Researchers at Baylor College of Medicine have combined high-throughput proteomics with artificial intelligence to discover and optimize molecular glues that target VAV1, an immune-cell signaling protein linked to blood cancers and autoimmune diseases. Published in Nature Communications, the study details how the workflow—dubbed GluePlex—accelerates the identification of compounds that direct the cell’s disposal machinery to eliminate disease-causing proteins entirely, rather than just blocking their function.
High-Throughput Proteomics Identifies VAV1-Degrading Compounds
Many scientists are shifting focus toward eliminating harmful proteins instead of inhibiting their activity. Because molecular glues remove the entire protein, the strategy produces a more complete therapeutic effect. To find compounds capable of degrading VAV1, the research team screened a library of molecules using high-throughput proteomics, a technology that assesses thousands of proteins simultaneously. Senior and co-corresponding author Jin Wang, director of Baylor’s Center for NextGen Therapeutics, explained the shift in approach. First and co-corresponding author Hanfeng Lin noted that this unbiased analysis revealed a series of compounds, including NGT-201-12, which caused VAV1 levels to drop while affecting relatively few other proteins.
Follow-up experiments showed that the degradation process depends entirely on the proteasome and cereblon, a critical component of the cell’s protein-degradation pathway. By using this pathway, the identified compounds effectively flag the target protein for destruction without disrupting the wider cellular environment.
AI-Driven Structural Modeling with GluePlex
To move past the limits of traditional screening, the Baylor team developed GluePlex, a computational workflow that integrates AI-based protein-structure prediction with physics-based modeling. Without relying on an experimental structure of the ternary complex, GluePlex successfully modeled how VAV1, cereblon, and the molecular glue assemble together. Lin pointed out that the model identified a specific region of VAV1, known as the SH3-2 domain, as essential for degradation. Subsequent experimental tests confirmed this prediction, pinpointing the exact spot the glue uses: a small surface loop on VAV1 that acts as a degradation signal, or “degron.” This loop differs distinctly from degradation signals commonly associated with cereblon-targeting molecular glues.
Did you know? Molecular glues act as matchmakers by bridging a disease-linked target protein with the cell’s own ubiquitin-proteasome system, essentially tricking the cell into clearing out rogue proteins.
Optimizing Degradation Potency with Free Energy Perturbation
Applying Free Energy Perturbation to predicted ternary structures yields cooperativity metrics that correlate directly with degradation potency. According to the study’s authors, this approach overcomes the limitations of standard docking, enabling the prospective ranking of analogs even when starting from weak initial binders. By adding halogen substitutions to restrict molecular flexibility, the team improved degradation efficiency and produced NGT-201-18. This optimized compound forms a stronger degradation complex, successfully reducing VAV1 levels and suppressing T-cell activation in primary human T cells.
However, dose-response proteomics also identified LIMD1 as an off-target protein carrying a canonical G-loop degron. This discovery demonstrates that a single molecular glue can engage structurally distinct degrons, highlighting the absolute necessity of proteome-wide profiling during early drug discovery.
Common Questions Regarding Molecular Glue Discovery
What is a molecular glue?
Molecular glues are small molecules that bring disease-linked proteins into close physical contact with the cell’s disposal machinery, marking those specific proteins for elimination.
What is VAV1 and why is it targeted?
VAV1 is an immune-cell signaling protein implicated in both blood cancers and autoimmune diseases. Eliminating it entirely offers a more complete therapeutic effect than merely inhibiting its functions.
What is GluePlex?
GluePlex is a computational workflow developed by Baylor researchers that integrates AI-based protein-structure prediction with physics-based modeling to map how molecular glues assemble with target proteins and cereblon.
Are these compounds ready for clinical use?
No. The compounds remain strictly preclinical, meaning researchers must conduct additional studies to assess their pharmacology, safety, selectivity, and activity in disease models.
Keep reading