AI-aided design and computational validation of a novel uranyl specific cyclic peptide motif.

The specific adsorption of uranyl ions is a key issue in the design of materials for Uranium Extraction from Seawater(UES). Cyclic peptides, with their flexible and tunable structures, can be engineered to selectively bind uranyl ions by adjusting their amino acid sequences. However, the vast chemical space of cyclic peptides presents a significant challenge for rapidly identifying motifs with high specificity for uranyl. To address this, we integrated artificial intelligence (AI), molecular dynamics (MD) simulations, and density functional theory (DFT) calculations to design cyclic peptides for specific uranyl binding. This approach led to the discovery of a novel cyclic peptide motif [sequence: cyc-DSDKSD(cyc-ASP-SER-ASP-LYS-SER-ASP)]. Independent 100-ns MD simulations confirmed stable uranyl binding, with a calculated binding energy of -70.4 kcal/mol. Wavefunction analysis revealed that the uranyl coordinates with oxygen atoms from carboxylate groups on the peptide, forming coordination bonds with lengths between 2.3 and 2.6 Å. Further analysis indicated that the binding is primarily driven by electrostatic and orbital interactions. We also systematically evaluated the binding modes and affinities of cyclic peptide with uranyl and other competing metal ions (K+, Ca2+, VO2+, Na+, Mg2+, Sr2+), with uranyl showing the strongest interaction (-649.6 kcal/mol). Alanine scanning results indicated that Asp is critical for maintaining uranyl coordination. This study demonstrates that AI-assisted design can significantly accelerate the development of materials for UES. The identified cyclic peptide motif can be grafted onto solid substrates to serve as an active binding site in next-generation materials for UES.