There are many proteins in humans, bacteria and viruses that would make great targets to shut down a pathway to treat a disease or stop an infection, but, for a variety of reasons, designing drugs to target those proteins has been difficult.

"Undruggable proteins tend to not have known small-molecule binding sites and/or they tend to be difficult to set up a screening method for – there's not a simple activity that you can measure in vitro," Matt Patricelli, head of early discovery at Vividion Therapeutics Inc., told BioWorld Insight.

J. Kevin Judice, founder and CEO of Dice Molecules SV LLC, said he sees disrupting protein-protein interactions, such as a growth factor binding to its receptor, as an interesting class of targets that has typically been thought of as undruggable. "You have two challenges: the first is getting anything to stick to the protein and the second is getting whatever sticks to the protein to block the attempted interaction by that protein with another large protein," Judice explained. "Small molecules – as the name implies – are small. So you're asking a small object to disrupt a very large interaction."

Target agnostic

Rather than starting with a target, San Diego-based Vividion starts with the small molecule and then looks for proteins that bind to the molecule. "We've turned the system upside down," Diego Miralles, CEO of Vividion, explained.

The discovery platform starts with a probe that can link to available amino acids, cystine residues for instance, to map all of the available residues on proteins in a cell using mass spectrometry. The company estimates it can hit about 40 percent of the proteome with its different probes over a variety of protein families.

To screen for targets, the drug candidate is incubated with the cells first, followed by the probe. Vividion can detect places where the drug is binding because it will block binding of the probe and the signal for that protein fragment will be dampened on the mass spectrometry reading for the drug plus probe compared to the probe alone.

In March, Vividion struck a deal with Celgene Corp. to develop drugs targeting E3 ubiquitin ligases that are responsible for targeting drugs for degradation, which could be useful for a range of oncology, inflammatory and neurodegenerative indications. (See BioWorld, March 6, 2018.)

Celgene had been working in the ubiquitin space and could "see how Vividion presented a unique opportunity," Miralles said. "Those proteins, by definition, are undruggable. It's really hard to drug E3 ligases."

Celgene, of Summit, N.J., paid $101 million up front, which includes an equity investment, for the four-year deal that has an option to be extended for an additional two years that would come with an additional undisclosed payment.

Vividion is responsible for the identification of the drug candidates, with Celgene having the right to opt in at the acceptance of IND applications. For some of the programs, Celgene will have worldwide rights with Vividion receiving milestone payments and up to double-digit royalties on sales, while other programs will be split equally, either in the U.S. or worldwide, between the two companies.

Panning

To look for small molecules that could bind to traditionally hard-to-target proteins, San Francisco-based Dice Molecules is taking a shotgun approach starting with libraries with complexity in the 1 billion to 10 billion range and then refining from there.

"You're just sifting a large collection of things to see what sticks and is valuable. For that to work well, you need to be able to do multiple repetitions," Judice said. "That really increases the signal to noise so that you can find interesting and rare solutions."

After the initial panning, the company creates a new library specific to the target "that really blows out the particular bits of chemical space that were discovered in the first campaign," Judice explained.

Over the last two years, Dice has struck discovery deals with Paris-based Sanofi SA and Genentech, a unit of Basel, Switzerland-base Roche Holding AG. The Sanofi deal was for up to 12 targets over a five-year period, providing Dice with $50 million in equity, up-front, target exclusivity, technology access fees and research services, and up to $184 million in research, clinical and regulatory milestone payments per target as well as royalties on future sales. Financial details of the multiyear Genentech deal weren't disclosed. (See BioWorld Today, March 17, 2016.)

The company is gearing up for a new screen in February for 22 targets, of which 10 are for targets Dice is interested in and the rest of the targets are split among the two partners.

"We can attend to the needs of those partnerships and do diligent work for our partners while parallel processing targets and molecules that will wind up in our own pipeline and that we own outright," Judice said.

Dice's lead work surrounds interleukin-17, including the different family members of the cytokine, with plans to enter the clinic in late 2020 or early 2021.

Aggregation

Aelin Therapeutics SA is using its Pept-in technology to target proteins that have typically been hard to drug. The technology takes advantage of short stretches of amino acids in proteins that induce aggregation. "Aggregated proteins can no longer do their function," Aelin CEO Els Beirnaert pointed out to BioWorld Insight.

The peptide-based Pept-in molecules work best on proteins with rapid turnover because they induce the aggregation by binding to their target during the protein folding step.

Potential aggregation sites occur every 100 amino acids on average, so larger proteins are more likely to have multiple sites amenable for targeting, but Aelin has identified the sites in 90 percent to 95 percent of proteins.

Aelin's lead program is targeting proteins in gram-negative bacteria. Depending on how conserved the target is between strains, Aelin could design antibiotics that are very specific or that are more broad spectrum. The company is shooting for having a lead candidate to go into preclinical testing at the end of 2019 or early 2020.

Last week, Aelin, of Leuven, Belgium, announced a €1 million (US$1.13 million) grant from the Flemish Government through Flanders Innovation & Entrepreneurship to explore using Pept-in for targets in oncology.

Beirnaert noted that targeting human proteins offers new challenges since the company needs to avoid homology to other human proteins and the "PK profile is different compared to bacterial infection."

With the first oncology targets identified, Aelin is shooting for starting a screen early next year. The company's goal is to discover peptides against multiple targets and then explore partnering opportunities at the end of next year and into early 2020.