AI-guided design and optimization of a novel KIM-1-targeted peptide for bFGF delivery in acute kidney injury repair.

Acute kidney injury (AKI) remains a major clinical challenge due to its high incidence and mortality rates. Growth factor therapy has emerged as a promising strategy in AKI treatment, but it is limited by low targeting efficiency. Kidney injury molecule (KIM-1) is up-regulated after AKI and serves as a potential target for growth factor delivery systems. In this study, de novo designed WEV peptide targeting KIM-1 was screened to systematically optimize the affinity and specificity using a structure-based computational 'anchor extension' strategy integrated with deep learning approaches. The WEV peptide was computationally predicted and experimentally validated to have superior specificity and binding affinity with KIM-1 in vitro and in vivo. Then, a KIM-1-targeted recombinant WEV-bFGF protein was constructed to direct bFGF specifically to the ischemic kidney, which highly expressed KIM-1.This targeted delivery of WEV-bFGF could protect ischemic kidney, decrease cell apoptosis and inflammatory response, alleviate fibrosis and improve renal function. This process was revealed to activate tissue regeneration-related genes, while inhibiting genes related to apoptosis and inflammation, as determined by transcriptome analysis. Therefore, this optimized WEV peptide held a potential for a KIM-1 targeted growth factor delivery system and provided a new therapeutic strategy for AKI treatment.