Predicting the Supramolecular Assembly of Amphiphilic Peptides from Comprehensive Coarse-Grained Simulations

Nanoscale molecular architectures that are pH-switchable have wide applications in biomedicine, nanoelectronics, and catalysis and can be realized through the self-assembly of peptides with charged amino acid groups into superstructures. We investigate the morphologies and assembly kinetics of stimuli-responsive supramolecular polymers combining experimental data from transmission electron microscopy and circular dichroism spectroscopy with coarse-grained molecular dynamics simulations. Two types of C 3-symmetric building blocksbased on either glutamic acid (acidic) or lysine (basic) groupsare studied in aqueous buffer solution at varying pH. Both are known to form aggregates at opposite pH conditions, indicating that for extremely acidic and basic conditions, one type of peptide favors the dispersed state, while the other assembles into homopolymers. This implies a pH-dependent homo-co-homopolymer transition in bidisperse solutions of acidic and basic peptides, which is difficult to identify experimentally. We develop a generic and extensible coarse-grained model for charged peptides and demonstrate that it efficiently reproduces the experimentally observed states for monodisperse solutions in a wide range of pH values. For equimolar mixtures, the simulations confirm a homo-to-copolymer transition, whereby copolymers exhibit predominantly alternate stacking of the two types of monomers. Our study demonstrates that the interplay of hydrophobic and electrostatic interactions determines pH-responsive supramolecular self-assembly of amphiphilic peptides with complementary charged residues.