Molecular Scale Structure and Kinetics of Layer-by-Layer Peptide Self-Organization at Atomically Flat Solid Surfaces.

Understanding the mechanisms of self-organization of short peptides into two- and three-dimensional architectures are of great interest in the formation of crystalline biomolecular systems and their practical applications. Since the assembly is a dynamic process, the study of structural development is challenging at the submolecular dimensions continuously across an adequate time scale in the natural biological environment, in addition to the complexities stemming from the labile molecular structures of short peptides. Self-organization of solid binding peptides on surfaces offers prospects to overcome these challenges. Here we use a graphite binding dodecapeptide, GrBP5, and record its self-organization process of the first two layers on highly oriented pyrolytic graphite surface in an aqueous solution by using frequency modulation atomic force microscopy in situ. The observations suggest that the first layer forms homogeneously, generating self-organized crystals with a lattice structure in contact with the underlying graphite. The second layer formation is mostly heterogeneous, triggered by the crystalline defects on the first layer, growing row-by-row establishing nominally diverse biomolecular self-organized structures with transient crystalline phases. The assembly is highly dependent on the peptide concentration, with the nucleation being high in high molecular concentrations, e.g., >100 µM, while the domain size is small, with an increase in the growth rate that gradually slows down. Self-assembled peptide crystals are composed of either singlets or doublets establishing P1 and P2 oblique lattices, respectively, each commensurate with the underlying graphite lattice with chiral crystal relations. This work provides insights into the surface behavior of short peptides on solids and offers quantitative guidance toward elucidating molecular mechanisms of self-assembly helping in the scientific understanding and construction of coherent bio/nano hybrid interfaces.

Data-driven design of a multiplexed, peptide-sensitized transistor to detect breath VOC markers of COVID-19.

Exhaled human breath contains a rich mixture of volatile organic compounds (VOCs) whose concentration can vary in response to disease or other stressors. Using simulated odorant-binding proteins (OBPs) and machine learning methods, we designed a multiplex of short VOC- and carbon-binding peptide probes that detect a characteristic "VOC fingerprint". Specifically, we target VOCs associated with COVID-19 in a compact, molecular sensor array that directly transduces vapor composition into multi-channel electrical signals. Rapidly synthesizable, chimeric VOC- and solid-binding peptides were derived from selected OBPs using multi-sequence alignment with protein database structures. Selective peptide binding to targeted VOCs and sensor surfaces was validated using surface plasmon resonance spectroscopy and quartz crystal microbalance. VOC sensing was demonstrated by peptide-sensitized, exposed-channel carbon nanotube transistors. The data-to-device pipeline enables the development of novel devices for non-invasive monitoring, diagnostics of diseases, and environmental exposure assessment.