An advance from the lab of chemist Thomas Kodadek, Ph.D. at The Herbert Wertheim UF Scripps Institute for Biomedical Innovation & Technology offers a cost-effective and efficient way to fish for enzymes with potentially useful properties, ones that enable the creation of medications called “molecular glues.”
The method enabled the Kodadek group to snag an elusive diabetes and cancer immunotherapy drug target, one that had been previously dismissed as “undruggable” due to its chemistry.
Molecular glues are a subject of much research right now, because they offer a creative way to address incurable diseases with known targets. They work by pulling together both the disease-causing drug target and a special cellular tag whose job is to mark unnecessary or toxic materials for recycling in the cell. This is designed to allow the cell to rid itself of the disease-driving substance. But a crucial step is finding the right molecular glues that can pull together these elements.
The Kodadek team’s new publication, in the Journal of the American Chemical Society, offers an innovative method for doing that. The paper’s co-first authors were Jiajun Dong, Ph.D., and Bo Li, Ph.D., who were both postdoctoral researchers in the Kodadek lab during the study.
“A major limitation in the development of molecular glues that trigger the post-translational modification of a target protein is the paucity of building blocks,” especially ones that attach to, but don’t degrade, the target, Kodadek said. “The vast majority of ligands for protein-modifying enzymes are inhibitors, which are not ideal for this application.”
They tackled the problem by developing a compound screening platform that strongly rejects discovery of inhibitors and instead connects with innocuous binders. Their system used libraries of potentially useful compounds linked to tiny beads prepared with markers for easy detection. The beads carried both a desirable connector and a “bad” one that modified undesirable inhibitors to make them not register.
“The assay is so sensitive, even very weak ligands can be identified,” Kodadek said. “Using this approach, Jiajun was able to identify a highly selective ligand for PTP1B, a previously ‘undruggable’ target,” said Kodadek, who is also a member of the UF Health Cancer Institute.
Scientists have tried but failed to make a drug that removes PTP1B, short for Protein Tyrosine Phophatase 1B. It’s a sought-after drug target, because it negatively regulates insulin and leptin signaling pathways, and affects inflammation and immune signaling. Removing it could be important for blood sugar control, appetite, inflammation, and possibly could improve the effectiveness of cancer immunotherapies, research suggests.
“The method is simple and does not require sophisticated instrumentation or reagents,” Kodadek said. “I hope that it will be adopted by many investigators interested in the discovery of novel molecular glues.”
The study, “Discovery of Noninhibitory Macrocyclic Ligands for Protein Tyrosine Phosphatase 1B Using a Function-Based, Iterative Screening Strategy,” was published July 27, 2026 in the Journal of the American Chemical Society. In addition to Kodadek, Dong and Li, Chung-Wei Fu contributed to the research.
Funders include the George T. Elmore Impact for Good Initiative, the University of Florida Research Opportunity Seed Fund and the National Institutes of Health. (R35GM15175.)