Identification of Microproteins in Saccharomyces cerevisiae under Different Stress Conditions

Abstract

Small open reading frame-encoded peptides (SEPs) are microproteins with a length of 100 amino acids or less, which may play a critical role in maintaining cell homeostasis under stress. Therefore, we used mass spectrometry-based proteomics to explore microproteins potentially involved in cellular stress responses in Saccharomyces cerevisiae. A total of 225 microproteins with 1920 unique peptides were identified under six culture conditions: normal, oxidation, starvation, ultraviolet radiation, heat shock, and heat shock with starvation. Among these microproteins, we found 70 SEPs with 75 unique peptides. The annotated microproteins are involved in stress-related processes, such as cell redox reactions, cell wall modification, protein folding and degradation, and DNA damage repair. It suggests that SEPs may also play similar functions under stress conditions. For example, SEP IP_008057, translated from a short coding sequence of YJL159W, may play a role in heat shock. This study identified stress-responsive SEPs in S. cerevisiae and provided valuable information to determine the functions of these proteins, which enrich the genome and proteome of S. cerevisiae and show clues to improving the stress tolerance of S. cerevisiae.

Biological Factors and Statistical Limitations Prevent Detection of Most Noncanonical Proteins by Mass Spectrometry

“…Its translation product was not identified in the Sun et al analysis. 25 A clear pattern of ribo-seq read triplet periodicity was observed in the frame of YNL155C-A (i.e., reads tend to match to the first position of a codon) before the overlap with YNL156C, indicating translation in this frame ( Figure 6B ). There also appears to be a triplet periodic pattern in a frame distinct from both YNL156C and YNL155C-A at the locus, suggesting that all three frames may be translated.…”

“…All mass spectrometry data files were taken from three studies. The He et al 2018 23 25 dataset PXD028623 was downloaded from IPROX. These datasets were searched using all proteins predicted to be encoded from the full reference translatome described in Wacholder et al 2021.…”

“…Using the MSFragger program 26 , we searched the three aforementioned published MS datasets optimized for detection of short, lowly expressed proteins [23][24][25] against a sequence dataset that included all 5,968 canonical yeast proteins on Saccharomyces Genome Database (SGD) 27 as well as predicted proteins from 18,947 noncanonical ORFs (including both unannotated ORFs and ORFs annotated as "dubious") inferred to be translated in Wacholder et al 2023 7 on the basis of ribosome profiling data. The spectra from the three studies were pooled and false discovery rates (FDR) were estimated separately for canonical and noncanonical ORFs using a target-decoy approach.…”

“…We do not rule out a role for YIL060W, but all considered evidence provides greater support for the biological significance of YIL059C. 25 We therefore constructed a sequence database consisting of all canonical ORFs as well as noncanonical ORFs that overlap canonical ORFs on the same strand, with ORFs determined only from the genome sequence rather than expression evidence.…”

“…Gao et al 2021 24 also used a combination of strategies to detect many small and low abundance proteins. Sun et al 2022 25 searched for unannotated microproteins in a variety of stress conditions and found 70, all expressed from alternative reading frames of canonical coding sequences. At the same time as these studies provided increased coverage of the yeast proteome, Wacholder et al 2023 7 integrated ribo-seq data from hundreds of experiments in over 40 published studies and assembled a high-confidence yeast reference translatome including 5372 canonical protein-coding genes and over 18,000 noncanonical ORFs.…”