The cell membrane is notoriously selective about what it lets in. Charged molecules? Mostly rejected at the door. That’s a problem, because some of the most promising drug targets sit behind that barrier, and reaching them requires chemistry the membrane won’t tolerate.
A multi-institutional team including Nathanael S. Gray at Stanford and Brenda A. Schulman at the Max Planck Institute of Biochemistry found a way around this. In a study published in Nature Chemical Biology in April 2026, they describe a compound called ZZ1 that enters cells in disguise, transforms once inside and then recruits a protein-disposal complex that had never previously been targeted by a drug. It’s part secret agent, part Trojan horse, and it could change how we think about destroying disease-causing proteins.
The Problem No One Had Solved
Targeted protein degradation (TPD) is one of the hottest areas in drug development. Instead of blocking a harmful protein, you trick the cell into throwing it away. Small molecules called molecular glue degraders (MGDs) make this happen by gluing a target protein to an E3 ubiquitin ligase, the enzyme that tags cellular trash for disposal.
The catch: nearly all known MGDs rely on the same small handful of ligases. Cancer cells can resist these drugs by simply dialing down the ligase. And many E3 ligases that recognize charged clients have been difficult to access, because cell-permeable charged small molecules are hard to develop.
The Screen That Found Something Strange
The team took a direct approach. They started with JQ1, a well-known inhibitor of bromodomain and extraterminal domain (BET) bromodomain proteins (epigenetic regulators), and bolted on a library of small chemical tags to see if any would turn it into a protein destroyer.
To measure whether BRD4 protein was actually disappearing from cells, they used the Nano-Glo® HiBiT Lytic Detection System. Using CRISPR-Cas9, they fused the small 11-amino-acid HiBiT tag directly to BRD4 so it would be expressed under endogenous regulation. Adding the detection reagent provides the complementing polypeptide LgBiT, which interacts with the HiBiT tag to reconstitute the bright, luminescent NanoBiT® enzyme. Luminescence intensity is directly proportional to the amount of HiBiT-tagged protein in the cell lysate. This simple add-mix-read assay gave the team a sensitive, quantitative way to screen compounds for degrader activity.
One compound, ZZ1, lit up the screen. It selectively destroyed BET family proteins. But here’s where things got interesting: CRL inhibition did not prevent degradation the way proteasome inhibition did, suggesting involvement of a novel non-CRL E3 ligase.
The Disguise
A CRISPR screen pointed to CTLH E3 ligase components, and biochemical reconstitution plus knockout experiments established that the WDR26-YPEL5-containing CTLH complex was necessary and sufficient for ZZ1-mediated BRD4 degradation. No drug had ever recruited this ligase for targeted protein degradation.
When the team determined cryo-EM structures of the BRD4BD1-bound YPEL5-CTLH recognition complex, the mechanism snapped into focus. ZZ1’s reactive sulfonyl fluoride group wasn’t doing what reactive groups usually do (forming a covalent bond). Instead, ZZ1 underwent metabolic/chemical activation to a sulfinic acid derivative, and metabolomic analyses identified ZZ1-SO₂H as the predominant intracellular species. That negatively charged sulfinic acid moiety interacts with a basic pocket on YPEL5, held in place by electrostatic attraction.
In other words, ZZ1 is a molecular Trojan horse. It slips through the membrane as a neutral, unremarkable molecule. Once inside, it sheds its disguise, revealing the charged group that actually does the work. The team called this new class of compounds “charged molecular glues,” or c-Glues.
Better by Design
The structural data revealed another surprise. Neither the active form of ZZ1 alone nor BRD4 alone has any meaningful affinity for YPEL5. Only the composite surface of BRD4 bound to the activated compound engages the ligase, with a dissociation constant of ~12 nM. This cooperative, target-first binding is the opposite of how existing drugs like thalidomide work (those bind the ligase first), and it means future campaigns could start from the target side for greater pharmacological control.
The structures also handed the team a blueprint for improvement. They spotted a vacant basic pocket on YPEL5, added a chlorine atom to fill it, and created ZZ2, a roughly three-fold more potent degrader (DC50 of 169 nM).
What This Unlocks
This work matters because it cracks open a door that was previously shut. The CTLH ligase is now a viable target for drug-induced proximity. The prodrug trick solves the membrane permeability problem for charged pharmacology. And because c-Glues require both drug and target to be present before the ligase engages, they sidestep the “hook effect” that weakens many bifunctional degraders at high doses. With YPEL5 essential in most cancer cell lines, this provides a structure-guided starting point for developing more potent YPEL5-dependent c-Glues, though extending the approach to new target classes remains an open challenge.
Reference
Zhuang, Z., Byun, W.S., Chrustowicz, J. et al. Charged molecular glue discovery enabled by targeted degron display. Nat Chem Biol (2026). https://doi.org/10.1038/s41589-026-02182-5
Sara Millevolte
Latest posts by Sara Millevolte (see all)
- Charged Molecular Glues: The Molecular Trojan Horse That Slipped Past Biology’s Toughest Bouncer - May 14, 2026
- From Gum Disease to Breast Cancer: An Oral Bacterium’s Unexpected Journey - February 10, 2026
- Brewing Immunity: The Vaccine Beer Experiment - January 13, 2026
