Selection of peptide biomarkers in human bone collagen for the evaluation of deamidation (N, Q) and oxidation (M) by targeted proteomics.
We used a dataset of (56) human bones from Herculaneum (22), Pompeii (11), Baia Scalandrone (12), Oplonti (8), and S.Paolo Belsito(3). A shotgun proteomic approach was applied to the samples, and collagen type I chains have always been identified with good scores and confidence. Screening of all the mass spectrometric data was performed in search of the most frequently detected and/or deamidated peptides and the results were analyzed on a statistical basis to select the best peptides to be later used in the quantification of deamidation.
Steps to reproduce
For protein identification: The acquired MS/MS spectra were transformed in Mascot Generic files (.mgf) format and used to query the SwissProt database 2015_04 (548,208 sequences; 195,282,524 residues), with Homo sapiens as taxonomy restriction. A licensed version of Mascot software (www.matrixscience.com) version 2.4.0. was used with trypsin as enzyme; 3, as allowed number of missed cleavage; 10 ppm MS tolerance and 0.6 Da MS/MS tolerance; peptide charge from +2 to +3. No fixed chemical modification was inserted, but possible oxidation of methionines, deamidation at asparagines and glutamines and hydroxylation on lysine and proline were considered as variable modifications. Only proteins presenting two or more peptides were considered as positively identified. Individual ion score threshold provided by Mascot software to evaluate the quality of matches in MS/MS data. Spectra with Mascot score below 25 were rejected. Evaluation of the deamidation N,Q and oxidation (M) levels in collagen type I chains :The acquired MS/MS spectra were re-analyzed in standard MaxQuant searches in a UniProt database (759,512 sequences, 37,179,137 residues) with Homo sapiens as taxonomic restriction (20199 sequences, 928,813 residues). Samples have been analyzed as group datasets (5) based on the archaeological site. Deamidation of asparagine and glutamine residues (Asn, Gln), oxidation of methionine residues (Met), and hydroxylation of proline and lysine residues (Pro, Lys) were set as variable modifications. Additionally, MaxQuant’s “evidence.txt” file was used for a positional evaluation of the relative deamidation and oxidation, using an in-house code available at https://github.com/ismaRP/MSMSdeamidation. For each asparagine , glutamine and methionine position the values are calculated by dividing the sum of intensities of the peptides containing the modification by the total intensity of all the peptides containing that position. Only peptides with high probability modification site assignments (>30) were considered. In this way it was possible to evaluate whether the most frequent detected as deamidated (N,Q) and oxidized (M) positions of the previous step were also significantly modified.