Influence of N7 Protonation on the Mechanism of the N-Glycosidic Bond Hydrolysis in2′-Deoxyguanosine. A Theoretical Study

Raquel Rios-Font, Luis Rodrıguez-Santiago,* Joan Bertran, and Mariona Sodupe*Departament de Quı´mica, UniVersitat Autonoma de Barcelona, 08193 Bellaterra (Barcelona), Spain

ReceiVed: January 31, 2007; In Final Form: March 16, 2007

The influence of N7 protonation on the mechanism of the N-glycosidic bond hydrolysis in 2′-deoxyguanosinehas been studied using density functional theory (DFT) methods. For the neutral system, two different pathways(with retention and inversion of configuration at the C1′ anomeric carbon) have been found, both of themconsisting of two steps and involving the formation of a dihydrofurane-like intermediate. The Gibbs freeenergy barrier for the first step is very high in both cases (53 and 46 kcal/mol for the process with inversionand with retention, respectively). However, the N7-protonated system shows a very different mechanismwhich consists of two steps. The first one leads to the formation of an oxacarbenium ion intermediate, witha Gibbs free energy barrier of 27 kcal/mol, and the second one corresponds to the nucleophilic attack of thewater molecule to the oxacarbenium ion and takes place with a barrier of 1.3 kcal/mol. Thus, these resultsagree with a stepwise SN1 mechanism (DN*A N), with a discrete intermediate formed between the leavinggroup and the nucleophile approach, and show that N7 protonation strongly catalyzes the hydrolysis of theN-glycosidic bond, making the guanine a better leaving group. Finally, kinetic isotope effects have beencalculated for the protonated system, and the results obtained are in very good agreement with experimentaldata for analogous systems.

1. Introduction

The genetic information of a cell is determined by thesequence of nucleobases along the DNA chain, and anyalteration of those bases may strongly affect the integrity ofthe hereditary material. Depurination is the loss of a purine base(guanine or adenine) as a consequence of N-glycosidic bondcleavage. It occurs spontaneously, at an estimated rate of about104 depurinations per mammalian cell per day,1 or is inducedby chemical damage, in both cases leading to abasic sites thatmay be potential sources of mutagenesis. On the other hand, itis known that depurination is involved in the so-called baseexcision repair (BER) pathways, in which DNA glycosylasescatalyze the removal of damaged bases by hydrolyzing theN-glycosidic bond.2 In the last 10 years, more than 50 crystalstructures of DNA glycosylases have been solved, and a greatdeal of experimental work has been done in order to elucidatethe enzymatic mechanism of the N-glycosidic bond hydrolysis.3-13

In addition, many experimental nonenzymatic studies haveanalyzed the intrinsic chemical properties of nucleoside glyco-sidic bond hydrolysis and evaluated the influence of environ-mental and chemical agents such as pH and alkylating agents.14-18

In this context, computational chemistry can be an excellenttool in the understanding and interpretation of these kinds ofprocesses and the fundamental chemistry behind them.19-24

There seems to be general agreement on the fact thatprotonation of the nucleobase catalyzes the hydrolytic cleavageof the N-glycosidic bond by making it a better leaving group,thus facilitating the nucleophilic substitution at the anomericcarbon of the deoxyribose.25-27 The reaction mechanism of thisprocess is still unclear, and much experimental work, especiallyrelated to transition state analysis using kinetic isotope effects

studies, is being done to clarify it.26,28-31 Nevertheless, theoreti-cal calculations on these kinds of systems are scarce,32-34 andto our knowledge, no quantum chemical studies performing anexhaustive exploration of the potential energy surface anddiscussing the influence of protonation in the mechanism ofthe N-glycosidic bond hydrolysis have been done. It is knownthat nucleophilic substitutions at the anomeric carbon of sugarsproceed through oxacarbenium ion intermediates or transitionstates. Moreover, nucleophilic substitutions can be classifiedas bimolecular (SN2) or stepwise (SN1) reactions. According tothe IUPAC nomenclature of glycosidic bond hydrolysis, SN2-type processes are named ANDN and SN1-type ones are knownas DN*A N (see Scheme 1). In the first case, there is nucleophileaddition (AN) and dissociation (DN) in the transition state, anddepending on the degree of bond order to the nucleophile andthe leaving group, the mechanism is further described to be“associative” if a significant bond order exists or “dissociative”if the bonding is small. Obviously, if the two bond orders havethe same value, the reaction is synchronous. In the stepwisemechanism, the leaving group dissociation (DN) is followed bynucleophile addition (AN), and the “*’’ sign indicates that ashort-lived intemediate oxacarbenium ion is formed.

In the present work, the reaction mechanism of the N-glycosidic bond hydrolysis in 2′-deoxyguanosine (dG) with asingle water molecule has been analyzed by means of compu-tational chemistry. Moreover, in order to evaluate the effectsof protonation on the energetics and the mechanism of theprocess, the hydrolysis of the N7-protonated system has alsobeen studied.

2. Computational Details

Methods based on the density functional theory (DFT) havebeen proved to be cost-effective methods to describe the

* Corresponding authors. E-mail: luis@qf.uab.es (L.R.-S.); mariona@qf.uab.es (M.S.).

6071J. Phys. Chem. B2007,111,6071-6077

10.1021/jp070822j CCC: $37.00 © 2007 American Chemical SocietyPublished on Web 05/04/2007

structure and thermochemistry of systems of considerably largesize. However, when considering energy barriers of nucleophilicsubstitutions such as SN2 reactions, they can lead to results thatsignificantly vary depending on the amount of exact exchangeintroduced in the functional.35,36The hydrolytic cleavage of theN-glycosidic bond is a nucleophilic substitution, which may beeither an SN2 reaction or a stepwise SN1 one. Thus, it isimportant to calibrate the functional used to describe theenergetics of the process. Because of that, we have firstperformed a calibration study for the hydrolytic cleavage of theN9-C bond in the 9-methylguanine (GCH3) model system usingthree DFT approaches: (i) the nonlocal hybrid three-parameterfunctional B3LYP (20% HF exchange), (ii) the hybrid metafunctional MPWB1K (44% HF exchange), and (iii) the nonlocalhybrid half and half functional BHandHLYP (50% HF ex-change). In addition, single point calculations using the highlycorrelated CCSD(T) method were performed at both the B3LYPand MPWB1K optimized geometries.

For both the model 9-methylguanine and dG systems, thepotential energy surface in the gas phase has been exploredwithout imposing any geometry constraints, the full geometryoptimizations and harmonic vibrational frequencies beingcomputed with the standard 6-31++G(d,p) basis set. Moreover,intrinsic reaction coordinate (IRC) calculations were carried outfor all of the transition structures in order to confirm theirconnectivity with the desired minima.

Finally, thermochemical corrections to the energy values havebeen computed using the standard rigid-rotor/harmonic oscillator

formulas.37 When considered, solvent effects have been intro-duced using the continuum CPCM model38 and performingsingle point calculations at the gas phase optimized geometries.Atomic charges and spin densities have been obtained fromnatural population analysis,39 and natural bond order analysishas also been carried out.40 All calculations have been performedwith the Gaussian 03 package.41

3. Results and Discussion

3.1. Benchmark Calculations.As mentioned, in order toanalyze the behavior of DFT methods for the hydrolysis of theN-C bond, we have performed calculations using the B3LYP,MPWB1K, and BHandHLYP density functional methods for9-methylguanine and compared them to single point CCSD(T)calculations. Figure 1 shows the energy profile obtained withall of these methods. It can be observed that all functionalsprovide essentially the same results, with the mechanismoccurring through two steps. In the first one, the N9-Cmethyl

bond is heterolytically broken while a new O-Cmethyl bond isformed as a consequence of the nucleophilic attack of the watermolecule. Protonation of the N3 position of guanine with an Hatom of the water molecule leads to an intermediate in whichthe methanol and guanine molecules interact through twohydrogen bonds. The second step is a proton transfer, from theN3 to the N9 position of guanine, assisted by the hydroxyl groupof the methanol molecule, leading to the canonical tautomer ofguanine.

Optimized structures of reactants, products, and transitionstates obtained with the three functionals are basically identical,the largest difference (0.09 Å) corresponding to the N9-Cmethyl

distance in TS1 (see Figure 2). In fact, root mean squaredeviation (rmsd) values considering all bond distances aresmaller than 0.01 Å for the three DFT methods in minimastructures. For transition state geometries, deviations slightlyincrease to 0.02 Å (see the Supporting Information). Relativeenergies obtained with the different functionals are quite similarexcept for the first step, which is around 8-9 kcal/mol lowerwith B3LYP than with MPWB1K or BHandHLYP, andcompare reasonably well with the single point CCSD(T) results.It should be mentioned that CCSD(T) values differ by less than0.3 kcal/mol regardless of whether we use the optimized B3LYP

SCHEME 1: IUPAC Nomenclature for the ReactionMechanisms of Glycosidic Bond Hydrolysis (Adaptedfrom ref 2)

Figure 1. Reaction mechanism and potential energies (kcal/mol) for the hydrolytic cleavage of the N9-Cmethyl bond in GCH3. The superscript “a”in the upper left-hand corner of the figure indicates the following: CCSD(T) single point calculations on the B3LYP geometries.

6072 J. Phys. Chem. B, Vol. 111, No. 21, 2007 Rıos-Font et al.

or MPWB1K geometries. Moreover, the CCSD(T) energybarrier for the first step, which is the rate determining step ofthe reaction, lies just between the B3LYP and MPWB1K values.Since the different functionals provide similar results and mostof the theoretical data published for nucleobases have beenobtained with B3LYP, we have chosen this functional forstudying the hydrolysis of dG.

3.2. 2′-Deoxyguanosine.In order to analize the influence ofN7 protonation on the hydrolysis of the N-glycosidic bond ofdG, we have explored the potential energy surface of both theneutral and N7-protonated system. Scheme 2 shows the systemconsidered in this study. Note that we have taken a slightlymodified dG, in which the hydroxyl group of C3′ has beensubstituted by OCH3, to avoid spurious hydrogen bonds,especially with the incoming water molecule. Since thesehydrogen bonds do not take place with the O5′ hydroxyl group,we have not used methyl capping on H5′. Moreover, Scheme 2

also shows the three different orientations of the water moleculeconsidered in the nucleophilic attack.

3.2.1. Neutral System.When the attack takes place on the iorientation, the O4′-C1′ endocyclic bond of the deoxyribosebreaks and the ring opens. This fact is not surprising, since ringopening is one of the steps involved in the catalytic cleavageof the N-glycosidic bond carried out by bifunctional DNAglycosylases.42,43 These enzymes use a terminal amino groupas a nucleophile, first opening the sugar ring and then forminga Schiff base, which eventually leads to the final products.However, in our case, the nucleophile is not an amino groupbut a water molecule, and thus, the formation of a Shiff basedoes not take place. Moreover, calculations show that the ring-opened intermediate that results from the nucleophilic attackin this orientation lies very high in energy (around 60 kcal/mol) from the asymptote, and thus, this mechanism has beendisregarded. In fact, experimental work carried out by Zoltewiczand co-workers27 on acid-catalyzed hydrolysis of some purinenucleosides provides good evidence against the pathway involv-ing C-O cleavage and strongly suggests that the mechanismproceeds through protonation of the nucleobase followed bycleavage of the N-glycosidic bond.

Nucleophilic attack on both the ii and iii orientations doescause the breakage of the N-glycosidic bond, the formerimplying an inversion of the configuration at the anomeric C1′carbon and the latter implying retention of its configuration. Inboth cases, the first step of the reaction produces a dihydro-furane-like intermediate and a guanine molecule by transferringthe Ha2′ proton from the deoxyribose to the N9 of guanine.The second step consists of the addition of a water molecule toa double C-C bond. Figures 3 and 4 show the Gibbs free energyprofiles and the optimized geometries of the reactant, transitionstate structures, intermediate, and product corresponding to eachprocess. Since the process of interest, that is, the cleavage ofthe N-glycosidic bond, takes place in the first step, in thefollowing discussion, we will focus on the energetic andgeometrical aspects of this step.

It can be observed that, in both orientations, the first transitionstate structure has the N9-C1′ bond already dissociated, thebond order being zero. Consequently, there is an important

Figure 2. Relevant bond distances (Å) for the stationary points.

SCHEME 2: Orientations Considered for theNucleophilic Attack of the Water Moleculea

a The numbering of the most relevant centers is indicated in red.

N-Glycosidic Bond Hydrolysis in 2′-Deoxyguanosine J. Phys. Chem. B, Vol. 111, No. 21, 20076073

charge separation between the two fragments. In the processwith inversion, the charges over the guanine and the deoxyriboseare -0.89 and+0.87, respectively, and in the process withretention, they are-0.77 and+0.80, respectively, in agreementwith previous results which analyze the dissociation of neutraldG.33,44 In both cases, the positive charge on the C1′ atom aswell as the short bond distances between O4′-C1′ and C1′-C2′ can be interpreted as a proof of the oxacarbenium ioncharacter of the deoxyribose in the transition state. There is alsoan important geometrical change when going from the reactantsto transition state structures, and it is related to the sugarconformation, usually known as puckering. One way ofquantitatively describing the puckering of a five-member ringis via the phase angle of pseudorotation,P (eq 1), which is afunction of the five internal torsion angles of the ring,45 withτ0 being the torsion angle about O4′-C1′, τ1 the one aboutC1′-C2′, τ2 the one about C2′-C3′, and so on in a clockwisemanner.

The most stable conformations found on the great majorityof riboses and deoxyriboses are C3′-endo (0° e P e 36°) andC2′-endo (144° e P e 180°). In our case, the conformation ofdG is C2′-endo, withP ) 149.3°. However, theP value for thefirst transition state structure of the process with inversion is196.2°, which corresponds to a C3′-exo conformation. Even

though this conformation is energetically less favorable thanC2′-endo by around 2 kcal/mol, this change in the puckeringallows a better hyperconjugative effect, which consists of acertain stabilization of the oxacarbenium ion in the transitionstate due to electron donation from theσ-bonded electrons atthe 2′ position to the p orbital of the C1′ atom. This has beenconfirmed by NBO analysis which shows that the stabilizationenergy gained by donation from C2′-Ha2′ to the vacant p orbitalon C1′ is 9.3 kcal/mol. For the process with retention, we alsoobserved a change on the sugar puckering from C2′-endo toC3′-exo, but in this case, the NBO analysis does not show anystabilization due to hyperconjugative effects. Anyway, in bothcases, the C1′-N9 distance is significantly large (2.98 and3.29 Å, respectively), which leads to an sp2 hybridization onC1′, thus hindering a C2′-endo conformation.

Regarding the intermediate, in both processes, we obtain adihydrofurane-like species as a consequence of the protontransfer from the sugar ring to the guanine. These kinds ofintermediates have also been localized in recent theoreticalworks, as a consequence of H atoms and OH radical attack onnucleobase19 or in the study of human DNA repair proteinHOGG1 activity.20

An important difference between the two processes is therole of the water molecule, which does not take part in theprocess with inversion, since the N9 position of the guaninebecomes protonated directly with the Ha2′ atom from the sugar.However, in the process with retention, the Ha2′ is transferredto the water molecule which subsequently protonates the N9

Figure 3. Gibbs free energy profile (kcal/mol) for the hydrolysis of dG with inversion and optimized geometries of the species involved in theprocess. Bond distances in Å.

tanP )(τ4 + τ1) - (τ3 + τ0)

2τ2(sin 36+ sin 72)(1)

6074 J. Phys. Chem. B, Vol. 111, No. 21, 2007 Rıos-Font et al.

position. This fact, which could be considered as a “solventassisted” effect, lowers the Gibbs free energy barrier of theprocess by 7 kcal/mol with respect to the one with inversion.The lowering is not as large as that found in other processes,46,47

since the energy cost of this step mainly arises from the N-Cbond dissociation. Note that the heterolytic bond dissociationof dG was computed to be around 140 kcal/mol.33,44In addition,due to the charge separation in the transition state structure,the inclusion of the dielectric medium by using a continuumsolvent model reduces the energy barrier by 3.3 kcal/mol.Nevertheless, the process still has a considerably high energybarrier, mainly due to the heterolytic bond cleavage of theN-glycosidic bond, and thus, a catalytic strategy is needed sothat the reaction can take place in biological systems. In fact,the formation of the dihydrofurane-like intermediate is a wayof preserving electroneutrality in the system, since chargeseparation in the gas phase is energetically very costly.

3.2.2. N7-Protonated System.As it has already been pointedout, enzymatic studies suggest that glycosylases can catalyzethe N-glycosidic bond cleavage by activating the leavingnucleobase. In fact, many of them have acidic residues that canestablish acid-base interactions with the N7 position of thepurine, even though in most cases it is unclear which the sourceof the catalytic proton is. In order to evaluate the possiblecatalytic effects of protonation on the hydrolysis mechanism,we have studied the reaction between N7-protonated dG and awater molecule. The Gibbs free energy profile and optimizedstructures are shown in Figure 5. A first try to follow the samemechanistic route as the two found for the neutral system wasunsuccessful, and even though we could locate a minimum

corresponding to the dihydrofurane-like intermediate reason-ably close to reactants (8.5 kcal/mol), no transition state structureconnecting both minima was found. All attempts to localize sucha structure collapsed to a transition state that did not connect tothis intermediate. Instead, IRC calculations indicated that sucha transition state structure connected the reactant with theintermediate shown in Figure 5, that is, 14.4 kcal/mol above.This intermediate can be viewed as a hydrogen bond complexbetween the positively charged oxacarbenium sugar and a neutralguanine tautomer. The fact that both intermediates are close inenergy could make us think about the possibility of having anequilibrium between them and, thus, two competitive processes.However, the addition of the water molecule to the dihydro-furane-like intermediate is likely to be a much more costlyprocess than the nucleophilic attack of the water to the C1′ atomof the oxacarbenium ion. In fact, the Gibbs free energy barrierof this process is only 1.3 kcal/mol, which is not surprisingsince the high electrophilic character of oxacarbenium ionsmakes them undergo nucleophilic attack with almost nobarrier.48,49 The second step of the process implies also theprotonation of the N3 position on the guanine. Note that in thiscase we do not obtain the canonical tautomer of guanine but adifferent one which lies 4.8 kcal/mol above it, in good agreementwith previous theoretical results at the B3LYP level.50

Concerning the conformational aspects on the sugar ring, thereis a change in the puckering when going from the reactant dG(C2′-endo,P ) 155.3°) to the first transition state (C4′-endo,P ) 230.6°). In the intermediate structure, the deoxyribose ringalso adopts a C3′-exo conformation (P ) 215.5°). NBO analysisshows that these conformational changes are better associated

Figure 4. Gibbs free energy profile (kcal/mol) for the hydrolysis of dG with retention and optimized geometries of the species involved in theprocess. Bond distances in Å.

N-Glycosidic Bond Hydrolysis in 2′-Deoxyguanosine J. Phys. Chem. B, Vol. 111, No. 21, 20076075

with the fact that the highly dissociated C1′-N9 bond involvesan sp2 hybridization on the C1′ atom, which hinders a C2′-endo conformation, than with hyperconjugative effects. Also,the bond distances between O4′-C1′and C1′-C2′ are shorterthan those in the reactants, indicating a certain double bondcharacter. All of these facts together clearly suggest a highoxacarbenium ion character on the intermediate structure, whichis confirmed by the NPA charge at this fragment (+0.82), inagreement with the results found upon considering the dissocia-tion of protonated dG.33

These results clearly indicate that the hydrolysis of theN-glycosidic bond is actually catalyzed by protonation at theN7 site of guanine. Moreover, the Gibbs free energy profileobtained for this reaction, as well as the fact that a discreteintermediate has been found, indicates that this process followsa DN*A N mechanism. As it has been mentioned previously,kinetic isotope effect (KIE) measurements are widely used toelucidate the mechanisms of these types of reactions, since theycan provide us information about the coordinate motion. Thus,harmonic vibrational frequencies have been calculated on thereactant and first transition state for different isotopic substitu-tions in order to calculate the KIEs on C1′, N9, andâ-Ha2′,and then compare them with experimental results. The primary15

N9 KIE, which usually falls in the range 1.00-1.03,28 is directlyproportional to the C-N bond cleavage, and in our case, it hasbeen found to be 1.03, in good agreement with the highlydissociative character of the transition state structure. Thesecondaryâ-2Ha2′ KIE is a measure of the extension of the

hyperconjugative effect and the oxacarbenium ion character, andis related to the puckering of the sugar. The computedâ-2Ha2′KIE is 1.10, very close to the commonly found values (1.08-1.2028), which is in good agreement with a high oxacarbeniumion character. The primary C1′ KIE is associated with thereaction coordinate motion of the carbon atom in the transitionstate, and the KIEs we got for14C and13C are 1.03 and 1.01,respectively. These values lie in the limit of the rangescorresponding to a dissociative ANDN mechanism (14C KIE,1.025-1.06;13C KIE, 1.013-1.03) and a DN*A N one (14C KIE,1.01-1.02; 13C KIE, 1.005-1.01),28 and agree with experi-mental KIE measurements for similar compounds.

4. Conclusions

In order to evaluate the catalytic effect of protonation, thereaction mechanisms of N-glycosidic bond hydrolysis in bothneutral and N7-protonated 2′-deoxyguanosine (dG) have beenstudied using DFT methods. Three DFT functionals (B3LYP,BHandHLYP, and MPWB1K) were calibrated against the highlycorrelated CCSD(T) methodology for the model 9-methylgua-nine system, the results showing that all functionals providesimilar results and compare well with CCSD(T), and thus, theB3LYP functional has been adopted. This study shows that thehydrolysis mechanisms for the neutral and protonated systemsare completely different. For the neutral system, we haveexplored two different pathways depending on the orientationof the water molecule attack. Both of them consist of two steps

Figure 5. Gibbs free energy profile (kcal/mol) for the hydrolysis of N7-protonated dG and optimized geometries of the species involved in theprocess. Bond distances in Å.

6076 J. Phys. Chem. B, Vol. 111, No. 21, 2007 Rıos-Font et al.

involving the formation of a dihydrofurane-like intermediate.The Gibbs free energy barrier of the first step is 53 kcal/molfor the process with inversion, but it decreases to 46 kcal/molin the process with retention because of the solvent assistedeffect due to the water molecule. However, the mechanism foundfor the N7-protonated system leads to the formation of anoxacarbenium ion intermediate in the first step, with a Gibbsfree energy barrier of 19 kcal/mol. Subsequently, the watermolecule performs a nucleophilic attack on the oxacarbeniumion to yield the final products guanine and deoxiribose, in anearly barrierless process. The energy profile of this process aswell as the formation of a discrete intermediate indicate thatthe hydrolysis of N7-protonated dG follows a DN*A N mecha-nism. Finally, kinetic isotope effects have been calculated, theresults being in very good agreement with experimental data.These results show that protonation of guanine strongly catalyzesthe hydrolysis of the N-glycosidic bond by making the nucleo-base a better leaving group, and give support to many enzymaticstudies which propose this way as the one followed by DNAglycosylases in the BER processes.

Acknowledgment. Financial support from MEC and DUR-SI, through the CTQ2005-08797-C02-02 and SGR2005-00244projects, and the use of the Catalonia Supercomputer Center(CESCA) are gratefully acknowledged. R.R-F. is indebted tothe Ministerio de Educacio´n y Ciencia for a doctoral FPUfellowship.

Supporting Information Available: Cartesian coordinatesof all optimized systems and bond distance rmsd values of themodel system associated with different DFT methods. Thismaterial is available free of charge via the Internet at http://pubs.acs.org.

References and Notes

(1) Lindahl, T.; Nyberg, B.Biochemistry1972, 11, 3610.(2) Stivers, J. T.; Jiang, Y. L.Chem. ReV. 2003, 103, 2729.(3) Banerjee, A.; Yang, W.; Karplus, M.; Verdine, G. L.Nature2005,

434, 612.(4) Fromme, J. C.; Banerjee, A.; Huang, S. J.; Verdine, G. L.Nature

2004, 427, 652.(5) Fromme, J. C.; Verdine, G. L.J. Biol. Chem.2003, 278, 51543.(6) Fromme, J. C.; Bruner, S. D.; Yang, W.; Karplus, M.; Verdine, G.

L. Nat. Struct. Biol.2003, 10, 204.(7) Norman, D. P. G.; Chung, S. J.; Verdine, G. L.Biochemistry2003,

42, 1564.(8) Eichman, B. F.; O’Rourke, E. J.; Radicella, J. P.; Ellenberger, T.

EMBO J.2003, 22, 4898.(9) Versees, W.; Steyaert, J.Curr. Opin. Struct. Biol.2003, 13, 731.

(10) Scharer, O. D.Angew. Chem., Int. Ed.2003, 42, 2946.(11) Drohat, A. C.; Kwon, K.; Krosky, D. J.; Stivers, J. T.Nat. Struct.

Biol. 2002, 9, 659.(12) Werner, R. M.; Stivers, J. T.Biochemistry2000, 39, 14054.(13) Kuo, C. F.; McRee, D. E.; Fisher, C. L.; O’Handley, S. F.;

Cunningham, R. P.; Tainer, J. A.Science1992, 258, 434.(14) Gates, K. S.; Nooner, T.; Dutta, S.Chem. Res. Toxicol.2004, 17,

839.(15) Laayoun, A.; De´cout, J. L.; Lhomme, J.Tetrahedron Lett.1994,

35, 4989.(16) Kumar, A. M.; Nayak, R.Biochem. Biophys. Res. Commun.1990,

173, 731.(17) Remaud, G.; Zhou, X. X.; Chattopadhyaya, J.; Oivanen, M.;

Lonnberg, H.Tetrahedron1987, 43, 4453.

(18) Lonnberg, H.; Lehikoinen, P.Nucleic Acids Res.1982, 10, 4339.(19) Zhang, R. B.; Eriksson, L. A.J. Phys. Chem. B2006, 110, 23583.(20) Schyman, P.; Danielsson, J.; Pinak, M.; Laaksonen, A.J. Phys.

Chem. A2005, 109, 1713.(21) Versees, W.; Loverix, S.; Vandemeulebroucke, A.; Geerlings, P.;

Steyaert, J.J. Mol. Biol. 2004, 338, 1.(22) Thomas, A.; Field, M. J.J. Am. Chem. Soc.2002, 124, 12432.(23) Dinner, A. R.; Blackburn, G. M.; Karplus, M.Nature2001, 413,

752.(24) Rick, S. W.; Abashkin, Y. G.; Hilderbrandt, R. L.; Burt, S. K.

Proteins1999, 37, 242.(25) O’Brien, P. J.; Ellenberger, T.Biochemistry2003, 42, 12418.(26) Horenstein, B. A.; Parkin, D. W.; Estupinan, B.; Schramm, V. L.

Biochemistry1991, 30, 10788.(27) Zoltewicz, J. A.; Clark, D. F.; Sharpless, T. W.; Grahe, G.J. Am.

Chem. Soc.1970, 92, 1741.(28) Berti, P. J.; McCann, J. A. B.Chem. ReV. 2006, 106, 506.(29) Berti, P. J.; Tanaka, K. S. E.AdV. Phys. Org. Chem.2002, 37,

239.(30) Schramm, V. L.Methods Enzymol.1999, 308, 301.(31) Mentch, F.; Parkin, D. W.; Schramm, V. L.Biochemistry1987,

26, 921.(32) Cysewski, P.; Bira, D.; Bialkowski, K.THEOCHEM2004, 678,

77.(33) Baik, M. H.; Friesner, R. A.; Lippard, S. J.J. Am. Chem. Soc.2002,

124, 4495.(34) Cavalieri, E. L.; Vauthier, E. C.; Cosse´-Barbi, A.; Fliszar, S.Theor.

Chem. Acc.2000, 104, 235.(35) Gonzales, J. M.; Cox, R. S.; Brown, S. T.; Allen, W. D.; Schaefer,

H. F. J. Phys. Chem. A2001, 105, 11327.(36) Gritsenko, O. V.; Ensing, B.; Schipper, P. R. T.; Baerends, E. J.J.

Phys. Chem. A2000, 104, 8558.(37) McQuarrie, D. A.Statistical Mechanics; Harper & Row: New York,

1986.(38) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V.J. Comput. Chem.

2003, 24, 669.(39) Reed, A. E.; Curtiss, L. A.; Weinhold, F.Chem. ReV. 1988, 88,

899.(40) Glendiening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F.

NBO, version 3.1.(41) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,

M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K.N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A.D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A.G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.Gaussian03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004.

(42) McCullough, A. K.; Dodson, M. L.; Lloyd, R. S.Annu. ReV.Biochem. 1999, 68, 255.

(43) Fromme, J. C.; Banerjee, A.; Verdine, G. L.Curr. Opin. Struct.Biol. 2004, 14, 43.

(44) Rios-Font, R.; Bertran, J.; Rodrı´guez-Santiago, L.; Sodupe, M.J.Phys. Chem. B2006, 110, 5767.

(45) Altona, C.; Sundaralingam, M.J. Am. Chem. Soc.1972, 94, 8205.(46) Constantino, E.; Solans-Monfort, X.; Sodupe, M.; Bertran, J.Chem.

Phys.2003, 295, 151.(47) Rodrı´guez-Santiago, L.; Vendrell, O.; Tejero, I.; Sodupe, M.;

Bertran, J.Chem. Phys. Lett.2001, 334, 112.(48) Richard, J. P.; Williams, K. B.; Amyes, T. L.J. Am. Chem. Soc.

1999, 121, 8403.(49) Richard, J. P.Tetrahedron1995, 51, 1535.(50) Colominas, C.; Luque, F. J.; Orozco, M.J. Am. Chem. Soc.1996,

118, 6811.

N-Glycosidic Bond Hydrolysis in 2′-Deoxyguanosine J. Phys. Chem. B, Vol. 111, No. 21, 20076077