De novo design of peptides that coassemble into β sheet–based nanofibrils

Abstract

Peptides' hierarchical coassembly into nanostructures enables controllable fabrication of multicomponent biomaterials. In this work, we describe a computational and experimental approach to design pairs of charge-complementary peptides that selectively coassemble into -sheet nanofibers when mixed together but remain unassembled when isolated separately. The key advance is a peptide coassembly design (PepCAD) algorithm that searches for pairs of coassembling peptides. Six peptide pairs are identified from a pool of ~10 6 candidates via the PepCAD algorithm and then subjected to DMD/PRIME20 simulations to examine their co-/self-association kinetics. The five pairs that spontaneously aggregate in kinetic simulations selectively coassemble in biophysical experiments, with four forming -sheet nanofibers and one forming a stable nonfibrillar aggregate. Solid-state NMR, which is applied to characterize the coassembling pairs, suggests that the in silico peptides exhibit a higher degree of structural order than the previously reported CATCH(+/−) peptides.

Engineering β-Sheet Peptide Coassemblies for Biomaterial Applications

“…The top six candidate peptides identified by the PepCAD algorithm were tested computationally and experimentally for coassembly. Five out of the six peptide pairs were confirmed to selectively coassemble, with four forming β-sheet nanofibers and one forming a stable nonfibrillar aggregate . NMR spectra of the four nanofiber-forming peptide pairs indicated a higher degree of structural order than had been observed for previous β-sheet coassemblies.…”

“…Evidence that nanofiber structure is not well-controlled in β-sheet peptide coassemblies inspired recent efforts to design new pairs computationally. Xiao et al developed a Monte Carlo (MC)-type peptide coassembly design (PepCAD) algorithm to discover potential selectively coassembling peptide pairs quickly and efficiently . Development of this algorithm follows recent success in computational designs of α-helical assemblies .…”

“…A low negative score during sequence evolution means that the peptide pair is more likely to form fibril-like coassemblies but not fibril-like self-assemblies. In applying the PepCAD algorithm, Xiao et al started with a pool of around 1 × 10 6 possible peptide pairs and identified six pairs of charge-complementary 11-mer peptides …”

“…Xiao et al developed a Monte Carlo (MC)-type peptide coassembly design (PepCAD) algorithm to discover potential selectively coassembling peptide pairs quickly and efficiently. 46 Development of this algorithm follows recent success in computational designs of α-helical assemblies. 47 The PepCAD algorithm proceeds by modifying amino acid sequences while evaluating a score that depends on the binding free energy and intrinsic self-aggregation propensity for a pair of peptides coassembled into a β-sheet scaffold.…”

“…In applying the PepCAD algorithm, Xiao et al started with a pool of around 1 × 10 6 possible peptide pairs and identified six pairs of charge-complementary 11-mer peptides. 46 Experimental and computational analysis of the computationally designed peptide pairs mostly indicate highly ordered βsheet nanofibers, but the PepCAD algorithm is not completely predictive of the final structure. The top six candidate peptides identified by the PepCAD algorithm were tested computationally and experimentally for coassembly.…”