Journal of Experimental Botany Advance Access published March 30, 2006 Journal of Experimental Botany, Page 1 of 7 Plant Proteomics Special Issue doi:10.1093/jxb/erj157
Relative and absolute quantitative shotgun proteomics: targeting low-abundance proteins in Arabidobsis thaliana Stefanie Wienkoop1 and Wolfram Weckwerth2,* 1
Proteome Factory AG, Dorotheenstr. 94, D-10117 Berlin, Germany
2
Max-Planck-Institute of Molecular Plant Physiology, D-14476 Potsdam-Golm, Germany
Received 10 April 2005; Accepted 8 February 2006
Abstract
Introduction
The plant system is a highly dynamic structure on all molecular levels, transcripts, proteins, and metabolites. Thus, protein analysis has to cope with a highly dynamic range of concentrations. A severe problem is the detection of low-abundance proteins in the presence of housekeeping proteins. Basically three approaches are facilitated to measure protein abundance in a comprehensive manner: 2DE and one- or multi-dimensional shotgun proteomics, with or without stable-isotope labelling. These comparative techniques allow for the identification of altered protein levels compared with a reference state. However, they are limited to the analysis of medium/high-abundance proteins. Using stable-isotope dilution techniques it is possible to target the quantitative analysis to lowabundance proteins and to measure absolute concentrations of proteins. Based on multi-dimensional non-gel shotgun proteomics in Arabidopis thaliana, a list of tryptic peptides comprising >1000 proteins was generated. A strategy for quantitative plant proteomics is proposed using this master-list for selecting signature peptides of proteins. To prove the concept, a liquid chromatography–high-resolution triple quadrupole multiple reaction monitoring–mass spectrometry technique is described to determine the absolute amount of a low-abundance sucrose synthase isoform out of an ultra-complex A. thaliana protein extract.
Mass spectrometry (MS) offers the opportunity to generate large amounts of protein sequence-dependent data that, combined with genomic information, gives the potential for high throughput analysis of the plant proteome (Zivy and de Vienne, 2000; Roberts, 2002; Whitelegge, 2004; Agrawal and Rakwal, 2005; Glinski and Weckwerth, 2005b). Facing proteomics, the ultimate benefit to biology will be determined by the reliability of relative and absolute protein expression measurements. Currently, three main techniques are used for quantitation in proteomics: two-dimensional (2D)-PAGE, stable-isotope labelling, and stable-isotope labelling–free shotgun proteomics (Fig. 1). A common argument in favour of 2D-PAGE is that a comparison can readily be made between two gels and thus proteome differences can be detected. However, 2D-PAGE is still problematic because of reproducibility, varying staining efficiency of individual gels, and bias against some protein classes such as membrane proteins. In shotgun proteomics a complex protein sample is tryptically digested and analysed using liquid chromatography–mass spectrometry (LC-MS) techniques (Yates, 2004). The heterogeneous sample with proteins of different chemical and physical behaviour is broken down to a mixture of peptides easily adaptable to classical reversed-phase LC-MS techniques (Fig. 1). However, the increased complexity of the sample due to the sheer number of tryptic peptides is a challenge to chromatographic and mass spectrometric resolution. The advantages of shotgun proteomics are the capacity for throughput and less bias against protein classes, as in the case of 2D-PAGE (Washburn et al., 2001). Nevertheless, shotgun proteomics is not a quantitative technique per se. The intensity of a peptide peak depends linearly on the concentration of the peptide. However, different peptides have different propensities
Key words: Absolute quantitation, Arabidopsis, high resolution, linear ion trap, multiple reaction monitoring (MRM), plant, quantitative proteomics, relative quantitation, shotgun proteomics, single reaction monitoring (SRM), stable-isotope labelling, SUSY, triple quadrupole.
* To whom correspondence should be addressed. E-mail:
[email protected] ª The Author [2006]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail:
[email protected]
2 of 7 Wienkoop and Weckwerth
Fig. 1. General scheme of the most-used quantitative approaches for proteomics. Starting with the comparison of two, A and B, or more samples, relative quantification is achieved in all three cases. (1) Comparative proteomics using traditional 2DE. Protein staining enables the comparison of relative protein abundance. Protein identification is achieved by cutting out the protein spot, tryptic digestion, and analysis using mass spectrometry. A step further is the use of extracted ion traces for signature peptides to quantify the proteins in a spot. This technique is complementary to protein staining in the gel and circumvents the problem of overlapping protein spots. (2) Differential stable-isotope labelling. Owing to a mass shift introduced using different stable isotope-labelled linkers the intensity ratios of peptide isomers in an MS analysis can be determined. The ratio depicts the differences in abundance in the two samples. (3) Direct quantification using shotgun protein LC-MS analysis. Peptides are quantified by integration of extracted ion traces and normalization to internal standards (Weckwerth et al., 2004; Morgenthal et al., 2005) or averaging the spectral count per protein (Liu et al., 2004). (for details, see text). Using the direct quantitative approach the measurement of many samples and statistical data mining becomes feasible (Weckwerth et al., 2004; Morgenthal et al., 2005).
for ionization. Therefore, two different peptides present in equimolar amounts may show substantially different intensities in the mass spectra. In the case of label-free shotgun proteomics, this results in a relative quantitative analysis used, for example, for an integrative metabolite/ protein profiling approach (Weckwerth, 2003; Weckwerth et al., 2004; Morgenthal et al., 2005). Recently, a simple and fast method for the rough estimation of relative protein abundance has been described by Liu et al. (2004). They found that the number of tandem mass spectra (‘spectral count’) collected from a peptide mixture displayed perfect linearity with respect to concentration. By contrast, percentage sequence coverage and number of peptides per protein did not show as good a linear correlation as a spectral count. Nevertheless, most quantitative techniques rely on modifying one of the samples with a stable isotope, which changes the molecular mass but not the mass spectrometric and chromatographic behaviour. Quantitative differences are then determined
directly as the difference in peak area between the two peptides in the mixed sample. There are several approaches for labelling peptides with stable isotopes (Fig. 1): metabolic labelling using isotope-enriched or -depleted media (Oda et al., 1999; Ong et al., 2002; Whitelegge et al., 18 2004); proteolytic labelling of peptides using H16 2 O=H2 O (Yao et al., 2001); protein/peptide derivatization with ICAT (isotope-coded affinity tag-labelling) (Smolka et al., 2001) or methanol/HCl (Goodlett et al., 2001); and, for instance, ITRAQ that uses a multiplex set of four amine-specific isobaric reagents (Ross et al., 2004) allowing four-way relative and absolute quantitation. Besides the relative quantification of proteins, there is a strong need for the analysis of low-abundance proteins and the determination of absolute quantities of proteins in ultra-complex mixtures. The highly dynamic proteome is a great challenge. As a promising approach stable-isotope dilution techniques in combination with shotgun proteomics are emerging (Barr et al., 1996; Gerber et al., 2003;
Quantitative shotgun proteomics
Zhang et al., 2004; Pan et al., 2005). Proteins of interest are tryptically digested in the presence of synthetic peptide standards of known concentration with an incorporated stable isotope (13C, 15N). These standards are identical to the analyte peptides of interest but are distinguished by mass difference. Stable isotope-labelled and unlabelled peptides co-migrate during chromatography and absolute quantification is achieved by comparison of the peak area abundances of the internal standard peptide with the corresponding native counterpart due to, for example, multiple reaction monitoring (MRM) via tandem MS (Barnidge et al., 2003). Since abundance and metabolic turnover of a protein are not correlated, many key enzymes are less abundant and therefore difficult to analyse. Sucrose synthase (SUSY) is a key enzyme involved in sucrose metabolism. This enzyme catalyses the reversible conversion of sucrose and UDP to UDP-glucose and fructose. Its activity has been studied in various plants and has been shown to play a major role in energy metabolism, controlling the mobilization of sucrose into various pathways important for the metabolic, structural, and storage functions of the plant cell (Hesse and Willmitzer, 1996). Several studies (proteomic studies included) indicate that SUSY exists in both a cytosolic and a plasma membrane/bacteroid membraneassociated form (Komina et al., 2002; Wienkoop and Saalbach, 2003). In the model plant Arabidopsis, the complete genome sequence reveals six putative members of the SUSY gene family (Barratt et al., 2001). Isoform At5g20830 and At5g49190 in leaves are known to be highly stress responsive (Dejardin et al., 1999). To date, no data relative to the other four isoforms are available (Baud et al., 2002). Here a strategy for absolute quantitative one-dimensional (1D) high-throughput shotgun proteomics is proposed, starting with a master-list of signature peptides identified in a shotgun proteomics experiment (Wienkoop et al., 2004) (Fig. 2). Based on the analysis of a signature peptide for low-abundance SUSY isoform At3g43190, this protein is detected and quantified out of an ultra-complex protein mixture of entire Arabidopsis tissue. SUSY is usually not detectable in a typical non-targeted 1D shotgun analysis of a complex plant protein extract. The targeted analysis allows not only low-abundance proteins to be quantified but also enables the throughput analysis of many samples. The proposed method can be expanded for quantitative examinations of many proteins in a single run.
Materials and methods Generation of a protein master-list for the identification of signature peptides Arabidopsis thaliana Col 0 plant protein was analysed using a multi-dimensional chromatographic approach coupled to ion trap
3 of 7
Fig. 2. Proposed strategy for the targeted analysis of interesting and low-abundance proteins. Using a continuously growing master-list of identified peptides, it is possible to select subsets of signature peptides for proteins and to use these sequences as stable isotope-labelled synthetic internal standards. Furthermore the absolute quantification of proteins is possible (Gerber et al., 2003).
MS. The whole procedure is described in a recent study by Wienkoop et al. (2004). Preparation and LC-MS analysis of a stable isotope-labelled internal standard peptide for SUSY A specific internal SUSY (At3g43190) peptide standard HVSNLDRLEA*RR was synthesized by stable-isotope (13C/15N) alanine labelling (Thermo Electron, Ulm, Germany). LC analysis of the triple charged precursor ion was performed. To achieve the most sensitive and specific signal possible in the presence of a complex matrix, four single reaction monitoring (SRM) transitions were monitored for the native and internal standard peptides and the best chosen: native 489.17!484.6 and standard 490.5!484.6 (m/z) with an optimized collision energy of 15. Protein purification and sample digestion of Arabidopsis plant tissue Hydroponic A. thaliana Col 0 cultures were grown in phytotrons under controlled light, gas, and temperature conditions. Arabidopsis tissue of 7-week-old plants was harvested and proteins were extracted according to Weckwerth et al. (2004). For tryptic digestion the ultra-complex protein mixture was solubilized in 10% ACN/25 mM Ambic/10 mM CaCl/8 M urea and divided into two portions. One portion was spiked with internal standard peptide (25 fmol peptide per 1 lg protein). In the first step of digestion 1:100 LysC was used for 5 h at 37 8C followed by a second digestion
4 of 7 Wienkoop and Weckwerth with 2 ll trypsin beats per 10 lg protein (Poroszyme, Applied Biosystem, Darmstadt, Germany) for 16 h at 30 8C. Prior to trypsin digestion the sample was diluted to an end concentration of 2 M urea using the following buffer: 10% ACN/25 mM Ambic/10 mM CaCl. Non-targeted and targeted peptide analysis using nano LC–ion trap-MS and nano-LC–triple quadrupole-MS A nano HPLC system (Agilent 1100) was used for controlled nanoflow rates (300 nl min1). Samples (20 lg) were loaded onto a 75 lm ID RP column (Zorbax 300SB-C18, 3.5 lm, Agilent) coupled on-line with the MS. A 90 min gradient was performed from 40% to 100% MeOH, 0.1% FA. MS/MS was performed using a ThermoFinnigan (San Jose, CA, USA) mass spectrometer (LTQ ion trap). MRM was performed using a high-resolution TSQ Quantum triple quadrupole (ThermoFinnigan). On the LTQ a triple play and a combination of triple play, together with a pseudo-SIM-MS/MS scan sequence according to Venable et al. (2004) was performed on both precursor ions (489.17 and 490.5) with a mass window of 10 m/z. The pseudoSIM-MS/MS sequence was programmed between 46 and 58 min during the gradient. The TSQ tuning and MRM of the target peptides were essentially performed according to Glinski and Weckwerth (2005a). Q1 was kept at 0.3 resolution and Q3 at 0.7 m/z with a dwell time of 50 ms (Glinski and Weckwerth, 2005a). For quantitation, a calibration curve was achieved using different amounts of standard peptide in the protein mixture from 0 to 0.5 pmol (0/10/50/100/250/500 fmol) end concentration.
Results and discussion Generation of signature peptides for proteins
A master-list of proteins was generated using a multidimensional chromatographic approach as demonstrated in Fig. 2. Total leaf protein was extracted and fractionated using ion exchange chromatography as described in Wienkoop et al. (2004). Each fraction was analysed using a two-dimensional chromatographic system comprising a strong cation-exchange column coupled to a silica-based reversed-phase C18 monolithic column. The chromatographic system was downscaled to nano-flow ESI to achieve highest sensitivity. Based on this analysis a master-list of proteins was obtained. From this list, peptides were chosen as signature peptides for proteins based on the highest score for identification. For proof of concept study, a SUSY isoform was chosen known to be of very low abundance. This protein was only detected in the study using extensive fractionation of proteins prior to 2D shotgun proteomics (Wienkoop et al., 2004). Target-peptide analysis using nano-flow liquid chromatography coupled to a high resolution triple quadrupole mass spectrometer Electrospray ionization exhibits highest ionization efficiency with very low flow rates down to 100 nl min1 or lower. For that purpose, the coupling of a nano-LC pump system to the MS instrument was facilitated (Wienkoop et al., 2004). To increase the performance further a triple quadrupole mass spectrometer was used in the MRM mode. In this mode the mass spectrometer is tuned to the
target-peptide thereby increasing selectivity and sensitivity of the analysis. Several SRM modes can be performed in one chromatographic separation. Thus, the simultaneous analysis of dozens of target-peptides can be performed in a single run (Glinski and Weckwerth, 2005a). Using internal standard peptides it is also possible to generate calibration curves for the absolute quantitation of peptides. Targeted analysis and absolute quantitation of SUSY out of an ultra-complex protein sample For absolute quantitation of biological samples stableisotope dilution techniques are standard methods. Nevertheless, quantification of less abundant proteins out of ultra-complex plant protein mixtures such as crude extracts from Arabidopsis thaliana using a non-gel 1D shotgun proteomics approach has never been shown before. SUSY, a key enzyme involved in sucrose metabolism in plants was chosen for initial absolute quantification studies. Its activity has been studied in various plants. However, specific isoforms have not been distinguished. Yet, the SUSY isoform At3g43190 was chosen, due to mass spectrometric peptide sequence information acquired by multi-dimensional peptide identification, as described above (Wienkoop et al., 2004). The tryptic peptide HVSNLDRLEARR was picked for standard synthesis since it reached highest spectrum quality. The native peptide can easily be distinguished from the chemically identical counterpart via a 4 Da mass shift and identical retention time (52 min; Fig. 3). However, fragmentation patterns are of great advantage for confirmation. In the first step, the quality of the internal standard peptide has to be evaluated such that the replacement of the residue containing stable isotopes after solid-phase peptide synthesis must be nearly 100%. Thus, no residual peak without stable-isotope label should be monitored, otherwise the absolute amount of the native peptide would be falsified (Fig. 4). For peak integration, three replicates of 20 lg digested Arabidopsis protein extract with and without 500 fmol spiked standard peptide each were analysed via LC-MRM-MS using a high-resolution triple quadrupole mass spectrometer (TSQ, Thermoelectron) (Fig. 3). Increasing the resolution of the first and the third quadrupole enhances the selectivity of the SRM. Thereby the signal to noise ratio is increased and, thus, also the sensitivity/purity of detection and quantification (Glinski and Weckwerth, 2005a). With the standard curve (Fig. 5), the absolute amount of the peptide (1.58% mass of total protein) was determined to be 4.5 fmol lg1 protein mixture, which correlates to 2.25 fmol mg1 fresh weight. Non-targeted analysis using a ‘pseudo-SIM-MS/MS’ scan on an ion trap MS enhances signal detection of low-abundance peptides in an ultra-complex sample
The analysis of the same protein mixture under exactly the same HPLC settings as for the triple quadrupole
Quantitative shotgun proteomics
5 of 7
Fig. 3. High-resolution triple-quadrupole MS (TSQ) analysis of the sucrose synthase within an ultra-complex Arabidopsis protein mixture. (A) Sample (20 lg protein mixture) without internal stable isotope-labelled standard peptide. (B) Sample (20 lg protein mixture) spiked with internal stable isotopelabelled standard peptide.
Fig. 4. Control of purity for the stable isotope-labelled peptide. The upper trace (489.17 m/z) is the native peptide ion; the lower trace (490.5 m/z) is the stable isotope-labelled peptide ion. Measurement was made with TSQ Quantum.
instruments performed with data-dependent scanning on a linear ion trap (LTQ, Thermoelectron) showed no detectable peak at 52 min, corresponding to the native SUSY-peptide ion trace (Fig. 6A). However, identification of a purified protein of even less than 10 fmol is usually possible but seems to be limited within an
ultra-complex protein mixture due to ion suppression. Even under extended gradient conditions, SUSY could not be detected. Intriguingly, after insertion of a datadependent MS/MS segment with a mass window of 10 Da, including the SUSY-peptide precursor m/z, a peak was found at 52 min, which could be identified as the
6 of 7 Wienkoop and Weckwerth
SUSY-peptide (Fig. 6Bii). This scan-mode can be interpreted as a pseudo-SIM or pseudo-SRM on a very wide mass window. Consequently, a range of precursor ions is selected for fragmentation including the targeted ion. The higher sensitivity can be explained by the fact that all fragment ions are summed up to a total ion intensity signal by contrast to a SRM scanning process on a triple Standard curve (HVSNLDRLEA*RR)
300
y = 505.57x
peak area* E5
250
R2 = 0.9989
200
Conclusion
150 100 50 0
quadrupole MS. Thus, the data-dependent MS/MS scan technique without dynamic exclusion, by contrast with full scan triple play analysis with dynamic exclusion, seems to be very powerful for searching for low-abundance proteins in a complex protein matrix and provides enough scans for quantification. However, the typical wide precursor selection mass window for an ion trap spectrum decreases selectivity by contrast to an SRM on a triple quadrupole MS (Glinski and Weckwerth, 2005a). Comparing signal/ noise intensities of both analyses between TSQ and LTQ the triple quadrupole MS analysis was about 2-fold higher (see also Glinski and Weckwerth, 2005a).
0
0.1
0.2
0.3
0.4
0.5
pmol
Fig. 5. Standard curve for the determination of absolute quantity of the HVSNLDRLEARR peptide from sucrose synthase isoform At3g43190.
One-dimension shotgun protein analysis with stableisotope dilution using internal standard peptides provides a powerful tool for high throughput quantitative proteome analysis of low-abundance proteins and absolute quantitation. These data will enable the different proteins and their concentrations in the plant tissue to be compared. However, for some very low-abundance proteins it may be necessary to scale-up the initial sample loading and extend
Fig. 6. Ion trap MS (LTQ) analysis of the sucrose synthase within an ultra-complex Arabidopsis protein mixture. Using a pseudo-SIM-MS/MS experiment on an LTQ instrument (see Materials and methods) it is possible to quantify the low-abundance SUSY peptide (Bii). This method can be used for non-targeted relative quantification of peptides/proteins in a complex sample. (A) Full MS ion trace of 489.2 m/z chromatogram after triple play analysis. (B) Combined triple play/data-independent MS/MS analysis. (Bi) Full MS ion trace of 489.2 m/z chromatogram with triple play. (Bii) Time segment where only the pseudo-SIM-MS/MS was performed. The precursor m/z 489.2 for the SUSY-peptide is now detected using this scan-mode (for details see Results and discussion).
Quantitative shotgun proteomics
the chromatography to two- (or more) dimensional fractionation prior to analysis (Wienkoop et al., 2004). Acknowledgement We thank Joel Louette, Annette Westermayer and Sven Klingel Thermo Electron (Ulm, Germany) for providing us with stable isotope labelled peptides.
References Agrawal GK, Rakwal R. 2005. Rice proteomics: a cornerstone for cereal food crop proteomes. Mass Spectrom Review 25, 1–53. Barnidge DR, Dratz EA, Martin T, Bonilla LE, Moran LB, Lindall A. 2003. Absolute quantification of the G protein-coupled receptor rhodopsin by LC/MS/MS using proteolysis product peptides and synthetic peptide standards. Analytical Chemistry 75, 445–451. Barr JR, Maggio VL, Patterson DG, Cooper GR, Henderson LO, Turner WE, Smith SJ, Hannon WH, Needham LL, Sampson EJ. 1996. Isotope dilution mass spectrometric quantification of specific proteins: model application with apolipoprotein A-I. Clinical Chemistry 42, 1676–1682. Barratt DHP, Barber L, Kruger NJ, Smith AM, Wang TL, Martin C. 2001. Multiple, distinct isoforms of sucrose synthase in pea. Plant Physiology 127, 655–664. Baud S, Boutin JP, Miquel M, Lepiniec L, Rochat C. 2002. An integrated overview of seed development in Arabidopsis thaliana ecotype WS. Plant Physiology and Biochemistry 40, 151–160. Dejardin A, Sokolov LN, Kleczkowski LA. 1999. Sugar/ osmoticum levels modulate differential abscisic acid-independent expression of two stress-responsive sucrose synthase genes in Arabidopsis. Biochemical Journal 344, 503–509. Gerber SA, Rush J, Stemman O, Kirschner MW, Gygi SP. 2003. Absolute quantification of proteins and phosphoproteins from cell lysates by tandem MS. Proceedings of the National Academy of Sciences, USA 100, 6940–6945. Glinski M, Weckwerth W. 2005a. Differential multisite phosphorylation of the trehalose-6-phosphate synthase gene family in Arabidopsis thaliana: a mass spectrometry-based process for multiparallel peptide library phosphorylation analysis. Molecular and Cellular Proteomics 4, 1614–1625. Glinski M, Weckwerth W. 2005b. The role of mass spectrometry in plant systems biology. Mass Spectrometry Review 25, 173–214. Goodlett DR, Keller A, Watts JD, Newitt R, Yi EC, Purvine S, Eng JK, von Haller P, Aebersold R, Kolker E. 2001. Differential stable isotope labeling of peptides for quantitation and de novo sequence derivation. Rapid Communications in Mass Spectrometry 15, 1214–1221. Hesse H, Willmitzer L. 1996. Expression analysis of a sucrose synthase gene from sugar beet (Beta vulgaris L). Plant Molecular Biology 30, 863–872. Komina O, Zhou Y, Sarath G, Chollet R. 2002. In vivo and in vitro phosphorylation of membrane and soluble forms of soybean nodule sucrose synthase. Plant Physiology 129, 1664–1673. Liu H, Sadygov RG, Yates 3rd JR. 2004. A model for random sampling and estimation of relative protein abundance in shotgun proteomics. Analytical Chemistry 76, 4193–4201. Morgenthal K, Wienkoop S, Scholz M, Selbig J, Weckwerth W. 2005. Correlative GC-TOF-MS-based metabolite profiling and LC-MS-based protein profiling reveal time-related systemic regulation of metabolite–protein networks and improve pattern recognition for multiple biomarker selection. Metabolomics 1, 109–121.
7 of 7
Oda Y, Huang K, Cross FR, Cowburn D, Chait BT. 1999. Accurate quantitation of protein expression and site-specific phosphorylation. Proceedings of the National Academy of Sciences, USA 96, 6591–6596. Ong SE, Blagoev B, Kratchmarova I, Kristensen DB, Steen H, Pandey A, Mann M. 2002. Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Molecular and Cellular Proteomics 1, 376–386. Pan S, Zhang H, Rush J, Eng J, Zhang N, Patterson D, Comb MJ, Aebersold R. 2005. High throughput proteome screening for biomarker detection. Molecular and Cellular Proteomics 4, 182–190. Roberts JKM. 2002. Proteomics and a future generation of plant molecular biologists. Plant Molecular Biology 48, 143–154. Ross PL, Huang YN, Marchese JN, et al. 2004. Multiplexed protein quantitation in Saccharomyces cerevisiae using aminereactive isobaric tagging reagents. Molecular and Cellular Proteomics 3, 1154–1169. Smolka MB, Zhou HL, Purkayastha S, Aebersold R. 2001. Optimization of the isotope-coded affinity tag-labeling procedure for quantitative proteome analysis. Analytical Biochemistry 297, 25–31. Venable JD, Dong MQ, Wohlschlegel J, Dillin A, Yates JR. 2004. Automated approach for quantitative analysis of complex peptide mixtures from tandem mass spectra. Nature Methods 1, 39–45. Washburn MP, Wolters D, Yates JR. 2001. Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nature Biotechnology 19, 242–247. Weckwerth W. 2003. Metabolomics in systems biology. Annual Review of Plant Biology 54, 669–689. Weckwerth W, Wenzel K, Fiehn O. 2004. Process for the integrated extraction, identification, and quantification of metabolites, proteins and RNA to reveal their co-regulation in biochemical networks. Proteomics 4, 78–83. Whitelegge JP. 2004. Mass spectrometry for high throughput quantitative proteomics in plant research: lessons from thylakoid membranes. Plant Physiology and Biochemistry 42, 919–927. Whitelegge JP, Katz JE, Pihakari KA, Hale R, Aguilera R, Gomez SM, Faull KF, Vavilin D, Vermaas W. 2004. Subtle modification of isotope ratio proteomics: an integrated strategy for expression proteomics. Phytochemistry 65, 1507–1515. Wienkoop S, Glinski M, Tanaka N, Tolstikov V, Fiehn O, Weckwerth W. 2004. Linking protein fractionation with multidimensional monolithic RP peptide chromatography/mass spectrometry enhances protein identification from complex mixtures even in the presence of abundant proteins. Rapid Communications in Mass Spectrometry 18, 643–650. Wienkoop S, Saalbach G. 2003. Proteome analysis: novel proteins identified at the peribacteroid membrane from Lotus japonicus root nodules. Plant Physiology 131, 1080–1090. Wienkoop S, Zoeller D, Ebert B, Simon-Rosin U, Fisahn J, Glinski M, Weckwerth W. 2004. Cell-specific protein profiling in Arabidopsis thaliana trichomes: identification of trichome-located proteins involved in sulfur metabolism and detoxification. Phytochemistry 65, 1641–1649. Yao X, Freas A, Ramirez J, Demirev PA, Fenselau C. 2001. Proteolytic 18O labeling for comparative proteomics: model studies with two serotypes of adenovirus. Analytical Chemistry 73, 2836–2842. Yates 3rd JR. 2004. Mass spectral analysis in proteomics. Annual Review of Biophysical and Biomolecular Structure 33, 297–316. Zhang FG, Bartels MJ, Stott WT. 2004. Quantitation of human glutathione S-transferases in complex matrices by liquid chromatography/tandem mass spectrometry with signature peptides. Rapid Communications in Mass Spectrometry 18, 491–498. Zivy M, de Vienne D. 2000. Proteomics: a link between genomics, genetics and physiology. Plant Molecular Biology 44, 575–580.
