Biochimica et Biophysica Acta 1764 (2006) 1788 – 1800 www.elsevier.com/locate/bbapap
Review
Analysis of posttranslational modifications exemplified using protein kinase A Frank Gesellchen, Oliver Bertinetti, Friedrich W. Herberg ⁎ Universität Kassel, FB 18 Naturwissenschaften, Abt. Biochemie, Heinrich-Plett-Str. 40, 34132 Kassel, Germany Received 14 June 2006; received in revised form 18 September 2006; accepted 5 October 2006 Available online 7 October 2006
Abstract With the completion of the major genome projects, one focus in biomedical research has shifted from the analysis of the rather static genome to the highly dynamic proteome. The sequencing of whole genomes did not lead to much anticipated insights into disease mechanisms; however, it paved the way for proteomics by providing the databases for protein identification by peptide mass fingerprints. The relative protein distribution within a cell or tissue is subject to change upon external and internal stimuli. Signal transduction events extend beyond a simple change in protein levels; rather they are governed by posttranslational modifications (PTMs), which provide a quick and efficient way to modulate cellular signals. Because most PTMs change the mass of a protein, they are amenable to analysis by mass spectrometry. Their investigation adds a level of functionality to proteomics, which can be expected to greatly aid in the understanding of the complex cellular machinery involved in signal transduction, metabolism, differentiation or in disease. This review provides an overview on posttranslational modifications exemplified on the model system cAMP-dependent protein kinase. Strategies for detection of selected PTMs are described and discussed in the context of protein kinase function. © 2006 Elsevier B.V. All rights reserved. Keywords: cAMP-dependent protein kinase; Phosphorylation; Myristoylation; Deamidation; Mass spectrometry
1. Introduction More than a decade has passed since the term “proteome” was first introduced into the scientific community by Marc Wilkins as the “total protein complement of a genome” [1,2]. A major hallmark – and a major challenge – of proteome analysis, as opposed to genome projects, is the dynamics underlying the protein composition of a cell, a tissue or an organism at a given time point under given external and internal conditions. By comparing two defined states of a sample (i.e. before and after treatment or healthy and diseased condition) one should be able to identify those proteins which differ between these states. This differential proteomics approach can provide an insight into the mechanism underlying the observed phenotypes, adding a level of functionality to proteomic analyses that is often missing from the merely descriptive genome projects. This analysis should not be limited to the up- or downregulation of gene products, ⁎ Corresponding author. Fax: +49 561 8044466. E-mail address:
[email protected] (F.W. Herberg). 1570-9639/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2006.10.001
but should also include the analysis of post-translational modifications (PTMs) in a proteome, since these covalent protein modifications can alter protein function and interactions significantly, as outlined below. The basic methodology of proteome analysis has remained largely unchanged in the last decade: due to the immense complexity of the proteome of even a single cell, the first step of proteome analysis is usually a separation step. In most cases this is achieved via two-dimensional gel electrophoresis [3,4], which still provides the highest resolution for separation of complex protein samples. The method has been continually refined, thus overcoming limitations of the earlier pioneering techniques and – most importantly – improving reproducibility [5,6]. Additionally, the use of fluorescent dyes can ameliorate the lack in dynamic range and linearity of common gel stains [7]. The use of two different fluorophores (e.g. Cy3 and Cy5) for the labelling of two samples allows for the comparison of these samples in a single gel, a technique known as difference gel electrophoresis (DIGE) [8]. Differences in protein levels are then reflected in the corresponding
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
spot colour and intensity, depending on which protein is more prominent. Identification of the proteins in question is then performed by mass spectrometry, usually by a tryptic digest and subsequent analysis of the resulting peptides via the peptide mass fingerprint (PMF). Based on MS/MS spectra of the peptides, sequence information can be obtained and – possibly – posttranslational modifications can be identified. As their name implies, they are covalent modifications introduced after – in some cases also during – protein biosynthesis. PTMs can range from the attachment of a methyl or phosphate group up to the addition of fatty acids, sugar moieties or – in the case of ubiquitination – even whole proteins. More than 200 different post-translational modifications of proteins have been described to date (for review see [9]). A common feature of these very different modifications is that they have an impact on protein function, for example on protein activity, localisation, protein interactions or on protein turnover. A proteomic study that aims to go beyond the quantitative description of the protein content of a sample has to include the analysis of post-translational modifications in order to obtain an idea on the investigated protein's function. This kind of analysis must not be restricted to a qualitative analysis of PTMs but should include some method for quantification, since some PTMs occur quantitatively, as stable modifications, while others affect only a subpopulation of an enzyme, or are only transient, still generating a massive cellular response. This is certainly true for the most important PTM in signal transduction, protein phosphorylation, which is mediated by the action of protein kinases. In cellular signal transduction [10,11] a small change in the phosphorylation status within a kinase cascade has a huge impact on the resulting cellular response. This can be attributed to conformational control exerted by phosphorylation that can function as an on–off switch (see Section 2.1 and Fig. 2) or change the binding behaviour of the modified protein. 1.1. The model system cAMP-dependent protein kinase (PKA) Several articles in this issue will focus on the most prominent PTMs. In the present article we would like to illustrate some of the strategies that can be applied for PTM analysis on a single protein that is considered paradigmatic for a whole class of enzymes: the cyclic AMP dependent protein kinase (PKA). PKA is a key enzyme in the modulation of intracellular processes in eukaryotes and is also implicated in several human diseases [12–14]. In the absence of the second messenger cAMP, this kinase is kept in an inactive state by a regulatory (R) subunit dimer, which binds and inhibits two catalytic (C) subunit monomers. This heterotetrameric holoenzyme is activated by cAMP binding to the R-subunits, which then release the active catalytic subunits [15,16]. Thus, the C-subunit can phosphorylate target proteins in the cytosol or the nucleus, thereby also influencing gene transcription. PKA- C-subunit has been extensively characterized since it was discovered in 1968 [17]. The original work on the protein had to be performed using preparations from mammalian tissues, due to the lack of
1789
recombinant techniques at that time, identifying two major isoforms termed type I and type II, according to their elution behaviour in ion exchange chromatography [18–20]. With the introduction of modern genetic methods, it became possible to overexpress this enzyme in several heterologous systems. However, when doing so, it became apparent that the PTM pattern of the PKA C-subunit was altered [21]. On the other hand, with the aid of recombinant techniques previously unknown isoforms of the catalytic, as well as the regulatory subunit were discovered. To date, four isoforms of the regulatory [22–24] and at least a dozen isoforms of the catalytic subunit [25–31] have been described in mammals, increasing complexity even further. The majority of cellular responses to cAMP seem to be attributable to the ubiquitously expressed Cα subunit, but distinct signals may be promoted by the less abundant isoforms. Using mass spectrometric analyses, it was shown much later, that enzyme preparations from porcine heart consist of a mixture of at least the catalytic subunits Cα and Cβ. Those are 93% identical on the amino acid level, still unique peptides from each subunit were identified using tandem mass spectrometry [32]. As the different isoforms cannot be separated by conventional biochemical methods, homogeneous preparations of C-subunit isoforms can only be obtained with the use of recombinant techniques. 2. Posttranslational modifications of the PKA C-subunit Several different post-translational modifications have been observed on this enzyme. Thus, PKA-C has been described to be multiply phosphorylated [33], N-terminally myristoylated (in fact it was the first protein shown to contain this PTM [34]) and, more recently, a deamidation has been described as well [32]. 2.1. Phosphorylation In eukaryotes, proteins can be phosphorylated on serine, threonine or tyrosine residues, allowing a distinction between Ser/Thr and Tyr kinases, according to their substrate specificity. Protein kinases do not only phosphorylate other proteins, thus altering their target's properties—they themselves are subject to phosphorylation, as a matter of fact most kinases need this PTM to be in an active state. In the case of the prototypic Cα subunit of PKA, belonging to the subfamily of Ser/Thr kinases, a threonine residue at position 197 in the so-called activation loop needs to be phosphorylated for the enzyme to assume an active conformation [21,35] (see Fig. 1). A second phosphorylation site at Ser 338 is obviously important for structural stability of the enzyme [35]. Both these phosphorylation sites are stable and remarkably resistant to phosphatase activity [36,37]. However, there is some evidence for a PP2A-like activity capable of dephosphorylating Thr 197 [38]. Mutant Cα-subunit lacking this phosphorylation site shows a dramatically decreased catalytic activity (500 fold [39,40]). While the Cα-subunit, like many protein kinases, is capable of autophosphorylation, there is also evidence for an upstream kinase performing this task in vivo [41,42]. Interestingly, even when expressed in E. coli, the major PKA C-subunits Cα and Cβ1 are capable of
1790
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
without having to rely on the energy consuming process of protein synthesis. 2.2. Myristoylation
Fig. 1. Posttranslational modifications of the PKA C-subunit. Ribbon view of the crystal structure of the myristoylated porcine PKA C-subunit (1CMK, [111]). Red: α-helices, blue: β-sheets, tan: myristic acid. Residues subject to post-translational modifications are highlighted in yellow (phosphorylation) or green (deamidation). Known and putative effects of PTMs are indicated. Visualisation of the crystal structure was performed using the Visual Molecular Dynamics software [112] (VMD, version 1.8.2, University of Illinois, UrbanaChampaign).
autophosphorylation and autoactivation, while the minor human PKA isoform PrKX is not [30]. Phosphorylation/dephosphorylation of the homologous Thr-residue in the activation loop is a general principle in the control of kinase activity [43,44] (see Figs. 1, 2), giving rise to the hierarchical phosphorylations in kinase signalling cascades (see respective articles in this issue). What makes phosphorylation so unique among PTMs? From a mechanistic view, it replaces a small polar residue (a hydroxyl group) by the bulkier, negatively charged phosphate. Obviously this constitutes a drastic change in the physicochemical properties of the affected seryl-, threonyl- or tyrosyl residue. Aside from the steric changes, it is now able to undergo electrostatic interactions and has changed from an H-bond donor to an H-bond acceptor. Thereby, in many cases, existing interaction networks (inter- and intramolecular) are disrupted or reconfigured and new interactions are formed. This is exemplified in the binding sites for SH2 domains generated by tyrosine phosphorylation [45], or the recognition of phosphoserine and -threonine by 14-3-3 proteins and other protein domains (reviewed in [46]). A very striking example of intramolecular rearrangements is the activation of the insulin receptor kinase. Upon binding of the ligand insulin this receptor tyrosine kinase dimerises and is activated by (auto)phosphorylation, resulting in an extended rearrangement of the active site (see Fig. 2). Another very prominent feature of protein phosphorylation is its dynamics. Protein phosphorylation is a reversible covalent modification, and is removed by phosphatases, thus efficiently terminating the effects elicited by the preceding phosphorylation. This is why phosphorylation has evolved to be the molecular switch in signalling networks, allowing cells to respond quickly and efficiently to changes in their environment,
When the complete amino acid sequence of the bovine PKA C-subunit was first determined, the N-terminus was found to be blocked by an unknown functional group [47], that was shortly afterwards identified as n-Tetradecanoyl [34], or myristic acid. Acylation of proteins in general can occur at several sites, either as myristoylation on an N-terminal glycine residue or in the polypeptide chain as S-acylation (palmitoylation) and prenylation (geranylgeranoylation or farnesylation) on cysteine residues. Some cell surface proteins have a glycolipid anchor (glycosylphosphatidylinostol, GPI) attached to their C-terminus. Protein acylation is usually associated with membrane targeting [48], as exemplified in the prenylation and Sacylation of small G-proteins [49–51] where protein localisation is closely related to protein function [52–54]. Recent evidence also indicates that myristoylation might be involved in the autoinhibition of protein kinases, for example in the case of the c-abl tyrosine kinase [55, 56]. In case of the PKA Csubunit, however, it is still unclear whether N-terminal myristoylation (see Fig. 1) is involved in membrane targeting. Although it has been shown that a myristoylated peptide,
Fig. 2. Structural rearrangement of a protein upon phosphorylation. Superposition of the kinase domains of insulin receptor tyrosine kinase in the inactive, nonphosphorylated (1IRK [113], blue) and the active, triply phosphorylated state (1IR3 [114], red). In the nonphosphorylated form the activation loop (cyan) blocks access to the ATP binding site of the kinase. Upon phosphorylation, the activation loop flips over (yellow) giving access to the ATP binding pocket (ATP molecule shown in green). The three (phospho)tyrosine residues in the activation loop are depicted in stick representation. Note also the downward movement of the entire upper lobe of the enzyme after phosphorylation, which is also critical to ATP-binding. Visualisation and alignment of the crystal structures was performed using the software YASARA View (http://www.yasara.org).
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
derived from the Cα N-terminus can mediate membrane association [57], and a chimeric protein replacing the 14 Nterminal residues of v-src by the corresponding (myristoylated) Cα sequence [58] localises to the plasma membrane, in vivo evidence for such an event is still missing. In contrast, in the case of PKA, anchoring generally occurs via a multitude of AKinase anchoring proteins (AKAPs), which tether the PKA holoenzyme via its R-subunits [59] to subcellular compartments. Still, some studies suggest an influence of Cα myristoylation in protein–protein or protein–membrane interactions [60,61]. Interestingly, phosphorylation of the C-subunit at Ser 10 seems to have a significant effect on the structure of the N-terminus: in the absence of a phosphorylation, it adopts a helical conformation, possibly extending the myristic acid into solution [60]. Taken together, this PTM warrants further investigation, since it becomes more and more obvious, that there is much more behind it than the structural stability that had been described as one of its first effects in earlier reports for the Cα subunit [62,63]. Acylation of proteins also highlights one of the problems that proteomics has to deal with: a subset of proteins that might simply be lost during sample preparation, especially since 2D gel electrophoresis is not well suited for the analysis of membrane proteins. Novel combinations of separation techniques, based on liquid chromatography, have been shown to be superior in this respect and can at least complement 2D-PAGE analysis [64,65]. This shotgun proteomics approach, termed MudPIT (multidimensional protein identification technology) has been shown to be very powerful when combined with protein fractionation to reduce sample complexity [66]. Reduction of sample complexity is necessary due to the relatively low sequence coverage obtained with complex protein mixtures, a situation that is generally undesirable for PTM analysis. 2.3. Deamidation A rather recently identified post-translational modification of the PKA C-subunit is the deamidation of the asparagine residue at position 2 to aspartic acid [32], although this modification had already been described some time ago as an isoelectric variant of the C-subunit [67]. Generally considered as an indication of protein aging [68], this modification turned out to be present in C-subunits prepared from bovine heart. In fact, deamidation was demonstrated to be responsible for a puzzling behaviour of C-subunits purified from mammalian tissue. It had been observed earlier, that in ion exchange chromatography, two fractions of electrophoretically homogeneous C-subunit were separated [69], termed CA and CB, however, the reason for the different migration behaviour of these otherwise identical enzyme fractions remained unclear. With the identification of the deamidated form of the C-subunit it became apparent that the charge difference from asparagine to the aspartic acid residue is responsible for this phenomenon. In microinjection experiments it was shown that deamidation has an effect on subcellular localisation, with the nonmodified form reaching higher nuclear concentrations than the deamidated enzyme [70].
1791
It is tempting to speculate that this would have an impact on nuclear signalling of PKA. 3. PTM analysis of PKA-Cα 3.1. General considerations One starting point in classical proteomics is the protein identification via peptide mass fingerprint (PMF). The detected mass, or rather the mass to charge ratio (m/z) of a peptide is compared to the masses of theoretical tryptic digests of all proteins of a defined database [71]. This search is made more powerful by including sequence information from MS/MS spectra, so-called peptide sequence tags [72]. In the MASCOT analysis software package [73], a probability based score is assigned to the MS/MS spectrum based on specific algorithms [74], which allow the user to judge the significance of the match. A different approach directly compares tandem mass spectra to theoretical mass spectra generated from a protein database to identify a protein [75]. This strategy, implemented for example in the SEQUEST software package, can also be used to detect modifications to a peptide [76]. Since most PTMs change the mass of a peptide, the user has to have some idea which PTMs he is expecting. The ExPASy proteomics server at the Swiss Institute for Bioinformatics [77] provides some helpful tools for the prediction of PTMs in proteins (http://www.expasy.org/tools/#ptm). Usually, common PTMs are already covered in MS analysis software packages, and can be assigned either as fixed or variable modifications. If other modifications are suspected to occur, they should be manually entered by the user or imported from databases, such as Unimod (http://www.unimod.org) or RESID (http://www. ebi.ac.uk/RESID/). It should be kept in mind, however, that this does not only increase data processing time, but can also increase the rate of false positive identifications, due to the higher number of possible peptide masses. For this reason, manual validation of MS/MS-spectra is still required for ambiguous matches. Obviously, PTM detection on a peptide level requires high sequence coverage. Failure to find a modification is not a proof of its absence! 3.2. Phosphorylation 3.2.1. Strategies for identification of phosphopeptides Two main strategies based on tandem MS are in use for the specific detection of phosphorylations: Ser or Thr-phosphorylated peptides can either be detected in a precursor ion scan, when looking at daughter ions with an m/z of 79 Da (PO3−) or in a neutral loss scan where they are detected as loss of 98 Da (H3PO4), for singly charged peptides. Compared to the neutral loss, the precursor ion scan has the advantage of being independent from the charge state of the phosphopeptide. Fig. 3 shows the identification of two PKA C-subunit phosphorylation sites by MS/MS on an ABI 4000 Q TRAP. The first phosphorylation, detected on a doubly charged parent ion, was unambiguously assigned to the seryl residue at
1792
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
position 3, corresponding to Ser 10 of the PKA Cα sequence (Fig. 3A). Only peptide fragments that contain this residue can show the characteristic loss of H3PO4, i.e. either upwards from the b3 ion or downwards from the y12-ion. If the second serine in this peptide (Ser 14) were the phosphorylated residue, the H3PO4 loss would not have been detectable on fragment ions b3 to b6 (or y12 to y9), as they do not contain this residue. The second phosphorylation site was confirmed to be Thr 197 (Fig.
3B); in this case the other putative phosphorylation sites (Thr 195, 201, Ser 212 as well as Tyr 204!) can be ruled out for analogous reasons. This observation is consistent with biochemical [21] and structural evidence [78,79], where Thr 197 was identified as the phosphorylation site critical for activation and for inhibitor binding conserved in most protein kinases. However, the corresponding peptide is not very well accessible to MS analysis, and the assignment of the correct
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
phosphoaminoacid residue requires careful inspection by the user (see Fig. 3). Furthermore, phosphorylation was detected on Ser 139 as well as Ser 338 (data not shown). These MS findings corroborate earlier findings based on microsequencing [21]. Interestingly, the peptides containing the stable P-sites on Ser 338 and Thr 197 are stoichiometrically phosphorylated, and it is known for a long time that the C-subunit is constitutively phosphorylated at these sites [33,47]. Nevertheless, those sites might not be readily detectable in mass spectrometry. This may be due to a less efficient ionization of phosphopeptides compared to unmodified peptides, although a recent study challenges this hypothesis [80]. In the case of PKA C however, it was shown that phosphorylation can interfere with tryptic cleavage [81]. This is most likely due to steric hindrance, because the PKA phosphorylation consensus site contains a basic residue at P-2 and/or P-3 and thus also a trypsin cleavage site. The ensuing miscleavage can therefore result in considerably larger peptides than expected. In the case of Ser338 we have only been able to detect the phosphorylation site on a peptide with a missed cleavage at the neighbouring Arg 336. This resulted in a phosphopeptide with an m/z of 2997 (containing two missed cleavages) instead of the expected 768. In order to circumvent this problem, an elastase digest in combination with neutral loss scans has been used successfully for characterization of PKA phosphorylation sites [81]. The usefulness of this approach for phosphopeptide mapping was recently shown in a study that used a combination of elastase, proteinase K and thermolysine together with phosphopeptide enrichment on a TiO2-column (see below). Using this strategy, Schlosser and co-workers were able to detect a total of 21 phosphorylation sites on the protein mPER2 (murine circadian protein period 2) compared to only six phosphopeptides in a tryptic digest [82]. Another interesting approach utilizes the selective chemical derivatisation of phosphoserine/-threonine residues to generate lysine mimics that can subsequently be cleaved using the endoproteinase Lys-C or trypsin [83,84]. These additional cleavage sites are very useful in phosphopeptide mapping, but inherently bear the risk of further increasing the complexity of the peptide mixture. In order to reduce background from unmodified peptides, it is advantageous to enrich the phosphopeptides in a separation step. Currently established technologies for phosphopeptide enrichment rely on immobilized metal affinity chromatography
1793
(IMAC) [85], using trivalent metal ions [86,87] or on titanium dioxide solid phase material (Titansphere [88]). Both techniques have been employed successfully for selective enrichment and subsequent MS detection of phosphopeptides down to the femtomolar level in the case of TiO2-columns [88,89]. For a more detailed discussion for phosphopeptide analysis see elsewhere in this issue. 3.2.2. Quantification As noted before, certain PTMs – most notably phosphorylation – can appear at low stoichiometries and still have a striking impact on physiology. One of the key challenges is therefore to determine the abundance of a phosphorylated peptide versus its unphosphorylated counterpart. These differences might easily go unnoticed in large scale analyses, especially if specific enrichment strategies for phosphopeptides are employed. Specific PTMs on the recombinant murine Cα have been investigated by several groups [32,81,90,91]. However, investigations dealing with absolute or even relative quantifications have been limited so far. Methodology for quantification includes stable isotope labelling of amino acids in cell culture (SILAC [92]), labelling of cysteine residues with isotope coded affinity tags (ICAT [93,94]), or labelling of primary amines with isobaric tags for relative and absolute quantification iTRAQ [95]. These approaches have in common, that two samples (or up to four in the case of iTRAQ) are differentially labelled and subsequently mixed prior to separation and MS/ MS analysis. This allows distinguishing and quantifying the differentially labelled peptides on the basis of the isotope or peptide label. While these methods are not specific for posttranslational modifications, they do not interfere with them, either. Therefore, they can be useful in quantitating posttranslationally modified peptides, especially when combined with specific enrichment strategies [96,97]. It can be useful to complement this method with the classical 2D-PAGE/MS analysis, since the latter also permits the visualisation of differently modified protein species and might correct for some of the bias for high MW proteins associated with LC/MS analysis [98]. Absolute quantification of a known post-translational modification can also be achieved by comparison with isotopically labelled peptide standards, which are spiked into the sample (during sample preparation) at known concentrations [99]. Triple quadrupole linear ion trap mass spectrometers are
Fig. 3. Identification of two autophosphorylation sites on the Cα-subunit of PKA. 5 μg of recombinantly expressed murine PKA Cα were subjected to a tryptic digest after one-dimensional SDS-PAGE according to standard protocols [115]. The peptide sample was separated on a nanoLC-Ultimate HPLC-system (LC Packings, Dionex, Amsterdam, The Netherlands) coupled online to a linear ion trap mass spectrometer (4000 QTRAP®, Applied Biosystems, Darmstadt, Germany). Sample was loaded onto an C18 pre-column (C18 PepMap 100, 300 μm ID × 5 mm, particle size 5 μm, Dionex, Idstein, Germany) in 0.1% formic acid for 6 min at a flow rate of 30 μL/min. Separation was performed on a C18 column (C18 PepMap100, 75 μm ID × 150 mm, particle size 3 μm, Dionex) at a flow rate of 200 nL/min using a binary gradient composed of solvent A (0.1% formic acid) and solvent B (0.1% formic acid in 84% acetonitrile) with a linear increase in solvent B from 5 to 50% in 35 min, and from 50 to 95% B in 5 min, followed by an isocratic run at 95% B for another 5 min, before re-equilibrating the column to 5% solvent B. (A) MS/MS spectrum of a doubly charged peptide of m/z 830.9 ([M + 2H]2+) showing the characteristic loss of H3PO4. Fragment ion b3 is the first ion to show the neutral loss, therefore phosphorylation can be assigned to the first serine residue (Ser10 of the PKA Cα sequence). Peptide sequence as well as expected bn and yn fragment ion masses are shown above the figure. Predicted and observed (mass tolerance: ±0.4 Da) fragment ion masses are printed in normal and italic fonts, respectively. Note that there are two missed cleavages in this peptide, with the first one (at the very N-Terminus) probably attributable to steric hindrance by the nearby phosphoserine [81]. (B) MS/MS spectrum of a doubly charged peptide of m/z 1107.9. Peptide sequence as indicated above the figure was confirmed by an overlapping series of b- and y-ions. Phosphorylation can unambiguously be assigned to Thr197, as only fragment ions from b3 upwards (but not b2) show a loss of H3PO4.
1794
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
especially suited for this kind of analysis, because they allow observing several defined fragmentation reactions quasi simultaneously in a process called multiple reaction monitoring (MRM). This involves operating the first quadrupole (Q1) in resolving mode to let pass ions of a defined m/z value, fragmenting them in Q2, and detecting a defined fragment ion in Q3. The peptide sequence and PTM locations can subsequently be confirmed in a linear ion trap MS/MS scan [100]. By predefining specific fragmentation patterns based on available biological information (i.e. primary sequence, known or predicted PTMs), it is possible to achieve a high selectivity even in complex samples. The method has been shown to be suitable for relative quantification of peptides, as the MRM signal scaled linearly over several orders of magnitude [100]. However, it is even more powerful with the inclusion of internal standard peptides, as shown in a study on the quantitative analysis on multi-site phosphorylation of cyclin dependent kinases [101]. 3.2.3. Top-down approaches A different strategy for determination of PTMs on proteins is the so-called top-down approach [102], where the molecular mass of the intact protein is determined via MS. Although this usually requires high resolution instrumentation (see below), valuable information can also be gained using instruments with lower mass accuracy, for example triple quadrupole mass spectrometers. In the case of PKA-C, this approach can be used to determine the phosphorylation status of the enzyme. In contrast to Cα subunit purified from mammalian tissue, recombinantly expressed protein can contain additional phosphorylations at Ser10 and Ser139 [91], that are attributable to autocatalytic activity of the C-subunit [35]. Since the phos-
phates are located on the enzyme surface (see Fig. 1), recombinant proteins can be quite easily separated by cation exchange chromatography [91]. In earlier experiments, less sophisticated instrumentation required this separation of the differently phosphorylated species before LC-ESI-MS analysis of the intact protein [91]. Recent advances in instrumentation allow this kind of analysis even with a mixture of the different phosphoforms of the same enzyme. Fig. 4 shows multiply charged species of PKA-Cα injected on an ABI 4000 Q TRAP instrument and the corresponding deconvoluted spectrum. The mass differences of ∼ 80 Da between the three major peaks indicate the existence of differentially phosphorylated species of the enzyme. Compared to the theoretical mass these peaks correspond to triply, quadruply and quintuply phosphorylated enzyme and possibly even a sixth phosphorylation site. A genuine top-down proteomics approach has to include the fragmentation of the analyte protein(s), in order to identify the protein and PTMs via mass fingerprints or MS/MS. The recent Fourier transform-ion cyclotron resonance (FT-ICR) instruments are ideally suited for this kind of analysis, because of their – as yet – unparalleled mass accuracy and resolving power ([103], for a recent review on instrumentation see also [104]). Very recently, the usefulness of the new “orbitrap” instruments for top-down proteomics has been established as well [105]. 3.3. The N-Terminus of some PKA C isoforms is subject to modification 3.3.1. Myristoylation The myristoylated N-Terminus of the PKA-Cα subunit can be detected via the peptide mass fingerprint and tandem MS of the N-terminal peptide. Compared to the theoretical mass of the
Fig. 4. Analysis of the phosphorylation pattern on an intact protein. The murine catalytic subunit Cα of the cAMP-dependent protein kinase was expressed in E. coli and purified via phosphocellulose chromatography as described elsewhere without a further separation into isoelectric variants [91]. 50 pmol of purified protein were subjected to liquid chromatography, as described in Fig. 3, and online MS analysis. The figure shows an Enhanced MS scan with multiply charged species of PKA Cα as indicated. The mass spectrum was deconvoluted (inset) using the Bayesian Protein Reconstruct® routine included in the Analyst® software package (Applied Biosystems). Resulting masses can be assigned to phosphoisoforms of the catalytic subunit (theoretical mass 40439 Da) with 3 to 6 phosphorylations (P) as indicated.
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
peptide sequence GNAAAAK (601.7 Da) the mass is increased by 210 Da for the myristoyl moiety attached to the glycine residue. This mass increase is also observed on the corresponding b-ions of the MS/MS spectrum (see Fig. 5). A recent study on protein acylation defined marker ions for the detection of lipid modifications, namely the a1 and b1-ions at m/z 240.4 and 268.4, respectively [106], both of which are detected well within the mass tolerance of the 4000 Q TRAP instrument. It should be noted that the myristoylation is only detected on some isoforms of the native PKA-C enzyme, but not on the same isoforms expressed in E. coli, since prokaryotes lack the enzyme N-myristoyltransferase (NMT) that catalyzes this posttranslational modification. When the PKA Cα-subunit is coexpressed in E. coli with the yeast NMT, however, Cα is correctly processed [62]. When using non-myristoylated enzyme for MS analysis, we were not able to detect the (nonmyristoylated) N-terminal peptide, suggesting that it is lost during sample preparation. Obviously, myristoylation changes the chromatographic behaviour of the peptide due to the significant increase in hydrophobicity. According to their primary sequence most PKA-C isoforms cannot be myristoylated, because they lack the N-terminal glycine residue. Generally, those isoforms are less abundant in most tissues compared to their myristoylated counterparts. 3.3.2. Deamidation Besides the myristoylation, another PTM has been observed at the N-Terminus. The amino acid at position 2 can be subject to modification by being deamidated from asparagine to aspartic acid. Although this PTM results in a mass difference of only 1 Da, it massively influences the chromatographic behaviour of the enzyme [67]. C-subunit preparations from porcine heart consist of a mixture of the myristoylated Cα and Cβ1 subunits, 30% of which were found to be deamidated at residue 2. A charged residue at position 2 usually prevents an N-terminal myristoylation, thus, the deamidation must occur posttranslationally after the myristoylation. Deamidation introduces a negative charge close to the myristic acid and might therefore interfere with a putative membrane association. As it is also speculated that phosphorylation of the neighbouring seryl residue at position 10 influences membrane binding [60], it is obvious that for a functional analysis these PTMs should be investigated in conjunction. Deamidation is not catalyzed by an enzymatic reaction but occurs at physiological pH via a βaspartyl shift mechanism, in which a succinimide ring structure is formed as an intermediate, that is subsequently hydrolyzed to yield either an isoaspartyl or a normal aspartyl residue [107]. Due to racemisation of the succinimide ring, the D-isomers of these amino acids can be formed as significant by-products. These non-natural amino acids can be “repaired” by the Protein L-Isoaspartyl/D-Aspartyl Methyltransferase (PIMT) [108]. It is regarded as a mechanism involved in protein aging and turnover, but in the case of the PKA C-subunit it has been shown to be of functional significance, influencing the intracellular distribution of the enzyme [70]. In a very interesting approach, Lehmann, Kinzel and co-workers have managed to detect intermediates of the deamidation process on
1795
synthetic peptides [109] as well as tryptic digests of bovine PKA C-subunits [110]. After an HPLC separation, the diastereomeric peptides were detected and distinguished from each other by MS/MS on the basis of their immonium ion pattern [109]. It should be noted here, that isoaspartate formation and racemisation do not lead to a change in molecular mass of a protein or peptide. 4. Conclusions The genomic projects of the past have provided the necessary background to establish databases for gene products of whole organisms, a prerequisite for proteomics research. Proteomics has to aim beyond the purely descriptive approach of listing which genes are differentially expressed under certain conditions. With the constantly developing proteomic techniques, we have the opportunity to assign functionality to these differences and gain valuable insights into the mechanistics of a cell or tissue. This is not only important for our general understanding of cellular events, but is also indispensable in uncovering disease mechanisms, biomarkers, and drug targets. While it is certainly useful to look at differences in gene expression—and thus protein content, it is of special importance to assess the post-translational modifications of proteins that change in response to certain stimuli. These changes might not be readily apparent on the classical 2D gel spot, either because they do not change migration behaviour of the protein or because of their low stoichiometry. Recent advances in chromatography and mass spectrometry permit the selective enrichment and subsequent analysis of the single most important PTM—phosphorylation. Other PTMs require different approaches, as demonstrated in this article on the prototypical catalytic subunit of PKA, and in greater detail in other articles in this issue. Quantification is an ever present issue in the field of proteomics in general, and is of even greater interest when dealing with PTMs. New developments using isotope labelling techniques allow for quantification of peptides in complex mixtures, but still require optimization procedures in many cases. While these analysis strategies rely on the tryptic digest of proteins, top-down proteomics starts from an intact protein yielding valuable information about possible PTMs. With the development of mass spectrometers of ever higher mass accuracy and resolving power this approach should gain more significance in the field. In the future, proteomics research, instead of just recording posttranslational modifications, has to unravel the associated physiological function. In a way, the situation in proteomics reflects the situation after finishing the first genome sequencing projects: a wealth of data only manageable by bioinformatics, but not necessarily providing meaningful biological information for basic or applied research. To further complicate the situation, many PTMs exist only transiently, are restricted to a certain cell type or to a subcellular compartment or are only affecting a subpopulation of cells of the same type. Proteomic data in combination with existing biological knowledge can
1796
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
Fig. 5. Detection of an N-terminally modified peptide. Tryptic peptides derived from the murine catalytic subunit (Cα) of PKA co-expressed in E. coli with yeast Nmyristoyltransferase [62] were chromatographed as described in the legend to Fig. 3 and subjected to MS/MS analysis. (A) The spectrum shows the complete y and b ion series of a singly charged peptide with m/z 812.5 at a retention time of 61.68 min indicating a high degree of hydrophobicity (see Fig. 3B). The myristoylation is conclusively shown by the characteristic fragment ion b1. Additionally, the a1 marker ion (bold print) at 240.6 indicates an N-terminal myristoylation [106]. The sequence of the peptide and the expected bn (N terminal fragment ions) and yn (C terminal fragment ions) masses are shown in the table. Predicted and observed (mass tolerance: ±0.4 Da) fragment ion masses are printed in normal and italic fonts, respectively. (B) Extracted Ion Chromatogram for m/z 812.2–812.7, showing the retention time for peptides of this particular m/z ratio.
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
serve as a basis for hypothesis driven research to focus on protein function in a cellular network. Those networks are based on non-covalent protein–protein interactions which – for a better understanding – need to be quantified in detail. Finally, in vivo confirmation in cells or animal systems, also with the help of genetic tools, is needed to generate a more complete picture of a functional entity, i.e. an organelle, a cell, a tissue or an organism, ultimately integrating information concerning the entire system of investigation. Acknowledgements The authors would like to thank Christof Lenz (Applied Biosystems) for expert help with MS experiments. Anke Prinz and Silke Hutschenreiter are acknowledged for critical reading of the manuscript. This work was financially supported by the German ministry for education and research (BMBF grant 2 01 GR 0441, NGFN2) and by the European Union (RICA 026008). References [1] V.C. Wasinger, S.J. Cordwell, A. Cerpa-Poljak, J.X. Yan, A.A. Gooley, M.R. Wilkins, M.W. Duncan, R. Harris, K.L. Williams, I. HumpherySmith, Progress with gene-product mapping of the mollicutes: mycoplasma genitalium, Electrophoresis 16 (1995) 1090–1094. [2] M.R. Wilkins, J.C. Sanchez, A.A. Gooley, R.D. Appel, I. HumpherySmith, D.F. Hochstrasser, K.L. Williams, Progress with proteome projects: why all proteins expressed by a genome should be identified and how to do it, Biotechnol. Genet. Eng. Rev. 13 (1996) 19–50. [3] J. Klose, Protein mapping by combined isoelectric focusing and electrophoresis of mouse tissues. A novel approach to testing for induced point mutations in mammals, Humangenetik 26 (1975) 231–243. [4] P.H. O'Farrell, High resolution two-dimensional electrophoresis of proteins, J. Biol. Chem. 250 (1975) 4007–4021. [5] J. Klose, U. Kobalz, Two-dimensional electrophoresis of proteins: an updated protocol and implications for a functional analysis of the genome, Electrophoresis 16 (1995) 1034–1059. [6] A. Gorg, W. Weiss, M.J. Dunn, Current two-dimensional electrophoresis technology for proteomics, Proteomics 4 (2004) 3665–3685. [7] W.F. Patton, A thousand points of light: the application of fluorescence detection technologies to two-dimensional gel electrophoresis and proteomics, Electrophoresis 21 (2000) 1123–1144. [8] M. Ünlü, M.E. Morgan, J.S. Minden, Difference gel electrophoresis. A single gel method for detecting changes in protein extracts, Electrophoresis 18 (1997) 2071–2077. [9] R.G. Krishna, F. Wold, Posttranslational modifications, in: R.H. Angeletti (Ed.), Proteins: Analysis and Design, Academic Press, San Diego, CA, 1998, pp. 121–207. [10] E.G. Krebs, Nobel lecture. Protein phosphorylation and cellular regulation i, Biosci. Rep. 13 (1993) 127–142. [11] R. Seger, E. Krebs, The mapk signaling cascade, FASEB J. 9 (1995) 726–735. [12] Y.S. Cho-Chung, S. Pepe, T. Clair, A. Budillon, M. Nesterova, Campdependent protein kinase: role in normal and malignant growth, Crit. Rev. Oncol./Hematol. 21 (1995) 33–61. [13] E.M. Aandahl, P. Aukrust, B.S. Skalhegg, F. Muller, S.S. Froland, V. Hansson, K. Tasken, Protein kinase a type i antagonist restores immune responses of t cells from HIV-infected patients, FASEB J. 12 (1998) 855–862. [14] G.M. Kammer, High prevalence of t cell type i protein kinase a deficiency in systemic lupus erythematosus, Arthritis Rheum. 42 (1999) 1458–1465.
1797
[15] S.S. Taylor, J. Yang, J. Wu, N.M. Haste, E. Radzio-, G. Anand, Pka: a portrait of protein kinase dynamics, Biochim. Biophys. Acta-Proteins Proteomics 1697 (2004) 259–269. [16] C. Hahnefeld, D. Moll, M. Goette, F.W. Herberg, Rearrangements in a hydrophobic core region mediate camp action in the regulatory subunit of pka, Biol. Chem. 386 (2005) 623–631. [17] D.A. Walsh, J.P. Perkins, E.G. Krebs, An adenosine 3′,5′-monophosphate-dependant protein kinase from rabbit skeletal muscle, J. Biol. Chem. 243 (1968) 3763–3765. [18] E.M. Reimann, C.O. Brostrom, J.D. Corbin, C.A. King, E.G. Krebs, Separation of regulatory and catalytic subunits of the cyclic 3′,5′adenosine monophosphate-dependent protein kinase(s) of rabbit skeletal muscle, Biochem. Biophys. Res. Commun. 42 (1971) 187–194. [19] D. Kubler, M. Gagelmann, W. Pyerin, V. Kinzel, Isolation of the catalytic subunits of cyclic amp-dependent protein kinases from different mammalian tissues on the basis of charge differences of their subunits, Hoppe-Seyler Z. Physiol. Chem. 360 (1979) 1421–1431. [20] J.D. Corbin, S.L. Keely, C.R. Park, The distribution and dissociation of cyclic adenosine 3′:5′-monophosphate-dependent protein kinases in adipose, cardiac, and other tissues, J. Biol. Chem. 250 (1975) 218–225. [21] W. Yonemoto, S. Garrod, S. Bell, S. Taylor, Identification of phosphorylation sites in the recombinant catalytic subunit of campdependent protein kinase, J. Biol. Chem. 268 (1993) 18626–18632. [22] T. Jahnsen, L. Hedin, V. Kidd, W. Beattie, S. Lohmann, U. Walter, J. Durica, T. Schulz, E. Schiltz, M. Browner, Molecular cloning, cDNA structure, and regulation of the regulatory subunit of type ii campdependent protein kinase from rat ovarian granulosa cells (published erratum appears in j biol chem 1988 mar 15;263(8):4041), J. Biol. Chem. 261 (1986) 12352–12361. [23] C.H. Clegg, G.G. Cadd, G.S. McKnight, Genetic characterization of a brain-specific form of the type i regulatory subunit of camp-dependent protein kinase, Proc. Natl. Acad. Sci. U. S. A. 85 (1988) 3703–3707. [24] K. Takio, S.B. Smith, E.G. Krebs, K.A. Walsh, K. Titani, Amino acid sequence of the regulatory subunit of bovine type ii adenosine cyclic 3′,5′-phosphate dependent protein kinase, Biochemistry 23 (1984) 4200–4206. [25] M.O. Showers, R.A. Maurer, A cloned bovine cDNA encodes an alternate form of the catalytic subunit of camp-dependent protein kinase, J. Biol. Chem. 261 (1986) 16288–16291. [26] S.J. Beebe, O. Oyen, M. Sandberg, A. Froysa, V. Hansson, T. Jahnsen, Molecular cloning of a tissue-specific protein kinase (c gamma) from human testis-representing a third isoform for the catalytic subunit of camp-dependent protein kinase, Mol. Endocrinol. 4 (1990) 465–475. [27] N. Reinton, S. Orstavik, T.B. Haugen, T. Jahnsen, K. Tasken, B.S. Skalhegg, A novel isoform of human cyclic 3′,5′-adenosine monophosphate-dependent protein kinase, c alpha-s, localizes to sperm midpiece, Biol. Reprod. 63 (2000) 607–611. [28] S. Orstavik, N. Reinton, E. Frengen, B.T. Langeland, T. Jahnsen, B.S. Skalhegg, Identification of novel splice variants of the human catalytic subunit cbeta of camp-dependent protein kinase, Eur. J. Biochem. 268 (2001) 5066–5073. [29] S. Wiemann, V. Kinzel, W. Pyerin, Isoform c beta 2, an unusual form of the bovine catalytic subunit of camp-dependent protein kinase, J. Biol. Chem. 266 (1991) 5140–5146. [30] B. Zimmermann, J.A. Chiorini, Y. Ma, R.M. Kotin, F.W. Herberg, Prkx is a novel catalytic subunit of the camp-dependent protein kinase regulated by the regulatory subunit type i, J. Biol. Chem. 274 (1999) 5370–5378. [31] A. Klink, K. Schiebel, M. Winkelmann, E. Rao, B. Horsthemke, H.J. Ludecke, U. Claussen, G. Scherer, G. Rappold, The human protein kinase gene pkx1 on xp22.3 displays xp/yp homology and is a site of chromosomal instability, Hum. Mol. Genet. 4 (1995) 869–878. [32] P.T. Jedrzejewski, A. Girod, A. Tholey, N. Konig, S. Thullner, V. Kinzel, D. Bossemeyer, A conserved deamidation site at asn 2 in the catalytic subunit of mammalian camp-dependent protein kinase detected by capillary lc-ms and tandem mass spectrometry, Protein Sci. 7 (1998) 457–469.
1798
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
[33] S. Shoji, K. Titani, J.G. Demaille, E.H. Fischer, Sequence of two phosphorylated sites in the catalytic subunit of bovine cardiac muscle adenosine 3′:5′-monophosphate-dependent protein kinase, J. Biol. Chem. 254 (1979) 6211–6214. [34] S.A. Carr, K. Biemann, S. Shoji, D.C. Parmelee, K. Titani, Ntetradecanoyl is the nh2-terminal blocking group of the catalytic subunit of cyclic amp-dependent protein kinase from bovine cardiac muscle, Proc. Natl. Acad. Sci. U. S. A. 79 (1982) 6128–6131. [35] W. Yonemoto, M. McGlone, B. Grant, S. Taylor, Autophosphorylation of the catalytic subunit of camp-dependent protein kinase in Escherichia coli, Protein Eng. 10 (1997) 915–925. [36] P.J. Bechtel, J.A. Beavo, E.G. Krebs, Purification and characterization of catalytic subunit of skeletal muscle adenosine 3′:5′-monophosphatedependent protein kinase, J. Biol. Chem. 252 (1977) 2691–2697. [37] Y.S. Chiu, M. Tao, Autophosphorylation of rabbit skeletal muscle cyclic amp-dependent protein kinase i catalytic subunit, J. Biol. Chem. 253 (1978) 7145–7148. [38] S. Liauw, R.A. Steinberg, Dephosphorylation of catalytic subunit of camp-dependent protein kinase at thr-197 by a cellular protein phosphatase and by purified protein phosphatase-2a, J. Biol. Chem. 271 (1996) 258–263. [39] J.A. Adams, M.L. McGlone, R. Gibson, S.S. Taylor, Phosphorylation modulates catalytic function and regulation in the camp-dependent protein kinase, Biochemistry 34 (1995) 2447–2454. [40] R.A. Steinberg, R.D. Cauthron, M.M. Symcox, H. Shuntoh, Autoactivation of catalytic (c alpha) subunit of cyclic amp-dependent protein kinase by phosphorylation of threonine 197, Mol. Cell. Biol. 13 (1993) 2332–2341. [41] R.D. Cauthron, K.B. Carter, S. Liauw, R.A. Steinberg, Physiological phosphorylation of protein kinase a at thr-197 is by a protein kinase a kinase, Mol. Cell. Biol. 18 (1998) 1416–1423. [42] X. Cheng, Y. Ma, M. Moore, B.A. Hemmings, S.S. Taylor, Phosphorylation and activation of camp-dependent protein kinase by phosphoinositide-dependent protein kinase, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 9849–9854. [43] L.N. Johnson, M.E. Noble, D.J. Owen, Active and inactive protein kinases: structural basis for regulation, Cell 85 (1996) 149–158. [44] S.S. Taylor, E. Radzioandzelm, 3 protein-kinase structures define a common motif, Structure 2 (1994) 345–355. [45] G. Waksman, D. Kominos, S.C. Robertson, N. Pant, D. Baltimore, R.B. Birge, D. Cowburn, H. Hanafusa, B.J. Mayer, M. Overduin, M.D. Resh, C.B. Rios, L. Silverman, J. Kuriyan, Crystal structure of the phosphotyrosine recognition domain sh2 of v-src complexed with tyrosine-phosphorylated peptides, Nature 358 (1992) 646–653. [46] M.B. Yaffe, A.E.H. Elia, Phosphoserine/threonine-binding domains, Curr. Opin. Cell Biol. 13 (2001) 131–138. [47] S. Shoji, D.C. Parmelee, R.D. Wade, S. Kumar, L.H. Ericsson, K.A. Walsh, H. Neurath, G.L. Long, J.G. Demaille, E.H. Fischer, K. Titani, Complete amino acid sequence of the catalytic subunit of bovine cardiac muscle cyclic amp-dependent protein kinase, Proc. Natl. Acad. Sci. U. S. A. 78 (1981) 848–851. [48] T. Magee, M.C. Seabra, Fatty acylation and prenylation of proteins: what's hot in fat, Curr. Opin. Cell Biol. 17 (2005) 190–196. [49] J.F. Hancock, H. Paterson, C.J. Marshall, A polybasic domain or palmitoylation is required in addition to the caax motif to localize p21ras to the plasma membrane, Cell 63 (1990) 133–139. [50] A.Q. Gomes, B.R. Ali, J.S. Ramalho, R.F. Godfrey, D.C. Barral, A.N. Hume, M.C. Seabra, Membrane targeting of rab gtpases is influenced by the prenylation motif, Mol. Biol. Cell 14 (2003) 1882–1899. [51] S.A. Armstrong, V.C. Hannah, J.L. Goldstein, M.S. Brown, Caax geranylgeranyl transferase transfers farnesyl as efficiently as geranylgeranyl to rhob, J. Biol. Chem. 270 (1995) 7864–7868. [52] I.A. Prior, C. Muncke, R.G. Parton, J.F. Hancock, Direct visualization of ras proteins in spatially distinct cell surface microdomains, J. Cell Biol. 160 (2003) 165–170. [53] M. Wherlock, A. Gampel, C. Futter, H. Mellor, Farnesyltransferase inhibitors disrupt egf receptor traffic through modulation of the rhob gtpase, J. Cell Sci. 117 (2004) 3221–3231.
[54] M. Calero, C.Z. Chen, W. Zhu, N. Winand, K.A. Havas, P.M. Gilbert, C.G. Burd, R.N. Collins, Dual prenylation is required for rab protein localization and function, Mol. Biol. Cell 14 (2003) 1852–1867. [55] B. Nagar, O. Hantschel, M.A. Young, K. Scheffzek, D. Veach, W. Bornmann, B. Clarkson, G. Superti-Furga, J. Kuriyan, Structural basis for the autoinhibition of c-abl tyrosine kinase, Cell 112 (2003) 859–871. [56] O. Hantschel, B. Nagar, S. Guettler, J. Kretzschmar, K. Dorey, J. Kuriyan, G. Superti-Furga, A myristoyl/phosphotyrosine switch regulates c-abl, Cell 112 (2003) 845–857. [57] J. Struppe, E.A. Komives, S.S. Taylor, R.R. Vold, 2h nmr studies of a myristoylated peptide in neutral and acidic phospholipid bicelles, Biochemistry 37 (1998) 15523–15527. [58] L. Silverman, M. Resh, Lysine residues form an integral component of a novel nh2-terminal membrane targeting motif for myristylated pp60v-src, J. Cell Biol. 119 (1992) 415–425. [59] L.K. Langeberg, J.D. Scott, A-kinase-anchoring proteins, J. Cell Sci. 118 (2005) 3217–3220. [60] C. Breitenlechner, R.A. Engh, R. Huber, V. Kinzel, D. Bossemeyer, M. Gassel, The typically disordered n-terminus of pka can fold as a helix and project the myristoylation site into solution, Biochemistry 43 (2004) 7743–7749. [61] M. Gangal, T. Clifford, J. Deich, X. Cheng, S.S. Taylor, D.A. Johnson, Mobilization of the a-kinase n-myristate through an isoformspecific intermolecular switch, Proc. Natl. Acad. Sci. 96 (1999) 12394–12399. [62] W. Yonemoto, M. McGlone, S. Taylor, N-myristylation of the catalytic subunit of camp-dependent protein kinase conveys structural stability, J. Biol. Chem. 268 (1993) 2348–2352. [63] F.W. Herberg, B. Zimmermann, M. McGlone, S.S. Taylor, Importance of the a-helix of the catalytic subunit of camp-dependent protein kinase for stability and for orienting subdomains at the cleft interface, Protein Sci. 6 (1997) 569–579. [64] M.P. Washburn, D. Wolters, J.R. Yates III, Large-scale analysis of the yeast proteome by multidimensional protein identification technology, Nat. Biotechnol. 19 (2001) 242–247. [65] E. Nagele, M. Vollmer, P. Horth, C. Vad, 2d-lc/ms techniques for the identification of proteins in highly complex mixtures, Expert Rev. Proteomics 1 (2004) 37–46. [66] E.I. Chen, J. Hewel, B. Felding-Habermann, J.R. Yates III, Large scale protein profiling by combination of protein fractionation and multidimensional protein identification technology (mudpit), Mol. Cell. Proteomics 5 (2006) 53–56. [67] A. Hotz, N. König, J. Kretschmer, G. Maier, H. Ponstingl, V. Kinzel, A sequence variant in the n-terminal region of the catalytic subunit of the camp-dependent protein kinase, FEBS Lett. 256 (1989). [68] H.T. Wright, Nonenzymatic deamidation of asparaginyl and glutaminyl residues in proteins, Crit. Rev. Biochem. Mol. Biol. 26 (1991) 1–52. [69] V. Kinzel, A. Hotz, N. Konig, M. Gagelmann, W. Pyerin, J. Reed, D. Kubler, F. Hofmann, C. Obst, H.P. Gensheimer, D. Goldblatt, S. Shaltiel, Chromatographic separation of two heterogeneous forms of the catalytic subunit of cyclic amp-dependent protein kinase holoenzyme type i and type ii from striated muscle of different mammalian species, Arch. Biochem. Biophys. 253 (1987) 341–349. [70] R. Pepperkok, A. Hotz-Wagenblatt, N. Konig, A. Girod, D. Bossemeyer, V. Kinzel, Intracellular distribution of mammalian protein kinase a catalytic subunit altered by conserved asn2 deamidation, J. Cell Biol. 148 (2000) 715–726. [71] W. Henzel, T. Billeci, J. Stults, S. Wong, C. Grimley, C. Watanabe, Identifying proteins from two-dimensional gels by molecular mass searching of peptide fragments in protein sequence databases, Proc. Natl. Acad. Sci. 90 (1993) 5011–5015. [72] M. Mann, M. Wilm, Error-tolerant identification of peptides in sequence databases by peptide sequence tags, Anal. Chem. 66 (1994) 4390–4399.
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800 [73] D.N. Perkins, D.J.C. Pappin, D.M. Creasy, J.S. Cottrell, Probabilitybased protein identification by searching sequence databases using mass spectrometry data, Electrophoresis 20 (1999) 3551–3567. [74] D.J. Pappin, P. Hojrup, A.J. Bleasby, Rapid identification of proteins by peptide-mass fingerprinting, Curr. Biol. 3 (1993) 327–332. [75] J.K. Eng, A.L. McCormack, J.R. Yates, An approach to correlate tandem mass-spectral data of peptides with amino-acid-sequences in a protein database, J. Am. Soc. Mass Spectrom 5 (1994) 976–989. [76] J.R. Yates, J.K. Eng, A.L. McCormack, D. Schieltz, Method to correlate tandem mass-spectra of modified peptides to amino-acid-sequences in the protein database, Anal. Chem. 67 (1995) 1426–1436. [77] E. Gasteiger, A. Gattiker, C. Hoogland, I. Ivanyi, R.D. Appel, A. Bairoch, Expasy: the proteomics server for in-depth protein knowledge and analysis, Nucleic Acids Res. 31 (2003) 3784–3788. [78] D.R. Knighton, J.H. Zheng, L.F. Ten Eyck, N.H. Xuong, S.S. Taylor, J.M. Sowadski, Structure of a peptide inhibitor bound to the catalytic subunit of cyclic adenosine monophosphate-dependent protein kinase, Science 253 (1991) 414–420. [79] D. Bossemeyer, R.A. Engh, V. Kinzel, H. Ponstingl, R. Huber, Phosphotransferase and substrate binding mechanism of the campdependent protein kinase catalytic subunit from porcine heart as deduced from the 2.0 a structure of the complex with mn2+ adenylyl imidodiphosphate and inhibitor peptide pki(5-24), EMBO J. 12 (1993) 849–859. [80] H. Steen, J.A. Jebanathirajah, J. Rush, N. Morrice, M.W. Kirschner, Phosphorylation analysis by mass spectrometry: myths, facts, and the consequences for qualitative and quantitative measurements, Mol. Cell. Proteomics 5 (2006) 172–181. [81] A. Schlosser, R. Pipkorn, D. Bossemeyer, W.D. Lehmann, Analysis of protein phosphorylation by a combination of elastase digestion and neutral loss tandem mass spectrometry, Anal. Chem. 73 (2001) 170–176. [82] A. Schlosser, J.T. Vanselow, A. Kramer, Mapping of phosphorylation sites by a multi-protease approach with specific phosphopeptide enrichment and nanolc-ms/ms analysis, Anal. Chem. 77 (2005) 5243–5250. [83] Z.A. Knight, B. Schilling, R.H. Row, D.M. Kenski, B.W. Gibson, K.M. Shokat, Phosphospecific proteolysis for mapping sites of protein phosphorylation, Nat. Biotechnol. 21 (2003) 1047–1054. [84] G. Arrigoni, S. Resjo, F. Levander, R. Nilsson, E. Degerman, M. Quadroni, L.A. Pinna, P. James, Chemical derivatization of phosphoserine and phosphothreonine containing peptides to increase sensitivity for maldi-based analysis and for selectivity of ms/ms analysis, Proteomics 6 (2006) 757–766. [85] J. Porath, J. Carlsson, I. Olsson, G. Belfrage, Metal chelate affinity chromatography, a new approach to protein fractionation, Nature 258 (1975) 598–599. [86] M.C. Posewitz, P. Tempst, Immobilized gallium(iii) affinity chromatography of phosphopeptides, Anal. Chem. 71 (1999) 2883–2892. [87] P. Cao, J.T. Stults, Phosphopeptide analysis by on-line immobilized metal-ion affinity chromatography-capillary electrophoresis-electrospray ionization mass spectrometry, J. Chromatogr. A 853 (1999) 225–235. [88] M.W. Pinkse, P.M. Uitto, M.J. Hilhorst, B. Ooms, A.J. Heck, Selective isolation at the femtomole level of phosphopeptides from proteolytic digests using 2d-nanolc-esi-ms/ms and titanium oxide precolumns, Anal. Chem. 76 (2004) 3935–3943. [89] M.R. Larsen, T.E. Thingholm, O.N. Jensen, P. Roepstorff, T.J.D. Jorgensen, Highly selective enrichment of phosphorylated peptides from peptide mixtures using titanium dioxide microcolumns, Mol. Cell. Proteomics 4 (2005) 873–886. [90] M.J. Chalmers, K. Håkansson, R. Johnson, R. Smith, J. Shen, M.R. Emmett, A.G. Marshall, Protein kinase a phosphorylation characterized by tandem Fourier transform ion cyclotron resonance mass spectrometry, Proteomics 4 (2004) 970–981. [91] F.W. Herberg, S.M. Bell, S.S. Taylor, Expression of the catalytic subunit of camp-dependent protein kinase in Escherichia coli: multiple isozymes reflect different phosphorylation states, Protein Eng. 6 (1993) 771–777.
1799
[92] S.-E. Ong, B. Blagoev, I. Kratchmarova, D.B. Kristensen, H. Steen, A. Pandey, M. Mann, Stable isotope labeling by amino acids in cell culture, silac, as a simple and accurate approach to expression proteomics, Mol. Cell. Proteomics 1 (2002) 376–386. [93] S.P. Gygi, B. Rist, S.A. Gerber, F. Turecek, M.H. Gelb, R. Aebersold, Quantitative analysis of complex protein mixtures using isotope-coded affinity tags, Nat. Biotechnol. 17 (1999) 994–999. [94] H. Zhou, J.A. Ranish, J.D. Watts, R. Aebersold, Quantitative proteome analysis by solid-phase isotope tagging and mass spectrometry, Nat. Biotechnol. 20 (2002) 512–515. [95] P.L. Ross, Y.N. Huang, J.N. Marchese, B. Williamson, K. Parker, S. Hattan, N. Khainovski, S. Pillai, S. Dey, S. Daniels, S. Purkayastha, P. Juhasz, S. Martin, M. Bartlet-Jones, F. He, A. Jacobson, D.J. Pappin, Multiplexed protein quantitation in Saccharomyces cerevisiae using amine-reactive isobaric tagging reagents, Mol. Cell. Proteomics 3 (2004) 1154–1169. [96] B. Blagoev, I. Kratchmarova, S.-E. Ong, M. Nielsen, L.J. Foster, M. Mann, A proteomics strategy to elucidate functional protein–protein interactions applied to egf signaling, Nat. Biotechnol. 21 (2003) 315–318. [97] A. Gruhler, J.V. Olsen, S. Mohammed, P. Mortensen, N.J. Faergeman, M. Mann, O.N. Jensen, Quantitative phosphoproteomics applied to the yeast pheromone signaling pathway, Mol. Cell. Proteomics 4 (2005) 310–327. [98] F. Schmidt, S. Donahoe, K. Hagens, J. Mattow, U.E. Schaible, S.H.E. Kaufmann, R. Aebersold, P.R. Jungblut, Complementary analysis of the mycobacterium tuberculosis proteome by two-dimensional electrophoresis and isotope-coded affinity tag technology, Mol. Cell. Proteomics 3 (2004) 24–42. [99] S.A. Gerber, J. Rush, O. Stemman, M.W. Kirschner, S.P. Gygi, Absolute quantification of proteins and phosphoproteins from cell lysates by tandem ms, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 6940–6945. [100] D.M. Cox, F. Zhong, M. Du, E. Duchoslav, T. Sakuma, J.C. McDermott, Multiple reaction monitoring as a method for identifying protein posttranslational modifications, J. Biomol. Tech. 16 (2005) 83–90. [101] V. Mayya, K. Rezaul, L. Wu, M.B. Fong, D.K. Han, Absolute quantification of multi-site phosphorylation by selective reaction monitoring mass spectrometry: determination of inhibitory phosphorylation status of cyclin dependent kinases, Mol. Cell. Proteomics 5 (2006) 1146–1157. [102] N.L. Kelleher, Top-down proteomics, Anal. Chem. 76 (2004) 197A–203A. [103] M.W. Senko, C.L. Hendrickson, M.R. Emmett, S.D.-H. Shi, A.G. Marshall, External accumulation of ions for enhanced electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry, J. Am. Soc. Mass Spectrom 8 (1997) 970–976. [104] B. Domon, R. Aebersold, Mass spectrometry and protein analysis, Science 312 (2006) 212–217. [105] B. Macek, L.F. Waanders, J.V. Olsen, M. Mann, Top-down protein sequencing and ms3 on a hybrid linear quadrupole ion trap-orbitrap mass spectrometer, Mol. Cell. Proteomics 5 (2006) 949–958. [106] M.D. Hoffman, J. Kast, Mass spectrometric characterization of lipidmodified peptides for the analysis of acylated proteins, J. Mass Spectrom 41 (2006) 229–241. [107] T. Geiger, S. Clarke, Deamidation, isomerization, and racemization at asparaginyl and aspartyl residues in peptides. Succinimide-linked reactions that contribute to protein degradation, J. Biol. Chem. 262 (1987) 785–794. [108] J. Lowenson, S. Clarke, Recognition of D-aspartyl residues in polypeptides by the erythrocyte L- isoaspartyl/D-aspartyl protein methyltransferase. Implications for the repair hypothesis, J. Biol. Chem. 267 (1992) 5985–5995. [109] W. Lehmann, A. Schlosser, G. Erben, R. Pipkorn, D. Bossemeyer, V. Kinzel, Analysis of isoaspartate in peptides by electrospray tandem mass spectrometry, Protein Sci. 9 (2000) 2260–2268. [110] V. Kinzel, N. Konig, R. Pipkorn, D. Bossemeyer, W. Lehmann, The amino
1800
F. Gesellchen et al. / Biochimica et Biophysica Acta 1764 (2006) 1788–1800
terminus of pka catalytic subunit-a site for introduction of posttranslational heterogeneities by deamidation: D-asp2 and d-isoasp2 containing isozymes, Protein Sci. 9 (2000) 2269–2277. [111] J. Zheng, D.R. Knighton, N. Xuong, S.S. Taylor, J.M. Sowadski, L.F. Ten-Eyck, Crystal structures of the myristylated catalytic subunit of camp-dependent protein kinase reveal open and closed conformations, Protein Sci. 2 (1993) 1559–1573. [112] W. Humphrey, A. Dalke, K. Schulten, Vmd: Visual molecular dynamics, J. Mol. Graph. 14 (1996) 33–8, 27–8.
[113] S.R. Hubbard, L. Wei, L. Ellis, W.A. Hendrickson, Crystal structure of the tyrosine kinase domain of the human insulin receptor, Nature 372 (1994) 746–754. [114] S.R. Hubbard, Crystal structure of the activated insulin receptor tyrosine kinase in complex with peptide substrate and atp analog, EMBO J. 16 (1997) 5572–5581. [115] A. Shevchenko, M. Wilm, O. Vorm, M. Mann, Mass spectrometric sequencing of proteins from silver-stained polyacrylamide gels, Anal. Chem. 68 (1996) 850–858.