(314b) Structural Characterization of an Effector-Biasing Interleukin-2 Immunocytokine

Background: Interleukin-2 (IL-2) has been used as a cancer therapeutic for roughly three decades. However, severe side-effects such as vascular leak syndrome combined with the vanishingly short serum half-life of the cytokine have limited the effectiveness of IL-2 therapy. Adding to this complexity, IL-2 signals through activation of either a high-affinity heterotrimeric receptor consisting of the IL-2 receptor-α (IL-2Rα, also CD25), IL-2Rβ, and common gamma (γ_C) chain or an intermediate affinity heterodimeric receptor that lacks IL-2Rα. IL-2Rβ and γ_C mediate signaling, whereas IL-2Rα regulates affinity of the receptor complex.Since T_Regs robustly express IL-2Rα, whereas naïve effector cells (Effs) do not, T_Regs have dramatically higher sensitivity to the IL-2 cytokine.

To address shortcomings of IL-2 therapy, previous studies have explored linking cytokines such as IL-2 to antibodies to improve both pharmacokinetics and targeting, in molecules called immunocytokines (ICs). Taking these ideas a step further, we have developed ICs which fuse antibodies against IL-2 to the cytokine itself. One of these molecules, denoted F10 IC, biases the IL-2 response toward immunostimulatory activity. Encouragingly, F10 IC has proven to be safe and effective as a cancer therapeutic in mouse models of melanoma. However, numerous molecular and mechanistic features of F10 IC have yet to be elucidated. Improved understanding of IL-2 ICs promises to advance their therapeutic design and development. Here, we present a structure-based characterization of the F10 IC molecule, and propose a model in which interactions between the antibody and cytokine result in a functionally relevant conformational change to IL-2.

Methods: High-resolution characterization of the interaction between the F10 antibody fragment and IL-2 was carried out using x-ray crystallography. Structural features of the full-length ICs were elucidated using cryogenic electron microscopy (cryo-EM). Single particle reconstruction methods were employed to generate intermediate-resolution maps into which IC models were built. Structural studies were supplemented with immune cell subset expansion profiling in mice following administration of our immunostimulatory IC versus IL-2 alone.

Results: Structural analysis revealed F10 (present as an scFv for crystallographic studies) as a direct steric inhibitor of the alpha subunit of the receptor complex. Further analysis demonstrated that F10 scFv binding resulted in a conformational change to IL-2, resulting in a conformation more similar to the receptor bound state. This feature was also observed in an immunostimulatory anti-IL-2 antibody that recognizes the mouse cytokine. Cellular subset and animal studies demonstrated the superiority of F10 IC over IL-2 alone in the selective expansion of CD8+ T cells and NK cells.

Implications: Our study demonstrates the mechanism by which the F10 IC biases IL-2 signaling in an immunostimulatory manner. The structural and functional insights our studies revealed will provide actionable data for further rational design of cytokine therapies. Future studies will focus on structure-informed improvements to the ICs, and therapeutic evaluation in animal models of cancer. Continued development of IL-2-containing ICs represents a fundamental advance for cytokine therapy, with improved pharmacokinetic properties and signaling specificity compared to unconjugated IL-2.

Acknowledgments: This research was funded by NIH R01EB029455, DoD Concept Award W81XWH-18-1-0735, and a Melanoma Research Alliance Young Investigator Award to Dr. Jamie Spangler.