Low-energy, low-yield photoionization, and production of 8-oxo-2a?2-deoxyguanosine and guanine from 2a?2-deoxyguanosine

Article abstract of DOI:10.1021/jp020649l

Experiments employing electron scavenging methods and high performance liquid chromatography with mass spectrometry detection indicate that electrons, formed via one-photon ionization, guanine (G) and small amounts of 8-oxo-2a?2-deoxyguanosine (8-oxo-dG)

Full text of DOI:10.1021/jp020649l

7704
J. Phys. Chem. B 2002, 106, 7704-7712
Low-Energy, Low-Yield Photoionization, and Production of 8-Oxo-2′-deoxyguanosine and
Guanine from 2′-Deoxyguanosine
George A. Papadantonakis,† Robert Tranter, Kenneth Brezinsky, Yanan Yang,
,‡
§
§
|
|
,†
Richard B. van Breemen, and Pierre R. LeBreton*
Department of Chemistry, UniVersity of Illinois at Chicago, Chicago, Illinois 60607, Department of Chemical
Engineering, UniVersity of Illinois at Chicago, Chicago, Illinois 60607, and Department of Medicinal
Chemistry and Pharmacognosy, College of Pharmacy, UniVersity of Illinois at Chicago,
Chicago, Illinois 60612
ReceiVed: March 7, 2002; In Final Form: May 28, 2002
Experiments employing electron scavenging methods and high performance liquid chromatography with mass
spectrometry detection indicate that electrons, formed via one-photon ionization, guanine (G) and small amounts
of 8-oxo-2′-deoxyguanosine (8-oxo-dG) are formed during 254 nm irradiation of deaerated alkaline
2
2
′-deoxyguanosine (dG) solutions containing N O. The G, 8-oxo-dG and electron yields vary in a similar
-
4
way when the pH changes. At a dG concentration of 1.2 × 10 M, the 254 nm photoionization quantum
yield is in the range 0.01 to 0.02 at pH 11.4. The low sensitivity with which photoionization can be monitored
in electron scavenging experiments does not permit the direct measurement of dG photoionization at near
neutral pH. At pH 6.3, the 254 nm photoionization quantum yield for dG is no larger than 0.003. At pH 11.4,
-
4
the yields of G and 8-oxo-dG formed from 254 nm irradiation of 1.2 × 10 M dG for 40 min are more than
5
× larger than that at pH 6.3. The similar pH dependence of the G, 8-oxo-dG and photoelectron yields, and
earlier reports linking G and 8-oxo-dG formation to dG photoionization provide evidence that, under the
present conditions, G and 8-oxo-dG quantum yields, which can be monitored with higher sensitivity than
hydrated electrons in electron scavenging experiments, parallel the photoionization quantum yield. From this
perspective, the wavelength dependence of G and 8-oxo-dG quantum yields indicates that dG ionization
threshold wavelengths at pH 11.4 and 6.3 are 266 ( 16 and 260 ( 16 nm, respectively. When the 260 nm
threshold of dG at pH 6.3 is adjusted to account for an approximately 0.5 eV lowering of the ionization
energy associated with incorporation of dG into native B DNA sequences containing multiple stacked guanines
(
Sugiyama, H.; Saito, I. J. Am. Chem. Soc. 1996, 118, 7063-7068. Zhu, Q.; LeBreton, P. R. J. Am. Chem.
Soc. 2000, 122, 12 824-12 834.), the present results provide evidence that low quantum-yield DNA
photoionization occurs near the short-wavelength cutoff (290 nm) of the ground-level solar spectrum.
Introduction
Earlier research focuses almost exclusively on products and
3
-5,11-13
mechanisms related to guanine photoionization.
There
The finding that UV radiation causes mutations and cancer,
and that, at ground level, the solar spectrum contains 0.3%
UV-B radiation in the range 280 to 320 nm has inspired interest
has been less experimental effort to evaluate threshold photon
energies for DNA photoionization in physiological environ-
4
,13,14
in the photochemistry of DNA.1,2 This is a rich chemistry in
ments, or even in simple aqueous solution.
which nucleic acids participate in many photochemical path-
ways, including reactions that involve direct DNA photoexci-
tation and reactions that are mediated by photosensitizers.
Threshold measurements are hindered by a combination of
factors, including the low photoionization quantum yields of
DNA components at physiological pH and the low sensitivity
with which aqueous photoionization is detected. In the photo-
ionization of guanine, and guanine containing nucleosides and
nucleotides with pulsed lasers, transient absorption spectroscopy
3
-5
Among carcinogenic DNA products associated with direct
UV-B excitation, a great deal of research has examined
cyclobutadipyrimidines and pyrimidine(6-4)pyrimidone photo-
adducts which are formed by reactions in the pyrimidine first
4
,13,15-18
has been used to monitor hydrated electrons.
Here, the
excited singlet (S1) and triplet (T1) states.1
,4,5b
Another family
high pulse power needed to detect low quantum-yield photo-
ionization has restricted the laser measurements primarily to
one or two-photon ionization at the easily accessible wave-
of reactions generally believed to be associated with wavelengths
4
below the UV-B range is initiated by direct photoionization
of DNA, and much work investigates the photoionization of
4
6-10
lengths of 266, 248, and 193 nm. At neutral pH, the one-photon
guanine (G), the DNA base with the lowest ionization energy.
1
93 nm ionization quantum yields of guanosine and 2′-
*
To whom correspondence should be addressed. E-mail: lebreton@
13
deoxyguanosine (dG) are 0.073. Experiments with continuous
Xe and Hg arc lamps frequently employ electron scavengers to
monitor photoionization. However, the low sensitivity and
uic.edu.
†
Department of Chemistry, University of Illinois at Chicago.
Current address: University of Chicago, Department of Biochemistry
‡
19
and Molecular Biology, Chicago, Illinois 60637.
inconvenience of electron scavenging experiments make it
impractical to use these methods to measure thresholds of the
low quantum yield associated with DNA photoionization.
§
Department of Chemical Engineering, University of Illinois at Chicago.
Department of Medicinal Chemistry and Pharmacognosy, College of
|
Pharmacy, University of Illinois at Chicago.
1
0.1021/jp020649l CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/17/2002

Production of 8-Oxo-2′-deoxyguanosine and Guanine
J. Phys. Chem. B, Vol. 106, No. 31, 2002 7705
-
One product from UV damage of DNA is 8-oxo-2′-deoxy-
For 5′-dGMP at neutral pH, results from these investigations
1
1,12,20
guanosine (8-oxo-dG), formed via photooxidation of dG.
yield an aqueous photoionization threshold energy of 4.9 ( 0.5
2
1
37
8
-Oxo-dG causes miscoding by DNA polymerase in vitro, is
eV.
22
mutagenic in bacterial and mammalian cells, and induces
guanine f thymine transversions that occur frequently in
mutated oncogenes and tumor suppressor genes,23 resulting in
The main goal of the present investigation is to employ a
tunable weak light source, with which one-photon photolysis
predominates, and high performance liquid chromatography
(LC) with mass spectrometric analysis to examine how changes
in pH and wavelength influence the yields of 8-oxo-dG and G
associated with dG photolysis. The experiments were carried
out under anaerobic conditions to reduce the number of potential
2
4
the implication of 8-oxo-dG in carcinogenic mechanisms. In
addition to 8-oxo-dG, other products associated with the
photooxidation of dG are 2,2-diamino-5-[2-deoxy-â-D-erythro-
pentofuranosyl]amino-5(2H)-oxazolone, 4-hydroxy-8-oxo-4,8-
dihydro-2′-deoxyguanosine and guanine.5
a,25,26
The relative
1
photolysis pathways by eliminating singlet O2( ∆g) photooxi-
yields of different dG photooxidation products depend on
dation. The dependence of 8-oxo-dG and G yields on pH was
compared to the dependence of the dG photoionization yield
on pH determined in electron scavenging experiments. The
dependence of the quantum yields for G and 8-oxo-dG formation
whether reaction is initiated by direct photoionization or by
photosensitizers and whether O2 is present.2
5,26
With photo-
sensitizers, dG oxidation proceeds via a dG photoionization
1
-
mechanism or via
a
singlet O2 ( ∆g) mediated
on wavelength was compared to dG and dG aqueous photo-
mechanism.3
,5,25,27-29
Interestingly, 5′-dGMP , itself, is a photo-
-
ionization thresholds obtained using gas-phase data and the
theoretical methods outlined above. These comparisons were
made in order to examine the relationship between dG photo-
ionization and mechanisms leading to the formation of 8-oxo-
dG and G.
1
30
sensitizer that produces O2 ( ∆g). Under some reaction
conditions, the same product is formed via more than one
mechanism. For example, in DNA, 266 nm, two-photon laser
experiments, 8-oxo-dG is formed via direct photoionization and
via reaction with OH radicals. Relative yields of 2′-deoxy-
guanosine reactions initiated by direct photoionization and by
photosensitizers also depend on structure. 8-Oxo-2′-deoxy-
guanosine yields from dG incorporated into DNA are larger
than from dG in solution.5
.
11
Experimental Section
Preparation of Solutions. 2′-Deoxyguanosine, G and 8-oxo-
dG were obtained from Sigma (St. Louis, MO) and were at
least 99% pure. Solutions were prepared in doubly distilled water
provided by a Corning Mega Pure System MP-1. The concen-
a,11
Guanine forms, with a low quantum yield (0.0002), when
-
dry guanosine-5′-monophosphate (5′-GMP ) is irradiated at 140
tration of all dG solutions used in photolysis experiments was
3
1
to 160 nm. Guanine formation also occurs in dG, guanosine,
-4
1
.2 × 10 M as determined by UV absorption measurements
-
and 2′-deoxyguanosine 5′-monophosphate (5′-dGMP ) solutions
with a Cary 14/OLIS spectrophotometer. Photolysis experiments
at pH 6.3 were carried out with unbuffered dG solutions. For
experiments at pH 11.4, 2 M NaOH was added dropwise. The
solutions were deaerated on a glass vacuum line, where each
solution was subjected to five freeze-pump-thaw cycles using
liquid N2.
26
irradiated at 254 nm. and in calf thymus DNA irradiated at
1
93 nm.32 Investigation of the mechanisms for G formation from
-
nucleosides, 5′-dGMP and DNA provides evidence that
26,32
photoionization of the guanine moiety is an initiating event.
At 254 nm, the quantum yield for G formation from 75 µM
guanosine was reported to decrease to a value of zero at pH
In all experiments, N2O (99.5% pure from AGA Gas,
Hammond, IN) was equilibrated with the dG solutions at 25
C. Prior to use, the N2O was passed through an Oxy-trap
2
6
1
1.9 a surprising result in view of the fact that the photoion-
-
ization quantum yields of dG, guanosine and 5′-GMP increase
°
(
with increasing pH.1
6,18a
Alltech Associates, Deerfield, IL) and then further purified on
At wavelengths longer than 193 nm, different experimental
measurements of the photoionization quantum yields of guano-
sine and dG at neutral pH have yielded different results. For
example, results from one transient-absorption, pulsed-laser
experiment indicate that at pH 7.0 the 266 nm one-photon
quantum yield for photoionization of 95 µM guanosine is
the vacuum line by three liquid N2 freeze-pump-thaw cycles.
Purified N2O was bubbled through the dG solutions, originally
39
at the vapor pressure of water, 21.08 Torr, until the final
pressure was 740 Torr. To ensure N2O saturation under these
conditions, the solution and the gas were frozen to 77 K and
then thawed to room temperature. This was repeated three times.
1
5
16,18a
0
.083. However, in other laser experiments
at 248 and
40
Using previously reported data, the concentration of the
-
266 nm, no one-photon ionization of dG, guanosine or 5′-GMP
dissolved N2O was estimated to be 0.02 M.
was observed near neutral pH for concentrations between 100
and 300 µM, whereas quantum yields between 0.01 and 0.07
were measured at pHs between 11.0 and 11.5. At neutral pH,
support for small quantum yields, significantly less than 0.083,
for photoionization between 248 and 266 nm is provided by
results from a steady-state measurement using a 254 nm Hg
arc lamp.15 For guanosine, this experiment provides a one-
photon quantum yield of 0.0058.
Irradiation of 2′-Deoxyguanosine Solutions. Steady-state
irradiation of deaerated dG solutions containing N2O was
performed with an ILC Technology (Sunnyvale, CA) 1000 W
Xe, Cermak-type lamp. The light was reflected on a concave,
Al(MgF2) coated mirror (100 mm focal length) from Oriel
Instruments (Stratford, CT) and directed to a McPherson
(Chelmsford, MA) Model 272, 0.2 m, f/2, grating monochro-
mator. To obtain detectable signals, the monochromator slits
were adjusted to provide a 32 nm bandwidth. The emerging
light beam was directed onto the sample contained in a custom-
made 8 mm × 12 mm o.d. Suprasil Quartz (Technical Glass
Products, Mentor, OH) cylindrical cuvette mounted on the
Gas-phase photoionization energies and conformations of
nucleotide bases, base pairs, stacked bases, nucleosides, and
mono-, di-, and oligonucleotides have been widely examined
using photoelectron, mass spectrometry and theoretical
7
,10,33-36
-4
2
methods.
Aqueous ionization energies have only been
vacuum line. The irradiation area was 1.3 × 10 m . The power
determined using complicated methods that involve the com-
bined use of gas-phase photoelectron data, results from elec-
tronic structure calculations at the SCF and post-SCF levels,
of the irradiating light on the sample was obtained from
4
1
actinometry measurements using 1.0 mL of potassium ferri-
oxalate in the sample cuvette. The power at different wave-
lengths is given in Table 1.
6
,37,38
and theoretical evaluations of free energies of hydration.

7706 J. Phys. Chem. B, Vol. 106, No. 31, 2002
Papadantonakis et al.
TABLE 1: Lamp Intensities in the Sample Cell
Results
wavelength
power
(watts)
Figure 1 shows results from LC analysis of samples of 2′-
deoxyguanosine irradiated at 254 nm for 40 min in 0.02 M N2O.
Results are given for experiments carried out at pH 11.4 (panel
A) and 6.3 (panel B). The figure also contains results from
samples that were not irradiated. At both pH values the figure
only shows portions of the chromatograms where dG, guanine
and 8-oxo-2′-deoxyguanosine elute. Retention times are given
above each chromatographic peak, and concentrations reported
in terms of mass ratios relative to dG are given in square
brackets.
In panels A and B, the G, 8-oxo-dG and dG results from the
irradiated solutions are given in the three chromatograms at the
top. The results from the nonirradiated solutions are given in
the three chromatograms on the bottom. Panel A demonstrates
that at pH 11.4, irradiation at 254 nm for 40 min results in
measurable yields of G and 8-oxo-dG. The G and 8-oxo-dG
mass ratios relative to dG are larger than the corresponding mass
ratios in the nonirradiated sample by 75 and 200 %, respectively.
The mass ratio of 8-oxo-dG is at least 10 times smaller than
that of G. The results in Panel B demonstrate that at pH 6.3 the
G and 8-oxo-dG yields are much smaller than at pH 11.4. At
pH 6.3, the G and 8-oxo-dG mass ratios for the irradiated sample
differ from those in the nonirradiated sample by less than 1 %.
The present results indicating that the yields of G and 8-oxo-
dG at pH 11.4 are at least 5 times larger than at pH 6.3 are
different from results obtained in earlier experiments26 employ-
ing higher radiation levels and smaller dG concentrations. There
it was found that the yield of G decreases with increasing pH.
(nm)
photons/second
1
1
1
1
3
3
4
5
-5
-5
-4
-4
2
40
54
66
80
5.61 × 10
7.85 × 10
5.45 × 10
1.05 × 10
4.64 × 10
6.13 × 10
4.07 × 10
7.45 × 10
2
2
2
N2 Analysis. In scavenging experiments, aqueous electrons,
aq, formed from photoionization yielded N2 via the reaction,
aq + N2O + H2O f N2 + OH + OH . After irradiation,
-
-
e
e
-
. 19d
N2 analysis of the gas above the dG solutions was carried out
at room temperature with an Agilent Technologies (Palo Alto,
CA) 6890 Series gas chromatograph equipped with a thermal
conductivity detector. 19091P-MS4 HP-PLOT MoleSieve 5A
and 19091J-413 HP-5 Cross-linked 5% PH ME siloxane
columns were used.
Liquid Chromatography-Mass Spectrometry Analysis.
On-line, high-performance liquid chromatography-mass spec-
trometry (LC-MS-MS), using positive ion electrospray ioniza-
tion, was carried out with a Waters (Milford, MA) 2690
chromatograph, a Waters 2487 UV detector, and a Micromass
Manchester, UK) Quattro II triple quadrupole mass spectrom-
eter. Nitrogen was used as a nebulizing gas with a flow rate of
.8 L/min, and the electrospray source was maintained at 140
C.
After irradiation, dG solutions at pH 11.4 were adjusted to
pH 7.0 by adding 5% acetic acid. Solutions irradiated at pH
.3 were analyzed without changing pH. Reverse-phase chro-
(
0
°
6
matographic separation of dG photolysis products was per-
formed on a YMC (Wilmington, NC) YAQ C18 (2.0 × 250
mm) column. The mobile phase consisted of methanol (solvent
A) and water (solvent B). The gradient varied linearly, and was
-2 min 94-90% B, 2-14 min 90% B, 14-14.5 min 90-
0% B, 14.5-23 min 80% B, 23-23.5 min 80-10% B and
3.5-26 min 10% B. The flow rate was 200 µL/min and the
4
4
Figure 2 contains results from N2O electron scavenging
experiments carried out at 240 nm. The figure shows chromato-
-
4
grams from N2 analysis of the gas above 1.2 × 10 M dG
solutions containing N2O (0.02 M) that were irradiated for 19
h. Each panel gives results from an irradiated sample and from
a sample that was not irradiated. The amount of N2 obtained in
an experiment in which a sample of distilled water containing
N2O at pH 11.4 without dG was irradiated for 19 h at 240 nm
was the same as that obtained from a sample that was not
irradiated.
0
8
2
injection volume was 25 µL.
′-Deoxyguanosine, which was the largest component, was
2
detected by UV absorbance at 260 nm. Guanine was detected
by mass spectrometry using selective ion recording at m/z 152.0
in the positive ion mode. Calibration curves for dG and G were
obtained from standard aqueous solutions that were analyzed
and detected by UV absorbance and MS, respectively. In
chromatograms of samples prepared in the same manner but
measured on different days the corresponding areas associated
with the dG absorbance, and the G ion signal differed by less
than 10%.
The left panel of Figure 2 shows data from N2 analysis of a
dG solution irradiated at pH 11.4. The right panel shows data
from a solution irradiated at pH 6.3. The results indicate that
the yield of hydrated electrons produced during photolysis of
dG at pH 11.4, as indicated by the N2 yield, is more than at pH
6.3. Furthermore, measurements of UV absorption spectra at
254 nm as a function of irradiation time indicate that the
photodestruction of dG at pH 11.4 is greater than at pH 6.3, in
8
-Oxo-2′-deoxyguanosine was monitored during HPLC by
using tandem mass spectrometry with collision-induced dis-
sociation (CID) and product ion scanning. The protonated
molecule of m/z 284 was selected as the precursor ion and the
product ion was m/z 168.42 CID was carried out using Ar as
1
5
agreement with earlier results. For example, the absorbance
at 254 nm decreases from 1.64 to 1.04 after 19 h of irradiation
at pH 6.3, whereas at pH 11.4 the decrease is from 1.28 to 0.36.
The increase in the 240 nm one-photon ionization quantum yield
of dG that occurs with increasing pH, indicated by the results
in Figure 2, is similar to that reported earlier from solvated
electron transient absorption measurements obtained in 266 and
248 nm pulsed laser experiments on dG and other guanine
-3
the collision gas at a pressure 1.5 × 10 mTorr and a collision
energy of 18 eV. 8-Oxo-2′-deoxyguanosine was quantified using
an internal standard. 10.0 µL of isotopically labeled 8-oxo-dG
1
3
15
(
[ C10, N5]-8-oxo-2′-deoxyguanosine, at a concentration of
7
0 ppb were added to 60.0 µL of each sample. The internal
1
6,18
standard was synthesized using a previously reported proce-
dure.43 The equation relating the ratio between the areas of the
ion signals associated with the labeled and unlabled 8-oxo-dG
to the concentration of unlabeled 8-oxo-dG was determined in
standardization experiments. These were carried out using
solution conditions employed in the present investigations with
known quantities of labeled and unlabled 8-oxo-dG. The
derivatives.
Figure 3 shows the mass concentration (ppm) of 8-oxo-dG
-4
and G produced when 1.2 × 10 dG solutions at pH 11.4 were
irradiated at 254 nm for times between 0 and 18 h. The figure
also shows how the dG mass concentration changes with time.
After 9 h, the concentration of dG has decreased 19-fold. The
maximum 8-oxo-dG concentration, which is more than 45×
smaller than the maximum G concentration, occurs after 1 h of
4
2
resulting equation was the same as that reported earlier.

Production of 8-Oxo-2′-deoxyguanosine and Guanine
J. Phys. Chem. B, Vol. 106, No. 31, 2002 7707
Figure 1. Chromatograms showing results from guanine (G), 2′-deoxyguanosine (dG) and 8-oxo-2′-deoxyguanosine (8-oxo-dG) analysis of deaerated
-
4
samples of 1.2 × 10 M dG irradiated at 254 nm for 40 min in 0.02 M N
Panel A shows results from a dG sample irradiated at pH 11.4. Panel B show results at pH 6.3. Each of the panels show three chromatograms
above) obtained from an irradiated sample and three chromatograms (below) obtained from a blank, nonirradiated sample. The figure shows
2
O. Analytes were detected by UV absorbance, MS or MS/MS. See text.
(
retention times, given above the chromatographic peaks, and mass ratios relative to dG, given in square brackets. Uncertainties in the mass ratios
represent the standard deviations of results from nine analyses obtained from three irradiation experiments.
irradiation. In contrast, the maximum G concentration occurs
after 6 h, indicating that, at pH 11.4, 8-oxo-dG undergoes 254
nm photodamage more rapidly than G. The inset in Figure 3
shows an N2 chromatogram obtained after 254 nm irradiation
for 9 h in an N2O electron scavenging experiment carried out
with 1.2 × 10 M dG. The results in the inset indicate that,
over the time period between 0 and 9 h, when the dG
concentration is highest, significant photoionization occurs.
from samples irradiated for 40 min at wavelengths of 280, 266,
and 254 nm.
As Figure 1 indicates, significant levels of G and 8-oxo-dG
were not observed after 40 min of irradiation at pH 6.3.
However, G and 8-oxo-dG were observed in solutions irradiated
for longer times. The lower panel of Figure 4 shows data
obtained at pH 6.3 from samples irradiated for 9 h at the same
wavelengths as in the upper panel, and at 240 nm. For the
different irradiation wavelengths used, the levels of dG depletion
that occurred over 40 min in the pH 11.4 experiments, and over
9 h in the pH 6.3 experiments resulted in absorbance changes
at the radiation wavelengths of less than 9 and 26%, respec-
-
4
Figure 4 shows quantum yields for formation of guanine and
-
4
8
-oxo-2′-deoxyguanosine from irradiated samples of 1.2 × 10
M dG measured at different wavelengths. The ordinates on the
left and right give quantum yields for formation of G and 8-oxo-
dG, respectively. The upper and lower panels show results
obtained at pH 11.4 and 6.3. At pH 11.4, data were obtained
4
5
tively. In Figure 4, the quantum yields reported at pH 6.3 are
the average from the analysis of six samples obtained in two

7708 J. Phys. Chem. B, Vol. 106, No. 31, 2002
Papadantonakis et al.
Figure 2. N
2
2
gas chromatograms measured after 240 nm irradiation of 2′-deoxyguanosine for 19 h in 0.02 M N O at pH 11.4 (left panel) and at
pH 6.3 (right panel). The left and right panels also contain blank N
2
chromatograms obtained from samples containing 0.02 M N
2
O without dG,
irradiated for 19 h at pH 11.4 and 6.3, respectively. The small difference in the N
flow pressure of the He carrier gas.
2
retention times in the two panels is due to a difference in the
Figure 3. Concentrations of 2′-deoxyguanosine (b), 8-oxo-2′-deoxyguanosine (9) and guanine (2) obtained after 254 nm irradiation of 1.2 ×
-
4
1
0
M 2′-deoxyguanosine at pH 11.4 for varying times. The dG, G and 8-oxo-dG concentration scales are given on the left ordinate, on the right
ordinate, and on the right ordinate inside the border, respectively. Concentrations of G and 8-oxo-dG associated in blank nonirradiated samples
were subtracted from concentrations in irradiated samples. The inset shows an N gas chromatogram measured after 9 h of 254 nm irradiation of
O at pH 11.4. The inset also shows a chromatogram of a blank containing 0.02 M N O at pH 11.4 measured after
h of 254 nm irradiation. A similar blank chromatogram was obtained from a nonirradiated sample of dG in 0.02 M N O at pH 11.4.
2
-
4
1
.2 × 10 M dG in 0.02 M N
2
2
9
2
irradiation experiments. Those reported at pH 11.4 are the
average from nine samples obtained in three irradiation experi-
ments. The error bars represent the standard deviations.
The inset in the upper panel of Figure 4 indicates how the
absorbance (A) of dG before irradiation at pH 11.4 and the lamp
intensity (B) change with wavelength. The quantum yields were
obtained by dividing concentrations of 8-oxo-dG and G
measured by the number of photons absorbed,46 and have not
been corrected for the dG or photoproduct damage that occurs
during the course of an experiment.45 The 8-oxo-dG and G
product concentrations were determined by subtracting back-
ground mass concentrations measured in nonirradiated samples
from mass concentrations measured in irradiated solutions.
Consideration was given to the possibility that the 8-oxo-dG
and G formed in the wavelength region examined in these
experiments was associated with the low monochomator resolu-
tion employed. For example, the low resolution might give rise
to low-intensity, low-wavelength (<200 nm) photons that occur
in the tail of the light distributions obtained with the mono-
chomator settings employed in Figure 4. It might then be that
these high-energy photons are responsible for the products
observed. However, the observation that the total 8-oxo-dG
signal decreases as the monochomator settings are reduced from
254 nm indicates that this is not the case for 8-oxo-dG
production. At pH 6.3, the 8-oxo-dG/dG mass ratios after 9 h
-
10
irradiation with 254 and 240 nm photons were 2.46 × 10

Production of 8-Oxo-2′-deoxyguanosine and Guanine
J. Phys. Chem. B, Vol. 106, No. 31, 2002 7709
-
4
Figure 4. Average quantum yields of 8-oxo-2′-deoxyguanosine (b) and guanine (2) formed from 1.2 × 10 M 2′-deoxyguanosine measured as
a function of wavelength. The upper and lower panels show results obtained at pH 11.4 and 6.3, where the radiation times were 40 min and 9 h,
respectively. Ordinates on the left give guanine quantum yields; ordinates on the right give 8-oxo-2′-deoxyguanosine quantum yields. The inset in
the upper panel shows the absorption spectrum of dG measured at pH 11.4 (A) and the lamp intensity (B) as a function of wavelength. Arrows
within the borders of each panel give estimates of experimental photoionization threshold energies of anionic and neutral dG obtained from the
data. Arrows below the upper and lower panels give ionization energies of anionic and neutral dG obtained by employing a combination of gas-
phase photoelectron data, and theoretically calculated gas-phase ionization potentials and aqueous solvation energies. See text and refs 37 and 38b.
and 1.66 × 10-10, respectively. In Figure 4, the increase in the
nm. These increases in the quantum yields provide evidence of
the existence of thresholds for photochemical production of
8-oxo-dG and G at wavelengths in the region 280-250 nm.
Small quantum yields were measured at 280 nm for 8-oxo-
dG and G formation at pH 11.4 and 6.3. However, these are
negligible on the scale at which Figure 4 is drawn, and their
occurrence may be associated with the large band-pass of the
monochromator ((16 nm) that is needed to obtain measurable
signals. The 254 nm quantum yield for G formation at pH 6.3,
8
-oxo-dG quantum yield with decreasing wavelength is due to
the decrease in light intensity that occurs as the wavelength
decreases. See curve B of the inset. The results from N2O
electron scavenging experiments carried out at pH 11.4 also
indicate that photoionization is not occurring exclusively from
high-energy photons in the tail of the light distribution. Here,
the amount of N2 formed decreased more than 3-fold when the
wavelength was reduced from 254 to 225 nm.
The results in both panels of Figure 4 show that at the longer
wavelengths the quantum yields of 8-oxo-dG and G are near
zero and constant. However, at both pH 11.4 and 6.3, the
quantum yields increase significantly at wavelengths below 266
-
6
reported in Figure 4 (6.02 ( 0.08 × 10 ), is similar to the 254
-6
nm quantum yield (3 × 10 ) for guanine release from double-
stranded calf-thymus DNA in aerated solution at pH 7.4
However, this value is smaller than expected, based on the

7710 J. Phys. Chem. B, Vol. 106, No. 31, 2002
Papadantonakis et al.
-
4
quantum yield (5.05 ( 0.09 × 10 ) reported earlier from 1.05
It is important to acknowledge that over the 19 h irradiation
time required to obtain the electron scavenging results in Figure
2, or the 9 h used to obtain the data in the inset of Figure 3,
photoionization can occur not only from dG but from dG
photoproducts such as G and 8-oxo-dG. Nevertheless, the results
in Figures 1 and 2, as well as those from pulsed-laser
-
4
16
×
10 M dG at neutral pH. Although the source of the
discrepancy between the present and earlier results is uncertain,
it may be related to the different irradiation procedures
employed. Mononchomator selected radiation with a wide
bandwidth, and a lamp, which has sharply decreasing intensity
between 270 and 240 nm were used in the present experiments.
The 254 nm line from an unfiltered low-pressure Hg lamp was
used in the earlier experiments. 2
1
6,18
experiments,
support the conclusion that the quantum yield
for photoionization of dG, as well as the quantum yields for G
and 8-oxo-dG production increase with increasing pH.
Detailed photochemical mechanisms for the formation of
8-oxo-dG and G from dG and guanosine are not yet available.
6, 15
The long radiation times required to obtain the results in
Figure 4, especially at pH 6.3, and the low levels of G and
-oxo-dG observed, which are near the detection limit, made it
2
6
32
8
For G release from guanosine and from calf thymus DNA,
impossible to carry out experiments at a sufficient number of
wavelengths to determine the thresholds with high precision.
Within this limitation, the results provide similar estimated
thresholds for G and 8-oxo-dG with values of 266 ( 16, and
it is proposed that the principal mechanisms are initiated by
photoionization. Results from investigations of G formation from
guanosine demonstrate that the relative importance of different
mechanisms initiated by photoionization changes at different
pH’s. For guanosine at neutral pH without N2O, the finding
that the addition of electron scavengers reduces the yield of G
supports the conclusion that the mechanism is initiated by
2
60 ( 16 nm at pH 11.4 and 6.3, respectively. These estimates
are indicated by arrows within the borders of the upper and
lower panels of the figure.
2
6
capture of an electron by neutral guanosine
Discussion
+
.
-
Figure 1 shows that guanine and 8-oxo-dG are easily observed
Guanosine + hν f Guanosine + e
aq
-
4
products formed at pH 11.4 when dG (1.2 × 10 M) is
irradiated for 40 min at 254 nm. At pH 6.3, the 8-oxo-dG and
G occurring after 40 min of irradiation are near background
levels. While 8-oxo-dG and G, formed at pH 6.3, are observed
at longer radiation times (9 h, Figure 4), the results in Figure 1
indicate that their quantum yields are small near physiological
pH, but increase significantly with increasing pH.
The low sensitivity of N2O electron scavenging experiments
does not permit the detection of photoionization products at the
short radiation time (40 min) used to obtain the results in Figure
Guanosine + e f Guanosine-. f f G
-
aq
With N2O, the proposed mechanism for G formation from
guanosine and DNA involves reaction of ground-state, closed-
shell guanosine or DNA with OH radicals formed via N2O
scavenging of guanosine or DNA photoelectrons.
Reactions of hydroxyl radicals formed via N2O scavenging
of dG photoelectrons may also account for the formation of
Alternatively, 8-oxo-dG may be produced via the
mechanism shown below that occurs through hydration of the
.
26,32
25,48
8-oxo-dG.
1
; however, the results in Figure 2, obtained after 19 h of
radiation, exhibit pH dependence similar to the 8-oxo-dG and
G yields. These results demonstrate that the 240 nm photoion-
ization yield at pH 6.3 is more than 4 times smaller than that at
pH 11.4. The results in the inset in Figure 3 obtained at 254
nm also demonstrate the occurrence of one-photon ionization
at high pH. When the 254 nm experiment in the inset in Figure
11,12
radical cation formed from direct photoionization of dG.
Although there is consensus that in DNA photolysis the
formation of 8-oxo-dG proceeds in this way,
5,11,12,27,28,49
there
are different views about the significance of this mechanism in
11-13,27,50,51
the photolysis of free dG in solution.
Assessment of
the mechanism in free dG is made difficult by the small 8-oxo-
3
was carried out at pH 6.3 the result was the same at that in
Figure 1. Nitrogen formation was negligible. Consideration of
11
dG yield compared to that in native DNA
1
6
results from laser experiments at 248 and 266 nm with dG
and guanosine at concentrations between 1.2 and 1.4 × 10
1
8a
-4
M and at pH 11.0 to 11.5 indicates that, at pH 11.4, the 254
nm photoionization quantum yield of dG is in the range 0.01
to 0.02. On the basis of this observation, the present electron
scavenging results indicate that, at pH 6.3, the 254 nm quantum
yield for dG photoionization is smaller than 0.003.
The increase in the dG photoionization quantum yield with
increasing pH correlates with the decrease in the dG rate
constant for electron scavenging. At natural pH, the dG
9
-1 -1
scavenging rate constant (k ) 6.0 × 10 M s ) is similar to
9
-1 -1 47
the N2O rate constant (k ) 8.7 × 10 M s ). If dG behaves
like G, the rate constant at pH 11.4 is significantly smaller. At
pH 7.0 and 11.0 the rate constants for G electron scavenging
are 1.2 × 10 and 2 × 10 M s , respectively. The large
guanosine and dG electron scavenging rate constants at natural
pH may account, in part, for the large difference in photoion-
ization quantum yields reported from different 266 nm laser
Doubt about the involvement of the dG radical cation in the
formation of 8-oxo-dG from free dG is provided by guanosine
pulsed radiolysis experiments employing a high energy γ source
and analysis involving LC with electrochemical detection.50 In
pulsed radiolysis of 1 mM guanosine, where the guanosine
radical cation was formed by employing the dibromide radical
or thallium (II), no evidence of the hydration reaction was
observed. An upper limit of the rate constant for hydration of
the radical cation or of the deprotonated cation was estimated
1
0
9
-1 -1
47
1
5,16
experiments
(
that were carried out at different concentrations
-
4
0.75 versus 1.2 × 10 M). Because of guanosine and dG
electron scavenging, the apparent quantum yields are expected
to decrease with increasing concentration both in transient
absorption experiments and in N2O electron scavenging experi-
ments.
-
1 50
to be 0.1 s .
Differences between reports about the importance of the dG
radical cation for 8-oxo-dG formation have been attributed, in

Production of 8-Oxo-2′-deoxyguanosine and Guanine
J. Phys. Chem. B, Vol. 106, No. 31, 2002 7711
part, to different wavelengths used in different UV photolysis
experiments, to different photooxidation conditions in UV
photolysis versus pulsed radiolysis experiments, and to the ease
event that will become more important if or when ground-level
solar UV intensities increase.
12
Conclusions
with which 8-oxo-dG undergoes secondary oxidation. Support
for an important role of the radical cation in formation of 8-oxo-
dG is provided by the observation that, in two-photon 248 nm
laser experiments carried out at 77 K in NaClO4 glasses, 8-oxo-
dG production from DNA and dG exhibit similar dose depen-
dence.12 On the other hand, two-photon 266 nm experiments at
room temperature indicate that the 8-oxo-dG yield from native
DNA in solution is approximately 100 times larger than that
from free dG.1 Nevertheless, in the two-photon 266 nm solution
experiments, 8-oxo-dG formation from free dG persists in the
The main conclusions of this investigation are as follows:
1
. The 254 nm irradiation of deaerated 2′-deoxyguanosine
solutions containing N2O at pH 11.4 gives rise to easily
monitored photoionization, as well as to the formation of
guanine and 8-oxo-2′-deoxyguanosine, all of which occur via
one-photon processes. 8-Oxo-2′-deoxyguanosine is a minor
tp hh aot t oo pf r Go d. uct which has a yield that is at least 10× smaller than
2. The yields of formation of 8-oxo-dG and G, as well as the
quantum yield for dG photoionization increase with increasing
pH. At pH 11.4, yields of G and 8-oxo-dG and the quantum
yield for photoionization are 4 or more times larger than at pH
1
.
presence of an OH scavenger, supporting the conclusion that
approximately 25% of the total 8-oxo-dG yield from free dG is
11
formed via hydration of the radical cation.
6
.3.
3
Uncertainty in the detailed mechanisms associated with G
and 8-oxo-dG formation from dG in the present experiments
adds uncertainty to interpretation of the results in Figure 4.
Nevertheless, the similar pH dependence of the quantum yields
for photoionization, and for G and 8-oxo-dG formation (Figures
. The quantum yields for G and 8-oxo-dG exhibit threshold
behavior with estimated thresholds of 266 ( 16 and 260 ( 16
nm at pH 11.4 and 6.3, respectively. These threshold values
are similar to ionization energies of dG and dG estimated from
gas-phase photoelectron data and theoretical calculations.
-
1
and 2), and the strong implication, from earlier work, that
4
. The similarity between the pH dependence of 8-oxo-dG,
photoionization initiates G and 8-oxo-dG formation provide
evidence that the threshold behavior exhibited in Figure 4
represents the threshold of the dG photoionization quantum
yield. This reasoning leads to the conclusion that, under the
present experimental conditions, G and 8-oxo-dG formation are
photoionization markers even when, as in the case of measure-
ments at pH 6.3, the low sensitivity of N2O scavenging prevents
the direct detection of electrons.
G and photoelectron yields, as well as the observance of
thresholds for 8-oxo-dG and G, near those expected for dG
photoionization support the conclusion that, under the present
experimental conditions, 8-oxo-dG and G formation is initiated
by dG photoionization. A consideration of the lowering of the
guanine ionization energy by approximately 0.5 eV that occurs
in G runs in DNA indicates that in physiological environments,
low quantum-yield ionization of guanine occurs at wavelengths
near the UV cutoff of the ground-level solar spectrum.
In Figure 4, the experimental thresholds for 8-oxo-dG and G
formation at pH 11.4 (4.7 ( 0.3 eV) and pH 6.3 (4.8 ( 0.3
eV) are similar to aqueous photoionization threshold energies
Acknowledgment. Support by the American Cancer Society
-
of anionic dG at pH 11.4 (4.5 ( 0.5 eV) and of neutral dG at
(RPG-91-024-09-CNE), the National Cancer Institute (CA
pH 6.3 (4.9 ( 0.5 eV) obtained via a previously described
7
0771), and the Research Board of the University of Illinois is
37,38b
combination of gas-phase data and theoretical methods.
The
gratefully acknowledged. The authors had helpful discussions
with Drs. Annette Bernas and Dora Grand (Universit e´ Paris-
Sud, Laboratoire de Chimie Physique), and Prof. David Crich
(UIC). We would also like to thank Dr. Emi Ikeda for assistance
in preliminary N O scavenging experiments and Mr. John Costa
for technical help. PRL thanks Drs. Ronald Harvey and Marsha
Rosner of the Ben May Institute of the University of Chicago
for hospitality during the sabbatical on which this manuscript
was prepared.
theoretical thresholds are given by the arrows below the
-
horizontal axes in Figure 4. The finding that the theoretical dG
and dG ionization thresholds are within the range of uncertainty
associated with the experimental thresholds for 8-oxo-dG and
G formation provides further evidence that, under the experi-
mental conditions employed here, 8-oxo-dG and G are markers
of dG photoionization.
2
If the 8-oxo-dG and G results in Figure 4 represent thresholds
for dG photoionization, the results lead to an interesting
conclusion about the energetics of guanine photoionization in
native double-stranded DNA. The data in Figure 4 indicate that
the photoionization threshold of free dG occurs at approximately
References and Notes
(1) (a) Ananthaswamy, H. N.; Pierceall, W. E. Photochem. Photobiol.
1990, 52, 1119. (b) Sage, E. Photochem. Photobiol. 1993, 57, 163.
(
(
2) Frederick, J. E. Photochem. Photobiol. 1993, 57, 175.
3) Cadet, J.; Vigny, P. In Bioorganic Photochemistry, Vol. 1,
260 nm at pH 6.3, and that photoionization of free dG is not an
environmentally significant event induced by solar radiation
because the solar spectrum at ground level has no contribution
Photochemistry of the Nucleic Acids; Morrison H., Ed.; Wiley: New York,
1990, Chapter 1.
2
(4) G o¨ rner, H. J. Photochem. Photobiol. B: Biol. 1994, 26, 117.
(
from wavelengths shorter than 290 nm. However, in native
5) (a) Cadet, J.; Berger, M.; Douki, T.; Ravanat, J.-L. ReV. Physiol.
DNA the situation may be different. Here, recent computational
Biochem. Pharmacol. 1997, 65, 1. (b) Cadet, J. In DNA Adducts:
Identification and Biological Significance; Hemminki, K., Dipple, A.,
Shuker, D. E. G., Kadlubar, F. F.; Segerb a¨ ch, D,; Bartsch, H., Eds.; IARC
Publication No. 125; Int. Agency for Res. on Cancer: Lyon, 1994; pp 245-
investigations9
,10,33
indicate that hydrogen-bonding and base-
stacking interactions, which occur in sequences containing
multiple repeating guanines (G runs), have threshold ionization
energies that are 0.5 to 0.7 eV smaller than threshold energies
of free guanine. Applying this lowering of the ionization energy
to the present data for free dG, leads to a threshold wavelength
of 290 nm or longer for guanine photoionization in G runs,
within the UV-B range occurring in the solar spectrum. The
present data surprisingly indicate that low quantum-yield DNA
photoionization may be a ubiquitous environmentally induced
276.
(6) Fernando, H.; Papadantonakis, G. A.; Kim, N. S.; LeBreton, P. R.
In Molecular Modeling of Nucleic Acids; Leontis, N.; SantaLucia, J., Jr.,
Eds.; ACS Symposium Series 682; American Chemical Society: Wash-
ington, DC, 1997; p 18.
(7) (a) Urano, S.; Yang, X.; LeBreton, P. R. J. Mol. Struct. 1989, 214,
3
15. (b) Orlov, V. M.; Smirnov, A. N.; Varshavsky, Ya., M. Tetrahedron
Lett. 1976, 4377. (c) Hush, N. S.; Cheung, A. S.; Chem. Phys. Lett. 1975,
4, 11.
(8) Steenken, S. Chem. ReV. 1989, 89, 503.
3

7712 J. Phys. Chem. B, Vol. 106, No. 31, 2002
Papadantonakis et al.
(
9) Zhu, Q.; LeBreton, P. R. J. Am. Chem. Soc. 2000, 122, 12 824.
(35) Padva, A.; Dinerstein, R. J.; Ridyard, J. N. A.; LeBreton, P. R.
Biochem. Biophys. Res. Commun. 1974, 60, 1262.
(
(
10) Sugiyama, H.; Saito, I. J. Am. Chem. Soc. 1996, 118, 7063.
11) Angelov, D.; Spassky, A.; Berger, M.; Cadet, J. J. Am. Chem. Soc.
997, 119, 11 373.
12) Cullis, P. M.; Malone, M. E.; Merson-Davies, L. A. J. Am. Chem.
Soc. 1996, 118, 2775.
(
36) Dolgounitcheva, O.; Zakrzewski, V G.; Ortiz, J. V. J. Am. Chem.
Soc. 2000, 122, 12 304.
37) Fernando, H.; Papadantonakis, G. A.; Kim, N. S.; LeBreton, P. R.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 5550.
38) (a) Kim, N. S.; LeBreton, P. R. J. Am. Chem. Soc. 1996, 118, 3694.
b) Kim, N. S.; Zhu, Q.; LeBreton, P. R. J. Am. Chem. Soc. 1999, 121,
1 516.
39) Handbook of Chemistry and Physics, 78th ed.; CRC Press: Boca
1
(
(
(
(
13) Candeias, L. P.; Steenken, S. J. Am. Chem. Soc. 1992, 114, 699.
14) von Sonntag, C. The Chemical Basis of Radiation Biology; Taylor
(
(
1
Francis: London, 1987; p 9.
15) Crespo-Hern a´ ndez, C. E.; Flores, S.; Torres, C.; Negr o´ n-Encarna-
ci o´ n, I.; Arce, R. Photochem. Photobiol. 2000, 71, 534.
16) Papadantonakis, G. A.; Stevenson, K. L.; LeBreton, P. R. Chem.
Phys. Lett. 2001, 346, 97.
17) Arce, R.; Rivera, J. J. Photochem. Photobiol. A: Chem. 1989, 49,
19.
18) (a) Song, Q.; Yao, S.; Wang, W.; Lin, N. J. Photochem. Photobiol.
A: Chem. 1997, 102, 197. (b) Song, Q.; Yao, S.; Lin, N. Radiat. Phys.
Chem. 1997, 50, 435.
19) (a) Feitelson, J. Photochem. Photobiol. 1971, 13, 87. (b) Amouyal,
E.; Bernas, A.; Grand, D. Photochem. Photobiol. 1979, 29, 147. (c) Zechner,
J.; K o¨ hler, G.; Getoff, N.; Tatischeff, I.; Klein, R. Photochem. Photobiol.
981, 34, 163. (d) Hart, E. J.; Anbar, M. The Hydrated Electron; Wiley-
Interscience: New York, 1970; p 109.
20) (a) Podmore, I. D.; Cooper, D.; Evans, M. A.; Wood, M.; Lunec,
J. Biochem. Biophys. Res. Comm. 2000, 277, 764. (b) Bartolini, W. P.;
Johnston, M. V. J. Mass Spectrom. 2000, 35, 408.
21) Shibutani, S.; Takeshita, M.; Grollman, A. P. Nature 1991, 349,
31.
22) Cheng, K. C.; Cahill, D. S.; Kasai, H.; Nishimura, S.; Loeb, L. A.
J. Biol. Chem. 1992, 267, 166.
23) Husain, S. P.; Harris, C. C. Cancer Res. 1998, 58, 4023.
24) (a) Tommasi, S.; Denissenko, M. F.; Pfeifer, G. P. Cancer Res.
997, 57, 4727. (b) Shigenaga, M. K.; Hagen, T. M.; Ames, B. N. Proc.
Natl. Acad. Sci. U.S.A. 1994, 91, 10 771.
25) Ravanat, J.-L.; Remaud, G.; Cadet, J. Arch. Biochem. Biophys. 2000,
74, 118.
26) Crespo-Hern a´ ndez, C. E.; Arce, R. Photochem. Photobiol. 2000,
1, 544.
27) Kasai, H.; Yamaizumi, Z.; Berger, M.; Cadet, J. J. Am. Chem. Soc.
992, 114, 9692.
28) Melvin, T.; Cunniffe, S.; Papworth, D.; Roldan-Arjone, T.; O’Neill,
P. Photochem. Photobiol. 1997, 65, 660.
29) Cadet, J.; Berger, M.; Buchko, G. W.; Joshi, P. C.; S e´ bastien, R.;
Ravanat, J.-L. J. Am. Chem. Soc. 1994, 116, 7403.
30) Mohammad, T.; Morrison, H. J. Am. Chem. Soc. 1996, 118, 1221.
31) Dodonova, N. Ya.; Kiseleva, M. N.; Remisova, L. A.; Tsyganenko,
N. M. Photochem. Photobiol. 1982, 35, 129.
32) Gurzadyan, G. G.; 5. G o¨ rner, H. Photochem. Photobiol. 1992, 56,
71.
33) Hutter, M.; Clark, T. J. Am. Chem. Soc. 1996, 118, 7574.
34) Gidden, J.; Bushnell, J. T.; Bowers, M. T. J. Am. Chem. Soc. 2001,
23, 5610.
(
(
Raton, FL, 1997; pp 6-8.
(40) Wilhelm, E.; Battino, R.; Wilcock, R. J. Chem. ReV. 1997, 77, 2219.
(41) Hatchard, C. G.; Parker, C. A. Proc. R. Soc. 1956, A235, 518.
(
(
2
(
42) Hua, Y.; Wainhaus, S. B.; Yang, Y.; Shen, L.; Xiong, Y.; Xu, X.;
Zhang, F.; Bolton, J.; van Breemen, R. B. J. Am. Soc. Mass Spectrom.
001, 12, 80.
43) Frenkel, K.; Zhong, Z.; Wei, H.; Karkoszka, J.; Patel, U.; Rashid,
K.; Georgescu, M.; Solomon, J. 1991, J. Anal. Biochem. 1991, 196, 126.
44) Differences between results presented here and in refs 15 and 26
(
2
(
(
(
may be related to the lower fluence used in the present experiments. Here,
the longer path length and the monochromator give rise to a 254 nm intensity
1
13
of 7.85 × 10 photons/second. In a 40 min experiment, the fluence is 1132
(
-
2
J m . The experiments of refs 15 and 26 employed a low-pressure Hg
lamp located 1 cm from the sample. Evidence of the higher fluence
employed in the earlier experiments is indicated by an examination of the
dG absorption spectra of samples measured before and after irradiation. In
results in Figure 1 of ref 15 obtained from experiments carried out at natural
pH, the UV absorption spectrum of dG exhibits more change after 28 min
of 254-nm irradiation than after 9 h of irradiation in the present experiments.
Furthermore, examination of the pH dependence in ref 26 was carried out
(
4
(
(
(
-
5
at a smaller dG concentration (7.5 × 10 M) than the present experiments
1
-
4
(
1.2 × 10 M), and in an N2 saturated solution without N2O.
(
(45) For experiments at pH 11.4, the initial absorbances at 280, 266
3
7
1
and 254 nm were 0.94, 1.43 and 1.38. The final absorbances were 0.94,
1.31 and 1.34, respectively. For experiments at pH 6.3, the initial
absorbances at 280, 266, 254 and 240 nm were 1.03, 1.27, 1.64, and 1.18.
The final absorbances were 1.02, 1.26, 1.22 and 0.91.
(46) The number of absorbed photons was approximately evaluated by
employing the absorbance at the irradiation wavelength of the dG sample
at the beginning of an experiment, and the lamp intensity obtained from
actinometry measurements.
(
(
(
(
(47) Buxton, G. V.; Greenstock, C. L.; Ross, A. B. J. Phys. Chem. Ref.
(
(
Data 1988, 17, 513.
(48) Dizdaroglu, M. Biochemistry. 1985, 24, 4476.
(49) Steenken, S.; Jovanovic, S. B.; Candeias, L. P.; Reynisson, J. Chem.
(
Eur. J. 2001, 7, 2829.
3
(50) Candeias, L. P.; Steenken, S. Chem. Eur. J. 2000, 6, 475.
(
(
(51) Ioele, M.; Bazzanini, R.; Chatgilialoglou, C.; Mulazzani, Q. G. J.
Am. Chem. Soc. 2000, 122, 1900.
1