Energetics and Dynamics of the Fragmentation Reactions of Protonated Peptides Containing Methionine Sulfoxide or Aspartic Acid via Energy- and Time-Resolved Surface Induced Dissociation

Decoding the impact of neighboring amino acids on ESI-MS intensity output through deep learning

Peptide-level quantification using mass spectrometry (MS) is no trivial task as the physicochemical properties affect both response and detectability. The specific amino acid (AA) sequence affects these properties, however the connection between sequence and intensity output remains poorly understood. In this work, we explore combinations of amino acid pairs (i.e., dimer motifs) to determine a potential relationship between the local amino acid environment and MS1 intensity. For this purpose, a deep learning (DL) model, consisting of an encoder-decoder with an attention mechanism, was built. The attention mechanism allowed to identify the most relevant motifs. Specific patterns were consistently observed where a bulky/aromatic and hydrophobic AA followed by a cationic AA as well as consecutive bulky/aromatic and hydrophobic AAs were found important for the prediction of the MS1 intensity. Correlating attention weights to mean MS1 intensities revealed that some important motifs, particularly containing Trp, His, and Cys, were linked with low responding peptides whereas motifs containing Lys and most bulky hydrophobic AAs were often associated with high responding peptides. Moreover, Asn-Gly was associated with low response. The model predicts MS1 response with a mean average percentage error of ∼11 % and a Pearson correlation coefficient of ∼0.64. While dimer representation of peptide sequences did not improve predictive capacity compared to single AA representation in earlier work, this work adds valuable insight for a better understanding of peptide response in MS analysis. SIGNIFICANCE: Mass spectrometry is not inherently quantitative, and the response of a compound relies not only on its concentration but also on the molecular composition. For mass spectrometry-based analysis of peptides, such as in bottom-up proteomics, this directly implies that the response cannot be used directly to quantify individual peptides. Moreover, the dependency of the response on the amino acid sequence of individual peptides remains poorly understood. Using a deep learning model based on a recurrent neural network with an attention mechanism, we here investigate how the presence of dimer motifs within a peptide affects the MS1 response through the analysis of intended equimolar peptide pools comprising almost 200,000 unique peptides in total. Not only do we identify certain dimer classes and specific dimers that substantially affect the MS1 response, but the model is also able to predict peptide intensity with low error rates within the independent test subset. The findings not only improve our understanding of the link between sequence and response for peptides but also highlight the potential of utilizing deep learning for developing methods allowing for absolute, label-free peptide quantification.

Transformation of [M + 2H](2+) Peptide Cations to [M - H](+), [M + H + O](+), and M(+•) Cations via Ion/Ion Reactions: Reagent Anions Derived from Persulfate

The gas-phase oxidation of doubly protonated peptides is demonstrated here using ion/ion reactions with a suite of reagents derived from persulfate. Intact persulfate anion (HS2O8(-)), peroxymonosulfate anion (HSO5(-)), and sulfate radical anion (SO4(-•)) are all either observed directly upon negative nanoelectrospray ionization (nESI) or easily obtained via beam-type collisional activation of persulfate into the mass spectrometer. Ion/ion reactions between each of these reagents and doubly protonated peptides result in the formation of a long-lived complex. Collisional activation of the complex containing a peroxymonosulfate anion results in oxygen transfer from the reagent to the peptide to generate the [M + H + O](+) species. Activation of the complex containing intact persulfate anion either results in oxygen transfer to generate the [M + H + O](+) species or abstraction of two hydrogen atoms and a proton to generate the [M - H](+) species. Activation of the complex containing sulfate radical anion results in abstraction of one hydrogen atom and a proton to form the peptide radical cation, [M](+•). This suite of reagents allows for the facile transformation of the multiply protonated peptides obtained via nESI into a variety of oxidized species capable of providing complementary information about the sequence and structure of the peptide.

Surface-induced dissociation: a unique tool for studying energetics and kinetics of the gas-phase fragmentation of large ions

Surface-induced dissociation (SID) is a valuable tool for investigating the activation and dissociation of large ions in tandem mass spectrometry. This account summarizes key findings from studies of the energetics and mechanisms of complex ion dissociation in which SID experiments were combined with Rice-Ramsperger-Kassel-Marcus modeling of the experimental data. These studies used time- and collision-energy-resolved SID experiments and SID combined with resonant ejection of selected fragment ions on a specially designed Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometer. Fast-ion activation by collision with a surface combined with the long and variable timescale of FT-ICR mass spectrometry is perfectly suited to studying the energetics and dynamics of complex ion dissociation in the gas phase. Modeling of time- and collision-energy-resolved SID enables the accurate determination of energy and entropy effects in the dissociation process. It has been demonstrated that entropy effects play an important role in determining the dissociation rates of both covalent and noncovalent bonds in large gaseous ions. SID studies have provided important insights on the competition between charge-directed and charge-remote fragmentation in even-electron peptide ions and the role of the charge and radical site on the energetics of the dissociation of odd-electron peptide ions. Furthermore, this work examined factors that affect the strength of noncovalent binding, as well as the competition between covalent and noncovalent bond cleavages and between proton and electron transfer in model systems. Finally, SID studies have been used to understand the factors affecting nucleation and growth of clusters in solution and in the gas phase.

Oxidation of methionine residues in polypeptide ions via gas-phase ion/ion chemistry

The gas-phase oxidation of methionine residues is demonstrated here using ion/ion reactions with periodate anions. Periodate anions are observed to attach in varying degrees to all polypeptide ions irrespective of amino acid composition. Direct proton transfer yielding a charge-reduced peptide ion is also observed. In the case of methionine and, to a much lesser degree, tryptophan-containing peptide ions, collisional activation of the complex ion generated by periodate attachment yields an oxidized peptide product (i.e., [M + H + O](+)), in addition to periodic acid detachment. Detachment of periodic acid takes place exclusively for peptides that do not contain either a methionine or tryptophan side chain. In the case of methionine-containing peptides, the [M + H + O](+) product is observed at a much greater abundance than the proton transfer product (viz., [M + H](+)). Collisional activation of oxidized Met-containing peptides yields a signature loss of 64 Da from the precursor and/or product ions. This unique loss corresponds to the ejection of methanesulfenic acid from the oxidized methionine side chain and is commonly used in solution-phase proteomics studies to determine the presence of oxidized methionine residues. The present work shows that periodate anions can be used to 'label' methionine residues in polypeptides in the gas phase. The selectivity of the periodate anion for the methionine side chain suggests several applications including identification and location of methionine residues in sequencing applications.

Modelling of the gas-phase phosphate group loss and rearrangement in phosphorylated peptides

The gas-phase dissociation of phosphorylated peptides was modelled using a combination of quantum mechanics and the Rice-Ramsperger-Kassel-Marcus theory. Potential energy surfaces and unimolecular reaction rates for several low-energy fragmentation and rearrangement pathways were estimated, and a general mechanism was proposed. The neutral loss of the phosphoric acid was mainly an outcome of the intramolecular nucleophilic substitution mechanism. The mechanism involves a nucleophilic attack of the phosphorylated amino acid N-terminal carbonyl oxygen on β-carbon, yielding a cyclic five-membered oxazoline product ion. Regardless of the proton mobility, the pathway was charge directed either by a mobile proton or by a positively charged side chain of some basic residue. Although the mechanistic aspects of the phosphate loss are not influenced by the proton mobility environment, it does affect ion abundances. Results suggest that under the mobile proton environment, the interplay between phosphoric acid neutral loss product ion and backbone cleavage fragments should occur. On the other hand, when proton mobility is limited, neutral loss product ion may predominate. The fragmentation dynamics of phosphoserine versus phosphothreonine containing peptides suggests that H(3)PO(4) neutral loss from phosphothreonine containing peptides is less abundant than that from their phosphoserine containing analogs. During the low-energy CID of phosphorylated peptides in the millisecond time range, typical for ion trap instruments, a phosphate group rearrangement may happen, resulting in an interchange between the phosphorylated and the hydroxylated residues. Unimolecular dissociation rate constants imply the low abundance of such scrambled product ions.

The relative charge ratio between C and N atoms in amide bond acts as a key factor to determine peptide fragment efficiency in different charge states

The influence of charge state on the peptide dissociation behavior in tandem mass spectrometry (MS/MS) is worthy of discussion. Comparative studies of singly- and doubly-protonated peptide molecules are performed to explore the effect and mechanism of charge state on peptide fragmentation. In view of the charge-directed cleavage of protonated peptides described in the mobile proton model, radiolytic oxidation was applied to change the charge distribution of peptides but retain the sequence. Experimental studies of collision energy-dependent fragmentation efficiencies coupled with quantum chemical calculations indicated that the cleavage of ARRA and its side-chain oxidation products with oxygen atoms added followed a trend that doubly-protonated peptides fragment more easily than singly-protonated forms, while the oxidation product with the guanidine group deleted showed the opposite trend. By analyzing the charge distribution around the amide bonds, we found that the relative charge ratios between C and N atoms (Q(C)/Q(N)) in the amide bonds provided a reasonable explanation for peptide fragmentation efficiencies. An increase of the Q(C)/Q(N) value of the amide bond means that a peptide fragments more easily, and vice versa. The results described in this paper provide an experimental and calculation strategy for predicting peptide fragmentation efficiency.

Role of 2-oxo and 2-thioxo modifications on the proton affinity of histidine and fragmentation reactions of protonated histidine

A combination of electrospray ionisation (ESI), multistage and high-resolution mass spectrometry experiments was used to compare the gas-phase chemistry of the amino acids histidine (1), 2-oxo-histidine (2), and 2-thioxo-histidine (3). Collision-induced dissociation (CID) of all three different proton-bound heterodimers of these amino acids led to the relative gas-phase proton affinity order of: histidine >2-thioxo-histidine >2-oxo-histidine. Density functional theory (DFT) calculations confirm this order, with the lower proton affinities of the oxidised histidine derivatives arising from their ability to adopt the more stable keto/thioketo tautomeric forms. All protonated amino acids predominately fragment via the combined loss of H(2)O and CO to yield a(1) ions. Protonated 2 and 3 also undergo other small molecule losses including NH(3) and the imine HN=CHCO(2)H. The observed differences in the fragmentation pathways are rationalised through DFT calculations, which reveal that while modification of histidine via the introduction of the oxygen atom in 2 or the sulfur atom in 3 does not affect the barriers against the loss of H(2)O+CO, barriers against the losses of NH(3) and HN=CHCO(2)H are lowered relative to protonated histidine.

Gas-phase acidities of cysteine-polyglycine peptides: the effect of the cysteine position

The sequence and conformational effects on the gas-phase acidities of peptides have been studied by using two pairs of isomeric cysteine-polyglycine peptides, CysGly(3,4)NH(2) and Gly(3,4)CysNH(2). The extended Cooks kinetic method was employed to determine the gas-phase acidities using a triple quadrupole mass spectrometer with an electrospray ionization source. The ion activation was achieved via collision-induced dissociation experiments. The deprotonation enthalpies (Delta(acid)H) were determined to be 323.9 +/- 2.5 kcal/mol (CysGly(3)NH(2)), 319.2 +/- 2.3 kcal/mol (CysGly(4)NH(2)), 333.8 +/- 2.1 kcal/mol (Gly(3)CysNH(2)), and 321.9 +/- 2.8 kcal/mol (Gly(4)CysNH(2)), respectively. The corresponding deprotonation entropies (Delta(acid)S) of the peptides were estimated. The gas-phase acidities (Delta(acid)G) were derived to be 318.4 +/- 2.5 kcal/mol (CysGly(3)NH(2)), 314.9 +/- 2.3 kcal/mol (CysGly(4)NH(2)), 327.5 +/- 2.1 kcal/mol (Gly(3)CysNH(2)), and 317.4 +/- 2.8 kcal/mol (Gly(4)CysNH(2)), respectively. Conformations and energetic information of the neutral and anionic peptides were calculated through simulated annealing (Tripos), geometry optimization (AM1), and single point energy calculations (B3LYP/6-31+G(d)), respectively. Both neutral and deprotonated peptides adopt many possible conformations of similar energies. All neutral peptides are mainly random coils. The two C-cysteine anionic peptides, Gly(3,4)(Cys-H)(-)NH(2), are also random coils. The two N-cysteine anionic peptides, (Cys-H)(-)Gly(3,4)NH(2), may exist in both random coils and stretched helices. The two N-cysteine peptides, CysGly(3)NH(2) and CysGly(4)NH(2), are significantly more acidic than the corresponding C-terminal cysteine ones, Gly(3)CysNH(2) and Gly(4)CysNH(2). The stronger acidities of the former may come from the greater stability of the thiolate anion resulting from the interaction with the helix-macrodipole, in addition to the hydrogen bonding interactions.

Comparison of collision- versus electron-induced dissociation of Pt(II) ternary complexes of histidine- and methionine-containing peptides

Incubation of the histidine-containing peptides (GH, HG, GGH, GHG, HGG) and methionine-containing peptides (GM, MG, GGM, GMG, MGG) with the platinum complexes [Pt(terpy)Cl](+) (A) and [Pt(dien)Cl](+) (B) followed by electrospray ionisation (ESI) led to a number of singly and doubly charged ternary platinum peptide complexes, including [Pt(L)M](2+) and [Pt(L)M-H](+) (where L = the ligand terpy or dien; M is a peptide). Each of the [Pt(L)M](2+) complexes was subjected to electron capture dissociation (ECD), collision-induced dissociation (CID) and electron-induced dissociation (EID), while each of the [Pt(L)M-H](+) complexes was subjected to CID and EID. Results from ECD suggest that the free electron is captured by the metal ion thus weakening the bonds to its ligands. In the case of the ligand terpy, which binds more strongly than dien, this weakening leads to the loss of the peptide. The minor products in the ECD spectra of [Pt(terpy)M](2+) complexes do show fragmentation along the peptide backbone, but the ions observed are of the a-, b-, and y-type. For the complexes with methionine-containing peptides, a marker ion, [Pt(L)SCH(3)](+), was found which is indicative of binding of Pt to the methionine side chain. For the histidine-containing peptides, an ion containing platinum, the auxiliary ligand, and the histidine imine was observed in many instances, thus indicating the binding of the histidine side chain to the metal, but other modes of Pt coordination (N-terminus) were also found to be competitive. These findings are consistent with a recent finding (Sze et al. J. Biol. Inorg. Chem. 2009; 14: 163) that Pt occupies the methionine-rich copper(I)-binding site rather than histidine-rich copper(II)-binding site in the CopC protein.

Fragmentation mechanisms of oxidized peptides elucidated by SID, RRKM modeling, and molecular dynamics

The gas-phase fragmentation reactions of singly charged angiotensin II (AngII, DR(+)VYIHPF) and the ozonolysis products AngII+O (DR(+)VY*IHPF), AngII+3O (DR(+)VYIH*PF), and AngII+4O (DR(+)VY*IH*PF) were studied using SID FT-ICR mass spectrometry, RRKM modeling, and molecular dynamics. Oxidation of Tyr (AngII+O) leads to a low-energy charge-remote selective fragmentation channel resulting in the b(4)+O fragment ion. Modification of His (AngII+3O and AngII+4O) leads to a series of new selective dissociation channels. For AngII+3O and AngII+4O, the formation of [MH+3O](+)-45 and [MH+3O](+)-71 are driven by charge-remote processes while it is suggested that b(5) and [MH+3O](+)-88 fragments are a result of charge-directed reactions. Energy-resolved SID experiments and RRKM modeling provide threshold energies and activation entropies for the lowest energy fragmentation channel for each of the parent ions. Fragmentation of the ozonolysis products was found to be controlled by entropic effects. Mechanisms are proposed for each of the new dissociation pathways based on the energies and entropies of activation and parent ion conformations sampled using molecular dynamics.

Correct identification of oxidized histidine residues using electron-transfer dissociation

Oxidative modification to the side chain of histidine can noticeably change the collision-induced dissociation (CID) pathways of peptides containing this oxidized residue. In cases where an oxidized peptide consists two or more isomers differing only in the site of modification, oxidation to histidine usually causes the other oxidized sites to be mis-assigned in CID spectra. These spectral misassignments can sometimes be avoided by using multiple stages of MS/MS (MS(n)) or via specially optimized liquid chromatographic separation conditions. In this manuscript, we demonstrate that these misassignments can be more readily and easily avoided by using electron-transfer dissociation (ETD) to dissociate the oxidized peptides. Furthermore, we find that the relative insensitivity of ETD to side-chain chemistry allows the extent of oxidative modification to be determined readily for peptide isomers having more than one site of oxidation. The current results along with previous studies of oxidized peptides suggest that ETD is probably a better technique than CID for obtaining correct sequence and modification information for oxidized peptides.

Gas-phase peptide fragmentation: how understanding the fundamentals provides a springboard to developing new chemistry and novel proteomic tools

This tutorial provides an overview of the evolution of some of the key concepts in the gas-phase fragmentation of different classes of peptide ions under various conditions [e.g. collision-induced dissociation (CID) and electron transfer dissociation (ETD)], and then demonstrates how these concepts can be used to develop new methods. For example, an understanding of the role of the mobile proton and neighboring group interactions in the fragmentation reactions of protonated peptides has led to the design of the 'SELECT' method. For ETD, a model based on the Landau-Zener theory reveals the role of both thermodynamic and geometric effects in the electron transfer from polyatomic reagent anions to multiply protonated peptides, and this predictive model has facilitated the design of a new strategy to form ETD reagent anions from precursors generated via ESI. Finally, two promising, emerging areas of gas-phase ion chemistry of peptides are also described: (1) the design of new gas-phase radical chemistry to probe peptide structure, and (2) selective cleavage of disulfide bonds of peptides in the gas phase via various physicochemical approaches.