A new screening platform may expand the discovery of molecular glue degraders. Using an unbiased multiplexed screening approach, researchers identified the small molecule M12, which undergoes glutathionylation before recruiting the E3 ligase DCAF11 to degrade the RNA helicase DDX18, revealing what the researchers describe as a previously unrecognized mechanism of targeted protein degradation.
A multiplexed immunoprecipitation-mass spectrometry platform was developed to identify molecular glue degraders across multiple E3 ligases without prior knowledge of ligase-substrate interactions. The assay combined pools of recombinant E3 ligases, pooled small molecules, and whole-cell lysates, allowing endogenous metabolic activity to participate in compound activation. Following validation with established molecular glue degraders, the researchers screened seven E3 ligases against approximately 5,000 compounds organized into 320-compound pools. The screen identified DDX18 as a protein selectively recruited to DCAF11, and sequential deconvolution of the compound pools identified M12 as the active molecule.
The investigators then demonstrated that M12 promoted DDX18 degradation through the ubiquitin-proteasome pathway. Quantitative proteomics and immunoblotting showed dose-dependent reductions in DDX18 protein abundance, whereas DDX18 messenger RNA levels remained unchanged, supporting posttranslational degradation rather than reduced gene expression. Pharmacologic inhibition of ubiquitin activation, cullin neddylation, or proteasome function prevented DDX18 degradation, while lysosomal inhibition did not. In genome-wide and ubiquitin-proteasome system-focused clustered regularly interspaced short palindromic repeats (CRISPR)-Cas9 screens, DCAF11 was the only E3 ligase receptor identified as essential for the degradation response, and DCAF11 knockout abolished M12-induced DDX18 degradation.
Mechanistic studies indicated that M12 itself was not sufficient to induce DCAF11-DDX18 complex formation. Recombinant DCAF11 and DDX18 did not form a ternary complex in the presence of M12 alone, whereas addition of cellular lysate enabled the interaction, suggesting that an endogenous enzymatic process was required. Mass spectrometry identified a glutathionylated metabolite, GSH-M12, consistent with replacement of the compound's reactive chloride group by glutathione. The researchers further showed that recombinant glutathione S-transferase (GST) catalyzed the same modification in vitro and that DCAF11-DDX18 binding occurred only when M12, glutathione, and GST were present. Chemically synthesized GSH-M12 reproduced the interaction without cellular lysate, supporting glutathionylation as the activating step.
Cryoelectron microscopy provided structural support for the proposed mechanism. The 2.3 Å structure showed the glutathione moiety occupying a conserved glutathione-binding pocket within the WD40 domain of DCAF11, while the exposed M12 moiety remodeled the ligase surface to recruit DDX18. Mutations disrupting residues involved in glutathione binding or substrate recruitment prevented DDX18 recruitment, consistent with the structural model.
As a proof of concept, the investigators incorporated M12 into experimental proteolysis-targeting chimeras (PROTACs). DCAF11-directed degraders targeting BRD4, SMARCA2, BRD9, and multiple kinases induced degradation of their intended targets in cultured cells, with DCAF11 dependence demonstrated for several of these degraders. Proteomic analyses of a multikinase degrader identified degradation of LIMK2, WEE1, and the CDK7-cyclin H-MAT1 complex, findings confirmed by immunoblotting.
The study was limited to biochemical assays, recombinant proteins, structural analyses, and cultured human cell lines. The study did not include efficacy or safety testing in animal models or patients. The investigators noted that whether GSH-mediated prodrug activation could enhance therapeutic selectivity remains to be determined.
"Our studies reveal that the GSH-binding pocket of DCAF11 is highly conserved and that GSH itself can directly bind to DCAF11, suggesting a physiological GST- and GSH-dependent regulatory mechanism involving DCAF11," wrote Hojong Yoon, of The University of Texas MD Anderson Cancer Center, and colleagues.
Disclosures: Benjamin L. Ebert, Eric S. Fischer, Katherine A. Donovan, and Jianwei Che reported financial relationships with biotechnology and pharmaceutical companies, including research funding, consulting, advisory, equity, and founder roles. The remaining researchers declared no competing interests.
Source: Nature
