Mining a tandem mass spectrometry database to determine the trends and global factors influencing peptide fragmentation

Fragmentation characteristics of collision-induced dissociation in MALDI TOF/TOF mass spectrometry

The identification of proteins by tandem mass spectrometry relies on knowledge of the products produced by collision-induced dissociation of peptide ions. Most previous work has focused on fragmentation statistics for ion trap systems. We analyzed fragmentation in MALDI TOF/TOF mass spectrometry, collecting statistics using a curated set of 2459 MS/MS spectra and applying bootstrap resampling to assess confidence intervals. We calculated the frequency of 18 product ion types, the correlation between both mass and intensity with ion type, the dependence of amide bond breakage on the residues surrounding the cleavage site, and the dependence of product ion detection on residues not adjacent to the cleavage site. The most frequently observed were internal ions, followed by y ions. A strong correlation between ion type and the mass and intensity of its peak was observed, with b and y ions producing the most intense and highest mass peaks. The amino acids P, W, D, and R had a strong effect on amide bond cleavage when situated next to the breakage site, whereas residues including I, K, and H had a strong effect on product ion observation when located in the peptide but not adjacent to the cleavage site, a novel observation.

De novo peptide sequencing using ion peak intensity and amino acid cleavage intensity ratio

MOTIVATION

Peptide-sequencing methods by mass spectrum use the following two approaches: database searching and de novo sequencing. The database-searching approach is convenient; however, in cases wherein the corresponding sequences are not included in the databases, the exact identification is difficult. On the other hand, in the case of de novo sequencing, no preliminary information is necessary; however, continuous amino acid sequence peaks and the differentiation of these peaks are required. It is, however, very difficult to obtain and differentiate the peaks of all amino acids by using an actual spectrum. We propose a novel de novo sequencing approach using not only mass-to-charge ratio but also ion peak intensity and amino acid cleavage intensity ratio (CIR).

RESULTS

Our method compensates for any undetectable amino acid peak intervals by estimating the amino acid set and the probability of peak expression based on amino acid CIR. It provides more accurate identification of sequences than the existing methods, by which it is usually difficult to sequence.

The disulfide bond pattern of salmon egg lectin 24K from the Chinook salmon Oncorhynchus tshawytscha

The disulfide bonds in the galactose-specific lectin SEL 24K from the egg of the Chinook salmon Oncorhynchus tshawytscha were determined by mass spectrometry. Four predictive in silico tools were used to determine the oxidation state of cysteines in the sequence and possible location of the disulfide bonds. A combination of tryptic digestion, HPLC separation, and chemical modifications were used to establish the location of seven disulfide bonds and one pair of free cysteines. After proteolysis, peptides containing one or two disulfide bonds were identified by reduction and mass spectral comparison. MALDI mass spectrometry was supported by chemical modification (iodoacetamide) and in silico digestion. The assignments of disulfide bonds were further confirmed by mass spectral fragmentation studies including in-source dissociation (ISD) and collision-induced dissociation (CID). The experimentally determined disulfide bonds and free Cys residues were only partially consistent with those generated by several automated public-domain algorithms.

Global proteomic profiling of phosphopeptides using electron transfer dissociation tandem mass spectrometry

Electron transfer dissociation (ETD) is a recently introduced mass spectrometric technique that provides a more comprehensive coverage of peptide sequences and posttranslational modifications. Here, we evaluated the use of ETD for a global phosphoproteome analysis. In all, we identified a total of 1,435 phosphorylation sites from human embryonic kidney 293T cells, of which 1,141 ( approximately 80%) were not previously described. A detailed comparison of ETD and collision-induced dissociation (CID) modes showed that ETD identified 60% more phosphopeptides than CID, with an average of 40% more fragment ions that facilitated localization of phosphorylation sites. Although our data indicate that ETD is superior to CID for phosphorylation analysis, the two methods can be effectively combined in alternating ETD and CID modes for a more comprehensive analysis. Combining ETD and CID, from this single study, we were able to identify 80% of the known phosphorylation sites in >1,000 phosphorylated peptides analyzed. A hierarchical clustering of the identified phosphorylation sites allowed us to discover 15 phosphorylation motifs that have not been reported previously. Overall, ETD is an excellent method for localization of phosphorylation sites and should be an integral component of any strategy for comprehensive phosphorylation analysis.

Methods for peptide identification by spectral comparison

Background

Tandem mass spectrometry followed by database search is currently the predominant technology for peptide sequencing in shotgun proteomics experiments. Most methods compare experimentally observed spectra to the theoretical spectra predicted from the sequences in protein databases. There is a growing interest, however, in comparing unknown experimental spectra to a library of previously identified spectra. This approach has the advantage of taking into account instrument-dependent factors and peptide-specific differences in fragmentation probabilities. It is also computationally more efficient for high-throughput proteomics studies.

Results

This paper investigates computational issues related to this spectral comparison approach. Different methods have been empirically evaluated over several large sets of spectra. First, we illustrate that the peak intensities follow a Poisson distribution. This implies that applying a square root transform will optimally stabilize the peak intensity variance. Our results show that the square root did indeed outperform other transforms, resulting in improved accuracy of spectral matching. Second, different measures of spectral similarity were compared, and the results illustrated that the correlation coefficient was most robust. Finally, we examine how to assemble multiple spectra associated with the same peptide to generate a synthetic reference spectrum. Ensemble averaging is shown to provide the best combination of accuracy and efficiency.

Conclusion

Our results demonstrate that when combined, these methods can boost the sensitivity and specificity of spectral comparison. Therefore they are capable of enhancing and complementing existing tools for consistent and accurate peptide identification.

Dehydration versus deamination of N-terminal glutamine in collision-induced dissociation of protonated peptides

Some of the most prominent "neutral losses" in peptide ion fragmentation are the loss of ammonia and water from N-terminal glutamine. These processes are studied by electrospray ionization mass spectrometry in singly- and doubly-protonated peptide ions undergoing collision-induced dissociation in a triple quadrupole and in an ion trap instrument. For this study, four sets of peptides were synthesized: (1) QLLLPLLLK and similar peptides with K replaced by R, H, or L, and Q replaced by a number of amino acids, (2) QLnK (n = 0, 1, 3, 5, 7, 9, 11), (3) QLnR (n = 0, 1, 3, 5, 7, 9), and (4) QLn (n = 1, 2, 3, 4, 8). The results for QLLLPLLLK and QLLLPLLLR show that the singly protonated ions undergo loss of ammonia and to a smaller extent loss of water, whereas the doubly protonated ions undergo predominant loss of water. The fast fragmentation next to P (forming the y5 ion) occurs to a larger extent than the neutral losses from the singly protonated ions but much less than the water loss from the doubly protonated ions. The results from these and other peptides show that, in general, when N-terminal glutamine peptides have no "mobile protons", that is, the number of charges on the peptide is no greater than the number of basic amino acids (K, R, H), deamination is the predominant neutral loss fragmentation, but when mobile protons are present the predominant process is the loss of water. Both of these processes are faster than backbone fragmentation at the proline. These results are rationalized on the basis of resonance stabilization of the two types of five-membered ring products that would be formed in the neutral loss processes; the singly protonated ion yields the more stable neutral pyrrolidinone ring whereas the doubly protonated ion yields the protonated aminopyrroline ring (see Schemes). The generality of these trends is confirmed by analyzing an MS/MS spectra library of peptides derived from tryptic digests of yeast. In the absence of mobile protons, glutamine deamination is the most rapid neutral loss process. For peptides with mobile protons, dehydration from glutamine is far more rapid than from any other amino acid. Most strikingly, end terminal glutamine is by far the most labile source of neutral loss in excess-proton peptides, but not highly exceptional when mobile protons are not available. In addition, rates of deamination are faster in lysine versus arginine C-terminus peptides and 20 times faster in positively charged than negatively charged peptides, demonstrating that these formal neutral loss reactions are not "neutral reactions" but depend on charge state and stability.

Emerging methods in proteomics: top-down protein characterization by multistage tandem mass spectrometry

"Top-down" mass spectrometry methods have emerged as an attractive alternative to conventional "bottom-up" approaches for the comprehensive characterization of co- and post-translational protein modifications. Here we present a brief overview of current strategies employed for top-down protein characterization and discuss the key technical challenges and solutions associated with their implementation on a range of mass spectrometry instrument platforms. For more specific details regarding the individual strategies described herein, interested readers are referred to the references cited at the end of this article.

Substituent effects on the gas-phase fragmentation reactions of sulfonium ion containing peptides

The multistage mass spectrometric (MS/MS and MS3) gas-phase fragmentation reactions of methionine side-chain sulfonium ion containing peptides formed by reaction with a series of para-substituted phenacyl bromide (XBr where X=CH2COC6H4R, and R=--COOH, --COOCH3, --H, --CH3 and --CH2CH3) alkylating reagents have been examined in a linear quadrupole ion trap mass spectrometer. MS/MS of the singly (M+) and multiply ([M++nH](n+1)+) charged precursor ions results in exclusive dissociation at the fixed charge containing side chain, independently of the amino acid composition and precursor ion charge state (i.e., proton mobility). However, loss of the methylphenacyl sulfide side-chain fragment as a neutral versus charged (protonated) species was observed to be highly dependent on the proton mobility of the precursor ion, and the identity of the phenacyl group para-substituent. Molecular orbital calculations were performed at the B3LYP/6-31+G** level of theory to calculate the theoretical proton affinities of the neutral side-chain fragments. The log of the ratio of neutral versus protonated side-chain fragment losses from the derivatized side chain were found to exhibit a linear dependence on the proton affinity of the side-chain fragmentation product, as well as the proton affinities of the peptide product ions. Finally, MS3 dissociation of the nominally identical neutral and protonated loss product ions formed by MS/MS of the [M++H]2+ and [M++2H]3+ precursor ions, respectively, from the peptide GAILM(X)GAILK revealed significant differences in the abundances of the resultant product ions. These results suggest that the protonated peptide product ions formed by gas-phase fragmentation of sulfonium ion containing precursors in an ion trap mass spectrometer do not necessarily undergo intramolecular proton 'scrambling' prior to their further dissociation, in contrast to that previously demonstrated for peptide ions introduced by external ionization sources.

Mechanisms for the selective gas-phase fragmentation reactions of methionine side chain fixed charge sulfonium ion containing peptides

To enable the development of improved tandem mass spectrometry based methods for selective proteome analysis, the mechanisms, product ion structures, and other factors influencing the gas-phase fragmentation reactions of methionine side-chain derivatized "fixed-charge" phenacylsulfonium ion containing peptide ions have been examined. Dissociation of these peptide ions results in the exclusive characteristic loss of the derivatized side chain, thereby enabling their selective identification. The resultant product ion(s) are then subjected to further dissociation to obtain sequence information for subsequent protein identification. Molecular orbital calculations (at the B3LYP/6-31 + G (d,p) level of theory) performed on a simple peptide model, together with experimental evidence obtained by multistage dissociation of a regioselectively deuterated methionine derivatized sulfonium ion containing tryptic peptide, indicate that fragmentation of the fixed charge containing peptide ions occurs via SN2 reactions involving the N- and C-terminal amide bonds adjacent to the methionine side chain, resulting in the formation of stable cyclic five- and six-membered iminohydrofuran and oxazine product ions, respectively. These studies further indicate that the rings formed via these neighboring group reactions are stable to further dissociation by MS3. As a consequence, the formation of b- or y-type sequence ions are "skipped" at the site of cyclization. Despite this, complete sequence information is still obtained because of the presence of both cyclic products.

Tandem mass spectrometry of amidated peptides

The behavior of C-terminal amidated and carboxylated peptides upon low-energy collision-induced dissociation (CID) was investigated. Two sets of 76 sequences of variable amino acid compositions and lengths were synthesized as model compounds. In most cases, C-terminal amidated peptides were found to produce, upon CID, an abundant loss of ammonia from the protonated molecules. To validate such MS/MS signatures, the studied peptides contained amino acids that can potentially release ammonia from their side chains, such as asparagine, glutamine, tryptophan, lysine and arginine. Arginine, and to a lesser extent lysine, was shown to induce a competitive fragmentation leading to the loss of ammonia from their side chains, thus interfering with the targeted backbone neutral release. However, when arginine or lysine was located at the C-terminal position mimicking a tryptic digest, losses of ammonia from the arginine side chain and from the peptide backbone were completely suppressed. Such results were discussed in the frame of peptidomic or proteomic studies in an attempt to reveal the presence of C-terminal amidated peptides or proteins.

De novo peptide sequencing using exhaustive enumeration of peptide composition

We introduce the use of a peptide composition lookup table indexed by residual mass and number of amino acids for de novo sequencing of polypeptides. Polypeptides of 1600 Daltons (Da) or more can be sequenced effectively through exhaustive compositional analysis of MS/MS spectra obtained by unimolecular decomposition (without CID) in a MALDI TOF/TOF despite a fragment mass accuracy of 50 mDa. Peaks are referenced against the lookup table to obtain a complete profile of amino acid combinations, and combinations are assembled into series of increasing length. Concatenating the differences between successive entries in compositional series yields peptide sequences that can be scored and ranked according to signal intensity. While the current work involves measurements acquired on MALDI TOF-TOF, such general treatment of the data anticipates extension to other types of mass analyzers.

MALDI-TOF/TOFde novo sequence analysis of 2-D PAGE-separated proteins fromHalorhodospira halophila, a bacterium with unsequenced genome

Because protein identifications rely on matches with sequence databases, high-throughput proteomics is currently largely restricted to those species for which comprehensive sequence databases are available. The identification of proteins derived from organisms with unsequenced genomes mainly depends on homology searching. Here, we report the use of a simplified, gel-based, chemical derivatization strategy for de novo sequence analysis using a MALDI-TOF/TOF mass spectrometer. This approach allows the determination of de novo peptide sequences of up to 20 amino acid residues in length. The protocol was applied on a proteomic study of 2-D PAGE-separated proteins from Halorhodospira halophila, an extremophilic eubacterium with yet unsequenced genome. Using three different homology-based search algorithms, we were able to identify more than 30 proteins from this organism using subpicomole quantities of protein.

PRIMA: peptide robust identification from MS/MS spectra

In proteomics, tandem mass spectrometry is the key technology for peptide sequencing. However, partially due to the deficiency of peptide identification software, a large portion of the tandem mass spectra are discarded in almost all proteomics centers because they are not interpretable. The problem is more acute with the lower quality data from low end but more popular devices such as the ion trap instruments. In order to deal with the noisy and low quality data, this paper develops a systematic machine learning approach to construct a robust linear scoring function, whose coefficients are determined by a linear programming. A prototype, PRIMA, was implemented. When tested with large benchmarks of varying qualities, PRIMA consistently has higher accuracy than commonly used software MASCOT, SEQUEST and X! Tandem.

Identification and stoichiometry of glycosylphosphatidylinositol-anchored membrane proteins of the human malaria parasite Plasmodium falciparum

Most proteins that coat the surface of the extracellular forms of the human malaria parasite Plasmodium falciparum are attached to the plasma membrane via glycosylphosphatidylinositol (GPI) anchors. These proteins are exposed to neutralizing antibodies, and several are advanced vaccine candidates. To identify the GPI-anchored proteome of P. falciparum we used a combination of proteomic and computational approaches. Focusing on the clinically relevant blood stage of the life cycle, proteomic analysis of proteins labeled with radioactive glucosamine identified GPI anchoring on 11 proteins (merozoite surface protein (MSP)-1, -2, -4, -5, -10, rhoptry-associated membrane antigen, apical sushi protein, Pf92, Pf38, Pf12, and Pf34). These proteins represent approximately 94% of the GPI-anchored schizont/merozoite proteome and constitute by far the largest validated set of GPI-anchored proteins in this organism. Moreover MSP-1 and MSP-2 were present in similar copy number, and we estimated that together these proteins comprise approximately two-thirds of the total membrane-associated surface coat. This is the first time the stoichiometry of MSPs has been examined. We observed that available software performed poorly in predicting GPI anchoring on P. falciparum proteins where such modification had been validated by proteomics. Therefore, we developed a hidden Markov model (GPI-HMM) trained on P. falciparum sequences and used this to rank all proteins encoded in the completed P. falciparum genome according to their likelihood of being GPI-anchored. GPI-HMM predicted GPI modification on all validated proteins, on several known membrane proteins, and on a number of novel, presumably surface, proteins expressed in the blood, insect, and/or pre-erythrocytic stages of the life cycle. Together this work identified 11 and predicted a further 19 GPI-anchored proteins in P. falciparum.

Differential exoprotease activities confer tumor-specific serum peptidome patterns

Recent studies have established distinctive serum polypeptide patterns through mass spectrometry (MS) that reportedly correlate with clinically relevant outcomes. Wider acceptance of these signatures as valid biomarkers for disease may follow sequence characterization of the components and elucidation of the mechanisms by which they are generated. Using a highly optimized peptide extraction and matrix-assisted laser desorption/ionization-time-of-flight (MALDI-TOF) MS-based approach, we now show that a limited subset of serum peptides (a signature) provides accurate class discrimination between patients with 3 types of solid tumors and controls without cancer. Targeted sequence identification of 61 signature peptides revealed that they fall into several tight clusters and that most are generated by exopeptidase activities that confer cancer type-specific differences superimposed on the proteolytic events of the ex vivo coagulation and complement degradation pathways. This small but robust set of marker peptides then enabled highly accurate class prediction for an external validation set of prostate cancer samples. In sum, this study provides a direct link between peptide marker profiles of disease and differential protease activity, and the patterns we describe may have clinical utility as surrogate markers for detection and classification of cancer. Our findings also have important implications for future peptide biomarker discovery efforts.

Automated protein identification by tandem mass spectrometry: issues and strategies

Protein identification by tandem mass spectrometry (MS/MS) is key to most proteomics projects and has been widely explored in bioinformatics research. Obtaining good and trustful identification results has important implications for biological and clinical work. Although well matured, automated software identification of proteins from MS/MS data still faces a number of obstacles due to the complexity of the proteome or procedural issues of mass spectrometry data acquisition. Expected or unexpected modifications of the peptide sequences, polymorphisms, errors in databases, missed or non-specific cleavages, unusual fragmentation patterns, and single MS/MS spectra of multiple peptides of the same m/z are so many pitfalls for identification algorithms. A lot of research work has been carried out in recent years that yielded new strategies to handle a number of these issues. Multiple MS/MS identification algorithms are now available or have been theoretically described. The difficulty resides in choosing the most adapted method for each type of spectra being identified. This review presents an overview of the state-of-the-art bioinformatics approaches to the identification of proteins by MS/MS to help the reader doing the spade work of finding the right tools among the many possibilities offered.

Optimal construction of theoretical spectra for MS/MS spectra identification

We derive the optimal number of peaks (defined as the minimum number that provides the required efficiency of spectra identification) in the theoretical spectra as a function of (i) the experimental accuracy, sigma, of the measured ratio m/z; (ii) experimental spectrum density; (iii) size of the database; (iv) number of peaks in the theoretical spectra; and (v) types of ions that the peaks represent. We show that if theoretical spectra are constructed including b and y ions alone, then for sigma = 0.5, which is typical for high-throughput data, peptide chains of eight amino acids or longer can be identified based on the positions of peaks alone, at a rate of false identification below 1%. To discriminate between shorter peptides, additional (e.g., intensity-inferred) information is necessary. We derive the dependence of the probability of false identification on the number of peaks in the theoretical spectra and on the types of ions that the peaks represent. Our results suggest that the class of mass spectrum identification problems, for which more elaborate development of fragmentation rules (such as intensity model) is required, can be reduced to the problems that involve homologous peptides.

ProbIDtree: an automated software program capable of identifying multiple peptides from a single collision-induced dissociation spectrum collected by a tandem mass spectrometer

In MS/MS experiments with automated precursor ion, selection only a fraction of sequencing attempts lead to the successful identification of a peptide. A number of reasons may contribute to this situation. They include poor fragmentation of the selected precursor ion, the presence of modified residues in the peptide, mismatches with sequence databases, and frequently, the concurrent fragmentation of multiple precursors in the same CID attempt. Current database search engines are incapable of correctly assigning the sequences of multiple precursors to such spectra. We have developed a search engine, ProbIDtree, which can identify multiple peptides from a CID spectrum generated by the concurrent fragmentation of multiple precursor ions. This is achieved by iterative database searching in which the submitted spectra are generated by subtracting the fragment ions assigned to a tentatively matched peptide from the acquired spectrum and in which each match is assigned a tentative probability score. Tentatively matched peptides are organized in a tree structure from which their adjusted probability scores are calculated and used to determine the correct identifications. The results using MALDI-TOF-TOF MS/MS data demonstrate that multiple peptides can be effectively identified simultaneously with high confidence using ProbIDtree.

‘Top-down’ characterization of site-directed mutagenesis products of Staphylococcus aureus dihydroneopterin aldolase by multistage tandem mass spectrometry in a linear quadrupole ion trap

The gas-phase fragmentation reactions of a series of site-directed mutagenesis products of Staphylococcus aureus dihydroneopterin aldolase have been examined by multistage tandem mass spectrometry (MS/MS and MS(3)) in a linear quadrupole ion trap in order to explore the utility of this instrumentation for routine 'top-down' recombinant protein characterization. Following a rapid low resolution survey of the fragmentation behavior of the precursor ions from the wild type (WT) protein, selected charge states were subjected to detailed structural characterization by using high resolution 'zoom' and 'ultrazoom' resonance ejection MS/MS product ion scans. Dissociation of the [M + 18H](18+) charge state yielded a range of product ions from which extensive sequence information could be derived. In contrast, dissociation of the [M + 20H](20+) charge state resulted in a single dominant y(96) product ion formed by fragmentation between adjacent Ile/Gly residues, with only limited sequence coverage. Further extensive sequence information was readily obtained however, by MS(3) dissociation of this initial product. From the combined MS/MS and MS(3) spectra an overall sequence coverage of 66.9%, with fragmentation of 85 of the 127 amide bonds within the WT protein, was obtained. MS/MS and MS(3) of three of the four site-directed mutagenesis products (E29A), (Y61F) and (E81A) were found to yield essentially identical product ion spectra to the WT protein, indicating that these modifications had no significant influence on the fragmentation behavior. The specific site of modification could be unambiguously determined in each case by characterization of product ions resulting from fragmentation of amide bonds on either side of the mutation site. In contrast, MS/MS and MS(3) of the K107A mutant led to significantly different product ion spectra dominated by cleavages occurring N-terminal to proline, which restricted the ability to localize the modification site to within only an 8 amino acid region of the sequence. This work highlights the need for further studies to characterize the charge state, sequence and structural dependence to the low energy collision induced dissociation reactions of multiply protonated intact protein ions.

Investigation of gas phase ion structure for proline-containing b(2) ion

Unusual fragmentation was observed for doubly charged VPDPR in which cleavage C-terminal to proline and N-terminal to aspartic acid yielded b(2) (+ a(2))/y(3) complementary ions. This unique fragmentation is contradictory to trends previously established by statistical analysis of peptide tandem mass (MS/MS) spectra. Substitution of alanine for aspartic acid (i.e., VPAPR) did not change the fragmentation, indicating the cleavage was not directed by aspartic acid. Fragmentation patterns for VPAPR and V(NmA)APR (NmA = N-methyl alanine) were compared to determine whether conformational constraints from proline's cyclic side-chain contribute to b(2) ion formation. While both peptide sequences fragmented to yield b(2)/y(3) ions, only VPAPR produced a(2) ions, suggesting the VP b(2) ion is structurally different from the V(NmA) b(2) ion. Instead, the V(NmA) b(2) ion was accompanied by an ion corresponding to formal loss of 71. The suspected structural differences were confirmed by isolation and fragmentation of the respective b(2) ions (i.e., MS(3) spectra). Evidence supporting a diketopiperazine structure for the VP b(2) ion is reported. Fragmentation patterns for the VP b(2) ion and a synthetic VP diketopiperazine showed great similarity. N-terminal acetylation of VPAPR prevented the formation of the VP b(2) ion, presumably by blocking nucleophilic attack by the N-terminal amine on the carbonyl oxygen of the protonation site. Acetylation of the N-terminus for V(NmA)APR did not prevent the formation of the V(NmA) b(2) ion, indicating the V(NmA) b(2) ion has a structure, presumably that of an oxazolone, which requires no attack by the N-terminus for formation. Finally, high-resolution, accurate mass measurements determined that the V(NmA) (b(2)-71) ion results from losing a portion of valine from oxazolone V(NmA) b(2) ion, rather than cross-ring cleavage of the alternate diketopiperazine.

Accelerated on-column lysine derivatization and cysteine methylation by imidazole reaction in a deuterated environment for enhanced product ion analysis

The combination of separation techniques and mass spectrometry (MS) for peptide investigation allows superior sensitivity of detection and richer fragmentation data than available by direct MS analysis of a complex mixture. In this regard, liquid chromatography (LC) coupled to electrospray ionization (ESI) and matrix-assisted laser desorption/ionization (MALDI) MS have evolved as versatile analytical tools in proteomics. Very often, however, the product ion mass spectrum is either incomplete or overfilled with ions, thus making sequence analysis difficult. Here we report overall ion intensity improvement of C-terminal lysine-containing peptides from Lys-C digest by on-column derivatization of lysines with 2-methoxy-4,5-dihydro-1H-imidazole. The method is simple, fast and exhibits 100% efficiency of the reaction. Additionally, post-source decay carried out on derivatized peptides gave rise almost exclusively to y-series ion formation, at 100% sequence coverage and high intensity. The novelty of the method resides in the side reaction of this derivatization process, namely the methylation of cysteines. This facilitates the estimation of the disulfide bridge position in a protein and the fragmentation of cysteine-containing peptide fragments. Additionally, by using this derivatization procedure, the loss of peptides, their degradation and/or oxidation, usually occurring in digest alkylation procedures, is greatly minimized. The new on-column derivatization protocol is designed to be carried out on C18 Spin Tubes or Cleanup C18 Pipette Tips. We observed that use of buffered D2O solvent prevented unwanted oxidation and degradation reactions with respect to the stationary phase. This may be due to the fact that a deuteron is less polar than a proton, and thus the bonded silica stationary phase saturated with deuterons does not affect the reaction between epsilon-amino or cysteine thiol groups and 2-methoxy-4,5-dihydro-1H-imidazole. Complete tagging of the peptides by on-column reaction could be obtained when using D2O, as compared to water-based reaction. Methylation of cysteine residues was enhanced when beta-mercaptoethanol was added in the reactant solution.

'Information-Based-Acquisition' (IBA) technique with an ion-trap/time-of-flight mass spectrometer for high-throughput and reliable protein profiling

Highly complex protein mixtures can be analyzed after proteolysis using liquid chromatography/mass spectrometry (LC/MS). In an LC/MS run, intense peptide ions originating from high-abundance proteins are preferentially analyzed using tandem mass spectrometry (MS(2)), so obtaining the MS(2) spectra of peptide ions from low-abundance proteins is difficult even if such ions are detected. Furthermore, the MS(2) spectra may produce insufficient information to identify the peptides or proteins. To solve these problems, we have developed a real-time optimization technique for MS(2), called the Information-Based-Acquisition (IBA) system. In a preliminary LC/MS run, a few of the most intense ions detected in every MS spectrum are selected as precursors for MS(2) and their masses, charge states and retention times are automatically registered in an internal database. In the next run, a sample similar to that used in the first run is analyzed using database searching. Then, the ions registered in the database are excluded from the precursor ion selection to avoid duplicate MS(2) analyses. Furthermore, real-time de novo sequencing is performed just after obtaining the MS(2) spectrum, and an MS(3) spectrum is obtained for accurate peptide identification when the number of interpreted amino acids in the MS(2) spectrum is less than five. We applied the IBA system to a yeast cell lysate which is a typical crude sample, using a nanoLC/ion-trap time-of flight (IT/TOF) mass spectrometer, repeating the same LC/MS run five times. The obtained MS(2) and MS(3) spectra were analyzed by applying the Mascot (Matrix Science, Boston, MA, USA) search engine to identify proteins from the sequence database. The total number of identified proteins in five LC/MS runs was three times higher than that in the first run and the ion scores for peptide identification also significantly increased, by about 70%, when the MS(3) spectra were used, combined with the MS(2) spectra, before being subjected to Mascot analysis.

Deviation from the mobile proton model in amino-modified peptides: implications for multiple reaction monitoring analysis of peptides

The study of peptide fragmentation is important to the understanding of chemical processes occurring in the gas phase and the more practical concern of peptide identification for proteomic analysis. Using the mobile proton model as a framework, we explore the effect of amino-group modifications on peptide fragmentation. Three aldehydes are used to transform the peptides' primary amino groups into either a dimethylamino or a heterocyclic structure (five- or six-membered). The observed fragmentation patterns deviate strongly from those observed for the analogous underivatised peptides. In particular, the a1 ion is the base peak in most tandem mass spectra of the derivatised peptides. The a1 ion intensity depends strongly on the N-terminal amino acid, with tyrosine and phenylalanine having the strongest enhancement. Despite the change in fragmentation patterns of the derivatised peptides, they still provide high-quality tandem mass spectra that, in many cases, are more amenable to database searching than the spectra of underivatised peptides. In addition, the reliable presence of the a1 ion facilitates rapid quantitative measurements using the multiple reaction monitoring approach.

Improved classification of mass spectrometry database search results using newer machine learning approaches

Manual analysis of mass spectrometry data is a current bottleneck in high throughput proteomics. In particular, the need to manually validate the results of mass spectrometry database searching algorithms can be prohibitively time-consuming. Development of software tools that attempt to quantify the confidence in the assignment of a protein or peptide identity to a mass spectrum is an area of active interest. We sought to extend work in this area by investigating the potential of recent machine learning algorithms to improve the accuracy of these approaches and as a flexible framework for accommodating new data features. Specifically we demonstrated the ability of boosting and random forest approaches to improve the discrimination of true hits from false positive identifications in the results of mass spectrometry database search engines compared with thresholding and other machine learning approaches. We accommodated additional attributes obtainable from database search results, including a factor addressing proton mobility. Performance was evaluated using publically available electrospray data and a new collection of MALDI data generated from purified human reference proteins.

Protein and peptide identification algorithms using MS for use in high-throughput, automated pipelines

Current proteomics experiments can generate vast quantities of data very quickly, but this has not been matched by data analysis capabilities. Although there have been a number of recent reviews covering various aspects of peptide and protein identification methods using MS, comparisons of which methods are either the most appropriate for, or the most effective at, their proposed tasks are not readily available. As the need for high-throughput, automated peptide and protein identification systems increases, the creators of such pipelines need to be able to choose algorithms that are going to perform well both in terms of accuracy and computational efficiency. This article therefore provides a review of the currently available core algorithms for PMF, database searching using MS/MS, sequence tag searches and de novo sequencing. We also assess the relative performances of a number of these algorithms. As there is limited reporting of such information in the literature, we conclude that there is a need for the adoption of a system of standardised reporting on the performance of new peptide and protein identification algorithms, based upon freely available datasets. We go on to present our initial suggestions for the format and content of these datasets.

An evaluation, comparison, and accurate benchmarking of several publicly available MS/MS search algorithms: sensitivity and specificity analysis

MS/MS and associated database search algorithms are essential proteomic tools for identifying peptides. Due to their widespread use, it is now time to perform a systematic analysis of the various algorithms currently in use. Using blood specimens used in the HUPO Plasma Proteome Project, we have evaluated five search algorithms with respect to their sensitivity and specificity, and have also accurately benchmarked them based on specified false-positive (FP) rates. Spectrum Mill and SEQUEST performed well in terms of sensitivity, but were inferior to MASCOT, X!Tandem, and Sonar in terms of specificity. Overall, MASCOT, a probabilistic search algorithm, correctly identified most peptides based on a specified FP rate. The rescoring algorithm, PeptideProphet, enhanced the overall performance of the SEQUEST algorithm, as well as provided predictable FP error rates. Ideally, score thresholds should be calculated for each peptide spectrum or minimally, derived from a reversed-sequence search as demonstrated in this study based on a validated data set. The availability of open-source search algorithms, such as X!Tandem, makes it feasible to further improve the validation process (manual or automatic) on the basis of "consensus scoring", i.e., the use of multiple (at least two) search algorithms to reduce the number of FPs. complement.

Investigation of neutral loss during collision-induced dissociation of peptide ions

MS/MS fragmentation of peptides is dominated by overlapping b and y ion series. However, alternative fragmentation possibilities exist, including neutral loss. A database was generated containing 8400 MS/MS spectra of tryptic peptides assigned with high probability to an amino acid sequence (true positives) and a set of certified false (true negative) assignments for analysis of the amino terminus. A similar database was created for analysis of neutral loss at the carboxy termini using a data set of chymotryptic peptides. The analysis demonstrated that the presence of an internal basic residue, limiting proton mobility, has a profound effect on neutral loss. Peptides with fully mobile protons demonstrated minimal neutral loss, with the exception of amide bonds with proline on the carboxy terminal side, which created an intense neutral loss peak. In contrast, peptides with partial proton mobility contained many amino acids on either side of the amide bond associated with a strong neutral loss peak. Most notable among these was proline on the carboxy terminal side of an amide bond and aspartic acid on the amino terminal side of a bond. All results were found to be consistent for doubly and triply charged peptides and after adjustment for pairings across the amide bonds with particularly labile residues. The carboxy terminal of chymotryptic peptides also demonstrated significant neutral loss events associated with numerous amino acid residues. Clarification of the rules that govern neutral loss, when incorporated into analysis software, will improve our ability to correctly assign spectra to peptide sequences.

Selective identification and quantitative analysis of methionine containing peptides by charge derivatization and tandem mass spectrometry

To enable the development of a tandem mass spectrometry (MS/MS) based methodology for selective protein identification and differential quantitative analysis, a novel derivatization strategy is proposed, based on the formation of a "fixed-charge" sulfonium ion on the side-chain of a methionine amino acid residue contained within a protein or peptide of interest. The gas-phase fragmentation behavior of these side chain fixed charge sulfonium ion containing peptides is observed to result in exclusive loss of the derivatized side chain and the formation of a single characteristic product ion, independently of charge state or amino acid composition. Thus, fixed charge containing peptide ions may be selectively identified from complex mixtures, for example, by selective neutral loss scan mode MS/MS methods. Further structural interrogation of identified peptide ions may be achieved by subjecting the characteristic MS/MS product ion to multistage MS/MS (MS3) in a quadrupole ion trap mass spectrometer, or by energy resolved "pseudo" MS3 in a triple quadrupole mass spectrometer. The general principles underlying this fixed charge derivatization approach are demonstrated here by MS/MS, MS3 and "pseudo" MS3 analysis of side chain fixed-charge sulfonium ion derivatives of peptides containing methionine formed by reaction with phenacylbromide. Incorporation of "light" and "heavy" isotopically encoded labels into the fixed-charge derivatives facilitates the application of this method to the quantitative analysis of differential protein expression, via measurement of the relative abundances of the neutral loss product ions generated by dissociation of the light and heavy labeled peptide ions. This approach, termed "selective extraction of labeled entities by charge derivatization and tandem mass spectrometry" (SELECT), thereby offers the potential for significantly improved sensitivity and selectivity for the identification and quantitative analysis of peptides or proteins containing selected structural features, without requirement for extensive fractionation or otherwise enrichment from a complex mixture prior to analysis.

Computational investigation and hydrogen/deuterium exchange of the fixed charge derivative tris(2,4,6-trimethoxyphenyl) phosphonium: implications for the aspartic acid cleavage mechanism

Aspartic acid (Asp)-containing peptides with the fixed charge derivative tris(2,4,6trimethoxyphenyl) phosphonium (tTMP-P+) were explored computationally and experimentally by hydrogen/deuterium (H/D) exchange and by fragmentation studies to probe the phenomenon of selective cleavage C-terminal to Asp in the absence of a "mobile" proton. Ab initio modeling of the tTMP-P+ electrostatic potential shows that the positive charge is distributed on the phosphonium group and therefore is not initiating or directing fragmentation as would a "mobile" proton. Geometry optimizations and vibrational analyses of different Asp conformations show that the Asp structure with a hydrogen bond between the side-chain hydroxy and backbone carbonyl lies 2.8 kcal/mol above the lowest energy conformer. In reactions with D2O, the phosphonium-derived doubly charged peptide (H+)P+LDIFSDF rapidly exchanges all 12 of its exchangeable hydrogens for deuterium and also displays a nonexchanging population. With no added proton, P+LDIFSDF exchanges a maximum of 4 of 11 exchangeable hydrogens for deuterium. No exchange is observed when all acidic groups are converted to the corresponding methyl esters. Together, these H/D exchange results indicate that the acidic hydrogens are "mobile locally" because they are able to participate in exchange even in the absence of an added proton. Fragmentation of two distinct (H+)P+LDIFSDF ion populations shows that the nonexchanging population displays selective cleavage, whereas the exchanging population fragments more evenly across the peptide backbone. This result indicates that H/D exchange can sometimes distinguish between and provide a means of separation of different protonation motifs and that these protonation motifs can have an effect on the fragmentation.

Fragmentation pathways of protonated peptides

The fragmentation pathways of protonated peptides are reviewed in the present paper paying special attention to classification of the known fragmentation channels into a simple hierarchy defined according to the chemistry involved. It is shown that the 'mobile proton' model of peptide fragmentation can be used to understand the MS/MS spectra of protonated peptides only in a qualitative manner rationalizing differences observed for low-energy collision induced dissociation of peptide ions having or lacking a mobile proton. To overcome this limitation, a deeper understanding of the dissociation chemistry of protonated peptides is needed. To this end use of the 'pathways in competition' (PIC) model that involves a detailed energetic and kinetic characterization of the major peptide fragmentation pathways (PFPs) is proposed. The known PFPs are described in detail including all the pre-dissociation, dissociation, and post-dissociation events. It is our hope that studies to further extend PIC will lead to semi-quantative understanding of the MS/MS spectra of protonated peptides which could be used to develop refined bioinformatics algorithms for MS/MS based proteomics. Experimental and computational data on the fragmentation of protonated peptides are reevaluated from the point of view of the PIC model considering the mechanism, energetics, and kinetics of the major PFPs. Evidence proving semi-quantitative predictability of some of the ion intensity relationships (IIRs) of the MS/MS spectra of protonated peptides is presented.

Microwave-assisted enzyme-catalyzed reactions in various solvent systems

The work describes the accelerated enzymatic digestion of several proteins in various solvent systems under microwave irradiation. The tryptic fragments of the proteins were analyzed by matrix-assisted laser desorption/ionization mass spectrometry. Under the influence of rapid microwave heating, these enzymatic reactions can proceed in a solvent such as chloroform, which, under traditional digestion conditions, renders the enzyme inactive. The digestion efficiencies and sequence coverages were increased when the trypsin digestions occurred in acetonitrile-, methanol- and chloroform-containing solutions that were heated under microwave irradiation for 10 min using a commercial microwave applicator. The percentage of the protein digested under microwave irradiation increased with the relative acetonitrile content, but decreased as the methanol content was increased. These observations suggest that acetonitrile does not deactivate the enzyme during the irradiation period; in contrast, methanol does deactivate it. In all cases, the digestion efficiencies under microwave irradiation exceed those under conventional conditions.

Proteomics strategies for protein identification

The information from genome sequencing provides new approaches for systems-wide understanding of protein networks and cellular function. DNA microarray technologies have advanced to the point where nearly complete monitoring of gene expression is feasible in several organisms. An equally important goal is to comprehensive survey cellular proteomes and profile protein changes under different cellular states. This presents a complex analytical problem, due to the chemical variability between proteins and peptides. Here, we discuss strategies to improve accuracy and sensitivity of peptide identification, distinguish represented protein isoforms, and quantify relative changes in protein abundance.

Bioinformatic methods to exploit mass spectrometric data for proteomic applications

The new technologies in mass spectrometric analysis of peptides and proteins necessary to accommodate proteomics-scale analyses require, in turn, concomitant development of informatics technologies suitable for very large-scale data handling and analysis. This chapter focuses on the data analysis tools available to the community for analysis of mass spectrometric proteomics data. Different database searching strategies are discussed for peptide and protein identification, and approaches and tools available for comparative quantitative analysis of samples are outlined.

Neighbouring group processes in the deamination of protonated phenylalanine derivatives

The gas-phase fragmentation of protonated phenylalanine and a series of its derivatives (tyrosine, 4-methylphenylalanine, 4-aminophenylalanine, 4-methoxyphenylalanine, 4-tert-butylphenylalanine, 4-fluorophenylalanine, 4-chlorophenylalanine, 4-bromophenylalanine, 4-iodophenylalanine, 4-cyanophenylalanine, 4-nitrophenylalanine, 3-fluorophenylalanine, and 3,4-dichlorophenylalanine) were examined using a combination of low energy CID in a quadrupole ion trap mass spectrometer as well as DFT calculations and RRKM modelling. In particular, the relationship between the electron-donating ability of the substituent and the competitive losses of H2O + CO and NH3 were explored through the application of the Hammett equation. It was found that electron-donating substituents promote the loss of NH3, while electron-withdrawing substituents suppress the loss of NH3 and favour the H2O + CO loss fragmentation channel instead. These observations are consistent with a neighbouring group pathway operating for the loss of NH3. Molecular orbital calculation (at the B3LYP/6-31+G(d,p) level of theory) were also performed for a range of derivatives to compare the relative transition state energy barriers for three competing mechanisms: (i) the combined loss of H2O + CO, which is triggered by an initial intramolecular proton transfer from the ammonium group to hydroxyl OH, followed by the combined loss of H2O and CO to form an immonium ion; (ii) loss of NH3 via an aryl assisted neighbouring group pathway to yield a phenonium ion; (iii) loss of NH3 via a 1,2-hydride migration process, which results in the formation of a benzyl cation. The relative energy barriers for H2O + CO loss remain nearly constant, while that for both NH3 pathways increase as the substituent moves from electron-donating to electron-withdrawing. The relative transition state energy for loss of NH3 via the aryl assisted neighbouring group pathway is always lower than that of the 1,2-hydride migration process. RRKM modelling of the DFT predicted barrier heights suggest that the rate constants for H2O + CO loss are insensitive to the substituent on the ring, while the NH3 loss channels are greatly affected by the substituent. These theoretical results are consistent with the experimental observation of the relative yields of the competing fragmentation channels. Finally, comparisons with published gas phase and condensed phase studies on related systems are made.

A case study of de novo sequence analysis of N-sulfonated peptides by MALDI TOF/TOF mass spectrometry

The simplicity and sensitivity of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry have increased its application in recent years. The most common method of "peptide mass fingerprint" analysis often does not provide robust identification. Additional sequence information, obtained by post-source decay or collision induced dissociation, provides additional constraints for database searches. However, de novo sequencing by mass spectrometry is not yet common practice, most likely because of the difficulties associated with the interpretation of high and low energy CID spectra. Success with this type of sequencing requires full sequence coverage and demands better quality spectra than those typically used for data base searching. In this report we show that full-length de novo sequencing is possible using MALDI TOF/TOF analysis. The interpretation of MS/MS data is facilitated by N-terminal sulfonation after protection of lysine side chains (Keough et al., Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 7131-7136). Reliable de novo sequence analysis has been obtained using sub-picomol quantities of peptides and peptide sequences of up to 16 amino acid residues in length have been determined. The simple, predictable fragmentation pattern allows routine de novo interpretation, either manually or using software. Characterization of the complete primary structure of a peptide is often hindered because of differences in fragmentation efficiencies and in specific fragmentation patterns for different peptides. These differences are controlled by various structural parameters including the nature of the residues present. The influence of the presence of internal Pro, acidic and basic residues on the TOF/TOF fragmentation pattern will be discussed, both for underivatized and guanidinated/sulfonated peptides.

Experimental Peptide Identification Repository (EPIR)

LC MS/MS has become an established technology in proteomic studies, and with the maturation of the technology the bottleneck has shifted from data generation to data validation and mining. To address this bottleneck we developed Experimental Peptide Identification Repository (EPIR), which is an integrated software platform for storage, validation, and mining of LC MS/MS-derived peptide evidence. EPIR is a cumulative data repository where precursor ions are linked to peptide assignments and protein associations returned by a search engine (e.g. Mascot, Sequest, or PepSea). Any number of datasets can be parsed into EPIR and subsequently validated and mined using a set of software modules that overlay the database. These include a peptide validation module, a protein grouping module, a generic module for extracting quantitative data, a comparative module, and additional modules for extracting statistical information. In the present study, the utility of EPIR and associated software tools is demonstrated on LC MS/MS data derived from a set of model proteins and complex protein mixtures derived from MCF-7 breast cancer cells. Emphasis is placed on the key strengths of EPIR, including the ability to validate and mine multiple combined datasets, and presentation of protein-level evidence in concise, nonredundant protein groups that are based on shared peptide evidence.

Effects of single amino acid substitution on the collision-induced dissociation of intact protein ions: Turkey ovomucoid third domain

Expanded understanding of the factors that direct polypeptide ion fragmentation can lead to improved specificity in the use of tandem mass spectrometry for the identification and characterization of proteins. Like the fragmentation of peptide cations, the dissociation of whole protein cations shows several preferred cleavages, the likelihood for which is parent ion charge dependent. While such cleavages are often observed, they are far from universally observed, despite the presence of the residues known to promote them. Furthermore, cleavages at residues not noted to be common in a variety of proteins can be dominant for a particular protein or protein ion charge state. Motivated by the ability to study a small protein, turkey ovomucoid third domain, for which a variety of single amino acid variants are available, the effects of changing the identity of one amino acid in the protein sequence on its dissociation behavior were examined. In particular, changes in amino acids associated with C-terminal aspartic acid cleavage and N-terminal proline cleavage were emphasized. Consistent with previous studies, the product ion spectra were found to be dependent upon the parent ion charge state. Furthermore, the fraction of possible C-terminal aspartic acid cleavages observed to occur for this protein was significantly larger than the fraction of possible N-terminal proline cleavages. In fact, very little N-terminal proline cleavage was noted for the wild-type protein despite the presence of three proline residues in the protein. The addition/removal of proline and aspartic acids was studied along with changes in selected residues adjacent to proline residues. Evidence for inhibition of proline cleavage by the presence of nearby basic residues was noted, particularly if the basic residue was likely to be protonated.

Influence of basic residue content on fragment ion peak intensities in low-energy collision-induced dissociation spectra of peptides

The primary utility of trypsin digestion in proteomics is that it cleaves proteins at predictable locations, but it is also notable for yielding peptides that terminate in basic arginine and lysine residues. Tryptic peptides fragment in ion trap tandem mass spectrometry to produce prominent C-terminal y series ions. Alternative proteolytic digests may produce peptides that do not follow these rules. In this study, we examine 2568 peptides generated through proteinase K digestion, a technique that produces a greater diversity of basic residue content in peptides. We show that the position of basic residues within peptides influences the peak intensities of b and y series ions; a basic residue near the N-terminus of a peptide can lead to prominent b series peaks rather than the intense y series peaks associated with tryptic peptides. The effects of presence and position for arginine, lysine, and histidine are explored separately and in combination. Arg shows the most dominant effects followed by His and then by Lys. Fragment ions containing basic residues produce more intense peaks than those without basic residues. Doubly charged precursor ions have generally been modeled as producing only singly charged fragment ions, but fragment ions that contain two basic residues may accept both protons during fragmentation. By characterizing the influence of basic residues on gas-phase fragmentation of peptides, this research makes possible more accurate fragmentation models for peptide identification algorithms.

MS(n) characterization of protonated cyclic peptides and metal complexes

MS(n) experiments involving low energy collisionally activated dissociation (CAD) in a quadrupole ion trap were used to characterize the fragmentation of alkali, alkaline earth and transition metal complexes of five cyclic peptides, and the results were compared with those obtained for protonated cyclic peptides. Complexes with metal ions produced enhanced abundances of the most diagnostic fragments for elucidating the primary structures. For cyclosporin A, nickel and lithium complexes gave additional sequence information compared with the protonated peptide. For depsipeptides, sodium and lead complexes were superior to the protonated peptide or other metal complexes for sequencing residues, and CAD of the lead complexes led to preferential cleavage of two residues at a time. For cyclic lipopeptides, complexes with silver, nickel and strontium ions provided enhanced abundances of key fragment ions.

Analysis, statistical validation and dissemination of large-scale proteomics datasets generated by tandem MS

Tandem mass spectrometry has been used increasingly for high-throughput analysis of complex protein samples. A major challenge lies in the consistent, objective and transparent analysis of the large amounts of data generated by such experiments and in their dissemination and publication. Here, we review currently available computational tools and discuss the need for statistical criteria in the analysis of large proteomics datasets.