| Summary: | COVID-19 pandemic has caused a catastrophic loss of human life. With only a few approved vaccine candidates and the slow rate of vaccine distribution, particularly in developing nations, there is a need for antiviral candidates and rapid diagnostic solutions. This thesis describes a hybrid pipeline that combines machine learning tools, energy-based simulations, and experimental validation to develop an ACE2-derived peptide that targets the viral spike protein receptor-binding domain (RBD). The peptide was derived utilizing the existing crystal structure of spike protein’s RBD and ACE2 to determine the linear peptide fragments that contributed the most to the binding energy of the complex. We tested these linear peptide fragments against the spike protein RBD using a degradation assay and identified a 23 amino acid length peptide fragment as a strong candidate for computational and experimental mutagenesis.
We also present a molecular beacon that detects SARS-CoV-2 spike protein through a conformational switch. Our molecular beacons contain two peptides that can form a parallel heterodimer and a binding ligand between them to detect the SARS-CoV-2 spike protein. A fluorophore-quencher pair is attached to the two ends of the heterodimer stems. In the absence of SARS-CoV-2 spike protein (OFF state), the peptide beacon has a hairpin conformation that opens upon binding to the spike protein and produces a fluorescence signal (ON state).
All of the pipelines developed as part of this thesis are applicable to other protein targets of interest.
|
|---|