Single molecule peptide sequencing

The proteome is a highly dynamic and complex set of proteins, specific not only to a particular organism, but to cell types and environmental conditions. Understanding proteome changes as they occur is especially important for molecular diagnostics and developing biomarkers. Currently, the primary technology for proteome wide identification and quantification is shotgun mass spectrometry; while powerful, it lacks high sensitivity and coverage. In this dissertation, I discuss my work in the development of a new technology, termed “fluorosequencing”, for sequencing peptides from a complex protein sample at the level of single molecules. The concept is to generate a positional information pattern of an amino acid(s) (such as xKxxK, where K is lysine and x can be any of the other amino acid residues). In order to obtain such a pattern, we proposed a scheme of (i) selectively labeling one or more amino acid(s) in the peptides, (ii) immobilizing millions of these individual fluorescently labeled peptides on a glass surface, (iii) monitoring their changing fluorescent pattern by TIRF microscopy as the (iv) N-terminal amino acid is sequentially cleaved by Edman chemistry and (v) using the resulting fluorescent signature (fluorosequences) to uniquely identify individual single molecule peptides in the mixture. We began by developing a computational framework to justify the feasibility of the concept. By modeling different sources of anticipated errors, we showed that the errors do not greatly affect the identification of proteins in the human proteome. Secondly, after screening fluorophores for their solvent stability, we used fluorescently labeled synthetic peptides covalently immobilized on beads to experimentally demonstrate the ability of the technique to determine the position of the fluorescently labeled residue in peptides. Finally, we translated the bead optimized chemistry procedures to a single molecule setup. We implemented the fluorosequencing method to sequence synthetic peptide molecules and provided evidence for the technique’s utility to discriminate peptides in a peptide mixture with single molecule sensitivity. By establishing the foundational work towards the proof-of-principle for fluorosequencing, we can now scale the method in order to realize the idea of single molecule proteome wide sequencing.