Back

A Dynamically Regulated Designed IL-2 for Tumor-localized Signaling

Moffett, H. F.; Weitzner, B. D.; Lissner, M. M.; Davenport, T. M.; Haworth, K. G.; Hayes, E. T.; Langan, R. A.; Tait, L. J.; Obenza, W. M.; Hammerson, B.; Baker, L. E.; Crowl, J. T.; Hammer, J. L.; Tan, T.; Montoya, V. R.; Chen, J. C.; Clausen, D. S.; Sample, P. J.; Ng, A. H.; Yuan, S.; Amin, R. H.; Foster, A. E.; Boyken, S. E.; Lajoie, M. J.

2026-09-22 synthetic biology
10.64898/2026.09.18.752703 bioRxiv
Show abstract

Cytokines mediate cell-cell communication to coordinate immune responses and hold clinical promise as immunotherapies. While natural cytokines are exquisitely regulated by immune cells, engineered cytokines have not yet matched this sophisticated regulation, limiting clinical efficacy and safety. IL-2 is an important cytokine involved in antitumor response, but its production is disrupted in the immune-suppressive tumor microenvironment. Here we designed a regulated IL-2 module, OUTSMART designed IL-2 (dIL-2), that functions robustly in solid tumors and reduces systemic activity that could trigger toxicity: we computationally redesigned IL-2 to have a novel topology that increases stability, preserves native IL-2R{beta}{gamma} interfaces, ablates IL-2R binding, and delivers IL-2 signaling to CD8+ immune-effector cells. This designed cytokine is genetically regulated by a T cell activation-responsive promoter, allowing dynamic production from chimeric antigen receptor (CAR) T cells. Like wild type IL-2, OUTSMART dIL-2 drives T cell proliferation and enhances effector function. Unlike wild type IL-2, it preferentially stimulates CD8+ T cells and Natural Killer (NK) cells, avoids T regulatory cells, and stimulates CAR-T proliferation while preserving stemness, achieving durable tumor elimination in two solid tumor animal models. These design principles can be applied to create other dynamically-regulated cytokine systems that address mechanisms of complex disease.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.