Elsevier

Journal of Proteomics

Volume 72, Issue 3, 13 April 2009, Pages 555-566
Journal of Proteomics

Validation of gel-free, label-free quantitative proteomics approaches: Applications for seed allergen profiling

https://doi.org/10.1016/j.jprot.2008.11.005Get rights and content

Abstract

Plant seeds provide a significant portion of the protein present in the human diet, but are also the major contributors of allergenic proteins that cause a majority of the reported cases of food-induced anaphylaxis. New varieties of grains and nuts as well as other seeds could be screened for allergen content before they are introduced as cultivars for food production using mass spectrometry-based quantitation approaches. Here, we present a practical comparison of gel-free and label-free methods, peak integration and spectral counting, using a linear trap mass spectrometer. The results show that both methods are linear and reproducible with protein standards from 5–200 ng, however, bioinformatic analysis for spectral counting is much simpler and therefore more amenable to high-throughput sample processing. We therefore applied spectral counting towards the analysis of transgenic peanut lines targeting the reduction of a prominent allergen. Spectral count analysis of an Ara h 2 (conglutin-7) RNA-silenced line confirmed reduction of this allergen as well as Ara h 6 (conglutin), which was further confirmed by quantitative immunoblotting. Other collateral changes include an increase in Ara h 10 (oleosin 1) in one of the three lines, a decrease in conarachin as well as increased 13-lipoxygenase and Ahy-3 (arachin) in two of three lines.

Introduction

Proteins are responsible for many allergic reactions suffered by people including the most threatening one, anaphylaxis, of which 80% are caused by peanuts or tree nuts in the U.S. [1]. It has been reported that as much as 6–8% of children 4 years old or younger are allergic to certain foods [2]. In peanut, Ara h 1–3 and 6 are considered the major allergens because they elicit immune responses in over 50% of peanut-allergic people [3], [4]. Because seeds (nuts and legumes) are the major provider of oils and bulk protein for the world, and new varieties are continually being created or discovered, a fast and accurate method for measuring seed allergen content before they enter the market is paramount.

Quantitation of peptides by mass spectrometry includes chemical labeling and label-free approaches with some variations on these two themes [5]. Quantitation by chemical labeling has been achieved by feeding stable isotopes of amino acids to cells in culture (SILAC) [6], or by covalently binding isotopically-coded affinity tags to cysteines of denatured proteins or peptides (ICAT) [6], [7], [8]. Peptide labeling has also been demonstrated by linking isobaric tags to primary amines (iTRAQ), or by 18O-labeling during proteolytic digestion [5], [9], [10]. These isotopic labeling methods benefit the researcher by allowing sample pooling thereby minimizing sample-to-sample variation and decreasing mass spectrometry use time [11]. However, chemical labeling is not 100% efficient, and depending on the method, does not evenly represent all proteins or peptides [12], [13]. Chemical labeling is also relatively expensive and offers a limited dynamic range of quantitation [8]. Label-free methods overcome some of these limitations and are therefore gaining acceptance as an alternative for quantitative proteomics. Label-free technology allows the researcher to quantitate peptides in a relative or absolute manner. Relative quantitation methods quantify peptides using spectral characteristics such as peak area (peak integration) or frequency of peptide fragment spectra during LC-MS/MS analysis (spectral counting). Absolute quantitation requires sample spiking with synthetic, stable-isotopic peptide standards for comparing ion current peak areas [9], [14]. Label-free methods are aided by recent advances in mass spectrometry instrumentation. Spectral counting benefits from faster scan rates, higher sensitivity and faster MS to MS/MS conversions, while peak integration benefits from stable and precise LC systems and high accuracy mass analyzers [5], [11], [15].

In the current study, we compared two label-free methods (peak integration and spectral counting) using a 40-fold dilution series of a protein standard. The methods were compared for response linearity, reproducibility (coefficient of variance) and ease of use. Spectral counting was comparable to peak integration for the first two criteria, however, in our hands post-analysis time for spectral counting was much lower than peak integration. Spectral counting was therefore employed to analyze transgenic peanut lines reduced in peanut allergens Ara h 2 and 6 by RNAi in order to test the effectiveness of spectral counting with a complex biological sample. Along with corroborating the knockdown of Ara h 2 and 6, spectral counting was able to provide a nearly complete view of the relative quantities of other major peanut allergens (Ara h 1, 2, 3/4, 6, 7, 8 and 10 found in allergen.org), showing drastic increases in Ara h 10 (oleosin 1) as well as collateral decreases in a conarachin (gi 52001225). However, most of these collateral changes were not observed in all three lines suggesting variation among transgenic events that will ultimately require confirmation from multiple seed from different generations. This proof-of-principle study demonstrates that gel-free, label-free proteomics could be a useful, high-throughput approach to profile seed allergens, enabling the researcher to perform rapid screening of numerous transgenic or breeding lines at minimal cost.

Section snippets

Preparation of protein standard for mass spectrometry

For protein standard curve serial dilutions a protein standard (PeppermintStick™, Molecular Probes, Eugene, OR) containing stoichiometrically equal amounts of pure proteins was diluted from stock (0.5 μg/μL) to 0.2 μg/μL with de-ionized 18 MΩ water, three separate times (biological replicates). Each standard was made to 8 M urea by adding urea powder. Disulfide bonds were reduced with 10 mM DTT (100 mM stock in 100 mM Tris–HCl pH 8.0), at 25 °C for 1 h. Reduced cysteines were alkylated with

Results

To be considered quantitative, spectral counting and peak integration must provide a linear signal across a serial dilution series, and the results must be reproducible. To test the efficacy of each approach we compared a 40-fold dilution series (5–200 ng) of a commercial protein standard containing six proteins of varying molecular masses (116.25–14.4 kDa) at equimolar concentrations. The dilutions were treated, digested and analyzed by LC-MS/MS. Spectral data were then analyzed with different

Error and R2 values for serial dilution analysis by spectral counting and peak integration

Our results for the serial dilution of a protein standard show that spectral counting and peak integration are both reproducible methods for relative quantitation using linear trap-mass spectrometry (LT-MS). The largest differences were seen in the % error. At best, spectral counting gave technical CV values of 9% considering unweighted spectral counts, while peak integration provided 3% (Supplemental Table 1). This might be because the same data set was used for both techniques with the scan

References (41)

  • I.H. Kang et al.

    Temporal and spatial expression of the major allergens in developing and germinating peanut seed

    Plant Physiol

    (2007)
  • A.E. Flinterman et al.

    Children with peanut allergy recognize predominantly Ara h2 and Ara h6, which remains stable over time

    Clin Exp Allergy

    (2007)
  • J.J. Thelen et al.

    Quantitative proteomics in plants: choices in abundance

    Plant Cell

    (2007)
  • S.P. Gygi et al.

    Quantitative analysis of complex protein mixtures using isotope-coded affinity tags

    Nat Biotechnol

    (1999)
  • M. John et al.

    A label-free quantification method by MS/MS TIC compared to SILAC and spectral counting in a proteomics screen

    Proteomics

    (2008)
  • H.P. Antoine et al.

    Comparative LC-MS: a landscape of peaks and valleys

    Proteomics

    (2008)
  • J.E. Melanson et al.

    High-coverage quantitative proteomics using amine-specific isotopic labeling

    Proteomics

    (2006)
  • W.W. Wu et al.

    Comparative study of three proteomic quantitative methods, DIGE, cICAT, and iTRAQ, using 2D gel- or LC-MALDI TOF/TOF

    J Proteome Res

    (2006)
  • S.A. Gerber et al.

    Absolute quantification of proteins and phosphoproteins from cell lysates by tandem MS

    Proc Natl Acad Sci U S A

    (2003)
  • Y.J. Kim et al.

    Reproducibility assessment of relative quantitation strategies for LC-MS based proteomics

    Anal Chem

    (2007)
  • Cited by (0)

    View full text