Formation of tricyclic [4.3.3.0] adducts between 8-oxoguanosine and tyrosine under conditions of oxidative DNA-protein cross-linking

Article abstract of DOI:10.1021/ja803896d

8-Oxo-7,8-dihydro-2′-deoxyguanosine (OG), a prevalent product of oxidative stress on cellular DNA, is readily further oxidized forming adducts with nucleophiles. In the presence of tyrosine or p-cresol, an unusual tricyclo[4.3.3.0] adduct has been characterized in both nucleoside and oligodeoxynucleotide studies. The adduct is more stable in oligomers than nucleosides and undergoes slow reversion and hydration to spiroiminodihydantoin. Copyright

Full text of DOI:10.1021/ja803896d

Published on Web 07/09/2008
Formation of Tricyclic [4.3.3.0] Adducts between 8-Oxoguanosine and
Tyrosine under Conditions of Oxidative DNA-Protein Cross-Linking
Xiaoyun Xu, Aaron M. Fleming, James G. Muller, and Cynthia J. Burrows*
Department of Chemistry, UniVersity of Utah, 315 South 1400 East, Salt Lake City, Utah 84112-0850
Received May 23, 2008; E-mail: burrows@chem.utah.edu
Scheme 1. Adduct Formation between OG and Tyrosine or
p-Cresol
The human genome undergoes thousands of chemical modifica-
tions per day, many of which result in a loss of fidelity, compromis-
ing the integrity of the genetic blue print.1 The intimate interaction
between DNA and myriad proteins results in DNA-protein cross-
links (DPCs) during some of the oxidative damage events.2
Oxidation of either nucleic acids or proteins generates electrophilic
species, which can then be trapped by a nucleophilic group of the
partner biopolymer, leading to covalent bond formation between
DNA and the bound protein. This class of DNA lesions remains
among the least well characterized.
Several groups have explored DPCs by employing either model
reaction systems or by investigating fragments obtained from DNA
and protein digestion. For example, a covalent adduct formed
between C8 of guanine and the ꢀ-amino group of lysine during
guanosine (dGuo) oxidation has been thoroughly characterized.3,4
Recently, our laboratory showed that the roles of nucleophilic and
electrophilic partners could be reversed, for example, lysine could
be oxidized to an ꢀ-aminium radical that attacks the guanine base
at C8, with the final product structure being dependent on the
mechanism of oxidation.5 The competition of reaction mechanisms
arises because of the ambiguity of which partner, DNA or protein,
is initially oxidized and which serves as the nucleophile to form
the covalent adduct.
A similar question arises with the phenol-containing amino acid
tyrosine, particularly when interacting with DNA containing the
oxidation-sensitive site 8-oxo-7,8-dihydro-2′-deoxyguanosine (dOG)
whose redox potential is slightly lower than that of Tyr. Although
some information is available on cross-linking of Tyr and Lys with
pyrimidines under photochemical6,7 and oxidative conditions,8–10
no adducts between Tyr and dGuo or dOG have been well
characterized, although their formation was implicated in studies
of DPC formation with single-stranded binding protein (SSB) and
dOG-containing oligonucleotides,11 and other examples suggest that
phenol radicals add to C8 of dGuo.12–14 Tyrosine residues play
critical roles in certain DNA-processing enzymes including topoiso-
merases15 and are prevalent in the DNA-binding region of high
copy number proteins such as SSB.16 We report here an unusual
tricyclic adduct as the major product obtained from oxidation of
dOG in the presence of Tyr or its analogue p-cresol (Scheme 1).
In the experiment, 3 mM dOG (1) was allowed to react with 30
mM N-acetyltyrosine (2) in the presence of 10 mM K3Fe(CN)6 at
37 °C for 30 min and then quenched by addition of 10 mM
ascorbate. HPLC analysis of the reaction mixture showed two new
peaks with identical UV-vis spectra (λmax at 194 and 289 nm),
assigned as diastereomers 3a and 3b on the basis of the analytical
data described herein (Scheme 1). To simplify the structural
characterization of the tyrosine adducts, p-cresol (4) was employed
as a model compound. HPLC analysis of an analogous reaction
conducted with p-cresol again showed two peaks that possessed
identical UV-vis spectra (λmax at 196 and 292 nm) similar to the
above and consistent with the formation of diastereomers 5a and
5b. MS and MS/MS analyses of the adducts formed with Tyr and
p-cresol reactions also indicate the products from these reactions
are very similar.
NMR data were collected on the diasteromeric products from
the p-cresol reaction with dOG. Three aromatic protons were
observed with H3′′ and H5′′ showing HMBC correlations with the
methyl carbon at 21 ppm. The third, H6′′, was a doublet and coupled
to H5′′. Furthermore H3′′ showed an HMBC correlation to C5 of
the base, clearly supporting the C-C bond between the ortho
position of the phenol and C5 of the base. In the HMBC study,
both C4 (111 ppm, coupled to H1′) and C5 (61 ppm, coupled to
H3′′) showed large upfield shifts (∼30 ppm), while C8 (157 ppm,
coupled to H1′) moved modestly downfield (5 ppm) compared to
dOG.17 To rationalize these large shifts and the MS data the tricyclic
structure 5a/5b is proposed (Figure 1).
In previous studies, oxidation of dOG in the presence of water
or lysine led initially to adducts at C5 that underwent either
rearrangement to spiroiminodihydantoin (Sp) or decomposition to
guanidinohydantoin (Gh) structures.11,18–20 In contrast, phenols
appear to add to both the C4 and C5 positions of dOG during
oxidation, initially without rearrangement. To lend further support
to this surprising structure, the 13C NMR resonances and relative
energies for the aglycons 5a/5b, as well as other likely isomers
Figure 1. HMBC NMR spectrum of 5a, 5b.
9
10080 J. AM. CHEM. SOC. 2008, 130, 10080–10081
10.1021/ja803896d CCC: $40.75
2008 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Calculated Relative Energies for C13H13N5O3 Isomers and
13C Chemical Shifts of C4, C5, and C8 for the Corresponding
2′-Deoxyribonucleoside
Supporting Information). Adducts in oligomers were somewhat
more stable with ∼30% degradation after 24 h incubation at pH
7.5. The predominant degradation process is hydration, yielding
compound 9 which could also be characterized by HPLC and MS
before slow loss of the phenol to form 10. Adducts 5a/5b have
longer half-lives at pH 7.0 and 6.0 (∼48 and 100 h, respectively)
and were stable for weeks in organic solvents such as acetone or
DMSO. Therefore, structures such as 5a/5b could have long
lifetimes in the context of a dOG-Tyr cross-link formed in a
DNA-protein complex that excludes bulk water.
In summary, we have characterized an unusual tricyclo[4.3.3.0]
adduct in oxidative guanine-tyrosine cross-linking. Such “paddlane”
structures are uncommon in organic chemistry, and the inherent
ring strain may account for the observed slow decomposition of
the adduct ultimately to yield the more common spirodihydantoin
10, a product of 4-e- oxidation of dGuo or 2-e- oxidation of dOG.
Overall, these explorations of small molecule reactions have
important implications in the chemical nature of DNA-protein
cross-links derived from oxidative stress.
Acknowledgment. We are grateful to the National Science
Foundation for support of this work (Grant CHE-0514612). We
thank Professor Ilya Zharov, Dr. Anita Orendt, and Dr. Yu Ye for
assistance with NMR calculations and experiments.
Scheme 2. Reaction Pathway for p-cresol With dOG
Supporting Information Available: Complete ref 21, experimental
procedures, HPLC analysis of the reactions of dOG and N-AcTyr and
p-cresol, MS/MS of 3a/3b, 5a/5b, and 9, 13C chemical shifts and NMR
spectrum of 5a/5b, and energies and geometries of calculated structures.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Sarasin, A. Mutat. Res. 2003, 544, 99–106.
(2) Barker, S.; Weinfeld, M.; Murray, D. Mutat. Res. 2005, 589, 111–135.
(3) Morin, B.; Cadet, J. J. Am. Chem. Soc. 1995, 117, 12408–12415.
(4) Perrier, S.; Hau, J.; Gasparutto, D.; Cadet, J.; Favier, A.; Ravanat, J. L.
J. Am. Chem. Soc. 2006, 128, 5703–5710.
(5) Xu, X.; Muller, J. G.; Ye, Y.; Burrows, C. J. J. Am. Chem. Soc. 2008, 130,
703–709.
(6) Shaw, A. A.; Falick, A. M.; Shetlar, M. D. Biochemistry 1992, 31, 10976–
10983.
(7) Sun, G.; Fecko, C. J.; Nicewonger, R. B.; Webb, W. W.; Begley, T. P.
Org. Lett. 2006, 8, 681–683.
with the molecular formula C13H13N5O3, were calculated using
Gaussian 98 with the B3LYP/6-31G(d)//B3LYP/6-311+G(2d,p)
method (Table 1).21,22 The relative energies were obtained at the
6-311+G(2d,p) level and corrected by zero-point vibrational
energies from B3LYP/6-31G(d) frequency calculations. The cal-
culated 13C chemical shifts for C-4, C-5, and C-8 in 5a/5b best fit
the experimental values of 112, 60, and 156 ppm, respectively,
supporting the assignment as a tricyclo[4.3.3.0] adduct.23
The reaction is proposed to be initiated by oxidation of dOG (1)
to the electrophilic quinoid dOGox (6) which then undergoes
nucleophilic attack at C5 by the phenol to form 7, analogous to
the addition of other nucleophiles such as lysine to this position
(Scheme 2). In the special case of a phenol, the phenolic oxygen is
now well positioned for a second nucleophilic addition to C4,
generating 3a/3b or 5a/5b. This reaction also parallels the addition
of phenol or naphthol to acid-generated cationic species, in which
the o-carbon reacts with an electrophilic site first, followed by
addition of the phenolic oxygen to another electrophilic site.24
In aqueous phosphate buffer at pH 7.5, the products 5a/5b have
a half-life of ∼20 h, undergoing slow hydration and decomposition
to ultimately yield Sp (10). However, analogous adducts could be
formed in a tetramer d(AGXC) (X ) 5a or 5b) or an 18-mer (see
(8) Margolis, S. A.; Coxon, B.; Gajewski, E.; Dizdaroglu, M. Biochemistry
1988, 27, 6353–6359.
(9) Dizdaroglu, M.; Gajewski, E. Cancer Res. 1989, 49, 3463–3467.
(10) Gajewski, E.; Dizdaroglu, M. Biochemistry 1990, 29, 977–980.
(11) Johansen, M. E.; Muller, J. G.; Xu, X.; Burrows, C. J. Biochemistry 2005,
44, 5660–5671.
(12) Muller, J. G.; Paikoff, S. J.; Rokita, S. E.; Burrows, C. J. J. Inorg. Biochem.
1994, 54, 199–206.
(13) Stemmler, A. J.; Burrows, C. J. J. Am. Chem. Soc. 1999, 121, 6956–6957.
(14) Weishar, J. L.; McLaughlin, C. K.; Baker, M.; Grabryelski, W.; Manderville,
R. A. Org. Lett. 2008, 10, 1839–1842.
(15) Dong, K. C.; Berger, J. M. Nature 2007, 450, 1201–1205.
(16) Raghunathan, S.; Kozlov, A. G.; Lohman, T. M.; Waksman, G. Nat. Struct.
Biol. 2000, 7, 648–652.
(17) Cho, B. P. Magn. Reson. Chem. 1993, 31, 1048–1053.
(18) Luo, W.; Muller, J. G.; Burrows, C. J. Org. Lett. 2001, 3, 2801–2804.
(19) Luo, W.; Muller, J. G.; Rachlin, E. M.; Burrows, C. J. Org. Lett. 2000, 2,
613–617.
(20) Luo, W.; Muller, J. G.; Rachlin, E. M.; Burrows, C. J. Chem. Res. Toxicol.
2001, 14, 927–938.
(21) Frisch, M. J.; et al. Gaussian 98, revision A6; Gaussian Inc.; Pittsburgh,
PA, 1998.
(22) McCallum, J. E. B.; Kuniyoshi, C. Y.; Foote, C. S. J. Am. Chem. Soc.
2004, 126, 16777–16782.
(23) The adduct 5a/5b was also formed in an 18-mer and characterized after
nuclease digestion to the deoxynucleoside (see Supporting Information).
(24) Goossens, R.; Lhomeau, G.; Forier, B.; Toppet, S.; Meervelt, L. V.; Dehaen,
W. Tetrahedron 2000, 56, 9297–9303.
JA803896D
9
J. AM. CHEM. SOC. VOL. 130, NO. 31, 2008 10081