Advances in covalent drug discovery

Abstract

Covalent drugs have been used to treat diseases for more than a century, but tools that facilitate the rational design of covalent drugs have emerged more recently. The purposeful addition of reactive functional groups to existing ligands can enable potent and selective inhibition of target proteins, as demonstrated by the covalent epidermal growth factor receptor (EGFR) and Bruton's tyrosine kinase (BTK) inhibitors used to treat various cancers. Moreover, the identification of covalent ligands through 'electrophile-first' approaches has also led to the discovery of covalent drugs, such as covalent inhibitors for KRAS(G12C) and SARS-CoV-2 main protease. In particular, the discovery of KRAS(G12C) inhibitors validates the use of covalent screening technologies, which have become more powerful and widespread over the past decade. Chemoproteomics platforms have emerged to complement covalent ligand screening and assist in ligand discovery, selectivity profiling and target identification. This Review showcases covalent drug discovery milestones with emphasis on the lessons learned from these programmes and how an evolving toolbox of covalent drug discovery techniques facilitates success in this field.

Fig. 1. Timeline of the development of major covalent drugs.

Each covalent drug is classified according to the drug type or type of disease it treats. Unless otherwise indicated, the date refers to the first approval by the US Food and Drug Administration. NSAID, non-steroidal anti-inflammatory drug.

Fig. 2. Progression of EGFR inhibitor structures.

Progression from first-generation (part a) to second-generation (part b) epidermal growth factor receptor (EGFR) inhibitors involved the addition of a reactive acrylamide electrophile (highlighted in red) to covalently bind a cysteine residue (Cys797) in EGFR. In the progression from second-generation to third-generation (part c) EFGR inhibitors, the quinazoline moiety is replaced with a pyrimidine unit to provide selectivity for the T790M mutant.

Fig. 3. Aligned structures of KRAS(G12C) co-crystallized with adagrasib (MRTX849) and sotorasib (AMG-510).

The covalent inhibitors adagrasib (PDB ID: 6UT0) and sotorasib (PDB ID: 6OIM) are bound to the switch II pocket, which is adjacent to the GDP-binding pocket^,.

Fig. 4. Nirmatrelvir in complex with SARS-CoV-2 main protease.

The nitrile group of nirmatrelvir (shown in green) reacts with Cys145 of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) main protease (shown in light blue) to form a covalent thioimidate adduct. Extensive hydrogen-bond interactions (yellow dashed lines) occur throughout the pocket (PDB ID: 7RFW).

Fig. 5. Isotopic tandem orthogonal proteolysis–activity-based protein profiling.

a | Example of a reactive probe for activity-based protein profiling (ABPP) designed with a broadly reactive electrophilic warhead linked to an analytical handle. b | Schematic of the competitive isoTOP-ABPP (isotopic tandem orthogonal proteolysis–activity-based protein profiling) methodology. Treatment of cells or lysate with a protein-reactive compound prevents subsequent binding of the pan-reactive probe, and this competitive ligand binding can be detected by tandem mass spectrometry (MS/MS) after an enrichment step, to indirectly identify covalent protein targets for the compound of interest. TEV, tobacco etch virus.