Mutations in the RAS oncogene are the most common activating mutation in human cancer.  This presentation will describe how a covalent inhibitor approach proved to be a well-suited strategy for inhibiting KRAS G12C driven tumors.  The medicinal chemistry team uncovered and exploited a cryptic pocket in the KRAS G12C protein to drive potency which ultimately led to the discovery of sotorasib, a first-in-class covalent inhibitor of KRAS G12C.

The throughput of conventional LC-MS limits its practical application to the kinetic characterization for routine SAR study of covalent inhibitors. We demonstrated that high-throughput kinetic analysis of covalent inhibitors with a wide range of potency can be achieved by using MS-based methodologies. The platform described can be used to obtain reliable kinact/Ki values to guide SAR interpretation and understand the molecular mechanism of inhibitors in covalent lead discovery.

Macrocyclic peptides hold promise for targeting protein-protein interactions (such as Ras-Raf interaction) due to their propensity to potently bind to diverse surfaces. Leveraging a known allosteric KRAS inhibitory peptide KRpep-2d, we developed permeable but non-cationic macrocycles with pan-KRAS activity in both biochemical and cellular assays. The permeability mechanism studies revealed that endocytosis-independent pathway including passive permeation could be fulfilled through SAR optimization.

RET signalling is implicated in a number of disease states. Overactivity may result in cancer, but stimulation of the system is being considered as treatment for neurodegenerative diseases such as Parkinson’s and Alzheimer’s disease.  We have determined the cryo-EM structure of the hexameric signalling complex formed between RET, the NRTN ligand and GFRa2 co-receptor. The structure highlights the importance of the cysteine-rich domain of RET for the complex assembly and signalling.

Drugs targeting the µ-opioid receptor (µOR) are the most effective analgesics but are associated with the severe side effects behind the current opioid epidemic. Here I will present our work on the structural and mechanistic basis of action of µOR agonists with different safety profiles. I will discuss how cryo-EM structures, molecular dynamics simulations and cell-based assays suggest that drugs engaging different parts of the µOR orthosteric pocket can lead to distinct signaling outcomes.

NMR-guided screening strategies to design potent and effective ligands targeting PPIs include our recently developed fHTS by NMR, that can guide the design of nanomolar agents even for challenging drug targets. Moreover, we also found that the design of potent ligands can be aided by introduction of properly juxtaposed electrophiles targeting surface Lys residues, that are more often found at protein interfaces compared to Cys residues. We found that incorporation of aryl-fluorosulfates can lead to pharmacologically viable irreversible agents that are cell permeable and orally bioavailable. Perhaps most importantly, the approach is also useful in devising novel E3-directed covalent degraders.

E3 ubiquitin ligases are part of the ubiquitin proteasome system (UPS). In humans, over 600 E3 ligases are described, but only a handful of E3 ligases, such as VHL, cereblon and IAP, are currently explored for the targeted degradation of disease relevant proteins using PROteolysis-Targeting Chimeras (PROTACs). Here, we focus on the structural characterization and functional validation of a cancer relevant E3 ligase for the PROTAC approach. We apply fragment-based screening approaches for the identification of previously unknown binding sites and optimize initial hits towards covalent PROTACs.

To query reversible engagement across KRAS hotspot mutants in cells, we developed pan-RAS target engagement probes.  The probes engaged prevalent KRAS mutants inside cells regardless of nucleotide state.  This assay profiled known RAS S2P ligands and identified weak affinity fragments from a small library.  We found that multiple KRAS mutants are surprisingly vulnerable to reversible S2P engagement.  Our findings support utilization of cell-based screening workflows to identify allosteric RAS inhibitors.

Classical drug-like compounds that largely abide by Lipinski's rules are generally unsuitable for direct competitive PPI inhibition due to the absence of ligandable pockets and in this context such targets are “Hard to drug.” While still challenging, drug-like inhibitors have been successfully developed that allosterically modulate PPIs and here we describe our biophysical approaches to supporting the development of such inhibitors using a variety of techniques.

We investigated the structural basis for regulating formation of human GPCR signaling complexes with partner proteins by membrane phospholipids and cholesterol. Using NMR spectroscopy of GPCR complexes prepared in nanodisc membrane mimetics, we show how the presence of specific endogenous phospholipids and cholesterol regulate mechanisms by which GPCRs recognize and interact with partner proteins. Results show how careful consideration of lipid composition is crucial when studying protein-protein interactions.

Targeted covalent inhibitors are an important class of drugs and chemical probes. However, relatively few electrophiles meet the criteria for successful covalent inhibitor design. Here we describe a-substituted methacrylamides as a new class of electrophiles suitable for targeted covalent inhibitors. In the context of the BTK inhibitor ibrutinib, these electrophiles showed lower intrinsic thiol reactivity and high potency. This Covalent Ligand Directed Release (CoLDR) chemistry helped to develop fluorogenic probes for BTK, EGFR, and K-RasG12C. CoLDR chemistry allowed site-specific labelling of proteins in cells which helped to find the half-life of protein and develop CoLDR PROTACS.

The identification of molecular glues stabilizing protein-protein interactions is facilitated by a detailed structural, biochemical, and biophysical understanding of the interaction.  Ambagon Therapeutics is a leader in developing a modular approach to molecular glue discovery, utilizing a significant data set of the interactions between the 14-3-3 family of adaptor proteins and their phosphoprotein clients, many of which play critical roles in diseases of unmet medical need.

NKG2D is a crucial immune receptor that is activated by various endogenous protein ligands (NKG2DLs) that are overexpressed on the surface of the stressed cells. The NKG2D-NKG2DL interface buries a large interface and lacks discernable druggable pockets, making inhibition of this PPI very challenging.  Implementing a multipronged hit-finding campaign identified compounds that disrupt the activation of NKG2D. Through structural studies we show that these small molecule inhibitors capture a cryptic pocket leading to a unique MoA.