Role of the guanine N1 imino proton in the migration and reaction of radical cations in DNA oligomers

Article abstract of DOI:10.1021/ja0573763

Oxidation of a guanine nucleobase to its radical cation in DNA oligomers causes an increase in the acidity of the N1 imino proton that may lead to its spontaneous transfer to N3 of the paired cytosine. This proton transfer is suspected of playing an important role in long-distance radical cation hopping in DNA and the decisive product-determining role in the reaction of the radical cation with H₂O or O₂. We prepared and investigated DNA oligomers in which certain deoxycytidines are replaced by 5-fluoro-2′-deoxycytidines (F⁵dC). The pK_a of F⁵C was determined to be 1.7 units below that of dC, which causes proton transfer from the guanine radical cation to be thermodynamically unfavorable. Photoinitiated one-electron oxidation of the DNA by UV irradiation of a covalently attached anthraquinone derivative introduces a radical cation that hops throughout the oligomer and is trapped selectively at GG steps. The introduction of F⁵dC does not affect the efficiency of charge hopping, but it significantly reduces the amount of reaction at the GG sites, as revealed by subsequent reaction with formamidopyrimidine glycosylase. These findings suggest that transfer of the guanine radical cation N1 proton to cytosine does not play a significant role in long-range charge transfer, but this process does influence the reactions with H₂O and/or O₂. Copyright

Full text of DOI:10.1021/ja0573763

Published on Web 03/15/2006
Role of the Guanine N1 Imino Proton in the Migration and Reaction of Radical
Cations in DNA Oligomers
Avik K. Ghosh and Gary B. Schuster*
School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332
Received October 28, 2005; E-mail: gary.schuster@cos.gatech.edu
Intensive investigation of the oxidation of duplex DNA has
shown that loss of an electron generates a radical cation (“hole”)
that migrates by a hopping mechanism until it is trapped irreversibly
in a chemical reaction with H2O or O2, which usually occurs at a
guanine or a Gn sequence.1-6 The relevance of these processes to
genetic mutation and to the potential applications of DNA in
molecular electronics has fueled interest in understanding the
detailed mechanisms of the hopping and trapping reactions.7 A
significant open question concerns the role played by the guanine
N1 imino proton in these processes.
Conversion of 2′-deoxyguanosine to its radical cation causes an
enormous increase in its acidity.8 On the basis of solution-phase
pKa data, Steenken9 concluded that oxidation causes the proton on
N1 of guanine in a DNA G/C base pair to shift spontaneously to
N3 of the cytosine (KEQ ) 100.4, ∆G ) -1.5 kcal/mol at room
-
Figure 1. Reaction of oxidized guanine with H2O, O2, and O2
.
temperature), see eq 1. However, DFT calculations in the gas phase,
Figure 2. Schematic representation of the DNA oligomers.
which is suggested to be a realistic model of stacked base pairs in
DNA (where solvent is excluded), indicate that the structure with
a proton on guanine N1 [C/(H)G•+] is more stable than [C+(H)/
G•] by 1.4 kcal/mol.10 In contrast, first-principles calculations on a
partially hydrated G/C base pair in DNA indicate that the proton
transferred form [C+(H)/G•] has an energy that is 4.0 kcal/mol below
that of [C/(H)G•+].11 Furthermore, an extensive calculation on a
related system indicates that charge transfer in oxidized duplex DNA
is coupled with proton transfer from guanine to cytosine.12 This
conclusion is consistent with experiments carried out in D2O that
reveal a kinetic isotope effect for guanine oxidation13 and for charge
transfer in DNA,14 both of which implicate a concerted proton-
coupled electron transfer involving the guanine N1 proton, and with
experiments that show inhibition of charge transfer when proton
loss from guanine is facilitated.15
The N1 proton is also thought to play the deciding role in the
chemical reactions of oxidized guanines.16-19 In solution, rapid loss
of this proton and subsequent reaction of the resulting guanine
radical with O2 leads eventually to an oxazolone (dZ, see Figure
1). However, proton loss from a guanine radical cation is slowed
when it is part of a base pair with cytosine in DNA, and in this
form it reacts with H2O to form 8-oxo-7,8-dihydroguanine (8-oxoG).
We report here an investigation of the one-electron oxidation of
a duplex DNA oligomer that contains 5-fluoro-2′-deoxycytidines
(F5dC) in selected positions complementary to GG steps. As
expected, F5dC is a weaker base than 2′-deoxycytidine, with ∆(pKa)
) 1.7 determined by titration in solution (see Supporting Informa-
tion). Consequently, proton transfer from a guanine radical cation
to F5dC is thermodynamically unfavorable. We find that replace-
ment of dC by F5dC has no measurable effect on long-distance
radical cation transport, but F5dC partially inhibits the reaction,
leading to strand cleavage at the complementary guanine.
The DNA oligomers examined in this work are shown in Figure
2. DNA(1) contains a series of six GG steps separated by TT
sequences, an anthraquinone group (AQ) linked covalently to a 5′-
terminus, and a 32P radiolabel (* in Figure 2) on the 3′-terminus of
the GG-containing strand. As expected, irradiation of DNA(1) (350
nm, only the AQ group absorbs) and its subsequent treatment with
formamidopyrimidine glycosylase (Fpg enzyme) results in strand
cleavage at each GG step, which is detected by PAGE and
autoradiography (see Figure 3A). Importantly, Fpg cleaves DNA
at sites that contain either a dZ or an 8-oxoG lesion,20 and F5dC
does not affect the ability of Fpg to induce strand cleavage (see
Supporting Information). The electronically excited state of AQ
causes the one-electron oxidation of an adjacent nucleobase to its
radical cation, which hops (khop) through the duplex until it either
is trapped (ktrap) by reaction with H2O or O2 or is consumed (ka) in
an annihilation reaction21 that leads to regeneration of the guanine.19
This experiment was carried out under single-hit conditions (low
conversion, see Supporting Information), where each DNA molecule
reacts once or not at all. In this circumstance, analysis of the linear
semilog plot of the distance-dependent distribution of strand
cleavage (see Figure 3B) gives the ratio of the rate of radical cation
hopping to its consumption by trapping and annihilation kratio
)
9
4172
J. AM. CHEM. SOC. 2006, 128, 4172-4173
10.1021/ja0573763 CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
solvent environment are in a conformation that enables reaction to
occur. Clearly, the decreased basicity of F5dC shifts KEQ and reduces
the amount of guanine radical available for reaction with O2, which
may cause a concomitant reduction in the magnitude of ktrap, and
thus a greater fraction of the radical cation will be consumed by
the annihilation reaction (ka, see Figure 1) that simply regenerates
dG and reduces the amount of strand cleavage.
The complexity of DNA is reflected in the analyses of the
hopping and trapping reactions. For example, the replacement of
dC by F5dC may modify the base pair hydration environment, or,
operating through the hydrogen bonds, it may affect the electronic
structure of the guanine.23 However, neither of these effects is
observed when 5-methyl-2′-deoxycytidines are paired with GG
steps.24 Clearly, the most significant consequence of replacing dC
with F5dC is the reduction in basicity, and the findings reported
here are interpreted on this basis. Additional experiments are
underway to test this conclusion.
Figure 3. (A) Autoradiogram showing the results of irradiation of DNA-
(1) and DNA(2). The six GG steps of these oligomers are indicated by the
numbered arrows at the right, and F indicates the presence of F5dC at the
complementary strand of DNA(2). The three lanes correspond to D (dark
control) and 4 and 8 min of irradiation, respectively. (B) Semilog plots of
the amount of strand cleavage at the GG steps of DNA(1) determined by
phosphorimagery (circles, the solid line is the least-squares fit) and DNA-
(2) (triangles, the dotted line connects the data points), as a function of
distance from the AQ. Strand cleavage at each GG step (n) is normalized
to the total (t) amount of reaction at all GG steps.
Acknowledgment. This work was supported by the National
Science Foundation and by the Vassar Wooley Foundation. Dr.
Sriram Kanvah prepared the DNA samples.
Supporting Information Available: Preparation and characteriza-
tion of DNA oligomers, synthesis of F5C, pH titrations of dC and F5C,
quantitative analysis of irradiation experiments, reactivity of Fpg with
DNA containing F5dC, melting temperature and CD data. This material
is available free of charge via the Internet at http://pubs.acs.org.
10. This indicates that hopping from GG step to GG step in DNA-
(1) is approximately 10 times faster than the irreversible reactions
that consume the guanine radical cation.22
References
(1) Ly, D.; Kan, Y.; Armitage, B.; Schuster, G. B. J. Am. Chem. Soc. 1996,
118, 8747-8748.
(2) Schuster, G. B. Acc. Chem. Res. 2000, 33, 253-260.
(3) Giese, B.; Spichty, M.; Wessely, S. Pure Appl. Chem. 2001, 73, 449-
453.
(4) O’Neill, M. A.; Barton, J. K. J. Am. Chem. Soc. 2004, 126, 11471-11483.
(5) Lewis, F. D. Photochem. Photobiol. 2005, 81, 65-72.
(6) Kawai, K.; Majima, T. Pure Appl. Chem. 2005, 77, 963-975.
(7) Schuster, G. B. Long-Range Charge Transfer in DNA I, II; Springer-
Verlag: Heidelberg, 2004; Vol. 236, p 237.
DNA(2) contains F5dC bases in place of dC at positions
complementary to three of the six GG steps (GG2,4,6). Its irradiation
and subsequent treatment with Fpg also results in strand cleavage
at each of the GG steps (see Figure 3A). Significantly, the relative
amount of cleavage observed at the GG steps opposite unmodified
cytosines (GG1,3,5) in DNA(2) is the same within experimental error
as for DNA(1), but strand cleavage is inhibited at the positions
that are complementary to F5dC (see Figure 3B). These findings
show that the efficiency of charge hopping is unaffected by the
50-fold decrease in basicity of cytosine that accompanies its fluorine
substitution, but this modification does influence the reactions of
the complementary guanine radical cation with H2O or O2.
The role played by the imino N1 proton in charge migration in
DNA and in reaction of the guanine radical cation is illuminated
by these experiments. The reduced basicity of F5dC would inhibit
hopping of the radical cation from one GG step to the next if this
process is strongly coupled to proton transfer from guanine to
cytosine. The experiments reported here show that substitution with
F5dC does not measurably affect the efficiency of hopping. This
suggests that the N1 proton remains primarily on the guanine radical
cation even in a normal G/C base pair, as was suggested by
calculation.10 In this case, the kinetic isotope effect observed in
charge-transfer experiments carried out in D2O solution must
involve other protons of the DNA or may be attributed to
participation of water molecules tightly bound to the DNA. In
contrast, the irreversible chemical trapping of the radical cation
that is revealed by Fpg-induced strand cleavage at damaged
guanines is partially inhibited by substitution with F5dC. Apparently,
proton transfer from the guanine radical cation to its cytosine partner
plays an important role when the DNA duplex and its immediate
(8) Candeias, L. P.; Steenken, S. J. Am. Chem. Soc. 1989, 111, 1094-1099.
(9) Steenken, S. Biol. Chem. 1997, 378, 1293-1297.
(10) Li, X.; Cai, Z.; Sevilla, M. D. J. Phys. Chem. B 2001, 105, 10115-10123.
(11) Gervasio, F. G.; Laio, A.; Iannuzzi, M.; Parinello, M. Chem. Eur. J. 2004,
10, 4846-4852.
(12) Gervasio, F. G.; Laio, A.; Parrinello, M.; Boero, M. Phys. ReV. Lett. 2005,
94, 158103.
(13) Weatherly, S. C.; Yang, I. V.; Armistead, P. A.; Thorp, H. H. J. Phys.
Chem. B 2003, 107, 372-378.
(14) Shafirovich, V.; Dourandin, A.; Geacintov, N. E. J. Phys. Chem. B 2001,
105, 8431-8435.
(15) Giese, B.; Wessely, S. Chem. Commun. 2001, 2001, 2108-2109.
(16) Kasai, H.; Yamaizumi, Z.; Berger, M.; Cadet, J. J. Am. Chem. Soc. 1992,
114, 9692-9694.
(17) Burrows, C. J.; Muller, J. G. Chem. ReV. 1998, 98, 1109-1154.
(18) Cadet, J.; Douki, T.; Gasparutto, D.; Ravanat, J.-L. Mutat. Res. 2003,
531, 5-23.
(19) Misiaszek, M.; Crean, C.; Joffe, A.; Geacintov, N. E.; Shafirovich, V. J.
Biol. Chem. 2004, 279, 32106-32115.
(20) Tretyakova, N. Y.; Wishnok, J. S.; Tannenbaum, S. R. Chem. Res. Toxicol.
2000, 13, 658-664.
(21) Henderson, P. T.; Jones, D.; Hampikian, G.; Kan, Y.; Schuster, G. B.
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 8353-8358.
(22) Liu, C.-S.; Hernandez, R.; Schuster, G. B. J. Am. Chem. Soc. 2004, 126,
2877-2884.
(23) Kawai, K.; Wata, Y.; Hara, M.; Tojo, S.; Majima, T. J. Am. Chem. Soc.
2002, 124, 3586-3590. These authors report a slightly accelerated rate
constant for electron-transfer quenching in CH2Cl2 solution by a modified
Br5C/G base pair compared with C/G, which is attributed to a reduced
Eox for G in Br5C/G. The pKa of C is reduced by bromine substitution.
(24) Kanvah, S.; Schuster, G. B. J. Am. Chem. Soc. 2004, 126, 7341-7344.
JA0573763
9
J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006 4173