J. Phys. Chem. B 2010, 114, 9617–9621
9617
The Fate of H Atom Adducts to 3′-Uridine Monophosphate Ran Wang,† Ru bo Zhang,*,† and Leif A. Eriksson*,‡ Institute for Chemical Physics, School of Science, Beijing Institute of Technology, Beijing 100081, China, and School of Chemistry, National UniVersity of Ireland, Galway, Ireland, and School of Science and Technology, ¨ rebro UniVersity, O ¨ rebro, Sweden O ReceiVed: January 06, 2010; ReVised Manuscript ReceiVed: May 06, 2010
The stabilities of the adducts deriving from H free radical addition to the O2, O4, and C5 positions of 3′uridine monophosphate (3′UMP) are studied by the hybrid density functional B3LYP approach. Upon H atom addition at the O2 position, a concerted low-barrier proton-transfer process will initially occur, followed by the potential ruptures of the N-glycosidic or β-phosphate bonds. The rupture barriers are strongly influenced by the rotational configuration of the phosphate group at the 3′ terminal, and are influenced by bulk solvation effects. The O4-H adduct has the highest thermal stability, as the localization of the unpaired electron does not enable cleavage of either the C1′-N1 or the C3′-O(P) bonds. For the most stable adduct, with H atom added to the C5 position, the rate-controlled step is the H2′a abstraction by the C6 radical site, after which the subsequent strand rupture reactions proceed with low barriers. The main unpaired electron densities are presented for the transient species. Combined with previous results, it is concluded that the H atom adducts are more facile to drive the strand scission rather than N-glycosidic bond ruptures within the nucleic acid bases. 1. Introduction It is a well-established fact that ionization radiation can interact with water molecules to produce OH free radicals, protons, and electrons. The electrons can become captured by nucleotides to form transient anion radicals, followed by protonation giving net hydrogen atom adducts.1-10 In addition, H free radicals can add directly to nucleobases at close to diffusion-controlled rates to form similar adducts. These are crucial transient intermediates in radiation damage of nucleic acids. Previous studies have shed light on the identity of these radical adducts, but their subsequent degradation reactions are relatively little known.11 Thus, further studies are needed to explore the process of free-radical induced damage of nucleic acids, and the relevant kinetic and thermodynamic properties. The protonation of nucleic acid bases and their electron adducts have been reported in earlier work.12-21 Relevant computations were performed to probe the potential mechanisms of unimolecular decomposition of the adducts. It was concluded that decomposition was activated by the addition of the unpaired electron. The N1-C2, N1-H1, C5-H5, and C4-C5 bonds could be broken in the case of the H• + uracil base adduct, with the most facile scission being the removal of the added H atom from the C5 position of uracil.16 The barrier was estimated to be 129 kJ/mol at the B3-PMP2/6-311+G(3df,2p) level, and 140 kJ/mol at the B3LYP/6-31+G(d,p) level of theory. The barriers for C4-C5 and N1-C2 bond breakage were larger by 8-16 kJ/mol, at the B3-PMP2/6-311+G(3df,2p) level. All reactions were quite endothermic. Adducts formed by H atom addition to the O4 and O2 atoms of uracil were also probed experimentally and theoretically,17 with removal of the added H atom again having the lowest barrier. At the B3-PMP2/6311+G(2d,p) level, these were estimated to be 111 kJ/mol for * To whom correspondence should be addressed. E-mail: zhangrubo@ bit.edu.cn and
[email protected]. † Beijing Institute of Technology. ‡ ¨ rebro University. National University of Ireland and O
the O4-H bond and 60 kJ/mol for the O2-H bond, respectively, which were considerably lower than those of the N1-C2 and C2-N3 bond cleavages. The computed barriers at the B3-PMP2/ 6-311+G(2d,p) level could be well reproduced by B3LYP/631+G(d,p) calculations.16,17 The results clearly show that the H• + base adducts are relatively stable. On the basis of this, excited electronic states resulting from vertical electron capture by the uracil + H+ system were thought to be the contributing factor to the bond rupture processes.16,17 Our previous studies have revealed that the reaction barrier is quite high for H2′a atom abstraction by C6 of the thymine base, the main reactive hotspot induced through H atom addition to the C5 position of 3′dTMP.22 The free radical adducts were essentially unable to generate DNA strand breaks or base scission.22,23 The main reason is the high stability of the initial adducts. However, with the capture of an extra electron by the H atom adduct, the carbanion became facile to abstract the H2′a atom, resulting in DNA strand breakage.24 Free radical adducts to RNA, on the other hand, are prone to undergo breakage of the ribose C3′-O(P) bond.25 The distinct difference is the acidity of the 2′OH proton of ribose in RNA, that can be transferred to one of the oxygen atoms of the phosphate group, concomitant with the transfer of electron spin from C2′ to C3′. From these previous studies, one question stands out: what is the possible fate of the system when an H atom is added to the uracil base of 3′-uridine monophosphate (3′UMP)? In the present paper, the reactions of adducts between an H atom and 3′UMP at the C5, O2, and O4 positions (cf. Scheme 1) are individually addressed in detail to provide further insight into the possible damage processes upon H atom attack to RNA. 2. Methodology All geometries were optimized at the hybrid Hartree-Fockdensity functional theory B3LYP level,26,27 in conjunction with the 6-31+G(d,p) basis set. The method is verified as reliable for these types of systems, based on comparisons with studies
10.1021/jp100116w 2010 American Chemical Society Published on Web 06/30/2010
9618
J. Phys. Chem. B, Vol. 114, No. 29, 2010
Wang et al.
SCHEME 1
at the ROMP2, QCISD, and B3-PMP2 levels.16,17 Frequency calculations were performed at the same level of theory, to confirm the correct nature of the stationary points and to extract
the zero-point vibrational energies (ZPE). For the N-glycosidic bond rupture processes, the potential-energy surfaces (PESs) were scanned from the initial H radical adducts by varying the
Figure 1. Selected geometric parameters of the stationary structures along the glycosidic and C3′-O(P) bond rupture reactions for O2-H adducts of 3′UMP (B3LYP/6-31+G(d,p) level).
Fate of H Atom Adducts to 3′-Uridine Monophosphate
J. Phys. Chem. B, Vol. 114, No. 29, 2010 9619
TABLE 1: Barriers and Reaction Energies (in kcal/mol) for the O2-H System O2-1 f O2-3 ∆E
qa
∆Ea a
c
3.7 (4.5) 14.3,d 13.2e -6.6 (-4.0)
O2-3 f Prod 1b 17.2 (17.3) 17.5 (-5.6)
O2-3′ f Prod 1b 23.6 (17.0) 5.2 (-10.2)
O2-3 f Prod 2b 18.0 (18.2) 10.8 (-19.7)
O2-3′ f Prod 2b 8.2 (8.6) -1.6 (-24.4)
b
Relative electronic energies (ZPE corrected). Prod 1 and Prod 2 are the sum of the dissociated fragments, produced from N-glycosidic and C3′-O(P) dissociation, respectively. c Values in bulk solution given in parentheses. d In the gas phase at the B3LYP/6-31+G(d,p) level, from ref 17. e In the gas phase at the B3-PMP2/6-311+G(2d,p) level, from ref 17.
N1-C1′ distance using a step length of 0.1 Å, and optimizing the remaining coordinates. The same procedure was used to determine the PES of β-phosphate bond dissociation of the radical adducts, by scanning the C3′-O3′ distance. Bulk solvation effects were included in the series of single-point energy calculations on the optimized structures, through the integral equation formalism of the polarized continuum model (IEF-PCM).28 The dielectric constant ε ) 78.4 was employed, to model aqueous solution. All calculations were performed with the Gaussian 03 package.29 Atomic labeling used in the text and tables throughout refers to Scheme 1. 3′UMP in its C3′-endo form was employed throughout. The RNA strand was truncated by hydrogen atoms at the C5′OH and the C3′-OP(O)2- OH ends. 3. Results and Discussions 3.1. O2-H Adduct. For the adduct with the H atom attached to O2 (O2-1 in Figure 1), the added H atom (H2) is oriented such that it forms a hydrogen bond to O2′ of the sugar moiety. The C2-O2 bond length is 1.222 Å in 3′UMP, which increases to 1.363 Å when the H2 atom is added. The unpaired electron is mainly located on C2 (0.96 e), which leads to the deviation of C2 from the base plane. A network of hydrogen bonds is formed at H2 · · · O2′-H2b′ · · · O(P) in O2-1, as seen in Figure 1, with the distances 1.682 and 1.517 Å, respectively, characteristic of low barrier hydrogen bonds (LBHB). Concerted proton transfer (PT) reactions take place in the above hydrogen bonding network, into a O2 · · · H2-O2′ · · · H2b′-O(P) hydrogen bonded structure in the PT product (O2-3 in Figure 1). The PT barrier is estimated to be 3.7 kcal/mol at the B3LYP/631+G(d,p) level, displayed in Table 1. The barrier is far lower than that (60 kJ/mol) of H atom dissociation from the H + uracil base adduct.17 The reaction is exothermic by 6.6 kcal/mol including ZPE corrections. For the PT product, the C2-O2 bond length is reduced to 1.265 Å, comparable to the 1.222 Å seen in 3′UMP. The change in bond length is attributed to the unpaired electron being transferred to the C5 (0.21 e) and C6 (0.40 e) atoms from C2 of the initial adduct. Through a rotation in the C2′-C3′-O-P dihedral angle of O2-3 with a step length of 0.5° (see Figure 2), the rotamer O2-3* is obtained, which is also shown in Figure 1 for comparison. The structural difference between O2-3 and O2-3* is the presence (O2-3) or absence (O2-3*) of a O2′ · · · HO(P) hydrogen bond, which can introduce effects on the transition structure toward β-phosphate bond rupture. The difference in free energies between the rotamers is 10.8 kcal/ mol and the relative electronic energy difference is 12.3 kcal/ mol in the gas phase, while it is 4.6 kcal/mol when including bulk solvation, in all cases favoring the hydrogen bonded O2-3 conformer. The transition structures during β-phosphate bond rupture show that the C3′ · · · O(P) distance is 1.859 Å in O2-4, but only 1.692 Å in O2-4*, cf. Figure 1. In addition, the leaving O(P) atom is stabilized by H2b′ in O2-4*. This is, however,
Figure 2. Energy curves for the C3′-O(P) bond rotation, obtained at the B3LYP/6-31+G(d,p) level (without ZPE corrections).
not observed in O2-4. This difference can explain why O2-4 is slightly more stable (by 2.5 kcal/mol) than O2-4* in the gas phase. The barrier toward scission of the C3′-O(P) bond is 18.0 kcal/mol for O2-3 and 8.2 kcal/mol for O2-3* in the gas phase, and 18.2 kcal/mol for O2-3 and 8.6 kcal/mol for O2-3* in bulk solvent, all of which are obtained at the B3LYP/ 6-31+G(d,p) level. The 3′-phosphate group of the initial adduct is in a configuration similar to that of O2-3. This is converted to O2-3* through the rotation around the C3′-O(P) bond with the barriers of 13.2 kcal/mol in the gas phase and 5.4 kcal/mol including bulk solvation. Since the barrier of the C3′-O(P) bond rupture of O2-3 is comparable to that of rotation of the phosphate group in the gas phase, these two changes should occur competitively. However, the barrier for the conversion from O2-3 to O2-3* is much lower than that of β-phosphate bond rupture in bulk solvation. Thus, the configuration conversion is largely favored, and the scission of the C3′-O(P) bond will subsequently occur from O2-3*. Thus, the local solvation of the phosphate and its relative orientation to the ribose will influence the H atominduced β-phosphate bond rupture. The unpaired electron can also drive the N-glycosidic bond rupture. For the transition structure O2-5 deriving from O2-3, the distance between the C1′ and N1 atoms is 1.855 Å. The leaving N atom is stabilized by HO(-C5′). In addition, the hydrogen bond O2′-H2′ · · · O2 is retained with a length of 1.837 Å ; this is not found in the transition structure O2-5* from O2-3*. In O2-5*, the distance between C1′ and N1 is 1.891 Å, and HO(-C5′) now interacts with the C5dC6 bond. The barrier of N-glycosidic bond rupture for the O2-3* f O2-5* reaction is estimated to be 23.6 kcal/mol, which is 6.4 kcal/ mol higher than the one for the O2-3 f O2-5 reaction. The latter is similar in magnitude to the barrier of the C3′-O(P) bond scission for the O2-3 f O2-4 reaction. The bulk solvation effects on the reactions are calculated by using the dielectric constant e ) 78.4 to model aqueous solution.
9620
J. Phys. Chem. B, Vol. 114, No. 29, 2010
Wang et al. TABLE 2: Barriers and Reaction Energies (in kcal/mol) for the C5-H System ∆Eq a ∆Ea
Figure 3. Selected geometric parameters of the stationary structures along the glycosidic and C3′-O(P) bond rupture reactions for O4-H adducts of 3′UMP (B3LYP/6-31+G(d,p) level).
As seen in Table 1, the reaction energies are changed significantly, and the scission reactions for both N-glycosidic and β-phosphate bonds are exothermic due to the influence of bulk solvation. 3.2. O4-H Adduct. The adduct deriving from H atom addition to O4 is also studied, and the optimized geometries are shown in Figure 3. This is slightly more stable than the O2-H adduct, with a Gibbs’ free energy difference of ∼3.7 kcal/mol. The unpaired electron is mainly localized on C6, C4, and C5, with the values 0.63, 0.44, and -0.18 e, respectively. The H atom addition allows for scission of the N-glycosidic bond, for which 24.8 and 22.7 kcal/mol barriers need to be overcome in the gas phase and in bulk solvation, respectively. The relative reaction energy is 16.8 kcal/mol in the gas phase, and 6.0 kcal/mol in bulk solvent. Bulk solvation effects hence favor the reaction to take place. The unpaired electron is entirely localized on the resulting ribose + phosphate fragment. C3′-O(P) bond breakage could not be observed from the H atom addition. On the basis of the above results, the adduct in which the H atom is added to O4 appears thermally and dynamically relatively stable and will not lead to strand breakage.
C5-1 f C5-3
C5-3 f Prod 3b
C5-3 f Prod 4b
16.2 (17.8)c 33.5,d 31.1e -6.8 (-6.7)
10.9 (19.0)
4.5 (5.8)
35.8 (26.7)
14.9 (-7.1)
a Relative electronic energies (ZPE corrected). b Prod 3 and Prod 4 are the sum of the dissociated fragments, produced from N-glycosidic and C3′-O(P) dissociation, respectively. c Values in bulk solution given in parentheses. d In the gas phase at the B3LYP/ 6-31+G(d,p) level, from ref 16. e In the gas phase at the B3-PMP2/ 6-311+G(2d,p) level, from ref 16.
3.3. C5-H Adduct. H atom addition to C5 of 3′UMP is energetically favored over the above-described adduct formations. The ZPE-corrected ordering of the adducts is C5-1 (-18.7 kcal/mol) < O4-1 (-3.8 kcal/mol) < O2-1 (0.0 kcal/ mol). The same order is also observed in bulk solvent. All structures relating to H atom addition at position C5 of uracil, including the initial adducts and the subsequent products, are shown in Figure 4. The radical center is localized at C6. The (O2′)H2′b hydrogen forms a weak hydrogen bond to the negatively charged oxygen atom of the phosphate group, with a distance of 1.732 Å. This again characterizes low-barrier hydrogen bonding (LBHB) and its significance for the types of reactions studied herein has been emphasized in earlier work.25 During the H2′a transfer process, the C6 · · · C2′ distance is shortened from 3.054 Å in C5-1 to 2.552 Å in the transition structure C5-2. In the TS, the distances from H2′a to C6 and C2′ are 1.481 and 1.309 Å, respectively. The O2′-H2′b bond undergoes a partial dissociation during the H2′a transfer process. The reaction energy profile is presented in Table 2. The activation barrier for H2′a atom transfer to C6 is in the gas phase estimated to be 16.2 kcal/mol, and 17.8 kcal/mol in bulk solvent. This is 5-7 kcal/mol higher than that of the corresponding OH radical-induced H atom transfer (11.2 kcal/mol).25 The reaction is exothermic by 6.8 kcal/mol. Note that the barrier of H
Figure 4. Selected geometric parameters of the stationary structures along the glycosidic and C3′-O(P) bond ruptures reactions for C5-H adducts of 3′UMP (B3LYP/6-31+G(d,p) level).
Fate of H Atom Adducts to 3′-Uridine Monophosphate abstraction is far lower than that of H atom dissociation from the C5-H adduct (140 kJ/mol).16 Thus, the abstraction reaction described herein is more favorable. The subsequent N-glycosidic bond rupture is also calculated. The ZPE-corrected barrier is 10.9 kcal/mol in the gas phase, which is comparable to the 12.4 kcal/mol seen for OH radical induced N-glycosidic bond rupture,25 and the reaction energy is estimated to be 35.8 kcal/ mol. The barrier of the β-phosphate bond rupture is 4.5 kcal/ mol and the reaction energy is 14.9 kcal/mol in the gas phase while it is exothermic by 7.1 kcal/mol under the influence of the solvent. Thus, bulk solvent could significantly influence the above reactions. 4. Conclusions The adducts, deriving from H atom addition to either of the O2, O4, and C5 positions of 3′UMP have been studied at the B3LYP/6-31+G(d,p) level to probe the possible fates of the H atom adducts. The stability ordering of the initial adducts is O2-H < O4-H < C5-H. For the different systems, the following observations are made: (1) For the O2-H adduct, the main low-barrier proton transfer from O2 to O2′ concomitantly with proton transfer from O2′ to the phosphate group leads to the unpaired electron being transferred from C2 of the initial adduct O2-1, to C5 and C6. The barrier to subsequent N-glycosidic bond rupture is 17.2 or 23.6 kcal/mol in the gas phase for the two rotamers O2-3 and O2-3*, respectively. The β-phosphate bond scission requires 18.0 or 8.2 kcal/mol to be overcome for the same two systems. The difference in barrier heights is due to the presence or absence of stabilizing hydrogen bonds. Bulk solvation does not influence the barriers to any significant extent, but makes the bond rupture reactions exothermic. (2) For the O4-H adduct, only the N-glycosidic bond rupture is observed, and its barrier is over 20 kcal/mol. Thus, the O4-H adduct is thermally the most stable of the systems studied. Bulk solvation effects favor the reaction. (3) For the C5-H adduct, the abstraction of H2′a by the C6 radical center needs to overcome a barrier of 16.2 kcal/mol. The unpaired electron is transferred to C2′, which drives the N-glycosidic and β-phosphate bond ruptures, the barriers of which are 10.9 and 4.4 kcal/mol, respectively. It is furthermore concluded that inclusion of bulk solvent could significantly influence the above reactions. Combined with the previous results outlined herein, the conclusions can be drawn that the H atom addition as such does not induce bond ruptures within the nucleic acid bases. Instead it is the transfer of the unpaired electron that enables the N-glycosidic and/or β-phosphate bond breakage reactions. Acknowledgment. The Swedish science research council ¨ rebro (VR), the faculty of Science and Technology at O University, and the National University of Ireland are gratefully acknowledged for financial support. This work is supported by the National Nature Science Foundation of China (Grant Nos.
J. Phys. Chem. B, Vol. 114, No. 29, 2010 9621 20643007 and 20703004) and Excellent Young Scholars Research Fund of the Beijing Institute of Technology. References and Notes (1) von Sonntag, C. In Physical and Chemical Mechanism in Molecular Radiation Biology; Glass, W. A., Varma, M. N., Eds.; Plenum Press: New York, 1991; pp 287-321. (2) Symons, M. C. R. J. Chem. Soc., Faraday Trans. 1 1987, 83, 1. (3) Steenken, S. Chem. ReV. 1989, 89, 503. (4) Sagstuen, E.; Hole, E. O.; Nelson, W. H.; Close, D. M. J. Phys. Chem. 1989, 93, 5974. (5) Ohta, N.; Tanaka, N.; Ito, S. J. Chem. Soc., Perkin Trans. 2 1999, 2597. (6) Colson, A.-O.; Becker, D.; Eliezer, I.; Sevilla, M. D. J. Phys. Chem. A 1997, 101, 8935. (7) Lyngdoh, R. H. D.; Schaefer, H. F. Acc. Chem. Res. 2009, 42, 563. (8) Evangelista, F. A.; Schaefer, H. F. ChemPhysChem 2006, 7, 1471. (9) Zhang, J. D.; Schaefer, H. F. J. Chem. Theory Comput. 2007, 3, 115. (10) Dabkowska, I.; Rak, J.; Gutowski, M. Eur. Phys. J. D 2005, 35, 429. (11) Wallace, K. B. In Free Radical Toxicology; Taylor and Francis, Washington, D.C., 1997. (12) Marino, T.; Milano, V.; Russo, N.; Toscano, M. THEOCHEM 1994, 112, 185. (13) Russo, N.; Toscano, M.; Grand, A.; Jolibois, F. J. Comput. Chem. 1998, 19, 989. (14) Adamo, C.; Heitzmann, M.; Meilleur, F.; Rega, N.; Scalmani, G.; Grand, A.; Cadet, J.; Barone, V. J. Am. Chem. Soc. 2001, 123, 7113. (15) Grand, A; Morell, C.; Labet, V.; Cadet, J.; Eriksson, L. A. J. Phys. Chem A 2007, 111, 8968. (16) Syrstad, E. A.; Vivekananda, S.; Turecek, F. J. Phys. Chem. A 2001, 105, 8339. (17) Wolken, J. K.; Turecek, F. J. Phys. Chem. A 2001, 105, 8352. (18) Turecek, F.; Yao, C. X. J. Phys. Chem. A 2003, 107, 9221. (19) Chen, X. H.; Syrstad, E. A.; Nguyen, M. T.; Gerbaux, P.; Turecek, F. J. Phys. Chem. A 2005, 109, 8121. (20) Yao, C. X.; Cuadrado-Peinado, M. L.; Polasek, M.; Turecek, F. Angew. Chem., Int. Ed 2005, 44, 6708. (21) Yao, C. X.; Turecek, F.; Polce, M. J.; Wesdemiotis, C. Int. J. Mass Spectrom. 2007, 265, 106. (22) Zhang, R. B.; Eriksson, L. A. J. Phys. Chem. B 2006, 110, 23583. (23) Zhang, R. B.; Gao, F. X.; Eriksson, L. A. J. Chem. Theory Comput. 2007, 3, 803. (24) Zhang, R. B.; Eriksson, L. A. Chem.sEur. J. 2008, 14, 2850. (25) Zhang, R. B.; Eriksson, L. A. Chem.sEur. J. 2009, 15, 2394. (26) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (27) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (28) Tomasi, J.; Persico, M. Chem. ReV. 1994, 94, 2027. (29) 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.; 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. Gaussian 03, Revision B.04; Gaussian, Inc.: Pittsburgh, PA, 2003.
JP100116W
