Visible Light-Assisted Coordination of a Rh(III)-BODIPY Complex to Guanine.

Most photodynamic therapeutics (PDTs) used in cancer treatment require oxygen to work efficiently to terminate cancer cells. These PDTs do not efficiently treat tumors in hypoxic conditions. Rh(III) polypyridyl complexes have been reported to have a photodynamic therapeutic effect in hypoxic conditions when exposed to UV light. UV light can damage tissue and cannot penetrate deep to reach cancer cells. This work proposes the coordination of a BODIPY fluorophore to a rhodium metal center to form a Rh(III)-BODIPY complex that enhances the reactivity of the rhodium under visible light. This complex formation is facilitated with the BODIPY as the highest occupied molecular orbital (HOMO), while the lowest unoccupied molecular orbital (LUMO) is localized on the Rh(III) metal center. Irradiation of the BODIPY transition at ∼524 nm can cause an indirect electron transfer from the orbital of the BODIPY-centered HOMO to the Rh(III)-centered LUMO, populating the dσ* orbital. In addition, photo binding of the Rh complex covalently coordinated to the N (7) position of guanine in an aqueous solution was also observed by mass spectrometry after chloride dissociation upon irradiation with green visible light (532 nm LED). Calculated thermochemistry values of the Rh complex reaction in methanol, acetonitrile, water, and guanine were determined using DFT calculations. All enthalpic reactions and Gibbs free energies were identified as endothermic and nonspontaneous, respectively. This observation supports the chloride dissociation using 532 nm light. This Rh(III)-BODIPY complex expands the class of visible light-activated Rh(III) photocisplatin analogs that may have potential photodynamic therapeutic activity for the treatment of cancers in hypoxic conditions.

Protonated α-N-Acetyl Galactose Glycopeptide Dissociation Chemistry.

We recently provided mass spectrometric, H/D labeling, and computational evidence of pyranose to furanose N-acetylated ion isomerization reactions that occurred prior to glycosidic bond cleavage in both O- and N-linked glycosylated amino acid model systems (Guan et al. Phys. Chem. Chem. Phys., 2021, 23, 23256-23266). These reactions occurred irrespective of the glycosidic linkage stereochemistry (α or ß) and the N-acetylated hexose structure (GlcNAc or GalNAc). In the present article, we test the generality of the preceding findings by examining threonyl α-GalNAc-glycosylated peptides. We utilize computational chemistry to compare the various dissociation and isomerization pathways accessible with collisional activation. We then interrogate the structure(s) of the resulting charged glycan and peptide fragments with infrared "action" spectroscopy. Isomerization of the original pyranose, the protonated glycopeptide [AT(GalNAc)A+H]+, is predicted to be facile compared to direct dissociation, as is the glycosidic bond cleavage of the newly formed furanose form, i.e., furanose oxazolinium ion structures are predicted to predominate. IR action spectra for the m/z 204, C8H14N1O5+, glycan fragment population support this prediction. The IR action spectra of the complementary m/z 262 peptide fragment were assigned as a mixture of the lowest-energy structures of [ATA+H]+ consistent with the literature. If general, the change to a furanose m/z 204 product ion structure fundamentally alters the ion population available for MS3 dissociation and glycopeptide sequence identification.

Evidence of gas-phase pyranose-to-furanose isomerization in protonated peptidoglycans.

Peptidoglycans are diverse co- and post-translational modifications of key importance in myriad biological processes. Mass spectrometry is employed to infer their biomolecular sequences and stereochemisties, but little is known about the critical gas-phase dissociation processes involved. Here, using tandem mass spectrometry (MS/MS and MSn), isotopic labelling and high-level simulations, we identify and characterize a facile isomerization reaction that produces furanose N-acetylated ions. This reaction occurs for both O- and N-linked peptidoglycans irrespective of glycosidic linkage stereochemistry (α/ß). Dissociation of the glycosidic and other bonds thus occur from the furanose isomer critically altering the reaction feasibility and product ion structures.

Gas-Phase Dissociation Chemistry of Deprotonated RGD.

We investigate the structure and dissociation pathways of the deprotonated amphoteric peptide arginylglycylasparic acid, [RGD-H]-. We model the pertinent gas-phase structures and fragmentation chemistry of the precursor anions and predominant sequence-informative bond cleavages (b2+H2O, c2, and z1 peaks) and compare these predictions to our tandem mass spectra and infrared spectroscopy experiments. Formation of the b2+H2O anions requires rate-limiting intramolecular back biting to cleave the second amide bond and generate an anhydride structure. Facile cleavage of the newly formed ester bond with concerted expulsion of a cyclic anhydride neutral generates the product structure. IR spectroscopy supports this b2+H2O anion having structures that are essentially identical to C-terminally deprotonated arginylglycine, [RG-H]-. Formation of the c2 anion is predicted to require concerted expulsion of CO2 from the aspartyl side chain carboxylate and cleavage of the N-Calpha bond to produce a proton-bound dimer of arginylglycinamide and acrylate. Proton transfers within the dimer then enable predominant detection of a c2 anion with the negative charge nominally on the central, glycine nitrogen (amidate structure) as the proton affinity of this structure is predicted to be lower than acrylate by ∼27 kJ mol-1. Alternate means of cleaving the same N-Calpha bond produce deprotonated cis-1,4-dibut-2-enoic acid z1 anion structures. These lowest energy processes involve C-H proton mobilization from the aspartyl side chain prior to N-Calpha bond cleavage consistent with proposals from the literature.

Sodium-cationized carbohydrate gas-phase fragmentation chemistry: influence of glycosidic linkage position.

We investigate the gas-phase structures and fragmentation chemistry of two isomeric sodium-cationized carbohydrates using combined tandem mass spectrometry, hydrogen/deuterium exchange experiments, and computational methods. Our model systems are the glucose-based disaccharide analytes cellobiose (ß-d-glucopyranosyl-(1 → 4)-d-glucose) and gentiobiose (ß-d-glucopyranosyl-(1 → 6)-d-glucose). These analytes show substantially different tandem mass spectra. We characterize the rate-determining barriers to both the glycosidic and structurally-informative cross-ring bond cleavages. Sodiated cellobiose produces abundant Y1 and B1 peaks. Our deuterium labelling and computational chemistry approach provides evidence for 1,6-anhydroglucose B1 ion structures rather than the 1,2-anhydroglucose and oxacarbenium ion structures proposed elsewhere. Unlike those earlier proposals, this finding is consistent with the experimentally observed Bn/Ym branching ratios. In contrast to cellobiose, sodiated gentiobiose primarily fragments by cross-ring cleavage to form various A2 ion types. Fragmentation is facilitated by ring-opening at the reducing end which enables losses of CnH2nOn oligomers. Deuterium labelling and theory enables rationalization of these processes. Theory and experiment also support the importance of consecutive fragmentation processes at higher collision energies.

Stereochemical Sequence Ion Selectivity: Proline versus Pipecolic-acid-containing Protonated Peptides.

Substitution of proline by pipecolic acid, the six-membered ring congener of proline, results in vastly different tandem mass spectra. The well-known proline effect is eliminated and amide bond cleavage C-terminal to pipecolic acid dominates instead. Why do these two ostensibly similar residues produce dramatically differing spectra? Recent evidence indicates that the proton affinities of these residues are similar, so are unlikely to explain the result [Raulfs et al., J. Am. Soc. Mass Spectrom. 25, 1705-1715 (2014)]. An additional hypothesis based on increased flexibility was also advocated. Here, we provide a computational investigation of the "pipecolic acid effect," to test this and other hypotheses to determine if theory can shed additional light on this fascinating result. Our calculations provide evidence for both the increased flexibility of pipecolic-acid-containing peptides, and structural changes in the transition structures necessary to produce the sequence ions. The most striking computational finding is inversion of the stereochemistry of the transition structures leading to "proline effect"-type amide bond fragmentation between the proline/pipecolic acid-congeners: R (proline) to S (pipecolic acid). Additionally, our calculations predict substantial stabilization of the amide bond cleavage barriers for the pipecolic acid congeners by reduction in deleterious steric interactions and provide evidence for the importance of experimental energy regime in rationalizing the spectra. Graphical Abstract ᅟ.

Cationized Carbohydrate Gas-Phase Fragmentation Chemistry.

We investigate the fragmentation chemistry of cationized carbohydrates using a combination of tandem mass spectrometry, regioselective labeling, and computational methods. Our model system is D-lactose. Barriers to the fundamental glyosidic bond cleavage reactions, neutral loss pathways, and structurally informative cross-ring cleavages are investigated. The most energetically favorable conformations of cationized D-lactose were found to be similar. In agreement with the literature, larger group I cations result in structures with increased cation coordination number which require greater collision energy to dissociate. In contrast with earlier proposals, the B n -Y m fragmentation pathways of both protonated and sodium-cationized analytes proceed via protonation of the glycosidic oxygen with concerted glycosidic bond cleavage. Additionally, for the sodiated congeners our calculations support sodiated 1,6-anhydrogalactose B n ion structures, unlike the preceding literature. This affects the subsequent propensity of formation and prediction of B n /Y m branching ratio. The nature of the anomeric center (α/ß) affects the relative energies of these processes, but not the overall ranking. Low-energy cross-ring cleavages are observed for the metal-cationized analytes with a retro-aldol mechanism producing the 0,2 A 2 ion from the sodiated forms. Theory and experiment support the importance of consecutive fragmentation processes, particularly for the protonated congeners at higher collision energies. Graphical Abstract ᅟ.

Benefits of multidimensional fractionation for the study and characterization of natural organic matter.

The belief that chromatographic separation of complex environmental mixtures or natural organic matter (NOM) produces featureless humps from which little, if anything, can be learned is still pervasive. Meanwhile improvements in chromatography and the use of information-rich detection methods have led to meaningful fractionation and revealed consistent data. Here, we build on this work and developed a robust, facile two-dimensional separation with high orthogonality between dimensions. We illustrate that re-injections of fractions (both in the first and in the second dimension) leads to individual peaks at the expected retention times and use information-rich detection to investigate the basis on which NOM is fractionated. We demonstrate unprecedentedly feature-rich chromatograms are observed even with standard UV detection for polar NOM fractions. The second stage of fractionation is demonstrated to separate isomers, providing a direct look at isomeric complexity in NOM as well as a tool to reduce it. Consistent with expectation, but confirmed for the first time through mass spectral data, radicals were detected for NOM components that were generally nonpolar and grouped in the condensed aromatic structure - like region of van Krevelen plots. High-resolution tandem mass spectral data, furthermore, suggests that many higher-MW components of fulvic acids (especially the highly oxidized ones) have formulas that do not match any known compounds in the literature, supporting the hypothesis that fulvic acids are a unique compound-class. Combined, the data illustrate that meaningful reduction in complexity reveals new compositional and structural detail and avails new avenues of investigation.

Proton Mobility in b2 Ion Formation and Fragmentation Reactions of Histidine-Containing Peptides.

A detailed energy-resolved study of the fragmentation reactions of protonated histidine-containing peptides and their b2 ions has been undertaken. Density functional theory calculations were utilized to predict how the fragmentation reactions occur so that we might discern why the mass spectra demonstrated particular energy dependencies. We compare our results to the current literature and to synthetic b2 ion standards. We show that the position of the His residue does affect the identity of the subsequent b2 ion (diketopiperazine versus oxazolone versus lactam) and that energy-resolved CID can distinguish these isomeric products based on their fragmentation energetics. The histidine side chain facilitates every major transformation except trans-cis isomerization of the first amide bond, a necessary prerequisite to diketopiperazine b2 ion formation. Despite this lack of catalyzation, trans-cis isomerization is predicted to be facile. Concomitantly, the subsequent amide bond cleavage reaction is rate-limiting.

Cα hydrogen atom transfer in post-cleavage radical-cation complexes: short and steep versus long winding road.

Recently, I explored structurally straightforward pathways to Cα hydrogen atom, H(•), transfer reactions in the radical cation complex following electron capture/transfer of a series of polyprotonated peptides (J. Phys. Chem. A 2013, 117, 1189-1196). Here, I extend my analysis to incorporate detailed rearrangement processes potentially occurring prior to H(•) transfer. This comprises intracomplex isomerization of the initial iminol-terminated (-C(OH)═NH) form of the cn' species to the energetically more favorable, amide-terminated form (-C(O)-NH2) prior to Cα H(•) abstraction by the zm(•) species. The data indicate that the previously published H(•) transfer barriers are more energetically demanding than those of this multistep alternative. The rate-determining step is typically the intracomplex iminol isomerization, consistent with the substantial energetic favorability of the amide form of the cn species. The barriers to H(•) transfer still rise steeply as a function of the charge state. In agreement with experiment, evidence for product separation without H(•) transfer at a higher charge state is also provided.

Unequivocal determination of site-specific protein disulfide bond reduction potentials by top-down FTICR MS: characterization of the N- and C-terminal redox-active sites in human thioredoxin 1.

We report the reliable determination of equilibrium protein disulfide bond reduction potentials (E°') by isotope-coded cysteine alkylation coupled with top-down Fourier transform ion cyclotron resonance mass spectrometry (FTICR MS). This technique enables multiple redox-active sites to be characterized simultaneously and unambiguously without the need for proteolysis or site-directed mutagenesis. Our model system was E. coli thioredoxin, and we determined E°' for its CGPC active-site disulfide as -280 mV in accord with literature values. E°' for the homologous disulfide in human thioredoxin 1 (Trx1) was determined as -281 mV, a value considerably more negative than the previously reported -230 mV. We also observed S-glutathionylation of Trx1 and localized that redox modification to Cys72; E°' for the intermolecular disulfide was determined as -186 mV. Intriguingly, that value corresponds to the intracellular glutathione/glutathione disulfide (GSH/GSSG) potential at the redox boundary between cellular differentiation and apoptosis.

Structural and some medicinal characteristics of the copper(II)-hydroxychloroquine complex.

In the first phase of this study, the binding of hydroxychloroquine to the copper(II) cation is examined using liquid chromatography-mass spectrometry (LC-MS), matrix assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF-MS), Fourier transform-ion cyclotron resonance spectrometry (FT-ICR) and nuclear magnetic resonance ((1)H and (13)C NMR) in one and two dimensions. The data suggest the metal-ligand complex is a polarity adaptive molecule. In the second phase of the study, the complexes activity is tested against the National Cancer Institute's 60 cell line panel. Its anti-cancer activity is compared to quinine, Cu(II)-quinine and hydroxychloroquine. It serves as a base line for future anti-cancer complexes in which hydroxychloroquine is utilized for its ability to impact cell autophagy.

To jump or not to jump? Cα hydrogen atom transfer in post-cleavage radical-cation complexes.

Conventionally, electron capture or transfer to a polyprotonated peptide ion produces an initial radical-cation intermediate which dissociates "directly" to generate complementary c(n)' and z(m)(•) sequence ions (or ions and neutrals). Alternatively, or in addition, the initial radical-cation intermediate can undergo H(•) migration to produce c(n)(•) (or c(n) - H(•)) and z(m)' (or z(m)(•) + H(•)) species prior to complex separation ("nondirect"). This reaction significantly complicates spectral interpretation, creates ambiguity in peak assignment, impairs effective algorithmic processing (reduction of the spectrum to solely (12)C m/z values), and reduces sequence ion signal-to-noise. Experimental evidence indicates that the products of hydrogen atom transfer reactions are substantially less prevalent for higher charge state precursors. This effect is generally rationalized on the basis of decreased complex lifetime. Here, we present a theoretical study of these reactions in post N-C(α) bond cleavage radical-cation complexes as a function of size and precursor charge state. This approach provides a computational estimate of the barriers associated with these processes for highly charged peptides with little charge solvation. The data indicate that the H(•) migration is an exothermic process and that the barrier governing this reaction rises steeply with precursor ion charge state. There is also some evidence for immediate product separation following N-C(α) bond cleavage at higher charge state.

Relative stability of peptide sequence ions generated by tandem mass spectrometry.

We report the use of unimolecular dissociation by infrared radiation for gaseous multiphoton energy transfer to determine relative activation energy (E(a,laser)) for dissociation of peptide sequence ions. The sequence ions of interest are mass-isolated; the entire ion cloud is then irradiated with a continuous wave CO(2) laser, and the first order rate constant, k(d), is determined for each of a series of laser powers. Provided these conditions are met, a plot of the natural logarithm of k(d) versus the natural logarithm of laser power yields a straight line, whose slope provides a measure of E(a,laser). This method reproduces the E(a) values from blackbody radiative dissociation (BIRD) for the comparatively large, singly and doubly protonated bradykinin ions (nominally y ( 9 ) and y ( 9 ) ( 2+ )). The comparatively small sequence ion systems produce E(a,laser) values that are systematic underestimates of theoretical barriers calculated with density functional theory (DFT). However, the relative E(a,laser) values are in qualitative agreement with the mobile proton model and available theory. Additionally, novel protonated cyclic-dipeptide (diketopiperazine) fragmentation reactions are analyzed with DFT. FT-ICR MS provides access to sequence ions generated by electron capture dissociation, infrared multiphoton dissociation, and collisional activation methods (i.e., b ( n ), y ( m ) , c ( n ), z ( m ) ( • ) ions).

Gas-phase structure and fragmentation pathways of singly protonated peptides with N-terminal arginine.

The gas-phase structures and fragmentation pathways of the singly protonated peptide arginylglycylaspartic acid (RGD) are investigated by means of collision-induced-dissociation (CID) and detailed molecular mechanics and density functional theory (DFT) calculations. It is demonstrated that despite the ionizing proton being strongly sequestered at the guanidine group, protonated RGD can easily be fragmented on charge directed fragmentation pathways. This is due to facile mobilization of the C-terminal or aspartic acid COOH protons thereby generating salt-bridge (SB) stabilized structures. These SB intermediates can directly fragment to generate b(2) ions or facilely rearrange to form anhydrides from which both b(2) and b(2)+H(2)O fragments can be formed. The salt-bridge stabilized and anhydride transition structures (TSs) necessary to form b(2) and b(2)+H(2)O are much lower in energy than their traditional charge solvated counterparts. These mechanisms provide compelling evidence of the role of SB and anhydride structures in protonated peptide fragmentation which complements and supports our recent findings for tryptic systems (Bythell, B. J.; Suhai, S.; Somogyi, A.; Paizs, B. J. Am. Chem. Soc. 2009, 131, 14057-14065.). In addition to these findings we also report on the mechanisms for the formation of the b(1) ion, neutral loss (H(2)O, NH(3), guanidine) fragment ions, and the d(3) ion.

Cyclization and rearrangement reactions of a(n) fragment ions of protonated peptides.

a(n) ions are frequently formed in collision-induced dissociation (CID) of protonated peptides in tandem mass spectrometry (MS/MS) based sequencing experiments. These ions have generally been assumed to exist as immonium derivatives (-HN(+)═CHR). Using a quadrupole ion trap mass spectrometer, MS/MS experiments have been performed and the structure of a(n) ions formed from oligoglycines was probed by infrared spectroscopy. The structure and isomerization reactions of the same ions were studied using density functional theory. Overall, theory and infrared spectroscopy provide compelling evidence that a(n) ions undergo cyclization and/or rearrangement reactions, and the resulting structure(s) observed under our experimental conditions depends on the size (n). The a(2) ion (GG sequence) undergoes cyclization to form a 5-membered ring isomer. The a(3) ion (GGG sequence) undergoes cyclization initiated by nucleophilic attack of the carbonyl oxygen of the N-terminal glycine residue on the carbon center of the C-terminal immonium group forming a 7-membered ring isomer. The barrier to this reaction is comparatively low at 10.5 kcal mol(-1), and the resulting cyclic isomer (-5.4 kcal mol(-1)) is more energetically favorable than the linear form. The a(4) ion with the GGGG sequence undergoes head-to-tail cyclization via nucleophilic attack of the N-terminal amino group on the carbon center of the C-terminal immonium ion, forming an 11-membered macroring which contains a secondary amine and three trans amide bonds. Then an intermolecular proton transfer isomerizes the initially formed secondary amine moiety (-CH(2)-NH(2)(+)-CH(2)-NH-CO-) to form a new -CH(2)-NH-CH(2)-NH(2)(+)-CO- form. This structure is readily cleaved at the -CH(2)-NH(2)(+)- bond, leading to opening of the macrocycle and formation of a rearranged linear isomer with the H(2)C═NH(+)-CH(2)- moiety at the N terminus and the -CO-NH(2) amide bond at the C terminus. This rearranged linear structure is much more energetically favorable (-14.0 kcal mol(-1)) than the initially formed imine-protonated linear a(4) ion structure. Furthermore, the barriers to these cyclization and ring-opening reactions are low (8-11 kcal mol(-1)), allowing facile formation of the rearranged linear species in the mass spectrometer. This finding is not limited to 'simple' glycine-containing systems, as evidenced by the IRMPD spectrum of the a(4) ion generated from protonated AAAAA, which shows a stronger tendency toward formation of the energetically favorable (-12.3 kcal mol(-1)) rearranged linear structure with the MeHC═NH(+)-CHMe- moiety at the N terminus and the -CO-NH(2) amide bond at the C terminus. Our results indicate that one needs to consider a complex variety of cyclization and rearrangement reactions in order to decipher the structure and fragmentation pathways of peptide a(n) ions. The implications this potentially has for peptide sequencing are also discussed.

Effect of the His residue on the cyclization of b ions.

The MS(n) spectra of the [M + H](+) and b(5) peaks derived from the peptides HAAAAA, AHAAAA, AAHAAA, AAAHAA, and AAAAHA have been measured, as have the spectra of the b(4) ions derived from the first four peptides. The MS(2) spectra of the [M + H](+) ions show a substantial series of b(n) ions with enhanced cleavage at the amide bond C-terminal to His and substantial cleavage at the amide bond N-terminal to His (when there are at least two residues N-terminal to the His residue). There is compelling experimental and theoretical evidence for formation of nondirect sequence ions via cyclization/reopening chemistry in the CID spectra of the b ions when the His residue is near the C-terminus. The experimental evidence is less clear for ions when the His residue is near the N-terminus, although this may be due to the use of multiple alanine residues in the peptide making identifying scrambled peaks more difficult. The product ion mass spectra of the b(4) and b(5) ions from these isomeric peptides with cyclically permuted amino acid sequences are similar, but also show clear differences. This indicates less active cyclization/reopening followed by fragmentation of common structures for b(n) ions containing His than for sequences of solely aliphatic residues. Despite more energetically favorable cyclization barriers for the b(5) structures, the b(4) ions experimental data show more clear evidence of cyclization and sequence scrambling before fragmentation. For both b(4) and b(5) the energetically most favored structure is a macrocyclic isomer protonated at the His side chain.

Proton-driven amide bond-cleavage pathways of gas-phase peptide ions lacking mobile protons.

The mobile proton model (Dongre, A. R., Jones, J. L., Somogyi, A. and Wysocki, V. H. J. Am. Chem. Soc. 1996, 118 , 8365-8374) of peptide fragmentation states that the ionizing protons play a critical role in the gas-phase fragmentation of protonated peptides upon collision-induced dissociation (CID). The model distinguishes two classes of peptide ions, those with or without easily mobilizable protons. For the former class mild excitation leads to proton transfer reactions which populate amide nitrogen protonation sites. This enables facile amide bond cleavage and thus the formation of b and y sequence ions. In contrast, the latter class of peptide ions contains strongly basic functionalities which sequester the ionizing protons, thereby often hindering formation of sequence ions. Here we describe the proton-driven amide bond cleavages necessary to produce b and y ions from peptide ions lacking easily mobilizable protons. We show that this important class of peptide ions fragments by different means from those with easily mobilizable protons. We present three new amide bond cleavage mechanisms which involve salt-bridge, anhydride, and imine enol intermediates, respectively. All three new mechanisms are less energetically demanding than the classical oxazolone b(n)-y(m) pathway. These mechanisms offer an explanation for the formation of b and y ions from peptide ions with sequestered ionizing protons which are routinely fragmented in large-scale proteomics experiments.

Infrared spectroscopy of fragments of protonated peptides: direct evidence for macrocyclic structures of b5 ions.

b ions are of fundamental importance in peptide sequencing using tandem mass spectrometry. These ions have generally been assumed to exist as protonated oxazolone derivatives. Recent work indicates that medium-sized b ions can rearrange by head-to-tail cyclization of the oxazolone structures generating macrocyclic protonated peptides as intermediates. Here, we show using infrared spectroscopy and density functional theory calculations that the b(5) ion of protonated G(5)R exists in the mass spectrometer as an amide oxygen protonated cyclic peptide rather than fleetingly as a transient intermediate. This assignment is supported by our DFT calculations which show this macrocyclic isomer to be energetically preferred over the open oxazolone form despite the entropic constraints the cyclic form introduces.

What is the structure of b(2) ions generated from doubly protonated tryptic peptides?

A recent statistical study (Savitski, M. M.; Falth, M.; Eva Fung, Y. M.; Adams, C. M.; Zubarev, R. A. J. Am. Soc. for Mass Spectrom.doi: 10.1016/j.jasms.2008.08.003) of a large spectral database indicated that the product ion spectra of doubly protonated tryptic peptides fall into two distinct classes. The main factor distinguishing the two classes is the relative abundance of the y(N-2) fragment: for Class I spectra y(N-2) is the most abundant y fragment while for Class II other y ions dominate the corresponding spectra. To explain the dominance of y(N-2) for Class I spectra formation of a nontraditional b(2) ion with a diketopiperazine (6-membered cyclic peptide) rather than an oxazolone structure was proposed. Here we present evidence from tandem mass spectrometry, hydrogen/deuterium exchange, and density functional calculations that do not support this proposal. Namely, that CID of doubly protonated YIGSR, YGGFLR, and YIYGSFK produce Class I product ion spectra, yet the b(2) fragment is shown to have the traditional oxazolone structure.