Technical Note pubs.acs.org/jpr
Performance of Isobaric and Isotopic Labeling in Quantitative Plant Proteomics Fábio C.S. Nogueira,†,# Giuseppe Palmisano,‡,# Veit Schwam ̈ mle,‡ Francisco A.P. Campos,§ ‡ †, Martin R. Larsen, Gilberto B. Domont, * and Peter Roepstorff‡,* †
Proteomic Unit, Institute of Chemistry, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil Department of Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark § Department of Biochemistry and Molecular Biology, Universidade Federal do Ceará, Fortaleza, Brazil ‡
S Supporting Information *
ABSTRACT: Mass spectrometry has become indispensable for peptide and protein quantification in proteomics studies. When proteomics technologies are applied to understand the biology of plants, twodimensional gel electrophoresis is still the prevalent method for protein fractionation, identification, and quantitation. In the present work, we have used LC−MS to compare an isotopic (ICPL) and isobaric (iTRAQ) chemical labeling technique to quantify proteins in the endosperm of Ricinus communis seeds at three developmental stages (IV, VI, and X). Endosperm proteins of each stage were trypsin-digested in-solution, and the same amount of peptides was labeled with ICPL and iTRAQ tags in two orders (forward and reverse). Each sample was submitted to nanoLC coupled to an LTQ-Orbitrap high-resolution mass spectrometer. Comparing labeling performance, iTRAQ was able to label 99.8% of all identified unique peptides, while 94.1% were labeled by ICPL. After statistical analysis, it was possible to quantify 309 (ICPL) and 321 (iTRAQ) proteins, from which 95 are specific to ICPL, 107 to iTRAQ, and 214 common to both labeling strategies. We noted that the iTRAQ quantification could be influenced by the tag. Even though the efficiency of the iTRAQ and ICPL in protein quantification depends on several parameters, both labeling methods were able to successfully quantify proteins present in the endosperm of castor bean during seed development and, when combined, increase the number of quantified proteins. KEYWORDS: ICPL, iTRAQ, quantitative proteomics, plant proteomics, high-resolution mass spectrometry
■
INTRODUCTION The application of strategies to quantify proteins in biological systems is indispensable in proteomic studies, as relative and absolute protein quantification frequently leads to new insights into the biology of the system under study.1 The development of novel methods for protein quantification in parallel with the use of mass spectrometry and bioinformatics techniques will help in widening the application of these technologies for achieving a better understanding of key aspects of the biology of plants.2 The presence of a wide diversity of secondary metabolites, carbohydrates, and lipids may prevent quantitative analysis of plant proteomes, as they interfere with the labeling techniques currently used. Until recently, two-dimensional gel electrophoresis combined with MALDI mass spectrometry (MS) has been the method of choice for plant protein fractionation and identification; however, state-of-the-art quantitative shotgun LC−MS techniques are slowly taking over.3,4 The use of MS for peptide and protein quantification is a well-established technology in many laboratories. Moreover, quantitative techniques, e.g., label-free approach,5,6 metabolic labeling7 and chemical labeling,8,9 are widespread in the © 2012 American Chemical Society
proteomic community and have been compared in several studies. The isotope-code protein labeling, ICPL, technology is a chemical labeling strategy designed to isotopically label all free amino groups of proteins by amine specific reagents8 in order to quantify at the MS level the relative peptide abundance before assembling them into protein ratio. Even though the feasibility of using this technology for peptide labeling in bottom-up proteomic approaches (SERVA Quadruplex-Kit, instruction manual, www.serva.de) has been suggested, only recently a detailed experimental protocol, called postdigest ICPL, has been produced,10,11 highlighting a better protein identification and quantification, when compared with the previous ICPL usage. Isobaric tag for relative and absolute quantitation, iTRAQ, has been successfully applied to quantify proteins on the basis of peptide labeling and quantification. iTRAQ is capable of multiplexing up to eight different samples.9,12 This method uses amine specific isobaric reagents to label the primary amines of peptides from different samples. The quantitation of iTRAQ occurs at the MS/MS level, where Received: February 28, 2012 Published: March 27, 2012 3046
dx.doi.org/10.1021/pr300192f | J. Proteome Res. 2012, 11, 3046−3052
Journal of Proteome Research
Technical Note
Figure 1. Experimental workflow. Proteins from endosperm of Ricinus communis seeds at stages IV, VI, and X were digested with trypsin, and the same amount of peptides from each stage were labeled with ICPL and iTRAQ in two orders and afterward mixed, generating four different samples: ICPL forward and reverse and iTRAQ forward and reverse. After cleaning, the samples were analyzed in four replicates by an EASY-LC coupled with a LTQ-Orbitrap XL mass spectrometer.
from plants grown under irrigation conditions. For this work, we chose seed filling stages IV (early), VI (medium), and X (late), in which oil and proteins are stored in the endosperm tissue from castor bean seeds, as defined by Greewood and Bewley.13 Endosperm at these selected developmental stages were dissected manually from seeds, cut into small pieces, defatted overnight in acetone, freeze-dried, ground to a fine powder, and kept at −20 °C until used.
the reporter ions present in a low-mass range are quantified on the basis of their relative intensities.9 Although postdigest ICPL and iTRAQ employ a similar strategy for labeling, they use different methodologies for quantification: one based in isotopic distribution pattern of labeled precursor ions at MS level (ICPL) and the other based in measurements of the reporter ions at MS/MS level (iTRAQ). Even though the equivalence of these two labeling techniques in terms of quantitation efficiency has been suggested,10 no experimental support for this has been produced. In this work, we applied and compared both techniques to quantify proteins in the endosperm of castor bean seeds at three developmental stages. The endosperm of castor bean seeds is a reserve tissue in which protein and oil reserves accumulate during seed development to be mobilized during germination to support the development of the growing embryo, thus providing a unique system with developmentally regulated patterns of protein deposition. The systematic comparison between isotopic (ICPL) and isobaric (iTRAQ) labeling reported in this study will help researchers in choosing the labeling strategy for specific purposes.
■
Protein Extraction and In-Solution Trypsin Digestion
Protein extraction was performed essentially as described by Nogueira et al.14 Endosperm powders were mixed with pyridine buffer (50 mM pyridine, 10 mM thiourea and 1% SDS, pH 5.0) and polyvinylpolypyrrolidone in the proportion of 1:40:2 (w/ v/w). The mixture was stirred for 2 h at 4 °C and centrifuged at 10000g for 40 min. Proteins from the supernatant were precipitated with cold 10% trichloroacetic acid (TCA) in acetone, and the pellet was washed with cold acetone three times. The last precipitate was then dried under vacuum and dissolved in 7 M urea, 2 M thiourea, and 200 mM TEAB. Protein concentration was determined in duplicate by amino acid analysis using a Biochrom 30+ Amino Acid Analyzer (Biochrom), following the manufacturer’s instructions. Trypsin digestion was performed after disulfide bridge reduction with 10 mM DTT for 1 h at 25 °C and free thiol alkylation with 40 mM iodacetamide for 40 min at room temperature in the dark. Samples were diluted to less than 1 M urea and digested with trypsin (Promega) (1:50, w/w) overnight at 25 °C. After digestion, formic acid was added to a final concentration of 2%, and the tryptic peptides were cleaned using Oasis cartridges (Millipore). Peptides were resuspended in 200 mM TEAB, and their concentration was measured using Qubit assay (Invitrogen). For samples of each seed developmental stage, aliquots of 33 μg were collected, dried down, and stored at −80 °C until used.
EXPERIMENTAL SECTION
Reagents
Pyridine, TCA, TEAB, and polyvinylpolypyrrolidone were from Sigma-Aldrich. Urea, thiourea, SDS, DTT, and IAA were from GE Healthcare. Modified trypsin was from Promega. Empore C18 disks (3 M) were from 3 M Bioanalytical Technologies. Ammonium solution (25%), NaOH, HCL, and formic acid were from Merck and Co. Inc. Ultrahigh quality (UHQ) water was from a Milli-Q system (Millipore). All reagents were HPLC grade or higher. Plant Material
Seeds of castor bean (Ricinus communis cv. Nordestina) were harvested between the months of July to September, 2010, 3047
dx.doi.org/10.1021/pr300192f | J. Proteome Res. 2012, 11, 3046−3052
Journal of Proteome Research
Technical Note
the Orbitrap.18 MS/MS for acquiring CID was set as described and MS/MS for acquiring HCD was set at 7500 at 400 m/z resolution, 30 000 signal threshold, 5-ms activation time at 48 normalized collision energy, and dynamic exclusion enabled for 30 s with a repeat count of 1. A 5% ammonia−water solution in a 500-μL Eppendorf tube with holes in the cover was put under the pulled needle to avoid the supercharged effect of the iTRAQ 4plex tag, phenomenon previously described.19
ICPL Labeling
Labeling was performed using the ICPL Quadruplex-Kit (SERVA) following the manufacturer’s instructions, adjusted to the “postdigest optimized protocol”.10 Briefly, 33 μg of peptides of each developmental stage were dissolved in 30 μL of 20 mM TEAB, pH 8.5, and submitted for labeling using 3 μL of ICPL-labeling reagent for 90 min at room temperature under argon atmosphere. A new 1.5 μL aliquot of reagent was then added and incubated for 90 min at room temperature under argon atmosphere. The other steps were in accordance with the ICPL kit instructions: stop solution provided by the kit for 20 min at room temperature was added, samples were mixed and vortexed, the pH was adjusted to 11.9 with 2 N NaOH, and after 20 min, 2 N HCl was added to neutralize the sample solution. Finally, an aliquot was analyzed by MALDI-TOFTOF to evaluate the labeling. Two labeling strategies were used, forward and reverse (Figure 1). Labeled peptides mixtures were purified by custom-made chromatographic Poros 50 R2 (PerSeptive Biosystems) and Poros Oligo R3 (Applied Biosystems) reverse phase microcolumns prior to MS analysis, as described in detail earlier.15,16
Data Analysis
Raw data were inspected in Xcalibur v.2.1 (Thermo Scientific). Database searches were performed using Proteome Discoverer with Mascot v.2.3 algorithm against a target and decoy (reverse) concatenated Ricinus communis database downloaded from Uniprot database January, 2011. The searches were performed with the following parameters: MS accuracy 10 ppm, MS/MS accuracy 0.6 Da for CID and 0.1 Da for HCD, trypsin digestion with two missed cleavages allowed, fixed carbamidomethyl modification of cysteine and variable modification of oxidized methionine. ICPL (label 0, monoisotopic mass = 105.02), ICPL:13C(6) (label 6, monoisotopic mass = 111.04), and ICPL:13C(6)2H(4) (label 10, monoisotopic mass = 115.06) were set as variable modification of N-terminus and Lys. iTRAQ 4plex (monoisotopic mass = 144.102) for Nterminus and Lys were set as variable modification. Number of proteins and protein groups and numbers of peptides were estimated using Proteome Discoverer, false discovery rates around 1%, and peptide rank 1 was applied as a cutoff limit. ICPL quantification was performed using Proteome Discoverer to integrate the extract ion chromatogram of every precursor. iTRAQ quantification was performed using Proteome Discoverer as well, but based in reporter ion integration within 20 ppm window.
iTRAQ Labeling
Labeling of the peptides (33 μg) with the 4-plex iTRAQ reagents (Applied Biosystems) was performed according to the manufacturer’s recommendations. The tryptic peptides were dissolved in 30 μL of 20 mM TEAB, pH 8.5. To each iTRAQ 4-plex reagent vial, 70 μL of ethanol was added, and the mixture was combined with the peptide sample and incubated at room temperature for 1 h. After incubation, samples were acidified using formic acid and dried. Two labeling strategies were used for iTRAQ, forward and reverse (Figure 1). Both peptide mixtures were cleaned up by TSK Amide-80 HILIC microHPLC system.17
Statistical Analysis
Reversed Phase NanoLC and Mass Spectrometry Analysis
A total of 16 samples consisting of four replicates forward “stage IV, stage VI, and stage X” and four replicates in the reverse “stage X, stage IV, stage VI” for each labeling method were analyzed (Figure 1). The entire statistical analysis was carried out taking the log2-values of the integration value of the precursor (ICPL) or the reporter ion (iTRAQ). In order to improve performance, we normalized on sets of four replicates. For each sample, the mean over all values and all stages was taken. The samples become comparable by subtracting this mean. Afterward, for each developmental stage separately, the values of the four replicates are merged, and the total median was used for normalization of each replicate. This normalization constant is more accurate as it relies on a larger number of data points. Quantitative values for the proteins are obtained with the RRollup method from the DanteR package (http://www.omics. pnl.gov) requiring a minimum of two unique peptides per protein. The final values are the averages of the different peptide measurements after scaling and removal of outliers with a Grubbs test. Quantifications of a protein that were measured in different runs of the mass spectrometer can be compared through their relative changes only. We therefore normalize each protein of a replicate by the mean taken over all three developmental stages. For the significance analysis, we consider both forward and reverse; i.e., we obtain a maximum of eight replicates per protein and labeling method. This analysis neglects effects coming from the usage of different labels. An ANOVA test yields p-values for the detection of significant changes between experimental stages. The values are calculated
Prior to MS analysis, peptide quantification after ICPL and iTRAQ labeling was performed by Qubit assay (Invitrogen). Labeled peptides mixtures were dissolved in 0.1% formic acid, and 1 μg of them was loaded onto a C18-reversed phase pulled needle capillary column (18 cm length, 100 μm inner diameter, packed in-house with ReproSil-Pur C18-AQ 3 μm resin). The samples were analyzed by an EASY-nano LC system (Proxeon Biosystems) coupled online to an ESI-LTQ-Orbitrap XL mass spectrometer (Thermo Fisher Scientific). Peptides were eluted using a gradient from 100% phase A (0.1% formic acid) to 35% phase B (0.1% formic acid, 95% acetonitrile) for 120 min, 35− 100% phase B for 5 min, and 100% B for 8 min (total of 133 min at a flow rate of 250 nL/min). After each run, the column was washed with 90% phase B and re-equilibrated with phase A. m/z spectra were acquired in a positive mode applying datadependent automatic survey MS scan and tandem mass spectra (MS/MS) acquisition. A MS scan (400−1800 m/z) in the Orbitrap mass analyzer set at resolution 60 000 at 400 m/z, 1 × 106 automatic gain control target, and 500-ms maximum ion injection time was followed by MS/MS of the five most intense multiply charged ions in the LTQ at 30 000 signal threshold, 20 000 gain control target, 300-ms maximum ion injection time, 2.5 m/z isolation width, 30-ms activation time at 35 normalized collision energy, and dynamic exclusion enabled for 30 s with a repeat count of 1 (applied in the ICPL samples). For iTRAQ samples, data-dependent CID MS/MS analysis of the three most intense ions was performed in the LTQ followed by HCD MS/MS analysis of the corresponding ions with detection in 3048
dx.doi.org/10.1021/pr300192f | J. Proteome Res. 2012, 11, 3046−3052
Journal of Proteome Research Table 1.
a
ICPL forward
ICPL reverse
iTRAQ forward
iTRAQ reverse
Technical Note
replicate
peptides
unique peptides
unique peptides labeled
% of unique peptides labeled
first second third fourth AVG SD first second third fourth AVG SD first
3084 3300 3719 3495 3400 271 3183 3162 3503 3492 3335 188 3660
1002 1081 1443 1323 1212 206 1080 1050 1413 1426 1242 205 1416
945 1028 1361 1229 1141 189 1022 1001 1326 1329 1170 183 1413
94.3 95.1 94.3 92.9 94.1
second third fourth AVG SD first
3544 3622 3294 3530 165 3814
1329 1341 1284 1343 55 1418
1326 1337 1282 1340 54 1415
99.8 99.7 99.8 99.8
second third fourth AVG SD
3825 3908 3504 3763 178
1403 1450 1367 1410 34
1400 1448 1364 1407 35
99.8 99.9 99.8 99.8
94.6 95.3 93.8 93.2 94.1 99.8
99.8
proteins
protein groups
quantified proteins
% of quantified proteinsb
543 553 697 643 609 74 572 564 672 720 632 77 652
301 311 388 363 341 42 310 304 386 396 349 49 354
482 487 621 574 541 68 431 443 544 536 489 60 647
88.8 88.1 89.1 89.3 88.8
617 617 600 622 22 634
335 337 332 340 10 339
613 611 595 617 22 628
99.4 99.0 99.2 99.2
636 665 639 644 14
341 355 355 348 9
632 657 635 638 13
99.4 98.8 99.4 99.1
75.3 78.5 81.0 74.4 77.3 99.2
99.1
a
For each labeling reagent, setup, and replicate, the number of peptides, unique peptides and unique peptides labeled, the percentage of the peptides labeled, the number of identified proteins, protein groups and quantified proteins, and the percentage of quantified proteins were measured. The average (AVG) and the standard deviation (SD) were calculated for each label setup. All values were obtained taking into account the parameters peptide rank 1 and 1% FDR. bProteins quantified with at least one peptide labeled with two tags.
Figure 2. (A) Percentage of proteins identified with 1, 2, and 3 or more unique peptides in each labeling setup. (B) Ion score distribution for each labeling setup plotted as a box-plot; the arithmetic mean and median are represented by “×” and “−”, respectively. (C) Percentages of the peptide charge for all identified peptides.
residues. For each reagent, labeling order, and replicate, we counted the number of total peptides, unique peptides, unique peptides labeled, proteins, protein groups, and quantified proteins (proteins quantified with at least one peptide labeled with two tags and providing at least one ratio). Using 1% FDR and peptide rank 1 as filter criteria, we have established a peptide score cutoff in each replicate, which for ICPL labeled samples was higher than that for iTRAQ labeled samples (Table S1, Supporting Information (SI)). As shown in Table 1, ICPL labeling led to identification of 3400 ± 271 (forward) and 3335 ± 188 (reverse) peptides, 1212 ± 206 (forward) and 1242 ± 205 (reverse) unique peptides, and 1141 ± 189
using the DanteR software and corrected for multiple testing using the Benjamini−Hochberg procedure.20
■
RESULTS
Evaluation of ICPL and iTRAQ Labeling
Our experimental setup comprised the labeling of peptide aliquots obtained from the endosperm of seeds at developmental stages IV, VI, and X (1:1:1) with ICPL and iTRAQ in two different orders, forward and reverse, and in four replicates for each labeling order as shown in Figure 1. Incorporation of the tags was evaluated by the number of peptides with the respective modification in the N-terminal and/or lysine 3049
dx.doi.org/10.1021/pr300192f | J. Proteome Res. 2012, 11, 3046−3052
Journal of Proteome Research
Technical Note
(forward) and 1170 ± 183 (reverse) unique peptides labeled. iTRAQ labeling led to the identification of 3530 ± 165 (forward) and 3763 ± 178 (reverse) peptides, 1343 ± 53 (forward) and 1410 ± 34 (reverse) unique peptides, and 1340 ± 54 (forward) and 1407 ± 35 (reverse) unique peptides labeled. Comparing labeling performance, iTRAQ was able to label 99.8% of the all identified unique peptides, while 94.1% were labeled by ICPL. The difference in the number of identified proteins and protein groups between the two reagents was low, but the number of proteins quantified by ICPL and by iTRAQ was significantly different (Table 1). Both labeling methods lead to similar amounts of proteins and protein group identifications. However, iTRAQ performed slightly better in terms of quantifications. In order to understand the underlying reasons, we evaluated the number of unique peptides per protein, the ion score for all identified peptides, and the charge state of the identified peptides. As shown in Figure 2, the majority of proteins were identified with a similar ion score and with more than 1 peptide, but a peptide charge-enhancement for iTRAQ was observed even in the presence of ammonia vapor under the electrospray needle as a proton scavenger. As shown in Table 1, ICPL shows a higher error rate than iTRAQ. In order to confirm this, we investigated the labeling reproducibility by determining the standard deviation distribution of all values (corresponding to the logarithm of the measured precursor intensities) for each tag used in the ICPL and iTRAQ multiplexing labeling (Figure S1 (SI)) and confirmed the higher error rate presented by ICPL, thus indicating the lower reproducibility of this labeling.
regulated proteins, and two proteins with inconsistent quantitation values between the two labeling strategies.
■
DISCUSSION In this work, we evaluated and compared isobaric (iTRAQ) and isotopic (ICPL) chemical labeling techniques in the quantitation of proteins from the endosperm of castor bean seeds at three stages of development. While the labeling principle of both methods is similar, i.e., incorporation of a tag at Nterminal and lysine residues of a peptide,8,9 the strategy to quantify derivatized peptides by MS is different. In ICPL, m/z of the precursor ions is obtained at high resolution in the Orbitrap, while fragments generated by CID are obtained at low resolution in the LTQ. For iTRAQ identification and quantification, the same precursor ion selected for fragmentation by CID is also fragmented by HCD to produce the reporter ions acquired in high resolution at the Orbitrap, as a strategy for circumventing the low mass cutoff limitation of the ion trap for fragmented peptides by CID.21 In our case, all samples were analyzed under the same conditions (RP-column, LC system, etc.) and in the same mass spectrometer, an ESILTQ-Orbitrap XL. Incidentally, this is the first time that this type of MS instrument was used to analyze postdigest ICPL labeled peptides. On the basis of the number of peptides labeled, almost 100% of the identified unique peptides were labeled by iTRAQ, while ICPL labeled around 94%. As both quantitation strategies are peptide-based, thus bearing in protein quantitation, that difference in labeling efficiency is important and is reflected in the higher percentage (99%) of proteins quantified by iTRAQ. The low values for ICPL (77−89%) were unexpected, since two previously published studies10,11 with similarly complex samples achieved a percentage of 95%. Evaluating the number of unique peptides per protein, the distribution of the ion scores of unique peptides and the charge states of the identified peptides, could not explain the lower values of ICPL. However, the fact that more peptides were quantified per protein using iTRAQ than when using ICPL might be a reason. The observed higher charge state distribution for iTRAQ has already been observed in a previous study.19 In this study, it was demonstrated that it could be partially bypassed by the use of ammonia vapor. Despite using this in the present study, a slightly higher average charge state for the iTRAQ labeled peptides was observed. On the basis of the distribution of the standard deviations of each label in four replicates and two different labeling orders, we can conclude that iTRAQ exhibits a smaller variance with standard deviations being less than half in average compared to the ones found for ICPL labeling. When comparing the forward and reverse order labeling, we obtained an unexpected result: the principal component analysis of the data shows that the data sets from the same experimental stages as well as the ones with the same label order cluster together, whereas in the ICPL labeling we can only identify groupings within the same experimental stage (Figure S2 (SI)). Therefore, iTRAQ shows a measurable effect with respect to the individual isobaric tag labels. This observation is puzzling, since it has been assumed22 that the isobaric tag does not influence the quantification. Taking into consideration the criteria established for the statistical analysis, we established the expression patterns of those proteins quantified both by iTRAQ and ICPL. For most of them, these patterns are essentially the same and only in a few cases label-dependent. We suggest that this may be related
Quantitative Analysis of Castor Bean Seed Development
A statistical analysis was performed to evaluate the quantification of proteins from the endosperm of three developmental stages of castor bean seeds. This analysis took into account both multiplex reagents and combines forward and reverse orders. This led to the quantification of 309 (ICPL) and 321 (iTRAQ) proteins, from which 95 were only identified in the ICPL experiment, 107 were only identified in the iTRAQ experiment, and 214 were common to both labeling strategies (Figure 3 and Table S2 (SI)). For those quantified proteins
Figure 3. Venn diagram for the quantified proteins quantified by iTRAQ and ICPL labeling. A total of 416 proteins were quantified taking into account the parameters used for statistical analysis.
common to both labeling and present in all three developmental stages, the deposition profiles could be established (Table S2 (SI)). For most of them, quantitation values obtained for the two labeling strategies were comparable, but in at least seven cases values were divergent. Figure 4 shows the protein profiles of two up-regulated proteins, two down3050
dx.doi.org/10.1021/pr300192f | J. Proteome Res. 2012, 11, 3046−3052
Journal of Proteome Research
Technical Note
Figure 4. Profiles of proteins present in endosperm of Ricinus communis seeds at stages IV, VI, and X. Each protein profile is reported for ICPL and iTRAQ labeling. Preproricin (D6MWM3), legumin A (B9SF35), sucrose synthase (B9RR41), G3PDH (B9RBN8), oleosin (B9RAW7), and ACP (B9REW6). G3PDH = glyceraldehyde 3-phosphate dehydrogenase; ACP = acyl carrier protein.
Figure S2: Principal component analysis (PCA) of data sets from the same experimental stage as well as the same label order. This material is available free of charge via the Internet at http://pubs.acs.org.
to drawbacks of each labeling method. In the case of ICPL, it is known that multiplex labeling increases the complexity at the MS1 level and also the isotopic effect of deuterated tags interferes with retention time of the labeled peptides during LC,23 complicating the subsequent quantitation in several database search programs. In iTRAQ labeling, eventually coeluted peptides could be isolated with the same precursor ion window, leading to errors in the measurements of reporter ions. In addition, peptides with post-translational modifications and/or presenting different patterns of tag incorporation could lead to data misinterpretations.24 Even though the efficiency of the iTRAQ and ICPL in protein quantification depends on several parameters, both labels were able to quantify successfully proteins present in the endosperm of castor bean during seed development. This kind of sample may be regarded as a challenging sample, since the high abundance storage proteins make the analysis of low abundance proteins a difficult task by considerably increasing the dynamic range.25 In conclusion, we have found that both the use of iTRAQ and ICPL as labeling reagents allows quantification of a comparable number of proteins in the present study and that the two methods when used in combination results in an increased number of quantified proteins.
■
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]; Fax: +45 65502404 (P.R.). E-mail:
[email protected]; Fax: +55-21-25627353 (G.R.). Author Contributions #
These authors contributed equally.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Lene A. Jakobsen is acknowledged for her technical assistance; F. C. S. Nogueira, G. B. Domont and F. A. P. Campos were supported by PETROBRAS and the Brazilian National Research Council (CNPq); Giuseppe Palmisano is supported by the Danish Medical Science Research Council (GP Grant No. 11-107551); Martin R. Larsen is supported by the Danish Natural Science Research Council (MRL Grant No. 21030167) and the Danish Strategic Research Council (MRL Young Investigator Award); Veit Schwämmle is supported by the Danish Council for Independent Research, Natural Sciences (FNU).
ASSOCIATED CONTENT
S Supporting Information *
Table S1: For each labeling reagent, setup and replicate, all evaluated values are shown. The average (AVG) and the standard deviation (SD) were calculated for each label setup. All values were obtained taking into account the parameters values shown in the table. Table S2: Proteins quantified by ICPL, iTRAQ, and both labeling strategies taking into account the rules imposed by the statistical analysis. Figure S1: Box plot of the standard deviation distribution of all values (corresponding to the logarithm of the measured precursor intensities) for each tag used in the ICPL and iTRAQ multiplexing labeling.
■
REFERENCES
(1) Ong, S. E.; Mann, M. Mass spectrometry-based proteomics turns quantitative. Nat. Chem. Biol. 2005, 1 (5), 252−62. (2) Bindschedler, L. V.; Cramer, R. Quantitative plant proteomics. Proteomics 2011, 11 (4), 756−75. (3) Thelen, J. J.; Peck, S. C. Quantitative proteomics in plants: choices in abundance. Plant Cell 2007, 19 (11), 3339−46.
3051
dx.doi.org/10.1021/pr300192f | J. Proteome Res. 2012, 11, 3046−3052
Journal of Proteome Research
Technical Note
(4) Oeljeklaus, S.; Meyer, H. E.; Warscheid, B. Advancements in plant proteomics using quantitative mass spectrometry. J. Proteomics 2009, 72 (3), 545−54. (5) Wang, W.; Zhou, H.; Lin, H.; Roy, S.; Shaler, T. A.; Hill, L. R.; Norton, S.; Kumar, P.; Anderle, M.; Becker, C. H. Quantification of proteins and metabolites by mass spectrometry without isotopic labeling or spiked standards. Anal. Chem. 2003, 75 (18), 4818−26. (6) Carvalho, P. C.; Fischer, J. S.; Chen, E. I.; Yates, J. R. 3rd; Barbosa, V. C. PatternLab for proteomics: a tool for differential shotgun proteomics. BMC Bioinf. 2008, 9, 316. (7) Ong, S. E.; Kratchmarova, I.; Mann, M. Properties of 13Csubstituted arginine in stable isotope labeling by amino acids in cell culture (SILAC). J. Proteome Res. 2003, 2 (2), 173−81. (8) Schmidt, A.; Kellermann, J.; Lottspeich, F. A novel strategy for quantitative proteomics using isotope-coded protein labels. Proteomics 2005, 5 (1), 4−15. (9) Ross, P. L.; Huang, Y. N.; Marchese, J. N.; Williamson, B.; Parker, K.; Hattan, S.; Khainovski, N.; Pillai, S.; Dey, S.; Daniels, S.; Purkayastha, S.; Juhasz, P.; Martin, S.; Bartlet-Jones, M.; He, F.; Jacobson, A.; Pappin, D. J. Multiplexed protein quantitation in Saccharomyces cerevisiae using amine-reactive isobaric tagging reagents. Mol. Cell. Proteomics 2004, 3 (12), 1154−69. (10) Leroy, B.; Rosier, C.; Erculisse, V.; Leys, N.; Mergeay, M.; Wattiez, R. Differential proteomic analysis using isotope-coded protein-labeling strategies: comparison, improvements and application to simulated microgravity effect on Cupriavidus metallidurans CH34. Proteomics 2010, 10 (12), 2281−91. (11) Fleron, M.; Greffe, Y.; Musmeci, D.; Massart, A. C.; Hennequiere, V.; Mazzucchelli, G.; Waltregny, D.; De Pauw-Gillet, M. C.; Castronovo, V.; De Pauw, E.; Turtoi, A. Novel post-digest isotope coded protein labeling method for phospho- and glycoproteome analysis. J. Proteomics 2010, 73 (10), 1986−2005. (12) Choe, L.; D’Ascenzo, M.; Relkin, N. R.; Pappin, D.; Ross, P.; Williamson, B.; Guertin, S.; Pribil, P.; Lee, K. H. 8-Plex quantitation of changes in cerebrospinal fluid protein expression in subjects undergoing intravenous immunoglobulin treatment for Alzheimer’s disease. Proteomics 2007, 7 (20), 3651−60. (13) Greenwood, J. S.; Bewley, J. D. Seed development in Ricinus communis (castor bean). I. Descriptive morphology. Can. J. Bot. 1982, 60 (9), 1751−60. (14) Nogueira, F. C.; Goncalves, E. F.; Jereissati, E. S.; Santos, M.; Costa, J. H.; Oliveira-Neto, O. B.; Soares, A. A.; Domont, G. B.; Campos, F. A. Proteome analysis of embryogenic cell suspensions of cowpea (Vigna unguiculata). Plant Cell Rep. 2007, 26 (8), 1333−43. (15) Gobom, J.; Nordhoff, E.; Mirgorodskaya, E.; Ekman, R.; Roepstorff, P. Sample purification and preparation technique based on nano-scale reversed-phase columns for the sensitive analysis of complex peptide mixtures by matrix-assisted laser desorption/ ionization mass spectrometry. J. Mass Spectrom. 1999, 34 (2), 105−16. (16) Larsen, M. R.; Cordwell, S. J.; Roepstorff, P. Graphite powder as an alternative or supplement to reversed-phase material for desalting and concentration of peptide mixtures prior to matrix-assisted laser desorption/ionization-mass spectrometry. Proteomics 2002, 2 (9), 1277−87. (17) Palmisano, G.; Lendal, S. E.; Engholm-Keller, K.; Leth-Larsen, R.; Parker, B. L.; Larsen, M. R. Selective enrichment of sialic acidcontaining glycopeptides using titanium dioxide chromatography with analysis by HILIC and mass spectrometry. Nat. Protoc. 2010, 5 (12), 1974−82. (18) Rewitz, K. F.; Larsen, M. R.; Lobner-Olesen, A.; Rybczynski, R.; O’Connor, M. B.; Gilbert, L. I. A phosphoproteomics approach to elucidate neuropeptide signal transduction controlling insect metamorphosis. Insect Biochem. Mol. Biol. 2009, 39 (7), 475−83. (19) Thingholm, T. E.; Palmisano, G.; Kjeldsen, F.; Larsen, M. R. Undesirable charge-enhancement of isobaric tagged phosphopeptides leads to reduced identification efficiency. J. Proteome Res. 2010, 9 (8), 4045−52.
(20) Benjamini, Y.; Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. 1995, 57 (1), 12. (21) Kocher, T.; Pichler, P.; Schutzbier, M.; Stingl, C.; Kaul, A.; Teucher, N.; Hasenfuss, G.; Penninger, J. M.; Mechtler, K. High precision quantitative proteomics using iTRAQ on an LTQ Orbitrap: a new mass spectrometric method combining the benefits of all. J. Proteome Res. 2009, 8 (10), 4743−4752. (22) Burkhart, J. M.; Vaudel, M.; Zahedi, R. P.; Martens, L.; Sickmann, A. iTRAQ protein quantification: a quality-controlled workflow. Proteomics 2011, 11 (6), 1125−34. (23) Brunner, A.; Keidel, E. M.; Dosch, D.; Kellermann, J.; Lottspeich, F. ICPLQuanta software for non-isobaric isotopic labeling proteomics. Proteomics 2010, 10 (2), 315−26. (24) Ow, S. Y.; Salim, M.; Noirel, J.; Evans, C.; Rehman, I.; Wright, P. C. iTRAQ underestimation in simple and complex mixtures: “the good, the bad and the ugly”. J. Proteome Res. 2009, 8 (11), 5347−55. (25) Krishnan, H. B.; Oehrle, N. W.; Natarajan, S. S. A rapid and simple procedure for the depletion of abundant storage proteins from legume seeds to advance proteome analysis: a case study using Glycine max. Proteomics 2009, 9 (11), 3174−88.
3052
dx.doi.org/10.1021/pr300192f | J. Proteome Res. 2012, 11, 3046−3052
