Influence of chlorine substitution on the hydrolytic stability of biaryl ether nucleoside adducts produced by phenolic toxins

Article abstract of DOI:10.1021/jo401122j

A kinetic study is reported for the acid-catalyzed hydrolysis of oxygen (O)-linked biaryl ether 8-2′-deoxyguanosine (dG) adducts produced by phenolic toxins following metabolism into phenoxyl radical intermediates. Strikingly, the reaction rate of hydrolysis at pH 1 decreases as electron-withdrawing chlorine (Cl) substituents are added to the phenoxyl ring. The Hammett plot for hydrolysis at pH 1 shows a linear negative slope with ρ_X = -0.65, implying that increased Cl-substitution diminishes the rate of hydrolysis by lowering N⁷ basicity. Spectrophotometric titration provided an N⁷H⁺ pK_a value of 1.1 for the unsubstituted adduct 8-phenoxy-dG (Ph-O-dG). Model pyridine compounds suggest N⁷H⁺ pK_a values of 0.92 and 0.37 for 4-Cl-Ph-O-dG and 2,6-dichloro-Ph-O-dG (DCP-O-dG), respectively. Density functional theory (DFT) calculations also highlight the ability of the 8-phenoxy substituent to lower N⁷ basicity and predict a preference for N ³-protonation for highly chlorinated O-linked 8-dG adducts in water. The calculations also provide a rationale for the hydrolytic reactivity of O-linked 8-dG adducts in the gas-phase, as determined using electrospray mass spectrometry (ESI-MS). The inclusion of our data now establishes that the order of hydrolytic reactivity at neutral pH for bulky 8-dG adducts is N-linked > C-linked > O-linked, which correlates with their relative ease of N ⁷-protonation.

Full text of DOI:10.1021/jo401122j

Article
pubs.acs.org/joc
Influence of Chlorine Substitution on the Hydrolytic Stability of
Biaryl Ether Nucleoside Adducts Produced by Phenolic Toxins
Michael S. Kuska,† Mohadeseh Majdi Yazdi,‡ Aaron A. Witham,† Heidi A. Dahlmann,§ Shana J. Sturla,*,§
Stacey D. Wetmore,*,‡ and Richard A. Manderville*,†
†Departments of Chemistry and Toxicology, University of Guelph, Guelph, Ontario, N1G 2W1, Canada
‡Department of Chemistry and Biochemistry, University of Lethbridge, Lethbridge, Alberta, T1K 3M4, Canada
§Institute of Food, Nutrition and Health, ETH Zurich, 8006 Zurich, Switzerland
̈
̈
S
* Supporting Information
ABSTRACT: A kinetic study is reported for the acid-catalyzed hydrolysis of
oxygen (O)-linked biaryl ether 8-2′-deoxyguanosine (dG) adducts produced by
phenolic toxins following metabolism into phenoxyl radical intermediates.
Strikingly, the reaction rate of hydrolysis at pH 1 decreases as electron-
withdrawing chlorine (Cl) substituents are added to the phenoxyl ring. The
Hammett plot for hydrolysis at pH 1 shows a linear negative slope with ρX =
−0.65, implying that increased Cl-substitution diminishes the rate of hydrolysis
by lowering N7 basicity. Spectrophotometric titration provided an N7H+ pKa
value of 1.1 for the unsubstituted adduct 8-phenoxy-dG (Ph-O-dG). Model
pyridine compounds suggest N7H+ pKa values of 0.92 and 0.37 for 4-Cl-Ph-O-dG
and 2,6-dichloro-Ph-O-dG (DCP-O-dG), respectively. Density functional theory
(DFT) calculations also highlight the ability of the 8-phenoxy substituent to
lower N7 basicity and predict a preference for N3-protonation for highly
chlorinated O-linked 8-dG adducts in water. The calculations also provide a
rationale for the hydrolytic reactivity of O-linked 8-dG adducts in the gas-phase, as determined using electrospray mass
spectrometry (ESI-MS). The inclusion of our data now establishes that the order of hydrolytic reactivity at neutral pH for bulky
8-dG adducts is N-linked > C-linked > O-linked, which correlates with their relative ease of N7-protonation.
Covalent modification of dG can also accelerate the rate of
depurination to afford abasic sites. In duplex DNA the N7
position of guanine is the most nucleophilic site and is
particularly susceptible to attack by numerous alkylating
agents.10 The resulting N7-guanine adduct contains a formal
positive charge on the guanine ring, and is a good leaving group
for depurination. Interestingly, the 8-site of dG is also
susceptible to covalent modification by numerous electrophiles.
Depending on the nature of the 8-substituent, the 8-guanine
adduct can be an excellent leaving group for the depurination
process. For example, reactive nitrogen species can generate the
8-NO2-dG adduct that has been detected in the peripheral
lymphocyte of humans exposed to cigarette smoke.11 The 8-
NO2-dG lesion is unstable and depurinates at neutral pH by
releasing 8-NO2-G with a half-life ranging from 20 to 31 h in
oligonucleotides.12 In general, attachment of electron-with-
drawing substituents to the 8-position of dG increases the rate
of depurination through stabilization of the developing negative
charge at N9 during rate-limiting cleavage of the glycosyl
bond.13,14
INTRODUCTION
■
The chemistry of DNA damage is diverse and complex. DNA is
not indefinitely stable, and numerous sources of DNA-
damaging agents of endogenous and exogenous origin can
contribute further to its instability.1 Depurination, the cleavage
of the 1′-N9 glycosidic bond between a purine base and its
deoxyribose sugar, to afford abasic sites is a common type of
damage suffered by genomic DNA.2 Abasic sites are produced
by exposure of DNA to oxidative stress, radiation, anticancer
drugs, and mutagens, and if left unrepaired, abasic sites are
mutagenic and cytotoxic.3
For unmodified 2′-deoxyguanosine (dG), exposure to acid
accelerates the rate of depurination.4 The acid-catalyzed
hydrolysis of dG (Figure 1) is known to proceed via a stepwise
mechanism.5−7 The first step involves protonation at N7,
defined by the acid dissociation constant Ka1; the pKa1 value for
N7H+-dG is 2.34.8 The second step is rate-limiting and involves
unimolecular cleavage of the glycosidic bond, defined by the
rate constant k1, to release protonated guanine as a good
leaving group. Following cleavage, the oxocarbenium ion
undergoes hydration to form the 1′-hydroxylated sugar. In
duplex DNA, the corresponding tautomeric aldehyde has the
potential to generate an interstrand DNA cross-link.9
Received: May 22, 2013
Published: July 2, 2013
© 2013 American Chemical Society
7176
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185

The Journal of Organic Chemistry
Article
Figure 1. (a) Acid-catalyzed hydrolysis of dG. (b) Structures of N-linked, C-linked, and O-linked 8-dG adducts.
Scheme 1. Synthesis of Biaryl Ether Adducts
Bulky aryl ring systems also attach to the 8-site of dG to
generate nitrogen-, carbon-, or oxygen-linked adducts (i.e.,
Figure 1b). The N-linked 8-dG adducts are the most common
lesions produced from reactions of DNA with nitrenium ion
metabolites from arylamine carcinogens.15−17 The N-linked 8-
dG adducts can undergo depurination under mildly acidic to
neutral conditions where dG shows little reactivity.18 In this
case, the 8-arylamino substituent is electron-donating and
increases pKa1 of the N7H+ nucleoside adduct by ∼2 pKa units
compared to unmodified N7H+-dG. Under acidic conditions
(pH < 2), the N-linked 8-dG adducts are only 2- to 5-fold more
reactive than dG, but between pH 3−6 the hydrolysis rates are
accelerated by 40- to 1300-fold.18 The rate increase is ascribed
to the decrease in ionization constant of the N-linked N7H+-dG
7177
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185

The Journal of Organic Chemistry
Article
adduct coupled with the release of steric strain upon removal of
the sugar moiety.
The corresponding C-linked adducts19 are produced by a
range of chemical mutagens that include phenols,20−22
polycyclic aromatic hydrocarbons (PAHs),23 arylhydrazines,24
estrogens25 and nitroaromatics.26 In this case, direct attachment
of the aryl ring to afford the C-linked adduct has little impact on
pKa1.19 However, in 0.1 M HCl (pH ∼ 1) the adducts are 5- to
45-fold more reactive than dG, which increases to 9- to 200-
fold at pH 4.19 Electron-withdrawing substituents at the para-
position of the attached phenyl ring caused the greatest increase
in hydrolysis rate relative to dG. Relief of steric strain upon
removal of the deoxyribose sugar moiety coupled with
stabilization of the developing negative charge at N9 by
electron-withdrawing para-substituents provided a rationale for
the relative depurination efficiencies of C-linked 8-dG
adducts.19
Finally, O-linked 8-dG adducts have been observed following
metabolism of phenols by peroxidase enzymes in vitro.27−29
Increased chlorination of the phenol ring favors O-linked
adduct formation by increasing electrophilicity of the phenolic
radical intermediate and decreasing the rate of bimolecular
phenolic radical coupling.28 Given that attachment of electron-
withdrawing groups to the 8-position of dG can increase rates
of hydrolysis to afford abasic sites,13,14 we were particularly
interested in establishing the hydrolytic stability of O-linked
phenolic 8-dG adducts. Adducts bearing electron-deficient
polychlorinated phenoxyl ring systems were expected to have
enhanced susceptibility to hydrolysis. In this paper we present a
kinetic and computational study of the hydrolysis of O-linked 8-
dG adducts and draw comparison to hydrolysis rates already
established for dG5−7 and the corresponding C- and N-linked
8-dG adducts.18,19
Figure 2. (a) Absorbance spectrum of Ph-O-dG (solid line) and Ph-O-
G (dashed line) in pH 7 MOPS buffer (0.05 M, 0.31 M NaCl) and Ph-
O-G (dotted line) in pH 1 KCl/HCl (0.2 M). (b) Changes in
absorbance spectra of Ph-O-dG in pH 1 buffer (0.2 M KCl/HCl)
recorded in 30 s intervals.
Ph-O-dG in pH 1 buffer. The arrows indicate loss of
absorbance at 250 nm for Ph-O-dG and gain of absorbance
at 275 nm, which correlated with λmax for Ph-O-G in pH 1
buffer (Figure 2a). These spectral changes were consistent with
acid-catalyzed hydrolysis of Ph-O-dG into the deglycosylated
product Ph-O-G, and this was confirmed by reverse-phase
HPLC analysis (Figure S1, Supporting Information).
RESULTS AND DISCUSSION
■
Synthesis of O-Linked 8-dG Adducts. A general method
for the synthesis of O-linked 8-dG adducts has been developed
by Dahlmann and Sturla.30 As outlined in Scheme 1a, the
synthesis involves base-promoted reactions of phenols with the
protected 8-Br-dG analogue 1 in xylenes at 135 °C followed by
TBAF-mediated desilylation of the 3′- and 5′-hydroxyls and
catalytic hydrogenation to remove the O6-benzyl (Bn)
protecting group. O-Linked adducts 4-Me-Ph-OdG, 4-OMe-
Ph-O-dG, 2,4,6-trichlorophenyl (TCP)-O-dG and 2,3,4,5,6-
pentachlorophenyl (PCP)-O-dG were obtained in this manner.
Further optimization of the synthesis was carried out by the
Manderville laboratory (Scheme 1b) and involved replacement
of the O6−Bn protecting group in 1 with the trimethylsilylethyl
group. Thus, with 2 as substrate (prepared according to
literature procedures31) nucleophilic displacement of Br− by
phenolate in xylenes was then followed by TBAF treatment for
removal of all protecting groups in a single step. This procedure
was used for the preparation of Ph-O-dG, 2-Cl-Ph-O-dG and
2,4-dichlorophenyl (DCP)-O-dG.
First order rate constants of hydrolysis (kobs) and half-lives
(t1/2) were initially determined for all O-linked adducts
depicted in Scheme 1 by monitoring in pH 1 buffer (0.2 M
KCl/HCl) at 37 °C the appearance of a peak corresponding to
the deglycosylated base (λmax ∼ 275 nm) as a function of time
(Table 1). All O-linked 8-dG adducts hydrolyzed at a faster rate
than dG. Using Ph-O-dG as the reference point, the derivatives
bearing electron-donating groups (4-Me- and 4-MeO-Ph-O-
dG) showed little deviation from the rate of Ph-O-dG, while
the addition of Cl-substituents retarded the rate of hydrolysis at
pH 1. The addition of one chlorine atom to the phenyl moiety
increased the half-life by ∼30%; upon increased chlorination
the half-lives increased further with PCP-O-dG having a half-life
∼8-fold larger than Ph-O-dG. Of the adducted phenolic
moieties, PCP is the most electron-withdrawing and was
expected to stabilize the developing negative charge on N9
(Figure 1), thus increasing the rate of hydrolysis.13,14,19
Conversely, however, the electronegativity of chlorinated
phenolic moieties would decrease the N7H+ pKa values,
resulting in a lower concentration of protonated adducts at
pH 1.0.13,14
Hydrolysis Kinetics and Activation Parameters. A
spectrophotometric procedure was used to determine rates of
hydrolysis for the O-linked 8-dG adducts, as previously carried
out for the corresponding C-linked analogues.19 At neutral pH
Ph-O-dG has λmax at ∼250 nm with a shoulder peak at ∼284
nm (bold trace, Figure 2a). The deglycosylated nucleobase Ph-
O-G shows greater absorbance at ∼280 nm at pH 7 (dashed
trace) that shifts to λmax at ∼275 nm at pH 1.0 (dotted trace,
Figure 2a). Figure 2b shows changes in absorbance spectra of
Gibbs energy of activation (ΔG‡), enthalpy of activation
(ΔH‡) and entropy of activation (ΔS‡) for the depurination of
the 8-phenoxide-subsitituted nucleosides (Table 1) were
extracted from Eyring plots (Figure S2, Supporting Informa-
7178
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185

The Journal of Organic Chemistry
Article
Table 1. Summary of First-Order Rate Constants (kobs), Half-lives (t1/2), and Activation Parameters for Hydrolysis of O-Linked
8-dG Adducts in pH 1 Buffer
b
c
adduct
Ph-O-dG
kobs (min−1), t1/2 (min)
k/k(dG)
ΔG‡ (kcal/mol)
ΔH‡ (kcal/mol)
ΔS‡ (eu)
1.11 0.04, 0.63
1.178 0.003, 0.59
1.152 0.006, 0.60
0.81 0.05, 0.85
0.871 0.008, 0.80
0.51 0.02, 1.36
0.260 0.008, 2.66
0.145 0.004, 4.77
28
30
29
21
22
13
6.7
3.7
1
18.2
18.1
18.1
18.3
18.3
18.6
19.0
19.5
21.2
19.8 0.5
19.7 0.3
19.8 0.1
20.1 0.1
20.0 0.1
19.7 0.1
20.0 0.1
21.6 0.3
22.5
5.5 1.5
5.4 0.7
5.6 0.3
6.1 0.8
5.5 0.6
3.7 0.2
3.5 0.2
7.1 1.1
4.3
4-Me-Ph-O-dG
4-MeO-Ph-O-dG
4-Cl-Ph-O-dG
2-Cl-Ph-O-dG
DCP-O-dG
TCP-O-dG
PCP-O-dG
a
dG
0.0391, 17.7
a
b
c
Data for dG are taken from ref 6. Determined in pH 1 buffer (0.2 M KCl/HCl) at 37 °C from an average of six kinetic runs. Calculated from ΔG‡
= ΔH‡ − TΔS‡ at 25 °C.
tion) of rate as a function of temperatures (Table S1,
Supporting Information). Activation parameters have been
determined previously for hydrolysis of dG in 0.1 M HCl (pH
∼ 1),6 and are shown in Table 1 for comparison. The ΔG‡
values for O-linked 8-dG adducts are all lower than for dG
(ΔG‡ = 21.2 kcal/mol), as the most stable adduct PCP-O-dG
hydrolyzed 3.7 times faster than dG, indicating a lower
activation barrier. The ΔS‡ values do not show any particular
trend; however, they are all small positive values, suggesting a
late transition state, whereby the glycosidic bond is breaking to
create more disorder in the system. Overall, the activation
parameters for the O-linked 8-dG adducts are consistent with
parameters for the hydrolysis of purine deoxynucleosides, and
the magnitudes are comparable.6
effects are largely absent. Under this scenario, ortho
substituents have similar electronic influences as para
substuents;34−37 therefore, σp values were also utilized for σo.
The resulting plot yielded a negative slope (ρ) of −0.65 with a
good linear correlation of R2 = 0.94. The negative reaction
constant ρ suggests that the reaction rate is determined by the
step that includes a buildup of positive charge, the pre-
equilibrium established in the first step of the acid-catalyzed
deglycosylation mechanism (Figure 1a). At pH 1 it is unlikely
that O-linked adducts form dicationic species given that pKa2
for dG is ∼ −2.5.38 Under conditions where involvement of the
diprotonated species can be ignored, k1 ≈ kobsKa1/[H+].5,6
Thus, log kobs = log k1 − log Ka1 + log [H+] and log kobs = σρ ≈
σρk1 − σρKa1 + constant (log [H+]). Electron-withdrawing
substituents favor bond cleavage (k1)14,19 as well as dissociation
of the acid (Ka1)37 and provide positive ρ values in Hammett
plots. The negative slope observed for the Hammett plot
(Figure 3) implies that σρKa1 is larger than σρk1 at pH 1.
Previously, Hammett analysis for the C-linked 8-dG adducts at
∼ pH 1 showed a small positive slope of 0.54.19 In this case the
pH of the solution was well below N7H+ pKa values for the C-
linked 8-dG adducts. However, at pH 4, where pH > pKa1, the
Hammett analysis for the C-linked 8-dG adducts provided a
negative slope.39 The Hammett plot shown in Figure 3 implied
surprisingly low N7H+ pKa values (<1) for the O-linked 8-dG
adducts.
Figure 3 shows a plot of log kobs/kobs(Ph‑O‑dG) versus Hammett
substituent constant σ for hydrolysis of the O-linked
Proton Affinity (PA). Efforts to determine ionization
constants for O-linked 8-dG adducts in a low pH range
(0.5−4 in aqueous buffered media) were carried out at 25 °C
using the spectrophotometric procedure. Overlay absorption
spectra for Ph-O-dG as a function of pH showed an increase in
absorbance at 273 nm as the solution acidity increased. By
plotting the initial absorbance at 273 nm as a function of pH for
Ph-O-dG (Figure S3, Supporting Information), a pKa value of
1.1 0.2 was obtained for the protonated species. This value is
more than 1 pH unit below the pKa of N7H+-dG (2.34).8 For
Figure 3. Hammett plot for O-linked phenolic 8-dG adducts using σ vs
log kobs/kobs(Ph‑O‑dG) in pH 1 buffer (0.2 M KCl/HCl) at 37 °C.
chlorophenolic 8-dG adducts using the rate data presented in
Table 1. For adducts containing more than one Cl-substituent,
σ is the sum of the individual contributions of each Cl in the
adduct:32 values of σm and σp of 0.37 and 0.23 are taken from
the literature.33 For ortho Cl groups it was assumed that steric
Table 2. Ionization Constants for 2-Phenoxypyridines and O-Linked 8-dG Adducts Used for Determination of First-Order Rate
Constants (k1) at pH 1
a
e
compound
NH+ pKa
ΔpKa
adduct
N7H+ pKa
Ka
k1 (min−1
)
b
a
Ph-O-pyr
2.24 0.05
2.05 0.07
1.51 0.06
−3.01
−0.18
−0.73
Ph-O-dG
1.1 0.2
0.079
0.12
0.43
0.87
0.97
2.2
c
c
d
4-Cl-Ph-O-pyr
DCP-O-pyr
4-Cl-Ph-O-dG
DCP-O-dG
0.92
d
0.37
a
b
Obtained from spectrophotometric titration at 25 °C from an average for three independent measurements. Compared to literature pKa value (ref
c
d
40) for pyridine. Compared to pKa value determined for Ph-O-pyr. Estimated value from pKa = pKa Ph-O-dG (1.1) + ΔpKa (−0.18 or −0.73).
e
Recorded at pH 1 where k1 ≈ kobsKa1/[H+].
7179
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185

The Journal of Organic Chemistry
Article
the chlorophenolic O-linked 8-dG adducts, accurate pKa values
could not be determined because of the weakly basic nature of
N7, coupled with competitive hydrolysis of the adduct in
strongly acidic media. To provide surrogate information on the
impact of the phenoxy substituent on N7 basicity, 2-
phenoxypyridine analogues (Ph-O-pyr, 4-Cl-Ph-O-pyr and
DCP-O-pyr) were prepared as model compounds, and their
proton affinity was analyzed using the spectrophotometric
procedure. The sp2-hybridized ring nitrogen of pyridine is
much more basic than N7 of dG, possessing a pKa of 5.25 for its
conjugate acid,40 and the pyridine derivatives do not undergo
hydrolysis under our experimental conditions. Thus, pKa values
for the conjugate acids of Ph-O-pyr, 4-Cl-Ph-O-pyr and DCP-
O-pyr were easier to determine with accuracy compared to
their corresponding monoprotonated 8-dG adducts. Spectro-
photometric titrations of the 2-phenoxypyridine analogues
(Figure S4, Supporting Information) provided the pKa values
for the conjugate acids given in Table 2. The pKa of the
conjugate acid of Ph-O-pyr was ∼2.24 for a ΔpKa value of
−3.01 relative to the conjugate acid of pyridine. 4-Cl-Ph-O-pyr
and DCP-O-pyr have pKa values for the conjugate acids of 2.05
and 1.51, respectively. For comparison, the pKa for the
conjugate acid of 2-chloropyridine is ∼0.5.40
The ΔpKa values of −0.18 and −0.73 for 4-Cl-Ph-O-pyr and
DCP-O-pyr relative to Ph-O-pyr (Table 2) provided a measure
on how these Cl-substituted phenoxy ring systems impact the
basicity of the adjacent sp2-hybridized N atom. Given that Ph-
O-dG possesses a pKa ∼ 1.1 for its conjugate acid, the ΔpKa
values were used to estimate the N7H+ pKa values for 4-Cl-Ph-
O-dG and DCP-O-dG (i.e., pKa = pKa Ph-O-dG (1.1) + ΔpKa
(−0.18 or −0.73)) and were calculated to be 0.92 and 0.37,
respectively. Utilizing the Ka values, k1 values of 0.87, 0.97, and
2.2 min−1 were determined for the three O-linked adducts at
pH 1 (Table 2). The increase in k1 upon addition of Cl groups
is the expected trend for addition of electron-withdrawing
substituents to the 8-position of dG.13,14,19
Our experimental findings suggested that increased Cl-
substitution of the phenyl ring in O-linked 8-dG adducts
diminishes the rate of hydrolysis at pH 1 by lowering N7
basicity. However, protonation at N3 may compete effectively
for protonation at N7 for some O-linked 8-dG adducts, and this
would influence the hydrolysis reaction. Given our inability to
distinguish N7 from N3 protonation using spectrophotometric
techniques, DFT calculations were used to determine structures
of the protonated species (i.e., Figure 4) and determine gas-
phase and solvent-phase (water) proton affinities (PA,
kcal·mol−1) for both N7H+ and N3H+ adducts (Table 3). For
dG, the anti-conformation is the most stable structure for both
the neutral and N7H+ species (Table 3). However, the syn-
conformation is favored for the N3H+ species due to H-bonding
interaction between 5′-O of the sugar moiety and N3H. For the
O-linked 8-dG adducts, the syn-conformation is the most stable
structure (Figure 4, Table 3), which reduces steric interactions
between the phenolic ring and sugar moiety and favors H-
bonding interactions between 5′-OH and N3 (Figure 4). On
the basis of terminology for describing biaryl ether
conformation,41 neutral Ph-O-dG (Figure 4), 4-Cl-Ph-O-dG
and DCP-O-dG (Table 3) were present in a planar
conformation, while neutral TCP-O-dG (Table 3) and PCP-
O-dG (Figure 4) adopted a skew conformation that diminishes
steric interactions between the dG moiety and the TCP and
PCP ring system. All O-linked 8-dG adducts adopt a skew
Figure 4. The most stable B3LYP/6-311+G(2df,p)//B3LYP/6-
31G(d) conformers for the Ph-O-dG and PCP-O-dG adducts, as
well as their N3-(N3H+) and N7-(N7H+) protonated analogues (Select
hydrogen bond lengths (Å) are provided).
a
Table 3. Low-Energy Conformations and Proton Affinities
(PA, kcal·mol−1) at the N3 and N7 Site of O-Linked 8-dG
b
adducts
N3
N3
N7
N7
adduct neutral N7H+ N3H+ PA(gas) PA(water) PA(gas) PA(water)
Ph-O-
dG
syn-
planar
syn-
skew
syn-
skew
219.8
217.9
257.8
257.4
227.8
225.2
259.6
258.8
4-Cl-
Ph-
syn-
planar
syn-
skew
syn-
skew
O-dG
DCP-
O-dG
TCP-
O-dG
PCP-
O-dG
dG
syn-
planar
syn-
skew
syn-
skew
219.1
219.1
218.0
216.6
259.1
257.4
257.4
255.7
223.9
223.2
221.3
232.1
258.0
254.8
253.8
260.8
syn-
skew
syn-
skew
syn-
skew
syn-
skew
syn-
skew
syn-
skew
anti
anti
syn
a
The most stable B3LYP/6-311+G(2df,p)//B3LYP/6-31G(d) con-
formers for the neutral, N7-(N7H+) and N3-(N3H+) protonated
b
analogues. N3 and N7 proton affinity (PA) was calculated as the
negative of the enthalpy change for protonation (kcal·mol−1).
conformation when protonated at both N7 and N3 (Table 3,
Figure 4).
In the gas-phase the calculated N7 PA for dG is 232.1
kcal·mol−1 (Table 3), which is similar to the experimental PA
(234.4 kcal·mol−1),42 and is ∼15.5 kcal·mol−1 above the gas-
phase N3 PA; a calculated value of ∼20 kcal·mol−1 has been
previously reported.43 In water the DFT calculations still
predict a preference for N7-protonation of dG, but the energy
difference between the PA of N7 and N3 is only ∼5 kcal·mol−1
(Table 3). Attachment of the phenoxy substituent to C8 of dG
to afford Ph-O-dG decreases the gas-phase N7 PA by 4.3
kcal·mol−1, and raises N3 PA by 3.2 kcal·mol−1. The gas-phase
7180
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185

The Journal of Organic Chemistry
Article
calculations indicate a strong preference for N7-protonation
(favored by 8 kcal·mol−1) of Ph-O-dG, while in water the
preference for N7-protonation is only favored by 1.8 kcal·mol−1.
Attachment of Cl-substituents to the phenyl ring further
diminishes N7 PA, with PCP-O-dG having a gas-phase N7 PA
of 221.3 kcal·mol−1, which is 6.5 kcal·mol−1 below N7 PA for
Ph-O-dG. In contrast, N3 PA is not strongly influenced by Cl-
substitution, and the gas-phase N3 PA is ∼218−220 kcal·mol−1
for all O-linked adducts, which is 1.4−3.2 kcal·mol−1 above N3
PA calculated for dG. This difference may stem from the fact
that dG is the only base that undergoes a conformational
change (anti → syn) upon N3-protonation. In water the
calculations predict N3 to have a greater PA than N7 for DCP-
O-dG, TCP-O-dG and PCP-O-dG. Thus, for these O-linked
adducts, protonation at N3 may compete effectively for
protonation at N7, and this would influence the hydrolysis
reaction.
Influence of Protonation Site on Deglycosylation.
Given that DFT calculations predict a greater N3 PA than N7
PA for DCP-O-dG, TCP-O-dG and PCP-O-dG (Table 3), it
was desirable to determine the influence of protonation site on
the deglycosylation barrier. Figure 5 shows the calculated
deglycosylation profile for Ph-O-dG (purple trace), 4-Cl-Ph-O-
dG (blue trace), DCP-O-dG (green trace), TCP-O-dG (orange
trace) and PCP-O-dG (red trace), as well as the corresponding
N3- and N7-protonated species (kJ·mol−1). The calculations
predict that N7-protonation has a much larger effect on the
barrier for deglycosylation than does N3-protonation. In water
(Figure 5a), the barrier for the N7-protonated O-linked adducts
is from ∼60−80 kJ·mol−1, while the barrier for the N3-
protonated adducts is ∼100−120 kJ·mol−1 that is further
increased to ∼140 kJ·mol−1 for the neutral adducts. For each
species (neutral, N3H+ and N7H+), the calculations predict
PCP-O-dG to possess the lowest barrier, followed by TCP-O-
dG, DCP-O-dG, 4-Cl-Ph-O-dG and finally Ph-O-dG. In the
gas-phase (Figure 5b), the deglycosylation barrier is slightly
diminished for the N7-protonated adducts versus the
corresponding barrier in water, while the barrier for the N3-
protonated adducts is increased and is comparable to the
neutral adducts. These calculations imply that the chlorophe-
noxyl substituents could influence the site of protonation to
generate N3H+ adducts that are relatively stable to hydrolysis
compared to their N7H+ counterparts. This possibility cannot
be ruled out and may provide a rationale for the relative
stability of PCP-O-dG in aqueous media at pH 1.
Gas-Phase Deglycosylation. Further insight into the
influence of Cl-substitution on the hydrolysis of the O-linked
8-dG adducts was accomplished using electrospray positive
ionization (ESI+) mass spectrometry. The deglycosylation
barriers predicted by DFT calculations suggested PCP-O-dG
to be the most reactive, while hydrolysis kinetics in aqueous
media at pH 1 showed PCP-O-dG to be the least sensitive to
acid-catalyzed hydrolysis of the O-linked adducts. In the DFT
calculations, barriers were determined for the monoprotonated
species, while the hydrolysis kinetics is influenced by the pre-
equilibrium established in the first step of the acid-catalyzed
deglycosylation mechanism. Thus, it was desirable to use an
experimental technique such as ESI-MS that can isolate
monoprotonated adducts and determine their relative stability
to deglycosylation. These relative gas-phase stabilities can be
expressed as the percentage of deglycosylated product as
determined by the relative peak intensities between the parent
ion [M + H]+ and deglycosylated product ion [M − 116 + H]+
by ESI+-MS/MS. It is important to note that in this experiment,
we assumed that the ionization potential of the parent ion and
deglycosylation product ion were the same across each adduct
tested.
Figure 6a shows the ESI+-MS/MS spectrum of 4-Cl-Ph-O-
dG. The adduct shows an [M + H]+ ion at 394 with loss of 116
mass units corresponding to the deoxyribose sugar moiety to
generate a positively charged nucleobase at m/z 278. This
observation was consistent with deglycosylation of the
monoprotonated adduct through cleavage of the N-glycosidic
bond.44−46 The percentage of deglycosylated product for 4-Cl-
Ph-O-dG was determined to be 28%. Under identical ESI+-MS/
MS conditions the percentage of deglycosylated product for
Ph-O-dG and the other chlorinated O-linked adducts was
determined (Figure 6b). Consistent with the DFT calculations
(Figure 5), the monoprotonated Ph-O-dG adduct was the least
reactive with only 1.2% deglycosylated product. All chlorinated
O-linked adducts had monoprotonated species more sensitive
to deglycosylation than monoprotonated Ph-O-dG (Figure 6b).
However, upon further chlorination of the phenyl ring system
the degree of deglycosylation decreased relative to that of 4-Cl-
Ph-O-dG with the monoprotonated species of PCP-O-dG
showing the lowest percentage of deglycosylated product
(9.9%) among the chlorinated analogues.
Figure 5. Constrained IEF-PCM-B3LYP/6-31G(d) deglycosylation
barriers calculated in (a) water and (b) the gas phase for Ph-O-dG
(purple), 4-Cl-Ph-O-dG (blue), DCP-O-dG (green), TCP-O-dG
(orange) and PCP-O-dG (red), as well as the corresponding N3-(N3
H+) and N7-(N7 H+) protonated species (kJ·mol−1).
7181
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185

The Journal of Organic Chemistry
Article
Information) were constructed and afforded straight lines,
indicating continuous first-order dependence on H+ activ-
ity.13,19 Experimental limitations for obtaining rates over a
broader range of pH values included the limited solubility of O-
linked adducts, especially chlorinated analogues, at pH values
≥2.5, and the inhibition of hydrolysis when using CH3CN as
cosolvent to enhance solubility. To provide an estimate of rates
at neutral pH for O-linked adducts, the pH-rate profiles were
extrapolated to pH 7 to render rates of 6 × 10−9 min−1 (t1/2
≈
220 years) and 3 × 10−10 min−1 (t1/2 ≈ 4395 years) for Ph-O-
dG and DCP-O-dG, respectively, at 55 °C. For comparison, a
similar estimate for the rate of hydrolysis of the C-linked 8-Ph-
dG nucleoside adduct at neutral pH is ∼1.90 × 10−5 min−1 (t1/2
≈ 25 days) at 37 °C.19 For N-linked 8-dG adducts, rates of
hydrolysis at neutral pH are as high as 5.2 × 10−7 s−1 (t1/2 ≈ 15
days) at 20 °C.18 A corresponding rate for dG at 20 °C is
estimated at ∼1.2 × 10−9 s−1 (t1/2 ≈ 6,684 days).18 The
inclusion of our data on the stability of O-linked adducts now
establishes that the order of hydrolytic reactivity at neutral pH
for bulky 8-dG adducts is N-linked > C-linked > O-linked,
which correlates with their relative ease of N7-protonation.
CONCLUSIONS
■
The current study has allowed us to conclude the following: (1)
O-linked 8-dG adducts produced by phenols are resistant to
hydrolysis under physiological conditions and are unlikely
intermediates leading to abasic site formation. (2) Their
resistance to hydrolysis stems from the ability of the phenoxy
substituent to diminish N7 proton affinity, which is supported
by both experimental pKa values and calculated proton affinities
(PAs) in gas- and solvent- (water)-phase. (3) In water at pH 1,
increased Cl-substitution diminishes the rate of hydrolysis by
disfavoring N7-protonation in the pre-equilibrium established in
the first step of the acid-catalyzed deglycosylation mechanism.
This trend is supported by the Hammett plot, which shows a
linear negative slope with ρX = −0.65. However, DFT
calculations show that the barrier to deglycosylation becomes
progressively smaller for N7H+ adducts bearing increased
numbers of Cl-substituents. This observation follows the
general trend for hydrolysis of 8-dG adducts; electron-
withdrawing 8-substituents increase the rate of depurination
through stabilization of the developing negative charge at N9
during rate-limiting cleavage of the glycosyl bond. (4) The
degree of deglycosylation induced by CID in ESI+-MS
demonstrate that all O-linked 8-dG adducts containing Cl-
substituents are more prone to deglycosylation than the
unsubstituted analogue Ph-O-dG in the gas-phase. However,
within the chlorinated analogues, 4-Cl-Ph-O-dG showed the
greatest percent deglycosylation, and progressive addition of
Cl-substituents to generate PCP-O-dG diminished the extent of
deglycosylation. DFT calculations suggest that N3-protonation
may compete effectively with N7-protonation for both TCP-O-
dG and PCP-O-dG, and monoprotonated N3H+ species possess
much greater barriers to deglycosylation than their N7-
protonated counterparts.
Figure 6. (a) ESI+-MS/MS spectrum of 4-Cl-Ph-O-dG. (b) Relative
gas-phase stabilities of O-linked 8-dG adducts are expressed as %
deglycosylated product by determining the relative peak intensities of
the parent ion [M + H]+ and deglycosylated product ion [M-116 +
H]+ by ESI+-MS/MS.
The mechanism of deglycosylation for protonated dG by
collision-induced dissociation (CID) in MS has recently been
explored using deuterium-enriched samples.44,45 These experi-
ments suggest that fragmentation proceeds from the syn-
conformation of the N7H+-dG species with initial proton
transfer from 5′-OH of the sugar moiety to N3 on the
nucleobase.44,45 This proton transfer generates a dicationic
nucleobase and an anionic sugar possessing a negatively
charged 5′-alkoxy group. Liu and co-workers suggest that the
5′-alkoxy group performs a nucleophilic attack at the 1′-site to
release the protonated nucleobase and form a neutral sugar
containing a new 5-membered ring.44 In contrast, Greisch and
co-workers favor a mechanism in which a 5′-alkoxy group
deprotonates a 2′-H, resulting in beta-elimination of the
nucleobase to produce a sugar byproduct as a neutral species.45
Regardless of the exact mechanism, the 8-phenoxy substituents
in the O-linked adducts appear well removed from the 1′-N9
reaction site (i.e., Figure 4). Thus, steric interference of
deglycosylation by increased Cl substitution is unlikely to play a
major role in the relative percent deglycosylated product
observed by ESI+-MS. Instead, electronic factors are proposed
to play a main role. For TCP-O-dG and especially PCP-O-dG,
parent ions may contain both N3H+ and N7H+ ions and
increased involvement of N3H+ ions will increase the
deglycosylation barrier (Figure 5) and hence decrease the
percent deglycosylated product (Figure 6).
EXPERIMENTAL SECTION
■
Materials and Methods. Anhydrous 1,4-dioxane was distilled
over Na. Xylenes and pyridine were distilled over CaH2 and stored
under nitrogen. Other commercial compounds were used as received.
O-Linked 8-dG adducts 4-Me-Ph-O-dG, 4-OMe-Ph-O-dG, TCP-O-
dG and PCP-O-dG were synthesized as previously reported and were
Hydrolysis at Neutral pH. To gain insight into hydrolysis
rates at neutral pH, hydrolysis kinetics between pH 0.8−2.5 for
Ph-O-dG and DCP-O-dG at 55 °C (Figure S5, Supporting
30
1
>95% pure by H NMR. The synthesis of 8-bromo-2′-deoxyguano-
7182
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185

The Journal of Organic Chemistry
Article
sine (8-Br-dG) was performed, as outlined previously,47 by treating dG
with N-bromosuccinimide in a water−acetonitrile mixture. 8-Bromo-
3′,5′-O-bis(tert-butyldimethylsilyl)-2′-deoxyguanosine was prepared
according to literature procedures by treating 8-Br-dG with excess
tert-butyl(chloro)dimethylsilane and imidazole in DMF.31 NMR
spectra were recorded on 300 and 600 MHz spectrometers in either
DMSO-d6, CDCl3 or CD2Cl2 referenced to the respective solvent.
High-resolution mass spectra (HRMS) were recorded on a Q-TOF
instrument, operating in nanospray ionization at 0.5 μL/min detecting
positive ions.
white solid (80% over two steps): mp 155 °C decomp.; 1H NMR (300
MHz, DMSO-d6) δ 10.64 (s, 1H), 7.81 (s 1H), 7.53, (m, 2H), 6.44,
(bs, 2H), 6.21 (t, J = 7.3 Hz, 1H), 5.25 (d, J = 4.1 Hz, 1H), 4.79 (t, J =
5.9 Hz, 1H), 4.32 (m, 1H), 3.76 (m, 1H), 3.47 (m, 1H), 3.43 (m, 1H),
2.92 (m, 1H), 2.15 (m, 1H); 13C NMR (151 MHz, DMSO-d6) δ
155.7, 153.5, 150.1, 148.5, 147.8, 130.6, 129.9, 128.7, 126.2, 124.5,
110.6, 87.5, 81.9, 70.9, 62.0, 36.6; HRMS (ESI) calcd for
C16H16Cl2N5O5 428.0528, found 428.0522 (MH+).
8-Phenoxy-guanine (Ph-O-G). 8-Phenoxy-2′-deoxyguanosine
(Ph-O-dG, 0.02 g, 0.05 mmol) was added to a round-bottom flask
along with 5 mL of 10% formic acid. The mixture was stirred and
heated to 75 °C for 1 h. After cooling, 10 mL of water was added, and
the pH was adjusted to 6 using 1 M NaOH. The isolated solid (14 mg,
96%) was the formate salt of Ph-O-dG: mp 237 °C decomp.; 1H NMR
(300 MHz, DMSO-d6) δ 10.93 (bs, 1H), 8.45 (s, 1H), 8.29 (s, 1H,
(HCOOH)), 7.39 (m, 2H), 7.25 (m, 2H), 7.23 (m, 1H), 6.40 (s, 2H);
13C NMR (151 MHz, DMSO-d6, partial assignment) δ 165.8
(HCOOH), 154.1, 153.1, 129.5, 124.4, 119.2; HRMS (ESI) calcd
for C11H10N5O2 244.0834, found 244.0835 (MH+).
General Method for Synthesis of 2-Phenoxypyridines. 2-
Phenoxypyridines were prepared using a microwave-assisted Ullmann
ether synthesis reported by D’Angelo and co-workers.48 2-Bromopyr-
idine and the desired phenol (1.5 equiv) were suspended in dry DMF
(∼4 mL) in a microwave vial. Copper powder (10%) and Cs2CO3 (3
equiv) were added, and the system was purged with argon and then
placed in the microwave reactor and allowed to react for 10 min at 100
°C and 60 W. The reaction was then cooled to room temperature and
diluted with CH2Cl2 (25 mL). The organic phase was washed with 1
M NaOH (60 mL) and then 100 mL water, dried over Na2SO4,
decanted, and concentrated in vacuo. The crude product was purified
by silica gel chromatography using (10:2 Hex:EtOAc) to yield the
desired 2-phenoxypyridines, which were characterized by NMR
spectroscopy.
2-Phenoxypyridine (Ph-O-pyr). Following the general proce-
dure, phenol (1.5 g, 4.6 mmol) and 2-bromopyridine (0.16 mL, 1.6
mmol) provided 0.273 g of Ph-O-pyr49 as a clear oil (100%): 1H NMR
(300 MHz, CDCl3) δ 8.11 (dd, J = 5.0, 1.1 Hz, 1H), 7.57 (td, J = 7.2,
2.0 Hz, 1H), 7.31 (t, J = 6.2 Hz, 2H), 7.28−7.03 (m, 3H), 6.88 (t, J =
5.9 Hz, 1H), 6.79 (d, J = 8.3 Hz, 1H); 13C NMR (151 MHz, CDCl3) δ
163.7, 154.1, 147.8, 139.4, 129.7, 124.6, 121.1, 118.4, 111.5.
8-Bromo-3′,5′-O-bis(tert-butyldimethylsilyl)-O6-(trimethylsi-
lylethyl)-2′-deoxyguanosine (2). Compound 2 was prepared by
treating 8-bromo-3′,5′-O-bis(tert-butyldimethylsilyl)-2′-deoxyguano-
sine (2 g, 3.48 mmol) with triphenylphosphine (4.6 g, 17.4 mmol),
2-(trimethylsilyl)-ethanol (2.5 mL, 17.4 mmol) and diisopropyl
azodicarboxylate (DIAD) (3.4 mL, 17.4 mmol) in anhydrous dioxane,
as outlined previously.31 The crude product was purified by silica gel
column chromatography (Hex:EtOAc 20:1 to 10:1) to yield 1.5 g
1
(64%) of a colorless oil: H NMR (300 MHz, CDCl3) δ 6.24 (t, J =
6.9 Hz, 1H), 4.75 (m, 1H), 4.70 (bs, 2H), 4.50 (m, 2H), 3.90 (m, 1H),
3.85 (m, 1H), 3.67 (dd, J = 10.2, 4.3 Hz, 1H), 3.52 (m, 1H), 2.14 (m,
1H) 1.18 (m, 2H), 0.91 (s, 9H), 0.83 (s, 9H), 0.12 (s, 6H), 0.06 (s,
9H), 0.01 (s, 3H), −0.03 (s, 3H); 13C NMR (151 MHz, CDCl3) δ
160.2, 158.5, 154.1, 125.6, 116.5, 87.3, 85.6, 72.3, 65.0, 62.7, 36.3, 25.8,
18.3, 18.0, 17.5, −1.5, −4.6, −4.7, −5.43, −5.5; MS (ESI) calcd for
C27H53BrN5O4Si3 674.3, found m/z 674.3 (MH+).
General Method for O-Linked 8-dG Adduct Synthesis.
Compound 2 was added to a round bottomed flask with 4 equiv of
the desired phenol and 2 equiv of finely ground K3PO4. The reaction
vessel was sealed and purged with argon, prior to adding 8 mL of
distilled xylenes. The reaction was then stirred and heated at 130 °C
for ∼17 h. After completion, 25 mL of EtOAc was added, and the
mixture was washed with 2 × 50 mL saturated bicarbonate solution,
followed by 50 mL of water. The organic layer was dried over Na2SO4
and concentrated in vacuo. To the crude product was added 3 equiv of
TBAF·3H2O and 5 mL of reagent grade THF. THF was removed in
vacuo, and the crude product was purified by silica gel chromatography
(10% methanol in methylene chloride) to yield the O-linked
1
nucleoside adduct in ≥95% purity, as evidenced by H NMR.
8-Phenoxy-2′-deoxyguanosine (Ph-O-dG). 8-Phenoxy-2′-deox-
yguanosine (Ph-O-dG)30 was synthesized from compound 2 (1 g, 1.48
mmol), phenol (0.56 g, 5.92 mmol), and K3PO4 (0.63g, 2.96 mmol)
followed by deprotection with TBAF·3H2O (1.4 g, 4.44 mmol) to
yield 0.38 g of a white solid (73% over two steps): mp 148 °C
decomp.; 1H NMR (600 MHz, DMSO-d6) δ 10.68 (bs, 1H), 7.44 (t, J
= 8.5 Hz, 2H), 7.31 (d, J = 7.8 Hz, 2H), 7.25 (t, J = 7.4 Hz, 1H), 6.45,
(s, 2H), 6.21 (t, J = 7.3 Hz, 1H), 5.27 (d, J = 4.0 Hz, 1H), 4.85 (t, J =
5.9 Hz, 1H), 4.34 (m, 1H), 3.77 (m, 1H), 3.50 (m, 1H), 3.44 (m, 1H),
2.91 (m, 1H), 2.15 (m, 1H); 13C NMR (151 MHz, DMSO-d6) δ
155.8, 153.5, 153.3, 149.8, 149.0, 129.7, 125.2, 119.8, 110.7, 87.4, 81.8,
70.9, 62.0, 36.6; HRMS (ESI) calcd for C16H18N5O5 360.1308, found
360.1304 (MH+).
2-(4-Chlorophenoxy)pyridine (4Cl-Ph-O-pyr). Following the
general procedure 2-bromopyridine (0.155 mL, 1.6 mmol) and 4-
chlorophenol (0.31 g, 2.4 mmol) afforded 0.204 g of 4Cl-Ph-O-pyr50
1
as a yellow oil (62%): H NMR (300 MHz, CDCl3) δ 8.07 (dd, J =
4.4, 1.3 Hz, 1H), 7.56 (td, J = 7.2, 2.0 Hz, 1H), 7.23 (d, J = 6.7 Hz,
2H), 6.98 (d, J = 5.9 Hz, 2H), 6.86 (t, J = 7.2 Hz, 1H), 6.79 (d, J = 8.3
Hz, 1H); 13C NMR (151 MHz, CDCl3) δ 163.3, 152.6, 147.6, 139.5,
129.7, 129.6, 122.6, 118.7, 111.6.
2-(2,4-Dichlorophenoxy)pyridine (DCP-O-pyr). Following the
general procedure 2-bromopyridine (0.155 mL, 1.6 mmol) and 2,4-
dichlorophenol (0.39 g, 2.4 mmol) yielded 0.157 g of DCP-O-pyr51 as
1
a clear oil (41%): H NMR (300 MHz, CD2Cl2) δ 8.15 (dd, J = 4.9,
8-(2-Chlorophenoxy)-2′-deoxyguanosine (2Cl-Ph-O-dG). 8-
(2-Chlorophenoxy)-2′-deoxyguanosine (2Cl-Ph-O-dG) was synthe-
sized from compound 2 (0.2 g, 0.30 mmol), 2-chlorophenol (0.15 g,
1.18 mmol), and K3PO4 (0.13 g, 0.59 mmol) followed by deprotection
with TBAF·3H2O (0.28 g, 0.88 mmol) to yield 0.08 g of a white solid
1.2 Hz, 1H), 7.74 (td, J = 7.2, 2.0 Hz, 1H), 7.49 (d, J = 2.5 Hz, 1H),
7.30 (dd, J = 8.6, 2.5 Hz 1H), 7.17 (d, J = 8.7 Hz, 1H), 7.05−7.00 (m,
2H); 13C NMR (151 MHz, CDCl3) δ 162.6, 148.5, 147.4, 139.6,
130.7, 130.3, 128.2, 128.0, 124.7, 118.8, 111.1.
1
Kinetics and pKa Determinations. The kinetic study was carried
out using a UV−vis spectrophotometer equipped with a constant
temperature water bath. Hydrolysis reactions were followed by
monitoring formation of the deglycosylated nucleobase at its
absorption maximum (λmax, e.g., 275 nm for Ph-O-G). The reaction
was initiated by injection of 20 μL of a 4 mM stock solution (DMSO)
of O-linked 8-dG adduct into a Teflon-capped 10 mm UV cell
containing 1980 μL of aqueous 0.2 M buffer (pH 0.8−2.5) that had
been incubated at the appropriate temperature (25, 37, 45, 55, or 65
°C) for 15 min. Measurements were conducted in parallel using the
multicell changer, with six first-order rate constant values for hydrolysis
obtained for each adduct in each set of conditions to allow for
determination of mean standard deviation.
(68% over two steps): mp 135 °C decomp.; H NMR (300 MHz,
DMSO-d6) δ 10.62 (bs, 1H), 7.62 (d, J = 6.6 Hz, 1H), 7.44 (m, 2H),
7.33 (m, 1H) 6.42, (bs, 2H), 6.22 (t, J = 7.1 Hz, 1H), 5.25 (d, J = 4.1
Hz, 1H), 4.79 (t, J = 5.9 Hz, 1H), 4.32 (m, 1H), 3.77 (m, 1H), 3.46
(m, 2H), 2.94 (m, 1H), 2.17 (m, 1H); 13C NMR (151 MHz, DMSO-
d6) δ 155.7, 153.4, 150.0, 148.8, 148.7, 130.4, 128.6, 127.3, 125.0,
123.2, 110.6, 87.5, 81.9, 70.9, 62.0, 36.6; HRMS (ESI) calcd for
C16H17ClN5O5 394.0918, found 394.0913 (MH+).
8-(2,4-Dichlorophenoxy)-2′-deoxyguanosine (DCP-O-dG). 8-
(2,4-Dichlorophenoxy)-2′-deoxyguanosine DCP-O-dG was synthe-
sized from compound 2 (0.22 g, 0.388 mmol), 2,4-dichlorophenol
(0.22 g, 1.34 mmol), and K3PO4 (0.14 g, 0.66 mmol) followed by
deprotection with TBAF·3H2O (0.42g, 1.3 mmol) to yield 0.13 g of a
7183
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185

The Journal of Organic Chemistry
Article
and N3H+ adducts). This material is available free of charge via
the Internet at http://pubs.acs.org.
UV−vis titrations for pKa determinations of Ph-O-dG, Ph-O-pyr, 4-
Cl-Ph-O-pyr and DCP-O-pyr were carried out at 25 °C. Solutions
were prepared using 20 μL of a 4 mM substrate stock solution
(DMSO) and 1980 μL of buffer consisting of pH 0.8−1.8 (0.2 M
KCl/HCl) or pH 2−5 (0.05 M citric acid, 0.31 M NaCl). pKa values
were obtained as outlined in detail previously,19 and error was
calculated from three independent measurements for determination of
mean standard deviation.
AUTHOR INFORMATION
Corresponding Author
*E-mail: sturla@ethz.ch (S.J.S.); stacey.wetmore@uleth.ca
■
(S.D.W.); rmanderv@uoguelph.ca (R.A.M.).
HPLC Analysis. Confirmation of deglycosylation of Ph-O-dG was
accomplished using reverse-phase HPLC. Samples of Ph-O-dG and
Ph-O-G standards were prepared from 1 μL of the 40 mM DMSO
adduct stock solution and 99 μL of purified water (18.2 MΩ). Samples
of Ph-O-dG that were subjected to acid before injection were prepared
using 20 μL of the DMSO adduct stock and 1980 μL of 0.1 M HCl.
The acidic solution was heated to 40 °C and allowed to react for 5 and
10 min; 100 μL was taken out at the desired time of analysis and
directly injected into the HPLC. The compounds were separated on a
C18 column (50 × 4.60 mm) with a flow rate of 0.75 mL/min using a
gradient running from 95% 50 mM aqueous triethylamine acetate
(TEAA, pH 7.2):5% acetonitrile to 30% 50 mM TEAA:70%
acetonitrile over 30 min. Adduct detection was accomplished by
monitoring the UV absorbance at 258 nm.
ESI-MS. Experiments were carried out using a quadrupole ion trap
mass spectrometer in the positive ionization mode with an
electrospray ionization source. The MSn spectra were obtained by
collision-induced dissociation (CID) with helium gas after isolation of
the appropriate precursor ions. Ionization was carried out using the
following settings on the ESI: nebulizer gas flow 40 psi, dry gas 10 L/
min, dry temperature 220 °C, spray voltage 4500 V. The scan range
was 100−1000 m/z and scan resolution was 8100 m/z/s. Collision
energies were 0.36 V for ESI+-MS/MS experiments. Adduct stock
solutions (6 mM in DMSO) were diluted to 0.03 mM with water and
introduced, through electrospray, to the quadrupole ion trap using a
syringe pump at flow rates of 3−5 μL/min. The collection time for
each experiment was 1.0 min (300 scans/min).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support for this research was provided by the Natural
Sciences and Engineering Research Council (NSERC) of
Canada, the Canada Research Chair program, the Canada
Foundation for Innovation, the Ontario Innovation Trust Fund,
and the European Research Council (260341). This research
has been facilitated by use of computing resources provided by
WestGrid and Compute/Calcul Canada.
REFERENCES
(1) Lindahl, T. Nature 1993, 362, 709−715.
■
(2) Evans, A. R.; Limp-Foster, M.; Kelley, M. R. Mutat. Res. 2000,
461, 83−108.
(3) Loeb, L. A.; Preston, B. D. Annu. Rev. Genet. 1986, 20, 201−230.
(4) Roger, M.; Hotchkiss, R. Proc. Natl. Acad. Sci. U. S. A. 1961, 47,
654−669.
(5) Zoltewicz, D.; Clark, D. F.; Sharpless, T. W.; Grahe, G. J. Am.
Chem. Soc. 1970, 92, 1741−1750.
(6) Hevesi, L.; Wolfson-Davidson, E.; Nagy, J. B.; Nagy, O. B.;
Bruylants, A. J. Am. Chem. Soc. 1972, 94, 4715−4720.
(7) Zoltewicz, J. A.; Clark, D. F. J. Org. Chem. 1972, 37, 1193−1197.
(8) Da Costa, C. P.; Sigel, H. Inorg. Chem. 2003, 42, 3475−3482.
(9) Dutta, S.; Chowdhury, G.; Gates, K. S. J. Am. Chem. Soc. 2007,
129, 1852−1853.
Computational Details. An internal coordinate Monte Carlo52
conformational search was initially performed using HyperChem53
with the AMBER molecular mechanics force field to determine the
preferred conformations of the neutral O-linked phenoxyl adduct and
the various Cl-substituted analogues, as well as the corresponding N3-
and N7-protonated equivalents. Partial atomic charges for the AMBER
calculations were obtained from the PM3 method. In each case, more
than 200 conformers were obtained. From these conformers, the 50
lowest energy conformers were fully optimized with B3LYP/6-
31G(d). Subsequently, frequency calculations were performed on
the 10 lowest energy B3LYP conformers. The global minimum from
this set was identified using B3LYP/6-311+G(2df,p) single-point
energies that included B3LYP/6-31G(d) zero-point vibrational energy
(ZPVE) corrections. The resulting gas-phase global minima were used
to obtain the corresponding solvent (water) phase structures from
IEF-PCM B3LYP/6-31G(d) optimizations. The lowest energy gas and
solvent phase structures were then used to determine the proton
affinity (PA) and the deglycosylation barrier of O-linked structures.
The B3LYP/6-311+G(2df,p) PA (including B3LYP/6-31G(d) ZPVE
corrections) at the N3 and N7 sites of the O-linked 8-dG adducts were
determined in the gas and solvent (water) phases as the negative of the
enthalpy change for protonation.42 To scrutinize the effects of the
phenoxyl moiety, Cl-substitution and protonation on sugar loss, the
deglycosylation reaction was investigated by altering and fixing the
glycosidic bond length (1′−N9) in 0.1 Å increments from 1.4 to 3.5 Å
in both the gas phase and water. All B3LYP calculations were
performed using Gaussian 09 revisions A.0254 and C.01.55
(10) Gates, K. S.; Nooner, T.; Dutta, S. Chem. Res. Toxicol. 2004, 17,
839−856.
(11) Hsieh, Y. S.; Chen, B. C.; Shiow, S. J.; Wang, H. C.; Hsu, J. D.;
Wang, C. J. Chem.−Biol. Interact. 2002, 140, 67−80.
(12) Suzuki, N.; Yasui, M.; Geacintov, N. E.; Shafirovich, V.;
Shibutani, S. Biochemistry 2005, 44, 9238−9245.
(13) Hovinen, J.; Glemarec, C.; Sandstrom, A.; Sund, C.;
̈
Chattopadhyaya, J. Tetrahedron 1991, 47, 4693−4708.
́
(14) Laayoun, A.; Decout, J.-L.; Lhomme, J. Tetrahedron Lett. 1994,
35, 4989−4990.
(15) Humphreys, W. G.; Kadlubar, K. K.; Guengerich, F. P. Proc.
Natl. Acad. Sci. U. S. A. 1992, 89, 8278−8282.
(16) Dipple, A. Carcinogenesis 1995, 16, 437−441.
(17) Patel, D. J.; Mao, B.; Gu, Z.; Hingerty, B. E.; Gorin, A.; Basu, A.
K.; Broyde, S. Chem. Res. Toxicol. 1998, 11, 391−407.
(18) Novak, M.; Ruenz, M.; Kazerani, S.; Toth, K.; Nguyen, T.-M.;
Heinrich, B. J. Org. Chem. 2002, 67, 2303−2308.
(19) Schlitt, K. M.; Sun, K. M.; Paugh, R. J.; Millen, A. L.; Navarro-
Whyte, L.; Wetmore, S. D.; Manderville, R. A. J. Org. Chem. 2009, 74,
5793−5802.
(20) Kikugawa, K.; Kato, T.; Kojima, K. Mutat. Res. 1992, 268, 65−
75.
(21) Dai, J.; Wright, M. W.; Manderville, R. A. J. Am. Chem. Soc.
2003, 125, 3716−3717.
(22) Kornyushyna, O.; Stemmler, A. J; Graybosch, D. M.; Bergenthal,
I.; Burrows, C, J. Bioconjugate Chem. 2005, 16, 178−183.
(23) Rogan, E. G.; Cavalieri, E. L.; Tibbels, S. R.; Cremonesi, P.;
Warner, C. D.; Nagel, D. L.; Tomer, K. B.; Cerny, R. L.; Gross, M. L. J.
Am. Chem. Soc. 1988, 110, 4023−4029.
(24) Kohda, K.; Tsunomoto, H.; Kasamatsu, T.; Sawamura, F.;
Terashima, I.; Shibutani, S. Chem. Res. Toxicol. 1997, 10, 1351−1358.
(25) Akanni, A.; Abul-Hajj, Y. J. Chem. Res. Toxicol. 1999, 12, 1247−
1253.
ASSOCIATED CONTENT
■
S
* Supporting Information
Figures S1−S5 and Table S1 described in the text, NMR
spectra of synthetic samples, and Tables S2−S37 (Cartesian
coordinates of global minimum structures for neutral, N7H+
7184
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185

The Journal of Organic Chemistry
Article
(26) Enya, T.; Kawanishi, M.; Suzuki, H.; Matsui, S.; Hisamatsu, Y.
Chem. Res. Toxicol. 1998, 11, 1460−1467.
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.;
Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador,
P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision C.01; Gaussian, Inc.: Wallingford, CT, 2010.
(27) Dai, J.; Wright, M. W.; Manderville, R. A. Chem. Res. Toxicol.
2003, 16, 817−821.
(28) Dai, J.; Sloat, A. L.; Wright, M. W.; Manderville, R. A. Chem. Res.
Toxicol. 2005, 18, 771−779.
(29) Vaidyanathan, V. G.; Villalta, P. W.; Sturla, S. J. Chem. Res.
Toxicol. 2007, 20, 913−919.
(30) Dahlmann, H. A.; Sturla, S. J. Eur. J. Org. Chem. 2011, 2987−
2992.
(31) Dumas, A.; Luedtke, N. W. J. Am. Chem. Soc. 2010, 132, 18004−
18007.
(32) Li, C.; Hoffman, M. Z. J. Phys. Chem. B 1999, 103, 6653−6656.
(33) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165−195.
(34) Hammett, L. P. Physical Organic Chemistry; McGraw-Hill Book
Co., Inc.: New York, 1940; p 206.
(35) Taft, R. W., Jr. In Steric Effects in Organic Chemistry; Newman,
M. S., Ed.; John Wiley & Sons, Inc.: New York, 1956; p 593.
(36) Charton, M. Can. J. Chem. 1960, 38, 2493−2499.
(37) Hammett, L. P. J. Am. Chem. Soc. 1937, 59, 96−103.
(38) Oivanen, M.; Lonnberg, H.; Zhou, X.-X.; Chattopadhyaya, J.
̈
Tetrahedron 1987, 43, 1133−1140.
(39) Rankin K. M. Ph.D. Thesis, University of Guelph, Guelph, ON,
December, 2012.
(40) Linnell, R. H. J. Org. Chem. 1960, 25, 290−290.
(41) Uno, B.; Iwamoto, T.; Okumura, N. J. Org. Chem. 1998, 63,
9794−9800.
(42) Greco, F.; Liguori, A.; Sindona, G.; Uccella, N. J. Am. Chem. Soc.
1990, 112, 9092−9096.
(43) Xia, F.; Xie, H.; Cao, Z. Int. J. Quantum Chem. 2008, 108, 57−
65.
(44) Liu, J.; Cao, S.; Jia, B.; Wei, D.; Liao, X.; Lu, J.; Zhao, Y. Int. J.
Mass Spectrom. 2009, 282, 1−5.
(45) Greisch, J.-F.; Leyh, B.; De Pauw, E. Eur. Phys. J. D 2009, 51,
89−96.
(46) Sagoo, S.; Beach, D. G.; Manderville, R. A.; Gabryelski, W. J.
Mass. Spectrom. 2011, 46, 41−49.
(47) Rankin, K. M.; Sproviero, M.; Rankin, K.; Sharma, P.; Wetmore,
S. D.; Manderville, R. A. J. Org. Chem. 2012, 77, 10498−10508.
(48) D’Angelo, N. D.; Peterson, J. J.; Booker, S. K.; Fellows, I.;
Dominguez, C.; Hungate, R.; Reider, P. J.; Kim, T.-S. Tetrahedron Lett.
2006, 47, 5045−5048.
(49) Zhang, Q.; Wang, D.; Wang, X.; Ding, K. J. Org. Chem. 2009, 74,
7187−7190.
(50) Yong, F.-F.; Teo, Y.-C.; Yan, Y.-K.; Chua, G.-L. Synlett 2012, 23,
101−106.
(51) Fujikawa, K.; Kondo, K.; Yokomichi, I.; Kimura, F.; Haga, T.;
Nishiyama, R. Agric. Biol. Chem. 1970, 34, 68−79.
(52) Chang, G.; Guida, W. C.; Still, W. C. J. Am. Chem. Soc. 1989,
111, 4379−4386.
(53) HyperChem 8.0.8; Hypercube, Inc.: Gainesville, FL, 2007.
(54) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, Jr., J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009.
(55) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
7185
dx.doi.org/10.1021/jo401122j | J. Org. Chem. 2013, 78, 7176−7185