A Non-immunogenic Bivalent d-Protein Potently Inhibits Retinal Vascularization and Tumor Growth

Abstract

We report a general approach to engineering multivalent D-proteins with antibody-like activities in vivo. Mirror-image phage display and structure-guided design were utilized to create a Dprotein that uses receptor mimicry to antagonize vascular endothelial growth factor A (VEGF-A). Selections against the D-protein form of VEGF-A using phage-displayed libraries of two different domain scaffolds yielded two proteins that bound distinct receptor interaction sites on VEGF-A. X-ray crystal structures of the D-protein/VEGF-A complexes were used to guide affinity maturation and to construct a heterodimeric D-protein VEGF-A antagonist with picomolar activity. The D-protein VEGF-A antagonist prevented vascular leakage in a rabbit eye model of wet age-related macular degeneration and slowed tumor growth in the MC38 syngeneic mouse tumor model with efficacies comparable to those of approved antibody drugs, and in contrast with antibodies, the D-protein was non-immunogenic during treatment and following subcutaneous immunizations.

Single-Shot Flow Synthesis of D-Proteins for Mirror-Image Phage Display

“…2). Taken together, these syntheses represent a major fraction of the total number of D-proteins reported in the past five years (12 out of 36 reported targets [2][3][4]11,12,[16][17][18][23][24][25][26][27][28][29][30][31][31][32][33][34][35][36][37] ). We aimed to apply a recently reported screening platform based on phage display to generate high-affinity conformationally constrained ("stapled") α-helical peptides, termed Helicons, in a single MIDP screening pass 38 .…”

“…First reported almost 30 years ago MIDP has been applied to a number of important protein targets, but has not yet been implemented broadly. As highlighted by the 2018 Nobel Prize awarded in part for the development of phage display, these display techniques have become ubiquitous, but examples of screening on D-proteins remain rare -to the best of our knowledge, MIDP has only been applied to nine protein targets 7,[9][10][11][12][13][14][15][16][17][18][19] .…”

D‐Peptide and D‐Protein Technology: Recent Advances, Challenges, and Opportunities**

“…Of note, RFX037.D was non‐immunogenic in mice, whereas the L‐enantiomer generated a strong immune response. In an extension of the MIPD against VEGF‐A, bivalent D‐protein ligands were also developed using orthogonal MIPD assays with two different scaffold mini‐proteins (53 and 58 residues) [75] . The two best scaffolds were connected via a covalent linkage to yield the bivalent D‐protein RFX‐V1a2a, with sub‐nanomolar ( K D =0.8 nM) affinity for VEGF‐A.…”

D-Peptide and D-Protein technology: Recent advances, challenges, and opportunities

“…In an extension of the MIPD against VEGF-A, bivalent D-protein ligands were also developed using orthogonal MIPD assays with two different scaffold mini-proteins (53 and 58 residues). [72] The two best scaffolds were connected via a covalent linkage to yield the bivalent D-protein RFX-V1a2a, with sub-nanomolar (KD = 0.8 nM) affinity for VEGF-A.…”

“…Particularly, native chemical ligation has facilitated the synthesis of numerous D-proteins, including enantiomeric venom toxins, [30, 36-38, 40, 46, 49, 51, 53, 58] growth factors [57,69,[71][72] and enzymes, [6,8,[11][12][13]117] including the 90 kDa split-enzyme D-Pfu. [15] The pursuit of more complex D-proteins is perplexed by their size and posttranslational modification status.…”