Background: Tumor necrosis factor receptor 2 (TNFR2) has emerged as a pivotal regulator of immunosuppressive functions and lineage stability in regulatory T cells (Tregs). Generation of a reliable and well-engineered TNFR2 agonist antibody can open new avenues in therapeutic intervention for autoimmune diseases.

Objectives: We computationally designed anti-TNFR2 antibodies with agonistic activity and assessed their in vitro pharmacological properties in primary human Tregs and in vivo efficacy in a skin inflammation humanized-mouse model.

Methods: Relying on structural models of the TNFR2, we computationally designed multiple antibodies to agonize it. For cellular receptor binding, human peripheral blood mononuclear cells (PBMCs) were incubated with an anti-TNFR2 agonistic antibody to measure Treg surface binding by flow cytometry. TNFR2 signaling activity was measured in a HEK-Blue NFkB reporter cell line and also by phosphorylated RelA (pRelA) induction in primary human Tregs. In functional assays, human PBMCs were incubated with the anti-TNFR2 agonistic antibody for 5 days for assessment of Treg proliferation and functional marker characterization. Human TNFR2 knock-in mice were utilized for evaluating in vivo TNFR2 receptor occupancy and in a KLH induced delayed type hypersensitivity (KLH-DTH) model. In the KLH-DTH model the mice were sensitized and challenged with KLH on Day 0 and Day7, respectively. A single dose of the anti-TNFR2 agonistic antibody was subcutaneously administered on Day 6. Anti-inflammatory effect was assessed by measuring ear thickness.

Results: In the reporter assay, 8 single chain fragment variables (scFvs) were identified as TNFR2 agonists, suggesting that the computationally designed variants agonize TNFR2 by stabilizing an active conformation of the receptor. Bivalent antibodies showed higher NFkB reporter activity compared to the parent scFv, suggesting that clustering the receptors further enhances the agonism. A lead antibody preferentially bound to TNFR2 on Tregs compared to other T cell subsets. In concordance with receptor binding, concentration-dependent pRelA induction was observed in Tregs but not in conventional CD4 and CD8 T cells. The lead antibody also promoted Treg proliferation and upregulation of FoxP3 and TIGIT in a primary human PBMC culture system as a single agent without the need for T cell receptor stimulation. The Treg-selectivity profile of the lead antibody observed in vitro was also consistent with the receptor occupancy profile seen in vivo. Finally, a single dose treatment of the agonistic antibody showed efficacy in suppressing antigen-induced skin inflammation in the KLH-DTH mouse model.

Conclusion: We have designed unique scFv-based TNFR2 binders recognizing a “signaling epitope”. The lead antibody has Treg-selective agonistic biological activities, which is a desirable property in therapeutic intervention for various autoimmune diseases.

REFERENCES: NIL.

Acknowledgements: NIL.

Disclosure of Interests: Takahiro Miyazaki Nektar Therapeutics, Mekhala Maiti Nektar Therapeutics, Inbar Amit Biolojic Design, Damon Hamel Nektar Therapeutics, Qiuxia Wu Nektar Therapeutics, Rhoneil Pena Nektar Therapeutics, Katherine Brendza Nektar Therapeutics, Thomas Chang Nektar Therapeutics, Vivian Guo Nektar Therapeutics, Saul Kivimae Nektar Therapeutics, Yanay Ofran Biolojic Design, Jonathan Zalevsky Nektar Therapeutics.