Abstact

Erythropoietin (EPO) initiates EPO receptor (EPOR) signaling in hematopoietic cells by binding to an asymmetric EPOR dimer through two different sites. We engineered dimeric diabody-Fc (Db-Fc) fusion proteins that appeared to act as potent agonists of human EPOR in cell proliferation assays. However, detailed analysis of their oligomeric forms revealed that the predominant Db-Fc species bound EPOR with high affinity but failed to induce cell proliferation. Instead, a minor oligomeric form, identified as a putative tetrabody (Tb) fused to two Fc domains (Tb-Fc2), proved to be the minimal active form. The existence of a tetrameric agonist was further supported by crystallography, which revealed an asymmetric Tb structure. Additionally, the structure of an antigen-binding fragment (Fab) bound to EPOR revealed an epitope distinct from the EPO binding sites, and structural modeling showed that engagement of two of the four binding sites on the Tb could form an asymmetric EPOR dimer nearly identical to the active conformation recruited by EPO. In a knock-in mouse model, where mouse EPOR was replaced by human EPOR, purified Tb-Fc2 stimulated erythropoiesis with greater potency, efficacy, and duration than darbepoetin, a recombinant EPO that is the leading therapeutic erythropoiesis-stimulating agent (ESA). Collectively, these findings demonstrate that asymmetric tetravalent antibodies such as Tb-Fc2 represent promising next-generation ESAs that provide enhanced potency, efficacy, and durability. Moreover, they may reduce the oncogenic and cardiovascular risks associated with the pleiotropy of EPO.