Mechanisms and energetics for N-glycosidic bond cleavage of protonated 2'-deoxyguanosine and guanosine

Nucleoside Cation Radicals: Generation, Radical-Induced Hydrogen Atom Migrations, and Ribose Ring Cleavage in the Gas Phase

Nucleoside ions that were furnished on ribose with a 2'-O-acetyl radical group were generated in the gas phase by multistep collision-induced dissociation of precursor ions tagged with radical initiator groups, and their chemistry was investigated in the gas phase. 2'-O-Acetyladenosine cation radicals were found to undergo hydrogen transfer to the acetoxyl radical from the ribose ring positions that were elucidated using specific deuterium labeling of 1'-H, 2'-H, and 4'-H and in the N-H and O-H exchangeable positions, favoring 4'-H transfer. Ion structures and transition-state energies were calculated by a combination of Born-Oppenheimer molecular dynamics and density functional theory and used to obtain unimolecular rate constants for competitive hydrogen transfer and loss of the acetoxyl radical. Migrations to the acetoxyl radical of ribose hydrogens 1'-H, 2'-H, 3'-H, and 4'-H were all exothermic, but product formation was kinetically controlled. Both Rice-Ramsperger-Kassel-Marcus (RRKM) and transition-state theory (TST) calculations indicated preferential migration of 4'-H in a qualitative agreement with the deuterium labeling results. The hydrogen migrations displayed substantial isotope effects that along with quantum tunneling affected the relative rate constants and reaction branching ratios. UV-vis action spectroscopy indicated that the cation radicals from 2'-O-acetyladenosine consisted of a mixture of isomers. Radical-driven dissociations were also observed for protonated guanosine, cytosine, and thymidine conjugates. However, for those nucleoside ions and cation radicals, the dissociations were dominated by the loss of the nucleobase or formation of protonated nucleobase ions.

Resolution of Identity in Gas-Phase Dissociations of Mono- and Diprotonated DNA Trinucleotide Codons by 15N-Labeling and Computational Structure Analysis

Dissociations of DNA trinucleotide codons as gas-phase singly and doubly protonated ions were studied by tandem mass spectrometry using ¹⁵N-labeling to resolve identity in the nucleobase loss and backbone cleavages. The monocations showed different distributions of nucleobase loss from the 5'-, middle, and 3'-positions depending on the nucleobase, favoring cytosine over guanine, adenine, and thymine in an ensemble-averaged 62:27:11:<1 ratio. The distribution for the loss of the 5'-, middle, and 3'-nucleobase was 49:18:33, favoring the 5'-nucleobase, but also depending on its nature. The formation of sequence w₂⁺ ions was unambiguously established for all codon mono- and dications. Structures of low-Gibbs-energy protomers and conformers of dAAA⁺, dGGG⁺, dCCC⁺, dTTT⁺, dACA⁺, and dATC⁺ were established by Born-Oppenheimer molecular dynamics and density functional theory calculations. Monocations containing guanine favored classical structures protonated at guanine N7. Structures containing adenine and cytosine produced classical nucleobase-protonated isomers as well as zwitterions in which two protonated bases were combined with a phosphate anion. Protonation at thymine was disfavored. Low threshold energies for nucleobase loss allowed extensive proton migration to occur prior to dissociation. Loss of the nucleobase from monocations was assisted by neighboring group participation in nucleophilic addition or proton abstraction, as well as allosteric proton migrations remote from the reaction center. The optimized structures of diprotonated isomers for dAAA²⁺ and dACA²⁺ revealed combinations of classical and zwitterionic structures. The threshold and transition-state energies for nucleobase-ion loss from dications were low, resulting in facile dissociations involving cytosine, guanine, and adenine.

Intramolecular hydrogen transfer in DNA induced by site-selective resonant core excitation

We present experimental evidence for soft X-ray induced intramolecular hydrogen transfer in the protonated synthetic tri-oligonucleotide d($^{\mbox{\footnotesize{F}}}$UAG) in the gas-phase. The trinucleotide cations were stored in a cryogenic ion trap...

What are the Potential Sites of DNA Attack by N-Acetyl-p-benzoquinone Imine (NAPQI)?

Metabolic bioactivation of small molecules can produce electrophilic metabolites that can covalently modify proteins and DNA. Paracetamol (APAP) is a commonly used over-the-counter analgesic, and its hepatotoxic side effects have been postulated to be due to the formation of the electrophilic metabolite N-acetylbenzoquinone imine (NAPQI). It has been established that NAPQI reacts to form covalent bonds to the side-chain functional groups of cysteine, methionine, tyrosine, and tryptophan residues. While there have been scattered reports that APAP can form adducts with DNA the nature of these adducts have not yet been fully characterised. Here the four deoxynucleosides, deoxyguanosine (dG), deoxyadenosine (dA), deoxycytidine (dC), and deoxythymidine (dT) were reacted with NAPQI and the formation of adducts was profiled using liquid chromatography–mass spectrometry with positive-ion mode electrospray ionisation and collision-induced dissociation. Covalent adducts were detected for dG, dA, and dC and tandem mass spectrometry (MS/MS) spectra revealed common neutral losses of deoxyribose (116 amu) arising from cleavage of the glyosidic bond with formation of the modified nucleobase. Of the four deoxynucleosides, dC proved to be the most reactive, followed by dG and dA. A pH dependence was found, with greater reactivity being observed at pH 5.5. The results of density functional theory calculations aimed at understanding the relative reactivities of the four deoxynucleosides towards NAPQI are described.

Impact of Sodium Cationization on Gas-Phase Conformations of DNA and RNA Cytidine Mononucleotides

Gas-phase conformations of the sodium-cationized forms of the 2'-deoxycytidine and cytidine mononucleotides, [pdCyd+Na]⁺ and [pCyd+Na]⁺, are examined by infrared multiple photon dissociation action spectroscopy. Complimentary electronic structure calculations at the B3LYP/6-311+G(2d,2p)//B3LYP/6-311+G(d,p) level of theory provide candidate conformations and their respective predicted IR spectra for comparison across the IR fingerprint and hydrogen-stretching regions. Comparisons of the predicted IR spectra and the measured infrared multiple photon dissociation action spectra provide insight into the impact of sodium cationization on intrinsic mononucleotide structure. Further, comparison of present results with those reported for the sodium-cationized cytidine nucleoside analogues elucidates the impact of the phosphate moiety on gas-phase structure. Across the neutral, protonated, and sodium-cationized cytidine mononucleotides, a preference for stabilization of the phosphate moiety and nucleobase orientation is observed, although the details of this stabilization differ with the state of cationization. Several low-energy conformations of [pdCyd+Na]⁺ and [pCyd+Na]⁺ involving several different orientations of the phosphate moiety and sugar puckering modes are observed experimentally.

Structures and Relative Glycosidic Bond Stabilities of Protonated 2'-Fluoro-Substituted Purine Nucleosides

The 2'-substituent is the primary distinguishing feature between DNA and RNA nucleosides. Modifications to this critical position, both naturally occurring and synthetic, can produce biologically valuable nucleoside analogues. The unique properties of fluorine make it particularly interesting and medically useful as a synthetic nucleoside modification. In this work, the effects of 2'-fluoro modification on the protonated gas-phase purine nucleosides are examined using complementary tandem mass spectrometry and computational methods. Direct comparisons are made with previous studies on related nucleosides. Infrared multiple photon dissociation action spectroscopy performed in both the fingerprint and hydrogen-stretching regions allows for the determination of the experimentally populated conformations. The populated conformers of protonated 2'-fluoro-2'-deoxyadenosine, [Adofl+H]⁺, and 2'-fluoro-2'-deoxyguanosine, [Guofl+H]⁺, are highly parallel to their respective canonical DNA and RNA counterparts. Both N3 and N1 protonation sites are accessed by [Adofl+H]⁺, stabilizing syn and anti nucleobase orientations, respectively. N7 protonation and anti nucleobase orientation dominates in [Guofl+H]⁺. Spectroscopically observable intramolecular hydrogen-bonding interactions with fluorine allow more definitive sugar puckering determinations than possible for the canonical systems. [Adofl+H]⁺ adopts C2'-endo sugar puckering, whereas [Guofl+H]⁺ adopts both C2'-endo and C3'-endo sugar puckering. Energy-resolved collision-induced dissociation experiments with survival yield analyses provide relative glycosidic bond stabilities. The N-glycosidic bond stabilities of the protonated 2'-fluoro-substituted purine nucleosides are found to exceed those of their canonical analogues. Further, the N-glycosidic bond stability is found to increase with increasing electronegativity of the 2'-substituent, i.e., H < OH < F. The N-glycosidic bond stability is also greater for the adenine nucleoside analogues than the guanine nucleoside analogues.

Relative glycosidic bond stabilities of naturally occurring methylguanosines: 7-methylation is intrinsically activating

The frequency and diversity of posttranscriptional modifications add an additional layer of chemical complexity beyond canonical nucleic acid sequence. Methylations are particularly frequently occurring and often highly conserved throughout the kingdoms of life. However, the intricate functions of these modified nucleic acid constituents are often not fully understood. Systematic foundational research that reduces systems to their minimum constituents may aid in unraveling the complexities of nucleic acid biochemistry. Here, we examine the relative intrinsic N-glycosidic bond stabilities of guanosine and five naturally occurring methylguanosines (O2'-, 1-, 7-, N2,N2-di-, and N2,N2,O2'-trimethylguanosine) probed by energy-resolved collision-induced dissociation tandem mass spectrometry and complemented with quantum chemical calculations. Apparent glycosidic bond stability is generally found to increase with increasing methyl substitution (canonical < mono- < di- < trimethylated). Many biochemical transformations, including base excision repair mechanisms, involve protonation and/or noncovalent interactions to increase nucleobase leaving-group ability. The protonated gas-phase methylguanosines require less activation energy for glycosidic bond cleavage than their sodium cationized forms. However, methylation at the N7 position intrinsically weakens the glycosidic bond of 7-methylguanosine more significantly than subsequent cationization, and thus 7-methylguanosine is suggested to be under perpetually activated conditions. N7 methylation also alters the nucleoside geometric preferences relative to the other systems, including the nucleobase orientation in the neutral form, sugar puckering in the protonated form, and the preferred protonation and sodium cation binding sites. All of the methylated guanosines examined here are predicted to have proton affinities and gas-phase basicities that exceed that of canonical guanosine. Additionally, the proton affinity and gas-phase basicity trends exhibit a roughly inverse correlation with the apparent glycosidic bond stabilities.

Proton Transfer Accounting for Anomalous Collision-Induced Dissociation of Proton-Bound Hoogsteen Base Pair of Cytosine and Guanine

To understand the anomalous collision-induced dissociation (CID) behavior of the proton-bound Hoogsteen base pair of cytosine (C) and guanine (G), C:H⁺∙∙∙G, we investigated CID of a homologue series of proton-bound heterodimers of C, 1-methylcytosine, and 5-methylcytosine with G as a common base partner. The CID experiments were performed in an energy-resolved way (ER-CID) under both multiple and near-single collision conditions. The relative stabilities of the protonated complexes examined by ER-CID suggested that the proton-bound complexes produced by electrospray ionization in this study are proton-bound Hoogsteen base pairs. On the other hand, in contrast to the other base pairs, CID of C:H⁺∙∙∙G exhibited more abundant productions of C:H⁺, the fragment protonated on the moiety with a smaller proton affinity, than that of G:H⁺. This appeared to contradict general prediction based on the kinetic method. However, further theoretical exploration of potential energy surfaces found that there can be facile proton transfers in the proton-bound Hoogsteen base pairs during the CID process, which makes the process accessible to an additional product state of O-protonated C for C:H⁺ fragments. The presence of an additional dissociation channel, which in other words corresponds to twofold degeneracy in the transition state leading to C:H⁺ fragments, effectively doubles the apparent reaction rate for production of C:H⁺. In this way, the process gives rise to the anomaly, the observed pronounced formation of C:H⁺ in the CID of the proton-bound Hoogsteen base pair, C:H⁺∙∙∙G. Graphical Abstract ᅟ.

Modified Quadrupole Ion Trap Mass Spectrometer for Infrared Ion Spectroscopy: Application to Protonated Thiated Uridines

Modifications to a Paul-type quadrupole ion trap mass spectrometer providing optical access to the trapped ion cloud as well as hardware and software for coupling to a table-top IR optical parametric oscillator laser (OPO) are detailed. Critical experimental parameters for infrared multiple photon dissociation (IRMPD) on this instrument are characterized. IRMPD action spectra, collected in the hydrogen-stretching region with this instrument, complemented by spectra in the IR fingerprint region acquired at the FELIX facility, are employed to characterize the structures of the protonated forms of 2-thiouridine, [s²Urd+H]⁺, and 4-thiouridine, [s⁴Urd+H]⁺. The measured spectra are compared with predicted linear IR spectra calculated at the B3LYP/6-311+G(d,p) level of theory to determine the conformers populated in the experiments. This comparison indicates that thiation at the 2- or 4-positions shifts the protonation preference between the 2,4-H tautomer and 4-protonation in opposite directions versus canonical uridine, which displays a roughly equal preference for the 2,4-H tautomer and O4 protonation. As found for canonical uridine, protonation leads to a mixture of conformers exhibiting C2'-endo and C3'-endo sugar puckering with an anti nucleobase orientation being populated for both 2- and 4-thiated uridine. Graphical Abstract ᅟ.

Structural and Energetic Effects of O2'-Ribose Methylation of Protonated Purine Nucleosides

The chemical difference between DNA and RNA nucleosides is their 2'-hydrogen versus 2'-hydroxyl substituents. Modification of the ribosyl moiety at the 2'-position and 2'-O-methylation in particular, is common among natural post-transcriptional modifications of RNA. 2'-Modification may alter the electronic properties and hydrogen-bonding characteristics of the nucleoside and thus may lead to enhanced stabilization or malfunction. The structures and relative glycosidic bond stabilities of the protonated forms of the 2'-O-methylated purine nucleosides, 2'-O-methyladenosine (Adom) and 2'-O-methylguanosine (Guom), were examined using two complementary tandem mass spectrometry approaches, infrared multiple photon dissociation action spectroscopy and energy-resolved collision-induced dissociation. Theoretical calculations were also performed to predict the structures and relative stabilities of stable low-energy conformations of the protonated forms of the 2'-O-methylated purine nucleosides and their infrared spectra in the gas phase. Low-energy conformations highly parallel to those found for the protonated forms of the canonical DNA and RNA purine nucleosides are also found for the protonated 2'-O-methylated purine nucleosides. Importantly, the preferred site of protonation, nucleobase orientation, and sugar puckering are preserved among the DNA, RNA, and 2'-O-methylated variants of the protonated purine nucleosides. The 2'-substituent does however influence hydrogen-bond stabilization as the 2'-O-methyl and 2'-hydroxyl substituents enable a hydrogen-bonding interaction between the 2'- and 3'-substituents, whereas a 2'-hydrogen atom does not. Further, 2'-O-methylation reduces the number of stable low-energy hydrogen-bonded conformations possible and importantly inverts the preferred polarity of this interaction versus that of the RNA analogues. Trends in the CID_50% values extracted from survival yield analyses of the 2'-O-methylated and canonical DNA and RNA forms of the protonated purine nucleosides are employed to elucidate their relative glycosidic bond stabilities. The glycosidic bond stability of Adom is found to exceed that of its DNA and RNA analogues. The glycosidic bond stability of Guom is also found to exceed that of its DNA analogue; however, this modification weakens this bond relative to its RNA counterpart. The glycosidic bond stability of the protonated purine nucleosides appears to be correlated with the hydrogen-bond stabilization of the sugar moiety.

The intrinsic basicity of the phosphate backbone exceeds that of uracil and thymine residues: protonation of the phosphate moiety is preferred over the nucleobase for pdThd and pUrd

The gas-phase conformations of the protonated forms of thymidine-5'-monophosphate and uridine-5'-monophosphate, [pdThd+H]⁺ and [pUrd+H]⁺, are investigated by infrared multiple photon dissociation (IRMPD) action spectroscopy and electronic structure calculations. The IRMPD action spectra of [pdThd+H]⁺ and [pUrd+H]⁺ are measured over the IR fingerprint and hydrogen-stretching regions using the FELIX free electron laser and an OPO/OPA laser system. Low-energy conformations of [pdThd+H]⁺ and [pUrd+H]⁺ and their relative stabilities are computed at the MP2(full)/6-311+G(2d,2p)//B3LYP/6-311+G(d,p) and B3LYP/6-311+G(2d,2p)//B3LYP/6-311+G(d,p) levels of theory. Comparisons of the measured IRMPD action spectra and B3LYP/6-311+G(d,p) linear IR spectra computed for the low-energy conformers indicate that the dominant conformers of [pdThd+H]⁺ and [pUrd+H]⁺ populated in the experiments are protonated at the phosphate oxo oxygen atom, with a syn nucleobase orientation that is stabilized by strong P[double bond, length as m-dash]OH⁺O2 and P-OHO4' hydrogen-bonding interactions, and C2'-endo sugar puckering. Minor abundance of conformers protonated at the O2 carbonyl of the nucleobase residue may also contribute for [pdThd+H]⁺, but do not appear to be important for [pUrd+H]⁺. Comparisons to previous IRMPD spectroscopy investigations of the protonated forms of thymidine and uridine, [dThd+H]⁺ and [Urd+H]⁺, and the deprotonated forms of pdThd and pUrd, [pdThd-H]^- and [pUrd-H]^-, provide insight into the effects of the phosphate moiety and protonation on the conformational features of the nucleobase and sugar moieties. Most interestingly, the thymine and uracil nucleobases remain in their canonical forms for [pdThd+H]⁺ and [pUrd+H]⁺, unlike [dThd+H]⁺ and [Urd+H]⁺, where protonation occurs on the nucleobases and induces tautomerization of the thymine and uracil residues.

IRMPD Action Spectroscopy, ER-CID Experiments, and Theoretical Studies of Sodium Cationized Thymidine and 5-Methyluridine: Kinetic Trapping During the ESI Desolvation Process Preserves the Solution Structure of [Thd+Na]<sup/>

Thymidine (dThd) is a fundamental building block of DNA nucleic acids, whereas 5-methyluridine (Thd) is a common modified nucleoside found in tRNA. In order to determine the conformations of the sodium cationized thymine nucleosides [dThd+Na]⁺ and [Thd+Na]⁺ produced by electrospray ionization, their infrared multiple photon dissociation (IRMPD) action spectra are measured. Complementary electronic structure calculations are performed to determine the stable low-energy conformations of these complexes. Geometry optimizations and frequency analyses are performed at the B3LYP/6-311+G(d,p) level of theory, whereas energies are calculated at the B3LYP/6-311+G(2d,2p) level of theory. As protonation preferentially stabilizes minor tautomers of dThd and Thd, tautomerization facilitated by Na⁺ binding is also considered. Comparisons of the measured IRMPD and computed IR spectra find that [dThd+Na]⁺ prefers tridentate (O2,O4',O5') coordination to the canonical 2,4-diketo form of dThd with thymine in a syn orientation. In contrast, [Thd+Na]⁺ prefers bidentate (O2,O2') coordination to the canonical 2,4-diketo tautomer of Thd with thymine in an anti orientation. Although 2,4-dihydroxy tautomers and O2 protonated thymine nucleosides coexist in the gas phase, no evidence for minor tautomers is observed for the sodium cationized species. Consistent with experimental observations, the computational results confirm that the sodium cationized thymine nucleosides exhibit a strong preference for the canonical form of the thymine nucleobase. Survival yield analyses based on energy-resolved collision-induced dissociation (ER-CID) experiments suggest that the relative stabilities of protonated and sodium cationized dThd and Thd follow the order [dThd+H]⁺ < [Thd+H]⁺ < [dThd+Na]⁺ < [Thd+Na]⁺. Graphical Abstract ᅟ.

Effects of sodium cationization versus protonation on the conformations and N-glycosidic bond stabilities of sodium cationized Urd and dUrd: solution conformation of [Urd+Na]+ is preserved upon ESI

Uridine (Urd) is one of the naturally occurring pyrimidine nucleosides of RNA. 2'-Deoxyuridine (dUrd) is a naturally occurring modified form of Urd, but is not one of the canonical DNA nucleosides. In order to understand the effects of sodium cationization on the conformations and energetics of Urd and dUrd, infrared multiple photon dissociation (IRMPD) action spectroscopy experiments and density functional theory (DFT) calculations are performed. By comparing the calculated IR spectra of [Urd+Na]⁺ and [dUrd+Na]⁺ with the measured IRMPD spectra, the stable low-energy conformers populated in the experiments are determined. Anti oriented bidentate O2 and O2' binding conformers of [Urd+Na]⁺ are the dominant conformers populated in the experiments, whereas syn oriented tridentate O2, O4', and O5' binding conformers of [dUrd+Na]⁺ are dominantly populated in the experiments. The 2'-hydroxyl substituent of Urd stabilizes the anti oriented O2 binding conformers of [Urd+Na]⁺. Significant differences between the measured IRMPD and calculated IR spectra for complexes of [Urd+Na]⁺ and [dUrd+Na]⁺ involving minor tautomeric forms of the nucleobase make it obvious that none are populated in the experiments. Survival yield analyses based on energy-resolved collision-induced dissociation (ER-CID) experiments suggest that the relative stabilities of protonated and sodium cationized Urd and dUrd follow the order: [dUrd+H]⁺ < [Urd+H]⁺ < [dUrd+Na]⁺ < [Urd+Na]⁺. The 2'-deoxy modification is found to weaken the glycosidic bond of dUrd versus that of Urd for the sodium cationized uridine nucleosides.

Gas-Phase Conformations and N-Glycosidic Bond Stabilities of Sodium Cationized 2'-Deoxyguanosine and Guanosine: Sodium Cations Preferentially Bind to the Guanine Residue

2'-Deoxyguanosine (dGuo) and guanosine (Guo) are fundamental building blocks of DNA and RNA nucleic acids. In order to understand the effects of sodium cationization on the gas-phase conformations and stabilities of dGuo and Guo, infrared multiple photon dissociation (IRMPD) action spectroscopy experiments and complementary electronic structure calculations are performed. The measured IRMPD spectra of [dGuo+Na]⁺ and [Guo+Na]⁺ are compared to calculated IR spectra predicted for the stable low-energy structures computed for these species to determine the most favorable sodium cation binding sites, identify the structures populated in the experiments, and elucidate the influence of the 2'-hydroxyl substituent on the structures and IRMPD spectral features. These results are compared with those from a previous IRMPD study of the protonated guanine nucleosides to elucidate the differences between sodium cationization and protonation on structure. Energy-resolved collision-induced dissociation (ER-CID) experiments and survival yield analyses of protonated and sodium cationized dGuo and Guo are performed to compare the effects of these cations toward activating the N-glycosidic bonds of these nucleosides. For both [dGuo+Na]⁺ and [Guo+Na]⁺, the gas-phase structures populated in the experiments are found to involve bidentate binding of the sodium cation to the O6 and N7 atoms of guanine, forming a 5-membered chelation ring, with guanine found in both anti and syn orientations and C2'-endo (²T₃ or ³T₂) puckering of the sugar. The ER-CID results, IRMPD yields and the computed C1'-N9 bond lengths indicate that sodium cationization activates the N-glycosidic bond less effectively than protonation for both dGuo and Guo. The 2'-hydroxyl substituent of Guo is found to impact the preferred structures very little except that it enables a 2'OH···3'OH hydrogen bond to be formed, and stabilizes the N-glycosidic bond relative to that of dGuo in both the sodium cationized and protonated complexes.

Influence of Sodium Cationization versus Protonation on the Gas-Phase Conformations and Glycosidic Bond Stabilities of 2'-Deoxyadenosine and Adenosine

The influence of noncovalent interactions with a sodium cation on the gas-phase structures and N-glycosidic bond stabilities of 2'-deoxyadenosine (dAdo) and adenosine (Ado), [dAdo+Na](+) and [Ado+Na](+), are probed via infrared multiple photon dissociation (IRMPD) action spectroscopy and energy-resolved collision-induced dissociation (ER-CID) experiments. ER-CID experiments are also performed on the protonated forms of these nucleosides, [dAdo+H](+) and [Ado+H](+), for comparison purposes. Complementary electronic structure calculations are performed to determine the structures and relative stabilities of the stable low-energy conformations of the sodium cationized nucleoside complexes and to predict their IR spectra. Comparison between the measured IRMPD action spectra and calculated IR spectra enables the conformations of the sodium cationized nucleosides present in the experiments to be elucidated. The influence of sodium cationization versus protonation on the structures and IR spectra is elucidated by comparison to IRMPD and theoretical results previously reported for the protonated forms of these nucleosides. The influence of sodium cationization versus protonation on the glycosidic bond stability of the adenine nucleosides is determined by comparison of the ER-CID behavior of these systems. All structures present in the experiments are found to involve tridentate binding of Na(+) to the N3, O4', and O5' atoms forming favorable 5- and 6-membered chelation rings, which requires that adenine rotate to a syn configuration. This mode of sodium cation binding results in moderate flexibility of the sugar moiety such that the sugar puckering of the conformations present varies between C2'-endo and O4'-endo. Sodium cationization is found to be less effective toward activating the N-glycosidic bond than protonation for both dAdo and Ado. Both the IRMPD yields and ER-CID behavior indicate that the 2'-hydroxyl substituent of Ado stabilizes the N-glycosidic bond relative to that of dAdo.

Tautomerization lowers the activation barriers for N-glycosidic bond cleavage of protonated uridine and 2'-deoxyuridine

The gas-phase conformations of protonated uridine, [Urd+H](+), and its 2'-deoxy form, protonated 2'-deoxyuridine, [dUrd+H](+), have been examined in detail previously by infrared multiple photon dissociation action spectroscopy techniques. Both 2,4-dihydroxy tautomers and O4 protonated conformers of [Urd+H](+) and [dUrd+H](+) were found to coexist in the experiments with the 2,4-dihydroxy tautomers dominating the population. In the present study, the kinetic energy dependence of the collision-induced dissociation behavior of [Urd+H](+) and [dUrd+H](+) are examined using a guided ion beam tandem mass spectrometer to probe the mechanisms and energetics for activated dissociation of these protonated nucleosides. The primary dissociation pathways observed involve N-glycosidic bond cleavage leading to competitive elimination of protonated or neutral uracil. The potential energy surfaces (PESs) for these N-glycosidic bond cleavage pathways are mapped out via electronic structure calculations for the mixture of 2,4-dihydroxy tautomers and O4 protonated conformers of [Urd+H](+) and [dUrd+H](+) populated in the experiments. The calculated activation energies (AEs) and heats of reaction (ΔHrxns) for N-glycosidic bond cleavage at both the B3LYP and MP2(full) levels of theory are compared to the measured values. The agreement between experiment and theory indicates that B3LYP provides better estimates of the energetics of the species along the PESs for N-glycosidic bond cleavage than MP2, and that the 2,4-dihydroxy tautomers, which are stabilized by strong hydrogen-bonding interactions, predominantly influence the observed threshold dissociation behavior of [Urd+H](+) and [dUrd+H](+).

O2 Protonation Controls Threshold Behavior for N-Glycosidic Bond Cleavage of Protonated Cytosine Nucleosides

IRMPD action spectroscopy studies of protonated 2'-deoxycytidine and cytidine, [dCyd+H](+) and [Cyd+H](+), have established that both N3 and O2 protonated conformers coexist in the gas phase. Threshold collision-induced dissociation (CID) of [dCyd+H](+) and [Cyd+H](+) is investigated here using guided ion beam tandem mass spectrometry techniques to elucidate the mechanisms and energetics for N-glycosidic bond cleavage. N-Glycosidic bond cleavage is observed as the major dissociation pathways resulting in competitive elimination of either protonated or neutral cytosine for both protonated cytosine nucleosides. Electronic structure calculations are performed to map the potential energy surfaces (PESs) for both N-glycosidic bond cleavage pathways observed. The molecular parameters derived from theoretical calculations are employed for thermochemical analysis of the energy-dependent CID data to determine the minimum energies required to cleave the N-glycosidic bond along each pathway. B3LYP and MP2(full) computed activation energies for N-glycosidic bond cleavage associated with elimination of protonated and neutral cytosine, respectively, are compared to measured values to evaluate the efficacy of these theoretical methods in describing the dissociation mechanisms and PESs for N-glycosidic bond cleavage. The 2'-hydroxyl of [Cyd+H](+) is found to enhance the stability of the N-glycosidic bond vs that of [dCyd+H](+). O2 protonation is found to control the threshold energies for N-glycosidic bond cleavage as loss of neutral cytosine from the O2 protonated conformers is found to require ∼25 kJ/mol less energy than the N3 protonated analogues, and the activation energies and reaction enthalpies computed using B3LYP exhibit excellent agreement with the measured thresholds for the O2 protonated conformers.

Mechanisms and energetics for N-glycosidic bond cleavage of protonated adenine nucleosides: N3 protonation induces base rotation and enhances N-glycosidic bond stability

N3 protonation induces base rotation and stabilizes the syn orientation of the adenine nucleobase of [dAdo+H]⁺ and [Ado+H]⁺via formation of a strong intramolecular N3H⁺⋯O5′ hydrogen-bonding interaction, which in turn influences the mechanisms and energetics for N-glycosidic bond cleavage.