Computational protocol: Integrated omics dissection of proteome dynamics during cardiac remodeling

Similar protocols

Protocol publication

[…] Protein turnover and abundance measurements were performed as described in detail in the open data descriptor. Briefly, mouse plasma was analyzed with gas chromatography mass spectrometry. Plasma sample from each time point (20 µL) was mixed with NaOH (10 N, 2 µL) and acetone (5% v/v, 5 µL) in acetonitrile. Standard curves were created by mixing to the acetone 0–20% molar ratio of 2H2O at 11 regular intervals in phosphate-buffered saline in lieu of mouse plasma sample. The sample mixtures were incubated at ambient temperature overnight. Acetone was extracted by adding chloroform (500 µL) and anhydrous sodium sulfate (500 mg). The extracted solution (1 µL) was analyzed on a gas chromatography mass spectrometer (Agilent 6890/5975) with a J&W DB17-MS capillary column (Agilent, 30 m × 0.25 mm × 0.25 µm) at the UCLA Molecular Instrumentation Center. The column temperature gradient was as follows: 60 °C initial, 20 °C·min−1 increase to 100 °C, 50 °C min−1 increase to 220 °C, 1 min hold. The mass spectrometer operated in the electron impact mode (70 eV) and selective ion monitoring at m/z 58 and 59 with 10 ms dwell time.First-dimension (high pH) separation was conducted by resolving peptides on a Phenomenex C18 column (Jupiter Proteo C12, 4 µm particle, 90 Å pore, 100 mm × 1 mm dimension) at high pH using a Finnigan Surveyor liquid chromatography system. The solvent gradient was as follows: 0–2 min, 0–5% B; 3–32 min, 5–35% B; 32–37 min, 80% B; 50 µL min−1; A: 20 mM ammonium formate, pH 10; B: 20 mM ammonium formate, 90% acetonitrile, pH 10. Fifty micrograms of proteolytic peptides were injected with a syringe into a manual 6-port/2-position switch valve. Twelve fractions from 16–40 min were collected, lyophilized and re-dissolved in 20 µL 0.5% formic acid with 2% acetonitrile prior to low-pH reversed-phase separation. On-line second-dimension (low-pH) reversed-phase chromatography was performed on all samples using an Easy-nLC 1000 nano-UPLC system (Thermo Scientific) on an EasySpray C18 column (PepMap, 3 µm particle, 100-Å pore; 75 µm × 150 mm dimension; Thermo Scientific) held at 50 °C. The solvent gradient was 0–110 min: 0–40% B; 110–117 min: 40–80% B; 117–120 min: 80% B; 300 nL min−1; A: 0.1% formic acid, 2% acetonitrile; B: 0.1% formic acid, 80% acetonitrile. Each high pH fraction was injected (10 µL) by the autosampler to the Easy-nLC 1000 nano-UPLC system.High-resolution Fourier-transform tandem mass spectrometry was performed on an LTQ Orbitrap Elite instrument (Thermo Scientific), coupled on-line to an Easy-nLC 1000 nano-UPLC system (Thermo Scientific) through a Thermo EasySpray interface. Signals were acquired in FT/IT mode: each FT MS1 survey scan was analyzed at 60,000 resolving power in profile mode, followed by IT MS2 scans on the top 15 ions. MS1 and MS2 target ion accumulation targets were 1.0E4 and 1.0E6, respectively. MS1 lock mass (m/z 425.120025) and dynamic exclusion (90 s) were used. Peptide identification was performed using the database search algorithm ProLuCID against a reverse-decoyed protein sequence database (Uniprot Reference Proteome, reviewed, accessed April-08–2014, 16,672 forward entries). Static cysteine carboxyamidomethylation (57.02146 Da) and ≤3 of the following variable modifications were allowed: methionine oxidation (15.9949 Da), lysine acetylation (42.0106 Da), serine/threonine/tyrosine phosphorylation (79.9663 Da), lysine ubiquitylation (114.0403 Da), and asparagine deamidation (0.9840 Da). Tryptic, semi-tryptic, and non-tryptic peptides within a 20-ppm mass window surrounding the candidate precursor mass were searched. Protein inference was performed using DTASelect (v.2.0), requiring ≤1% global peptide false discovery rate and 2 unique peptides per protein. Modified or non-tryptic peptides were subjected to separate statistical filters to limit false discovery.Mass spectra were converted into mzML format using ProteoWizard (v.2.1), then input for analysis in ProTurn (v.2.0.5). ProTurn retrieved identified peptides that were uniquely assigned to proteins for area integration. Extracted ion chromatographs were processed with Savitzky-Golay filters, and the areas-under-curves within 60 ppm of the peptide mass at the retention time of the identified peptides were integrated. Peptides that were explicitly identified and integrated in ≥4 time points were accepted for the calculation of protein abundance and turnover. The areas-under-curve of up to the first six peptide isotopomers for each qualifying peptide at each time point are integrated over retention time space for abundance and turnover calculation. To calculate abundance, the summed areas of all isotopomers within each peptide envelope were normalized against total spectral intensity, then normalized against the number of possible tryptic peptides (6–30 amino acids) from in silico digestion of the protein. Abundance changes are presented as log2 ratios of abundance at each hypertrophy time points over control day 0 samples. To calculate turnover rates, the fractional abundance of the first mass isotopomer from each integrated time point (A0) was modeled with a compound kinetic curve and the best-fit parameter was optimized using the Nelder-Mead algorithm. The A0 values measured by the mass spectrometer from each of seven time-points served as independent biological replicates to allow estimation of the best-fitted values as well as variance of the turnover rate constant (k). Only a subset of peptide isotopomer time-series with goodness-of-fit ≥0.8 or standard error of estimate ≤0.05 were accepted for turnover calculation. Protein-level turnover rate was reported as the median and median absolute deviation of the best-fitted turnover rate constant of each accepted constituent peptide that is unique in the proteome. Protein turnover flux is reported as the product between the turnover rate constant and fractional abundance.Tandem mass tags (10-plex) were purchased from Thermo Scientific. Peptides were labeled according to manufacturer’s instruction. The TMT10plex quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Only peptide spectra containing all reporter ions were designated as quantifiable. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the peptides pertaining to that protein. […]

Pipeline specifications

Software tools ProLuCID, DTASelect, ProteoWizard, Proteome Discoverer
Application MS-based untargeted proteomics
Organisms Mus musculus