Advances in Quantitative Phosphoproteomics - Analytical Chemistry

---

REVIEW pubs.acs.org/ac

Advances in Quantitative Phosphoproteomics Carol L. Nilsson* Department of Pharmacology & Toxicology, University of Texas Medical Branch, 301 University Blvd., Galveston, Texas 77555-0617, United States

’ CONTENTS

’ SAMPLE PREPARATION

Sample Preparation Chromatographic Methods for Phosphoproteomics Combined Chromatographic Approaches Metal-Based Enrichment Strategies Phosphotyrosine Enrichment Mass Spectrometry Analysis and Data Processing Quantitation From Data Analysis to Signaling Pathways New Biological Insights Derived from Quantitative Phosphoproteomic Applications Perspectives Author Information Biography Acknowledgment References

735

Chromatographic Methods for Phosphoproteomics. All quantitative, global studies of phosphoproteins require some type of sample preparation before analysis (Figure 1). Because of the high background of nonphosphorylated peptides in a proteomic digest, separation of phosphopeptides prior to introduction into the mass spectrometer is needed to obtain optimal proteome coverage. Most approaches described in this Review utilize chromatography for phosphoprotein or phosphopeptide enrichment and for the final separation prior to mass spectrometric analysis. With the exception of phosphotyrosine signaling studies, nearly all report the use of chromatographic methods as the initial workflow step. Recently, exhaustive reviews of phosphoprotein and phosphopeptide enrichment strategies have been published.1 4 Thus, basic terminology will be explained in this Review followed by a focus on newer applications and enrichment strategies as applied to phosphoproteomics. Because the phosphate group on phosphopeptides and phosphoproteins has a low pKa value, chromatographic enrichment methods that are based on ion exchange mechanisms are widely used, even though they typically suffer from lower peak capacity than other separation methods.5 In strong cation-exchange chromatography (SCX), proteins or peptides are retained on a column that contains a hydrophilic, anionic resin through the affinity to positively charged groups on amino acids. A salt gradient is applied, and bound molecules elute off the column in the order of increasing isoelectric point. Phosphopeptides, by nature more acidic than their unmodified counterparts, elute in earlier fractions along with peptides enriched in acidic amino acids. A modified method that involves lowering the pH of the gradient to 2.7 can be applied to increase the enrichment of phosphopeptides relative to unmodified peptides in initial SCX fractions.6 Still, the method is not highly specific for phosphopeptides but also enriches peptides with other acidic modifications such as sialic acid and acetylation. Strong anion exchange has also been successfully applied to phosphopeptide separation7,8 and used in combined chromatographic approaches.9 11 Hydrophilic interaction chromatography (HILIC)12 separates proteins and peptides based on their polarity. The chromatographic approach is to use a polar stationary phase and a partly hydrophilic mobile phase. Peptides are retained in proportion to their hydrophilicity, just the opposite of reverse-phase chromatography. The addition of chaotropes can decrease retention

735 736 737 737 737 739 740 741 742 743 743 743 743

R

eversible O-phosphorylation of proteins is of central importance to cell signaling in eukaryotic systems. Protein phosphorylation may increase or decrease protein activity and alter protein localization and protein protein interactions. Global studies of protein phosphorylation continue to present analytical challenges, especially in the areas of sample preparation, data acquisition, and data analysis. Key regulatory phosphoproteins are typically minor components of the proteome. The nature of protein phosphorylation in response to signaling in transduction cascades is transitory, and quantitative measurements are required to fully characterize the biological response. In contrast to assays based on immunological methods, such as reversed-phase phosphoproteomics (RPPA), quantitative mass-spectrometry-based investigations can provide data at a high level of sensitivity and specificity that does not require a priori knowledge about nodes in networks or antibodies toward specific targets. However, optimization of tandem mass spectrometric techniques is necessary to obtain maximum sequence information and high-accuracy assignment of phosphorylation sites. Phosphopeptides exist in substoichiometric quantities and are heterogeneous with respect to sites of phosphorylation. Analysis can be hampered by ion suppression and insufficient sequence ion information obtained in mass spectrometry-based workflows. Once data sets are acquired and processed, extracting biological knowledge from the data is a nontrivial procedure. Because the number of bioanalytical studies that employ quantitative phosphoproteomics continues to grow at a rapid pace, this Review focuses on major advances in the field that have been published in the most recent years (2009 2011). r 2011 American Chemical Society

Special Issue: Fundamental and Applied Reviews in Analytical Chemistry Published: November 01, 2011 735

dx.doi.org/10.1021/ac202877y | Anal. Chem. 2012, 84, 735–746

Analytical Chemistry

REVIEW

Figure 1. Quantitative phosphoproteomics encompasses a comprehensive, modular analytical approach that includes chromatography, phosphopeptide enrichment, a strategy for protein/peptide quantitation, (nano)-LC-MS and -MS/MS, and data analysis.

times of highly hydrophilic analytes. HILIC use has been reported occasionally in recent years in phosphoproteomic studies.13 19 It has the advantage of not requiring the addition of involatile salts to the samples (unlike SCX) and thus can be coupled online to MS analysis.20 The closely related technique electrostatic repulsion-hydrophilic interaction liquid chromatography (ERLIC) was introduced by Alpert.21 Separation can be performed isocratically. The column matrix (weak anion exchange) carries a similar charge as the analytes, but the solvent contains enough organic phase to force the solutes to remain on the column despite electrostatic repulsion. This mixed mode of separation allows for the enrichment of phosphopeptides from other peptide components in a proteomic digest. Many phosphoproteins are also glycosylated, and glycopeptides, which also are more hydrophilic relative to their unmodified counterparts, can be separated with ERLIC. Hao et al. demonstrated the application of ERLIC followed by RP-LC-MS in a single assay to characterize the phospho- and glycoproteome of rat kidney.22 Out of a total of nearly three thousand proteins, many of which were low-abundance or membrane proteins, 583 phosphorylation sites and 722 N-glycosylation sites were determined. The application of ERLIC, a recently developed chromatographic technique with unique features, to phosphoproteomics either alone or in combination with other techniques, appears to be gaining in popularity.15,23 25 In a recent comprehensive study of HeLa cell proteins, fractionation by ERLIC, HILIC, and SCX, each followed by TiO2 bead enrichment, was evaluated in a global phosphoproteomic approach applied to tryptic digests of proteins.15 SCX was found to yield the largest number of nonredundant (3913) phosphopeptides. ERLIC and HILIC yielded 1683 and 1273 phosphopeptides, respectively. ERLIC yielded 69% multiphosphorylated peptides; SCX yielded 37%, and HILIC yielded 17%. When followed by TiO2 enrichment of fractions, the highest number of phosphopeptides were detected in SCX fractions, but ERLIC fractions contained the highest number of multiply phosphorylated peptides. Remarkably, less than 10% of the identified phosphopeptides were detected in all three separation modalities, underscoring the importance of the use of complementary methods to obtain the optimum coverage of the phosphoproteome. Combined Chromatographic Approaches. In a biological fluid such as plasma, the dominance of a few major proteins (albumin and immunoglobulins, for instance) make the detection of phosphorylated peptides extremely difficult. A novel protein

and peptide fractionation workflow was applied to clinical serum samples from patients with benign prostate hyperplasia.26 Highpressure size-exclusion chromatography (SEC) was performed on a pooled sample that contained nearly 15 mg of total serum protein, followed by protein digestion and offline peptide fractionation by zwitterion-ion hydrophilic interaction chromatography (ZIC-HILIC). The separation method (ZIC) is based on separation of zwitterionic peptides according to a hydrophilic partitioning mechanism superimposed on weak electrostatic interactions, followed by HILIC. Peptides were observed to elute in order of increasing hydrophilicity. Fractions were analyzed by reversed-phase chromatography and nanoelectrospray ionization coupled to an ion trap instrument. The data yielded the identity of more than 1900 proteins, including 375 phosphoproteins. The method was demonstrated to be superior in direct comparison to serum immunodepletion methods coupled to SCX or ZIC-HILIC without SEC, as determined by the number of proteins identified. However, Petricoin et al. did employ immunodepletion and serial TiO2-enrichment of peptides derived from 2 mg of serum and were able to identify approximately 100 phosphopeptides.27 Phosphoproteomic studies of serum are particularly challenging, because phosphoproteins are much more rare in this biofluid, compared to cell lysates or solid tissues. Larsen and co-workers devised a multidimensional strategy for detection of more than 4700 phosphopeptides from 400 μg of digested proteins from metabolically labeled control and EGFstimulated HeLa cells.28 Phosphopeptides were separated into mono- and multiphosphorylated pools by sequential elution from IMAC (SIMAC), HILIC separation of the peptides, TiO2 chromatography of the HILIC fractions, and online nano-LCMS/MS analysis. The EGF pathway was previously studied by global quantitative phosphoproteomics,29 but Larsen’s method yielded 636 unique phosphorylation sites regulated by 5 min EGF treatment, compared to 120 in the previous work. Ficarro et al. devised two novel multidimensional fractionation strategies to analyze phosphopeptides.9 In the first approach, aliphatic ion pairing reagents were used to improve retention of phosphopeptides at high pH in two-dimensional reversed phase(RP)-RP. The first RP separation was performed at pH 10. The second approach was to construct a three-dimensional separation by adding strong anion exchange (SAX) as the second step in 3DRP-SAX-RP. The second approach proved to have high peak separation capacity, efficient retention of phosphopeptides, and 736

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry good reproducibility. When applied to just 400 μg of material from a cell line characterized by oncogenic FLT3 signaling in acute myeloid leukemia, 7787 phosphopeptides were identified, a 10-fold increase over the number identified by RP alone.30 Metal-Based Enrichment Strategies. Metal-based enrichment strategies are dependent upon the Lewis base (electron-pair donating) properties of the phosphate group. Phosphorylated conjugates bind readily to metals that carry a positive charge, such as iron. Immobilized metal affinity chromatography (IMAC) involves immobilization of a metal ion (Fe3+, Ti4+, Zr4+, etc.) to a matrix material and is widely used to enrich phosphopeptides. In addition, metal oxides (TiO2 and ZrO2) are frequently used in metal oxide affinity chromatography (MOAC) approaches. Comparative studies of metal and metal oxide enrichment media have been analyzed for applications in phosphopeptide enrichment. Gates et al.31 compared commercially available products that included Fe-NTA IMAC, titanium dioxide resin, prepacked MOAC enrichment tips, and magnetic titanium beads, for enrichment of phosphopeptides from a casein digest and concluded that the titanium MOAC materials performed best. Ficarro et al. evaluated enrichment of phosphopeptides derived from human myeloid K562 cells with magnetic beads coupled with metal ions of Fe, Ga, ZrO, Zn, Cu, or Al, chelated with either nitrilotriacetic acid (NTA) or iminodiacetic acid (IDA). Their results demonstrated that all metal ions coupled to NTA provided superior selectivity compared to the IDA-coupled metal ions. Fe-NTA and Ga-NTA performed best, with Ga-NTA providing more multiphosphorylated peptides. IMAC is the most frequently used technique to enrich phosphopeptides, with about forty reports published in the past two years. Binding of nonphosphorylated peptides that contain high numbers of acidic amino acid residues, such as glutamic and aspartic acid, represents a major limitation of this method. In addition, enrichment of phosphopeptides by IMAC is influenced by the nature of the sorbent material, coupled chelating compound, and binding/elution conditions. Many of the recent studies focus on refinements to the technique to address its shortcomings. Novotna et al. devised magnetic nonporous microspheres modified with IDA and immobilized Fe3+ or Ga3+ ions32 and found that the beads showed low retention of the coating of the magnetic support and no preference of either Fe-IDA or Ga-IDA toward singly or doubly phosphorylated peptides, unlike those metals’ NTA conjugates.33 This contrasts to literature stating that both iron and gallium have strong preferential binding for phosphopeptides bearing one or multiple phosphate groups.34 Jensen and co-workers developed an optimized protocol (IMAC IMAC) for phosphopeptide recovery that entailed the use of Fe-NTA resin, optimization of the percentage of acetonitrile and pH to minimize phosphopeptide loss, and a tandem set of consecutive IMAC enrichments.35 The consecutive IMAC IMAC procedure nearly doubled the number of phosphopeptides identified from a mouse cell lysate, compared to a single IMAC experiment. Figeys et al. demonstrated high sensitivity of analysis when Ti-IMAC was performed on minute amounts of material isolated from liver subcellular organelles in a phosphoproteomics reactor followed by nano-LC-MS/ MS.36 The group was able to assign thousands of phosphorylation sites on about 600 proteins with the improved workflow. Hydroxyapatite (HAP) is formed from the alkali earth metal calcium combined with phosphate (molecular formula Ca10(PO4)6(OH)2). Due to strong attraction between Ca 2+ and phosphate, HAP has been reported to selectively enrich mono- and multiply

REVIEW

phosphorylated peptides without coenrichment of acidic peptides, which occurs in other metal-based strategies.37,38 The mineral has also been incorporated in a monolithic column for more rapid protein separation and phosphopeptide enrichment.39 Phosphotyrosine Enrichment. Phosphotyrosine-containing peptides are much scarcer that phosphoserine or phosphothreonine peptides in a proteomic digest. Thus, large-scale studies of phosphotyrosine signaling pathways usually require enrichment of phosphotyrosine proteins or peptides. In contrast to phosphoserine and phosphothreonine peptides, immunoaffinity enrichment of phosphotyrosine-containing proteins or peptides is a viable option, because commercial antibodies of good quality are available.40 Boersema et al.41 employed phosphopeptide antibody purification and stable isotope dimethyl label quantitation to identify and quantify over 1100 phosphopeptides, of which eighty percent were tyrosine-phosphorylated peptides, from 4 mg of HeLa cells. Signaling pathways mediated by tyrosine kinases are prominent in several disease states, including cancer. Because phosphotyrosine-containing peptides are rare in comparison to other phosphopeptides, most investigators in quantitative phosphoproteomics have embraced the combination of antiphosphotyrosine antibodies and IMAC for sample enrichment.42 A recent application of note from the White group was quantitative analysis of tyrosine phosphorylation in U87MG glioblastoma multiforme cells that expressed an epidermal growth factor receptor (EGFR) mutant form (EGFRvIII) mutated at various tyrosine phosphorylation sites.43 Mutation at Y1173 led to increased phosphorylation of EGFRvIII and the identification of crosstalk between EGFRvIII and another oncogenic signaling pathway (c-Met). A second report from this group described the study of T-cell receptor signaling in diabetes prone and resistant mice by immunoprecipitation and IMAC purification of phosphotyrosine peptides with quantitation by iTRAQ. The data set provided insights into the underlying mechanisms of T-cell dysregulaton in the mouse model of diabetes. In another report, mutant FLT3 receptor tyrosine kinase signaling in acute myeloid leukemia44 was studied by iTRAQ-based quantification, and a robust error model was applied to the quantitative data. In order to accurately determine sites and regulation of phosphorylation, assignment of p and q values and confidence intervals were assigned to every peptide. Few novel enrichment tools have surfaced recently for phosphoproteomics. A novel aptamer-based approach was developed45,46 on the basis of polyamidoamine dendrimers functionalized with titanium ions (PolyMAC) and applied sequentially with antibodies to the tyrosine phosphoproteome of breast cancer cells. The combined method provided high phosphopeptide recovery from the samples and allowed for identification of nearly 800 tyrosine phosphorylation sites. Most of those phosphorylation events were correlated to the activity of Syk, a protein tyrosine kinase which acts as a tumor suppressor.47

’ MASS SPECTROMETRY ANALYSIS AND DATA PROCESSING Successful mass spectrometric analysis of the phosphoproteome demands data which can identify the sequence of peptides and site-specific phosphorylation with certainty as well as accurately determine quantitative changes in both site-specific phosphorylation and the phosphoprotein from which it was derived. High-sensitivity of analysis is typically obtained by liquid 737

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry

REVIEW

chromatography coupled to nanoelectrospray ionization, a technique which (in positive mode) efficiently generates multiply protonated gas-phase peptide ions. Recently, dramatic improvement of phosphoproteomic analysis by use of a miniaturized LCelectrospray was demonstrated by Ficarro et al.,48 who manufactured fused silica analytical columns with integrated emitter tips. They investigated phosphotyrosine-containing peptides isolated from human embryonic stem cells and showed a 2.7fold increase in the number of phosphopeptides identified at 1% FDR when a 25 μm column was employed, compared to a 75 μm column. Moreover, they observed higher intensities for low m/z iTRAQ reporter ions when the 25 μm column was used, which improved the accuracy of quantitation. Quantitative phosphoproteomic workflows typically employ a hybrid-type mass spectrometer. Hybrid mass spectrometers allow for precursor ion isolation followed by dissociation to obtain MS/ MS spectra from which peptide sequence information can be derived. Quadrupole-time-of-flight (Q-TOF), quadrupole-ion trap, ion trap-Orbitrap, and ion trap-Fourier transform ion cyclotron resonance mass spectrometers are among the most popular instrument configurations.49 54 Higher resolution, high mass accuracy instruments are beneficial to proteomic analysis in general, because peptide measurements made in MS mode with high accuracy decrease the number of theoretical peptide matches in a hypothetical digest of a proteome.55 In high resolution mass spectrometers, baseline separation of phosphorylated and sulfated peptides can be achieved as well.56 When isobaric tags (iTRAQ or TMT) are employed as a quantitation mode, high resolution in MS/MS mode has the advantage of distinguishing reporter ions from low m/z interferences.14,57 There are several MS/MS techniques that may be applied in quantitative phosphoproteomics, each with a distinctive profile. For an excellent and recent in-depth review of tandem mass spectrometry strategies for phosphoproteome analysis, see Palumbo et al.58 Collision-induced dissociation (CID) is a widely used technique for inducing fragmentation of gas-phase peptide and phosphopeptide ions.29,58 62 CID typically involves inelastic collisions of excited ions with neutral atoms or molecules. In the case of unmodified peptides, extensive b- and y-ions may be generated from which amino acid sequences can be derived. However, CID-MS/MS spectra derived from phosphoserine and phosphothreonine containing peptides show facile fragmentation of the phosphate group and dominance of neutral phosphate (HPO3, 80 Da, and H3PO4, 98 Da) losses from the precursor ion.63 This phenomenon can lead to relatively uninformative MS/MS data that are poor in sequence information. One other caveat to phosphopeptide MS/MS by CID is the risk of phosphate migration, especially in peptide precursor ions with immobile or partially mobile protons.64 However, this phenomenon is disputed in the setting of proteomics.65 In a large scale study of phosphopeptides generated from human proteins by use of trypsin or Lys-N hydrolytic enzymes, little evidence was found for phosphate rearrangement by CID.66 Instead, site specific phosphate assignment ambiguities observed in about one-quarter of the sequenced phosphopeptides were attributed to inefficient fragmentation in the MS/MS events. Pulsed Q collision-induced dissociation (PQD) is an MS/MS approach developed for linear ion trap mass spectrometers in order to overcome the low m/z cutoff phenomenon67 and enable detection of reporter ions from isobarically tagged peptides. Compared to the number of published reports in which CID was used as the MS/MS method, there are few that describe PQD.68 71

The combination of CID and HCD has been reported to outperform PQD and HCD combined for quantitative proteomic studies that employ isobaric tagging reagents.72 The less popular PQD method is not as efficient for ion fragmentation compared to CID and has lacked software for determining and reporting protein abundance ratios from isobarically tagged proteomic samples. Two recent reports address these issues. Wu et al.73 performed a systematic comparative study of CID and PQD of peptides and phosphopeptides by varying instrument parameters and analyzing fragmentation patterns, intensities, and bioinformatic scores obtained from database searches to improve PQD data sets. In another report, a new and freely available mzXML-compatible software pipeline (iQuant) was described that automates calculations of protein weighted average abundance ratios and p-values derived from MS/MS spectra of iTRAQ-labeled peptides.74 The PQD dissociation method can be combined with other MS/MS methods, such as electron transfer dissociation (ETD)75 to improve confidence of phosphopeptide site identification and reporter ion quantification. Higher energy C-trap dissociation (HCD)76,77 in the orbitrap mass analyzer provides CID-type fragmentation with more extensive fragmentation due to the higher energy of collisions. Consequently, higher quality results and more informative spectra compared to those obtained by conventional CID methods are obtained. Because HCD is performed in a collision cell and the ions are sent back through the C-trap for detection in the orbitrap, the technique has advantages over typical ion trap MS/MS approaches, including high resolution, high mass accuracy, and no low m/z cutoff. The latter is especially important for measurement of reporter ions generated from isobarically labeled peptides during MS/MS.14,76,78,79 In earlier versions of the LTQ-Orbitrap, these advantages were offset by the relatively slow acquisition of MS/MS data by HCD compared to CID. However, the newer configurations of the instrument allow for acquisition of high resolution MS/MS data at higher rates than were possible previously.80 Another important feature of HCD compared to CID is the reduced frequency of phosphate transfer81 in phosphoserine and phosphothreonine-containing peptides, as well as the formation of an atypical 4,5-dihydrooxazolyl x-ion which pinpoints the phosphorylation site of the peptide. Comparisons of HCD and CID fragmentation have been evaluated in the linear ion trap-orbitrap for application in phosphoproteomic studies, with divergent results. Mann and co-workers compared high resolution precursor mass measurement in the orbitrap followed by low resolution CID in the ion trap with HCD MS/MS and determined that the HCD approach was superior to CID for large-scale phosphoproteomics82 and used the technique to identify roughly 16 000 phosphorylation sites of metabolically labeled HeLa cell proteins. The conclusions presented in this paper were opposed by the results of Gygi and coinvestigators,83 who compared back-to-back HCD and CID experiments on a pTyr peptide immunoprecipitate and whole phosphoproteome digest. This group determined that, despite higher-scoring peptide data obtained by HCD, CID was superior because the speed of data acquisition allowed for nearly twice as many phosphopeptide assignments in total. Differences between the acquisition parameter sets used, especially the exclusion width mass tolerance for CID, were believed to contribute to the discrepancy between the two studies. Other combinations of MS/MS strategies for protein quantification have been described, including the application of HCD solely for iTRAQ 738

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry reporter ion data and CID/ETD for peptide sequence ion data.66,84 CID methodologies have long been applied to phosphoproteomics, but electron-driven dissociation methods can provide complementary sequence information, often with retention of site-specific phosphorylation. However, these methods induce charge-state reduction of the precursor and can only be applied to ions of z g 2. In electron capture dissociation (ECD),85,86 positive ions are irradiated with low energy electrons (2 and low m/z), whereas CID yields better quality MS/MS data for doubly protonated peptide ions at higher m/z ratios. In addition, low energy collisional activation of ions undergoing ECD or ETD can improve the efficiency of dissociation.88,89 Because both ECD90 and ETD91 can be performed on a chromatographic time scale, they have been successfully applied to phosphoproteomics, especially in approaches that combine CID with electron-induced dissociation.27,92 97 The information-rich MS/MS data sets produced by ECD and ETD experiments have posed new challenges to informatics searches. Search algorithms match MS/MS data to in silico-based theoretical gas-phase cleavages of peptides calculated from models of collision-induced dissociation. Because electron induced dissociation is very different from CID, peptide matching and scoring suffered, with differences between the best and the worst search engines as high as 70%.98 Recent advances in search algorithms have achieved improved peptide identification from ETD data, by removing ETD-specific features from spectra which interfere with standard informatics procedures.99 Zhang developed a model based on kinetics and fragmentation patterns described in the literature100 to predict peptide ETD and ECD spectra. A new database search algorithm based on the probabilistic modeling of shared peaks count and shared peaks intensity between the ETD spectra and the peptide sequences101 was developed and shown to be complementary to standard CIDbased approaches. Chalkley and co-workers improved upon Protein Prospector software performance for ETD data set analysis by weighting results depending upon precursor ion charge state and peptide sequence.102 Because the combined CID/ETD approach continues to increase in popularity, Pandey et al. evaluated data analysis derived from a complex mixture of phosphopeptides.103 They observed that merging CID and ETD spectra prior to database searches reduced the number of spectral matches compared to separate searches of the data sets. Thus, further enhancements provided by new algorithms are still needed to improve analysis of large scale studies based on CID/ETD data sets. The size of data sets (on the order of 106 peptide identifications) that need to be integrated into protein identifications, with quantitation and site-specific modifications, presents significant challenges. Inference of protein identities from large data sets of

REVIEW

tandem MS data derived from peptides continue to run the risk of high false discovery rates (FDR) unless controlled by targetdecoy database searches.104 Furthermore, not all MS/MS search tools may be compliant with the target-decoy database search approach.105

’ QUANTITATION Several strategies to derive changes of abundances of phosphoproteins in response to various circumstances (environment, drug treatment, etc.) have been developed to provide accurate quantitation of peptide phosphorylation. The introduction of stable isotope labels either in vivo or in vitro yields peptide isoforms with identical physicochemical properties that permit simultaneous separation and analysis. The most commonly used methods in large-scale studies are stable isotope labeling by metabolic incorporation of amino acids (SILAC),106,107 chemical modification with isobaric tags,108,109 dimethyl labeling,41 and label-free quantitation.110 Metabolic labeling by SILAC entails the culture of cells in media that contain stable-isotope amino acids, which become incorporated into proteins by cellular synthetic enzymes. An acceptable level of isotopic incorporation requires several cell passages. Up to three samples can be multiplexed. Cell extracts can be combined prior to downstream enrichment and purification steps, which helps to minimizes errors in quantitation. Unlike isobarically tagged peptides, isotopically labeled peptides can be detected in MS mode. In its tenth year at the time of this Review, the SILAC technique continues to be applied with great success in large scale studies of phosphoprotein signaling, contributing to new biological insights in both basic and cancer cell biology.111 114 Recently, an advance in the use of SILAC for in vivo studies by a spike-in SILAC method was reported which allowed for the quantitative comparison of 10 000 phosphorylation sites in a murine response to insulin treatment.115 In their study of metabolically labeled, EGF stimulated HeLa cells, Larsen et al. compared data processing software for phosphopeptide identification28 from a large scale data set. They processed the LTQ-Orbitrap data with DTASupercharge (http:// msquant.sourceforge.net) to generate peaklists that were subsequently searched using Mascot, then further processed and quantified by MSQuant.116 Those results were compared to results obtained from fully automated software tools, Proteome Discoverer from Thermo Scientific and MaxQuant.117 They found that, at the same false discovery rate, DTASupercharge and Proteome Discoverer yielded similar numbers of peptide spectrum matches while the number for MaxQuant was about one-third higher. However, the larger number of matches in MaxQuant were attributed to peptides with Mascot scores below 22, and by manual inspection of the data, it became obvious that the spectra were not of sufficient quality to support the spectral matches. In contrast to SILAC, not only can isobaric tagging reagents (iTRAQ,109 TMT108) be used to encode up to eight different samples derived cell culture but also the technique is readily applied to tissue samples. The reagents contain a reactive group, usually an N-hydroxysuccinimide ester that reacts with peptidyl amines to form a covalent bond. Because each set of chemical tags is isobaric, having the same nominal (iTRAQ) or exact mass (TMT), MS/MS analysis is required to generate sets of reporter ions (m/z range 144 131, depending on the reagent) and decode the relative quantitative changes between sample states or phenotypes. iTRAQ reagents are available in 4- and 8-plex sets 739

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry

REVIEW

Table 1. Partial List of Publicly Available Phosphoproteomic Tools and Databases name

URL

description

Gene Ontology

http://www.geneontology.org/GO.tools.shtml

gene ontology tools

Global Proteome Machine Database

http://gpmdb.thegpm.org/

public repository of proteomics data

Phosida

http://www.phosida.de/

protein phosphorylation data

PhosphoElm

http://phospho.elm.eu.org/

database of S,T,Y phosphorylation sites

Phosphomouse

https://gygi.med.harvard.edu/phosphomouse/index.php

mouse protein and phosphorylation data

PhosphoSite

http://phosphosite.org

curated phosphorylation site database

Pride

http://www.ebi.ac.uk/pride/

public repository of proteomics data

SH2 Domain STRING Database

http://sh2.uchicago.edu/ http://string-db.org/

mouse and human SH2-domain protein information functional protein association network database

Tranche

https://trancheproject.org/

repository of peptide data

and TMT in a 6-plex format. Derivatization of phosphopeptides with isobaric tags was reported to yield higher charge states during electrospray ionization and reduce identification efficiency by as much as 50%.118 If this proves to be a general phenomenon, it could provide one explanation why SILAC-based global phosphoproteomic studies usually provide higher numbers of protein identities compared to those that employ isobaric tags. Quantitation using SILAC entails abundance measurements made in high resolution MS mode. In workflows that employ isobaric tags, peptide precursor ions are usually isolated for MS/MS at lower resolution and concurrent isolation of multiple precursors with similar m/z ratios is quite common, even when LC separation is applied. This phenomenon can degrade the accuracy of quantitation. To address this issue, Wenger et al. applied a fast protontransfer ion ion reaction to essentially perform a gas-phase purification of mixtures of precursor ions prior to HCD MS/ MS and demonstrated increased dynamic range and improved precision of quantification.57 Another major challenge for quantitation based on reporter ions derived from isobaric tagging reagents is the inhomogeneity of variance in the reporter ion intensities, especially if the reporter ion intensities are low.44,119,120 A comparison of the performance of the reagents was studied by Pichler et al.121 on both standard proteins and a complex biological mixture with a hybrid CID/HCD method in an LTQ-Orbitrap. They found that the 4-plex reagents gave the highest average number of identified and quantified peptides; 8-plex iTRAQ showed the poorest results, and TMT reagents were intermediate. The recent addition of an axial electric field in the HCD fragmentation cell improves data precision and yields higher numbers of MS/MS spectra that contain full sets of reporter ions.79 In bottom-up protein sequencing approaches, phosphoprotein quantitation may be skewed by changes in absolute protein amounts.122,123 This is due to the site-specificity of protein phosphorylation, which leads to heterogeneity at the protein level. Gygi and co-workers applied a protein normalization strategy to accurately characterize changes in protein phosphorylation status in a study of the yeast MAPK pathway.122 The group found that nearly one-quarter of regulated phosphopeptide abundance changes depended upon changes in the underlying protein abundances. Their results underscore the need for protein normalization to be applied universally in quantitative phosphoproteomic workflows. Isotope dimethyl labeling of peptides by formaldehyde targets primary N-terminal amino groups and lysine side chains. It is a relatively fast, efficient, and inexpensive method for in vitro labeling. Dimethylation can be applied to studies of cells and tissues.

Up to three different sample states have been analyzed using this approach,124 which was recently applied successfully to studies of tyrosine41,125 and global16 phosphorylation. One benefit of the dimethylation strategy is the low cost of the reagents. Label-free approaches to quantitation involve the integrated measurement of peak areas (area under the curve) for peptides in LC-MS runs126 and have recently been reviewed thoroughly by Neilson et al.110 The measurement of the peak area is reportedly linearly proportional to the peptide concentration in the range of 10 fmol to 100 pmol. Samples must be analyzed separately, and LC retention times and ion counts are compared between treatments or states. Identification of the peptide eluting at a specific retention time is performed by MS/MS for verification. The use of label-free approaches requires rigorous attention to reproducibility of intersample preparation and chromatographic separations and the use of automated methods for alignment of data from multiple LC-MS runs. Two recent studies performed on TiO2-enriched phosphopeptides from zebrafish127 and an acute myeloid leukemia cell line128 demonstrated good precision and linearity.

’ FROM DATA ANALYSIS TO SIGNALING PATHWAYS Once large quantitative data sets have been processed and the peptide and protein identities, modifications, and quantitative changes assigned with confidence, the challenge of mining biological insights begins. In addition to commercial software solutions, there are many open access resources (Table 1) that greatly speed up the process of deriving knowledge from data. Beside protein data repositories such as Tranche and Phosida, resources such as Gene Ontology and STRING can aid in assigning biological changes associated with large data sets. Even with advanced informatics tools for data analysis, manual interpretation and literature searches are still required to glean knowledge from quantitative phosphoproteomic data. The human genome predicts more than 500 kinases,129 and the number of potential phosphorylation sites in the human proteome is perhaps on the order of the hundreds of thousands.130 Large scale studies of protein phosphorylation show that the distribution of phosphorylated residues are typically 80 85% pS, 10 15% pT, and 2% pY.29,95 In general, phosphotyrosine signaling is better studied than serine and threonine phosphorylation. Phosphotyrosinemediated signaling differs from phosphoserine and phosphothreonine signaling (Figure 2) because of the extra signaling modulator represented by SH2-domain containing proteins. Phosphotyrosine-mediated signaling encompasses tyrosine kinases that specifically target Tyr residues, phosphotyrosine phosphatases that 740

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry

REVIEW

Figure 2. Protein phosphorylation is a reversible process mediated by kinases and phosphatases (dephosphorylation) with adenosine triphosphate (ATP) or guanine triphosphate (GTP) as cofactors. Signal transduction cascades modulate protein activity, subcellular localization, and protein protein interactions. In phosphotyrosine signaling (right panel), SH2-domain-containing proteins that recognize and bind to sites of tyrosine phosphorylation on other proteins comprise an extra module in the signaling cascade.

dephosphorylate Tyr, and Src Homology 2 (SH2) domains131 that recognize and bind to phosphorylated tyrosine sites. The network connections become even more complex, because some proteins contain both SH2 and kinase domains, such as the human proto-oncogene tyrosine kinase src.132,133 Proteins that contain SH2 domains play important roles in diseases such as infection,134 cancer,135,136 and diabetes.137 A bioinformatics toolbox (Table 1) has been developed to better characterize mouse and human proteins that contain SH2-domains, on a global scale (the “SH2-ome”). Signaling pathways are often represented as linear cascades. However, complex interactions exist between signaling cascades, such as in the case of the various signal transduction and activator of transcription (STAT) signaling cascades (Figure 3). Signaling occurs in networks of complicated interconnections that integrate multiple cascade initiators, potentiate the responses, and terminate them. In addition, kinase cascades are known to interact with lipid signaling and GTPases. Cross-talk between signaling pathways are frequently detected in quantitative phosphoproteomic studies, such as the identification of an EGFRSTAT5 signaling link in glioma.96 Current MS-based discovery approaches to quantitative phosphoproteomics can yield quantitative data for ∼104 phosphopeptides.29 Because each phosphorylated residue represents the result of the action of a kinase, one should be able to deduce the identity of upstream kinase activation. However, not all substrates of kinases are known. At the moment, there is no single analytical method to fully align the data with existing pathway knowledge. Recently, a targeted quantitative phosphoproteomic method that may aid the identification of kinase substrates was reported.138 This novel approach combined chemical genetics and proteomics in a screening platform. It was applied to the study of the substrates of the kinase Mek1 as a proof of concept experiment. Activated Mek1 was incubated with a cell extract from Mek1-deficient cells. As expected, the downstream targets Erk1 and Erk2 were identified as direct Mek1 targets. The workflow has high specificity.

Figure 3. Canonical representation of the STAT1 and STAT3 signaling pathways. In response to cytokines and growth factors, STAT family members are phosphorylated by the receptor associated kinases and then form homo- or heterodimers that translocate to the cell nucleus where they act as transcription activators. Phospho-STAT activity is decreased by protein inhibitor of STAT (PIAS), a family of sumo-ligases. Like many other signaling pathways, extensive crosstalk exists and can be explored by quantitative phosphoproteomics.

’ NEW BIOLOGICAL INSIGHTS DERIVED FROM QUANTITATIVE PHOSPHOPROTEOMIC APPLICATIONS Because phosphoprotein-mediated signaling systems occur in organisms as diverse as bacteria, plants, and vertebrates, the number of reports that apply phosphoproteomic analysis continues to grow. The application of phosphoproteomics to plant signal transduction was reviewed recently.139 Global phosphoproteomic studies have been performed in Arabidopsis, rice, and Medicago (untreated and phytohormone-stimulated, see Mithoe et al.139). Protein phosphorylation has been studied in bacteria as well (for review, see Kobir et al140). Bacteria can produce histidine and aspartic acid phosphorylated proteins in addition to phosphorylation of serine, threonine, and tyrosine. Bacterial kinases mediate both physiological and pathogenic processes. Lemeer et al.141 used an online TiO2-based enrichment of phosphopeptides coupled to LC-MS/MS in a study of zebrafish embryos. The phosphopeptide data were mined for kinase motifs and yielded the possible identities of kinases involved in zebrafish embryonic development. Tissue-specific protein phosphorylation in nine mouse tissues was described by Huttlin et al.59 This comprehensive study employed SCX-IMAC or SDS-PAGE followed by in-gel digestion and analysis in a hybrid linear ion trap-Orbitrap instrument. In total, 12 000 proteins were identified and 36 000 phosphorylation 741

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry sites assigned in mouse brain, brown fat, heart, liver, lung, kidney, pancreas, spleen, and testis. Even though a large number of phosphorylation sites were assigned, several phosphotyrosine sites previously observed by the group in mouse were not detected in the study, probably because phosphotyrosine enrichment by phosphopeptide antibodies was not used in the enrichment step. Not surprisingly, tissue-specific patterns of protein phosphorylation were observed. The authors were also able to infer the activity of several kinases in mouse tissues, on the basis of the frequency of identified phosphopeptide sequences. Derivation of deep biological knowledge from large-scale studies is quite challenging. The group applied hierarchical clustering and heat maps of the data based on spectral counting to visualize differences between tissues, explored gene ontology to find enriched functional categories of proteins, and created a tissue specific atlas of mouse phosphoproteins (Phosphomouse, Table 1). The broad resource that this work, performed on normal tissues, represents is expected to be very useful in future studies of protein phosphorylation studies in mouse models of disease. Quantitative phosphoproteomic studies continue to contribute to novel biological insights in medicine at an accelerated rate. Recently, two original articles that both employed quantitative phosphoproteomics to identify targets of mTOR, a cellular enzyme that controls cell growth and division, were highlighted in the journal Science.142 144 mTOR is a serine/threonine kinase whose function has not been completely elucidated. It is, however, a protein target of interest to the pharmaceutical industry. Rapamycin, a naturally occurring macrolide compound produced by a bacterium, is a partial mTOR inhibitor. Previous use of rapamycin as a chemical probe of the mTOR signaling pathway led to the identification of a few mTOR substrates, but rapamycin analogues designed for anticancer therapy have been met with limited clinical success. New details from the two recent studies point to the importance of a new candidate target of mTOR signaling, Grb10, a receptor tyrosine kinase with an SH2 domain. The methods employed in the two reports were different; Yu et al.143 used SILAC for quantitation and SCX-IMAC for phosphopeptide enrichment, whereas Hsu et al.144 applied iTRAQ to quantify peptide changes and magnetic metal-chelated agarose beads in the phosphopeptide enrichment step. Results obtained through different techniques in the two different studies converged upon Grb10 as a possible therapeutic target. A second important area in biomedicine that is being propelled by new insights from quantitative phosphoproteomics is stem cell research. Definition of the signaling pathways that drive selfrenewal and pluripotency is the key to understanding stem cell differentiation, both in normal development and in cancer. Rigbolt et al.145 applied combined SILAC and SCX-TiO2 approaches to characterize the global proteomic and phosphoproteomic responses in human embryonic stem cells during differentiation. New molecular links between dynamic phosphorylation and interacting partners of gene-silencing DNA methyltransferases were identified. Similar analytical workflows have been applied to proteomic and phosphoproteomic analysis of human embryonic stem cells in response to the growth factor FGF-2, providing new insights into how this cytokine helps maintain the stem-like state.146,147 In oncology research, cancer stem cells are hypothesized to represent a reserve of cells that are pluripotent and resistant to radiation and chemotherapies designed for differentiated cancer cells.148 Quantitative phosphoproteomics can help define pathways of resistance mediated by phosphoprotein signaling. Recently, a combined phosphoprotein

REVIEW

and phosphopeptide enrichment workflow that employed TMT quantitation was developed to study glioma cancer stem cell (gCSC) responses to perturbations in the STAT3/IL-6/HIF1α autocrine signaling loop.14 The study identified expected pathway changes in response to treatments but also generated new hypotheses for the roles of NO synthase and ubiquitination in gCSCs in response to a STAT3 phosphorylation inhibitor, hypoxia, and IL-6 stimulation. One trend that is beginning to emerge is the use of quantitative phosphoproteomics combined with other system-wide measurements in comprehensive biological studies. By the use of combined enrichment approaches such as ERLIC22,149,150 and lectin affinity-TiO2 enrichment,16,151 glycoproteomic and phosphoproteomic studies can be combined into one study. The potential lack of response of gCSCs to traditional cytotoxic and radiation therapies demands novel research approaches in tumor biology and therapeutics. Phenotypic changes in gCSCs induced by pharmacological treatment can be studied by a variety of systems biology approaches. Global quantitative phosphoproteomics has been employed to define changes in the STAT3-mediated signaling pathway.14 Glycan expression is regulated through glycosyltransferases and hydrolases and can influence downstream phosphorylation signaling cascades. Glycans mediate key pathophysiological events during tumorigenesis, including altered cell adhesion and invasivity, receptor activation, and intracellular signal transduction. A second report on the glycolipid and glycotranscriptomic responses of gCSCs to STAT3 phosphorylation was published recently.152 Taken together, the two papers represent the largest systems biology investigation of gCSCs to date. Because glioma tumors represent a complex system of migrating cells of variable phenotypes that adapt and interact to the environment in complex ways,153 system-wide studies are needed (Figure 4) to further our understanding of their pathobiology. Global quantitative phosphoproteomics is a key technology in this effort, but glycomic and glycotranscriptomic data sets for gCSCs in response to pharmacological treatments and mathematical and computational innovations for graphical modeling to reconstruct cell signaling events are expected to yield significant insights in future investigations. Such insights have importance in translational medicine, including determination of mechanisms of cancer stem cell resistance and mechanism of action of therapeutic agents, and patient stratification.

’ PERSPECTIVES In the past few years, improved methods applied to quantitative phosphoproteomics have led to advances in biology. Technical challenges remain in the areas of accurate global phosphosite identification, reproducibility between technical replicates and experimental configurations, and complexity of data analysis. Tens of thousands of phosphorylations can be detected and quantified, outstripping the pace of assigning biological implications of every phosphorylation site. With the large increase in phosphoproteomic reports based on bottom-up MS approaches, some limitations of the methodology become obvious. Contextual information, such as complete phosphorylation patterns of intact proteins, is lost when hydrolysis of proteins (chemical or enzymatic) is performed prior to MS analysis. A method to derive site-specific phosphorylation in a quantitative phosphoproteomic study of mitosis154 represents an important advancement of this area, but more work is needed. Quantitative 742

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry

REVIEW

Figure 4. Cell signaling and systems biology. Cell signaling processes are modulated by not only phosphoproteins but also by polar lipid expression in the cell membrane and changes in the transcriptome (left). For instance, STAT3 phosphorylation inhibition results in profound changes in phosphoprotein signaling and glycolipid expression in gCSCs.14,152 The cell surface glycan GalNAcβ1 4GlcNAc is required for STAT3 signaling required for mouse embryonic stem cell maintenance.157 When analyzed simultaneously, data sets from quantitative global analysis of phosphoproteomes and glycomes can yield new insights into biological response patterns, when analyzed by interrelated Gaussian Graph Networks (right).158

phosphoproteomics performed on intact proteins is still an intractable issue in global proteomic analysis,155,156 even if topdown analysis can be efficiently performed in targeted assays of proteins in the mid to low mass ranges. In the future, improved analytical methodologies and instrumentation, processing software for large MS data sets, the acquisition of complementary systems biology data, and mathematical graphing techniques will improve both the scope and quality of comprehensive studies by quantitative phosphoproteomics.

State University, and Dr. Jan Krumsiek, Helmholz Zentrum Munich, Germany.

’ REFERENCES (1) Eyrich, B.; Sickmann, A.; Zahedi, R. P. Proteomics 2011, 11, 554–570. (2) Leitner, A.; Sturm, M.; Lindner, W. Anal. Chim. Acta 2011, 703, 19–30. (3) Kosako, H.; Nagano, K. Expert Rev. Proteomics 2011, 8, 81–94. (4) Harsha, H. C.; Pandey, A. Mol. Oncol. 2010, 4, 482–495. (5) Gilar, M.; Olivova, P.; Daly, A. E.; Gebler, J. C. J. Sep. Sci. 2005, 28, 1694–1703. (6) Gauci, S.; Helbig, A. O.; Slijper, M.; Krijgsveld, J.; Heck, A. J. R.; Mohammed, S. Anal. Chem. 2009, 81, 4493–4501. (7) Han, G.; Ye, M.; Zhou, H.; Jiang, X.; Feng, S. S.; Jiang, X.; Tian, R.; Wan, D.; Zou, H.; Gu, J. Proteomics 2008, 8, 1346–1361. (8) Wang, F.; Han, G.; Yu, Z.; Jiang, X.; Sun, S.; Chen, R.; Ye, M.; Zou, H. J. Sep. Sci. 2010, 33, 1879–1887. (9) Ficarro, S. B.; Zhang, Y.; Carrasco-Alfonso, M. J.; Garg, B.; Adelmant, G. O.; Webber, J. T.; Luckey, C. J.; Marto, J. A. Mol. Cell. Proteomics 2011, 10, 1 19. (10) Hennrich, M. L.; Groenewold, V.; Kops, G. J; Heck, A. J. R.; Mohammed, S. Anal. Chem. 2011, 83, 7137–7143. (11) Dai, J.; Wang, L.-S.; Wu, Y.-B.; Sheng, Q.-H.; Wu, J.-R.; Shieh, C.-H.; Zeng, R. J. Proteome Res. 2008, 8, 133–141. (12) Alpert, A. J. J. Chromatogr. 1990, 9, 177–196. (13) Dillon, R.; Nilsson, C. L.; Shi, S. D. H.; Lee, N.; Krastins, B.; Greig, M. J. J. Proteome Res. 2011, 10, 5084–5094. (14) Nilsson, C. L.; Dillon, R.; Devakumar, A.; Rogers, J. C.; Krastins, B.; Rosenblatt, M. M.; Majo, M.; Kaboord, B. J.; Sarracino, D.; Rezai, T.; Prakash, A.; Lopez, M.; Ji, Y.; Priebe, W.; Colman, H.; Lang, F. F.; Conrad, C. A. J. Proteome Res. 2010, 9, 430–443. (15) Zarei, M.; Sprenger, A.; Metzger, F.; Gretzmeier, C.; Dengjel, J. J. Proteome Res. 2011, 10, 3474–3483. (16) Wu, C.-J.; Chen, Y.-W.; Tai, J.-H.; Chen, S.-H. J. Proteome Res. 2011, 10, 1088–1097. (17) Saleem, R. A.; Aitchison, J. D. J. Vis. Exp. 2009, 32, 1–3. (18) McNulty, D. E.; Annan, R. S. Methods Mol. Biol. 2009, 527, 93–105. (19) McNulty, D. E.; Annan, R. S. Mol. Cell. Proteomics 2008, 7, 971–980.

’ AUTHOR INFORMATION Corresponding Author

*Phone: 409-747-1840. Fax: 409-772-9648. E-mail: carol.nilsson@ utmb.edu.

’ BIOGRAPHY Carol L. Nilsson is Professor of Pharmacology and Toxicology and CPRIT Scholar in Cancer Research at the University of Texas Medical Branch. After undergraduate studies in chemistry at U.C. Berkeley, she obtained her M.D. and Ph.D in Clinical Neurochemistry from Goteborg University (GU), Sweden. She was awarded a postdoctoral scholarship from Knut and Alice Wallenberg Foundation and was an Associate Professor of Medical Biochemistry at GU. Prior to her current position, Dr. Nilsson was Director of the Ion Cyclotron Resonance User Program at the National High Magnetic Field Laboratory in Tallahassee, FL, and Senior Principal Scientist at Pfizer Global Research and Development in San Diego, CA. The focus of her research is the use of quantitative phosphoproteomics and related systems biological techniques in neuro-oncology research. ’ ACKNOWLEDGMENT Funding from Cancer Prevention Research Institute of Texas (CPRIT) and the University of Texas is gratefully acknowledged. John Helms (UTMB) is acknowledged for the graphics in this article. Figure 4 was provided by Dr. Anke Meyer-Baese, Florida 743

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry

REVIEW

(51) Yates, J. R.; Cociorva, D.; Liao, L.; Zabrouskov, V. Anal. Chem. 2005, 78, 493–500. (52) Kristensen, D. B.; Imamura, K.; Miyamoto, Y.; Yoshizato, K. Electrophoresis 2000, 21, 430–439. (53) Belov, M. E.; Nikolaev, E. N.; Anderson, G. A.; Udseth, H. R.; Conrads, T. P.; Veenstra, T. D.; Masselon, C. D.; Gorshkov, M. V.; Smith, R. D. Anal. Chem. 2000, 73, 253–261. (54) Syka, J. E.; Coon, J. J.; Schroeder, M. J.; Shabanowitz, J.; Hunt, D. F. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9528–9533. (55) He, F.; Emmett, M. R.; Hakansson, K.; Hendrickson, C. L.; Marshall, A. G. J. Proteome Res. 2004, 3, 61–7. (56) Bossio, R. E.; Marshall, A. G. Anal. Chem. 2002, 74, 1674–1679. (57) Wenger, C. D.; Lee, M. V.; Hebert, A. S.; McAlister, G. C.; Phanstiel, D. H.; Westphall, M. S.; Coon, J. J. Nat. Meth. 2011, 8, 933 935. (58) Palumbo, A. M.; Smith, S. A.; Kalcic, C. L.; Dantus, M.; Stemmer, P. M.; Reid, G. E. Mass Spectrom. Rev. 2011, 30, 600– 625. (59) Huttlin, E. L.; Jedrychowski, M. P.; Elias, J. E.; Goswami, T.; Rad, R.; Beausoleil, S. A.; Villen, J.; Haas, W.; Sowa, M. E.; Gygi, S. P. Cell 2010, 143, 1174–1189. (60) Dephoure, N.; Zhou, C.; Villen, J.; Beausoleil, S. A.; Bakalarski, C. E.; Elledge, S. J.; Gygi, S. P. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 10762–10767. (61) Holt, L. J.; Tuch, B. B.; Villen, J.; Johnson, A. D.; Gygi, S. P.; Morgan, D. O. Science 2009, 325, 1682–1686. (62) Villen, J.; Beausoleil, S. A.; Gerber, S. A.; Gygi, S. P. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 1488–1493. (63) Carr, S. A.; Huddleston, M. J.; Annan, R. S. Anal. Biochem. 1996, 239, 180–192. (64) Palumbo, A. M.; Reid, G. E. Anal. Chem. 2008, 80, 9735–9747. (65) Aguiar, M.; Haas, W.; Beausoleil, S. A.; Rush, J.; Gygi, S. P. J. Proteome Res. 2010, 9, 3103–3107. (66) Mischerikow, N.; Altelaar, A. F. M.; Navarro, J. D.; Mohammed, S.; Heck, A. J. R. Mol. Cell. Proteomics 2010, 9, 2140–2148. (67) Schwartz, J. C.; Syka, J. E. P.; Quarmby, S. T. In The 53rd ASMS Conference on Mass Spectrometry and Allied Topics, San Antonio, TX, June 6 10, 2005. (68) Guo, T.; Gan, C. S.; Zhang, H.; Zhu, Y.; Kon, O. L.; Sze, S. K. J. Proteome Res. 2008, 7, 4831–4840. (69) Bantscheff, M.; Boesche, M.; Eberhard, D.; Matthieson, T.; Sweetman, G.; Kuster, B. Mol. Cell. Proteomics 2008, 7, 1702–1713. (70) Griffin, T. J.; Xie, H.; Bandhakavi, S.; Popko, J.; Mohan, A.; Carlis, J. V.; Higgins, L. A. J. Proteome Res. 2007, 6, 4200–4209. (71) Armenta, J.; Hoeschele, I.; Lazar, I. J. Am. Soc. Mass Spectrom. 2009, 20, 1287–1302. (72) Kocher, T.; Pichler, P.; Schutzbier, M.; Stingl, C.; Kaul, A.; Teucher, N.; Hasenfuss, G.; Penninger, J. M.; Mechtler, K. J. Proteome Res. 2009, 8, 4743–4752. (73) Wu, W. W.; Wang, G.; Insel, P. A.; Hsiao, C.-T.; Zou, S.; Maudsley, S.; Martin, B.; Shen, R. F. J. Am. Soc. Mass Spectrom. 2011, 22, 1753–1762. (74) Onsongo, G.; Stone, M. D.; Van Riper, S. K.; Chilton, J.; Wu, B.; Higgins, L. A.; Lund, T. C.; Carlis, J. V.; Griffin, T. J. Proteomics 2010, 10, 3533–3538. (75) Yang, F.; Wu, S.; Stenoien, D. L.; Zhao, R.; Monroe, M. E.; Gritsenko, M. A.; Purvine, S. O.; Polpitiya, A. D.; Tolic, N.; Zhang, Q.; Norbeck, A. D.; Orton, D. J.; Moore, R. J.; Tang, K.; Anderson, G. A.; Pasa-Tolic, L.; Camp, D. G.; Smith, R. D. Anal. Chem. 2009, 81, 4137–4143. (76) McAlister, G. C.; Phanstiel, D.; Wenger, C. D.; Lee, M. V.; Coon, J. J. Anal. Chem. 2009, 82, 316–322. (77) Olsen, J. V.; Macek, B.; Lange, O.; Makarov, A.; Horning, S.; Mann, M. Nat. Methods 2007, 4, 709–712. (78) Boja, E. S.; Phillips, D.; French, S. A.; Harris, R. A.; Balaban, R. S. J. Proteome Res. 2009, 8, 4665–4675. (79) Pichler, P.; Kocher, T.; Holzmann, J.; Mohring, T.; Ammerer, G.; Mechtler, K. Anal. Chem. 2011, 83, 1468–1474.

(20) Singer, D.; Kuhlmann, J.; Muschket, M.; Hoffmann, R. Anal. Chem. 2010, 82, 6409–6414. (21) Alpert, A. J. Anal. Chem. 2008, 80, 62–76. (22) Hao, P.; Guo, T.; Sze, S. K. PLoS One 2011, 6, e16884. (23) Chien, K.-Y.; Liu, H.-C.; Goshe, M. B. J. Proteome Res. 2011, 10, 4041–4053. (24) Singh, P.; Gan, C. S.; Guo, T.; Phang, H.-Q.; Sze, S. K.; Koh, C. G. Proteomics 2011, 11, 2891–2900. (25) Gan, C. S.; Guo, T.; Zhang, H.; Lim, S. K.; Sze, S. K. J. Proteome Res. 2008, 7, 4869–4877. (26) Garbis, S. D.; Roumeliotis, T. I.; Tyritzis, S. I.; Zorpas, K. M.; Pavlakis, K.; Constantinides, C. A. Anal. Chem. 2011, 83, 708–718. (27) Zhou, W.; Ross, M. M.; Tessitore, A.; Ornstein, D.; VanMeter, A.; Liotta, L. A.; Petricoin, E. F. J. Proteome Res. 2009, 8, 5523–5531. (28) Engholm-Keller, K.; Hansen, T. A.; Palmisano, G.; Larsen, M. R. J. Proteome. Res. 2011, DOI: 10.1021/pr200641x. (29) Olsen, J. V.; Blagoev, B.; Gnad, F.; Macek, B.; Kumar, C.; Mortensen, P.; Mann, M. Cell 2006, 127, 635–648. (30) Zhang, Y.; Ficarro, S. B.; Li, S.; Marto, J. A. J. Am. Soc. Mass Spectrom. 2009, 20, 1425–1434. (31) Gates, M.; Tomer, L., K. J. Am. Soc. Mass Spectrom. 2010, 21, 1649–1659. (32) Novotna, L.; Emmerova, T.; Horak, D.; Kucerova, Z.; Ticha, M. J. Chromatogr., A 2010, 1217, 8032–8040. (33) Ficarro, S. B.; Adelmant, G. O.; Tomar, M. N.; Zhang, Y.; Cheng, V. J.; Marto, J. A. Anal. Chem. 2009, 81, 4566–4575. (34) Sykora, C.; Hoffman, R. C.; Hoffmann, P. Protein Pept. Lett. 2007, 14, 489–496. (35) Ye, J.; Zhang, X.; Young, C.; Zhao, X.; Hao, Q.; Cheng, L.; Jensen, O. N. J. Proteome Res. 2010, 9, 3561–3573. (36) Zhou, H.; Elisma, F.; Denis, N. J.; Wright, T. G.; Tian, R.; Zhou, H.; Hou, W.; Zou, H.; Figeys, D. J. Proteome Res. 2010, 9, 1279–1288. (37) Mamone, G.; Picariello, G.; Ferranti, P.; Addeo, F. Proteomics 2010, 10, 380–393. (38) Vormbrock, I.; Kaber, G.; Hartwig, S.; Eckel, J.; Schr€or, K.; Lehr, S. Arch. Physiol. Biochem. 2010, 116, 181–187. (39) Krenkova, J.; Lacher, N. A.; Svec, F. Anal. Chem. 2010, 82, 8335–8341. (40) Rush, J.; Moritz, A.; Lee, K. A.; Guo, A.; Goss, V. L.; Spek, E. J.; Zhang, H. i.; Zha, X.-M.; Polakiewicz, R. D.; Comb, M. J. Nat. Biotechnol. 2005, 23, 94–101, http://www.nature.com/nbt/journal/v23/n1/suppinfo/nbt1046_S1.html . (41) Boersema, P. J.; Foong, L. Y.; Ding, V. M. Y.; Lemeer, S.; van Breukelen, B.; Philp, R.; Boekhorst, J.; Snel, B.; den Hertog, J.; Choo, A. B. H.; Heck, A. J. R. Mol. Cell. Proteomics 2010, 9, 84–99. (42) Del Rosario, A. M.; White, F. M. Curr. Opin. Genet. Dev. 2010, 20, 23–30. (43) Huang, P. H.; Miraldi, E. R.; Xu, A. M.; Kundukulam, V. A.; Del Rosario, A. M.; Flynn, R. A.; Cavenee, W. K.; Furnari, F. B.; White, F. M. Mol. Biosyst. 2010, 6, 1227–1237. (44) Zhang, Y.; Askenazi, M.; Jiang, J.; Luckey, C. J.; Griffin, J. D.; Marto, J. A. Mol. Cell. Proteomics 2010, 9, 780–790. (45) Iliuk, A. B.; Martin, V. A.; Alicie, B. M.; Geahlen, R. L.; Tao, W. A. Mol. Cell. Proteomics 2010, 9, 2162–2172. (46) Tao, W. A.; Wollscheid, B.; O’Brien, R.; Eng, J. K.; Li, X.-J.; Bodenmiller, B.; Watts, J. D.; Hood, L.; Aebersold, R. Nat. Methods 2005, 2, 591–598. (47) Coopman, P. J. P.; Do, M. T. H.; Barth, M.; Bowden, E. T.; Hayes, A. J.; Basyuk, E.; Blancato, J. K.; Vezza, P. R.; McLeskey, S. W.; Mangeat, P. H.; Mueller, S. C. Nature 2000, 406, 742–747. (48) Ficarro, S. B.; Zhang, Y.; Lu, Y.; Moghimi, A. R.; Askenazi, M.; Hyatt, E.; Smith, E. D.; Boyer, L.; Schlaeger, T. M.; Luckey, C. J.; Marto, J. A. Anal. Chem. 2009, 9, 3440–3447. (49) Makarov, A.; Denisov, E.; Kholomeev, A.; Balschun, W.; Lange, O.; Strupat, K.; Horning, S. Anal. Chem. 2006, 78, 2113–2120. (50) McAlister, G. C.; Berggren, W. T.; Griep-Raming, J.; Horning, S.; Makarov, A.; Phanstiel, D.; Stafford, G.; Swaney, D. L.; Syka, J. E. P.; Zabrouskov, V.; Coon, J. J. J. Proteome Res. 2008, 7, 3127–3136. 744

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry

REVIEW

(109) 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. Mol. Cell. Proteomics 2004, 3, 1154–1169. (110) Neilson, K. A.; Ali, N. A.; Muralidharan, S.; Mirzaei, M.; Mariani, M.; Assadourian, G.; Lee, A.; van Sluyter, S. C.; Haynes, P. A. Proteomics 2011, 11, 535–553. (111) Ge, F.; Xiao, C.-L.; Bi, L.-J.; Tao, S.-C.; Xiong, S.; Yin, X.-F.; Li, L.-P.; Lu, C.-H.; Jia, H.-T.; He, Q.-Y. PLoS One 2010, 5, e13095. (112) Ma, D.; Li, S.-J.; Wang, L.-S.; Dai, J.; Zhao, S.; Zeng, R. Cell Res 2009, 19, 651–664, http://www.nature.com/cr/journal/v19/n5/suppinfo/ cr200946s1.html. (113) Malik, R.; Lenobel, R.; Santamaria, A.; Ries, A.; Nigg, E. A.; K€orner,, R. J. Proteome Res. 2009, 8, 4553–4563. (114) Hilger, M.; Bonaldi, T.; Gnad, F.; Mann, M. Mol. Cell. Proteomics 2009, 8, 1908–1920. (115) Monetti, M.; Nagaraj, N.; Sharma, K.; Mann, M. Nat. Methods 2011, 8, 655–658, http://www.nature.com/nmeth/journal/v8/n8/abs/ nmeth.1647.html#supplementary-information. (116) Mortensen, P.; Gouw, J. W.; Olsen, J. V.; Ong, S.-E.; Rigbolt, K. T. G.; Bunkenborg, J.; Cox, J.; Foster, L. J.; Heck, A. J. R.; Blagoev, B.; Andersen, J. S.; Mann, M. J. Proteome Res. 2010, 9, 393–403. (117) Cox, J.; Mann, M. Nat. Biotechnol. 2008, 26, 1367–1372, http:// www.nature.com/nbt/journal/v26/n12/suppinfo/nbt.1511_S1.html. (118) Thingholm, T. E.; Palmisano, G.; Kjeldsen, F.; Larsen, M. R. J. Proteome Res. 2010, 9, 4045–4052. (119) Karp, N. A.; Huber, W.; Sadowski, P. G.; Charles, P. D.; Hester, S. V.; Lilley, K. S. Mol. Cell. Proteomics 2010, 9, 1885–1897. (120) Casado, P.; Cutillas, P. R. Mol. Cell. Proteomics 2011, 10, 1–11. (121) Pichler, P.; Kocher, T.; Holzmann, J.; Mazanek, M.; Taus, T.; Ammerer, G.; Mechtler, K. Anal. Chem. 2010, 82, 6549–6558. (122) Wu, R.; Dephoure, N.; Haas, W.; Huttlin, E. L.; Zhai, B.; Sowa, M. E.; Gygi, S.; , Mol. Cell. Proteomics 2011, 10, 1 12. (123) Wu, R.; Haas, W.; Dephoure, N.; Huttlin, E. L.; Zhai, B.; Sowa, M. E.; Gygi, S. P. Nat. Methods 2011, 8, 677–683, http://www.nature.com/ nmeth/journal/v8/n8/abs/nmeth.1636.html#supplementary-information. (124) Boersema, P. J.; Raijmakers, R.; Lemeer, S.; Mohammed, S.; Heck, A. J. R. Nat. Protocols 2009, 4, 484–494. (125) Ding, V. M. Y.; Boersema, P. J.; Foong, L. Y.; Preisinger, C.; Koh, G.; Natarajan, S.; Lee, D.-Y.; Boekhorst, J.; Snel, B.; Lemeer, S.; Heck, A. J. R.; Choo, A. B. H. PLoS One 2011, 6, e17538. (126) Chelius, D.; Bondarenko, P. J. Proteome Res. 2002, 1, 317– 323. (127) Soderblom, E. J.; Philipp, M.; Thompson, J. W.; Caron, M. J.; Moseley, A. M. Anal. Chem. 2011, 83, 3758–3764. (128) Montoya, A.; Beltran, L.; Casado, P.; Rodriguez-Prados, J.-C.; Cutillas, P. R. Methods 2011, 54, 340 378. (129) Manning, G.; Plowman, G. D.; Hunter, T.; Sudarsanam, S. Trends Biochem. Sci. 2002, 27, 514–520. (130) Tan, C. S. H.; Pasculescu, A.; Lim, W. A.; Pawson, T.; Bader, G. D.; Linding, R. Science 2009, 325, 1686–1688. (131) Pawson, T. Curr. Opin. Cell Biol. 2007, 19, 112–116. (132) Lim, W. A.; Pawson, T. Cell 2010, 142, 661–667. (133) Pawson, T. Cell 2004, 116, 191–203. (134) Cannons, J. L.; Tangye, S. G.; Schwartzberg, P. L. Annu. Rev. Immunol. 2011, 29, 665–705. (135) De Braekeleer, E.; Douet-Guilbert, N.; Rowe, D.; Bown, N.; Morel, F.; Berthou, C.; Ferec, C.; De Braekeleer, M. Eur. J. Haematol. 2011, 86, 361–371. (136) Hamilton, M. J.; Ho, V. W.; Kuroda, E.; Ruschmann, J.; Antignano, F.; Lam, V.; Krystal, G. Exp. Hematol. 2011, 39, 2–13. (137) Morris, D. L.; Cho, K. W.; Rui, L. Endocrinology 2010, 151, 3643–3651. (138) Morandell, S.; Grosstessner-Hain, K.; Roitinger, E.; Hudecz, O.; Lindhorst, T.; Teis, D.; Wrulich, O. A.; Mazanek, M.; Taus, T.; Ueberall, F.; Mechtler, K.; Huber, L. A. Proteomics 2010, 10, 2015–2025. (139) Mithoe, S. C.; Menke, F. L. H. Phytochemistry 2011, 72, 997 1006.

(80) Pekar Second, T.; Blethrow, J. D.; Schwartz, J. C.; Merrihew, G. E.; MacCoss, M. J.; Swaney, D. L.; Russell, J. D.; Coon, J. J.; Zabrouskov, V. Anal. Chem. 2009, 81, 7757–7765. (81) Kelstrup, C. D.; Hekmat, O.; Francavilla, C.; Olsen, J. V. J. Proteome Res. 2011, 10, 2937–2948. (82) Nagaraj, N.; D’Souza, R. C. J.; Cox, J.; Olsen, J. V.; Mann, M. J. Proteome Res. 2010, 9, 6786–6794. (83) Jedrychowski, M. P.; Huttlin, E. L.; Haas, W.; Sowa, M. E.; Rad, R.; Gygi, S. P.; ,. Mol. Cell. Proteomics 2011, DOI: 10.1074/mcp. M111.009910. (84) Swaney, D. L.; McAlister, G. C.; Coon, J. J. Nat. Methods 2008, 5, 959–964, http://www.nature.com/nmeth/journal/v5/n11/suppinfo/ nmeth.1260_S1.html. (85) Zubarev, R. A.; Horn, D. M.; Fridriksson, E. K.; Kelleher, N. L.; Kruger, N. A.; Lewis, M. A.; Carpenter, B. K.; McLafferty, F. W. Anal. Chem. 2000, 72, 563–573. (86) Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W. J. Am. Chem. Soc. 1998, 120, 3265–3266. (87) Coon, J. J.; Ueberheide, B.; Syka, J. E. P.; Dryhurst, D. D.; Ausio, J.; Shabanowitz, J.; Hunt, D. F. Proc. Nat. Acad. Sci. U.S.A. 2005, 102, 9463–9468. (88) Swaney, D. L.; McAlister, G. C.; Wirtala, M.; Schwartz, J. C.; Syka, J. E. P.; Coon, J. J. Anal. Chem. 2006, 79, 477–485. (89) Horn, D. M.; Ge, Y.; McLafferty, F. W. Anal. Chem. 2000, 72, 4778–4784. (90) Creese, A.; Cooper, H. J. J. Am. Soc. Mass Spectrom. 2007, 18, 891–897. (91) Udeshi, N. D.; Compton, P. D.; Shabanowitz, J.; Hunt, D. F.; Rose, K. L. Nat. Protoc. 2008, 3, 1709–1717. (92) Sweet, S. M. M.; Bailey, C. M.; Cunningham, D. L.; Heath, J. K.; Cooper, H. J. Mol. Cell. Proteomics 2009, 8, 904–912. (93) Bergstr€om Lind, S.; Molin, M.; Savitski, M. M.; Emilsson, L.; Åstr€om, J.; Hedberg, L.; Adams, C.; Nielsen, M. L.; Engstr€om, Å.; Elfineh, L.; Andersson, E.; Zubarev, R. A.; Pettersson, U. J. Proteome Res. 2008, 7, 2897–2910. (94) Wu, S.-L.; Huhmer, A. F. R.; Hao, Z.; Karger, B. L. J. Proteome Res. 2007, 6, 4230–4244. (95) Swaney, D. L.; Wenger, C. D.; Thomson, J. A.; Coon, H, J. J. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 995–1000. (96) Chumbalkar, V.; Latha, K.; Hwang, Y. H.; Maywald, R.; Hawley, L.; Sawaya, R.; Diao, L.; Baggerly, K.; Cavenee, W. K.; Furnari, F. B.; Bogler, O. J. Proteome Res. 2011, 10, 1343–1352. (97) Deng, N.; Zhang, J.; Zong, C.; Wang, Y.; Lu, H.; Yang, P.; Wang, W.; Young, G. W.; Wang, Y.; Korge, P.; Lotz, C.; Doran, P.; Liem, D. A.; Apweiler, R.; Weiss, J. N.; Duan, H.; Ping, P. Mol. Cell. Proteomics 2011, 10, 1 14. (98) Kandasamy, K.; Pandey, A.; Molina, H. Anal. Chem. 2009, 81, 7170–7180. (99) Good, D. M.; Wenger, C. D.; Coon, J. J. Proteomics 2010, 10, 164–167. (100) Zhang, Z. Anal. Chem. 2010, 82, 1990–2005. (101) Sadygov, R. G.; Good, D. M.; Swaney, D. L.; Coon, J. J. J. Proteome Res. 2009, 8, 3198–3205. (102) Baker, P. R.; Medzihradszky, K. F.; Chalkley, R. J. Mol. Cell. Proteomics 2010, 9, 1795–1803. (103) Kim, M.-S.; Zhong, J.; Kandasamy, K.; Delanghe, B.; Pandey, A. Proteomics 2011, 11, 2568–2572. (104) Elias, J. E.; Gygi, S. P. In Proteome bioinformatices; Hubbard, S. J.; Jones, A. R., Eds.; Humana Press: New York, 2010; Vol. 604, pp 55 71. (105) Gupta, N.; Bandeira, N.; Keich, U.; Pevzner, P. A. J. Am. Soc. Mass Spectrom. 2011, 22, 1111–1120. (106) Jiang, H.; English, A. M. J. Proteome Res. 2002, 1, 345–350. (107) Ong, S.-E.; Blagoev, B.; Kratchmarova, I.; Kristensen, D. B.; Steen, H.; Pandey, A.; Mann, M. Mol. Cell. Proteomics 2002, 1, 376–386. (108) Thompson, A.; Sch€afer, J.; Kuhn, K.; Kienle, S.; Schwarz, J.; Schmidt, G.; Neumann, T.; Hamon, C. Anal. Chem. 2003, 75, 4942– 4942. 745

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746

Analytical Chemistry

REVIEW

(140) Kobir, A.; Shi, L.; Boscovic, A.; Grangeasse, C.; Franjevic, D.; Mijakovic, I. Biochim. Biophys. Acta 2011, 1810, 989 994. (141) Lemeer, S.; Pinkse, M. W. H.; Mohammed, S.; Van Breukelen, B.; den Hertog, J.; Slijper, M.; Heck, A. J. R. J. Proteome Res. 2008, 7, 1555–1564. (142) Yea, S. S.; Fruman, D. A. Science 2011, 332, 1270–1271. (143) Yu, Y.; Yoon, S.-O.; Poulogiannis, G.; Yang, Q.; Ma, X. M.; Villen, J.; Kubica, N.; Hoffman, G. R.; Cantley, L. C.; Gygi, S. P.; Blenis, J. Science 2011, 332, 1322–1326. (144) Hsu, P. P.; Kang, S. A.; Rameseder, J.; Zhang, Y.; Ottina, K. A.; Lim, D.; Peterson, T. R.; Choi, Y.; Gray, N. S.; Yaffe, M. l. B.; Marto, J. A.; Sabatini, D. M. Science 2011, 332, 1317–1322. (145) Rigbolt, K. T. G.; Prokhorova, T. A.; Akimov, V.; Henningsen, J.; Johansen, P. T.; Kratchmarova, I.; Kassem, M.; Mann, M.; Olsen, J. V.; Blagoev, B. Sci. Signal. 2011, 4, rs3. (146) Zoumaro-Djayoon, A. D.; Ding, V.; Foong, L.-Y.; Choo, A.; Heck, A. J. R.; Mu~noz, J. Proteomics 2011, 11, 3962–3971. (147) Munoz, J.; Heck, A. J. R. Methods Mol. Biol. 2011, 767, 297–312. (148) Eyler, C. E.; Rich, J. N. J. Clin. Oncol. 2008, 26, 2839–2845. (149) Zhang, H.; Guo, T.; Li, X.; Datta, A.; Park, J. E.; Yang, J.; Lim, S. K.; Tam, J. P.; Sze, S. K. Mol. Cell. Proteomics 2010, 9, 635–647. (150) Hao, P.; Guo, T.; Li, X.; Adav, S. S.; Yang, J. L.; Wei, M.; Sze, S. K. J. Proteome Res. 2010, 9, 3520–3526. (151) Mausbacher, N.; Schreiber, T. B.; Daub, H. Mol. Cell. Proteomics 2010, 9, 2337–2253. (152) He, H.; Nilsson, C. L.; Emmett, M. R.; Marshall, A. G.; Kroes, R. A.; Moskal, J. R.; Ji, Y.; Colman, H.; Priebe, W.; Lang, F. F.; Conrad, C. A. J. Proteome Res. 2010, 9, 2098–2108. (153) McLendon, R. E.; Rich, J. N. J. Oncol. 2011, 2011, 397195. (154) Olsen, J. V.; Vermeulen, M.; Santamaria, A.; Kumar, C.; Miller, M. L.; Jensen, L. J.; Gnad, F.; Cox, J.; Jensen, T. S.; Nigg, E. A.; Brunak, S.; Mann, M. Sci. Signal 2010, 3, ra3. (155) Compton, P. D.; Zamdborg, L.; Thomas, P. M.; Kelleher, N. L. Anal. Chem. 2011, 83, 6868–6874. (156) Tipton, J. D.; Tran, J. C.; Catherman, A. D.; Ahlf, D. R.; Durbin, K. R.; Kelleher, N. L. J. Biol. Chem. 2011, 286, 25451–25458. (157) Sasaki, N.; Shinomi, M.; Hirano, K.; Ui-Tei, K.; Nishihara, S. Stem Cells 2011, 29, 641–650. (158) Mueller, N.; Krumsiek, J.; Boehm, C.; Theis, F.; Meyer-Baese, A. SPIE Symp. Comput. Intell. 2011.

746

dx.doi.org/10.1021/ac202877y |Anal. Chem. 2012, 84, 735–746