Systematic Platform Accelerates Identification of Novel Protein Degradation Drug Candidates

Investigators at Dana-Farber Cancer Institute have developed a platform for systematically discovering molecular glues that could become protein degradation drug candidates. The platform could help drug developers dramatically expand the range of disease-related proteins that can be therapeutically targeted for elimination via protein degradation.

The platform also enabled their discovery of the first molecular glue degrader that is metabolically activated, suggesting that molecular glues could be more context dependent and potentially tunable than previously thought.

The study was published in Nature.

This novel platform is an exciting scalable approach to the discovery of molecular glues that could help drive the significant expansion of molecular glue applications for the treatment of cancer and other diseases."

Eric Fischer, Co-Senior Author, Dana-Farber Cancer Institute

Protein degraders eliminate unwanted proteins from inside a cell by disposing of them using the cell's built-in recycling system. Specifically, a molecular glue degrader binds an E3 ligase, an enzyme that tags proteins for destruction, and redirects it to tag a disease-related protein for disposal.

In 2014, Benjamin Ebert, MD, PhD, president and CEO of Dana-Farber, found the mechanism of action behind the multiple myeloma drug lenalidomide to be a molecular glue degrader of a transcription factor. Because transcription factors tend to be smooth with few pockets for inhibitor drugs to bind to, they were thought to be "undruggable." Degrading transcription factors opened a new way of thinking about the treatment of cancer.

Since then, several protein degraders have entered clinical testing. However, these degraders only leverage a small handful of the 600 E3 ligases in the human genome.

"There is an incredible range of opportunity for discovering new molecular glue degraders," says Ebert a co-senior author. "This systematic approach could help accelerate the discovery of novel degraders that could change the way we think about the treatment of cancer."

The discovery platform's systematic process begins with a screen, devised by co-first author Hojong Yoon, PhD, who was a postdoctoral fellow in Ebert's lab and is now at MD Anderson Cancer Center. The screen fixes a subset of E3 ligases to magnetic beads in a well and bathes them in cellular lysate (the full complement of proteins found in a cell) and a library of drug compounds.

A hit occurs when one of the drugs binds to one of the E3 ligases and increases its affinity for a given cellular protein. This change causes instances of that protein to collect nearby. The team used mass spectrometry to determine which cellular proteins have affinity for the drug-bound E3 ligase and would be likely to be tagged for disposal inside a cell.

The team tested the system by screening seven E3 ligases. They found that a protein called DDX18 was drawn to the E3 ligase DCAF11, which is understudied. To determine which drug compound had enabled the connection, they repeatedly narrowed the pool of drugs in the screen and zeroed in on a molecule called M12.

Surprisingly, when the team tried to validate M12 as a molecular glue degrader in a cellular model to degrade DDX18, it did not work. To understand why, co-first author Franziska Wachter, MD, a Dana-Farber scientist and pediatric oncologist, used cryo-electron microscopy to take a closer look at the complex of DCAF11 and M12 identified by the screen.

She found that M12 had been altered by a metabolic process called glutathionylation. M12 would only act as a molecular glue inside cells with elevated levels of metabolites related to oxidative stress in the cell-something that is more common in cancer cells than normal cells.

"This was a huge surprise, and it is the first observation of a molecular glue that has been activated metabolically by glutathionylation," says Wachter. "Our in-house ability to do structural biology using cryo-electron microscopy was essential for us to quickly understand what was going on."

Further exploration of activated M12 revealed that it is a versatile DCAF11-paired degrader. By binding additional proteins to the complex, the team was able to tune the system to degrade multiple other proteins, including cancer-related protein targets such as SMARCA2, WEE1 and CDK7. These findings are a proof-of-principal for the screening approach, and more research is needed to identify the best molecular glue degrader to use as a drug candidate.

"This is a fabulous example of how powerful our combined expertise in cancer genomics, cell biology, structural biology and protein biochemistry can be," Ebert says. "This systematic approach to discovering novel molecular glue degraders opens up the possibility for expanding the number of proteins that can be targeted for degradation as a treatment for cancer."