Reaction of Hydrated Electrons with Guanine Derivatives

Jan 22, 2009 - ... 101, 40129 Bologna, Italy, and University of Oxford, Gray Cancer Institute, Mount Vernon Hospital, Northwood, Middlesex, HA6 2JR, U...
1 downloads 0 Views 2MB Size
Download PDF
2170

J. Phys. Chem. B 2009, 113, 2170–2176

Reaction of Hydrated Electrons with Guanine Derivatives: Tautomerism of Intermediate Species Mila D’Angelantonio,† Marialuisa Russo,†,‡ Panagiotis Kaloudis,† Quinto G. Mulazzani,† Peter Wardman,‡ Maurizio Guerra,† and Chryssostomos Chatgilialoglu†,* ISOF, Consiglio Nazionale delle Ricerche, Via P. Gobetti 101, 40129 Bologna, Italy, and UniVersity of Oxford, Gray Cancer Institute, Mount Vernon Hospital, Northwood, Middlesex, HA6 2JR, U.K. ReceiVed: October 21, 2008; ReVised Manuscript ReceiVed: NoVember 26, 2008

Here, we show that two tautomers are produced by the protonation of the guanine-electron adduct. The fate of electron adducts of a variety of substituted guanosines was investigated by radiolytic methods and addressed computationally by means of time-dependent DFT (TD-B3LYP/6-311G**//B1B95/6-31+G**) calculations. The reaction of eaq- with guanosine and 1-methylguanosine produces two transient species, whereas the reaction with N2-ethylguanosine and N2,N2-diethylguanosine produces only one. The two short-lived intermediates, which show a substantial difference in their UV-visible spectra, are recognized to be two purine tautomers (i.e., iminic 18 and aminic 19 forms). The tautomerization 18 f 19 occurs with a rate constant of ca. 1.5 × 106 s-1, and theory suggests that it is a water-assisted process. Introduction One-electron reduction of nucleosides or oligonucleotides is of fundamental importance for better understanding the reactive intermediates related to DNA.1 Examples are one-electron reduction of thymine dimer lesion related to biological repair2 or the excess electron transfer in oligonucleotides3 potentially related to DNA chip technology.4 Nucleosides react with hydrated electrons (eaq-) at practically diffusion-controlled rates to give radical anions. In particular, radical anions of purine nucleosides are rapidly protonated by water because the pKa values are very high.5 In guanosine or 2′-deoxyguanosine (1), it was reported by Candeias et al.6 that, after fast protonation of the corresponding radical anion at the heteroatom E(O6, N3, or N7) with k g 107 s-1, the intermediate neutral radical undergoes a rapid transformation with ktaut ) 1.2 × 106 s-1 which has been attributed to protonation at C8 (Scheme 1). Indeed, the spectrum of radical 2 was formed after a delay, thus implying the existence of a precursor transient species (indicated as G(EH)•, in red in Scheme 1), possibly a tautomeric form of radical 2, but they were unable to record its spectrum. It was also reported that the rate of the reaction decreased by a factor of 8 in D2O,6 which suggest a proton transfer as the rate-determining step. An alternative path of formation of the same conjugated aminyl radical 2 involves addition of H-atom to 1 that selectively occurs at the C8 position (Scheme 1). This assignment is also based upon solid-state EPR data.7 A pKa value of 5.4 is associated with reversible protonation of adduct 2. One-electron reduction of 8-bromoguanine derivatives has been studied in some detail in our laboratories by chemical radiation techniques.8,9 These derivatives were found to give a radical anion that undergoes a fast protonation at C8 by proton transfer from solvent to afford a complex, that successively debrominates giving the corresponding one-electron-oxidized * Author to whom correspondence should be addressed. E-mail: [email protected]. † Consiglio Nazionale delle Ricerche. ‡ University of Oxford.

SCHEME 1: Reaction Mechanism for One-Electron Reduction of Guanosine Proposed by Candeias et al.6 based on Pulse Radiolysis Studies with Optical and Conductimetric Detection (G(EH)•, in Red, Is the Unidentified Tautomer)

guanine. These observations led us to reinvestigate the oneelectron reduction of guanosine in D2O and extend the same approach to a variety of substituted guanosine derivatives for a better comprehension of the mechanism in Scheme 1. Experimental Section Materials. All chemicals and solvents were purchased from Sigma-Aldrich and used as received. Solutions were freshly prepared using water purified with a Millipore (Milli-Q) system; D2O (Aldrich, 99.8 atom % D) was used. Substituted guanosines 3, 4, and 5 were prepared from guanosine using literature procedures.10,11 Pulse Radiolysis. Experiments were performed either at the Consiglio Nazionale delle Ricerche (CNR) or at the Gray Cancer Institute (GCI). The pulse radiolysis facilities at CNR12 and at GCI13 have been described. Pulse radiolysis with optical absorption detection was performed by using a 12 MeV linear accelerator at CNR (6 MeV at GCI), which delivered 20-200 ns electron pulses with doses between 5 and 50 Gy, by which OH•, H•, and eaq- are generated with 1-20 µM concentrations.

10.1021/jp809386c CCC: $40.75  2009 American Chemical Society Published on Web 01/22/2009

Reaction of Hydrated Electrons with Guanine Derivatives

Figure 1. Absorption spectra obtained from the pulse radiolysis of Ar-purged D2O solutions containing 1 mM guanosine at natural pD with 0.25 M t-BuOH, recorded 2 µs (black circles) and 20 µs (red circles) after the pulse; optical path ) 2.0 cm, dose per pulse ) 22 Gy. Inset: Time dependence of absorption at 300 nm from pulse radiolysis of 1 mM guanosine, 0.25 M t-BuOH unbuffered Ar-purged D2O solution, optical path ) 2.0 cm, dose ) 22 Gy; the solid line represents the linear fit to the data.

The pulse irradiations were performed at room temperature (22 ( 2 °C) on samples contained in Spectrosil quartz cells of 2 cm optical path length. Solutions were protected from the analyzing light by means of a shutter and appropriate cutoff filters. The bandwidth used throughout the pulse radiolysis experiments was 5 nm. The radiation dose per pulse was monitored by means of a charge collector placed behind the irradiation cell and calibrated with an N2O-saturated solution containing 0.1 M HCO2- and 0.5 mM methylviologen, using Gε ) 9.66 × 10-4 m2 J-1 at 602 nm.14 G(X) represents the number of moles of species X formed, consumed or altered per joule of energy absorbed by the system. The thiocyanate dosimeter (10 mM KSCN/air) was used at GCI, using calibration factors based on the recommendations of Buxton and Stuart.15 Continuous Radiolyses. Continuous radiolyses were performed at room temperature (22 ( 2 °C) on 10 mL samples using a 60Co-Gammacell with a dose rate of ∼10 Gy min-1. The exact absorbed radiation dose was determined with the Fricke chemical dosimeter, by taking G(Fe3+) ) 1.61 µmol J-1.16 The reactions of guanosine with eaq- and H• were investigated using vacuum degassed solutions containing 0.3 mM substrate and 0.25 M t-BuOH that were irradiated with appropriate doses. Quantitative HPLC studies were performed on a Waters Associates Unit equipped with a model 600 multisolvent delivery system and controller, a 486 tunable absorbance detector, and a Millenium 2010 data workstation. The aqueous solutions of the samples were loaded onto a reverse-phase analytical column (Zorbax SB-C18 column, 4.6 × 150 mm), eluted with triethylammonium acetate buffer (20 mM, pH 7) containing 0-25% acetonitrile in a nonlinear gradient over 30 min with a flow rate of 1.0 mL/min (detection at 254 nm). Absolute concentrations of guanosine were interpolated from external standard calibration curves. Computational Details. Hybrid meta DFT calculations with the B1B95 (Becke8817-Becke9518 one-parameter model for thermochemistry) functional19 were carried out using the Gaussian 03 system of programs.20 This HMDFT model was found to give excellent performance for thermochemistry19,21,22 and molecular geometries.21,22 Unrestricted wave functions were used for radical species. Optimized geometries and total energies were

J. Phys. Chem. B, Vol. 113, No. 7, 2009 2171 obtained employing the valence double-ζ basis set supplemented with polarization functions and augmented with diffuse functions on heavy atoms (6-31+G**). The nature of the ground (zero imaginary frequency) and transition (one imaginary frequency) states was verified by frequency calculations. Total energies were corrected for the zero point vibrational energy (ZPVE) computed from frequency calculations using a scaling factor of 0.9735 to account for anharmonicity.21 Thermal contributions to enthalpy H were computed by using the rotor/harmonic oscillator approximation and standard expressions for an ideal gas in the canonical ensemble at 298.15 K and 1 atm. The B3LYP method employing a triple-ζ basis set (B3LYP/6311G**//B1B95/6-31+G**) was used to compute optical transitions, since this HDFT method was found by us to provide reliable optical transitions in nucleosides.8,9,23-25 In Tables 1-3 and Scheme 2, we have reported the electronic transitions having an oscillator strength f greater than 0.01 and a wavelength λ greater than 280 nm. Results and Discussion Generation of Radicals.26,27 Radiolysis of neutral water leads to the species eaq-, HO•, and H•, as shown in eq 1 where the values in parentheses represent the chemical radiation yields (G) expressed in µmol J-1. The reactions of eaq- with substrates were studied by irradiating deoxygenated solutions containing 0.25 M t-BuOH or 0.13 M i-PrOH. The HO• radicals are scavenged efficiently by t-BuOH (eq 2, k2 ) 6.0 × 108 M-1 s-1) or i-PrOH (eq 3, k3 ) 1.9 × 109 M-1 s-1), whereas H• atoms react with i-PrOH (eq 3, k3 ) 7.4 × 107 M-1 s-1) much faster than t-BuOH (eq 2, k2 ) 1.7 × 105 M-1 s-1).

H2O ' eaq-(0.27), HO•(0.28), H•(0.062)

(1)

HO• /H• + t-BuOH f (CH3)2C(OH)CH2• + H2O/H2

(2) HO•/H• + i-PrOH f (CH3)2C•(OH) + H2O/H2

(3)

Reaction of eaq- with Guanosine (1). A rate constant of 3.3 × 109 M-1 s-1 was determined for the reactions of eaq- with 1 in heavy water (D2O) by measuring the rate of the decrease of the absorbance of eaq-, 28 which is similar to the corresponding value in H2O.6 A deaerated D2O solution of guanosine (1.0 mM) and t-BuOH (0.25 M) at natural pD was pulse irradiated, affording the spectra illustrated in Figure 1 at 2 µs (black circles) and 20 µs (red circles) after the pulse. The latter is identical to the one reported by Candeias et al.6 in H2O, exhibiting a maximum at 300 nm (ε ) 10 700 M-1 cm-1) with a shoulder and a weaker band at 460 nm (ε ) 2600 M-1 cm-1), and corresponds to the final C8 deuterated adduct. The spectrum at 2 µs contains two bands centered at 300 and 340 of similar intensity and a weak band around 440 nm. This spectrum is assigned to the first transient species obtained after the fast deuteration of the guanosine radical anion and it was not reported previously (cf. Scheme 1). The time for the transformation of the first to second transient species at 300 nm (see inset of Figure 1) followed first-order kinetics with a rate constant of (1.8 ( 0.1) × 105 s-1. Therefore, by replacing H2O with D2O, as the reaction medium, a kinetic isotope effect ktaut(H2O)/ ktaut(D2O) of 6.7 was calculated. This value is an indication of a proton transfer in the reaction path. In order to obtain information on the structure of the first intermediate, and the

2172 J. Phys. Chem. B, Vol. 113, No. 7, 2009

Figure 2. Absorption spectra obtained from the pulse radiolysis of Ar-purged H2O solutions containing 0.5 mM 1-methylguanosine (3) at natural pH with 0.25 M t-BuOH, recorded 2 µs (black circles) and 20 µs (red circles) after the pulse; optical path ) 2.0 cm, dose per pulse ) 22 Gy. Inset: Plot of kobs for the buildup observed at 300 nm versus the concentration of 3; the plateau value represents the rate constant for the tautomerization process, i.e., ktaut ) 1.7 × 106 s-1.

Figure 3. Absorption spectra obtained from the pulse radiolysis of Ar-purged D2O solutions containing 0.5 mM 1-methylguanosine (3) at natural pD with 0.25 M tBuOH, recorded 2 µs (black circles) and 20 µs (red circles) after the pulse; optical path ) 2.0 cm, dose per pulse ) 22 Gy. Inset: Time dependence of absorption at 300 nm from the pulse radiolysis of 1.52 mM 3, 0.25 M t-BuOH unbuffered Ar-purged D2O solution, optical path ) 2.0 cm, dose ) 12.4 Gy; the solid line represents the linear fit to the data.

possible tautomerization sites, differently substituted guanosine derivatives were then studied. Reaction of eaq- with 1-Methylguanosine (3). A rate constant of 2.9 × 109 M-1 s-1 was determined for the reactions of 3 with eaq- at natural pH in H2O. The optical absorption spectra obtained from the reaction of 3 with eaq- at 2 µs (black circles) and 20 µs (red circles) after the pulse are shown in Figure 2. It is worth emphasizing that the spectrum strongly resembles the one obtained from the analogous reaction of guanosine. The rates for the transformation of the first to the second transient species at 300 nm were found to increase slightly with increasing nucleoside concentrations, reaching a plateau at 3.0 mM of 3 (see inset of Figure 2). This behavior indicates that at this concentration the observed transformation

D’Angelantonio et al. is no longer limited by reaction with eaq-. The plateau value corresponds to ktaut ) 1.7 × 106 s-1, which is in accordance with the one previously reported for guanosine (1.2 × 106 s-1).6 These findings suggest that tautomerization also occurs in the case of 1-methylguanosine, i.e., in the absence of the H at the N1 position. To further investigate this reaction, an analogous experiment was performed in D2O solutions. A rate constant of 3.2 × 109 M-1 s-1 was determined for the reactions of 3 with eaq- at natural pD. The optical absorption spectra obtained after the pulse of a deaerated D2O solution containing 3 (0.5 mM) and t-BuOH (0.25 M) are presented in Figure 3. The spectrum registered at 20 µs after the pulse (red circles) is very similar to the analogous spectrum in H2O solution, whereas the spectrum obtained at 2 µs (black circles) is substantially different around 300 nm (cf. Figure 2). The time for the transformation of the first to second transient species at 300 nm (see inset of Figure 3) followed first-order kinetics with a rate constant of (2.8 ( 0.1) × 105 s-1. Therefore, by replacing H2O with D2O, as the reaction medium, a kinetic isotope effect ktaut(H2O)/ktaut(D2O) of 6.1 was calculated. This value is similar to the above-reported value of 6.7 for guanosine. Reaction of eaq- with N2-Ethylguanosine (4) and N2,N2Diethylguanosine (5). Rate constants of 2.8 × 109 and 2.7 × 109 M-1 s-1 were determined for the reactions of eaq- with 4 and 5, respectively, at natural pH in H2O. The optical absorption spectrum recorded at 4 µs after the pulse from the reaction of 5 with eaq- at natural pH in H2O solutions is presented in Figure 4. The shape of this spectrum resembles the spectra assigned to the second tautomer for both guanosine and 1-methylguanosine. The time profile of the formation of this transient at 305 nm (upper inset) or the disappearance of eaq- at 720 nm (lower inset) leads to the same pseudo-first-order rate constant (kobs ) 1.3 × 106 s-1), indicating that if any intermediate exists its half-life will be below a few nanoseconds (Figure 4). This is a strong indication that a tautomerization process does not take place. An analogous experiment was performed in D2O solutions. A rate constant 2-fold lower than the identical reaction in H2O was determined independently from the decay of the hydrated electron or the growth of the transient, confirming the absence of a tautomerization process. Since dialkylation of the NH2 group seems to influence the occurrence of a tautomerization, it was imperative to determine whether monoalkylation of this moiety could also produce the same effect. For this purpose, deaerated aqueous solutions containing N2-ethylguanosine 4 (0.5 mM) and t-BuOH (0.25 M) at natural pH were pulse irradiated and the resulting optical absorption spectrum obtained 4 µs after the pulse is reported in Figure 5. The time profile of the formation of this transient at 300 nm (upper inset) or the disappearance of eaq- at 720 nm (lower inset) leads to the same pseudo-first-order rate constant (kobs ) 1.4 × 106 s-1), denoting a single process that does not support the occurrence of tautomerization. Moreover, when H2O was replaced by D2O, a solvent kinetic isotope effect of 2.6 was obtained. As in the case of 5, none of the experiments performed on 4 seem to support a transformation process for the reaction of this substrate with eaq-. DFT Calculations and Proposed Tautomerization Process. Time-dependent DFT calculations (TD-B3LYP/6-311G**// B1B95/6-31+G**) were carried out in order to compute vertical optical transitions of possible intermediates and to compare with the experimental UV-visible spectra. This theoretical method was shown to predict optical transitions in nucleosides with reliability.8,9,23-25 TD-DFT calculations were carried out on

Reaction of Hydrated Electrons with Guanine Derivatives

J. Phys. Chem. B, Vol. 113, No. 7, 2009 2173

TABLE 1: Vertical Optical Transitions of the Hydrogen Atom Adduct of Guanosine (2) Computed at the TD-B3LYP/ 6-311G**//B1B95/6-31+G** Level

a

The dot in radical 2 is located on the atom with highest calculated spin density. b Oscillator strengths (f) for transition >0.010.

SCHEME 2: Vertical Optical Transitions (Wavelength λ and Oscillator Strength f) of the Hydrogen Atom Adduct of Guanosine Derivatives Computed at the TD-B3LYP/ 6-311G**//B1B95/6-31+G** Levela

a For the kind of excitation, see Table 1; oscillator strengths (f) for transition >0.010.

Figure 4. Absorption spectrum obtained after pulse radiolysis at an unbuffered pH of 0.5 mM 5, 0.25 M t-BuOH, N2-purged, optical path 2 cm, dose ) 10.8 Gy, 4 µs after the pulse. Inset: Time dependence of absorption at 305 and 720 nm from the pulse radiolysis of 0.5 mM 5, 0.25 M t-BuOH at natural pH, N2-purged, optical path ) 2.0 cm, dose ) 11 Gy. The solid lines represent the first-order kinetic fit to the data.

guanosine derivatives. In particular, we considered hydrogen atom adducts at C8 of various substituted guanosines as well as the possible tautomeric forms of the protonated guanosine radical anion. Table 1 shows the absorption spectra computed for the hydrogen atom adduct of guanosine (2). For this radical, the calculations predict a main transition at 286 nm with three weak transitions at 319, 377, and 460 nm in good accord with the experimental spectrum recorded 20 µs after the pulse displayed in Figure 1 that presents a main peak at about 300 nm with a shoulder and a weak band at about 460 nm. The two lower wavelength transitions are due to excitations from the higher occupied MOs to the SOMO, whereas the two higher wavelength transitions are due to excitations from the SOMO localized to the pπ orbital of N7 to the antibonding π* orbitals. Scheme 2 shows the analogous hydrogen atom adducts of substituted guanosines. All of them show the same kind of optical transitions reported in Table 1. Methyl substitution at the N1 position has a negligible effect on the transition location of radical 6 in agreement with experimental observations. The wavelength λ and oscillator strength f computed for the ethylsubstituted radicals 7 and 8 are also substantially similar. These

Figure 5. Absorption spectrum obtained after pulse radiolysis at an unbuffered pH of 0.5 mM 4, 0.25 M t-BuOH, N2-purged, optical path 2 cm, dose ) 6.9 Gy, 4 µs after the pulse. Inset: Time dependence of absorption at 300 and 720 nm from the pulse radiolysis of 0.5 mM 4, 0.25 M t-BuOH at natural pH, N2-purged, optical path ) 2.0 cm, dose ) 8 Gy. The solid lines represent the first-order kinetic fit to the data.

2174 J. Phys. Chem. B, Vol. 113, No. 7, 2009

D’Angelantonio et al.

TABLE 2: Vertical Optical Transitions of Radicals 9, 10, and 11 Computed at the TD-B3LYP/6-311G**//B1B95/6-31+G** Level

a

The dot in radicals is located on the atom with highest calculated spin density. b Oscillator strengths (f) for transition >0.010.

TABLE 3: Vertical Optical Transitions of Various Iminic Form Tautomers of the 8H-Adduct of Guanosine Computed at the TD-B3LYP/6-311G**//B1B95/6-31+G** Level

a

The dot in radicals is located on the atom with highest calculated spin density. b Oscillator strengths (f) for transition >0.010.

findings confirm once again that the TD-DFT method is reliable for predicting optical transitions in nucleoside radicals. The species responsible for the spectrum recorded 2 µs after the pulse could be a protonated form of the guanine radical anion at O6, N3, or N7, as previously suggested by Candeias

et al.6 Table 2 shows the calculations for these three radicals. The experimental spectrum presents two bands at 300 and 340 nm of similar intensity with a weak band at about 440 nm (cf. black circles in Figure 1), whereas Table 2 shows that the TDDFT method computes a main transition at about 300 nm for

Reaction of Hydrated Electrons with Guanine Derivatives

J. Phys. Chem. B, Vol. 113, No. 7, 2009 2175

SCHEME 3: Proposed Mechanism for the Reaction of eaq- with Guanosine and 1-Methylguanosine

all three protonated radicals with weak transitions at longer wavelengths. Therefore, TD-DFT calculations do not support this assignment. Next, we considered tautomers of the hydrogen atom adduct 2 (cf. Table 1). In particular, TD-DFT calculations were carried out on the three possible iminic tautomers 12, 13, and 14 shown in Table 3. The optical transitions computed for the O6Htautomer 12 show that agreement with experiment is poor, since the two main transitions are computed above 420 nm. A similar conclusion applies to the N3H-tautomer 13, since the stronger transition is computed at 468 nm and only one transition is computed in the range 280-400 nm. DFT calculations for the N7H-tautomer 14 are in better agreement with experiment. Indeed, a main transition is predicted at 324 nm and a weak band at 304 nm. This is in good accord with the trend observed

Figure 6. SOMO of the aminic (A) and iminic (B) forms of the C8protonated one-electron-reduced guanosine computed at the B3LYP/ 6-311G** level.

in the 8D-adduct of 1-methylguanosine where the first band at 300 nm is much smaller than the second one (cf. black circles in Figure 3). Furthermore, these findings suggest that in Figure 1 the spectrum recorded 2 µs after the pulse should be due to a superposition of spectra due to both the 8D-adduct of guanosine and its iminic N7D-tautomer (cf. adducts 2 and 14, respectively). The TD-DFT calculations suggest that the first observed transient species is the iminic form 14 that undergoes tautomerization to the aminic form 2. Figure 6 shows the calculated distribution of the unpaired electron in the SOMO of these two radicals. In the aminic form (A), the unpaired electron is mainly localized at N7, although substantial spin density is also at the N9, C4, C5, and O6 atoms. In the iminic form (B), the unpaired electron is mainly localized at the C5 and N7, with significant spin density at the N3, N10, and O6 atoms. Conceptually, these two tautomers are related to our previous work on the protonation of the radical anion of 8-haloguanosines.8,9 It was suggested that the radical anion was rapidly protonated at C8 to form a π-complex, with the Br atom situated above the molecular plane, that it facilitates ejection of Br-. A transition state was localized in which the initial deprotonation at N10 by means of H2O could occur during elimination of Brfrom the π-complex as HBr. We suggest that an analogous protonation/deprotonation step by the H2O solvent plays an important role in the present work. We propose the mechanism illustrated in Scheme 3 for the reaction of eaq- with guanosine or 1-methyl guanosine. The electron adduct 15 is not observed because it is rapidly protonated (k > 1 × 107 s-1) at the N7 position to give 16. Indeed, the N7-protonated guanosine radical anion is computed to be more stable than the O6- and N3-protonated forms by 5.1 and 11.0 kcal mol-1, respectively, at the B1B95/6-31+G** level. We suggest that radical 16 is SCHEME 4: Proposed Mechanism for the Regeneration of Guanosine

Figure 7. Irradiation of guanosine (1.0 mM) in the presence of 0.25 M t-BuOH at natural pH and at a dose rate of ∼10 Gy min-1.

2176 J. Phys. Chem. B, Vol. 113, No. 7, 2009 not the first observable species; rather, a rapid protonation at C8 invoking a transition state such as 17, as well as deprotonation by the H2O solvent, gives the first observable transient species which is 18. The subsequent tautomerization 18 f 19 occurs with ktaut ≈ 1.5 × 106 s-1 through a complex transition state involving a water molecule. The large kinetic isotope effect is in favor of the tautomerization. The Fate of the Conjugated Aminyl Radical 2. Deaerated aqueous solutions containing guanosine (1.0 mM) and t-BuOH (0.25 M) at natural pH were irradiated under stationary state conditions with a dose rate of ca. 10 Gy min-1 followed by RP-HPLC analysis. Surprisingly, the only detectable product in both cases was the initial nucleoside, with no measurable loss of guanosine even after 2 kGy, as illustrated in Figure 7. In the presence of t-BuOH, only eaq- (G ) 0.27) and H• (G ) 0.062) react with the substrate; thus, under these conditions, a dose of 2 kGy generates ca. 0.66 mM radicals in the irradiated solution. It has been established through pulse radiolysis experiments that both eaq- and H• react with guanosine, so by assuming G ) 0.33 for the consumption of guanosine, a conversion of ∼66% would be expected after 2 kGy, in contrast with the experimental results. Similar results were obtained by replacing t-BuOH (0.25 M) with i-PrOH (0.13 M) for scavenging both HO• and H• species. Our mechanistic proposal to explain this unexpected finding involves a cross disproportionation reaction between the nucleosidic neutral radical and the •CH2C(CH3)2OH or •C(OH)(CH3)2 radical, which is formed after the reaction between HO• or H• and t-BuOH or i-PrOH, respectively (Scheme 4).29 As a result, both the starting nucleoside and t-BuOH or i-PrOH are regenerated. This mechanism accounts for the lack of novel products in the irradiated solution, even after irradiation with high doses, and the zero consumption of the starting nucleoside. Conclusions Radical anions of DNA bases have held the interest of chemists for years, and a large body of results exist for nucleosides as simple models.1 Previously, an UV-visible spectrum obtained by one-electron reduction of guanosine was correctly assigned to the H-adduct at C8 position 19 (Scheme 4).6 Our work here reveals the existence of its tautomer 18 that is initially formed and rapidly converted (via a water-assisted process) to 19 with a rate constant of ktaut ≈ 1.5 × 106 s-1. Acknowledgment. This work was supported in part by the European Community’s Marie Curie Research Training Network under contract MRTN-CT-2003-505086 [CLUSTOXDNA]. The support and sponsorship by COST Action CM0603 on “Free Radicals in Chemical Biology (CHEMBIORADICAL)” are kindly acknowledged. We thank M. Lavalle, A. Monti, and A. Martelli for assistance with the pulse radiolysis experiments.

D’Angelantonio et al. References and Notes (1) von Sonntag, C. Free-Radical-Induced DNA Damage and Its Repair; Springer-Verlag: Berlin, 2006. (2) Chatgilialoglu, C.; Guerra, M.; Kaloudis, P.; Houe´e-Le´vin, C.; Marignier, J.-L.; Swaminathan, V. N.; Carell, T. Chem.sEur. J. 2007, 13, 8979. (3) Manetto, A.; Breeger, S.; Chatgilialoglu, C.; Carell, T. Angew. Chem., Int. Ed. 2006, 45, 318. (4) For a review, see: Wagenknecht, H.-A. Nat. Prod. Rep. 2006, 23, 973. (5) Steenken, S. Chem. ReV. 1989, 89, 503. (6) Candeias, L. P.; Wolf, P.; O’Neill, P.; Steenken, S. J. Phys. Chem. 1992, 96, 10302. (7) Rakvin, B.; Herak, J. N.; Voit, K.; Hu¨ttermann, J. Radiat. EnViron. Biophys. 1987, 26, 1. (8) Chatgilialoglu, C.; Caminal, C.; Guerra, M.; Mulazzani, Q. G. Angew. Chem., Int. Ed. 2005, 44, 6030. (9) Chatgilialoglu, C.; Caminal, C.; Altieri, A.; Mulazzani, Q. G.; Vougioukalakis, G. C.; Gimisis, T.; Guerra, M. J. Am. Chem. Soc. 2006, 128, 13796. (10) Yamauchi, K.; Tanabe, T.; Kinoshita, M. J. Org. Chem. 1979, 44, 638. (11) Sako, M.; Kawada, H.; Hirota, K. J. Org. Chem. 1999, 64, 5719. (12) Hutton, A.; Roffi, G.; Martelli, A. Quad. Area Ric. Emilia Romagna 1974, 5, 67. (13) Wrona, M.; Wardman, P. Free Radical Biol. Med. 2006, 41, 657. (14) Mulazzani, Q. G.; D’Angelantonio, M.; Venturi, M.; Hoffman, M. Z.; Rodgers, M. A. J. J. Phys. Chem. 1986, 90, 5347. (15) Buxton, G. V.; Stuart, C. R. J. Chem. Soc., Faraday Trans. 1995, 91, 279. (16) Spinks, J. W. T.; Woods, R. J. An Introduction to Radiation Chemistry, 3rd ed.; Wiley: New York, 1990; p 100. (17) Becke, A. D. Phys. ReV. A 1988, 38, 3098. (18) Becke, A. D. J. Chem. Phys. 1996, 104, 1040. (19) Zhao, Y.; Pu, J.; Lynch, B. J.; Truhlar, D. G. Phys. Chem. Chem. Phys. 2004, 6, 673. (20) Frisch, M. J.; et al. Gaussian 03, revision D.02; Gaussian, Inc.: Wallingford CT, 2004. (21) Zhao, Y.; Lynch, B. J.; Truhlar, D. G. J. Phys. Chem. A 2004, 108, 2715. (22) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2004, 108, 6908. (23) Chatgilialoglu, C.; Ferreri, C.; Bazzanini, R.; Guerra, M.; Choi, S.-Y.; Emanuel, C. J.; Horner, J. H.; Newcomb, M. J. Am. Chem. Soc. 2000, 122, 9525. (24) Chatgilialoglu, C.; Guerra, M.; Mulazzani, Q. G. J. Am. Chem. Soc. 2003, 125, 3839. (25) Kaloudis, P.; D’Angelantonio, M.; Guerra, M.; Gimisis, T.; Mulazzani, Q. G.; Chatgilialoglu, C. J. Phys. Chem. B 2008, 112, 5209. (26) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data 1988, 17, 513, and references cited therein. (27) Ross, A. B.; Mallard, W. G.; Helman, W. P.; Buxton, G. V.; Huie, R. E.; Neta, P. NDRL-NIST Solution Kinetic Database - Ver. 3, Notre Dame Radition Laboratory, Notre Dame, IN, and NIST Standard Reference Data, Gaithersburg, MD (1998). (28) For eaq- at 720 nm, ε ) 1.9 × 104 M-1 cm-1; Hug, G. L. Nat. Stand. Ref. Data Ser., Nat. Bur. Stand. 1981, No. 69. (29) Rate constants of 3.8 × 108 and 6.0 × 108 M-1 s-1 are reported for the radical-radical reactions of (CH3)2C(OH)CH2• and (CH3)2C•(OH), respectively, in aqueous solutions.27 On the basis of our proposed mechanism, the rate constant for the reaction in Scheme 4 should be a few times faster, i.e., g3× 109 M-1 s-1.

JP809386C