Chemoproteomics

Since their early days in the second half of the 19th century, small molecule drugs (SMOLs) targeting disease-related proteins have become a mainstay of the pharmaceutical industry, and today, treatments based on SMOLs make up about 90 percent of all drugs sold worldwide.

Despite this, the percentage of the human proteome – the entire set of proteins that can be expressed by a cell at any given time – that can be targeted by these drugs has remained comparatively small. The great majority of proteins are still considered “undruggable”, due to their lack of clearly defined binding pockets that can be effectively targeted by small molecules, or limitations of chemistry to design suitable compounds. This “undruggable space,” contains many proteins involved in the development of a variety of diseases, and approaches to address this undruggable space are critical for furthering the scope of modern drug discovery.

Chemoproteomics offers a way to overcome this obstacle and access these previously “undruggable” proteins. By rethinking the drug discovery process, it opens-up new treatment options for some of the most serious diseases affecting millions of patients all over the world.

Reinventing drug discovery

Traditionally, the process of drug discovery has followed one of two trajectories: In classical pharmacology, or phenotypic drug discovery (PDD), different small molecules were introduced into cellular or animal models of specific diseases and screened for desirable changes in the disease phenotype. Only after a therapeutic effect was observed for a compound, and often after drugs were approved, would researchers attempt to identify the protein target responsible for the drugs’ effects. Target identification for drugs resulting from phenotypic screening is a long and often unsuccessful process, and many popular drugs are still in use that act through unclear or unknown targets.

In the late 20th century, a new approach called target-based drug discovery (TDD) tried to overcome these limitations. This process, also called “reverse pharmacology,” begins with the identification of potentially promising target proteins, followed by the search for compounds that modulate the function of the target with the help of high-throughput screening methods. But this approach also has limitations, as the majority of disease-related proteins cannot be addressed by TDD, leaving a large number of targets undruggable.

Chemoproteomics, in particular activity-based protein profiling (ABPP), is yet again rethinking the drug discovery process and offers a key to unlock the undruggable space. Instead of beginning the search for new drugs by focusing on a specific compound (PDD) or target (TDD), this new approach takes an unbiased look at the entire human proteome, using reactivity-based chemical probes and high-resolution mass spectrometry to screen the interactions of different compounds with all of the approximately 20,000 proteins coded by the human genome in their native cellular environments.

This approach comes with various advantages over both PDD and TDD. By observing interactions of individual compounds with the entire proteome, ABPP has the potential to exponentially expand the druggable space by highlighting previously undiscovered binding sites that might function as starting points for new treatments of diseases. ABPP provides researchers with both the compound and the target involved in a binding event up front, eliminating the challenging search for targets. What is left is the biological step of the drug discovery process: identifying whether the binding event has a potential therapeutic effect.

The return of the covalent modifier

The biggest advantage of a chemoproteomics drug discovery approach like ABPP, however, is that it allows for the targeted identification of one of the most powerful kinds of small molecule drugs: covalent modifiers. Forming a specific and often irreversible bond with proteins, covalent modifiers are characterized by a higher potency and a longer-lasting effect than other small molecules and can target shallower binding pockets and even protein complexes.

The concept of covalent modifiers is not new. The first medication of this kind, aspirin, was synthesized by Bayer in the 1890s. Since then, covalent drugs have had an immense impact on human health. In 2009, three of the ten most successful medicines sold in the US were covalent drugs, and according to estimates by Nature, the 26 covalent drugs for which data is available collectively account for more than 33 billion US dollars in worldwide sales.

Many of the covalent drugs available today were discovered by chance, as pharmaceutical companies have hardly engaged actively in the development of new covalent drugs in the past, owing to their complexity and risk of failure. But with chemoproteomics a new era is dawning for covalent drug, and Bayer is aspiring to be a leader in this field.

Joining forces to expand the druggable space

Modern chemoproteomics tools allow us to develop novel, targeted covalent modifiers by taking a broad look at the entire proteome and mitigating many of the risks currently involved in drug discovery. Utilizing ABPP allows us to rapidly expand the druggable space and bring new, first-in-class precision treatments to patients suffering from as-yet intractable diseases.

Bayer has invested significantly in chemoproteomics and the development of Targeted Covalent Modifiers, externally as well as in-house. With the acquisition of Vividion Therapeutics, the company added a player with unmatched pioneering achievements in the field of activity-based proteomics that brings with it not only an industry-leading chemoproteomics screening platform, but also a library of cysteine-reactive small molecules that perfectly complements Bayer’s own. To date, Vividion’s technology has already proven its applicability pre-clinically in oncology and immune-related diseases, and Bayer is working to expand its capabilities into additional indications.

Cornerstone of success will be a rich and modern SMOL library with diverse high-quality compounds to generate the leads for future clinical candidates. Within the framework of the “Next Generation Library Initiative,” Bayer is building such a modern, diverse covalent compound library. Over the course of the last years, we have synthesized more than 12,000 high-quality, covalent compounds, and our unique, fully automated synthesis platform will allow us to increase this number flexibly and cost-effectively to over 60,000 in the near future. And Vividion’s proprietary library of over 15,000 diverse covalent SMOLs that are specifically designed to form covalent bonds with accessible cysteine residues in shallow protein binding pockets, is a valuable asset for the chemoproteomics screening.

Combining Vividion’s chemoproteomics technology with Bayer’s capabilities in bringing new small molecule medicines to patients at scale, we are in an unparalleled position to address undruggable targets and generate first-in-class novel compounds, unlocking the undruggable space for the benefit of patients with high unmet medical needs.