X-ray facilitating the design of optimized molecules. leadXpro implemented grating-coupled interferometry (GCI) to measure affinity, stoichiometry and kinetic parameters.

In this talk, case studies of small molecule-GPCR (e.g. CCR2, A₂B) and -ion channel (e.g. TRPV4) interaction studies are presented as well as binding studies of synthetic nanobodies raised against membrane protein targets.

The University of California Drug Discovery Consortium (UCDDC) strives to leverage the biomedical research and commercialization strengths of the University of California (UC) system to accelerate the development of life-saving therapies and to translate discoveries into knowledge driven commercial enterprises that stimulate California’s economy. The UCDDC Executive Committee works with each UC-campus to advance drug development by providing expertise in drug discovery and by building relationships between industry and academics.

Combining the use of pharmacophores with the theory of protein hotspots we developed a novel design protocol for fragment libraries. The SpotXplorer approach compiles small fragment libraries that maximize the coverage of experimentally confirmed binding pharmacophores at the most preferred hotspots of the target proteins. The efficiency of this approach is demonstrated with a pilot library screened against traditional and challenging targets. SpotXplorer screening retrieved an average of 70% of known pharmacophores for popular target classes and provided hits against recently established challenging targets including SARS-CoV-2 proteins.

Today fragment screening and GPCRs is still a very challenging combination. It requires large investments in the assay development and heavy modifications of the GPCR target and most of the times is not a winning combination. Here single molecule microscopy based dISA offers a unique opportunity not only to screen fragments against wildtype GPCRs but also to shorten and simplify the assay development dramatically.

I will introduce some new approaches to hit identification and optimisation which combine the sensitivity of DNA-encoded libraries with the power of fragments to sample chemical space.  The approach will be demonstrated with some model studies on a conventional target (PAK4 kinase) and a previously undrugged and challenging target, 2-epimerase.

NMR provides high-quality data for the detection/validation of hits. However, its poor sensitivity slows down the throughput and makes it require an important amount of material. Furthermore, a state-of-the-art NMR saturation transfer diffusion experiment only detects weak binders at slow throughput (50-100 cpd/d). We developed a technology using sensitivity enhanced NMR able to screen up to several thousands of compounds per day, at low micromolar concentration, and able to identify weak and strong binders. We show how such a performance improvement is possible and their expected impact.

Crystallographic fragments screening now routinely delivers hits along with the 3D-structural information of the fragment's binding mode. Via our Frag4Lead workflow, the 3D-information can be used as anchor for virtual pre-screening of suitable candidates from chemical catalogs. Therefore, the first fragment growing step can be achieved efficiently and conveniently without the need for custom synthesis and minimal virtual screening expertise.

One of the major obstacles in drug discovery is related to the automation of the hit-to-lead optimization process. I present an integrated strategy for the hit-to-lead optimization phase and the automated design of chemical probes. Our approach “Diversity Oriented Target-focused Synthesis” (DOTS) consists of an in silico chemical library design coupled with a de novo diversity-focused robotic synthesis and an automated in vitro high throughput screening evaluation platform.