Proteome discover

Multiple strategies were employed to identify peptides. Proteome Discover (ThermoFisher version 2.1) was used to identify peptides from LC-MS/MS data. Enzyme specificity was set to unspecific. The mass tolerance for fragments and peptides were 20 ppm and 0.02 Da, respectively. The sequences were searched against the Zea mays sequences in the UniProt database (http://www.uniprot.org/, accessed on 2 March 2021, UPID: UP000007305) and concatenated to the database of proteomics research. The fixed modifications were carbamidomethyl-Cys, 6-plex TMT at the N-termini and Lys-6-plex TMT, but Met oxidation was variable. Each peptide quantification value was exported to an Excel output file, and the average peptide ratios or fold change (FC = treatment/control) were determined by dividing the quantification value of each peptide in the treated samples by the quantification value of control samples. Differentially expressed peptides (DEPs) represented peptides with FC ≥ 1.25 (up-regulated) or ≤0.8 (down-regulated), along with significantly different abundances (p-value < 0.05). The peptidomics data were uploaded to iProX database (iProX ID: IPX0005992000).

The uniprot worms database (20161228, 17,392 sequences) was searched for raw data using Proteome Discover (Version 1.4, Thermo Fisher Scientific, Bremen, Germany) followed by scoring with the Mascot server (Version 2.3, Matrix Science, London, UK). The searched parameters were set as follows: 10 ppm for the precursor ion mass deviation and 0.05 Da for the allowed fragment mass deviation; up to two miscleavage sites; cysteine carbamidomethylation, iTRAQ 8-plex on lysine and the N-terminal set as fixed modifications; methionine oxidation; and STY (serine, threonine and tyrosine) phosphorylation as variable modifications. The Target Decoy PSM verification program incorporated in the Proteome Discoverer was uitilized to verify the search results using only matches with FDR ≤ 0.01. Functional classifications of phosphoproteins were analyzed using the David database, as previously described [64 (link)]. Subcellular locations were predicted through Wolf PSORT [65 (link)]. Protein–protein interactions were detected by STRING [32 (link)]. The motifs of protein phosphorylation were derived from our phosphoproteomic data using the Motif-X algorithm [66 (link)]. The kinase-substrate phosphorylation network (KSPN) was constructed by submitting differentially expressed phosphopeptides by using iGPS 1.0 software (the CUCKOO workgroup, DICP, Dalian, China) [35 (link)].

All generated raw files were submitted to Proteome Discover (PD, version 2.3, Thermo Fisher Scientific, Germany) with label-free quantitation (LFQ) analysis. The human and mouse protein sequence databases were downloaded from UniProt database in May 2019 (http://www.uniprot.org). Database searching was performed with the following parameters: cysteine carbamidomethylating (C, +57.0215 Da) as a fixed modification; methionine oxidation (M, +15.9949 Da) and N-terminal acetylation (+42.010565 Da) as variable modifications; up to two missed cleavage sites were permitted for trypsin digestion; the tolerances of precursor and fragment masses were set at 10 ppm and 0.02 Da, respectively; 1% protein false discovery rate (FDR) was used as the filter for both protein and peptide identification, and at least two peptide-spectrum matches (PSMs) were required for a peptide identification. Label-free method in PD was used for relative quantification of proteins among samples. LFQ was performed for calculation of protein abundances. Protein ratios were calculated as the median of all pairwise ratios calculated between the three replicates of all peptide abundances.

The Ψ-PTC mRNAs were compared to uridine containing (non-Ψ) PTC mRNAs (Supplementary Table S8). Flag-eGFP peptides translated in HEK-293T cells were purified with anti-Flag M2 magnetic beads (Sigma). Pulled down proteins were extensively washed with 50 mM ammonium acetate and were directly digested on the beads adding Trypsin and alkylated with iodoacetamide (55 mM). Peptides were analyzed using a Dionex, UltiMate 3000 nano-HPLC system (Germering, Germany) coupled via nanospray ionization source to a Thermo Scientific Q Exactive HF mass spectrometer (Vienna, Austria) using instrument settings as described previously (44 (link)). The database search was performed using ProteomeDiscover (Version 2.1, Thermo Scientific). Sensitivity of the assay was calculated by the median abundance of the 10 least abundant peptides in the least sensitive sample, out of the total abundance of all of the C-terminal peptides in any of the samples.

Micrographs of Lacusarx virus and identification of phage proteins from purified CsCl gradient centrifugation was performed as described previously^{30 (link)}. Briefly, 75 µl of the purified phage solution was mixed with 75 µl of 1% SDS and incubated for 30 min at 80 °C followed by TCA precipitation. The proteins were re-solubilized in 8 M urea, 45 mM DTT, and 50 mM Tris, pH 8.0, reduced and alkylated. The proteins were digested with trypsin and the resulting peptides were analysed using a Dionex 3000 RSLC UHPLC system (ThermoFisher Scientific, Hvidovre, Denmark) with an Aeris PEPTIDE 1.7 µm XB-C18, 150 × 2.1 mm column (phenomenex, Vaerloese, Denmark) coupled with a Q Exactive mass spectrometer (ThermoFisher Scientific, Hvidovre, Denmark). The resulting data was analysed with Proteome Discover (version 1.4, ThermoFisher Scientific, Hvidovre, Denmark) using a homemade protein database based on the obtained DNA sequences. The search results were filtered in Proteome Discover with the integrated Target decoy PSM validator algorithm to a q-value of <0.01, which ensures a peptide-spectrum match false discovery rate less than 0.01.

Raw MS data from 22 fractions were searched against Swiss-Prot database (March 8, 2015; Homo sapiens; 54,7964 sequences—after human taxonomy filter applied—20,203) using the MASCOT software (Matrix Science Ltd, version 2.2) through Proteome Discover software (Thermo Fisher Scientific, version 1.4). Mascot significance threshold was set to 0.05. Trypsin/P was specified as the cleavage enzyme allowing up to two missed cleavages. The database search included the following parameters: MS1 Tolerance: 10 ppm, MS2 Tolerance: 0.06 Da, fixed modification: Carbamidomethyl (C). Variable Modification: Oxidation (M), Dioxidation (M), Acetyl (N-term), Gln->pyro-Glu (N-term Q), hydroxyl (P), TMT10plex (N-term), and TMT10plex (K). The data were filtered by applying a 1% false discovery rate. Quantified proteins were filtered if the absolute fold-change difference between conditions to respective DMSO of the same biological replicate was ≥1.3 for upregulated proteins and ≤0.7 for downregulated proteins.