Precursor ion exclusion for enhanced identification of plasma biomarkers

Metabolism Unit, National Institute on Aging, National Institutes of Health, Baltimore, MD 21224, USA.
PROTEOMICS - CLINICAL APPLICATIONS (Impact Factor: 2.96). 06/2012; 6(5-6):304-8. DOI: 10.1002/prca.201100107
Source: PubMed


Our study aims to establish a plasma biomarker analysis workflow, with fewer fractionation steps, for enhanced identification of plasma biomarkers by precursor ion exclusion (PIE).
Plasma samples were depleted for highly abundant proteins, then further fractionated by molecular weight (MW), before trypsinization for LTQ-Orbitrap mass analysis. Data-dependent acquisition (DDA) was used for baseline analysis. PIE involves the re-injection of samples with exclusion of the previously identified peptides. We compared analyses using multiple PIE iterations, compared to DDA, for plasma interrogation
A higher percentage of unique plasma peptides was identified with PIE, compared to DDA. The first PIE iteration reveals an increase of 75-112% more peptides than the DDA method alone. PIE can interrogate complex plasma samples with the percentage of peptides identified successively increasing with even ≥4 iterations. The total number of peptides identified increases rapidly across the first three PIE iterations and then continues more slowly up to nine iterations.
Iterative injections with PIE resulted in many more peptide identifications in plasma samples of varying degrees of complexity, compared to re-injections using similar DDA parameters. PIE methods may therefore expand our ability to recover plasma peptides for plasma biomarker discovery.

Download full-text


Available from: Stuart Maudsley
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To fully appreciate the diversity and specificity of complex cellular signaling events, such as arrestin-mediated signaling from G protein-coupled receptor activation, a complex systems-level investigation currently appears to be the best option. A rational combination of transcriptomics, proteomics, and interactomics, all coherently integrated with applied next-generation bioinformatics, is vital for the future understanding of the development, translation, and expression of GPCR-mediated arrestin signaling events in physiological contexts. Through a more nuanced, systems-level appreciation of arrestin-mediated signaling, the creation of arrestin-specific molecular response "signatures" should be made simple and ultimately amenable to drug discovery processes. Arrestin-based signaling paradigms possess important aspects, such as its specific temporal kinetics and ability to strongly affect transcriptional activity, that make it an ideal test bed for next-generation of drug discovery bioinformatic approaches such as multi-parallel dose-response analysis, data texturization, and latent semantic indexing-based natural language data processing and feature extraction.
    Full-text · Article · Jun 2013 · Progress in molecular biology and translational science
  • [Show abstract] [Hide abstract]
    ABSTRACT: Liquid chromatography-mass spectrometry (LC-MS) based proteomics is one of the most widely used analytical platforms for global protein discovery and quantification. One of the challenges is the difficulty of identifying low abundance biomarker proteins from limited biological samples. Extensive fractionation could expand proteomics dynamic range, however, at the cost of high sample and time consumption. Extensive fractionation would increase the sample need and the labeling cost. Also quantitative proteomics depending on high resolution MS have the limitation of spectral acquisition speed. Those practical problems hinder the in-depth quantitative proteomics analysis such as tandem mass tag (TMT) experiments. We found the joint use of hydrophilic interaction liquid chromatography (HILIC) and strong cation exchange Chromatography (SCX) prefractionation at medium level could improve MS/MS efficiency, increase proteome coverage, shorten analysis time and save valuable samples. In addition, we scripted a program, Exclusion List Convertor (ELC), which automates and streamlines data acquisition workflow using the precursor ion exclusion (PIE) method. PIE reduces redundancy of high abundance MS/MS analyses by running replicates of the sample. The precursor ions detected in the initial run(s) are excluded for MS/MS in the subsequent run. We compared PIE methods with standard data dependent acquisition (DDA) methods running replicates without PIE for their effectiveness in quantifying TMT-tagged peptides and proteins in mouse tears. We quantified a total of 845 proteins and 1401 peptides using the PIE workflow, while the DDA method only resulted in 347 proteins and 731 peptides. This represents a 144% increase of protein identifications as a result of PIE analysis.
    No preview · Article · Jan 2015 · Journal of Proteomics & Bioinformatics
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Spectral count analysis via data-dependent acquisition (DDA) mode mass spectrometry is used as label-free protein quantification. However, combination of the DDA mode with exclusion list based DDA (DDA-EL) for the similar purpose has not yet been tested. Therefore, we have taken the initiative to check the protein abundance using DDA-EL and measured their suitability. To check the protein abundance correlation between different samples, multiple replicates of mass spectrometric analysis of peptides were conducted primarily in DDA mode. Subsequently, peptides were analyzed in multiple replicates in DDA-EL mode with an exclusion mass list prepared from the previous DDA analyses. The normalized spectral abundance factor (NSAF) for each identified protein was compared among replicated datasets of single DDA, DDA-EL, merged two DDAs, and merged DDA + DDA-EL or between different types of datasets. A strong and linear NSAF correlation with an average correlation coefficient of 0.939 was observed in the comparison between each pair of DDA data. Similar connotation was also monitored in the comparison among DDA-EL data (r =0.928) or among merged DDA + DDA-EL data (r =0.960) while a reduced correlation coefficient (r =0.892) with increased deviation was marked between DDA and DDA-EL data. Evaluation of protein abundance patterns from different cellular states can successfully be conducted by DDA-EL-based mass spectrometric analysis. Therefore, the new workflow, DDA-EL merged to DDA mode, is a potential alternative to protein identification and quantification method. Copyright © 2014 John Wiley & Sons, Ltd. Copyright © 2014 John Wiley & Sons, Ltd.
    Full-text · Article · Jan 2015 · Rapid Communications in Mass Spectrometry