Briefings in Functional Genomics Advance Access originally published online on February 7, 2006
Briefings in Functional Genomics 2006 4(4):363-376; doi:10.1093/bfgp/eli007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Technique Review |
Characterizing phosphoproteins and phosphoproteomes using mass spectrometry
Dr Michael B. Goshe, Department of Molecular and Structural Biochemistry, North Carolina State University, 128 Polk Hall, Campus Box 7622, Raleigh, NC 27695-7622, USA. Tel: +1 919 513 7740; Fax: +1 919 515 2047; E-mail: michael_goshe{at}ncsu.edu
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONThe reversible phosphorylation of proteins plays a major role in many vital cellular processes by modulating protein function and transmitting signals within cellular pathways and networks. Because phosphorylation is dynamic and the sites of modification cannot be predicted by an organism's genome, proteomic measurements are required to identify sites of and changes in the phosphorylation state of proteins. The low stoichiometry of phosphorylation sites that accompany the multifarious nature of protein phosphorylation in biological systems continues to challenge the dynamic range of present mass spectrometry (MS) technologies and proteomic measurements, despite the preponderance of research and analytical methods devoted to this area. This review addresses some of the strategies and limitations involving the use of MS to map and quantify changes in protein phosphorylation sites for samples that range from a single protein to an entire proteome, and presents several compelling reasons as to why comprehensive phosphorylation site analysis has proven to be so elusive without a hypothesis-driven experimental approach to elicit more meaningful and confident results.
Keywords: phosphoprotein, phosphoproteomics, proteomics, liquid chromatography, mass spectrometry, affinity labeling, stable isotope coding, quantification, quantitation
|
INTRODUCTION |
|---|
The reversible phosphorylation of proteins is a critical post-translation modification that is used to control protein function. The presence or absence of phosphate groups on amino acid side chains, such as serine, threonine and tyrosine, is used to modulate protein activity and propagate signals within cellular pathways and networks [1, 2]. Since phosphorylation is dynamic, the sites of phosphorylation cannot be predicted by an organism's genome, and thus requires analytical measurements to ascertain the sites of modification. Phosphorylation site analysis using mass spectrometry (MS) can be quite challenging when analysing just one protein and quickly becomes a daunting task when attempting to perform proteome-wide measurements (termed ‘phosphoproteomics’). The goal of phosphoproteomics is to use MS to identify sites of, and quantify changes in, all protein phosphorylation events. Despite the tremendous interest in understanding the nature of phosphorylation, in both the plant and animal kingdoms, and the enormous attention it has been given in the field of proteomics and MS analysis, a true proteome-wide study chronicling phosphorylation events has yet to be realized. This review addresses some of the reasons why comprehensive phosphorylation site analysis has proven to be so elusive and tries to promote a more hypothesis-driven experimental approach using existing methods in order to elicit more meaningful and confident results.
|
BIOCHEMICAL ANALYSIS OF PROTEIN PHOSPHORYLATION |
|---|
As more molecular and cellular biologists are turning their attentions away from microarray analysis to study their favorite genes towards investigations regarding the activity of the corresponding proteins (i.e. gene products), the need to understand how and when these proteins are phosphorylated is often deemed necessary, especially if one is to ascertain if the proteins play a substantial role in signal transduction and transcriptional control. Initially a considerable amount of traditional biochemical analysis is performed to characterize the protein or proteins of interest. Often this is accomplished by the use of [32P]ATP to characterize kinase activity and potential substrates. Western blot analysis using phosphoimaging based on the incorporation of the 32P-radiolabel or antibodies recognizing amino acid residues modified with a phosphate moiety is used to measure phosphorylation. With these methods the site or sites of protein phosphorylation are not determined nor are the phosphorylation site stoichiometries. Since [32P]ATP is added exogenously to the sample, it can only be used to ascertain phosphorylation events that occur after this addition, thus providing no insight regarding the state of protein phosphorylation prior to its introduction. This is where the use of MS can prove to be quite informative, since it can be used to study phosphorylation events both in vivo and in vitro without the use of [32P]ATP, enabling autophosphorylation events to be characterized. Unfortunately, simply applying conventional MS techniques that are used for protein identification may not provide sufficient measurements regarding phosphorylation, thus necessitating more advanced labeling and fractionation techniques to promote more efficient MS analysis. With so many methods already existing and well documented in the literature, it can be quite an arduous task to decide on the ‘best’ MS method to study protein phosphorylation. However, this selection process can be performed logically based on hypothesis-driven investigations to study phosphorylation. In this manner, samples and instrument time can be effectively utilized to yield more meaningful results.
|
MASS SPECTROMETRY ANALYSIS OF PHOSPHOPROTEINS |
|---|
A tremendous amount of protein and proteomic analysis relies on liquid chromatography-tandem MS (LC/MS/MS) to detect and quantify constituent peptides of enzymatically digested proteins (Figure 1) [3]. In this ‘bottom-up’ method, proteins are proteolytically digested (usually with trypsin); the peptides are separated by reversed-phase LC and the eluting peptides are captured and fragmented by the mass spectrometer. Each MS/MS spectrum is a collection of ions produced by collision-induced dissociation (CID) of the intact peptide. The peptide fragmentation preferentially occurs at peptide bonds to generate N-terminal fragments (b ions) and C-terminal fragments (y ions) at specific mass-to-charge (m/z) ratios and intensities that provide information regarding amino acid sequence and sites of modification. Although phosphorylation is merely an additional mass modification added to a subset of amino acid side chains of proteins, its analysis using LC/MS/MS can challenge the analytical mettle of the most seasoned mass spectrometrist due to the biological and chemical nature of protein phosphorylation.
|
Although it is predicted that nearly 30% of all proteins are phosphorylated [1, 2], phosphoproteins tend to be of low abundance often with multiple phosphorylation sites of varying stoichiometries and thus presents several challenges to MS analysis. First, protein phosphorylation stoichiometry is typically greater than four orders of magnitude, a dynamic range not usually achieveable with commercial MS instruments. Second, because phosphorylation occurs at solvent-accessible sites or sites that are accessible to kinases and phosphatases, complete sequence coverage of a protein using a ‘bottom-up’ approach is warranted. A three-dimensional structure of the protein could be used to determine solvent-accessible sites, but serve a limited purpose since proteins are dynamic molecules that change conformation upon binding with other proteins, ligands and allosteric regulators that can expose previously buried residues to phosphorylation. Third, the presence of other proteins in the sample can also impede the analysis of the phosphoprotein. Thus, characterizing phosphorylation events, even for a modest number of proteins, requires some method of isolation or enrichment prior to MS analysis.
The most widely used method to separate proteins has been sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) which can be used in conjunction with MS to identify and characterize phosphoproteins. The use of reversed-phase LC/MS/MS to identify in vitro sites of phosphorylation on peptides obtained from individual phosphoproteins can be achieved using hypothesis-driven approaches. If the protein to be studied has been biochemically determined to be phosphorylated via a kinase assay, then a set of expression vectors yielding FLAG- or GFP-tagged proteins can be used for both in vitro and in vivo functional analysis where immunoprecipitation and SDS-PAGE can be effectively used in conjunction with MS to map a number of in vivo phosphorylation sites from proteins obtained from tissues [4]. These directed and hypothesis-driven studies can be quite informative, but as shown by Wang et al. [4] complete phosphorylation site mapping for a given protein studied by this technique requires more than one simple measurement. For example, performing the proteolytic digestion using only trypsin may not produce small enough peptides for LC/MS/MS analysis (depending on the MS instrumentation used) thus requiring another enzyme (e.g. Glu-C) to be used separately or in combination with trypsin to get sufficient coverage. Also, the LC separation may have to be performed using alternative gradients (e.g. a combination of steep and shallow gradients) or stationary phases (e.g. C18 and C30) to enhance detection of phosphopeptides by LC/MS/MS. These multiple analyzes in turn require more material, and depending upon the LC and MS capabilities of a given lab, or the amount per analysis charged by core facilities, comprehensive identification of protein phosphorylation can also be quite costly in both time and money.
Although SDS-PAGE is used to separate phosphoproteins present in immunoprecipitates, it can present problems when coupled with MS analysis. Typically, at least two unique peptides are required for a positive identification of a protein from a gel slice, coverage that is inadequate for phosphorylation site analysis. To circumvent this, the amount of protein used during SDS-PAGE is increased. This may result in increased protein coverage during subsequent in-gel digestion analysis and permit new sites of phosphorylation to be identified; however, some regions of the protein may still not be identified due to low peptide extraction efficiencies from the gel matrix. This becomes more problematic when performing kinetic measurements of protein phosphorylation for two reasons. First, for a kinetic time course, several measurements are required, which will increase sample requirements and may limit the amount of phosphoprotein for each time point. Second, to accurately study the phosphorylation events, it is imperative that the same amount of phosphoprotein be analyzed in each time point measurement, a difficult task to perform for in vivo phosphorylation measurements due to the variability of the immunoprecipitation of specific proteins and subsequent SDS-PAGE MS analysis. If more immunoprecipitated protein is used for one time point measurement than the previous or subsequent ones, a phosphopeptide may be identified and attributed to the variable being studied rather than the consequence of more protein being available and permitting the low stoichiometry of phosphorylation to be identified by MS. To normalize these measurements, stable isotopes can be used as internal controls to more accurately quantify phosphorylation (see sections that follow).
In addition to their low abundance and varying stoichiometries, phosphoproteins present another unique challenge to MS/MS analysis. Peptide fragmentation for MS/MS analysis is almost exclusively performed using CID. However, the absolute identification of phosphoseryl (pSer) and phosphothreonyl (pThr) peptides and to a lesser extent phosphotyrosyl (pTyr) peptides by MS/MS analysis is highly dependent on peptide sequence, since the phosphate group (i.e. H3PO4 or HPO3) is extremely labile under the CID conditions most often employed for peptide fragmentation [5]. Although partial loss of the phosphate moiety can be tolerated, it can prevent unequivocal identification of the phosphorylated residue and can prevent phosphopeptides from being quantified with the precision and accuracy necessary to study protein phosphorylation at the proteome level. The use of electron capture dissociation (ECD) [6] holds much promise for LC/MS/MS phosphoproteomics because the phosphate moiety remains intact during peptide fragmentation. At the present time commercial MS instrumentation is not equipped with ECD capabilities, so most measurements still rely upon conventional CID techniques which typically involve manual inspection of the data to verify product ion assignments and neutral losses or the use of alternative analysis procedures [7]. The application of ‘top-down’ proteomics to study phosphorylation at the protein level is also being examined with ECD techniques [8].
|
GEL-BASED APPROACHES |
|---|
Although comprehensive phosphorylation site determination of a protein using SDS-PAGE MS is difficult, it has not deterred the use of two-dimensional (2D) gel analysis for proteomic applications [9–11]. During 2D-PAGE, proteins are separated by isoelectrophoresis in the first dimension and SDS-PAGE in the second dimension (Figure 2). The separated proteins are visualized by implementing a variety of staining techniques using various fluorescence dyes to probe either protein or phosphoprotein abundance. Each dye produces a ‘spotted’ pattern proportional to the protein abundance profile for a given sample. When comparing two samples, the differences in staining intensities for defined spots are quantified by densitometry measurements. For stained spots of interest, gel slices are excised and the proteins are in-gel proteolytically digested (normally using trypsin) to produce peptides that are extracted from the gel matrix and analyzed by peptide mass fingerprinting using matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) MS. With MALDI, the peptide ionization (solid-to-gas phase transition) is facilitated by matrix molecules that profoundly affect the ionization of phosphopeptides and subsequent identification. An alternative approach for identification would be to use LC/MS/MS employing electrospray ionization (ESI) (liquid-to-gas phase transition) via an applied voltage. Because MALDI and ESI are based on different chemical and physical principles, both methods seldom produce the same ionization efficiency for a given phosphopeptide. Thus to ensure the most comprehensive coverage of protein phosphorylation events, both ionization modes may be required for effective analysis.
|
The use of 2D-PAGE has garnered a dubious reputation in the field of proteomics for a variety of reasons, primarily due to analytical limitations for its application in comprehensive proteomic analysis [12]. Highly acidic and basic proteins as well as very small and very large proteins cannot be effectively separated, and hydrophobic membrane-bound proteins or proteins of low abundance are usually not detected. The occurrence of multiple protein forms per spot or multiple spots per protein, typically a result of phosphorylation, makes quantitative analysis difficult and tends to produce low sequence coverage of in-gel digested proteins for MS analysis. Together these limitations affect exhaustive phosphoprotein analysis. However, 2D-PAGE may have its greatest utility in rapidly generating a ‘snapshot’ of a phosphoproteome, making it extremely useful as an initial screening technique. For example, in setting up a series of kinetic experiments to monitor proteomic phosphorylation changes, the number and time intervals between measurements can be most effectively determined using 2D-gel spot density analysis. Time points eliciting the greatest detectable changes can be probed by in-gel digestion analysis with the remaining samples analyzed by more sensitive LC/MS-based labeling and quantification techniques as described next.
|
ISOTOPE CODING AND MS DETECTION FOR RELATIVE QUANTIFICATION |
|---|
In addition to detecting and cataloging the proteins present in a proteome, the quantification of differences in the protein component of tissues, cells or organelles of different species or states is increasingly being recognized as a key objective of MS-based proteomics research. This is because differential analysis of protein expression, degradation and modification will provide a more accurate and comprehensive view of the dynamic changes occurring within a proteome and can be used as a discovery tool to provide new avenues of biochemical research. This is especially true for phosphorylation events since transcriptome analyzes using cDNA microarrays fail to adequately describe the physiological responses of cellular function based on crucial and decisive phosphorylation events. To achieve these proteomic quantitative measurements, especially for low abundance phosphoproteins, extensive sample fractionation is required, which include isolation of cellular organelles or protein complexes coupled with multi-dimensional LC or the use of affinity-based isotope-coded MS labeling techniques [13, 14] to qualitatively and quantitatively determine the phosphorylation states of proteins [15–17].
Isotope coding of proteins is the process by which proteins (or their corresponding peptides) are labeled with a combination of stable isotopes that are used to differentiate control from treated samples while permitting relative quantification between two distinct proteins or proteomes to be measured. This ‘chemical tagging’ strategy, outlined in Figure 3, involves the modification of protein functional groups such as amino acid side chains (e.g. ICAT, isotope-coded affinity tag [18]) and has been developed for phosphoprotein analysis [13, 14]. In this gel-free approach chemical modifications are used to attach an affinity tag (e.g. a biotin moiety or solid phase reagent) to the targeted functional group and permit the isolation of peptides containing the modification. Affinity tagging combined with stable isotope labeling of proteins/peptides permits the reduction of sample complexity using chromatography while providing a means for simultaneous identification and relative quantification using MS analysis.
|
) (containing 1H, 12C, 14N or 16O atoms), and the other sample with the ‘heavy’ tag (•) (containing, 2H, 13C, 15N or 18O atoms) to introduce a covalent modification to a preferred functional group or phosphorylated residue (R). The two samples are then combined in a 1:1 ratio and proteolytically digested. For alternative strategies proteolytic digestion is performed prior to isotope labeling R at the peptide level, then combined. The labeled peptides are selectively enriched using affinity chromatography or an alternative selection technique, such as capture on a solid-phase support, based on the incorporated functionality of the labeling reagent. The isotope-coded peptides are subjected to liquid chromatography-tandem mass spectrometry analysis (LC/MS/MS). Labeled peptides are identified by matching the MS/MS spectra (lower right) against a translated genomic database. By integrating the signal intensity for each isotope-coded peptide (lower left), the ratio of protein or phosphorylation site abundances (L/H) between the two samples can be determined. The mass difference between the two samples (due to the incorporation of different isotopes) produces readily measurable changes in mass-to-charge (m/z) ratios. Because the light and heavy forms serve as mutual internal standards, the relative intensities of the mass-differentiated forms should accurately reflect the ratios of the peptides (and therefore the proteins or extent of phosphorylation) present in the original samples. The precision of quantitative proteomic measurements made with stable isotope labeling can be on the order of 20% error using a variety of mass analyzers. |
AFFINITY CHROMATOGRAPHY APPROACHES |
|---|
One approach to increase MS detection of phosphoproteins is to enrich sample preparations from proteomic mixtures by utilizing noncovalent binding and recognition of phosphate groups. Enrichment of phosphoproteins from complex mixtures can be performed by affinity chromatography using immobilized antibodies specific for pSer, pThr and pTyr residues. Although monoclonal antibodies are used, nonspecific binding of nonphosphorylated proteins tends to occur in addition to the inevitable losses of phosphoproteins that occur prior to elution. Thus to ensure coverage all flow through and washing steps need to be monitored for the presence of phosphoproteins.
Another technique more amenable to LC/MS/MS analysis is the use of immobilized metal affinity chromatography (IMAC) for phosphopeptide enrichment. IMAC relies on the affinity of phosphate groups for certain metal ions (e.g. Fe3+ or Ga3+) bound to tethered chelating reagents present on solid-phase supports. After proteolytic digestion, phosphopeptides are isolated by IMAC and subsequently analyzed using LC/MS/MS. However, IMAC also enriches aspartyl- and glutamyl-peptides due to the prevalence of the carboxylic acid side chains that promote nonspecific binding. This may not be problematic for a sample containing just a few proteins; however, a more acidic protein (i.e. one with a low pI) present in the sample could complicate downstream MS analysis. Hunt and co-workers have addressed this issue by converting the carboxylic acid groups of peptides to their corresponding methyl esters in order to attain higher specific binding of phosphopeptides using IMAC [19]. The esterification is amenable to stable isotope labeling with d0/d3-methanol and can be used to perform comparative proteomic studies [20] (Figure 4). A possible limitation for effective quantification of phosphorylation using this isotope coding method is that partial ester hydrolysis can occur during various stages of sample workup and chromatography that would produce a mixture of differentially esterified forms for each phosphopeptide thus compromising quantitative measurements. It should also be noted that depending on the esterification procedure used, the presence of residual water may promote deamination of asparaginyl and glutaminyl residues that can be subsequently converted into the corresponding methyl esters [21] and lead to incorrect identification and quantification. In addition, since IMAC exploits the affinity of the first row transition metals (e.g. Fe3+) to functional groups containing oxygen, sulfur and nitrogen, it is also capable of enriching for sulfonated peptides and other contaminates through nonspecific interactions regardless whether esterification is used. The most salient observation is that IMAC with esterification can significantly enrich for phosphopeptides, but depending on the nature of the protein sample being analyzed, it can enrich for some other unexpected components or be modified to enrich for another peptide class. For example, Ren et al. [22] utilized IMAC columns loaded with Cu2+ and N-acetylation of amines to enrich for histidyl-peptides from protein digests.
|
Because of its ease of use, the IMAC procedure has been applied to a number of phosphoproteomic studies. Ficarro et al. [20] mapped more than 60 phosphorylated sequences in a whole protein digest from capacitated sperm by MS/MS. Using d0/d3-methanol coding in combination with IMAC, He et al. [21] made comparative measurements of enriched phosphopeptides obtained from in vitro cultured human lung cells under control and starvation conditions. Salomon et al. [23] applied this method to identify pTyr sites during the activation of human T cells and to measure phosphorylation changes occurring in chronic myelogenous leukemia cells upon treatment with the inhibitor of oncogenic BCR-ABL kinase activity. Using IMAC, Brill et al. [24] assigned 70 tyrosine-phosphorylated peptides from human T cell lysate, and Bieber et al. [25] used immobilized metal-affinity pipettes to enrich for phosphoproteins obtained from human saliva. Recently, an IMAC/LC/ESI-MS system has been developed to automate and further enhance phosphopeptide analysis [26].
|
CHEMICAL TAGGING APPROACHES |
|---|
The strategy of using chemical labeling of phosphate groups to enable enrichment of phosphopeptides through covalent modification has also been explored by several groups. One approach utilizes high affinity avidin-biotin binding with immobilized supports to allow removal of nonphosphorylated peptides during washing. The method reported by Chait and coworkers [27] involves base-catalyzed ß-elimination of the phosphate group from pSer and pThr residues and subsequent Michael addition of ethanedithiol. The new thiolate moiety can be covalently modified with a biotinylated affinity tag containing a maleimide group, thereby allowing avidin affinity chromatography to be used to isolate the modified phosphopeptides. The main disadvantages of this phosphoproteomic method are that it is not applicable to tyrosine phosphorylation and the maleimidyl portion of the tag undergoes partial hydrolysis, resulting in two products for each labeled peptide. In an alternative approach, dithiothreitol (DTT) is used to generate the Michael adduct; the DTT-labeled peptides are covalently attached via disulfide exchange to an immobilized thiolate resin and subsequently released using an excess of DTT and analyzed by MALDI-MS [28].
In the solid-phase method developed by Zhou et al. [29] phosphopeptide capture is independent of the nature of the phosphorylated side chain, making it applicable for studying serine, threonine and tyrosine phosphorylation. This method involves blocking of amino groups of proteolytically digested peptides using tBoc (tert-butoxycarbonyl) chemistry. This is followed by carbodiimide catalyzed condensation of ethanolamine with the phosphate and carboxylate groups of the peptide to form phosphoramidate and amide bonds, respectively. Phosphate groups are regenerated by selective cleavage of ethanolamine with diluted trifluoroacetic acid (TFA). Following the addition of cysteamine to the regenerated phosphate groups and reduction of the disulfide bonds to release free thiolate groups, the phosphopeptides are captured with glass beads containing a thiolate-reactive immobilized iodoacetamide functionality. Cleavage of phosphoramidate bonds with concentrated TFA releases the captured phosphopeptides and simultaneously removes the tBoc-protecting groups from the amines while the modified carboxylate groups remain intact. Although this covalent coupling strategy permits stringent washing of noncovalently bound peptides, resulting in highly enriched mixtures of phosphopeptides, it does require a significant number of derivatization steps that could affect phosphopeptide yields. Although not performed by the authors, isotopic variants of ethanolamine to block the carboxylate groups of the proteins (similar to the isotope coding strategy used in IMAC) could be used to quantify the relative phosphopeptide abundance.
Stable isotope coding using affinity tags to quantify protein levels, as demonstrated with ICAT, has been extended to phosphoproteins (Figure 5) as described by Goshe et al. with the development of a phosphoprotein isotope-coded affinity tag (PhIAT) [30, 31]. In this method, cysteinyl-residues are blocked by performic acid oxidation, and the phosphate groups of pSer and pThr residues are removed by hydroxide ion-mediated ß-elimination to produce thiolate-reactive sites (i.e. 
, ß-unsaturated double bonds). When comparing the relative phosphorylation states of phosphopeptides from two distinct samples, these thiolate-reactive sites are modified with isotopic versions of 1,2-ethanedithiol (EDT) that contain either four alkyl hydrogens (d0-EDT) or four alkyl deuteriums (d4-EDT). Once isotopically labeled, the d0/d4-EDT-labeled residues are biotinylated using (+)-biotinyl-iodoacetamidyl-3,6-dioxaoctanediamine. The phosphorylated peptides are purified and concentrated using affinity chromatography and analyzed by MS. The relative stability of the PhIAT-label during CID enables the site of phosphorylation to be identified from the MS/MS spectrum and permits the state of phosphorylation to be quantified [31].
|
The PhIAT approach for phosphoprotein analysis has several limitations as shared with the ICAT approach. Due to the use of immobilized avidin during affinity chromatography, there is difficulty in removing all nonspecifically bound peptides, and sample recovery is reduced because of irreversible binding of a subpopulation of the biotinylated peptides [32]. In addition, the use of deuterium atoms in isotope coding causes differential elution of isotopomers during LC [33], and peptides containing multiple pSer and pThr residues are difficult to fragment and analyze by CID due to the mass of the additional PhIAT labels [34]. The hydrophobic biotin moiety of both the ICAT and PhIAT labels can influence the retention behavior of the peptides during reversed-phase chromatographic separation and produce a relatively narrow elution zone of all tagged peptides [35].
To address these issues, the Goshe and Smith laboratories have reported an improved solid-phase based version of PhIAT, termed phosphoprotein isotope-coded solid-phase tag (PhIST) [36]. The tagging strategy is similar to the PhIAT approach: ß-elimination of the phosphate moiety followed by the Michael addition of 1,2-ethanedithiol. However, the biotin affinity tag is replaced by an isotope-coded solid-phase reagent containing either light (12C6, 14N) or heavy (13C6, 15N) stable isotopes and a photocleavable linker that is used to capture and label the phosphopeptides in a single step. The captured peptides are released from the solid-phase support by UV photocleavage and analyzed by LC/MS/MS. The efficiency and sensitivity of the PhIST labeling approach for identification of phosphopeptides from mixtures were determined using a mixture of casein proteins, and the method was applied to the quantification of soluble phosphoproteins from a human breast cancer cell line.
Chemical tagging approaches for labeling pSer and pThr phosphorylation sites using the ß-elimination/Michael addition approaches can be problematic for several reasons. Depending on the reaction conditions, the labeling of some protein samples can be sub-stoichiometric and difficulties associated with the low solubility of the EDT compound in aqueous solutions can promote protein precipitation [15]. Recently, Goshe and coworkers have addressed these issues by improving their PhIST approach to yield effective quantitative blocking of cysteinyl residues while promoting quantitative labeling of both pSer and pThr residues at the peptide level using Ba2+-catalyzed ß-elimination and Michael addition of (R,R)-dithiothreitol as the thiolate linker [37]. In addition, the improved PhIST labeling method using [12C6, 14N]leucine and [13C6, 15N]leucine isotope-coded solid-phase reagents was applied to an in vitro model of Parkinson's disease [38]. The use of the ß-elimination/Michael addition chemistry with cysteamine has been applied by Knight et al. [39] to convert pSer and pThr residues into lysine analogs (aminoethylcysteine and beta-methylaminoethylcysteine, respectively), which are cleaved with a lysine-specific protease to map sites of phosphorylation. As mentioned by nearly all the authors reporting the use of ß-elimination/Michael addition for pSer and pThr analysis, the labeling techniques could be applied to identify and quantify O-linked glycoproteins, reveling the need to use glycosidases to remove the O-linked modifications prior to specific labeling of pSer and pThr sites.
|
OTHER APPROACHES INVOLVING STABLE ISOTOPE LABELING |
|---|
In addition to the IMAC and chemical tagging methods already discussed, there are many other stable isotope labeling methods that can be used to characterize protein phosphorylation [13]. Isotope coding of proteins at the translational level by culturing cells on 14N-minimal media or 15N-enriched media has been used to study in vivo phosphorylation [40]. In this method, an equivalent number of cells for each sample is combined and processed for LC/MS analysis. The relative quantification of the phosphorylation state of proteins is determined by measuring the intensity ratio (14N/15N) of the detected unphosphorylated and phosphorylated peptides. In an analogous method, stable isotope labeling by amino acids in cell culture (SILAC) [41] implements in vivo labeling of proteins using stable isotope-coded amino acids and can be applied to protein phosphorylation studies. For example, d0/d3-serine coding and immunoprecipitation have been used to identify protein phosphorylation in a histone protein (H2A.X) from human skin fibroblast cells in response to low dose radiation [42]. Although the in vivo labeling techniques using cultured cells are quite useful to study phosphorylation events, they cannot be directly applied to serum and tissue samples that require alternative labeling techniques.
Enzymatic digestion of proteins can be used to isotopically code proteome samples by performing the proteolysis in the presence of ‘normal’ water (predominately as the H216O isotope) or ‘heavy’ water (enriched in H218O). This encoding strategy labels every C-terminus produced during proteolytic digestion, after which the samples are combined and analyzed by LC/MS. When samples from each experiment are compared, an inverse labeling pattern reflecting a characteristic mass shift is observed between the two parallel analyzes for proteins that are differentially expressed. Since it can be readily applied, enzymatic 16O/18O-labeling has been used for a wide variety of applications from studying serotypes Ad2 and Ad5 of adenovirus [43] to characterizing the changes in the human plasma proteome upon administration of lipopolysaccharides [44]. To promote more accurate measurements, quantitative 18O-labeling has been achieved by using immobilized trypsin after in-solution digestion in H218O to promote exchange of both oxygen atoms in the C-termini of peptides [45]. This is important since incomplete incorporation of 18O would limit and complicate the peptide mass shift between the two samples and lead to significant overlap of the isotopic distribution for the light and heavy forms [14].
To facilitate more confident identification of phosphorylation events inverse stable isotope coding of two distinct samples can be performed in a manner analogous to the dye labeling techniques used in transcriptome analysis. In this ‘inverse labeling’ approach two converse isotope coding experiments are conducted in parallel (lightlabel-control x heavylabel-experimental and heavylabel-control x lightlabel-experimental). When samples from each experiment are compared, an inverse labeling pattern reflecting a characteristic mass shift is observed between the two parallel analyzes for proteins that are differentially expressed. This strategy provides more effective screening of peptides that dramatically change (3-fold or greater) since ambiguities associated with extreme changes in protein abundance are eliminated. It has been employed with both 16O/18O-enzymatic [46] and 14N/15N-metabolic labeling [47] and can be used with other isotope labeling strategies described in this review. Since the presence of other peptides makes analysis of only the phosphopeptides more difficult, these proteome-wide labeling techniques coupled with IMAC and/or chemical tagging and multi-dimensional LC separations prior to MS analysis will provide substantial improvements in phosphoproteomic analysis and will assist in the identification of critical phosphorylation events involved in complex biological processes.
|
ABSOLUTE QUANTIFICATION OF PHOSPHOPROTEINS USING INTERNAL STABLE ISOTOPE-CODED STANDARDS |
|---|
The stable isotope coding techniques described thus far in this review can be used to determine the relative concentration of many phosphoproteins in one sample versus another by measuring the abundance of isotopically labeled proteolytic peptides. Although these stable isotope coding techniques are useful in characterizing changes that are occurring from one sample to another, they do not provide information regarding the absolute peptide abundance levels contained in each sample due to the lack of true internal standards. The absolute concentration of a protein in the cellular pool is an important quantitative measurement for many biological investigations. It can be used to determine the specific activity of a kinase in different tissues and to further validate the identification of candidate genes as disease markers during the study of expression profiles at the mRNA level.
With the ionization of peptide mixtures, the intensity of a peptide ion signal is not necessarily an accurate representation of the amount of a peptide in a sample due to a number of variables and thus requires a method of normalization. According to stable isotope dilution theory, two peptides with identical chemical structure that differ in isotopic composition are thought to generate the same response detected by the mass spectrometer such that their relative ion signals reflect their relative concentrations in the sample. A method of isotope dilution using synthetic isotope-coded peptides as internal standards for absolute quantification of proteins by MS has been developed and is referred to as AQUA (absolute quantification) [48]. In the AQUA approach, peptides identical to their native counterparts formed by proteolysis of the protein of interest are synthesized with stable isotope-coded amino acids that provide a detectable mass shift. Prior to quantitative measurements, the synthetic peptides are evaluated by LC/MS/MS to obtain qualitative information regarding peptide retention time during reversed-phase LC, ionization efficiency and fragmentation propensity during CID in order to eliminate any errors resulting from peptide mass degeneracy. The basic strategy involves the introduction of a known quantity of the AQUA peptide standard to the sample at the stage of proteolytic digestion. Peptides are separated by reversed-phase LC and analyzed by MS using selected reaction monitoring to selectively detect the peptides of interest and increase the sensitivity for low abundance measurements. Since a known quantity of the synthetic peptide is added, the measured ratio of the synthetic to endogenous peptide intensity can be used to determine the absolute amount of the peptide present in the sample, which is reflective of absolute protein abundance. The AQUA internal standards can also be synthesized with covalent modifications to probe post-translational modifications, such as adding phosphate groups to peptides in order to measure in vivo phosphorylation events. The AQUA strategy has been used to quantify two low abundance yeast proteins involved in gene silencing, quantify cell-cycle dependent phosphorylation of Ser-1126 of human separase protein, and identify the kinase that phosphorylates at Ser-1501 of separase [48]. Undoubtedly, the AQUA approach to study protein phosphorylation will be used for specific hypothesis-driven studies.
|
CONCLUSIONS |
|---|
The characterization of protein phosphorylation using MS is a multifaceted undertaking, a reflection of the biological ramifications of how phosphorylation is used to modulate protein function. The application of chemical, metabolic or enzymatic incorporation of stable isotopes into phosphopeptides or phosphoproteins provides a platform to quantify both relative and absolute abundance changes of phosphorylation events using MS, and in some instances, provides a means for differential phosphorylation site mapping. With the onset of more advanced separation and MS innovations, comprehensive biological investigations into the phosphoprotein-mediated processes involving both proteome-wide and targeted approaches for hypothesis-driven investigations may be realized. Regardless of the method chosen to measure and quantify protein phosphorylation, each requires the use of phosphatase, kinase and protease inhibitors to preclude measuring artifacts of sample preparation and to ensure that true biological phosphorylation events are being elucidated. The ability to quantify changes in protein phosphorylation, especially for those events occurring at low stoichiometries, will continue to be an analytical challenge for proteomics and systems biology for many years to come.
| Key Points
|
|
Acknowledgements |
|---|
I would like to thank the research agencies of North Carolina State University and the North Carolina Agricultural Research Service for continued support of my biochemistry research program and MS laboratory. Portions of this work were supported by grants from the National Science Foundation (MCB-0419819) and the United States Department of Agriculture (NRI 2004-35304-14930 and NRI 2005-35604-15420).
|
FOOTNOTES |
|---|
Mike Goshe is an assistant professor of Molecular and Structural Biochemistry and is the co-director of the Advanced Biomolecular Interaction Resource at North Carolina State University. His current research interests involve the development and application of mass spectrometry techniques for mapping and quantifying protein phosphorylation sites to fundamental studies in cell signalling and viral infection and characterizing protein–protein and protein–nucleic acid interactions.
|
References |
|---|
- Cohen P. Signal integration at the level of protein kinases, protein phosphatases and their substrates. Trends Biochem Sci 1992; 17:408–13.[CrossRef][Web of Science][Medline]
- Cohen P. The regulation of protein function by multisite phosphorylation – a 25 year update. Trends Biochem Sci 2000; 25:596–601.[CrossRef][Web of Science][Medline]
- Aebersold R, Mann M. Mass spectrometry-based proteomics. Nature 2003; 422:198–207.[CrossRef][Medline]
- Wang X, Goshe MB, Soderblom EJ, et al. Identification and functional analysis of in vivo phosphorylation sites of the arabidopsis brassinosteroid-insensitive1 receptor kinase. Plant Cell 2003; 17:1685–703.
- DeGnore JP, Qin J. Fragmentation of phosphopeptides in an ion trap mass spectrometer. J Am Soc Mass Spectr 1998; 9:1175–88.[CrossRef][Web of Science][Medline]
- McLafferty FW, Horn DM, Breuker K, et al. Electron capture dissociation of gaseous multiply charged ions by fourier-transform ion cyclotron resonance. J Am Soc Mass Spectrom 2001; 12:245–9.[CrossRef][Web of Science][Medline]
- Schroeder MJ, Shabanowitz J, Schwartz JC, et al. A neutral loss activation method for improved phosphopeptide sequence analysis by quadrupole ion trap mass spectrometry. Anal Chem 2004; 76:3590–8.[Medline]
- Meng F, Forbes AJ, Miller LM, et al. Detection and localization of protein modifications by high resolution tandem mass spectrometry. Mass Spectrom Rev 2005; 24:126–34.[CrossRef][Medline]
- Patton WF. Proteome analysis. Ii. Protein subcellular redistribution: linking physiology to genomics via the proteome and separation technologies involved. J Chromatogr B Biomed Sci Appl 1999; 722:203–23.[CrossRef][Medline]
- Yan JX, Sanchez JC, Binz PA, et al. Method for identification and quantitative analysis of protein lysine methylation using matrix-assisted laser desorption/ionization–time-of-flight mass spectrometry and amino acid analysis. Electrophoresis 1999; 20:749–54.[Medline]
- Newsholme SJ, Maleeff BF, Steiner S, et al. Two-dimensional electrophoresis of liver proteins: characterization of a drug-induced hepatomegaly in rats. Electrophoresis 2000; 21:2122–8.[CrossRef][Web of Science][Medline]
- Lill J. Proteomic tools for quantitation by mass spectrometry. Mass Spectrom Rev 2003; 22:182–94.[CrossRef][Medline]
- Goshe MB, Smith RD. Stable isotope-coded proteomic mass spectrometry. Curr Opin Biotechnol 2003; 14:101–9.[CrossRef][Web of Science][Medline]
- Leitner A, Lindner W. Current chemical tagging strategies for proteome analysis by mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci 2004; 813:1–26.[Medline]
- Mann M, Ong SE, Gronborg M, et al. Analysis of protein phosphorylation using mass spectrometry: deciphering the phosphoproteome. Trends Biotechnol 2002; 20:261–8.[CrossRef][Web of Science][Medline]
- Conrads TP, Issaq HJ, Veenstra TD. New tools for quantitative phosphoproteome analysis. Biochem Biophys Res Commun 2002; 290:885–90.[CrossRef][Medline]
- Loughrey Chen S, Huddleston MJ, Shou W, et al. Mass spectrometry-based methods for phosphorylation site mapping of hyperphosphorylated proteins applied to Net1, a regulator of exit from mitosis in yeast. Mol Cell Proteomics 2002; 1:186–96.
[Abstract/Free Full Text] - Gygi SP, Rist B, Gerber SA, et al. Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat Biotechnol 1999; 17:994–9.[CrossRef][Web of Science][Medline]
- Ficarro SB, McCleland ML, Stukenberg PT, et al. Phosphoproteome analysis by mass spectrometry and its application to saccharomyces cerevisiae. Nat Biotechnol 2002; 20:301–5.[CrossRef][Web of Science][Medline]
- Ficarro S, Chertihin O, Westbrook VA, et al. Phosphoproteome analysis of capacitated human sperm. Evidence of tyrosine phosphorylation of a kinase-anchoring protein 3 and valosin-containing protein/p97 during capacitation. J Biol Chem 2003; 278:11579–89.
[Abstract/Free Full Text] - He T, Alving K, Feild B, et al. Quantitation of phosphopeptides using affinity chromatography and stable isotope labeling. J Am Soc Mass Spec 2004; 15:363–73.[CrossRef][Medline]
- Ren D, Penner NA, Slentz BE, et al. Evaluating immobilized metal affinity chromatography for the selection of histidine-containing peptides in comparative proteomics. J Proteome Res 2003; 2:321–9.[Medline]
- Salomon AR, Ficarro SB, Brill LM, et al. Profiling of tyrosine phosphorylation pathways in human cells using mass spectrometry. Proc Natl Acad Sci USA 2003; 100:443–8.
[Abstract/Free Full Text] - Brill LM, Salomon AR, Ficarro SB, et al. Robust phosphoproteomic profiling of tyrosine phosphorylation sites from human T cells using immobilized metal affinity chromatography and tandem mass spectrometry. Anal Chem 2004; 76:2763–72.[Medline]
- Bieber AL, Tubbs KA, Nelson RW. Metal ligand affinity pipettes and bioreactive alkaline phosphatase probes: tools for characterization of phosphorylated proteins and peptides. Mol Cell Proteomics 2004; 3:266–72.
[Abstract/Free Full Text] - Ficarro SB, Salomon AR, Brill LM, et al. Automated immobilized metal affinity chromatography/nano-liquid chromatography/electrospray ionization mass spectrometry platform for profiling protein phosphorylation sites. Rapid Commun Mass Spec 2005; 19:57–71.[CrossRef][Web of Science][Medline]
- Oda Y, Nagasu T, Chait BT. Enrichment analysis of phosphorylated proteins as a tool for probing the phosphoproteome. Nat Biotechnol 2001; 19:379–82.[CrossRef][Web of Science][Medline]
- McLachlin DT, Chait BT. Improved beta-elimination-based affinity purification strategy for enrichment of phosphopeptides. Anal Chem 2003; 75:6826–36.[Medline]
- Zhou H, Watts JD, Aebersold R. A systematic approach to the analysis of protein phosphorylation. Nat Biotechnol 2001; 19:375–8.[CrossRef][Web of Science][Medline]
- Goshe MB, Conrads TP, Panisko EA, et al. Phosphoprotein isotope-coded affinity tag approach for isolating and quantitating phosphopeptides in proteome-wide analyses. Anal Chem 2001; 73:2578–86.[Medline]
- Goshe MB, Veenstra TD, Panisko EA, et al. Phosphoprotein isotope-coded affinity tags: application to the enrichment and identification of low-abundance phosphoproteins. Anal Chem 2002; 74:607–16.[Medline]
- Moseley MA. Current trends in differential expression proteomics: isotopically coded tags. Trends Biotechnol 2001; 19:S10–6.[CrossRef][Medline]
- Zhang R, Sioma CS, Wang S, et al. Fractionation of isotopically labeled peptides in quantitative proteomics. Anal Chem 2001; 73:5142–9.[Medline]
- Borisov OV, Goshe MB, Conrads TP, et al. Low-energy collision-induced dissociation fragmentation analysis of cysteinyl-modified peptides. Anal Chem 2002; 74:2284–92.[Medline]
- Li J, Steen H, Gygi SP. Protein profiling with cleavable isotope-coded affinity tag (cICAT) reagents: the yeast salinity stress response. Mol Cell Proteomics 2003; 2:1198–204.
[Abstract/Free Full Text] - Qian WJ, Goshe MB, Camp DG, et al. Phosphoprotein isotope-coded solid-phase tag approach for enrichment and quantitative analysis of phosphopeptides from complex mixtures. Anal Chem 2003; 75:5441–50.[Medline]
- Soderblom EJ, Cawthon D, Duhart H, Xu ZA, et al. An improved labeling method using phosphoprotein isotope-coded solid-phase tags for neuronal cell applications. Int J Neuroprotec Neuroregen 2005; 1:91–7.
- Cawthon D, Soderblom EJ, Xu ZA, et al. Quantitative analysis of phosphoproteins in a parkinson's disease model using phosphoprotein isotope-coded solid-phase tags. Int J Neuroprotec Neuroregen 2005; 1:98–106.
- Knight ZA, Schilling B, Row RH, et al. Phosphospecific proteolysis for mapping sites of protein phosphorylation. Nat Biotechnol 2003; 21:1047–54.[CrossRef][Web of Science][Medline]
- Oda Y, Huang K, Cross FR, et al. Accurate quantitation of protein expression and site-specific phosphorylation. P Natl Acad Sci USA 1999; 96:6591–6.
[Abstract/Free Full Text] - Ong SE, Blagoev B, Kratchmarova I, et al. Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol Cell Proteomics 2002; 1:376–86.
[Abstract/Free Full Text] - Zhu H, Hunter TC, Pan S, et al. Residue-specific mass signatures for the efficient detection of protein modifications by mass spectrometry. Anal Chem 2002; 74:1687–94.[Medline]
- Yao X, Freas A, Ramirez J, et al. Proteolytic 18o labeling for comparative proteomics: model studies with two serotypes of adenovirus. Anal Chem 2001; 73:2836–42.[Medline]
- Qian WJ, Monroe ME, Liu T, et al. Quantitative proteome analysis of human plasma following in vivo lipopolysaccharide administration using 16O/18O labeling and the accurate mass and time tag approach. Mol Cell Proteomics 2005; 4:700–9.
[Abstract/Free Full Text] - Brown KJ, Fenselau C. Investigation of doxorubicin resistance in mcf-7 breast cancer cells using shot-gun comparative proteomics with proteolytic 18O labeling. J Proteome Res 2004; 3:455–62.[CrossRef][Medline]
- Wang YK, Ma Z, Quinn DF, et al. Inverse 18O labeling mass spectrometry for the rapid identification of marker/target proteins. Anal Chem 2001; 73:3742–50.[Medline]
- Wang YK, Ma Z, Quinn DF, et al. Inverse 15N-metabolic labeling/mass spectrometry for comparative proteomics and rapid identification of protein markers/targets. Rapid Commun Mass Spec 2002; 16:1389–97.[Medline]
- Gerber SA, Rush J, Stemman O, et al. Absolute quantification of proteins and phosphoproteins from cell lysates by tandem MS. Proc Natl Acad Sci USA 2003; 100:6940–5.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Blackburn and M. B. Goshe Challenges and strategies for targeted phosphorylation site identification and quantification using mass spectrometry analysis Briefings in Functional Genomics, March 1, 2009; 8(2): 90 - 103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. F. Wu and D. S. Chu Sperm Chromatin: Fertile Grounds for Proteomic Discovery of Clinical Tools Mol. Cell. Proteomics, October 1, 2008; 7(10): 1876 - 1886. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Yu, N. Ge, M. Xie, W. Sun, S. Burlingame, A. K. Pass, J. G. Nuchtern, D. Zhang, S. Fu, M. D. Schneider, et al. Phosphorylation of Thr-178 and Thr-184 in the TAK1 T-loop Is Required for Interleukin (IL)-1-mediated Optimal NF{kappa}B and AP-1 Activation as Well as IL-6 Gene Expression J. Biol. Chem., September 5, 2008; 283(36): 24497 - 24505. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B.-J. M. Webb-Robertson and W. R. Cannon Current trends in computational inference from mass spectrometry-based proteomics Brief Bioinform, September 1, 2007; 8(5): 304 - 317. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. J. Grant and C. C. Wu Advances in neuromembrane proteomics: efforts towards a comprehensive analysis of membrane proteins in the brain Briefings in Functional Genomics, March 12, 2007; (2007) elm001v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||