Abstract
<jats:p>We propose that supramolecular peptide networks can respond to multivalent metabolites through competing or cooperative sidechain interactions. This approach could offer a route to use metabolites to tune network-level properties. Here, we provide computational analysis, using atomistic molecular dynamics simulations showing how a peptide system enables selective recognition of two metabolites: serotonin and kynurenic acid through re-distribution of multivalent interactions. Mixtures of four (XZXZ) tetrapeptides combining one of three aromatic (X = phenylalanine (F), tyrosine (Y), or tryptophan (W)) and combinations of charged (Z = glutamic acid (E) or lysine (K)) residues assemble, with aromatic identity and peptide sequence dictating system-level distribution of interactions. Introduction of kynurenic acid or serotonin reorganizes peptide-peptide contacts into distinct, metabolite-specific sidechain interaction patterns by integrating metabolite into the network. Recognition proceeds through, multivalent ensembles, with early aromatic association followed by electrostatic stabilization. The tryptophan-based systems show network response, with K-carboxylate and E-amine electrostatic contacts complemented by aromatic interactions with WKWK and WEWE respectively dominating kynurenic acid and serotonin interactions. This demonstrates that computational analysis can be used to show how minimal, flexible peptide mixtures achieve selective recognition through network reorganization, providing a systems chemistry framework for distinguishing related metabolites without binding pockets.</jats:p>