The vertical ionization energies of the four nucleotides have been computed. Geometries have been chosen to mimic orientations as they appear in B-DNA. The negative charge on the phosphate was neutralized by protonation, and also by the inclusion of counterions. Calculations have been performed with electron propagator methods (P3), Møller-Plesset second-order perturbation theory, and density functional theory to determine the nature of the orbitals associated with the highest lying ionization energies. Calculations at the MP2/6-311G(d,p)//P3/6-311G(d,p) level of theory yield vertical ionization energies for 5'-dTMP 9.05 eV, for 5'-dCMP 8.40 eV, for 5'-dAMP 8.16 eV and for 5'-dGMP 7.96 eV. In all cases the highest occupied molecular orbital resides on the base moieties.

Single crystals of the phosphorylated amino acid L-O-serine phosphate were X-irradiated and studied at 10 K and at 77 K using EPR, ENDOR, and EIE techniques. Two radicals, R1(10 K) and R1(77 K), were detected and characterized as two different geometrical conformations of the protonated reduction product >CH-C(OH)(2). R1(10 K) is only observed after irradiation at 10 K, and upon heating to 40 K, R1(10 K) transforms rapidly and irreversibly into R1(77 K). The transition from R1(10 K) to R1(77 K) strongly increases the isotropic hyperfine coupling of the C-CH(beta) coupling (Delta = 32 MHz) and the major C-OH(beta) coupling (Delta = 47 MHz), in sharp contrast to the their much reduced anisotropic hyperfine couplings after the transition. An umbrella-like inversion of the carboxylic acid center, accompanied by minor geometrical adjustments, explains the changes of these observed isotropic and anisotropic couplings. DFT calculations were done on the reduced and protonated L-O-serine phosphate radical at the B3LYP/6-311+G(2df,p)//B3LYP/6-31+G(d) level of theory in order to support the experimental observations. Two different conformations of the anion radical, related by an inversion at the carboxylic center, could be found within the single molecule partial energy-optimization scheme. These two conformations reproduce the experimental hyperfine couplings from radicals R1(10 K) and R1(77 K). A third radical, radical R2, was observed experimentally at both 10 and 77 K and was shown to be due to the decarboxylated L-O-serine phosphate oxidation product, a conclusion fully supported from the DFT calculations. Upon thermal annealing from 77 to 295 K, radicals R1(77 K) and R2 disappeared and all three previously observed room-temperature radicals could be observed. No phosphate-centered radicals could be observed at any temperatures, indicating that the phosphate-ester bond break for one of the room-temperature radicals does not occur by dissociative electron capture at the phosphate group.

Single crystals of 5-methylcytosine hemihydrate and 5-methylcytosine hydrochloride were X-irradiated and studied at 10 K and at higher temperatures using X- and K-band EPR, ENDOR and EIE spectroscopy. In the hemihydrate crystals, four radicals were identified at 10 K, one of them being the recently reported N1-deprotonated one-electron oxidation product (Krivokapić et al., J. Phys. Chem. A 113, 9633-9640, 2009). The other radicals were the 3alphaH radical and the C6 and C5 H-addition radicals (the 5-yl and 6-yl radicals, respectively). After irradiation at 295 K, only the 3alphaH and the 5-yl radicals were observed. In the hydrochloride crystals, at least seven different radicals were present after irradiation at 10 K. These were the N1-deprotonated one-electron oxidation product, the 3alphaH radical, three different one-electron reduction products, and the 5- and 6-yl radicals. DFT calculations were used to assist in assigning the observed couplings. The 3alphaH and 5-yl radicals were dominant after thermal annealing to room temperature. In neither crystal system did the N1-deprotonated oxidation product transform into the 3alphaH radical upon warming. The radical yield was significantly greater after irradiation at 300 K compared to that after irradiation at 10 K followed by warming to 300 K and was also considerably greater in the hydrochloride crystals than in the hemihydrate crystals.

Single crystals of L-cysteic acid monohydrate were X-irradiated and studied at 295 K using EPR, ENDOR, and EIE techniques. Three spectroscopically different radicals were observed. These were a deamination radical reduction product (R1), and two oxidation products formed by hydrogen abstraction (radicals R2, R3). R2 and R3 were shown to exhibit the same chemical structure while exhibiting very different geometrical conformations. Cluster DFT calculations at the 6-31G(d,p) level of theory supported the experimental observations for radicals R1 and R2. It was not possible to simulate the R3 radical in any attempted cluster; hence, for this purpose a single molecule approach was used. The precursor radicals for R1, R2, and R3, identified in the low-temperature work on L-cysteic acid monohydrate by Box and Budzinski, were also investigated using DFT calculations. The experimentally determined EPR parameters for the low-temperature decarboxylated cation could only be reproduced correctly within the cluster when the carboxyl group remained in the proximity of the radical. Only one of the two observed low-temperature carboxyl anions (stable at 4 and 48 K) could be successfully simulated by the DFT calculations. Evidence is presented in support of the conclusions that the carboxyl reduction product already is protonated at 4 K and that the irreversible conversion between the two reduction products is brought forward by an umbrella-type inversion of the carboxyl group.

The primary oxidation product in X-irradiated single crystals of 5-methylcytosine hemihydrate and 5-methylcytosine hydrochloride has been studied at 10 K, using electron paramagnetic resonance (EPR), electron-nuclear double resonance (ENDOR), and ENDOR-induced EPR (EIE) spectroscopies. The radical is characterized by large couplings to the methyl protons and appears to be deprotonated at N1 in both crystal systems. In the hydrochloride crystal the methyl group is completely frozen at 10 K, whereas in the hemihydrate crystal it undergoes tunneling rotation. For the hemihydrate crystal, four ENDOR lines associated with transitions within the A and E rotational states were followed in three planes of rotation. Large ENDOR shifts as measured by saturation of the high- and low-field parts of the EPR spectrum indicate that the rotation is rather slow. Sidebands due to mixing of A and E rotational states are expected for slow rotation and were observed in both the EPR and the EIE spectra. The ENDOR shifts and the sideband frequencies indicate a tunneling splitting between 40 and 60 MHz. Estimates of the barrier to rotation in both crystalline systems were calculated using cluster and single-molecule density functional theory methods, and the results are consistent with those obtained by analysis of the experimental results.