Peptides and PTM identifications are very critical for studies related to food, and lots of non-specific cleavages are observed as a result of food-grade chemicals used. As a result, customized programming can play a crucial role in peptide identification. To study this effect, gelatine and one industrial gelatine hydrolysate from beef, pork and fish were investigated.
The results showed that when enzyme used (in this case Trypsin) is specific to 1 or 2 cleavage sites, the setting of the software plays a crucial role in the quality and quantity of identified peptides. The results showed that, irrespective of the species, with “no enzyme” selected, identified peptides were two times less in numbers. This finding is in contrast to the traditional thinking that without selecting the enzyme types, the number of identified peptides will increase. When the cleavage sites are highly specific, the difference between choosing and not choosing the enzyme type diminishes. This information is very important for agrifood industries.
The findings also confirm that the PTM setting is very important for agrifood studies. When 6 PTMs per peptides was chosen, the number of identified peptides was higher compared to traditional 3 PTMs per peptide setting. The final conclusion was that for the best outcome, one might need to know some details about the research or may need to optimize the settings of the software.
How was PEAKS used?
Database searches were performed using PEAKS Studio X+ software using the UniProtKB/Swiss-Prot databases restricted to Bos taurus (accessed March 2021—46,766 entries), Sus scrofa (accessed March 2021—120,911 entries), or Oreochromis niloticus (accessed July 2021—75,940 entires). Mass tolerance was set at 35 ppm for precursors and 0.2 Da for fragment ions. The data searches were performed, specifying trypsin as the enzyme, and three missed cleavage sites were allowed (for Tryp samples) or without specifying an enzyme (for other samples).
Search was performed in 2 forms: 1) methionine oxidation as variable PTM with maximum of 3 PTMs per peptide, 2) methionine oxidation and hydroxyproline as variable PTM with maximum of 6 PTMs per peptide. Data were filtered considering a p-value of 0.05 (p < 0.05) and a false discovery rate (FDR) < 1%.
Ambli, Mouna, et al. “Impact of Bioinformatics Search Parameters for Peptides’ Identification and Their Post-Translational Modifications: A Case Study of Proteolysed Gelatines from Beef, Pork, and Fish.” Foods 12.13 (2023): 2524. https://doi.org/10.3390/foods12132524
Abstract
Bioinformatics software, allowing the identification of peptides by the comparison of peptide fragmentation spectra obtained by mass spectrometry versus targeted databases or directly by de novo sequencing, is now mandatory in peptidomics/proteomics approaches. Programming the identification software requires specifying, among other things, the mass measurement accuracy of the instrument and the digestion enzyme used with the number of missed cleavages allowed. Moreover, these software algorithms are able to identify a large number of post-translational modifications (PTMs). However, peptide and PTM identifications are challenging in the agrofood field due to non-specific cleavage sites of physiological- or food-grade enzymes and the number and location of PTMs. In this study, we show the importance of customized software programming to obtain a better peptide and PTM identification rate in the agrofood field. A gelatine product and one industrial gelatine hydrolysate from three different sources (beef, pork, and fish), each digested by simulated gastrointestinal digestion, MS-grade trypsin, or both, were used to perform the comparisons. Two main points are illustrated: (i) the impact of the set-up of specific enzyme versus no specific enzyme use and (ii) the impact of a maximum of six PTMs allowed per peptide versus the standard of three. Prior knowledge of the composition of the raw proteins is an important asset for better identification of peptide sequences.