Interleukin-2 (IL-2) is a multifunctional cytokine that is essential to the differentiation, growth, and activity of both stimulatory and suppressive cells of the immune system. Its key role in regulating immune homeostasis renders IL-2 an attractive therapeutic target for a wide range of immune-linked disorders such as autoimmune disease, cancer, and infectious disease. Unfortunately, efforts to develop IL-2 as a therapeutic have been limited by its concurrent promotion of immune effector cells (such as memory-phenotype [MP] CD8+ T and natural killer [NK] cells) and regulatory T (TReg) cells, which leads to undesirable off-target effects and dose-limiting toxicity. It would therefore be of tremendous therapeutic value to isolate and selectively modulate the immunostimulatory and immunosuppressive effects of IL-2 in order to bias immune homeostasis for particular disease applications. We have implemented a structure-guided antibody engineering approach to achieve selective bias of IL-2 activity through engagement of particular cell subsets based on their surface receptor expression patterns. Our work represents a major advance in IL-2 therapeutic development and provides a molecular blueprint for the design of other cytokine-targeted drugs.
Whereas most cytokines signal through engagement of only two receptor chains, IL-2 signals through formation of either a high-affinity quaternary complex (Kd≈10 pM) with the interleukin-2 receptor-α (IL-2Rα, also known as CD25), IL-2Rβ, and IL-2Rγ (also known as common gamma, γc) chains, or an intermediate-affinity ternary complex (Kd≈1 nM) with only the IL-2Rβ and γc chains. Consequently, expression of the non-signaling IL-2Rα subunit regulates cytokine sensitivity whereas the IL-2Rβ and γc chains mediate signaling. IL-2Rα is robustly expressed on TReg cells but virtually absent from naïve effector cells, resulting in differential IL-2 responsiveness of these subsets. It was recently noted that immunocomplexes consisting of mouse IL-2 (mIL-2) and certain anti-cytokine antibodies preferentially stimulate particular immune cell subsets, presenting an enticing strategy for selective modulation of immune homeostasis. Immunocomplexes comprised of IL-2 and the JES6-1 antibody actively induce proliferation of IL-2RαHigh TReg cells but not IL-2RαLow effector cells. In contrast, immunocomplexes comprised of IL-2 and the S4B6 antibody potentiate growth of all immune cells, but particularly favor IL-2RαLow/IL-2RβHigheffector cells. Subsequent work has validated a vast array of therapeutic applications for these two IL-2-targeted antibodies: JES6-1 immunocomplexes promote graft tolerance and show efficacy in preclinical models of autoimmune disease and S4B6 immunocomplexes exhibit potent anti-tumor activity without inducing toxicity. However, in the absence of structural or molecular characterization, the mechanistic basis for selective cytokine potentiation was unknown.
We combined crystallographic, biophysical, and functional data to elucidate the molecular rationale for antibody-induced bias of cytokine activity. Through determination of the high-resolution molecular structures of both the IL-2:JES6-1 and IL-2:S4B6 immunocomplexes and implementation of surface plasmon resonance-based competition studies, we found that JES6-1 and S4B6 exert complex steric and allosteric effects on IL-2 to alter its molecular properties and, consequently, its functional behavior. JES6-1 sterically blocks binding of IL-2Rβ and γc and also allosterically impedes binding of IL-2Rα to the cytokine. Allosteric disruption of the IL-2:IL-2Rα interaction results in exquisite IL-2 sensitivity of IL-2RαHigh TReg cells but not IL-2RαLow effector cells, as high amounts of IL-2Rα expression are required to overcome this disruption to allow for IL-2 complex formation and signal activation. In vivo immune cell subset expansion studies revealed that selective proliferation of TReg cells is perpetuated by a transcriptional feedback loop. IL-2:JES6-1 immunocomplex treatment increases immune cell surface expression of IL-2Rα through two concurrent effects: (1) Increased IL-2Rα transcription, which has been shown to occur in response to IL-2 signaling, and (2) Selective proliferation of cells with particularly high surface densities of IL-2Rα. This elevated IL-2Rα expression in turn results in enhanced sensitivity to the IL-2:JES6-1 immunocomplex, further heightening selectivity for IL-2RαHigh TReg cell expansion. In the case of the IL-2:S4B6 immunocomplex, we found that S4B6 sterically occludes IL-2Rα and mildly obstructs IL-2Rβ binding to the cytokine but it also allosterically strengthens the IL-2:IL-2Rβ interaction by inducing an affinity-enhancing conformational change in IL-2. The net result of these structural effects is desensitization to IL-2Rα expression as S4B6-bound IL-2 signals equivalently through the IL-2 ternary complex on all IL-2 responsive immune cells, and susceptibility to IL-2 signaling is instead governed by IL-2Rβ expression. Allosteric enhancement of IL-2:IL-2Rβ interaction by S4B6 leads to increased stimulatory activity that particularly favors IL-2RβHigheffector cells, resulting in the observed bias toward immunostimulatory effects following IL-2:S4B6 immunocomplex treatment. These insights provide, for the first time, a direct link between the molecular interactions of cytokine-antibody complexes and their unique functional outcomes, establishing a new paradigm for antibody-mediated modulation of IL-2 behavior.
We harnessed our structural insights to engineer variants of the JES6-1 antibody with enhanced functional properties by combining molecular engineering tools with immunological characterization techniques. We pursued two objectives: (1) Development of antibodies that accentuate the TReg cell-biased activity of JES6-1 and (2) Development of a JES6-1 variant that cross-reacts with the human IL-2 (hIL-2) cytokine. Currently, JES6-1 is the gold standard antibody used ubiquitously to expand TReg cells, but there is significant room for improvement as IL-2:JES6-1 complexes are effective prophylactically in mouse models of autoimmune disease, but actually accelerate pathogenesis if administered after disease onset. This results from IL-2:JES6-1 immunocomplex-induced upregulation of IL-2Rα on activated effector cells, which heightens their sensitivity to IL-2 signaling and counteracts the TReg cell bias. Furthermore, JES6-1 is presently limited to mouse studies as it does not interact with hIL-2. We leveraged our exclusive structural insights to design yeast surface displayed libraries of JES6-1 that mutagenized residues in the cytokine-interacting epitope, and evolved these libraries to fulfill our two objectives. To develop antibodies with enhanced TReg cell selectivity, we evolved JES6-1 variants that bind IL-2 with a range of affinities and inhibit the IL-2:IL-2Rα interaction to varying extents. We then directly evaluated the relative in vivo TReg:effector cell expansion ratio elicited by these variants to determine the optimal IL-2:JES6-1 interaction affinity for TRegcell-biased growth. To isolate a JES6-1 variant that interacts with hIL-2, we selected and functionally characterized antibody clones that interact with both the mouse and human cytokines.
Collectively, our work establishes structure-based engineering as a general approach toward developing effective cytokine-biasing antibodies and introduces novel therapeutic molecules with immediate clinical relevance. Furthermore, our insights and methodologies can readily be applied to other systems to evolve cytokine- or growth factor-directed antibodies with optimized receptor binding and competitive properties to address a host of immunotherapeutic objectives.
See more of this Group/Topical: Meet the Faculty Candidate Poster Session – Sponsored by the Education Division