AI-assisted design of a VEGFR2 agonistic peptide that promotes angiogenesis and wound repair.

Vascular endothelial growth factor (VEGF) serves as a principal regulator of neovascularization and is essential for vascular regeneration. In this study, we identified and evaluated VEGF Mimetic Peptide3 (VMP3), a novel synthetic peptide developed as a potential VEGFR-2/VEGF-A modulator. VMP3 was generated using an AI-assisted peptide design framework incorporating long short-term memory (LSTM) networks, which enabled learning of sequence-level patterns within known VEGF agonist peptides and assisted the motif-guided and structure-informed design of new candidates. Molecular dynamics simulations suggested a stable interaction profile between VMP3 and VEGFR2, supporting the feasibility of receptor engagement. Experimental validation was performed using in vitro, ex vivo, and in vivo methods. In vitro functional assays showed that VMP3 promoted endothelial cell proliferation, migration, and invasion in a manner similar to that of native VEGF. Ex vivo aortic ring assays supported these findings, demonstrating micro vessel sprouting following VMP3 treatment. An in vivo Matrigel plug assay in C57BL/6J mice revealed increased neovascularization following VMP3 treatment, as confirmed by histological and immunohistochemical analyses (H&E and CD31 staining). In addition, in a diabetic mouse wound healing model, VMP3 significantly accelerated wound closure, and tissue-level analysis of wound sites confirmed enhanced re-epithelialization and neovascularization, as evidenced by Masson's trichrome, Picrosirius Red, H&E, and CD31 staining. Collectively, these results suggested that VMP3 functions in a manner similar to that of VEGF, promoting angiogenesis.