Proton-coupled electron transfer in the oxidation of guanines by an aromatic pyrenyl radical cation in aqueous solutions

Article abstract of DOI:10.1039/b000212g

Electron transfer reactions between nucleic acid residues in DNA and strong oxidants are often the critical initial steps that initiate oxidative, irreversible DNA damage. Employing laser flash photolysis transient absorption spectroscopic techniques, we investigated the characteristics of electron transfer reactions in aqueous solutions between the 2'- deoxynucleoside 5'-monophosphates, dGMP, dAMP, dCMP and dTMP and a representative one-electron oxidant. The latter was a radical cation of a pyrene derivative with enhanced water-solubility, 7,8,9,10- tetrahydroxytetrahydrobenzo[a]pyrene (BPT). The BPT radical cation BPT⁺, was generated by intense nanosecond laser pulse (308 or 355 nm, 50-70 mJ pulse^{- 1} cm^-2) by a non-linear consecutive two-photon absorption process. No electron transfer reactions were observed with dAMP, dTMP and dCMP, consistent with their unfavorable redox potentials. However, BPT⁺ efficiently oxidized dGMP with a rate constant k(b) = 1.7 ± 0.1) x 10⁹ M^{- 1} s^-1, which is smaller than the diffusion-controlled value by a factor of only ~ 3. The dGMP(-H) neutral radicals formed on time scales of a few microseconds, were identified by their characteristic transient absorption spectrum (λ(max) ~ 310 mn). The rate constant of electron transfer from dGMP to BPT⁺ was smaller in D₂O by a factor of ~ 1.5 than in H₂O. This kinetic isotope effect indicates that the electron transfer reaction from dGMP to BPT⁺ is accompanied by the deprotonation of dGMP⁺, and therefore appears to be a proton-coupled electron transfer reaction.

Full text of DOI:10.1039/b000212g

View Article Online / Journal Homepage / Table of Contents for this issue
Proton-coupled electron transfer in the oxidation of guanines by an
aromatic pyrenyl radical cation in aqueous solutions
Vladimir A. Kuzmin,¤ Alexander Dourandin, Vladimir ShaÐrovich* and Nicholas E. Geacintov
Chemistry Department and Radiation and Solid State L aboratory, 31 W ashington Place, New
Y ork University, New Y ork, New Y ork, 10003-5180, USA. E-mail: vs5=is3.nyu.edu;
Fax: ]1 212 998 8421; T el: ]1 212 998 8456
Received 10th January 2000, Accepted 8th February 2000
Electron transfer reactions between nucleic acid residues in DNA and strong oxidants are often the critical
initial steps that initiate oxidative, irreversible DNA damage. Employing laser Ñash photolysis transient
absorption spectroscopic techniques, we investigated the characteristics of electron transfer reactions in
aqueous solutions between the 2@-deoxynucleoside 5@-monophosphates, dGMP, dAMP, dCMP and dTMP and
a representative one-electron oxidant. The latter was a radical cation of a pyrene derivative with enhanced
water-solubility, 7,8,9,10-tetrahydroxytetrahydrobenzo[a]pyrene (BPT). The BPT radical cation BPT~`, was
generated by intense nanosecond laser pulse (308 or 355 nm, 50È70 mJ pulse~1 cm~2) by a non-linear
consecutive two-photon absorption process. No electron transfer reactions were observed with dAMP, dTMP
and dCMP, consistent with their unfavorable redox potentials. However, BPT~` efficiently oxidized dGMP
with a rate constant k \ 1.7 ^ 0.1) ] 109 M~1 s~1, which is smaller than the di†usion-controlled value by a
b
factor of only D3. The dGMP(-H)~ neutral radicals formed on time scales of a few microseconds, were
identiÐed by their characteristic transient absorption spectrum (j D 310 mn). The rate constant of electron
max
transfer from dGMP to BPT~` was smaller in D O by a factor of D1.5 than in H O. This kinetic isotope
2
2
e†ect indicates that the electron transfer reaction from dGMP to BPT~` is accompanied by the deprotonation
of dGMP~`, and therefore appears to be a proton-coupled electron transfer reaction.
transfer reactions have been measured directly from a nucleic
1. Introduction
acid electron donor to a BPT~` radical cation electron accep-
tor in aqueous solutions. The pyrenyl radical cation, BPT,~`
was generated by intense 308 or 355 nm laser pulse excitation
of BPT via a consecutive two-photon ionization mechanism.
We show that the BPT~` radical cation oxidizes dGMP (2@-
deoxyguanosine 5@-monophosphate) by a one-electron transfer
mechanism, but does not oxidize either of the other three 2@-
deoxynucleoside 5@-monophosphates, dAMP, dCMP or
dTMP. Furthermore, the rate constant of oxidation of BPT~`
by dGMP is slower in D O than in H O solution, suggesting
OxidationÈreduction reactions involving nucleic acids in
aqueous solutions are currently receiving considerable atten-
tion.1h3 Interest in this area has been further stimulated by the
observations of long-distance migration of oxidative damage
in double-stranded DNA molecules,4h15 which has been
attributed to the base-to-base migration of holes.9,10,13h15
Aromatic Ñuorophores, e.g. those with pyrenyl16h23 and other
residues24h28 can interact with the nucleic acid bases by elec-
tron transfer reactions. In previous studies, we have focused
on the mechanisms of Ñuorescence quenching of the pyrene
2
2
that the electron transfer is coupled to a proton transfer in
these microsecond time scale reactions. While dAMP is not
oxidized by BPT~`, it forms noncovalent complexes with
BTP~`. The BTP~` radical cations in [BTP~`É É ÉdAMP] com-
plexes are still capable of oxidizing dGMP, although with a
lower rate constant than free BTP~`, indicating that steric
hindrance factors inhibit electron transfer from dGMP to
BPT~` in these hydrophobic dAMP complexes.
derivative
7,8,9,10-tetrahydroxytetrahydrobenzo[a]pyrene
(BPT) by electron donorÈacceptor interactions.17h19 This
pyrene derivative is most suitable for studies with nucleic
acids in aqueous solutions. Its spectroscopic and redox char-
acteristics are very similar to those of pyrene, but its solubility
in water is signiÐcantly enhanced by the presence of the four
hydroxy groups in the cyclohexanyl ring. We have shown
recently18 that the quenching rate of the singlet excited state
of this pyrenyl tetraol, 1BPT, by the 2@-deoxynucleosides dG,
dC, and dT is less efficient by factors of 1.5È2.0 in D O than
2
in H O. This kinetic solvent isotope e†ect was accounted for
2
2. Experimental
in terms of a non-radiative energy dissipation mechanism that
Transient absorption spectra were recorded using a kinetic
spectrometer system (D7 ns response time) described earlier.29
This system consists of a pulsed Lambda Physik EMG 160
MSC XeCl excimer laser (308 nm, fwhm \ 12 ns, D70 mJ
competes with the radiative decay of the 1BPT singlet excited
state. In aqueous solutions, this mechanism involves the for-
mation of short-lived charge transfer intermediates by a
proton-coupled electron transfer reaction.18 In this work,
using transient absorption spectroscopic methods, electron
pulse~1 cm~2) or
a Q-switched Continuum Surelite II
Nd:YAG laser (355 nm fwhm \ 4 ns, D60 mJ pulse~1 cm2)
excitation sources operated at 10 Hz, and a 75 W xenon
pulsed lamp as the transient absorbance probe Ñash. The
wavelengths of the probe light Ñash were varied using a
¤ On leave from the Institute of Biochemical Physics, Russian
Academy of Sciences, Moscow V-334, Russia.
DOI: 10.1039/b000212g
Phys. Chem. Chem. Phys., 2000, 2, 1531È1535
This journal is ( The Owner Societies 2000
1531

View Article Online
computer-controlled McPherson monochromator. The tran-
sient absorption signals were recorded at 2 nm intervals in the
260È750 nm spectral range. The probe Ñash signal was moni-
tored by a Hamamatsu R928 phomultiplier, and its output
was digitized and recorded using a Tektronix TDS 620
oscilloscope. The transient absorption spectra measured at
Ðxed, di†erent time intervals after the excitation Ñash, *t, were
constructed by sampling the amplitudes of each decay curve
at the same item interval *t. All experiments, including data
collection and analysis, were controlled by a computer.
The 2-deoxynucleoside 5@-monophosphates (99% purity)
were obtained from Sigma Chemical Corp. (St Louis, MO)
and were used as received. The BPT were prepared by hydro-
lysis of racemic anti-BPDE (7r,8t-dihydroxy-t9,10-epoxy-7,8,9,
10-tetrahydrobenzo[a]pyrene, a gift from Dr. S. Amin, Amer-
ican Health Foundation, Valhalla, NY).16 The absorbances of
the BPT phosphate bu†er solutions (pH 7) were generally
D0.1 cm~1 (308 nm) or 0.07 cm~1 355 nm), corresponding to
D25 lM concentrations of BPT (the concentration was esti-
mated from the molar extinction coefficient30 of BPT, e
\
343
2.9 ] 104 M~1 cm~1). The sample solutions, saturated with
N or O , were forced through a 0.5 ml quartz Ñow cell using
Fig. 1 (a): Transient absorption spectrum of BPT (65 lM) in deoxy-
genated 20 mM phosphate bu†er (pH 7) solutions measured with dif-
ferent delay times after the 355 nm Nd : YAG laser pulse excitation
(60 mJ pulse~1 cm~2). (b): Normalized transient absorption spectra:
the 3BPT spectrum was recorded 20 ls after excitation with a low
energy (8 mJ pulse~1 cm2) in deoxygenated solution; the BPT~` spec-
trum was recorded 2 ls after excitation with high energy (60 mJ
pulse~1 cm~2) laser pulses in oxygenated aqueous solutions. The
BPT~~ spectrum was generated by subtraction of appropriately
weighted 3BPT and BPT~` spectra from the experimental spectrum
measured at *t \ 1.4 ls [panel (a)].
2
2
positive gas pressure at a Ñow rate of D2 ml min~1.29 The
optical axis of the cell (pathlength 1 cm) was aligned parallel
to the direction of the probe light beam. The laser beam was
focused (excimer laser) with a quartz lens (focal length, f P 1
m) or expanded (Nd:Yag) by a cylindric quartz lens ( f \ [30
cm) and was passed through a 3 ] 9 mm rectangular aperture
to obtain a beam of uniform intensity (D70 mJ pulse~1 cm~2)
at the sample cell position. The transient absorbance of the
sample solutions was probed by a light beam passing through
a 3 mm aperture, oriented perpendicular to the laser light
beam, from a pulsed Xe lamp. All laser Ñash photolysis experi-
ments were performed at 20 ¡C.
cm~2; while the absorption spectrum of BPT~` was recorded
in oxygen-saturated solution in order to scavenge the
hydrated electrons formed (see below). The superposition of
the 3BPT and BPT~` spectra accounts for all of the spectral
features of the experimentally observed spectra, except for the
sharp band at 494 nm [Fig. 1, panel (a)]. The spectrum gener-
ated by the subtraction of these superimposed 3BPT and
BPT~` spectra from the upper trace (*t \ 1.4 ls) in Fig. 1(a),
is shown after normalization in panel (b) of Fig. 1. This spec-
trum, showing a single narrow absorption maximum at 494
nm, is very close to the spectrum of the BPT radical anion,
BPT~~, observed in N,N@-dimethylformamide solutions.17 A
decrease in the initial concentration of BPT, or laser pulse
energy, results in a pronounced reduction of the BPT~~
radical anion yields. These observations are consistent with a
mechanism of the BPT~~ formation via reaction of hydrated
electrons with BPT.
3. Results and discussion
3.1. Photoionization of BPT in aqueous solutions
Laser pulse photoexcitation of BPT in aqueous solutions
induces formation of the singlet excited state, 1BPT, which
undergoes intersystem crossing to the triplet excited state,
3BPT.17h19 In addition, the radical cation BPT~` is also
formed at high laser pulse energies. The lifetime of 1BPT
varies from 200 ns in oxygen-free solution to 135 ns in air-
equilibrated solutions.31 This lifetime is sufficiently long to
allow for the absorption of a second photon by 1BPT within a
nanosecond-duration laser pulse according to the consecutive
two-photon absorption mechanism:
The primary products of BPT photoionization are BPT~`
radical cations and hydrated electrons. In order to utilize
BPT~` as an oxidant of nucleic acids, it is important to scav-
enge the hydrated electrons in order to avoid their potential
1BPT ] hl ] BPT~` ] e~
(1)
The absorption of a second photon within the lifetime of
1BPT can occur when the energies of the nanosecond laser
pulses exceed17,18 10È20 mJ pulse~1 cm2. In aqueous solu-
tions, photoionization of BPT can be easily monitored via the
structureless absorption band (from D500 to 750 nm) of
hydrated electrons32 and the rather sharp absorption band
(455 nm) of BPT radical cations, BPT~`.17,18
Typical transient absorption spectra of BPT in deoxy-
genated aqueous solutions (pH 7.0) recorded at various delay
times after the 355 nm laser pulse excitation, are shown in Fig.
1. The spectra are characterized by the appearance of distinct
absorption bands in the 350È550 spectral range [panel (a)]. In
agreement with previous observations,17h19 two of these
bands can be assigned to 3BPT (distinct maximum at 418 nm
and shoulders at 396 and 515 nm), and BPT~` (peaks at 455
and 372 nm). Panel (b) shows the individual absorption
spectra of these intermediates, each normalized to the same
value at the absorption maxima. The spectrum of 3BPT was
recorded at relatively low laser energy, E \ 8 mJ pulse~1
reactions with the substrates. Molecular oxygen, O , is a very
2
efficient scavenger of hydrated electrons,33 and thus can be
used for this purpose. The O ~ radical anions formed in reac-
2
tion with hydrated electrons are relatively unreactive, and do
not react with nucleic acid bases, at least on time scales of
milliseconds, or less.34,35 Thus, in solutions saturated with
O , the decay of [e ~] (absorbance at 650 nm) is greatly
2
h
accelerated leading to a diminished formation of BPT~~ as
well, while the transient absorption spectra of BPT~` are not
a†ected (data not shown).
Typical transient absorption spectra of BPT in oxygen-
saturated aqueous solutions subjected to intense laser pulse
excitation, are shown in Fig. 2. The spectra are characterized
by a bleaching of the BPT ground state absorption bands at
267, 278, 313, 327 and 343 nm, and the appearance of new
absorption bands at 372 and 455 nm (BPT~`), and at 396, 418
and 515 nm (3BPT). The decay of the 3BPT transient absorp-
1532
Phys. Chem. Chem. Phys., 2000, 2, 1531È1535

View Article Online
3BPT state thus becomes the unique short-lived intermediate
observed on microsecond time scales. Finally, we have demon-
strated that 3BPT is not quenched by reaction with guanine,
because the energy level of the 3BPT is lower than the energy
of the corresponding radical ion pairs 3[BPT~~É É ÉdG~`].
Thus, the only important reactive intermediate that can react
with guanine derivatives on microsecond time scales is the
BPT~` radical cation.
In the presence of dGMP, the absorbance at 455 nm decays
more rapidly than in the absence of this nucleotide. Thus, the
lifetime of the BPT~` radical cations generated by laser pulse
excitation of BPT in oxygen-saturated solutions is signiÐ-
cantly shortened in the presence of dGMP. The decay proÐles
of BPT~` recorded at 455 nm can be analyzed in terms of a
single exponential decay law with a pseudo-Ðrst order rate
constant, k . The value of k increases linearly with increasing
Fig. 2 Transient absorption spectra of BPT (25 lM) in oxygenated
20 mM phosphate bu†er (pH 7) solutions measured with di†erent
delay times after the 308 nm XeCl excimer laser pulse excitation (70
mJ pulse~1 cm2).
a
a
concentration of dGMP (Fig. 3) according to the equation:
k \ k ] k [dGMP],
(2)
a
0
b
tion occurs within D2.5 ls and is accompanied by a partial
recovery of the BPT ground state absorption bands. In con-
trast, the BPT~` radical cation absorbance at 455 nm decays
more slowly (within D50 ls), but there is no corresponding
recovery of the ground state absorption bands at 300È350 nm
(Fig. 2). Thus, the BPT~` radical cations result in the forma-
tion of unidentiÐed and relatively long-lived intermediates
that are characterized by the loss of the aromaticity of the
pynenyl ring system and a broad absorbance in the 350È600
nm spectral range. These long-lived intermediates may be the
precursors of more stable reaction products. As in the case of
the irradiation of pyrene in air-saturated aqueous solutions,36
the analogous ultimate oxidation products of BPT could be
the 1-hydroxy- and the 1,6-quinone derivatives. The mecha-
nisms of decay of the BPT~` radical cations in the absence of
nucleic acids (Fig. 2) were not investigated here, but have been
recently discussed in the case of other polynuclear aromatic
hydrocarbon cations in aqueous solution.37
where k \ 1/q (q is the lifetime of BPT~` in the absence of
dGMP) and k is the second order rate constant of the oxida-
0
0
0
b
tion of dGMP by BPT~` (Fig. 3). The value of k calculated
b
from the slope of the k vs. [dGMP] plot in Fig. 3 is
a
(1.7 ^ 0.1) ] 109 M~1 s~1 (Table 1). This value is lower by a
factor of 3 than the value of the di†usion-controlled rate con-
stant, k \ 6.5 ] 109 M~1 s~1 calculated from the Smolu-
dif
chowski equation.38
The decay of the BPT~` radical cation in the absence of
dGMP is not accompanied by the regeneration of the BPT
ground state (Fig. 2). In contrast, the enhancement of the
BPT~` decay rate by reaction with dGMP is accompanied by
the nearly complete recovery of the BPT ground state absorp-
tion bands (Fig. 4). The transient absorption spectra of a 25
lM BPT and 2 mM dGMP solution recorded at time delays
*t \ 0.6È2 ls after the laser Ñash, are compared to one
another in this Ðgure. The transient spectrum at *t \ 0.6 ls is
typical of the superposition of the BPT~` (372 and 455 nm)
and 3BPT (396, 418 and 515 nm) spectra (compare Fig. 4 with
Fig. 2). All of the transient absorption spectra measured at
3.2. Selective oxidation of dGMP by BPT~‘ radical cations
The photoexcitation of BPT with intense laser pulses gener-
ates the short-lived intermediates, 1BPT, 3BPT and BPT~`.
While the focus of our work here is on the reactions of BPT~`
with 2@-deoxynucleotide 5@-monophosphates, it is necessary to
consider potential reactions of the neutral excited states 1BPT
and 3BPT with these nucleic acids as well. The reactions of
these neutral BPT excited states with nucleic acid bases have
been extensively studied previously in our laboratory.17h19
The singlet excited state 1BPT is efficiently quenched by dG
via electron transfer mechanisms.17,18 However, the electron
transfer products, BPT~~ and dG~` have been observed only
in non-aqueous, strongly polar organic solvents.17 In aqueous
solutions, hydrophobic interactions prevent the dissociation of
the singlet radical ion pairs, 1[BPT~~É É ÉdG~`]. Intersystem
crossing to the triplet radical ion pair states,
3[BPT~~É É ÉdG~`] is followed by their rapid and efficient
recombination to the [3BPTÉ É ÉdG] triplet states.17h19 The
Fig. 3 Dependence of the rate constant, k (decay of BPT~`) on the
a
concentration of dGMP in H O (…) and D O (L) solutions. The
2
2
BPT radical cations were generated by 308 nm XeCl excimer laser
pulse excitation (70 mJ pulse~1 cm~2) of BPT (25 lM) in oxygenated
20 mM phosphate bu†er solution (pH 7).
Table 1 Kinetic parameters of nucleoside 5@-monophosphates oxidation by free BPT~` radical cation and (BPT~`É É ÉdAMP) noncovalent
complex in H O and D O phosphate bu†er, solutions (20 mM NaH PO /Na HPO , pH 7)
2
2
2
4
2
4
dGMP
k /109 M~1 s~1
b
dAMP, dCMP, dTMP
Radical cation
H O
D O
k
/k
k /M~1 s~1
2
2
H2O D20
b
Free BPT~`
Complex
(BPT~`É É ÉdAMP)a
1.7 ^ 0.1
1.1 ^ 0.1
1.6 ^ 0.2
1.5 ^ 0.2
\105
0.32 ^ 0.02
0.22 ^ 0.02
a [dAMP] \ 0.01È0.03 M.
Phys. Chem. Chem. Phys., 2000, 2, 1531È1535
1533

View Article Online
cm~2). These spectra are characteristic of both the guanine
radical cation, dGMP~` and the neutral guanine radical,
dGMP(-H)~.34,35 Since, in neutral aqueous solutions, depro-
tonation of dGMP~` occurs with a rate constant39 k B 2
] 106 s~1, the guanine radicals observed in the microsecond
time scale transient absorption spectrum (Fig. 5) are mostly
neutral radicals, dGMP(-H)~.
Lack of oxidative interactions between BPT~‘ radical cations
and dAMP, dCMP and dTMP. Analogous transient absorp-
tion experiments with dGMP replaced by dAMP, dCMP or
dTMP (up to 100 mM concentrations), showed that the
BPT~` decay kinetics measured at 455 nm were not a†ected
by these nucleotides. Thus, in contrast to dGMP, the other
three nucleotides are not oxidized by BPT~` radical cations.
Based on these results, the rate constants of oxidation of
dAMP or pyrimidine bases by the BPT~` radical cations are
O105 M~1 s~1. The negligible reactivity of BPT~` in reac-
tions with adenosine and pyrimidine bases is correlated with
the redox potentials of these nucleic acid bases. The redox
potential of the neutral guanosine radicals in neutral
Fig. 4 Transient absorption spectrum of BPT (25 lM) and dGMP (2
mM) in oxygenated 20 mM phosphate bu†er (pH 7) solutions mea-
sured with di†erent delay times after the 308 nm XeCl excimer laser
pulse excitation (70 mJ pulse~1 cm~2).
delay times *t \ 0.8È2 ls spectra (Fig. 4) clearly indicate that
at this dGMP concentration (2 mM), the decay of both
BPT~` and 3BPT results in the almost complete recovery of
the BPT ground state absorption bands (327 and 343 nm).
Analysis of the results shown in Fig. 4 and similar data, indi-
cates that dGMP enhances the rate of decay of BPT~` only.
In agreement with previous observations17h19 the lifetime of
3BPT is not a†ected by dGMP (data not shown).
We conclude that the recovery of the BPT ground state
absorption band in the presence of dGMP (Fig. 4) is related to
electron transfer from dGMP to BPT~`. The product of this
electron transfer reaction should be the dGMP radical cation,
dGMP~`, or its deprotonation product, the dGMP(-H)~
neutral radical. Indeed, new and weak absorption bands in the
range of 300È330 and 350È500 nm become apparent at the
longer *t values (Fig. 4). However, due to incomplete
reduction of BPT~` by the dGMP at the nucleotide concen-
trations employed, the Ðnal spectrum at *t \ 2 ls contains
the residual absorption of the BPT~` decay products and thus
obscures the absorption spectrum of the dGMP radicals
formed.
To provide for the complete reduction of BPT~` by dGMP,
we increased the concentration of dGMP up to 5 mM. At this
relatively high dGMP concentration, the transient absorption
spectrum recorded at *t \ 2 ls shows a sharp absorption
band at 310 nm and two less intense broad absorption bands
at 390 and 500 nm (Fig. 5, trace A). A similar transient
absorption spectrum (trace B) was obtained after direct
photoionization of pure dGMP solutions (in the absence of
BPT) with intense 266 nm laser pulses (D200 mJ pulse~1
solutions, E [G(-H)~/G] \ 1.29 V vs. NHE, is smaller than
7
the redox potentials of neutral adenosine radicals,
E [A(-H)~/A] \ 1.42 V vs. NHE, and the pyrimidine radicals,
7
7
E B 1.7 V (thymidine) and E B 1.6 V (2-deoxycytidine).40
7
3.3. Noncovalent complex formation of BPT~‘ radical cations
with dAMP and quenching by dGMP
While dAMP does not reduce the BPT~` radical cations,
there is nevertheless an interaction between these two species.
This is evident from the transient absorption spectra of BPT~`
in 0.01È0.05 M solutions of dAMP. In general, the negative
transient absorption spectrum in the 300È360 nm region is
essentially the inverted absorption spectrum of the BPT
ground state arising from the bleaching e†ect, i.e., decreased
concentrations of BPT molecules in the ground state. The
transient absorption spectra of BPT~` in the presence of 30
mM (trace A) and in the absence (trace B) of dAMP are com-
pared in Fig. 6. The former spectrum (trace A) is red-shifted,
just like the ground state absorption spectrum of BPT in the
presence of dAMP (trace C), which is attributed to the forma-
tion of ground state complexes16,18 of BPT with dAMP,
[BPTÉ É ÉdAMP]. The absorption spectrum of BPT in the
absence of dAMP is also shown for reference (trace D). In the
presence of dAMP, the characteristic absorption band of
BPT~` in the 400È500 nm range is shifted to longer wave-
lengths (trace A) and is broader than that of the free BPT~`
Fig. 5 Transient absorption spectrum of dGMP(-H~) neutral radicals
generated by one-electron oxidation of dGMP by BPT~` (trace A),
direct two-photon photoionization of dGMP (trace B), in the absence
of BPT, in oxygenated 20 mM phosphate bu†er (pH 7) solutions.
Trace A: the transient absorption spectrum of BPT (25 lM) and
dGMP (3 mM) measured at *t \ 2 ls after 308 nm XeCl excimer
laser pulse excitation (70 mJ pulse~1 cm~2). Trace B: the transient
absorption spectrum of dGMP (0.1 mM) measured at *t \ 10 ls after
266 nm Nd:YAG laser pulse excitation (D200 mJ pulse~1 cm~2).
Fig. 6 E†ect of 2@-deoxyadenosine 5@-monophosphate on the tran-
sient absorption spectra of the BPT~` radical cations (traces A and B)
measured at *t \ 2 ls after 308 nm XeCl excimer laser pulse excita-
tion (70 mJ pulse~1 cm~2). The optical absorption spectra of the BPT
ground state molecules in the absence (trace D) and presence (trace C)
of dAMP in oxygenated 20 mM phosphate bu†er (pH 7) solutions.
Traces A and C: [dAMP] \ 30 mM; traces B and D: [dAMP] \ 0.
The spectra were normalized to one another at their respective
maxima.
1534
Phys. Chem. Chem. Phys., 2000, 2, 1531È1535

View Article Online
11 B. Li, B. Mao, T.-M. Liu, J. Xu, A. Dourandin, S. Amin and N. E.
Geacintov, Chem. Res. T oxicol., 1995, 8, 396.
12 N. E. Geacintov, K. Solntsev, L. W. Johnson, J. Chen, A. D.
Kolbanovskiy, T.-M. Liu and V. Ya. ShaÐrovich, J. Phys. Org.
Chem., 1998, 11, 1.
13 E. Meggers, D. Kusch, M. Spichty, U. Wille and B. Giese, Angew.
Chem., Int. Ed., 1998, 37, 460.
14 M. Bixon, B. Giese, S. Wessely, T. Langenbacher, M. E. Michel-
Beyerle and J. Jortner, Proc. Natl. Acad. Sci. USA, 1999, 96,
11713.
15 B. Giese, S. Wessely, M. Spormann, U. Lindemann, E. Meggers
and M. E. Michel-Beyerle, Angew. Chem., Int. Ed., 1999, 38, 996.
16 N. E. Geacintov, R. Zhao, V. A. Kuzmin, S. K. Kim and L. J.
Pecora, Photochem. Photobiol., 1993, 58, 185.
radical cations (trace B). These e†ects of dAMP on the BPT~`
transient absorption spectra suggest that [BPT~`É É ÉdAMP]
noncovalent complexes are formed in which BPT~` does not
oxidize dAMP.
The BPT~` radical cations in the [BPT~`É É ÉdAMP] com-
plexes can undergo electron transfer reactions with dGMP,
although with a markedly lower rate constants. The value of
k calculated from the slope of a k vs. [dGMP] linear plot for
b
a
these [BPT~`É É ÉdAMP] complexes (as in Fig. 3, data not
shown) is lower by a factor of 5 than k measured in the case
b
of free BPT~` (Table 1).
17 V. Ya. ShaÐrovich, P. P. Levin, V. A. Kuzmin, T. E. Thorgeirs-
son, D. S. Kliger and N. E. Geacintov, J. Am. Chem. Soc., 1994,
116, 63.
3.4. Deuterium isotope e†ects and the proton-coupled electron
transfer mechanisms
18 V. Ya ShaÐrovich, S. H. Courtney, N. Ya and N. E. Geacintov, J.
Am. Chem. Soc., 1995, 117, 4920.
In H O solutions, the oxidation of dGMP by BPT~` is more
2
rapid than in D O solutions (Fig. 3), and the values of k in
2
b
19 V. Ya. ShaÐrovich, V. A. Kuzmin and N. E. Geacintov, Chem.
Phys. L ett., 1994, 22, 185.
H O are larger than those in D O solutions by a factor of
2
2
D1.5 (Table 1). This kinetic isotope e†ect suggests that the
electron transfer from dGMP to free BPT~` (or BPT~` in
noncovalent complexes [BPT~`É É ÉdAMP], Table 1) is
coupled to a deprotonation of the primary electron transfer
radical cation product, dGMP~`. This proton-coupled elec-
tron transfer step leads to a lowering of the overall free energy
of reaction, thus favoring electron transfer.41,42 Since this
reaction occurs on time scales of D0.1È10 ls, the Ðnal product
is the neutral radical dGMP(-H)~, which is a product of the
known, rapid deprotonation34,35 of dGMP~`. Hence, the oxi-
dation of dGMP (neutral molecules) by BPT~` with the con-
comitant formation of dGMP(-H)~ (neutral radicals), and BPT
(neutral molecules), can be viewed in terms of electron transfer
from dGMP to BPT~` accompanied by the deprotonation of
dGMP~` (proton transfer from G~` to water), i.e. a proton-
coupled electron transfer reaction.18,43
20 D. OÏConnor, V. Ya. ShaÐrovich and N. E. Geacintov, J. Phys.
Chem., 1994, 98, 9831.
21 T. L. Netzel, K. NaÐsi, J. Headrick and B. E. Eaton, J. Phys.
Chem., 1995, 99, 17948.
22 T. L. Netzel, M. Zhao, K. NaÐsi, J. Headrick, M. S. Sigman and
B. E. Eaton, J. Am. Chem. Soc., 1995, 117, 9119.
23 E. Zahavy and M. A. Fox, J. Phys. Chem. B, 1999, 103, 9321.
24 A. Harriman, Angew. Chem., Int. Ed., 1999, 38, 945.
25 C. M. Seidel, A. Schulz and M. H. M. Sauer, J. Phys. Chem.,
1996, 100, 5541.
26 K. Fukui and K. Tanaka, Angew. Chem., Int. Ed., 1998, 37, 158.
27 F. D. Lewis, T. Wu, Y. Zhang, R. L. Letsinger, S. R. GreenÐeld
and M. R. Wasielewski, Science, 1997, 277, 673.
28 F. D. Lewis, X. Liu, Y. Wu, S. E. Miller, M. R. Wasielewski, R. L.
Letsinger, R. Sanishvili, A. Joachimiak, V. Tereshko and M. Egli,
J. Am. Chem. Soc., 1999, 121, 9905.
29 V. ShaÐrovich, A. Dourandin, W. Huang, N. P. Luneva and N. E.
Geacintov, J. Phys. Chem. B, 1999, 103, 10924.
30 Unpublished measurements in this laboratory.
31 N. E. Geacintov, D. Zinger, V. Ibanez, R. Santella, D. Grunber-
ger and R. G. Harvey, Carcinogenesis, 1987, 8, 925.
32 E. J. Hart and J. W. Boag, J. Am. Chem. Soc., 1962, 84, 4090.
33 B. H. J. Bielski, Photochem. Photobiol., 1978, 28, 645.
34 L. P. Candeias and S. Steenken, J. Am. Chem. Soc., 1989, 111,
1094.
Acknowledgements
This work was supported by the National Science Founda-
tion, Grant CHE-9700429 and by a grant from the Kresge
Foundation.
35 L. P. Candeias and S. Steenken, J. Am. Chem. Soc., 1992, 114,
699.
References
36 M. E. Sigman, P. F. Schuler, M. M. Ghosh and T. T. Dabestani,
Environ. Sci. T echnol., 1998, 32, 3980.
1
2
3
4
W. K. Pogozelski and T. D. Tullius, Chem. Rev., 1998, 98, 1089.
C. J. Burrows and J. G. Muller, Chem. Rev., 1998, 98, 1109.
B. Armitage, Chem. Rev., 1998, 98, 1171.
37 S. Fukuzumi, I. Nakanishi and K. Tanaka, J. Phys. Chem. A,
1999, 103, 11212.
38 M. Von Smoluchowski, Z. Phys. Chem., 1917, 92, 129.
39 This constant was estimated from K \ k /k , with pK \ 3.9
L. P. Candeias, P. OÏNeill, G. D. D. Jones and S. Steenken, Int. J.
Radiat. Biol., 1992, 61, 15.
a
`
~
a
for the guanosine radical cation, and k \ 1.8 ] 1010 M~1 s~1
5
6
7
8
9
T. Melvin, S. Botchway, A. W. Parker and P. OÏNeill, J. Chem.
Soc., Chem. Commun., 1995, 653.
D. B. Hall, R. E. Holmlin and J. K. Barton, Nature, 1996, 382,
731.
`
for the reaction of the 1-methylguanosine neutral radical with
protons.34
40 S. Steenken and S. V. Jovanovic, J. Am. Chem. Soc., 1997, 119,
617.
S. R. Rajski, S. Kumar, R. J. Roberts and J. K. Barton, J. Am.
Chem. Soc., 1999, 121, 5615.
41 D. Rehm and A. Weller, Isr. J. Chem., 1970, 8, 259.
42 S. J. Atherton and A. Harriman, J. Am. Chem. Soc., 1993, 115,
B. Armitage, C. Yu, C. Devados and G. B. Schuster, J. Am. Chem.
Soc., 1994, 116, 9847.
1816.
43 R. I. Cukier and D. G. Nocera, Annu. Rev. Phys. Chem., 1998, 49,
P. T. Henderson, D. Jones, G. Hampikian, Y. Kan and G. B.
Schuster, Proc. Natl. Acad. Sci. USA, 1999, 96, 8353.
337.
10 D. Ly, L. Sanii and G. B. Schuster, J. Am. Chem. Soc., 1999, 121,
9400.
Paper b000212g
Phys. Chem. Chem. Phys., 2000, 2, 1531È1535
1535