Dehydration of peptide [M + H]+ ions in the gas phase

The loss of water from protonated peptides was studied using [¹⁸O]-labeling of the C-terminal carboxyl group. The structures (including the location of the isotopic label) of first-generation product ions were examined by sequential product ion scanning (MS³ and MS⁴) using a hybrid sector/quadrupole mass spectrometer. Water loss may involve carboxylic acid groups, side-chain hydroxyls, or peptide backbone oxygens. Although one of these three pathways often predominates, more than one dehydration route can be operative for a single peptide structure. When peptide backbone oxygen is lost, the dehydration can occur at one or two primary sites along the backbone, with the location of the site(s) varying among peptides. When water loss involves the C-terminal carboxyl group, the resulting ion may undergo extensive intraionic oxygen isotope exchange. This evidence for complex intraionic interactions further emphasizes the significance of gas-phase conformation in determining the fragmentations of peptide ions.     

C-terminal amino acid residue loss for deprotonated peptide ions containing glutamic acid, aspartic acid, or serine residues at the C-terminus

Deprotonated peptides containing C-terminal glutamic acid, aspartic acid, or serine residues were studied by sustained off-resonance irradiation collision-induced dissociation (SORI-CID) in a Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometer with ion production by electrospray ionization (ESI). Additional studies were performed by post source decay (PSD) in a matrix-assisted laser desorption ionization/time-of-flight (MALDI/TOF) mass spectrometer. This work included both model peptides synthesized in our laboratory and bioactive peptides with more complex sequences. During SORI-CID and PSD, [M - H]- and [M - 2H]2- underwent an unusual cleavage corresponding to the elimination of the C-terminal residue. Two mechanisms are proposed to occur. They involve nucleophilic attack on the carbonyl carbon of the adjacent residue by either the carboxylate group of the C-terminus or the side chain carboxylate group of C-terminal glutamic acid and aspartic acid residues. To confirm the proposed mechanisms, AAAAAD was labelled by 18O specifically on the side chain of the aspartic acid residue. For peptides that contain multiple C-terminal glutamic acid residues, each of these residues can be sequentially eliminated from the deprotonated ions; a driving force may be the formation of a very stable pyroglutamatic acid neutral. For peptides with multiple aspartic acid residues at the C-terminus, aspartic acid residue loss is not sequential. For peptides with multiple serine residues at the C-terminus, C-terminal residue loss is sequential; however, abundant loss of other neutral molecules also occurs. In addition, the presence of basic residues (arginine or lysine) in the sequence has no effect on C-terminal residue elimination in the negative ion mode.     

Fragmentation chemistry of [M + Cu]+ peptide ions containing an N-terminal arginine

[M + Cu]+ peptide ions formed by matrix-assisted laser desorption/ionization from direct desorption off a copper sample stage have sufficient internal energy to undergo metastable ion dissociation in a time-of-flight mass spectrometer. On the basis of fragmentation chemistry of peptides containing an N-terminal arginine, we propose the primary Cu+ ion binding site is the N-terminal arginine with Cu+ binding to the guanidine group of arginine and the N-terminal amine. The principal decay products of [M + Cu]+ peptide ions containing an N-terminal arginine are [a(n) + Cu - H]+ and [b(n) + Cu - H]+ fragments. We show evidence to suggest that [a(n) + Cu - H]+ fragment ions are formed by elimination of CO from [b(n) + Cu - H]+ ions and by direct backbone cleavage. We conclude that Cu+ ionizes the peptide by attaching to the N-terminal arginine residue; however, fragmentation occurs remote from the Cu+ ion attachment site involving metal ion promoted deprotonation to generate a new site of protonation. That is, the fragmentation reactions of [M + Cu]+ ions can be described in terms of a "mobile proton" model. Furthermore, proline residues that are adjacent to the N-terminal arginine do not inhibit formation of [b(n) + Cu - H]+ ion, whereas proline residues that are distant to the charge carrying arginine inhibit formation of [b(n) + Cu - H]+ ions. An unusual fragment ion, [c(n) + Cu + H]+, is also observed for peptides containing lysine, glutamine, or asparagine in close proximity to the Cu+ carrying N-terminal arginine. Mechanisms for formation of this fragment ion are also proposed.     

Criteria for distinguishing between [M]+· and [M+H]+ ions in field desorption mass spectra

Ten criteria are introduced to distinguish between molecular ions and protonated parent molecules in field desorption mass spectrometry.     

Collisional activation spectra of [M + Li]+, [M + Na]+ and [M + K]+ ions formed by field desorption of some monosaccharides

The collisional activation spectra of monosaccharide ions formed by [Li]⁺, [Na]⁺ and [K]⁺ ion attachment under field desorption conditions are reported. It is shown that the elimination of the alkali ions is determined by the alkali ion affinities of the molecules (M) and competes with a fragmentation of M which is almost independent of the alkali ion attached. Correspondingly the alkali ion is predominantly retained in the fragment ions. The usefulness of this method for the differentiation of underivatized isomers is demonstrated.     

Comparison of intensities between doubly charged ions [M + 2H]2+ and singly charged ions [M + H]+ of gramicidin S by electrospray mass spectrometry.

The reason why the intensity of doubly charged ions [M + 2H]2+ of gramicidin S is higher than that of singly charged ions [M + H]+ in electrospray is investigated by ion evaporation theory. As a result of comparison between the total free energies of extracting [M + 2H]2+ and [M + H]+ from a charged droplet to infinity, it is found that the total free energy of [M + 2H]2+ is estimated to be lower than that of [M + H]+. This clearly supports the experimental result. In addition, the importance of the electrostatic contribution in electrospray is demonstrated by showing the result that the total free energy of [M + 2H]2+ without electrostatic contribution is higher than that of [M + H]+.     

Unimolecular decomposition of M+. and [M+H]+ species of some fluorosubstituted acyclic nucleoside analogs.

Electron ionization and fast-atom bombardment mass spectrometry are shown to provide a valid analytical tool for the structural characterization of the title compounds. In fact, diagnostic fragmentation pathways were observed depending on the presence of different substituents (benzyloxy, (benzyloxy, p-tolylthio, p-tolylsulphinyl) as well as of different bases. Regioisomeric compounds could be differentiated by kinetic energy release measurements.     

Metastable [C4H8O]+˙ ions: Additional decompositions via the [2-butanone]+˙ ion

The losses of methyl and ethyl through the intermediacy of the [2-butanone]⁺˙ ion are shown to be the dominant metastable decomposition of 14 of 19 [C₄H₈O]⁺˙ ions examined. The ions that decompose via the [2-butanone]⁺˙ structure include ionized aldehydes, unsaturated and cyclic alcohols and enolic ions. [Cyclic ether]⁺˙ [cyclopropylmethanol]⁺˙ and [2-methyl-1-propen-1-ol]⁺˙ ions do not decompose through ionized 2-butanone. The rearrangements of various [C₄H₈O]⁺˙ ions the the 2-butanone ion were investigated by means of deuterium labeling. Those pathways involve up to eight steps. Ions with the oxygen on the end carbon rearrange to a common structure or mixture of structures. Those ions which ultimately rearrange to the [2-butanone]⁺˙ ion then undergo oxygen shifts from the terminal to the second and third carbons at about equal rates. However, this oxygen shift does not precede the losses of water and ethylene. Losses of water and ethylene were unimportant for ions with the oxygen initially on the second carbon. Ionized n-butanal and cyclobutanol, but not other [C₄H₈O]⁺˙ ions, undergo reversible hydrogen exchange between the oxygen and the terminal carbon. Rearrangement of ionized n-butanal to the [cyclobutanol]⁺˙ ion is postulated.     

A mass spectrometric and ab initio study of the pathways for dehydration of simple glycine and cysteine-containing peptide [M+H]+ ions

The gas phase fragmentation reactions of the [M+H]⁺ and [M+H?H²O]⁺ ions of glycylglycine, glycylcysteine, N-acetylglycine, N-acetylcysteine, their corresponding methyl esters, as well as several other related model systems have been examined by electrospray ionization (ESI) tandem mass spectrometry (MS ⁿ ) using triple quadrupole and quadrupole ion trap mass spectrometers. Two discrete gas phase fragmentation pathways for the loss of water from glycine-containing peptides, corresponding to retro-Koch and retro-Ritter type reactions were observed. Two pathways were also observed for the loss of water from C-terminal cysteine-containing peptides: a retro-Koch type reaction and an intramolecular nucleophilic attack at the carbonyl of the amide bond by the cysteinyl side chain thiol. Various intermediates involved in these reactions, derived from the [M+H?H²O]⁺ ions of N-formylglycine and N-formylcysteine, were modeled using ab initio calculations at the MP2(FC)/6-31G*//HF/6-31G* level of theory. These calculations indicate that: (i) the retro-Koch reaction product is predicted to be more stable than the product from the retro-Ritter reaction for N-formylglycine, and (ii) the intramolecular nucleophilic attack product is preferred over the retro-Koch and retro-Ritter reaction products for N-formylcysteine. The results from these ab initio calculations are in good agreement with the experimentally determined ion abundances for these processes.     

Competitive formation of M+˙ and [M + H]+ ions under fast atom bombardment conditions

The competitive formation of molecular ions M^+˙ and protonated molecules [M + H]⁺ under fast atom bombardment (FAB) conditions was examined using various kinds of organic compounds. The use of protic/hydrophilic matrices such as thioglycerol and glycerol resulted in relatively large values of the peak intensity ratio I([M + H]⁺)/I(M^+˙) compared with the use of relatively aprotic/hydrophobic matrices such as m-nitrobenzyl alcohol and o-nitrophenyl octyl ether. The change of matrix from thiol-containing such as thioglycerol and dithiothreitol to alcoholic such as glycerol and pentamethylene glycol increased the I([M + H]⁺)/I(M^+˙) ratio. Furthermore, the change of matrix increased the peak intensity ratio of the doubly charged ion [M + 2H]²⁺ to [M + H]⁺ in the FAB mass spectra of angiotensin I and gramicidin S. The addition of acids to the matrix solution increased the I([M + H]⁺)/I(M^+˙) ratio, although such an effect did not always occur. The acetylation of simple aniline compounds markedly increased the I([M + H]⁺)/I(M^+˙) ratio. It was concluded from these results that the hydrogen bonding interaction between hydroxyl groups(s) of the matrix and basic site(s) of analyte molecules in solution acts advantageously as a quasi-preformed state for [M + H]⁺ formation, and that the presence of significant proton acceptor(s) such as carbonyl group in analytes hinder the M^+˙ formation which may generally occur under FAB conditions. The formation of M^+˙ and [M + H]⁺ ions seemed to occur competitively, reflecting or according to the interaction or solvation states between the analyte and matrix molecules in solution and the structural characteristics of the analytes.     

Structure of [C3H5O]+ ions produced from unimolecular decomposition of several [C4H8O]+˙ metastable ions

It is demonstrated by means of collisionally activated decomposition (CAD) that [C₃H₅O]⁺ originating from metastable [C₄H₈O]^+˙ ions are either acylium [C₂H₅CO]⁺ (a) or hydroxycarbenium [CH₂CHCHOH]⁺ (b). Butanone gives exclusively a but 2-methyl-2-propen-1-ol, 2-buten-1-ol, 3-buten-1-ol, butanal and 2-methylpropanal lead to ion b. Both structures a and b are produced from 3-buten-2-ol. These results are discussed in conjunction with experimental and calculated (MINDO/3) thermodynamic data.     

Collision-induced decompositions of [M – H]− and [M + Li]+ ions from fast atom bombardment of dinucleoside phenylphosphonates

The collision-induced decompositions of the [M – H]^? and [M + Li]⁺ ions of a few dinucleoside phenylphosphonates were studied using fast atom bombardment and linked scanning at constant B/E. Deprotonation takes place on the base or sugar moieties. The [M – H]^? ion decomposes mainly by cleavage on either side of the phosphonate linkage, leading to the formation of mononucleotide fragment ions and also by cleavage of the basesugar bond. Rupture of the 3′-phosphonate bond is preferred. Unlike the normal charged nucleotides, these neutral nucleotides do not eliminate a neutral base from the [M – H]^? ion. However, the mononucleotide fragment ions which can have the charge on the phosphorus oxygen eliminate neutral bases by charge-remote fragmentation. The 4,4′-dimethoxytrityl (DMT)-protected nucleotides show the additional fragmentation of loss of DMT. Li⁺ attachment can occur at several sites in the molecule. As observed for the [M – H]^? ion, the major cleavage occurs on either side of the phosphonate bond in the fully deprotected nucleotides, cleavage of the ester bond on C(3′) being preferred. Cleavage of the 5′-phosphonate bond is not observed in the DMT-protected nucleotides. Many of the fragmentations observed can be explained as arising from charge-remote reactions.     

Electrospray ionization mass spectrometric analysis of microcystins, cyclic heptapeptide hepatotoxins: modulation of charge states and [M+H]+ to [M+Na]+ ratio

Electrospray ionization mass spectrometry was used to develop a rapid, sensitive, and accurate method for determination and identification of hepatotoxic microcystins, cyanobacterial cyclic heptapeptides. To optimize the electrospray ionization conditions, factors affecting charge state distribution, such as amino acid components of sample, proton affinity of the additives, and additive concentration, were investigated in detail and a method for controlling charge states was developed to provide molecular-related ions for assignment of molecular weight and reasonably abundant precursor ions for MS/MS analysis. A procedure for identification of microcystins consisting of known amino acids was proposed: for microcystins giving abundant [M + 2H]2+ ions, the addition of nitrogen-containing bases to the aqueous sample solution is effective to obtain an increased intensity of [M + H]+ ions, whereas the addition of Lewis acids containing nitrogen can produce increased abundances of [M + 2H]2+ ions for microcystins giving weak [M + 2H]2+ ions. Microcystins possessing no arginine residue always give sodium adduct ions [M + Na]+ as the base peak, and these are difficult to fragment via low energy collision-induced dissociation to yield structurally informative products; the addition of oxalic acid increases [M + H]+ ion abundances, and these fragment readily.     

Metastable decompositions of cyclopentanol ions and [C5H10O]+˙ ions with the oxygen on the first carbon

Ionized cyclopentanol and [C₅H₁₀O]+˙ ions with the oxygen on the first carbon lose methyl, ethylene, ethyl, ethane and water in their metastable decompositions. We show by collisionally activated decompositions of the products that the losses of ethyl form CH₃CH₂C?O⁺, the losses of ethylene form , and the losses of methyl probably yield . Deuterium labeling indicates that ethyl loss from ionized cyclopentanol occurs following α-cleavage of the ring, isomerization to the enol isomer of ionized n-pentanal and subsequent isomerization to the 3-pentanone ion.     

Collisional activation spectra. Behavior of [C3H7O]+ and other [CnH2n+1O]+ ions

Collisional activation spectra have identified (i) as Stable ion structures. Evidence is presented for a variety of pathway for their formation, including anchimeric assistance and hydrogen migration in less stable isomers. The fragmentation behavior of a number of [C_nH_2n+1O]⁺ isomers of n = 2 to 5 shows that extensive rearrangements are common, but that some reactions appear to be useful for ion structure elucidation. One reaction identified is unusual in that it represents the decomposition of an even-electron ion to yied an odd-electron ion product in significant abundance.     

Energetics of the loss of H2 from [C3H7]+

The internal energies of [C₃H₇]⁺ ions contributing to the metastable peak [C₃H₇]⁺ → [C₃H₅]⁺ + H₂ are higher (by perhaps > 100 kJ mol^?1) than those of the ion contributing to the threshold current in appearance energy measurements on [C₃H₅]⁺. The measured appearance energy may lead to an underestimation of the activation energy, i.e. negative ‘kinetic shift’, due to quantum, mechanical tunnelling. The distribution of energy released in the decomposition can be explained on the basis that much of the reverse activation energy and a statistical proportion of the excess energy is released as translation.