Copper/H2O2-Mediated oxidation of 2′-deoxyguanosine in the presence of 2-naphthol leads to the formation of two distinct isomeric adducts

Article abstract of DOI:10.1021/jo201423n

Exposure of cells to phenolic compounds through exogenous and endogenous sources can lead to deleterious effects via nucleobase modifications of DNA occurring under oxidative conditions. 2′-Deoxyguanosine (dG) is the most electron rich of the four canonical bases and includes many nucleophilic sites; it is also susceptible to oxidation with numerous reactive oxygen species. In these studies, dG was allowed to react with 2-naphthol in the presence of copper or iron salts yielding two principal isomeric products. Spectroscopic analysis and reactions with alkylated nucleosides support the assignment of compound 1a/1b as a pair of atropisomer N² adducts and compound 2a/2b as a diastereomeric mixture of tricyclic [4.3.3.0] adducts. Both products are the result of an overall four-electron oxidation process and consequently have the same masses, though drastically different structures, providing mechanistic insight into their formation. Thus, dG alkylation by 2-naphthol under oxidative conditions yields products whose structural properties are altered, leading to potentially mutagenic effects in genomic DNA.

Full text of DOI:10.1021/jo201423n

ARTICLE
pubs.acs.org/joc
Copper/H2O2-Mediated Oxidation of 20-Deoxyguanosine in the
Presence of 2-Naphthol Leads to the Formation of Two Distinct
Isomeric Adducts
Aaron M. Fleming, Arunkumar Kannan, James G. Muller, Yi Liao, and Cynthia J. Burrows*
Department of Chemistry, University of Utah, 315 South 1400 East, Salt Lake City, Utah 84112-0850, United States
S Supporting Information
b
ABSTRACT: Exposure of cells to phenolic compounds through
exogenous and endogenous sources can lead to deleterious
effects via nucleobase modifications of DNA occurring under
oxidative conditions. 20-Deoxyguanosine (dG) is the most
electron rich of the four canonical bases and includes many
nucleophilic sites; it is also susceptible to oxidation with
numerous reactive oxygen species. In these studies, dG was
allowed to react with 2-naphthol in the presence of copper or
iron salts yielding two principal isomeric products. Spectroscopic analysis and reactions with alkylated nucleosides support the
assignment of compound 1a/1b as a pair of atropisomer N2 adducts and compound 2a/2b as a diastereomeric mixture of tricyclic
[4.3.3.0] adducts. Both products are the result of an overall four-electron oxidation process and consequently have the same masses,
though drastically different structures, providing mechanistic insight into their formation. Thus, dG alkylation by 2-naphthol under
oxidative conditions yields products whose structural properties are altered, leading to potentially mutagenic effects in genomic DNA.
’ INTRODUCTION
Phenols provide two modes of reactivity with the bases of
nucleic acids under conditions of oxidation, either as nucleophiles
or as radical-generating species. 20-Deoxyguanosine (dG) is the
most electron rich of the four canonical nucleosides and provides
a reactive heterocyclic platform for coupling to phenols.1,2
Oxidation of aromatic compounds to either hydroquinones or
catechols provides activated phenols that can enter into a stepwise
Figure 1. Points of alkylation/arylation on dG by electrophilic phenols.
two-electron redox cycle in the presence of a metal catalyst and
reductant. Phenolic species known to arylate dG include phenoxyl
dG under oxidative conditions through their corresponding
radicals (PhO•), quinones (Q), semiquinone radicals (SQ•), and
phenoxyl radicals (Figure 1).16ꢀ18 These studies highlight the
quinone methides (QM).2,3 SQ• and Q electrophiles result from
fact that phenol electrophiles can specifically arylate many of the
redox-active hydroquinones and catechols that can react with dG
possible sites on dG. A consequence of dG being electron rich is
at N2, N7 and C8 as exemplified by the products observed from
that it has the lowest standard reduction potential (E0 = 1.29 V vs
3,4-catecholic estrones in the presence of dG under oxidative
NHE, pH 7) of the four canonical bases and is prone to oxidation
conditions (Figure 1).4ꢀ6 In contrast, 4-hydroxyequilenin SQ• is
reactions itself as well as alkylation/arylation reactions.19 Thus,
proposed to couple initially at the N2 of dG; next, N1 of dG acts as
oxidative coupling of purines with phenolic compounds provides
a nucleophile to form a new ring between the WatsonꢀCrick face
diverse products depending on which partner is oxidized first and
of dG and C2 of 4-hydroxyequilenin.5 2,3-Catecholic estrones
which serves as the nucleophile or electrophile.
when oxidized diverge from the previous path due to their ability
Nucleoside dG oxidation with radiolytically formed HO•,
to tautomerize to QM that are attacked by N2 of dG yielding
•-
SO4 , or photoexcited riboflavin yields many oxidation
stable products.7 Further studies have shown that QMs are
products.1,20 Products characterized from chemistry at C5 of
predominantly attacked by N2 of dG, but alkylation was also
dG include imidazolone (Iz) and itshydrolysisproduct oxazolone
observed at N1 and N7 of dG.8,9 Hydroquinones are oxidized to
(Z), which has been observed in vivo.21 In the case of dG
electrophilic p-quinones that are nucleophilically attacked by N2
oxidation by CuII/H2O2/reductant,22 NiCR/KHSO5,23 Mn-
of dG followed by a second nucleophilic attack by N1 yielding
TMPyP/KHSO5,20 or the epoxidizing agent dimethyldioxirane,24
unique tricyclic compounds.10ꢀ13 However, in the case of cate-
chols, they are oxidized to o-quinones that undergo conjugate
addition by N7 of dG; these have all been characterized as the
Received: July 11, 2011
Published: August 25, 2011
aglycone products.7,14,15 Lastly, hindered phenols react with C8 of
r
2011 American Chemical Society
7953
dx.doi.org/10.1021/jo201423n J. Org. Chem. 2011, 76, 7953–7963
|

The Journal of Organic Chemistry
ARTICLE
Scheme 1. Products Observed from the Initial Oxidation of dG at C5 or C8
oxidation at C5 was observed to yield 5-carboxamido-5-forma-
mido-2-iminohydantion (2Ih, Scheme 1).
Scheme 2. Proposed Pathway for Forming the Electrophile
dOGox from dG and dOG
Oxidation of dG by HO• in DNA yields 8-oxo-7,8-dihydro-
20-deoxyguanosine (dOG) as one of the major products under
aerobic oxidative conditions (Scheme 1),25,26 and its in vivo con-
centration is monitored to determine oxidative stress levels.27,28 In
addition, 2,6-diamino-4-hydroxy-5-formamidopyrimidine (Fapy-G)
is observed upon exposure of dG to HO• in the presence of
reductants or anaerobic reaction conditions (Scheme 1).29
A
striking feature of dOG is its dramatically decreased reduction
potential (E0 = 0.74 V vs NHE, pH 7) relative to dG, causing
dOG oxidation to be much more facile than that of dG.30 The
initial one-electron oxidation of dOG yields a radical cation/
radical pair (dOG•+/dOG•) with a pKa of ∼6.6 and a lifetime
that is dependent on the reaction context and the level of
oxidative stress (Scheme 2).31,32 In the presence of excess
oxidant, a second one-electron oxidation of dOG•+/dOG• can
occur yielding the proposed iminoquinone intermediate dOGox
that is trapped at C5 by H2O serving as a nucleophile yielding a
second intermediate 5-HO-dOG (Scheme 2). dOGox is the
proposed intermediate from the four-electron oxidation of dG by
1O2 that yields the intermediate 5-HO-dOG.33 5-HO-dOG is
short-lived and rearranges or decomposes through three possible
pathways:34 (1) a 1,2-acyl migration yielding spiroiminodihy-
dantoin (Sp, Figure 2); (2) hydration at C6 followed by
decarboxylation to yield guanidinohydantoin (Gh, Figure 2);
or (3) addition of a second nucleophilic H2O at C4, followed
by decomposition to 4-hydroxy-2,5-dioxo-imidazolidine-4 car-
boxylic acid (HICA, Figure 2). Sp is the major nucleoside
product observed at pH 7,35ꢀ41 Gh is the major product obser-
ved under slightly acidic conditions or duplex DNA,36,42ꢀ45 and
HICA is a minor product at neutral pH that is unstable and
readily decomposes.46 Overall, the oxidation of purines leads to
formation of electron-deficient species that are subject to attack
by nucleophiles, such as water. Because oxidation reactions do
not occur in isolation under in vivo conditions, it has been of
interest to see the structural effects that other nucleophiles have
on the products derived from dOG oxidation.
hydration and decomposition to afford the carboxylic acid end
product (Figure 2).46 Oxidation of dOG with Na2IrCl6 in the
presence of a polyamine, such as spermine (Spm), initially forms
the proposed intermediate 5-Spm-dOG that has limited stability
and decomposes to a Ua lesion (Figure 2).48 When dOG
nucleoside is oxidized by Na2IrCl6 in the presence of a primary
amine, such as Nα-acetyl-lysine (Lys), the final product observed
is an analog of Sp, with the ε-amino nitrogen of lysine covalently
linked to C5 of dOG yielding, 5-Lys-Sp (Figure 2).49 Both amine
adducts are proposed to be the result of the amine nucleophile
adding to C5 of dOGox, in which the primary amine (Lys) can
undergo the 1,2-acyl migration to the spirocycle. However, the
polyamine (Spm) initially forms a short-lived intermediate that
then decomposes through a spiroaminal intermediate to a Ua
lesion.48 Finally, when dOG is oxidized with K3Fe(CN)6 in the
presence of tyrosine or p-cresol, the product characterized was a
tricyclic [4.3.3.0] species (collectively named 4,5-PhO-dOG)
with the phenol covalently bound at both C4 and C5 of dOG
(Figure 2).50 The tricyclic product is proposed to result from the
ambidentate nature of the nucleophilic phenolate anion. This
product showedhigh stability in organicsolvents but decomposed
in aqueous solutions.50
Several adducts of dOG oxidation have already been charac-
terized. Studies with the oxidant ONOOꢀ at high concentrations
suggestthat ONOOꢀ is the nucleophile thataddsto C5 of dOGox
ultimately yielding dehydroguanidinohydantoin (DGh) as an
intermediate on the pathway to the decomposition product
ribosyl-urea (Ua, Figure 2).47 Oxidation of dOG by ONOOꢀ
in the presence of 30% MeOH provides a HICA analog, MICA,
bearing a methoxy group at C5. MICA is proposed to arise from
the coupling of MeO• to C5 of dOG• that then undergoes
These studies have shown that dOG is reactive with radicals at
C5, and most of the products are similar in structure to those
derived from nucleophilic addition (e.g., MICA and 5-Lys-Sp),
while nucleophiles that have more than one reactive site (e.g.,
Spm and phenols) yield unique products that have limited
stability and readily decompose in aqueous conditions. The work
reported herein involves the oxidation of dG with CuII/H2O2 or
FeIII/EDTA/H2O2/ascorbate, as well as the reaction of dOG
7954
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963

The Journal of Organic Chemistry
ARTICLE
Figure 2. Products characterized from dOG oxidation in the presence of various nucleophiles. The products Sp, Gh, and HICA are H2O adducts; when
dOG was oxidized in the presence of MeO•, MICA was observed, nucleophilic ONOOꢀ yielded DGh, oxidation in the presence of Spm yielded 5-Spm-
OG and its decomposition product Ua, while the presence of Lys yielded 5-Sp-Lys, and the phenols p-cresol and Tyr yielded 4,5-PhO-dOG.
with Na2IrCl6 in the presence of 2-naphthol. 2-Naphthol was
selected for study because its redox potential is very similar to
dOG, posing interesting questions about the interplay of com-
peting pathways. In addition, 2-naphthol mimics certain B-ring
aromatized metabolic products of estrogens, such as equilenin.
All of the reactions with 2-naphthol yield the same products with
the two major products being a newly observed stable adduct to
the exocyclic amine of dOG (compounds 1a/1b), and the other
products are a diastereomeric pair of tricyclic [4.3.3.0] adducts of
2-naphthol to C4 and C5 of dOG (compounds 2a/2b). The
mass balance is completed by dG oxidation products for which
2-naphthol did not participate in the reaction.
HPLC retention time and masses as those seen in the dG reaction
with CuII/H2O2 and 2-naphthol. Purification of the two peaks
followed by ESI+-MS/MS analysis of the free base adducts
((M + H)+ = 310) from both reactions supports the observation
that the products from the arylation of dG and dOG by
2-naphthol under oxidative conditions are the same. Additional
support for both reactions yielding the same products was
1
established by H NMR spectroscopy. The details of these
experiments will be discussed in greater detail while examining
the data that led to the proposed structures. Due to the higher
yields produced, the reaction with dOG was chosen for prepara-
tion and purification of sufficient quantities for further spectro-
scopic analysis. This analysis led to the conclusion that
compounds 1a/1b are an N2 adduct of dOG that exists as a pair
of atropisomers, while compounds 2a/2b are a diastereomeric
pair of the tricyclic [4.3.3.0] adducts 2a (4S,5S) and 2b (4R,5R)
formed between the dOG nucleobase and 2-naphthol. The 2a/2b
products mirror those recently characterized from the oxidation
of dOG in the presence of tyrosine or p-cresol(4,5-PhO-dOG).50
Spectroscopic Analysis for 1a/1b. Structural characteriza-
tion of 1a/1b was achieved through a combination of UVꢀvis,
MS, and NMR analysis after purification, and was aided by
mechanistic insight into dG alkylation. The UVꢀvis spectrum of
1a/1b retained the long wavelength absorption of the dOG
chromophore (λmax = 295 nm) suggesting that the aromaticity of
the nucleobase had remained intact. ESI+-MS/MS analysis for
the free base (M + H)+ = 310) of 1a/1b yielded very little
structural information due to the presence of only one fragment
that is consistent with the loss of the 2-naphthol ((M + H)+ =
166). The 1H NMR spectrum indicated the loss of one aromatic
proton from 2-naphthol, consistent with one covalent bond
being formed between dOG and a carbon atom of 2-naphthol
(Figure 3). Through COSY, HSQC, and HMBC assignments, it
was determined that the missing aromatic proton and point of
connectivity to dOG was at C1 of 2-naphthol (C100 in 1a/1b).
The anomeric proton of 1a/1b (6.17 ppm) presented as a
doublet of doublets in the 1H NMR spectrum. This observation
ruled out adducts that would exist between 2-naphthol and dOG
at the two carbon atoms C4 and C5; in these cases, formation of
an additional stereocenter would present these resonances as a
set of two doublets of doublets. Because 1a/1b eluted from the
HPLC as two peaks and the anomeric proton was a single
doublet of doublets (Figure 3), we concluded that both peaks
’ RESULTS AND DISCUSSION
When the nucleoside dG and 2-naphthol were allowed to react
in the presence of CuII and H2O2, two new products were
identified by reversed-phase HPLC, in which each elutes as two
peaks assumed to represent a pair of stereoisomers. The first pair
of peaks, compounds 1a/1b, and the second pair of peaks,
compounds 2a/2b each represented a yield of ∼40ꢀ45%. The
mass balance was completed with the dG oxidation products Sp,
Gh, Z, and 2Ih, each with yields <5%. For 1a/1b, the ratio of the
two peaks was essentially 1:1, while for 2a/2b, the presumed
diastereomeric ratio was about 60:40. These compounds were
quantified through their extinction coefficients following pre-
viously established procedures (Supporting Information).22 The
ESI+-MS analysis for the new peaks showed that both new
products had neutral masses of 425 (dG + 2-naphthol +16 - 2)
suggesting arylation of dG by 2-naphthol plus the formal addition
of an oxygen atom had occurred under the oxidative conditions.
The additional oxygen atom could be located on the phenol or
the nucleobase. An oxygen atom addition to the phenol was ruled
out by spectroscopic data discussed later. Furthermore, our
experience with the oxidation of dG suggested that the oxygen
atom could be the result of a two-electron oxidation of dG to
dOG followed by further reaction of dOG to yield the
adducts.35,42 To substantiate this hypothesis, a reaction was
conducted in which dOG and 2-naphthol were allowed to react
in the presence of the oxidant Na2IrCl6, since this transition
metal oxidant has been shown previously to oxidize selectively
dOG in high yields.35,42 The Na2IrCl6 initiated reaction with
dOG and 2-naphthol furnished two products that had the same
7955
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963

The Journal of Organic Chemistry
ARTICLE
Figure 4. Key HMBC (500 MHz, DMSO-d6, 298 K) correlations for
1a/1b. The arrows show correlations that establish C100 as the site of
adduction on 2-naphthol. Also, the C4 and C8 resonances were not
perturbed compared to dOG that were determined through their
correlation with H10, thus establishing that the aromaticity of the purine
moiety remains intact.
Figure 3. Partial 1H NMR spectra (400 MHz, D2O, 298 K) for 1a/1b,
and 2a/2b. Adducts to dOG that form a new stereocenter (2a/2b) show
H10 as two doublets of doublets. In the aromatic region, both 1a/1b and
2a/2b show only six resonances indicating adduction via a ring carbon of
2-naphthol.
however, the pKa of N7 of dOG (11.3) is much higher than that of
N1 (8.6), so if arylation occurs through this mechanism, N1
should be the preferred site of reaction.30,55 Another way to
deprotonate N7 is through oxidation of dOG to dOGox, but this
product is electrophilic and adds nucleophiles at C5, as previously
discussed.30,56 Furthermore, if arylation had occurred at N7, a
large 13C chemical shift change (>1.7 ppm) would be expected for
C8 and C4 of the base54 that was not observed.
are atropisomers of the same species, and not two similar products
(e.g., constitutional isomers) eluting closely in the HPLC. The
HMBC spectrum showed correlations of the anomeric proton to
C4 and C8 of 1a/1b, and their chemical shifts were only slightly
different from the parent nucleoside dOG (Figure 4).51 The
observed 13C chemical shifts corresponded to upfield perturba-
tions of 0.1 ppm (147.1 to 147.0 ppm) for C4 and 1.1 ppm (151.6
to 150.5 ppm) for C8.51 This provided two critical insights: (1)
thatthe sugar andbasearestillcovalently attached, and(2) thatthe
chemical environments for C4 and C8 have not changed drama-
tically after arylation by 2-naphthol. The 1H NMR spectrum for
1a/1b in DMSO-d6 showed conflicting data with respect to the
number of exchangeable protons that were present, and the acidic
phenol proton overlapped with the alcohol protons on the sugar.
To gain an accurate count of these protons, ESI+-MS analysis in
deuterated solvents was used to count the six exchangeable
protons in the presumptive structure, 1a/1b, in which dOG and
2-naphthol have a total of seven exchangeable protons. The fact
that one exchangeable proton was missing helped confirm that the
covalent linkage between dOG and 2-naphthol is positioned at a
nitrogen of the purine. Further support for arylation at nitrogen is
derived from the reactions discussed below with N1-methyl-dOG.
When N1-methyl-dOG and 2-naphthol were allowed to react in
the presence of Na2IrCl6, analogous compounds to 1a/1b were
observed (discussed below).
N1 and N2 are the other possible points of bond formation
between dOG and 2-naphthol. Because the 1H NMR spectrum
did not provide sufficient data to rule out one adduct over the
other, and the 13C NMR data could not be used to uniquely
identify C2, C5, and C6 of 1a/1b, N-alkyl dOG derivatives were
synthesized. N1-Methyl-dOG and a N2-diethyl-dOG derivative
were prepared and allowed to react with 2-naphthol in the
presence of Na2IrCl6. Under these conditions N1-methyl-
dOG led to formation of two compounds (3a/3b and 4a/4b)
with 2-naphthol (Scheme 3). Compounds 3a/3b led to an ESI+-
MS/MS fragmentation pattern similar to that observed for
compounds 1a/1b, suggesting that these two compounds are
very similar and that arylation did not occur at N1 of dOG. The
other compounds, 4a/4b, gave the same ESI+-MS/MS fragmen-
tation pattern as observed with 2a/2b that helped aid in
identifying some of the 2a/2b fragments (Supporting In-
formation). When N2 was blocked by ethylation, the HPLC
indicated major products (5a/5b) that had a ESI+-MS/MS
fragmentation patterns consistent with a tricyclic adduct similar
to 2a/2b, but there was also an unknown minor product (6)
observed (Scheme 3 and Supporting Information). Compound 6
was not characterized due to its instability suggesting it is not
similar to compounds 1a/1b, because, as will be discussed later,
1a/1b is stable. The only position that would still provide
arylation when N1 is blocked and decrease when N2 is blocked
would be N3, but as stated this does not fit the NMR data. We
conclude that products 1a/1b are the atropisomers formed from
oxidative addition of 2-naphthol to N2 of dOG.
The identity of this heteroatom attachment was determined by
13
observing the C NMR shifts for C100 and C200 in the phenol
moiety of the proposed structure, 1a/1b. HMBC analysis aided
in the identification of C100 (114.6 ppm) through its correlations
to H300 and H800, and C200 (152.8 ppm) through its correlation to
H400, where all aromatic protons were identified by COSY and
HSQC. Comparison of these shifts to 2-naphthols with aryl
amines or aryl ethers covalently linked to C1 suggested that 1a/
1b has phenol carbons with chemical shifts that are consistent
with an aryl amine attached at C100 of 1a/1b,52,53 ruling out
arylation of dOG at either O6 or O8 of the heterocyclic base. The
remaining possible points of covalent attachment between C1 of
2-naphthol and dOG include the nitrogen atoms N1, N2, N3, and
N7. N3 arylation was ruled out for two reasons: (1) N3 is
sterically inaccessible for 2-naphthol addition, and (2) N3
arylation is expected to cause a significant perturbation of the
13C resonance at C4 of the base due to a change in the base
aromaticity, which was not observed.54
Spectroscopic Analysis for 2a/2b. The structure of com-
pounds 2a/2b was also determined by a combination of UVꢀvis,
ESI+-MS, MS/MS fragmentation and NMR analysis after HPLC
purification. The UVꢀvis spectrum of 2a/2b looked similar to
that of 2-naphthol (λmax = 227 nm), and the long-wavelength
absorption for the dOG chromophore was not observed, sug-
gesting significant perturbation to the aromatic base. The ESI+-
MS/MS fragmentation pattern for the free base of 2a/2b,
released during facile fragmentation of the glycosidic bond,
presented many daughter fragments that aided in the structural
Next, a possible N7 adduct for 1a/1b was ruled out by
mechanistic and 13C NMR chemical shift considerations. For a
ring nitrogen to act as the nucleophile, it must be unprotonated;55
7956
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963

The Journal of Organic Chemistry
ARTICLE
Scheme 3. Products Observed from Oxidation of dOG Alkyl Derivatives in the Presence of 2-Naphthol
analysis ((M + H)+ = 310, 269, 225, 182, and 86, where the alkyl
dOG derivatives were used to help identify the fragments; see
1
Supporting Information). The H NMR spectrum of 2a/2b
showed the presence of six aromatic protons suggesting covalent
attachment to one of the phenol carbons (Figure 3). The proton
assignments were confirmed by COSY, HSQC, and HMBC
correlations that supported arylation of dOG by C1 of
2-naphthol. The anomeric protons (5.75 and 5.66 ppm), also
identified by COSY and HSQC, presented as two doublets of
doublets suggesting one or more stereocenters are present in the
base. HMBC correlations from the anomeric proton to C8 and
C4 of the base show that C8 has shifted downfield by 4.4
ppm (151.6 to 156.2 ppm), while C4 shifted upfield by 35.4
ppm (147.1 to 111.7 ppm) compared to dOG (Figure 5).51 Due
to the slow relaxation time for C4, it was not identified in the 13C
NMR spectra. Also, C2, C5, and C6 were not unambiguously
identifiable from the NMR data. While C5 is important as the
point of covalent attachment between dOG and 2-naphthol, its
chemical shift is proposed by comparison to a previous study in
our laboratory.50 The adduct characterized between Tyr and
dOG under related reaction conditions gave a similar product,
4,5-PhO-dOG (Figure 2), for which C5 was identifiable through
an HMBC correlation. HMBC analysis for the previous phenol
adduct showed that C5, through its correlation to H300 had
shifted downfield by 37.5 ppm (98.5 to 61.0 ppm) compared to
dOG.50,51 Using this as a guide, C5 of 2a/2b is proposed to be
the peak at 63.2 ppm that has a comparable downfield shift of
35.3 ppm (98.5 to 63.2 ppm). From the comparison of NMR
data between 2a/2b and 4,5-PhO-dOG, and the observation
that 2a/2b and 4,5-PhO-dOG give similar ESI+-MS/MS frag-
mentations, the following structure for the tricyclic [4.3.3.0] 2a/
2b compounds is proposed to be comprised of the 4S,5S 2a and
4R,5R 2b diastereomers.
Two-Electron Oxidation that Initiates Adduct Formation.
To arrive at arylated products 1a/1b and 2a/2b from the
nucleoside dG, a four-electron oxidation had to occur. Com-
pounds 1a/1b are a result of a two-electron oxidation on both dG
and 2-naphthol, yielding a dOG aryl adduct. The structure of
compounds 2a/2b implies that a four-electron oxidation of dG
has occurred with 2-naphthol serving as a nucleophile. Standard
Figure 5. Key HMBC (500 MHz, DMSO-d6, 298 K) correlations for
2a/2b. The arrows show correlations that establish C100 as the site of
adduction on 2-naphthol. C4 and C8 resonances were assigned that both
show large perturbations, suggesting that the purine aromaticity has
been disrupted.
reduction potentials of each species indicate that 2-naphthol
should be the sole oxidized species in the reaction
(Table 1),30,35,57ꢀ60 and one of the main products from this
reaction was indeed 2-naphthol dimers and tetramers, as identi-
fied by HPLC and ESI+-MS. From LC-ESI+-MS and the mass
balance, all guanosine-derived products had a +16 additional
mass, indicative of the oxidation of dG to dOG. The CuII/H2O2
oxidant system is unusual in dG oxidation due to the coordina-
tion of copper to N7 of dG, wherein the bound copper
specifically directs oxidation on dG.22,61,62 It is proposed that
2-naphthol reduces the CuII to CuI, at which point CuI and H2O2
can mediate the two-electron oxidation of dG to dOG following
a Fenton-like reaction; the details of this reaction are still under
debate.1,63,64 In an attempt to better understand this step, reac-
tions were conducted utilizing the oxidant system FeII/EDTA/
H2O2/ascorbate,65 a known hydroxyl radical generating system,
for which compounds 1a/1b and 2a/2b were found, but in lower
yields than observed with the CuII/H2O2 system. The decreased
7957
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963

The Journal of Organic Chemistry
ARTICLE
Table 1. Standard Reduction Potentials for Chemical Species
of Interest
Pathway B. It is also possible that dOG is oxidized by
one-electron to dOG• (the pKa for this radical is ∼6.6),30
while 2-naphthol undergoes a one-electron oxidation to the
2-naphthoxyl radical (2-NpO•). Next, these two radicals couple
yielding a neutral intermediate 7 that enolizes to the more stable
final products 1a/1b (Scheme 4). The 2-NpO• has been shown
by EPR spectroscopy to have the largest population of unpaired
spin density at C1,82 which is consistent with the spectroscopic
data for compounds 1a/1b. A similar mechanism has been
proposed by the Tannenbaum laboratory for the generation of
MICA in which MeO• couples with dOG• at C5.36 DFT
calculations conducted on dOG• suggest that the N2 centered-
aminyl radical represents a smaller population of the unpaired
electron spin density when compared to C5 of dOG•.56 It is
possible that due to the steric bulk of the 2-NpO• and the
observation that this radical is longer lived than a single ring
phenoxyl radical, that coupling at the exocyclic amine of dOG• is
the preferred point of arylation.5 Interestingly, it has been
proposed that the dG aminyl radical couples with the o-semi-
quinone radical of equilenin, a biaryl catechol, yielding the initial
covalent attachment between these two compounds.5 This would
suggest a unique feature of naphthyl radicals to couple with the
exocyclic aminyl radicals of dG and dOG.
Compd
OH•,OHꢀ
dG+•,dG
PhO•,PhOꢀ
C6Cl5O•,C6Cl5Oꢀ
p-Cresol•,p-Cresolate
Na2IrCl6 (IrIV,IrIII
2-NpO•,2-NpOꢀ
Na2IrBr6 (IrIV,IrIII
dOG+•,dOG
1-NpO•,1-NpOꢀ
K3Fe(CN)6 (FeIII,FeII)
E0a
pKa
ref
1.9
62
19
60
60
60
57
58
57
30
58
57
1.29
1
3.3, 9.4
10
0.99
0.9
4.7
10.3
)
0.9
0.82b
0.82
0.74
0.72b
0.42
9.6
)
8.6, 11.3
9.3
a All values are in V vs NHE, pH 7. b The pH adjusted E0 values for 1- and
2-naphthol were calculated using Eo values found at pH 13 from
reference 58 that were corrected for pH 7 utilizing the method outlined
in reference.60 Because the 2-naphthoxyl radical is more resonance
stabilized compared to phenol, it is possible that the calculated E0 values
are an overestimate.
Pathway C. Another possibility is the addition of dOG• to the
2-NpOꢀ ring system yielding the radical anion intermediate 8
that goes through a second oxidation and enolizes to the final
stable products 1a/1b (Scheme 4). Once again, the N2 centered
dOG• aminyl radical is not calculated to provide the largest radical
population,56 but sterics could be driving arylation at the exocyclic
amine. At first it was thought that radical addition to the 2-NpOꢀ
would not be specific for C1, but studies have shown that the
photoinduced phenylation of 2-naphthol occurs specifically at C1
through a possible SRN1-like mechanism.83 In this study it was
concluded thatsterics of the arylating group and the dominance of
the C1-centered 2-NpOꢀ provide the site specificity,83 a feature
that would also be consistent with the proposed structure of
compounds 1a/1b.
The reaction between dG and 2-naphthol in the presence of
Na2IrCl6 does not yield any detectable amount of adducts.
Instead the 2-naphthol is preferentially oxidized yielding poly-
mers that rapidly precipitate. This observation suggests that the
mechanism has to involve at least a one-electron oxidation of
dOG, ruling out any possibility of a mechanism not invoking
dOG oxidation and supporting all three proposed mechanisms.
On the basis of the observation that reactions with tyrosine and p-
cresol, which have higher reduction potentials (Table 1) and did
not yield significant amounts of an analogous N2 product,50 it is
concluded that the possible mechanism should involve both
oxidation of the dOG and 2-naphthol, supporting Pathway B.
Possible Reaction Pathways Leading to 2a/2b from dOG.
Compounds 2a/2b are the products observed from the four-
electron oxidation product of dG in the presence of 2-naphthol;
the focus here will be on the events thatmust occur to arrive at 2a/
2b. After the initial two-electron oxidation of the nucleoside dG to
dOG, the species present include dOG, 2-naphthol, and oxidant
(where the oxidant can be CuII/H2O2 or Na2IrCl6). Both dOG
and 2-naphthol have similar reduction potentials (Table 1); thus,
oxidation can be envisioned to occur from either or both
species.19,30,58,60,84,85 From the structure of 2a/2b, there exist
two possible pathways for product formation to be initiated.
Pathway D. dOG and 2-naphthol each undergo one-electron
oxidation to dOG• and 2-NpO•. These radicals can couple
yields are due to increased background oxidation leading to many
unidentified products according to HPLC analysis. Regardless of
the oxidant, dG can undergo a two-electron oxidation to dOG by
these metal oxidant systems,1,63,66,67 and further studies have
shown that DNA damage by these complexes is enhanced in the
presence of phenolic compounds.68,69 Oxidation reactions in-
volving dG and 2-naphthol initiated by a weaker oxidant
(Na2IrCl6) that is incapable of oxidizing dG under these condi-
tions did not yield any detectable products, again suggesting
that dOG is the reactive nucleoside involved in the arylation
chemistry.
Interestingly, 2-naphthol in the presence of CuII does not
precipitate out of solution,70,71 whereas when reactions were
conducted with other phenols, such as 1-naphthol and p-cresol,
under similar conditions, arylation products were not observed in
high yields due to precipitation of the corresponding CuII
phenolate complexes.71 This observation suggests a unique
feature associated with 2-naphthols and the ability to arylate dG
under oxidative conditions.
Possible Reaction Pathways Leading to Compounds 1a/1b
from dOG. The mechanistic pathway to yield compounds 1a/1b
is proposed to occur initially through the two-electron oxidation
of dG to the reactive intermediate dOG, as previously stated. At
this point the reaction can proceed through three possible
mechanistic pathways yielding 1a/1b, because of the similarity
in reduction potential values for dOG and 2-naphthol (Table 1).
Pathway A. In this mechanism, dOG proceeds through a two-
electron oxidation to the proposed quinonoid imine dOGox
(shown as the exocyclic amine tautomer); this species could
tautomerize to an exocyclic imine form of dOGox that is attacked
by the 2-naphtholate anion (2-NpOꢀ) at N2 yielding a neutral
intermediate 7 that enolizes to the stable products 1a/1b
(Scheme 4). This mechanism does not explain why the N2
adduct is phenol specific, as will be discussed later, unless the
point of arylation has some degree of control determined by
steric properties.50 Also, there is no literature precedent for
N-alkylation of iminoquinones; the preferred site of nucleophilic
attack is carbon centered that would yield compounds 2a/2b
(see below).72ꢀ81
7958
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963

The Journal of Organic Chemistry
ARTICLE
Scheme 4. Possible Pathways to Form Compound 1a/1b from dOG
yielding an intermediate species 9 with a covalent bond formed
between C1 of 2-naphthol and C5 of dOG (Scheme 5).
Scheme 5. Possible Pathways to Form 2a/2b from dOG
Pathway E. In this pathway, dOG undergoes a two-electron
oxidation to the electrophilic quininoid species dOGox, drawn as
the exocyclic amine tautomer, or equally likely, the exocyclic
imino tautomer, that is attacked at C5 by the C1 centered 2-
NpOꢀ, again giving the proposed intermediate 9 with a newly
formed covalent bond between C5 of dOG and C1 of 2-naphthol
(Scheme 5). Previous studies have shown that the preferred site
for the 2-NpOꢀ to alkylate is at C1 in aqueous solution86 that is
further supported by the spectroscopic data under both possible
mechanisms. The analogous intermediate from H2O attack, 5-
HO-dOG, is short-lived and undergoes further rearrangement/
decomposition to the hydantions Sp or Gh, or a second nucleo-
philic attack by H2O can occur at C4 followed by decomposition
to HICA in which all of these compounds are products under
different reaction conditions.33,35,46 Intermediate 9 has the phe-
nol positioned such that the oxygen centered 2-NpOꢀ nucleo-
phile can participate in an intramolecular cyclization at C4
yielding the tricyclic [4.3.3.0] products 2a/2b (Scheme 5).
The two diastereomers of 2a/2b can be rationalized through
geometric considerations in which the 4S,5S (2a) diastereomer is
formed when 2-naphthol is positioned on the si face and attacks
C4, and the 4R,5R (2b) diastereomer arises when 2-naphthol
attacks C4 from the re face. This mechanism is similar to that of
the two-electron oxidation product of dOG to HICA, but after
the second nucleophilic attack at C4, the HICA-like structure is
not observed because carbonyl formation is abolished at C4 in
2a/2b preventing further degradation. At this time, it is impos-
sible to determine the most likely pathway for formation of 2a/
2b, but Pathway E best explains dOG oxidation in the presence
of H2O,36,43 lysine,49 spermine,48 and peroxynitrite.87
oxidant system that was thought to initiate the oxidation reaction
on the phenol moiety and not on dOG.12,16 These reactions were
conducted with the tri-O-acetyl-OG nucleoside so that Sp, a
dOG oxidation product, could be retained on a reversed-phase
HPLC column allowing a complete mass balance of the reaction
in one HPLC run. Control studies with tri-O-acetyl-OG, HRP, anda
10-fold excess of H2O2 led to <5% oxidation of the nucleoside. The
HRP/H2O2 oxidant system yields of 1a/1b, 2a/2b, and Sp with
respect to pH (6.4, 7.4, and 8.4) were determined and compared to
analogous reactions with Na2IrCl6 as the oxidant (Figure 6).
When HRP/H2O2 (2.5 mM tri-O-acetyl-OG, 0.5 mM
2-naphthol, 2.5 U HRP, and 0.5 mM H2O2) was utilized
to initiate the reaction, regardless of pH, the major product
Effect of Oxidant on the Product Distribution. To better
understand the oxidation mechanisms leading to bond formation
between the reactive dOG nucleoside and 2-naphthol, reactions
were conducted with the Horse Radish Peroxidase (HRP)/H2O2
7959
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963

The Journal of Organic Chemistry
ARTICLE
Figure 6. Relative product distributions observed when tri-O-acetyl-OG
was allowed to react with 2-naphthol in the presence of either Na2IrCl6 or
HRP/H2O2. Each reaction was run at 2.5 mM tri-O-acetyl-OG, 0.5 mM
2-naphthol, and either 2.5 U HRP and 0.5 mM H2O2, or 0.5 mM
Na2IrCl6. Absolute product yields were ∼10%; each trial was run in
triplicate, giving an average error of 3% (error = one standard deviation).
Figure 7. Effect of the phenol on the relative-product distributions while
utilizing the HRP/H2O2 oxidant system. Each reaction was conducted
with 2.5 mM tri-O-acetyl-OG, 0.5 mM 2-naphthol, and either 2.5 U HRP
and 0.5 mM H2O2, or 0.5 mM Na2IrCl6. Absolute product yields were
∼10%; each trial was conducted in triplicate, giving an average error of 3%
(error = one standard deviation).
observed was Sp; if Na2IrCl6 (2.5 mM tri-O-acetyl-OG, 0.5 mM
2-naphthol, and 0.5 mM Na2IrCl6) initiated the oxidation, then
compounds 1a/1b and 2a/2b were the major products. The
HRP/H2O2 oxidant system appears to initiate the reaction
through formation of a 2-NpO• that is capable of oxidizing
dOG, based on the appearance of Sp. Interestingly, the amount
of 2-naphthol adducts did not increase, in fact they decreased,
suggesting that when the flux of 2-NpO• is high the radicals
couple, outcompeting adduct formation that is supported by
increased amounts of 2-naphthol dimers and tetramers observed
by HPLC. Another possibility that explains the increased amount
of Sp when HRP/H2O2 was used is the formation of O2•ꢀ from
Effect of the Phenol on the Product Distribution. In a
previous study conducted in our laboratory the 4,5-PhO-dOG
adduct that is similar in structure to 2a/2b was characterized from
the oxidation of dOG in the presence of either tyrosine or p-cresol
using the one-electron oxidant K3Fe(CN)6.50 In this study, the
phenol was substituted with 2-naphthol, and the product dis-
tribution included compounds 1a/1b that were not previously
observed for simple phenols. The goals of the following studies
were to determine if this new type of adduct (1a/1b) observed
with 2-naphthol was a result of 2-naphthol having a lower
standard reduction potential than tyrosine or p-cresol, or if it is
the naphthalene ring system that is influencing the product
distribution (Figure 7). The HRP/H2O2 oxidant system was
chosen to conduct these studies. When pentachlorophenol
(C6Cl5OH), a phenol with a high standard reduction potential
(Table 1) but resistant to carbon alkylation, was oxidized in the
presence of tri-O-acetyl-OG, the only product observed was Sp,
providing more evidence that phenoxyl radicals can oxidize the
OG base. When phenol or p-cresolwas oxidized by HRP/H2O2 in
the presence of tri-O-acetyl-OG, adduct formation was observed
but the major product was still Sp (Figure 7). In the studies with
the phenols that are not blocked at the ortho carbon (phenol, p-
cresol, 2-naphthol, and 1-naphthol) tricyclic adducts were ob-
served. The 4,5-PhO-dOG adduct is a defining product of phenol
arylation of the OG base observed through the course of an
oxidation reaction. The N2 adducts were only observed in the case
of the naphthalene-based phenols, 1- and 2-naphthol, which is a
defining product observed when these phenols arylate the OG
base under oxidative conditions. The key difference between the
simple phenols, phenol and p-cresol, and the naphthalene-based
phenols, 1- and 2-naphthol, is the stability of the intermediate
phenolic radical, that is, the naphthoxyl radical is more resonance
stabilized than the phenoxyl radicals.58 The appearance of the N2
adducts with naphthalene-based phenols supports a phenolic-
radical intermediate as proposed in Pathway B of Scheme 4.
Comparison of phenol reduction potentials to that of dOG
suggests that phenoxyl radicals are produced with HRP/H2O2
in the presence of dOG, the purine base acts as a reductant that
quenches the radical species, while being oxidized to Sp.
•-
HRP. Because O2 reacts with phenoxyl radicals at diffusion
controlled rates,88 2-naphthol cannot participate in adduct for-
mation. When a nonselective oxidant, such as Na2IrCl6 was used,
adduct formation increased because the overall flux of 2-NpO•
decreased due to oxidation of the dOG base. This result suggests
that 2-naphthol adducts to dOG require that dOG be activated
through at least a one-electron oxidation to yield 2-naphthol
adducts, therefore ruling out any adduct formation mechanism
that has dOG acting strictly as a nucleophile. This observation
explains why adducts were never observed to dG without con-
comitant oxidation at C8 to dOG, since 2-naphthol oxidation will
always occur in preference to dG oxidation due to its ∼500 mV
lower standard reduction potential (Table 1).
The next sets of studies were aimed at evaluating the nature of
the oxidant with respect to product formation. The weaker one-
electron oxidants Na2IrBr6 and K3Fe(CN)6 were utilized to
initiate the reaction between dOG and 2-naphthol. In these
studies it was found that product yields and pH dependence were
the same for these oxidants as was observed with the stronger
one-electron oxidant, Na2IrCl6 (Supporting Information). Since
dOG and 2-naphthol have similar redox potentials, the non-
selective oxidants will likely react with both reactants similarly,
providing the same overall relative amounts of radical species.
Effect of pH on the Product Distribution. The reaction
product yields were shown to be pH dependent, but the nature
of the oxidant did not have much of an effect on the product
distributions (Supporting Information). This observation is com-
plicated by the fact that dOG and 2-naphthol are both weak acids
and show the same pH dependency in their reduction
potentials.30,58 The yield of 1a/1b increases with increasing pH
suggesting that 2-naphthol oxidation dominates at higher pH and
at lower pH 2a/2b yields increase, and accordingly, dOG oxida-
tion dominates at lower pH.
Product Stability Studies. The stabilities of 1a/1b and 2a/2b
in aqueous buffer (75 mM NaPi, pH 7.4, 22 °C) are dramatically
different. Under these conditions, 1a/1b was unaltered after 14
days. This supports 1a/1b as an adduct to N2 because adducts to
the exocyclic amine are considered to be relatively stable.1,89
7960
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963

The Journal of Organic Chemistry
ARTICLE
In an attempt to decompose 1a/1b, it was found that this
compound is susceptible to further oxidation by Na2IrCl6,
yielding an early eluting product and many later eluting products
in the HPLC (Supporting Information). The early eluting
product can arise from two possible oxidation events. The
nucleobase of 1a/1b can still undergo a two-electron oxidation
to the hydantoins Sp or Gh in H2O. Alternatively, the 2-naphthol
of 1a/1b is positioned to undergo a two-electron oxidation to an
iminoquinone. The iminoquinone is electrophilic and suscepti-
ble to nucleophilic attack, in which the nucleophile can be water
or an intramolecular cyclization reaction can occur if N1 of the
base serves as the nucleophile. This has been observed with p-
quinones covalently linked to N2 of dG.5,10,12 This new product
gave a mass of 435 (1a/1b + 9). Attempts to further study this
product were not possible due to the instability of this new
compound. The later eluting peaks in the HPLC are most likely
polymers since oxidation of the phenol portion of 1a/1b yields a
radical that can polymerize. These polymers are likely composed
of oxidized nucleobases and multiple cross-links between the
2-naphthols, so attempts to characterize these structures where
not pursued. This phenomenon has recently been observed for
phenols covalently linked to C8 of dG.90
unique structures in which both an oxidation and arylation event
had occurred in tandem. Compounds 1a/1b are similar to other
products previously observed from the arylation of dG because N2
is the site most preferred by resonance-stabilized electrophiles;1
however, the base has also undergone a two-electron oxidation by
CuII/H2O2 yielding dOG. Compounds 2a/2b, also base oxidation
products of dG, are tricyclic compounds that result from the
ambidentate nature of 2-naphthol, causing more disruption in the
base structure. Both 1a/1b and 2a/2b have the same masses but
1
differ in their RP-HPLC, H NMR, and ESI+-MS/MS profiles,
providing the means to uniquely identify each. In the cellular
context, 1a/1b is expected to be the dominant arylation product of
guanine residues under these conditions due to steric considera-
tions. In either case, if 1a/1b or 2a/2b are present in the cell, the
consequences would be deleterious, because both would most
likely represent polymerase stop points in replication or transcrip-
tion. The two unique constitutional isomer adducts 1a/1b and 2a/
2b observed under oxidative conditions show a stark contrast to
the previously identified compounds Sp, Gh, and HICA for which
water serves as the nucleophile or when lysine (5-Lys-Sp),
spermine (5-Spm-OG), and peroxynitrite (DGh) form adducts
to dG and dOG.
Compounds 1a/1b elute as two peaks in the HPLC that we
1
assign as atropisomers. In the H NMR spectrum, only one
’ EXPERIMENTAL SECTION
anomeric proton was observed indicating that the two isomers
are very similar. To determine if the two atropisomers can
interconvert, the compounds were purified from one another
by HPLC, and then heated at 75 °C for 12 h. The chromato-
graphically enriched samples did not change during this time
period; this result was not too surprising because some atropi-
somers can have interconversion temperatures up to 200 °C.91
These studies were conducted with nucleoside adducts that
retain the 2-deoxyribose moiety. The effect of this group on
the interconversion of atropisomers cannot easily be determined.
Furthermore, it is anticipated that the free base adduct would not
be sufficiently water-soluble enough to allow its study.
Compounds 2a/2b were prone to decomposition under the
same conditions (75 mM NaPi, pH 7.4, 22 °C). By following the
loss of starting material, we observed that the two diastereomers
gave different stabilities, where one of the diastereomers has a
t1/2 ≈ 49 h, and the other was ∼69 h (Supporting Information).
Also, 2a/2b showed greater stability at lower pH (pH 6.4
the t1/2 increased ∼46 h) than at higher pH (pH 8.4 the
t1/2 decreased ∼37 h). RP-HPLC analysis of the decomposition
of 2a/2b yielded many peaks, and LC-ESI+-MS of this sample
gave products with masses of 462 (2a/2b + 36), 444 (2a/2b +
18), 416 (2a/2b - 10), 399 (2a/2b ꢀ 27), and 323 (2a/2b ꢀ
103). Due to low product yields and limited stability, further
characterization of these products was not pursued.
Oxidation of dG and 2-Naphthol in the Presence of a
Metal Catalyst and H2O2. A 200-μL solution of dG (3.0 mM,
0.60 μmoles, 0.16 mg) and 2-naphthol (1.0 mM, 0.20 μmoles, 0.03 mg)
was stirred in sodium phosphate buffer (75 mM, pH 7.4). The reaction
was initiated by the simultaneous addition of copper(II) acetate
(2.0 mM, 0.40 μmoles, 0.07 mg) and H2O2 (10.0 mM, 2.0 μmoles,
0.07 mg). The solution was kept at 22 °C for 24 h and then quenched
with Na2EDTA (10.0 mM, 2.0 μmoles, 0.50 mg). Alternatively, a 200-μL
solution of dG (3.0 mM, 0.60 μmoles, 0.16 mg) and 2-naphthol
(1.0 mM, 0.20 μmoles, 0.03 mg) was mixed with FeII/EDTA
(0.50 mM, 0.10 μmoles, 0.04 mg) H2O2 (1.0 mM, 0.2 μmoles, 0.05
mg), and ascorbate (1.0 mM, 0.20 μmoles, 0.04 mg) and incubated for
1 h. The FeII/EDTA complex was made fresh by mixing Fe(NH4)2(SO4)2
and Na2EDTA in a 1:2 ratio 30 min prior to reaction. The reaction
mixtures were analyzed by HPLC (see Supporting Information) giving
two product peaks. 1a/1b. HRMS (ESI+-TOF) calcd mass for
C20H19N5O6Na [M + Na+] = 448.1233, exp mass = 448.1241. ESI+-
MS/MS for the free base [(M + H)+ = 310] = 166. tR = 18.2 min, and 2a/
2b. HRMS (ESI+-TOF) calcd mass for C20H19N5O6Na [M + Na+] =
448.1233, exp mass = 448.1240. ESI+-MS/MSforthefreebase[(M+H)+ =
310] = 293, 267, 249, 225, 182, 166, and 86. tR = 21.7 and 22.3 min.
Oxidation of dOG and Base Analogs with 2-Naphthol in
the Presence of Na2IrCl6. A 200-μL solution of dOG (3.0 mM,
0.60 μmoles, 0.17 mg) and 2-naphthol (10 mM, 2.0 μmoles, 0.30 mg) was
mixed in sodium phosphate buffer (75.0 mM, pH 6.4, 7.4, or 8.4). The
reaction was initiated by the addition of Na2IrCl6 (6.0 mM, 1.0 μmoles,
0.67 mg). The solution was kept at 22 °C for 30 min and then quenched
with Na2EDTA (30 mM, 6 μmoles, 1.5 mg). The reaction mixture was
purified by HPLC and then concentrated to furnish 1a/1b (5.9 mg, 15%
yield) tR = 18.2 min as a white solid (mp =194.2ꢀ196.5, decomposed),
and 2a/2b (3.8 mg, 10% yield) tR = 21.7 and 22.3 as a white solid (mp =
170.8ꢀ175.4, decomposed). 1a/1b 1H NMR (500 MHz, DMSO-d6) δ
11.01 (b, 1 H), 8.47 (b, 1 H), 7.84 (d, J = 7.74 Hz, 1 H), 7.80 (d, J = 9.69
Hz, 1 H), 7.38 (t, J = 6.34 Hz, 1 H), 7.30 (t, J = 7.19 Hz, 1 H), 7.25 (d, J =
7.74 Hz, 1 H), 7.23 (d, J = 6.34 Hz, 1 H), 6.90 (b, 1 H), 6.18 (dd, J = 7.16
and 6.63 Hz, 1 H), 5.45 (s, 1 H), 5.19 (b, 2 H), 4.35 (ddd, J = 5.94, 3.31,
and 2.93 Hz, 1 H), 3.80 (td, J = 4.57 Hz, 1 H), 3.58 (m, J = 4.57 Hz, 1 H),
3.01 (ddd, J = 6.46 Hz, 1 H), 2.00 (ddd, J = 6.46 Hz, 1 H). 13C NMR (125
’ CONCLUSIONS
Genomic DNA interacts with many potential reactive species
in the cell. Chelated CuII is observed to be present in micro-
molar concentrations,92 H2O2 is a metabolic byproduct,93 and
2-naphthol has been proposed to reach micromolar concentra-
tions in cells that are exposed to high levels of this phenol or its
precursor naphthalene.94 20-Deoxyguanosine provides a platform
for unique chemical transformations to occur in comparison to the
other bases since this nucleobase is particularly electron rich. dG
and 2-naphthol were shown to react under oxidative conditions to
yield stereoisomeric pairs of two major constitutional isomers with
7961
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963

The Journal of Organic Chemistry
ARTICLE
MHz, DMSO-d6) δ 159.7, 155.5, 152.8 (C200), 150.5 (C8), 147.0 (C4),
132.7, 132.6, 129.2 (C400), 127.8 (C500), 126.7 (C700), 122.6 (C600), 121.3
(C800), 119.0 (C300), 114.6 (C100), 100.8, 87.6 (C40), 81.8 (C10), 71.6
(C30), 62.6 (C50), 36.3 (C20). UVꢀvis: 220 nm (6.8 ꢁ 104 Mꢀ1 cmꢀ1),
226 nm (7.5 ꢁ 104 Mꢀ1 cmꢀ1), 297 nm (2.5 ꢁ 103 Mꢀ1 cmꢀ1). HRMS
(ESI+-TOF): [M + Na]+ calcd for C20H19N5O6Na, 448.1233, found
448.1229; ESI+-MS/MS for the free base [(M + H)+ = 310] = 166. 2a/2b
1H NMR (500 MHz, DMSO-d6) δ 8.97 (s, 1 H), 8.59 (d, J = 8.21 Hz,
1H), 7.89 (s, 1 H), 7.86 (d, J = 8.03 Hz, 1 H), 7.49 (t, J = 8.14 Hz, 1 H),
7.35 (t, 1 H), 7.15 (d, 1 H), 6.64 (b, 2 H), 5.75 (dd, J = 6.33 and 6.01 Hz, 1
H), 5.65 (dd, J = 8.07 and 6.03 Hz, 5.39 (s, 1 H), 5.02(b, 1 H), 4.24 (ddd,
J = 5.70, 5.43, and 4.50 Hz, 1 H), 3.70 (td, J = 3.81 Hz, 1 H), 3.51 (m, 2
H), 2.81 (ddd, J = 6.05 Hz, 1 H), 1.85 (ddd, 1 H). 13C NMR (125 MHz,
DMSO-d6) δ 159.7, 156.2 (C8), 155.9 (C200), 155.7, 132.6 (C400), 130.0
(C900), 129.5 (C1000), 128.6 (C500), 127.2 (C700), 124.4 (C800), 123.6
(C600), 118.1 (C100), 112.2 (C300), 86.4 (C40), 82.5 (C10), 71.5 (C30),
63.1, 62.5 (C50), 36.8 (C20). UVꢀvis: 220 nm (3.9 ꢁ 103 Mꢀ1 cmꢀ1),
227 nm (4.6 ꢁ 103 Mꢀ1 cmꢀ1) HRMS (ESI+-TOF) [M + Na]+ calcd for
C20H19N5O6Na, 448.1233, found 448.1229; ESI+-MS/MS for the free
base [(M + H)+ = 310] = 293, 267, 249, 225, 182, 166, and 86. 3a/3b
HRMS (ESI+-TOF) [M + Na]+ calcd for C21H21N5O6Na 462.1390,
found 462.1360; ESI+-MS/MS for the free base [(M + H)+ = 324] = 180.
4a/4b HRMS (ESI+-TOF) [M + Na]+ calcd for C21H21N5O6Na
462.1390, found 462.1391; ESI+-MS/MS for the free base [(M + H)+
= 324] = 307, 267, 225, 182, 100. 5a/5b HRMS (ESI+-TOF) [M + Na]+
calcd for C24H27N5O6Na 504.1859, found 504.1863; ESI+-MS/MS for
the free base [(M + H+)+ = 366] = 323, 222, 182, 142.
Oxidation with HRP/H2O2. Literature protocols were followed
to synthesize tri-O-acetyl-OG.95 A 200-μL solution of tri-O-acetyl-
OG (3 mM, 0.60 μmoles, 0.17 mg) and each individual phenol
(0.5ꢀ5 mM, 0.10ꢀ0.50 μmoles, 0.014ꢀ0.072 mg) was mixed in sodium
phosphate buffer (75 mM, pH 6.4, 7.4, 8.4) and H2O2 (0.5ꢀ5 mM,
0.10ꢀ0.50 μmoles, 3ꢀ14 μg). The reaction was initiated by the addition
of HRP (2.5 U) and kept at 22 °C for 60 min. The reaction mixture was
analyzed by HPLC.
Synthesis of N1-methyl-dOG. To a solution of dOG (50 mg,
177 μmoles) in 1 mL of DMSO at 22 °C was added K2CO3 (2.0 mg,
14 μmoles), followed by addition of four mole equivalents of CH3I
(0.1 g, 708 μmoles), and the mixture was stirred for 48 h. The reaction
afforded N1-methyl-dOG as a white solid (mp = 225.4ꢀ227.0, decom-
posed), 10 mg (34 μmoles, 20% yield). 1H NMR (DMSO-d6) δ 10.60
(b, 1 H), 7.05 (b, 2 H), 6.02 (dd, J = 7.03 Hz, 1 H), 5.11 (b, 1 H), 4.71 (t,
J = 5.08 Hz, 1H), 4.30 (s, 1 H), 3.70 (s, 1 H), 3.54 (m, J = 4.89 Hz, 1H),
3.29 (s, 3 H), 3.23 (ddd, J = 4.30 Hz, 1 H), 2.96 (ddd, J = 7.81 Hz, 1 H),
1.89 (ddd, J = 6.45 Hz, 1 H). 13C NMR (125 MHz, DMSO-d6) δ 153.6,
151.7, 150.9, 145.5, 97.9, 87.2, 80.9, 71.3, 62.3, 28.1. HRMS (ESI+-TOF)
[M + H]+ calcd for C11H16N5O5 298.1151, found 298.1162.
2a/2b, 3a/3b, 4a/4b, and 5a/5b, N1-methyl-dG, and N2-
diethyl-dOG. HPLC and MS data for the decomposition of
1a/1b and 2a/2b and the pH dependence of the decomposition
of 2a/2b. This material is available free of charge via the Internet
at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: (801) 585-7290. Fax: (801) 585-0024. E-mail: burrows@
chem.utah.edu.
’ ACKNOWLEDGMENT
This work was supported by a grant by the NSF 0809483. We
are grateful for the instrumental and technical assistance pro-
vided by Dr. Charles Mayne, Dr. Xiaoyun Xu, and Dr. Yu Ye.
’ REFERENCES
(1) Burrows, C. J.; Muller, J. G. Chem. Rev. 1998, 98, 1109–1152.
(2) Bolton, J. L.; Trush, M. A.; Penning, T. M.; Dryhurst, G.; Monks,
T. J. Chem. Res. Toxicol. 2000, 13, 135–160.
(3) Bolton, J. L. Toxicology 2002, 177, 55–56.
(4) Nutter, L. M.; Wu, Y. Y.; Ngo, E. O.; Sierra, E. E.; Gutierrez, P. L.;
Abul-Hajj, Y. J. Chem. Res. Toxicol. 1994, 7, 23–28.
(5) Shen, L.; Qiu, S.; van Breemen, R. B.; Zhang, F.; Chen, Y.;
Bolton, J. L. J. Am. Chem. Soc. 1997, 119, 11126–11127.
(6) Akanni, A.; Abul-Hajj, Y. J. Chem. Res. Toxicol. 1999, 12, 1247–
1253.
(7) Stack, D. E.; Byun, J.; Gross, M. L.; Rogan, E. G.; Cavalieri, E. L.
Chem. Res. Toxicol. 1996, 9, 851–859.
(8) Veldhuyzen, W. F.; Lam, Y. F.; Rokita, S. E. Chem. Res. Toxicol.
2001, 14, 1345–1351.
(9) Pande, P.; Shearer, J.; Yang, J.; Greenberg, W. A.; Rokita, S. E.
J. Am. Chem. Soc. 1999, 121, 6773–6779.
(10) Nguyen, T. N. T.; Bertagnolli, A. D.; Villalta, P. W.; Buhlmann,
P.; Sturla, S. J. Chem. Res. Toxicol. 2005, 18, 1770–1776.
(11) Vaidyanathan, V. G.; Villalta, P. W.; Sturla, S. J. Chem. Res.
Toxicol. 2007, 20, 913–919.
(12) Dai, J.; Sloat, A. L.; Wright, M. W.; Manderville, R. A. Chem. Res.
Toxicol. 2005, 18, 771–779.
(13) Jowa, L. W., G.; Snyder, R.; Winkle, S.; Kalf, G. F. J. Appl.
Toxicol. 1990, 10, 47–54.
(14) Penning, T. M.; Burczynski, M. E.; Hung, C. F.; McCoull, K. D.;
Palackal, N. T.; Tsuruda, L. S. Chem. Res. Toxicol. 1999, 12, 1–18.
(15) McCoull, K. D.; Rindgen, D.; Blair, I. A.; Penning, T. M. Chem.
Res. Toxicol. 1999, 12, 237–246.
(16) Dai, J.; Wright, M. W.; Manderville, R. A. Chem. Res. Toxicol.
2003, 16, 817–821.
(17) Stemmler, A. J.; Burrows, C. J. J. Am. Chem. Soc. 1999, 121,
Synthesis of N2-diethyl-dOG. To a solution of dOG (15 mg,
53 μmoles) in 40:60 MeOH/H2O was added 80 μL acetaldehyde
(63.1 mg, 1.4 mmol) and NaBH3CN (35 mg, 558 μmoles). The resulting
mixture was refluxed for 48 h. The reaction afforded N2-diethyl-dOG as a
white solid (mp = 182.3ꢀ185.1, decomposed), 5 mg (15 μmoles, 33%
yield). 1H NMR (500 MHz, DMSO-d6) δ 10.62 (b, 2 H), 6.02 (dd, J =
7.81 Hz, 1 H), 5.39 (s, 1 H), 5.14 (d, J = 3.12 Hz, 1 H), 4.72 (t, J = 4.69 Hz,
1 H), 4.32 (d, J = 3.12 Hz, 1 H), 3.70 (q, J = 4.68 Hz, 1 H), 3.52 (q, J = 7.81
Hz, 1 H), 3.04 (dt, J = 6.25 Hz, 4 H), 1.94 (m, J = 3.12 Hz, 1 H), 1.09 (t, J =
6.25 Hz, 6 H). 13C NMR (125 MHz, DMSO-d6) δ 166.62, 152.55, 146.38,
87.52, 84.75, 81.22, 71.77, 63.08, 36.19, 42.53, 13.67. HRMS (ESI+-TOF)
[M + H]+ calcd for C14H21N5O5Na 362.1440, found 362.1418.
6956–6957.
(18) Muller, J. G.; Kayser, L. A.; Paikoff, S. J.; Duarte, V.; Tang, N.;
Perez, R. J.; Rokita, S. E.; Burrows, C. J. Coord. Chem. Rev. 1999,
185ꢀ186, 761–774.
(19) Steenken, S.; Jovanovic, S. V. J. Am. Chem. Soc. 1997, 119,
617–618.
(20) Pratviel, G.; Meunier, B. Chem.—Eur. J. 2006, 12, 6018–6030.
(21) Matter, B.; Malejka-Giganti, D.; Csallany, A. S.; Tretyakova, N.
Nucleic Acids Res. 2006, 34, 5449–5460.
(22) Fleming, A. M.; Muller, J. G.; Ji, I.; Burrows, C. J. Org. Biomol.
Chem. 2011, 9, 3338–3348.
(23) Ghude, P.; Schallenberger, M. A.; Fleming, A. M.; Muller, J. G.;
Burrows, C. J. Inorg. Chim. Acta 2011, 369, 240–246.
(24) Ye, W.; Sangaiah, R.; Degen, D. E.; Gold, A.; Jayaraj, K.;
Koshlap, K. M.; Boysen, G.; Williams, J.; Tomer, K. B.; Mocanu, V.;
’ ASSOCIATED CONTENT
Supporting Information. All UVꢀvis, HPLC, ESI+-MS,
S
b
ESI+-MS/MS, NMR spectral data for compounds 1a/1b and
7962
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963

The Journal of Organic Chemistry
ARTICLE
Dicheva, N.; Parker, C. E.; Schaaper, R. M.; Ball, L. M. J. Am. Chem. Soc.
2009, 131, 6114–6123.
(58) Das, T. N.; Neta, P. J. Phys. Chem. A 1998, 102, 7081–7085.
(59) Dizdaroglu, M. Biochemistry 1985, 24, 4476–4481.
(25) Neeley, W. L.; Essigmann, J. M. Chem. Res. Toxicol. 2006,
19, 491–505.
(60) Li, C.; Hoffman, M. Z. J. Phys. Chem. B 1999, 103, 6653–6656.
(61) Prutz, W. A. B., J.; Land, E. J. Int. J. Radiat. Biol. 1990, 58,
(26) Cadet, J.; Douki, T.; Ravanat, J.-L. Free Radic. Biol. Med. 2010,
49, 9–21.
(27) Shigenaga, M. K.; Gimeno, C. J.; Ames, B. N. Proc. Natl. Acad.
215–234.
(62) Sorokin, V. A.; Valeev, V. A.; Gladchenko, G. O.; Sysa, I. V.;
Blagoi, Y. P.; Volchok, I. V. J. Inorg. Biochem. 1996, 63, 79–97.
(63) Kennedy, L. J.; Moore, K.; Caulfield, J. L.; Tannenbaum, S. R.;
Dedon, P. C. Chem. Res. Toxicol. 1997, 10, 386–392.
(64) Oikawa, S.; Kawanishi, S. Biochemistry 1996, 35, 4584–4590.
(65) Pogozelski, W. K.; Tullius, T. D. Chem. Rev. 1998, 98, 1089–
1108.
Sci. U.S.A. 1989, 86, 9697–9701.
(28) Mangal, D.; Vudathala, D.; Park, J.-H.; Lee, S. H.; Penning,
T. M.; Blair, I. A. Chem. Res. Toxicol. 2009, 22, 788–797.
(29) Pouget, J. P. F., S.; Ravanat, J. L.; Testard, I.; Odin, F.; Cadet, J.
Radiat. Res. 2002, 157, 589–595.
(30) Steenken, S.; Jovanovic, S. V.; Bietti, M.; Bernhard, K. J. Am.
(66) Frelon, S.; Douki, T.; Favier, A.; Cadet, J. Chem. Res. Toxicol.
Chem. Soc. 2000, 122, 2373–2374.
2003, 16, 191–197.
(31) Misiaszek, R.; Uvaydov, Y.; Crean, C.; Geacintov, N. E.;
Shafirovich, V. J. Biol. Chem. 2005, 280, 6293–6300.
(32) Steenken, S.; Jovanovic, S. V.; Bietti, M.; Bernhard, K. J. Am.
Chem. Soc. 2000, 122, 2373–2374.
(33) Ye, Y.; Muller, J. G.; Luo, W.; Mayne, C. L.; Shallop, A. J.; Jones,
R. A.; Burrows, C. J. J. Am. Chem. Soc. 2003, 125, 13926–13927.
(34) McCallum, J. E. B.; Kuniyoshi, C. Y.; Foote, C. S. J. Am. Chem.
Soc. 2004, 126, 16777–16782.
(67) Bruskov, V. I.; Malakhova, L. V.; Masalimov, Z. K.; Chernikov,
A. V. Nucleic Acids Res. 2002, 30, 1354–1363.
(68) Li, Y.; Trush, M. A. Cancer Res. 1994, 54, 1895s–1898.
(69) Zimmerman, R. P.; Jia, Z.; Zhu, H.; Vandjelovic, N.; Misra,
H. P.; Wang, J.; Li, Y. Toxicology 2011, 280, 71–76.
(70) Sorokin, V. A.; Valeev, V. A.; Gladchenko, G. O.; Sysa, I. V.;
Blagoi, Y. P.; Volchok, I. V. J. Inorg. Biochem. 1996, 63, 79–91.
(71) Sass, S.; Kaufman, J. J.; Kiernan, J. Anal. Chem. 1957, 29,
143–144.
(35) Luo, W.; Muller, J. G.; Rachlin, E. M.; Burrows, C. J. Org. Lett.
2000, 2, 613–616.
(72) Lee, H. H.; Palmer, B. D.; Denny, W. A. J. Org. Chem. 1988,
(36) Niles, J. C.; Wishnok, J. S.; Tannenbaum, S. R. Chem. Res.
Toxicol. 2004, 17, 1510–1519.
(37) Sugden, K. D.; Campo, C. K.; Martin, B. D. Chem. Res. Toxicol.
2001, 14, 1315–1322.
(38) Adam, W.; Arnold, M. A.; Grune, M.; Nau, W. M.; Pischel, U.;
Saha-Moller, C. R. Org. Lett. 2002, 4, 537–540.
(39) Niles, J. C.; Wishnok, J. S.; Tannenbaum, S. R. Org. Lett. 2001,
3, 963–966.
(40) Joffe, A.; Geacintov, N. E.; Shafirovich, V. Chem. Res. Toxicol.
2003, 16, 1528–1538.
(41) Crean, C.; Geacintov, N. E.; Shafirovich, V. Angew. Chem., Int.
Ed. 2005, 44, 5057–5060.
(42) Luo, W.; Muller, J. G.; Rachlin, E. M.; Burrows, C. J. Chem. Res.
53, 6042–6047.
(73) Powers, J. C.; Oleksyszyn, J.; Narasimhan, S. L.; Kam, C. M.;
Radhakrishnan, R.; Meyer, E. F. Biochemistry 1990, 29, 3108–3118.
(74) Dryhurst, G.; Anne, A.; Wrona, M. Z.; Lemordant, D. J. Am.
Chem. Soc. 1989, 111, 719–726.
(75) Shudo, K.; Orihara, Y.; Ohta, T.; Okamoto, T. J. Am. Chem. Soc.
1981, 103, 943–944.
(76) Corbett, J. F. J. Chem. Soc. B 1969, 213–216.
(77) Tong, L. K. J. J. Phys. Chem. 1954, 58, 1090–1097.
(78) Boger, D. L.; Zarrinmayeh, H. J. Org. Chem. 1990, 55, 1379–1390.
(79) Tao, W.; Barra, M. J. Org. Chem. 2001, 66, 2158–2160.
(80) Badashkeyeva, A. G.; Gall, T. S.; Efimova, E. V.; Knorre, D. G.;
Lebedev, A. V.; Mysina, S. D. FEBS Lett. 1983, 155, 263–275.
(81) Rappoport, Z. The Chemistry of the Quinonoid Compounds; John
Wiley and Sons: New York, 1988; Vol. 2.
Toxicol. 2001, 14, 927–938.
(43) Ye, Y.; Muller, J. G.; Luo, W.; Mayne, C. L.; Shallop, A. J.; Jones,
R. A.; Burrows, C. J. J. Am. Chem. Soc. 2003, 125, 13926–13927.
(44) Duarte, V.; Muller, J.; Burrows, C. Nucleic Acids Res. 1999,
27, 496–502.
(82) Dixon, W. T.; Foster, W. E. J.; Murphy, D. J. Chem. Soc. Perkin II
1973, 2124–2127.
(83) Arguello, J. E.; Penenory, A. B. J. Org. Chem. 2003, 68,
(45) Gremaud, J. N.; Martin, B. D.; Sugden, K. D. Chem. Res. Toxicol.
2010, 23, 379–385.
(46) Niles, J. C.; Wishnok, J. S.; Tannenbaum, S. R. Chem. Res.
Toxicol. 2004, 17, 1501–1509.
2362–2368.
(84) Stanbury, D. M. Adv. Inorg. Chem. 1998, 33, 69–138.
(85) George, P.; Hanania, G. I. H.; Irvine, D. H. J. Chem. Soc.
1957, 3048–3052.
(47) Niles, J. C.; Burney, S.; Singh, S. P.; Wishnok, J. S.; Tannenbaum,
S. R. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 11729–11734.
(48) Hosford, M. E.; Muller, J. G.; Burrows, C. J. J. Am. Chem. Soc.
2004, 126, 9540–9541.
(49) Xu, X.; Muller, J. G.; Ye, Y.; Burrows, C. J. J. Am. Chem. Soc.
2008, 130, 703–709.
(86) Kornblum, N.; Seltzer, R.; Haberfield, P. J. Am. Chem. Soc. 1963,
85, 1148–1154.
(87) Niles, J. C.; Burney, S.; Singh, S. P.; Wishnok, J. S.; Tannenbaum,
S. R. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 11729–11734.
(88) Jonsson, M.; Lind, J.; Reitberger, T.; Eriksen, T. E.; Merenyi, G.
J. Phys. Chem. 1993, 97, 8229–8233.
(50) Xu, X.; Fleming, A. M.; Muller, J. G.; Burrows, C. J. J. Am. Chem.
Soc. 2008, 130, 10080–10081.
(89) Sage, E.; Haseltine, W. A. J. Biol. Chem. 1984, 259, 11098–11102.
(90) Weishar, J. L.; McLaughlin, C. K.; Baker, M.; Gabryelski, W.;
Manderville, R. A. Org. Lett. 2008, 10, 1839–1842.
(91) Cooke, A. S.; Harris, M. M. J. Chem. Soc. 1963, 2365–2373.
(92) Burkitt, M. J.; Lester, P. In Methods in Enzymology; Academic
Press: New York, 1994; Vol. 234, p 66.
(51) Cho, B. P. Magn. Reson. Chem. 1993, 31, 1048–1053.
(52) Didenko, A. L., V.; Trukhin, A.; Petrov, M. Russ. J. Org. Chem.
2007, 43, 1092–1095.
(53) de Silva, S. A. S.; Souto, A. L.; Agra, M. d. F.; da-Cunha, V. L.;
Barbosa-Filho, J. M.; da Silva, M. S.; Braz-Filho, R. ARKIVOC 2004,
54–58.
(93) Gonzꢀalez-Flecha, B.; Demple, B. J. Biol. Chem. 1995, 270,
13681–13687.
(54) Chang, C.-j. A., D.J.; Chern, L.-J.; Gomes, J. D.; Lee, C.-G.;
Mou, P. W.; Narayan, R. Org. Magn. Reson. 1984, 22, 671–675.
(55) Rizkalla, B. H. R., R.K.; Broom, A. D. Biochim. Biophys. Acta
1969, 195, 285–293.
(94) Palmatier, R. N.; Rennard, H. S. I.; Beckmann, J. D. Xenobiotica
1997, 27, 1201–1214.
(95) Nair, V.; Ma, X.; Shu, Q.; Zhang, F.; Uchil, V.; Cherukupalli,
G. R. Nucleosides, Nucleotides Nucleic Acids 2007, 26, 651–654.
(56) Munk, B. H.; Burrows, C. J.; Schlegel, H. B. J. Am. Chem. Soc.
2008, 130, 5245–5256.
(57) Hickerson, R. P.; Prat, F.; Muller, J. G.; Foote, C. S.; Burrows,
C. J. J. Am. Chem. Soc. 1999, 121, 9423–9428.
7963
dx.doi.org/10.1021/jo201423n |J. Org. Chem. 2011, 76, 7953–7963