A Proteogenomic Approach to Discovering Translational Peptides from Untranslated Regions

Abstract

Translation initiation and termination are key steps in translation and expression of gene products, because they contribute to the diversity of protein forms from a single gene, which affect the functionality of the protein. In molecular genetics, translation usually initiates at a start codon (usually AUG) and terminates at a stop codon (usually UAA, UAG or UGA). However, this mechanism is overlooked in terms of protein expression, regulation, and interaction. In this paper, we found out the alternative translation initiation and termination at untranslated regions (UTRs) by assuming the exceptional mechanisms, such as leaky scanning or translational read-through. To identify alternative selection of translation initiation and termination, we constructed a novel protein sequence database, “tUTRDB (translated UTR database),” which included in silico translation of sequences from UTR. In doing so, we assumed that the START codons applied with single nucleotide variants (SNVs) in 5’-UTR can serve as an alternative START codon. Similarly, we assumed that SNVs to the STOP codon can possibly give a new signal to adopt an alternative STOP codon. Over 100 million mass spectra obtained from cancer patients were searched against tUTRDB. A multi-stage search method was used to effectively identify novel peptides while controlling false-positives. The study was conducted using 80 tissue samples from patients with early onset gastric cancer and 454 new unique peptides (290 in 5’-UTR and 164 in 3’-UTR regions) were identified as potential translated UTR peptides. Further bioinformatics analysis shows that these newly found 5’-UTR peptides were likely to be expressed due to alternative translation initiation.