Phosphate Migration versus the Loss of Phosphoric Acid in Protonated Phosphopeptides: A Computational Study.

Residue-specific phosphorylation is a protein post-translational modification that regulates cellular functions. Experimental determination of the exact sites of protein phosphorylation provides an understanding of the signaling and processes at work for a given cellular state. Any experimental artifact that involves migration of the phosphate group during measurement is a concern, as the outcome can lead to erroneous conclusions that may confound studies on cellular signal transduction. Herein, we examine computationally the mechanism by which a phosphate group migrates from one serine residue to another serine in monoprotonated pentapeptides [BA-pSer-Gly-Ser-BB + H]+ → [BA-Ser-Gly-pSer-BB + H]+ (where BA and BB are different combinations of the three basic amino acids, histidine, lysine, and arginine). In addition to moving the phosphate group, the overall mechanism involves transferring a proton from the N-terminal amino acid, BA, to the C-terminal amino acid, BB. This is not a synchronous process, and there is a key high-energy intermediate, structure C, that is zwitterionic with both the basic amino acids protonated and the phosphate group attached to both serine residues and carrying a negative charge. The barriers to moving the phosphate group are calculated to be in the range of 219-274 kJ mol-1 at the B3LYP/6-31G(d) level. These barriers are systematically slightly lower and in good agreement with single-point energy calculations at both M06-2X/6-311++G(d,p) and MP2/6-31++G(d,p) levels. The competitive reaction, loss of phosphoric acid from the protonated pentapeptides, has a barrier in the range of 176-202 kJ mol-1 at the B3LYP/6-31G(d) level. Extension of the theory to M06-2X/6-311++G(d,p)//B3LYP/6-31G(d) and MP2/6-31++G(d,p)// B3LYP/6-31G(d) gives higher values for the loss of phosphoric acid, falling in the range of 196-226 kJ mol-1; these are comparable to the barriers against phosphate migration at the same levels of theory. For larger peptides His-pSer-(Gly)n-Ser-His, where n has values from 2 to 5, the barriers against the loss of phosphoric acid are higher than those against the phosphate group migration. This difference is most pronounced and significant when n = 4 and 5 (the differences are approximately 80 kJ mol-1 under the single-point energy calculations at the M06-2X and MP2 levels). Energy differences using two more recent functionals, M08-HX and MN15, on His-pSer-(Gly)n-Ser-His, where n = 1 and 5, are in good agreement with the M06-2X and MP2 calculations. These results provide the mechanistic rationale for phosphate migration versus other competing reactions in the gas phase under tandem mass spectrometry conditions.

Cations Derived from Fentanyls Generated by Atmospheric Pressure Photoionization in the Presence of Ammonia: An IMS-MS Study.

A set of fentanyl molecules when subjected to vacuum UV atmospheric pressure photoionization (VUV-APPI) in the presence of dopants (ammonia and anisole) shows two major bands in the ion mobility-mass spectrometry (IMS-MS) spectrum corresponding to (a) the protonated fentanyl, [M+H]+ and (b) a unique [M-74]+ ion. For the parent fentanyl, the [M-74]+ ion is at m/z 262 but, in the absence of ammonia, the product ion is shifted to m/z 245, corresponding to a difference of NH3. Collision-induced dissociations (CID) of the [M-74]+ ions for all the different fentanyls examined here show the same pattern of neutral losses, namely NH3 and HN=CH2, and the dominant product ion is at m/z 84 (shifted to m/z 98 for 3-methylfentanyl and m/z 142 and 231 for carfentanyl). Dissociation of the [M-74-NH3]+ ion derived from the fentanyls yields the same product ions as found in the electron impact (EI) ionization spectra of the fentanyls. The dissociation products of the [M-74-HN=CH2]+ ion are different, include the ion at m/z 84, and correspond to the fragmentation products of protonated norfentanyls. Theoretical modeling supports the opening of new fragmentation channels as a result of the reaction of the initially formed iminium cation with ammonia at atmospheric pressure.

Generation, Characterization, and Dissociation of Radical Cations Derived from Prolyl-glycyl-glycine.

Radical cations of an aliphatic tripeptide prolyl-glycyl-glycine (PGG•+) and its sequence ions [a3 + H]•+ and [b2 - H]•+ have been generated by collision-induced dissociation of the [Cu(Phen)(PGG)]•2+ complex, where Phen = 1,10-phenanthroline. Infrared multiple photon dissociation spectroscopy, ion-molecule reaction experiments, and theoretical calculations have been used to investigate the structures of these ions. The unpaired electron in these three radical cations is located at different α-carbons. The PGG•+ radical cation has a captodative structure with the radical at the α-carbon of the proline residue and the proton on the oxygen of the first amide group. This structure is at the global minimum on the potential energy surface (PES). By contrast, the [a3 + H]•+ and [b2 - H]•+ ions are not the lowest-energy structures on their respective PESs, and their radicals are formally located at the C-terminal and second α-carbons, respectively. Density functional theory calculations on the structures of the ternary copper(II) complex ion suggest that the charge-solvated isomer of the metal complex is the precursor ion that dissociates to give the PGG•+ radical cation. The isomer of the complex in which PGG is bound as a zwitterion dissociates to give the [a3 + H]•+ and [b2 - H]•+ ions.

Dissociative electron transfer of copper(ii) complexes of glycyl(glycyl/alanyl)tryptophan in vacuo: IRMPD action spectroscopy provides evidence of transition from zwitterionic to non-zwitterionic peptide structures.

We report herein the first detailed study of the mechanism of redox reactions occurring during the gas-phase dissociative electron transfer of prototypical ternary [CuII(dien)M]˙2+ complexes (M, peptide). The two final products are (i) the oxidized non-zwitterionic π-centered [M]˙+ species with both the charge and spin densities delocalized over the indole ring of the tryptophan residue and with a C-terminal COOH group intact, and (ii) the complementary ion [CuI(dien)]+. Infrared multiple photon dissociation (IRMPD) action spectroscopy and low-energy collision-induced dissociation (CID) experiments, in conjunction with density functional theory (DFT) calculations, revealed the structural details of the mass-isolated precursor and product cations. Our experimental and theoretical results indicate that the doubly positively charged precursor [CuII(dien)M]˙2+ features electrostatic coordination through the anionic carboxylate end of the zwitterionic M moiety. An additional interaction exists between the indole ring of the tryptophan residue and one of the primary amino groups of the dien ligand; the DFT calculations provided the structures of the precursor ion, intermediates, and products, and enabled us to keep track of the locations of the charge and unpaired electron. The dissociative one-electron transfer reaction is initiated by a gradual transition of the M tripeptide from the zwitterionic form in [CuII(dien)M]˙2+ to the non-zwitterionic M intermediate, through a cascade of conformational changes and proton transfers. In the next step, the highest energy intermediate is formed; here, the copper center is 5-coordinate with coordination from both the carboxylic acid group and the indole ring. A subsequent switch back to 4-coordination to an intermediate IM1, where attachment to GGW occurs through the indole ring only, creates the structure that ultimately undergoes dissociation.

Doubly Charged Small Organic Fragments Derived from [Ce(tripeptide)(CH3CN)m]3+ Complexes: Observation of the Elusive [bn + H]2+ Ions.

[a3 + H]2+ ions generated from Ln3+/tripeptide complexes, where Ln = La or Ce, have similar structures to the linear [an]+ ions but with protonation at both the terminal NH2 and N═CH2 groups. Ion stability is favored by having the basic secondary amine of the proline residue at the N-terminus and by an amino acid residue accommodating one of the protons on the side chain. Dissociation of [a3 + H]2+ ions derived from peptides containing only aliphatic residues is by cleavage of the second amide bond to give [b2]+ or [a2]+ ions along with internal [a1]+ ions. For [a3 + H]2+ ions containing a tryptophan residue in the central location, in addition to cleavage of the amide bond, losses of neutrals NH3, HN═CHR, (NH3 + CO), and HNCO were observed. Dissociations of some unsolvated Ln3+/tripeptide complexes gave [b3 + H]2+ ions in low abundance; formation of these [b3 + H]2+ ions was favored by the presence of a proline residue at the N-terminus and by either a histidine or tryptophan residue in the central position. Dissociation of these [b3 + H]2+ ions was by the loss of (H2O + CO) and not only CO, indicating that these ions did not have the same type of oxazolone structure as found for [bn]+ ions. Density functional theory calculations suggest that the observed [b3 + H]2+ ions of ProGlyGly were formed from [Ce(ProGlyGly)]3+ complexes in which the peptide was bound to the metal ion as an enolate. Dissociation of the slightly lower-energy complex, where the peptide is bound in the keto form, would produce an oxazolone but the high barrier required to create this isomer of the [b3 + H]2+ ion would be sufficient to result in further dissociation. Two isomers of the [b3 + H]2+ ion of ProHisGly have been created, one from the [Ce(ProHisGly)]3+ complex that characteristically dissociates by the combined loss of (H2O + CO) and the other by the loss of glycine from [ProHisGlyGly + 2H]2+. The [b3 + H]2+ ion derived from [ProHisGlyGly + 2H]2+ dissociated by the loss of only CO.

Structures and Dissociation Products of Ce/Peptide Complexes: Competition between Coordination and Charge Delocalization.

Structures of [Ce(GGG)]3+ and [Ce(GGG ? H)]2+ have been investigated by DFT calculations. The two lowest-energy structures of the triply charged metal complex have the peptide in either the iminol or conventional zwitterionic form, and these ions have almost identical energies. In the doubly charged complex, the iminol and charge-solvated structures are the best structures on the potential energy surface, but the latter is favored. In both iminol structures, the metal ion coordinates to the iminol oxygen rather than to the nitrogen, unlike in previously reported iminol-containing complexes. Triply charged [Ce(peptide)]3+ complexes are fragile and not easily isolated in a mass spectrometer, whereas the doubly charged [Ce(peptide ? H)]2+ complexes are more robust. Here, we studied the fragmentations of 37 [Ce(peptide ? H)]2+ and 30 [Ce(peptide)(peptide ? H)]2+ complexes and the results are systematically summarized. Losses of CO and/or H2O are the most commonly observed fragmentation channels for [Ce(peptide ? H)]2+ complexes and these dissociation pathways are modeled by DFT calculations. For [Ce(peptide)(peptide ? H)]2+ complexes the neutral peptide plays the role of a solvent molecule but, unlike in the dissociations of [Ce(CH3CN)(peptide ? H)]2+ complexes, the loss of the solvent molecule is not observed. Instead, fragmentation occurs by cleavage of the second amide bond of the solvating peptide molecule.

Hydrogen atom transfer in the radical cations of tryptophan-containing peptides AW and WA studied by mass spectrometry, infrared multiple-photon dissociation spectroscopy, and theoretical calculations.

Two types of radical cations of tryptophan-the π-radical cation and the protonated tryptophan-N radical-have been studied in dipeptides AW and WA. The π-radical cation produced by removal of an electron during collision-induced dissociation of a ternary Cu(II) complex was only observed for the AW peptide. In the case of WA, only the ion corresponding to the loss of ammonia, [WA-NH3] •+, was observed from the copper complex. Both protonated tryptophan-N radicals were produced by N-nitrosylation of the neutral peptides followed by transfer to the gas phase via electrospray ionization and subsequent collision-induced dissociation. The regiospecifically formed N• species were characterized by infrared multiple-photon dissociation spectroscopy which revealed that the WA tryptophan-N• radical remains the nitrogen radical, while the AW nitrogen radical rearranges into the π-radical cation. These findings are supported by the density functional theory calculations that suggest a relatively high barrier for the radical rearrangement (N• to π) in WA (156.3 kJ mol-1) and a very low barrier in AW (6.1 kJ mol-1). The facile hydrogen atom migration in the AW system is also supported by the collision-induced dissociation of the tryptophan-N radical species that produces fragments characteristic of the tryptophan π-radical cation. Gas-phase ion-molecule reactions with n-propyl thiol have also been used to differentiate between the π-radical cations (react by hydrogen abstraction) and the tryptophan-N• species (unreactive) of AW.

Loss of water from protonated polyglycines: interconversion and dissociation of the product imidazolone ions.

Collision-induced dissociation of isotopically labelled protonated pentaglycine produced two abundant [b5]+ ions, the products of the loss of water from the first and second amide groups, labelled [b5]+I and [b5]+II. IRMPD spectroscopy and DFT calculations show that these two [b5]+ ions feature N1-protonated 3,5-dihydro-4H-imidazol-4-one structures. 15N-Labelling established that some interconversion occurs between these two ions but dissociations are preferred. For both ions, DFT calculations show that the barrier to interconversion is slightly higher than those to dissociation. Dehydration of protonated hexaglycine produced three imidazolone ions. Ions [b6]+I and [b6]+II exhibit analogous CID spectra to those from [b5]+I and [b5]+II; however, the spectrum of the [b6]+III ion was dramatically different, showing losses predominantly of a further water molecule or cleavage of the second amide bond to give the glycyloxazolone (a deprotonated [b2]+ ion, labelled GlyGlyox (114 Da)) from the N-terminus. Protonated polyglycines [Glyn + H]+, where n = 7-9, all readily lose at least one water molecule. The corresponding [bn]+ ions lose either a further water molecule, an oxazolone from the N-terminus or a truncated peptide from the C-terminus. The number of amino acid residues in the latter two eliminated neutral molecules provides insight into the location of the imidazolone in the peptide chain and which oxygen was lost in the initial dehydration reaction. From this analysis, it appears that water loss from the longer protonated polyglycines is predominantly from the central residues.

Interconversion between 4-Imidazolone Ions; Isomers of [b4]+ Derived from Protonated Tetraglycine.

Collision-induced dissociations of isotopically labeled protonated tetraglycines establish that the [b4]+ ion formed by loss of water from the second amide bond (structure II) rearranges to form N1-protonated 3,5-dihydro-4H-imidazol-4-one (structure I), the product of water loss from the first amide bond. Structure II is slightly higher in energy than I (ΔH at 0 K is 5.1 kJ mol-1, as calculated at M06-2X/6-311++G-(d,p)), and the barrier to interconversion is 139.8 kJ mol-1 above I. The dominant dissociation pathway is the loss of methanimine (HN=CH2) from ion I with a barrier of 167.1 kJ mol-1, giving [GlyGlyGlyGly + H - H2O - HN=CH2]+, ion III; a minor channel, loss of NH3, has a slightly higher barrier (181.5 kJ mol-1). Using labeled glycine (13Cα) it was determined that loss of the imine is from the same residue as that from which water was initially lost. The collision-induced dissociation spectra of ion III derived from both I and II were identical, and their energy-resolved curves were also very similar. Ion III fragments by losses of a glycine molecule (the dominant channel), a water molecule, and a glycine residue (57 Da), giving ions IV, V, and VII, respectively. Isotopic labeling established the origins of each of the neutral molecules that are lost. Using glycine (2,2 D2), rapid deuterium exchange was observed for both ions I and II for the α-hydrogens that are from the same residue as that from which the water had been eliminated.

Isomerization versus dissociation of phenylalanylglycyltryptophan radical cations.

Four isomers of the radical cation of tripeptide phenylalanylglycyltryptophan, in which the initial location of the radical center is well defined, have been isolated and their collision-induced dissociation (CID) spectra examined. These ions, the π-centered [FGWπ˙]+, α-carbon- [FGα˙W]+, N-centered [FGWN˙]+ and ζ-carbon- [Fζ˙GW]+ radical cations, were generated via collision-induced dissociation (CID) of transition metal-ligand-peptide complexes, side chain fragmentation of a π-centered radical cation, homolytic cleavage of a labile nitrogen-nitrogen single bond, and laser induced dissociation of an iodinated peptide, respectively. The π-centered and tryptophan N-centered peptide radical cations produced almost identical CID spectra, despite the different locations of their initial radical sites, which indicated that interconversion between the π-centered and tryptophan N-centered radical cations is facile. By contrast, the α-carbon-glycyl radical [FGα˙W]+, and ζ-phenyl radical [Fζ˙GW]+, featured different dissociation product ions, suggesting that the interconversions among α-carbon, π-centered (or tryptophan N-centered) and ζ-carbon-radical cations have higher barriers than those to dissociation. Density functional theory calculations have been used to perform systematic mechanistic investigations on the interconversions between these isomers and to study selected fragmentation pathways for these isomeric peptide radical cations. The results showed that the energy barrier for interconversion between [FGWπ˙]+ and [FGWN˙]+ is only 31.1 kcal mol-1, much lower than the barriers to their dissociation (40.3 kcal mol-1). For the [FGWπ˙]+, [FGα˙W]+, and [Fζ˙GW]+, the barriers to interconversion are higher than those to dissociation, suggesting that interconversions among these isomers are not competitive with dissociations. The [z3 - H]˙+ ions isolated from [FGα˙W]+ and [Fζ˙GW]+ show distinctly different fragmentation patterns, indicating that the structures of these ions are different and this result is supported by the DFT calculations.

Radical-induced dissociation leading to the loss of CO2 from the oxazolone ring of [b5- H]˙(+) ions.

Macrocyclization is commonly observed in large bn(+) (n≥ 4) ions and as a consequence can lead to incorrect protein identification due to sequence scrambling. In this work, the analogous [b5- H]˙(+) radical cations derived from aliphatic hexapeptides (GA5˙(+)) also showed evidence of macrocyclization under CID conditions. However, the major fragmentation for [b5- H]˙(+) ions is the loss of CO2 and not CO loss, which is commonly observed in closed-shell bn(+) ions. Isotopic labeling using CD3 and (18)O revealed that more than one common structure underwent dissociations. Theoretical studies found that the loss of CO2 is radical-driven and is facilitated by the radical being located at the Cα atom immediately adjacent to the oxazolone ring. Comparable energy barriers against macrocyclization, hydrogen-atom transfer, and fragmentations are found by DFT calculations and the results are consistent with the experimental observations that a variety of dissociation products are observed in the CID spectra.

Nucleophilic substitution by amide nitrogen in the aromatic rings of [zn - H]˙⁺ ions; the structures of the [b2 - H - 17]˙⁺ and [c1 - 17]⁺ ions.

Peptide radical cations that contain an aromatic amino acid residue cleave to give [zn - H]˙⁺ ions with [b2 - H - 17]˙⁺ and [c1 - 17](+) ions, the dominant products in the dissociation of [zn - H]˙⁺, also present in lower abundance in the CID spectra. Isotopic labeling in the aromatic ring of [Yπ˙GG](+) establishes that in the formation of [b2 - H - 17]˙⁺ ions a hydrogen from the δ-position of the Y residue is lost, indicating that nucleophilic substitution on the aromatic ring has occurred. A preliminary DFT investigation of nine plausible structures for the [c1 - 17](+) ion derived from [Y(π)˙GG](+) shows that two structures resulting from attack on the aromatic ring by oxygen and nitrogen atoms from the peptide backbone have significantly better energies than other isomers. A detailed study of [Y(π)˙GG](+) using two density functionals, B3LYP and M06-2X, with a 6-31++G(d,p) basis set gives a higher barrier for attack on the aromatic ring of the [zn - H]˙⁺ ion by nitrogen than by the carbonyl oxygen. However, subsequent rearrangements involving proton transfers are much higher in energy for the oxygen-substituted isomer leading to the conclusion that the [c1 - 17](+) ions are the products of nucleophilic attack by nitrogen, protonated 2,7-dihydroxyquinoline ions. The [b2 - H - 17]˙⁺ ions are formed by loss of glycine from the same intermediates involved in the formation of the [c1 - 17](+) ions.

Alkali-Metal-Ion-Assisted Hydrogen Atom Transfer in the Homocysteine Radical.

Intramolecular hydrogen atom transfer (HAT) was examined in homocysteine (Hcy) thiyl radical/alkali metal ion complexes in the gas phase by combination of experimental techniques (ion-molecule reactions and infrared multiple photon dissociation spectroscopy) and theoretical calculations. The experimental results unequivocally show that metal ion complexation (as opposed to protonation) of the regiospecifically generated Hcy thiyl radical promotes its rapid isomerisation into an α-carbon radical via HAT. Theoretical calculations were employed to calculate the most probable HAT pathway and found that in alkali metal ion complexes the activation barrier is significantly lower, in full agreement with the experimental data. This is, to our knowledge, the first example of a gas-phase thiyl radical thermal rearrangement into an α-carbon species within the same amino acid residue and is consistent with the solution phase behaviour of Hcy radical.

Cysteine Radical/Metal Ion Adducts: A Gas-Phase Structural Elucidation and Reactivity Study.

The formation and investigation of sulfur-based cysteine radicals cationized by a group 1A metal ion or Ag+ in the gas phase are reported. Gas-phase ion-molecule reactions (IMR) and infrared multiple-photon dissociation (IRMPD) spectroscopy revealed that the Li+ , Na+ , and K+ adducts of the cysteine radical remain S-based radicals as initially formed. Theoretical calculations for the three alkali metal ions found that the lowest-energy isomers are Cα -based radicals, but they are not observed experimentally owing to the barriers associated with the hydrogen-atom transfer. A mechanism for the S-to-Cα radical rearrangement in the metal ion complexes was proposed, and the relative energies of the associated energy barriers were found to be Li+ >Na+ >K+ at all levels of theory. Relative to the B3LYP functional, other levels of calculation gave significantly higher barriers (by 35-40 kJ mol-1 at MP2 and 44-47 kJ mol-1 at the CCSD level) using the same basis set. Unlike the alkali metal adducts, the cysteine radical/Ag+ complex rearranged from the S-based radical to an unreactive species as indicated by IMRs and IRMPD spectroscopy. This is consistent with the Ag+ /cysteine radical complex having a lower S-to-Cα radical conversion barrier, as predicted by the MP2 and CCSD levels of theory.

Investigation of Fragmentation of Tryptophan Nitrogen Radical Cation.

This work describes investigation of the fragmentation mechanism of tryptophan N-indolyl radical cation, H3N(+)-TrpN(•) (m/z 204) studied via DFT calculations and several gas-phase experimental techniques. The main fragment ion at m/z 131, shown to be a mixture of up to four isomers including 3-methylindole (3MI) π-radical cation, was found to undergo further loss of an H atom to yield one of the two isomeric m/z 130 ions. 3-Methylindole radical cation generated independently (via CID of [Cu(II)(terpy)3MI](•2+)) displayed gas-phase reactivity partially similar to that of the m/z 131 fragment, further confirming our proposed mechanism. CID of deuterated tryptophan N-indolyl radical cation (m/z 208) suggested that up to six H atoms are involved in the pathway to formation of the m/z 131 ion, consistent with hydrogen atom scrambling during CID of protonated Trp.

Radical-induced, proton-transfer-driven fragmentations in [b(5)-H]˙(+) ions derived from pentaalanyl tryptophan.

The collision-induced dissociation (CID) of [b5 - H]˙(+) ions containing four alanine residues and one tryptophan give identical spectra regardless of the initial location of the tryptophan indicating that, as proposed for b5(+) ions, sequence scrambling occurs prior to dissociation. Cleavage occurs predominantly at the peptide bonds and at the N-Cα bond of the alanine residue that is attached to the N-terminus of the tryptophan residue. The product of the latter pathway, an ion at m/z 240, is the base peak in all the mass spectra. With the exception of one minor channel giving a b3(+) ion, the product ions retain both the tryptophan residue and the radical. Experiments with one trideuterated alanine established the sequences of loss of alanine residues. Formation of identical products implies a common intermediate, a [b5 - H]˙(+) ion that has a 'linear' structure in which the tryptophan residue is present as an α-radical located in the oxazolone ring, structure Ie. Density functional theory calculations show this structure to be at the global minimum, 14.6 kcal mol(-1) below the macrocyclic structure, ion II. Loss of CO from the [b5 - H]˙(+) ions is inhibited by the presence of the radical centre in the oxazolone ring and migration of the proton from the oxazolone ring onto the peptide backbone induces cleavage of an N-Cα or peptide bond. Three calculated structures for the ion at m/z 240 all have an oxazolone ring. Two of these structures may be formed from Ie, depending upon which proton migrates onto the peptide chain prior to the dissociation. The barrier to interconversion between these two structures requires a 1,3-hydrogen atom shift and is high (51.0 kcal mol(-1)), but both can convert into a third isomer that readily loses CO2 (barrier 38.7 kcal mol(-1)). The lowest barrier to the loss of CO, the usual fragmentation path observed for protonated oxazolones, is 47.0 kcal mol(-1).

Fragmentation of peptide radical cations containing a tyrosine or tryptophan residue: structural features that favor formation of [x(n-1) + H]˙⁺ and [z(n-1) + H]˙⁺ ions.

Peptide radical cations A(n)Y(•+) (where n = 3, 4, or 5) and A5W(•+) have been generated by collision-induced dissociation (CID) of [Cu(II)(tpy)(peptide)](•2+) complexes. Apart from the charge-driven fragmentation at the N-Cα bond of the hetero residue producing either [c + 2H](+) or [z - H](•+) ions and radical-driven fragmentation at the Cα-C bond to give a(+) ions, unusual product ions [x + H](•+) and [z + H](•+) are abundant in the CID spectra of the peptides with the hetero residue in the second or third position of the chain. The formation of these ions requires that both the charge and radical be located on the peptide backbone. Energy-resolved spectra established that the [z + H](•+) ion can be produced either directly from the peptide radical cation or via the fragment ion [x + H](•+). Additionally, backbone dissociation by loss of the C-terminal amino acid giving [b(n-1) - H](•+) increases in abundance with the length of the peptides. Mechanisms by which peptide radical cations dissociate have been modeled using density functional theory (B3LYP/6-31++G** level) on tetrapeptides AYAG(•+), AAYG(•+), and AWAG(•+).

Structures of a(n)* ions derived from protonated pentaglycine and pentaalanine: results from IRMPD spectroscopy and DFT calculations.

Infrared multiple-photon dissociation (IRMPD) spectroscopy and DFT calculations have been used to probe the most stable structures of a3(*) and a4(*) ions derived from both protonated pentaglycine (denoted G5) and pentaalanine (A5). The a3(*) and a4(*) ions derived from protonated A5 feature a CHR=N-CHR'- group at the N-terminus and an oxazolone ring at the C-terminus, as proposed previously [J. Am. Soc. Mass Spectrom. 19, 1788-1798 (2008)]. The isomeric a4(*) ion derived from A5 with a 3,5-dihydro-4H-imidazol-4-one ring structure was calculated to have a slightly better energy than the oxazolone, but the barrier to its formation is higher and there was no evidence of this ion in the IRMPD spectrum. By contrast, the a4(*) and [a4 - H2O](+) (denoted a4(0)) ions from G5 gave strikingly similar IRMPD spectra and both have the 3,5-dihydro-4H-imidazol-4-one ring structure similar to that recently reported for the [GGGG + H - H2O](+) ion [Int. J. Mass Spectrom. 316-318, 268-272 (2012)]. In the absence of a solvent molecule, the pathway to the oxazolone is calculated to be lower than those to thermodynamically more stable products, the a4(0) and the a4(*) with the 3,5-dihydro-4H-imidazol-4-one ring structure. Incorporation of one water molecule is sufficient to reduce the barrier to formation of the a4(0) of G5 to below that for formation of the oxazolone. On the equivalent potential energy surface for protonated A5 the barrier to formation of the a4(0) ion is 12.3 kcal mol(-1) higher than that for oxazolone formation and the a4(0) ion is not observed experimentally.

Metal ion complexes with HisGly: comparison with PhePhe and PheGly.

Gas-phase complexes of five metal ions with the dipeptide HisGly have been characterized by DFT computations and by infrared multiple photon dissociation spectroscopy (IRMPD) using the free electron laser FELIX. Fine agreement is found in all five cases between the predicted IR spectral features of the lowest energy structures and the observed IRMPD spectra in the diagnostic region 1500-1800 cm(-1), and the agreement is largely satisfactory at longer wavelengths from 1000 to 1500 cm(-1). Weak-binding metal ions (K(+), Ba(2+), and Ca(2+)) predominantly adopt the charge-solvated (CS) mode of chelation involving both carbonyl oxygens, an imidazole nitrogen of the histidine side chain, and possibly the amino nitrogen. Complexes with Mg(2+) and Ni(2+) are found to adopt iminol (Im) binding, involving the deprotonated amide nitrogen, with tetradentate chelation. This tetradentate coordination of Ni(II) is the preferred binding mode in the gas phase, against the expectation under condensed-phase conditions that such binding would be sterically unfavorable and overshadowed by other outcomes such as metal ion hydration and formation of dimeric complexes. The HisGly results are compared with corresponding results for the PheAla, PheGly, and PhePhe ligands, and parallel behavior is seen for the dipeptides with N-terminal Phe versus His residues. An exception is the different chelation pattern determined for PhePhe versus HisGly, reflecting the intercalation-type cation binding pocket of the PhePhe ligand. The complexes group into three well-defined spectroscopic patterns: nickel and magnesium, calcium and barium, and potassium. Factors leading to differentiation of these distinct spectroscopic categories are (1) differing propensities for choosing the iminol binding pattern, and (2) single versus double charge on the metal center. Nickel and magnesium ions show similar gas-phase binding behavior, contrasting with their quite different patterns of peptide interaction in condensed phases.

Structure and reactivity of the distonic and aromatic radical cations of tryptophan.

In this work, we regiospecifically generate and compare the gas-phase properties of two isomeric forms of tryptophan radical cations-a distonic indolyl N-radical (H3N(+) - TrpN(•)) and a canonical aromatic π (Trp(•+)) radical cation. The distonic radical cation was generated by nitrosylating the indole nitrogen of tryptophan in solution followed by collision-induced dissociation (CID) of the resulting protonated N-nitroso tryptophan. The π-radical cation was produced via CID of the ternary [Cu(II)(terpy)(Trp)](•2+) complex. CID spectra of the two isomeric species were found to be very different, suggesting no interconversion between the isomers. In gas-phase ion-molecule reactions, the distonic radical cation was unreactive towards n-propylsulfide, whereas the π radical cation reacted by hydrogen atom abstraction. DFT calculations revealed that the distonic indolyl radical cation is about 82 kJ/mol higher in energy than the π radical cation of tryptophan. The low reactivity of the distonic nitrogen radical cation was explained by spin delocalization of the radical over the aromatic ring and the remote, localized charge (at the amino nitrogen). The lack of interconversion between the isomers under both trapping and CID conditions was explained by the high rearrangement barrier of ca.137 kJ/mol. Finally, the two isomers were characterized by infrared multiple-photon dissociation (IRMPD) spectroscopy in the ~1000-1800 cm(-1) region. It was found that some of the main experimental IR features overlap between the two species, making their distinction by IRMPD spectroscopy in this region problematic. In addition, DFT theoretical calculations showed that the IR spectra are strongly conformation-dependent.