Discovering Bioactive Peptides and Characterizing the Molecular Pathways that Control Their Activity

Abstract

Bioactive peptides constitute a major class of signaling molecules in animals and have been shown to play a role in diverse physiological processes, including hypertension, appetite and sleep. As a result, knowing the identity of these molecules and understanding the mechanisms by which they are regulated has basic and medical significance. In this dissertation, I describe the development and application of novel methods for discovering bioactive peptides and the molecular pathways that control their activity. Recent analyses of mammalian RNAs have revealed the translation of numerous short open reading frames (sORFs). However, it is unknown whether these translation events produce stable polypeptide products that persist in the cell at functionally relevant concentrations. In Chapter 1, I describe a study in which we used a novel mass spectrometry-based strategy to directly detect sORF-encoded polypeptides (SEPs) in human cells. This analysis identified 115 novel SEPs, which is the largest number of mammalian SEPs discovered in a single study by more than a factor of 25. We observed widespread translation of SEPs from non-canonical RNA contexts, including polycistronic mRNAs and sORFs defined by non-AUG start codons. We also found that SEPs possess properties characteristic of functional proteins, such as stable expression, high cellular copy numbers, post-translational modifications, sub-cellular localization, the ability to participate in specific protein-protein interactions and the ability to influence gene expression. Taken together, these findings provide the strongest evidence to date that coding sORFs constitute a significant human gene class. In chapter 3, I describe a study in which we combine quantitative in vivo peptidomics, classical biochemical experiments and pharmacological studies in animal models to elucidate the metabolism of the neuropeptide substance P in the spinal cord. We identified two physiological substance P metabolites: the N- terminal fragments SP(1-9) and SP(1-7). Focusing our efforts on the SP(1-9)- producing pathway, we determined that an activity sensitive to the inhibitor GM6001 is the dominant SP(1-9)-generating activity in the spinal cord. We also show that GM6001 treatment causes a nearly three-fold increase in endogenous substance P levels in the spinal cords of mice, highlighting the functional relevance of the pathway blocked by this inhibitor.