Chemical radiation studies of 8-bromo-2′ -deoxyinosine and 8-bromoinosine in aqueous solutions

Article abstract of DOI:10.1002/chem.200600040

The reactions of hydrated electrons (e_aq^-) with 8-bromo-2′-deoxyinosine (8) and 8-bromoinosine (12) have been investigated by radiolytic methods coupled with product studies and have been addressed computationally by means of BB1K-HMDFT calculations. Pulse radiolysis revealed that one-electron reductive cleavage of the C-Br bond gives the C8 radical 9 or 13 followed by a fast radical translocation to the sugar moiety. Selective generation of a C5′ radical occurs in the 2′-de-oxyribo derivative, whereas in the ribo analogue the reaction is partitioned between the C5′ and C2′ positions with similar rates. Both C5′ radicals undergo cyclizations, 10→11 and 14→15, with rate constants of 1.4 × 10⁵ and of 1.3 × 10⁴ s^-1, respectively. The redox properties of radicals 10 and 11 have also been investigated. A synthetically useful photoreaction has also been developed as a one-pot procedure that allows the conversion of 8 to 5′,8-cyclo-2′- deoxyinosine in a high yield and a diastereoisomeric ratio (5′R)/ (5′S) of 4:1. The present results are compared with data previously obtained for 8-bromoadenine and 8-bromoguanine nucleosides. Theory suggests that the behavior of 8-bromopurine derivatives with respect to solvated electrons can be attributed to differences in the energy gap between the π*- and σ*-radical anions.

Full text of DOI:10.1002/chem.200600040

DOI: 10.1002/chem.200600040
Chemical Radiation Studies of 8-Bromo-2’-deoxyinosine and 8-Bromoinosine
in Aqueous Solutions
Marialuisa Russo,[a] Liliana B. Jimenez,[a, b] Quinto G. Mulazzani,[a]
Mila DꢀAngelantonio,[a] Maurizio Guerra,[a] Miguel A. Miranda,[b] and
Chryssostomos Chatgilialoglu*[a]
Abstract: The reactions of hydrated
electrons (eaqꢀ) with 8-bromo-2’-deoxy-
inosine (8) and 8-bromoinosine (12)
have been investigated by radiolytic
methods coupled with product studies
and have been addressed computation-
ally by means of BB1K-HMDFT calcu-
lations. Pulse radiolysis revealed that
one-electron reductive cleavage of the
analogue the reaction is partitioned be-
tween the C5’ and C2’ positions with
similar rates. Both C5’ radicals undergo
cyclizations, 10!11 and 14!15, with
rate constants of 1.4”105 and of 1.3”
104 sꢀ1, respectively. The redox proper-
ties of radicals 10 and 11 have also
useful photoreaction has also been de-
veloped as a one-pot procedure that
allows the conversion of 8 to 5’,8-cyclo-
2’-deoxyinosine in a high yield and a
diastereoisomeric ratio (5’R)/ACHTREUNG(5’S) of
4:1. The present results are compared
with data previously obtained for 8-
bromoadenine and 8-bromoguanine
nucleosides. Theory suggests that the
behavior of 8-bromopurine derivatives
with respect to solvated electrons can
be attributed to differences in the
energy gap between the p*- and s*-
radical anions.
been investigated.
A
synthetically
ꢀ
C Br bond gives the C8 radical 9 or 13
Keywords: cyclization
functional calculations
·
density
· nucleo-
followed by a fast radical translocation
to the sugar moiety. Selective genera-
tion of a C5’ radical occurs in the 2’-de-
oxyribo derivative, whereas in the ribo
sides · pulse radiolysis · radical
reactions
Introduction
constant (kc) of 1.6”105 sꢀ1. In the analogous reaction with
8-bromoadenosine, the intramolecular hydrogen abstraction
by the initially formed C8 radical is partitioned between two
channels to generate both the C5’ and C2’ radicals with simi-
lar rate constants.[3] On the other hand, 8-bromo-2’-deoxy-
guanosine (5) captures electrons with the formation of radi-
cal anion 6 that undergoes protonation at C8 to afford the
one-electron oxidized 2’-deoxyguanosine 7.[6] Similar reac-
tions have been observed with 8-bromoguanosine.[4,5]
Information on the two distinct reaction paths was also
obtained by means of UB3LYP/6-31G* calculations. In the
frozen radical anion of 1 the unpaired electron is substan-
tially localized at C8 and, upon relaxation of the structural
parameters, the unpaired electron tends to occupy the anti-
bonding s*C8-Br molecular orbital favoring the loss of Brꢀ
and formation of radical 2.[2] On the other hand, radical
anion 6 is computed to be stable and is found to be proto-
nated at C8 with loss of Brꢀ to give radical 7.[5,6]
The reactions of hydrated electrons (eaqꢀ) with 8-bromoade-
nine[1–3] and 8-bromoguanine nucleosides[4–6] have recently
been investigated by radiolytic methods. It was found that 8-
bromo-2’-deACHTREUNGoxyACHTREUNGadenosine (1) captures electrons and rapidly
loses a bromide ion to give the corresponding C8 radical 2.
This intermediate intramolecularly abstracts a hydrogen
atom from the C5’ position to selectively afford the 2’-de-
oxy
adenosin-5’-yl radical 3 (Scheme 1).[1,2,7] This allowed the
reactivity of 3 to be studied properly for the first time, par-
ticularly the cyclization step 3!4, which occurs with a rate
[a] Dr. M. Russo, L. B. Jimenez, Dr. Q. G. Mulazzani,
Dr. M. DꢀAngelantonio, Dr. M. Guerra, Dr. C. Chatgilialoglu
ISOF, Consiglio Nazionale delle Ricerche
Via P. Gobetti 101, 40129 Bologna (Italy)
Fax : (+39)051-639-8349
E-mail: chrys@isof.cnr.it
8-Bromo derivatives 1 and 5 were also incorporated in a
variety of single- or double-stranded oligonucleotides and G
quadruplexes as a detection system for excess electron-
transfer processes.[8–11] Interestingly, the 8-bromo-2’-deoxy-
[b] L. B. Jimenez, Prof. M. A. Miranda
Departamento de Quimica, Universidad PolitØcnica de Valencia
Camino de Vera s/n, 46022 Valencia (Spain)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
AHCTREaUNG denosine moieties in a series of DNA hairpins containing a
7684
ꢁ 2006Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7684 – 7693

FULL PAPER
9.40 for 2’-deoxyguanosine[15a]
and 9.25 for guanosine.[15]
Therefore, the acid–base prop-
erties of the purine moiety
appear to be influenced by the
electron-withdrawing bromine
atom at the C8 position.
Reaction of hydrated electrons
(eaqꢀ) with 8 and 12: Radiolysis
ꢀ
of neutral water leads to eaq
,
C
C
HO , and H , as shown in
Equation (1). The values in pa-
rentheses represent the radia-
tion chemical yields (G) in
mmolJꢀ1.[16a] The reactions of
Scheme 1. Chemical studies of hydrated electrons with 8-bromo-2’-deoxyadenosine (1) and 8-bromo-2’-deoxy-
guanosine (5) show two different reaction paths.
ꢀ
eaq with the substrates were
studied in deoxygenated solu-
C
light-dependent flavin electron injector in the loop region of
the hairpin were found to capture electrons with quantita-
tive formation of the corresponding debrominated oligonu-
cleotides, as do the analogous 8-bromo-2’-deoxyguanosine
tions containing 0.25m tBuOH. As the rate constants of HO
[17]
radical[16] and H atoms with tBuOH [Eq. (2)] are 6.0”108
C
and 1.2”106 mꢀ1 sꢀ1, respectively, the substrate concentration
was chosen so that the majority of the HO and H species
C
C
derivatives.[10] In G quadruplexes, the reaction with eaq indi-
are scavenged by tBuOH.[18]
ꢀ
cated, for the first time, that excess electron transfer is also
H O radiolysis
e
ð0:27Þ, HOC ð0:28Þ, HC ð0:062Þ
effective in this supramolecular arrangement.[9]
ꢀ
ƒƒƒƒ!
ð1Þ
ð2Þ
2
aq
In the present work, we extended our studies to 8-bromo-
hypoxanthine nucleosides. In particular, 8-bromo-2’-deoxyi-
nosine (8) and 8-bromoinosine (12) have been investigated
by using radiolytic methods coupled with product studies.
Furthermore, BB1K-HMDFT calculations were carried out
on radical anions of 8-bromo-9-methyl purines in order to
understand the influence of heteroaromatic rings.
HOC=HC þ tBuOH ! ðCH3Þ2CðOHÞCH2 þ H2O=H2
C
ꢀ
The rate constants for the reaction of eaq with 8 or 12 were
determined to be (1.6Æ0.1)”1010 and (1.7Æ0.1)”
1010 mꢀ1 sꢀ1, respectively, by measuring the rate of the optical
ꢀ
density decrease of eaq
at l=720 nm (e=1.9”
104 mꢀ1 cmꢀ1)[19] as a function of the nucleoside concentration
(see the Supporting Information). These values are very
similar to the analogous reactions with 8-bromoadenine[2,3]
and 8-bromoguanine[4,6] nucleosides.
Results and Discussion
Synthesis and acid–base properties of 8-bromo-2’-deoxyino-
sine (8): Compound 8 was prepared by using the procedure
of Holmes and Robins for the synthesis of 8-bromoino-
sine,[12] although some experimental conditions were
changed in order to improve the yield. Briefly, a suspension
of the corresponding adenine derivative 1 in acetic acid was
treated with sodium nitrite. After basic workup, the residue
was purified by using reverse-phase silica gel chromatogra-
phy followed by recrystallization to give pure 8 in a 30%
overall yield.
At pHꢁ7, the UV spectrum of 8 exhibits two bands at l
ꢁ210 and 255 nm that are typical of other purine nucleo-
sides, with a molar absorption coefficient (e) of 1.35”
104 mꢀ1 cmꢀ1 at 255 nm. The spectrum showed a pH depend-
ence, and a pKa value of 8.7 was obtained from the analysis
of the optical density versus pH curves. A similar experi-
ment performed for 2’-deoxyinosine and inosine gave pKa
values of 9.2[13] and 8.8, respectively.[14] For comparison, a
pKa of 8.4 has been reported for 8-bromoguanosine,[4]
whereas pKa values for the related debrominated derivatives
have been found to be substantially higher, for example,
The reaction of an aqueous solution of 8 (1 mm) and
tBuOH (0.25m) at pHꢁ7 with eaq in the absence of O2 was
ꢀ
complete within ꢁ300 ns. At this time, no significant absorp-
tion was detected in the l=300–700 nm region. However, a
spectrum containing three bands centered at 310, 350, and
500 nm developed in ꢁ20 ms (Figure 1). The time profile for
the formation of the transient at lmax =350 nm (inset a) fol-
lowed first-order kinetics with a rate constant that was inde-
pendent of [8] in the concentration range 0.1–1 mm and
which slightly increased with the dose/pulse ratio. This dose
dependence is attributed to the mixing of first-order growth
and second-order decay of the species. Extrapolation to zero
dose gave a rate constant of kc =(1.4Æ0.1)”105 sꢀ1 at 208C.
Inset b shows the effect of the dose on the apparent molar
absorption coefficient at 350 nm (eapp), calculated by assum-
ing a radiation chemical yield of 0.27 mmolJꢀ1, which is the
G of hydrated electrons.[16a] A molar absorption coefficient
of 5000Æ100mꢀ1 cmꢀ1 at 350 nm was calculated by extrapo-
lating to zero dose.
In analogy to the reaction of 8-bromo-2’-deoxyadenosine
1,[2] we assigned the transient shown in Figure 1 to the con-
Chem. Eur. J. 2006, 12, 7684 – 7693
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7685

C. Chatgilialoglu et al.
Figure 2. Transient absorption spectrum obtained from the pulse radioly-
sis of an Ar-purged solution containing 12 (1 mm) and tBuOH (0.25m) at
pH ꢁ7, taken 170 ms after the pulse; dose=23.8 Gy, optical path=
2.0 cm. Insets: a) Time-dependence of absorption at l=310 nm; dose=
22.6Gy; b) Dependence of kobs (see text) from the radiation dose at l=
310 nm.
Figure 1. Transient absorption spectrum obtained from the pulse radioly-
sis of an Ar-purged solution containing 8 (1 mm) and tBuOH (0.25m) at
pHꢁ7, taken 20 ms after the pulse; dose=25.5 Gy, optical path=2.0 cm.
Insets: a) Time-dependence of the absorption at l=350 nm; the solid
line represents the first-order kinetic fit to the data; b) Dependence of
eapp (see text) at l=350 nm on the radiation dose.
5900Æ100mꢀ1 cmꢀ1 at 310 nm was calculated by extrapolat-
ing to zero dose and assuming G=0.27 mmolJꢀ1.
In analogy to the reaction of the 2’-deoxyribo derivative
8, we assigned the transient in Figure 2 to the conjugated
aminyl radical 15, and the observed rate to the cyclization of
radical 14 (Scheme 3).
jugated aminyl radical 11 and the observed first-order
growth to the cyclization of radical 10 (Scheme 2).
ꢀ
A comparison of the reactions of eaq with 8 and 12 shows
that the transient spectra are developed in 20 and 170 ms, re-
spectively, and accordingly, the cyclization rate constant for
the reaction 10!11 is 11 times faster than that of the reac-
tion 14!15. This is probably a consequence of the confor-
mational changes between ribo and 2’-deoxyribo derivatives.
*
Figure 3 compares the spectra of aminyl radicals 11 ( )
~
and 15 ( ). The absorbance at the maxima (310 and
350 nm) of the two species are quite different: one being
60% of the other. The shape of both spectra strongly resem-
!
ble that assigned to the isostructural radical 19 ( ), ob-
tained from the reaction of inosine with H atoms.[14] Howev-
C
er, the unexpected similarities of re-
ported e values for 19 and our radical
15 forced us to reinvestigate the reac-
Scheme 2. Proposed mechanism for the reaction of hydrated electrons
with 8-bromo-2’-deoxyinosine (8).
C
tion of inosine with H atoms.
The reaction of an N2O-saturated
aqueous solution of inosine (2 mm)
and tBuOH (0.2m) at pHꢁ7 afforded
ꢀ
Under the same conditions, the reaction of eaq with 12
gave a similar spectrum that developed more slowly, namely
170 ms (Figure 2). The time profile for the formation of the
transient at lmax =310 nm (Figure 2, inset a) follows first-
order kinetics with a rate constant (kobs) that is independent
of 12 in the concentration range 0.2–1 mm. As expected
from the mixing of the first-order growth and the second-
order decay, kobs increased with the dose/pulse ratio
(Figure 2, inset b). An empirical expression was used to fit
the experimental data and extrapolate to zero dose[20] to
give a rate constant of kc =(1.3Æ0.1)”104 sꢀ1 at 208C. The
absorbance at l=310 nm is also found to vary substantially
with the dose. An apparent extinction coefficient (eapp) of
^
the spectrum ( ) shown in Figure 3,
which has an e value 60–70% higher
than that reported in the literature.[14]
The similarities in shape and e values
*
^
between radical 11 ( ) and 19 ( ) are gratifying. Therefore,
the difference in e values between radical 11 and 15 suggests
that only ꢁ60% of the radicals produced by the reaction of
the hydrated electron with 12 leads to radical 15
(Scheme 3).
ꢀ
More information on the reaction of eaq with 12 was ob-
tained by experiments in the presence of N,N,N’,N’-tetra-
7686
www.chemeurj.org
ꢁ 2006Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7684 – 7693

Chemical Radiation Studies
FULL PAPER
Scheme 3. Proposed mechanism for the reaction of hydrated electrons with 8-bromoinosine (12).
ments,[24] we concluded that the TMPD radical cation is
+
C
ꢀ
formed from intermediates generated by the reaction of eaq
with 12.[25]
+
C
The yield of TMPD (corrected for its decay) increased
on increasing the TMPD concentration, varying between
17% for [TMPD]=30 mm to 35% for [TMPD]=120 mm,
with respect to the yield of eaqꢀ. The induction period ob-
+
C
served for the formation of TMPD was satisfactorily fitted
with a two consecutive reaction model. The first component
is independent of TMPD concentration and occurs with a
rate constant of kf ꢁ7”104 sꢀ1. The second component de-
pends on the concentration of TMPD, and the rate constant
kTMPD ꢁ2”108 mꢀ1 sꢀ1 is assigned to the species reacting with
TMPD. We suggest that the initially produced radical 13
partitioned its translocation step between two channels to
afford the C5’-radical 14 and the C2’-radical 16 (Scheme 3),
similar to the case of 8-bromoadenosine.[3] The rate constant
for heterolytic cleavage of the glycosidic bond, which pro-
duces the radical cation 17 and hypoxanthine 18, is slightly
smaller than the analogous reaction of the 2’-adenosinyl rad-
ical (kf =1.1”105 sꢀ1).[3] Radical cation 17 oxidizes TMPD
with a rate constant of kTMPD ꢁ2”108 mꢀ1 sꢀ1.
Figure 3. Transient absorption spectra obtained from the pulse radiolysis
*
of solutions containing ( ) 8 (1 mm) and tBuOH (0.25m) at pHꢁ7, Ar-
purged, taken 20 ms after the pulse, assuming G=0.27 mmolJꢀ1 and ex-
~
trapolating to zero dose; ( ) 12 (1 mm) and tBuOH (0.25m) at pHꢁ7,
Ar-purged, taken 170 ms after the pulse, assuming G=0.27 mmolJꢀ1 and
!
extrapolating to zero dose; ( ) taken from ref. [14] and refers to inosine
(0.2 mm) and tBuOH (0.1m) at pHꢁ7, N2O-saturated, taken 40 ms after
^
the pulse; ( ) inosine (2 mm) and tBuOH (0.2m) at pHꢁ7, N2O-saturat-
ed, taken 9 ms after the pulse, dose=19.6Gy, assuming G=
0.084 mmolJꢀ1
.
Redox properties of radicals 10 and 11: Next we investigat-
ed the reactivity of the C5’ radical 10 and aminyl radical 11
with respect to [Fe(CN)6]3ꢀ, methyl viologen (MV2+), and
molecular oxygen in analogy to the reactions observed for
radicals 3 and 4.[2]
A
+
C
methyl-p-phenylenediamine (TMPD, E (TMPD /TMPD)=
0.27 V),[21] as previously reported for the case of 8-bromoa-
denosine.[3] Pulsing O2-free aqueous solutions of 12 (1 mm)
containing tBuOH (0.25m) and different concentrations of
TMPD (30–120 mm) at pHꢁ7 led to the oxidation of
In the case of [Fe(CN)6]3ꢀ, for which EA([Fe(CN)6]3ꢀ/
[Fe(CN)6]4ꢀ)=0.36V, [21a,26] this was tested by submitting
deaerated solutions of 8 (1 mm) containing tBuOH (0.25m)
and different concentrations of K3[Fe(CN)6] (25–100 mm) at
TMPD. The rate constants for this oxidation and the associ-
+
C
ated formation of TMPD were measured at l=565 nm
(e=1.25”104 mꢀ1 cmꢀ1), which represents one of the absorp-
+
ꢀ
C
tion maxima of TMPD
(see the Supporting Informa-
pH 7 to pulse radiolysis. Under these conditions, eaq reacts
tion).[22,23] Because HO radicals are trapped by tBuOH
under these experimental conditions and the reaction of
TMPD with the carbon-centered radicals derived from
tBuOH is unimportant on the timescale of our experi-
exclusively
with
8,
as
(eaqꢀ+[Fe(CN)6]3ꢀ)=3.1”
kACHTREUNG
C
109 mꢀ1 sꢀ1.[16] The absorption at l=350 nm was found to de-
crease by increasing the concentration of [Fe(CN)6]3ꢀ (see
inset of Figure 4). A Stern–Volmer type of approach gave
Chem. Eur. J. 2006, 12, 7684 – 7693
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7687

C. Chatgilialoglu et al.
+
[21]
C
k
oxA
MV )=ꢀ0.45 V, was also investigated. The experiments
were performed by adding different concentrations of MV2+
(50, 100, or 150 mm) to Ar-purged solutions containing 8
(1 mm) and tBuOH (0.25m). Under these conditions, 8 and
of this data with the kc value yielded k CTHREUNG
(10/[Fe(CN)6]3ꢀ)=
(2.0Æ0.3)”109 mꢀ1 sꢀ1, which is attributed to a reaction of
the C5’ radical 10 with [Fe(CN)6]3ꢀ to afford the correspond-
ing cation.
MV2+ should compete for eaq because k(eaqꢀ+MV2+)=
ꢀ
7.2”1010 mꢀ1 sꢀ1.[16] Indeed, a two-component increase of the
absorption at l=605 nm, characteristic of the methyl violo-
gen radical cation (MV+ , e605 nm =1.37”104 mꢀ1 cmꢀ1),[27] was
C
observed upon irradiation. For example, using 100 mm MV2+
,
the instantaneous formation of MV+ accounted for 29% of
eaqꢀ, whereas the slow growth accounted for 8.5%. As the
C
Me2C(OH)CH2 radical does not react with MV2+, this dem-
C
onstrated an electron transfer between the precursor of the
aminyl radical (i.e., C5’ radical 10) and MV2+ with a rate
constant of k
oxA
+
C
worth mentioning that, under these conditions, the MV
decays by second-order kinetics (kd =3.1”109 mꢀ1 sꢀ1), which
probably involves a back-electron transfer to the aminyl
C
(11) or to the Me2C(OH)CH2 radical.
The reactivity towards molecular oxygen was investigated
by pulsing solutions of 8 (1 mm) containing tBuOH (0.25m)
and different concentrations of O2 (0–140 mm) at pH 7. The
absorption at 350 nm was found to decrease on increasing
the O2 concentration (Figure 6). This decrease could be at-
Figure 4. Dependence of 1/DA at l=350 nm on [K3[Fe(CN)6]] obtained
from the pulse radiolysis of Ar-purged solutions containing 8 (1 mm) and
tBuOH (0.25m) at pHꢁ7; optical path=2.0 cm, dose=24.6Gy. Inset:
Dependence of DA at l=350 nm on [K3[Fe(CN)6]]. The solid line repre-
sents an exponential fit to the data.
ꢀ
tributed to the reactions of eaq with 8 and O2 because k-
AHCTREUNG
(eaqꢀ+O2)=1.9”1010 mꢀ1 sꢀ1.[16] Taking into consideration the
ꢀ
percentage of eaq captured by O2 and the use of a Stern–
Volmer type of approach (Figure 6, inset a), we obtained
On the other hand, the decay of the transient at 350 nm
followed first-order kinetics in the presence of [Fe(CN)6]3ꢀ
(see inset of Figure 5). From the slope of kd(11) versus
[K3[Fe(CN)6]], the bimolecular rate constant was found to
k CHTREUNG
(10/O2)/kc =8.8”103 mꢀ1. Combination of this data with
oxA
the kc value yielded k
the other hand, the first-order growth (kobs) increased linear-
ly by increasing the oxygen concentration, that is, kobs =kc +
oxA
be k
C
k CHTREUNG
oxA
molecular rate constant was found to be k
0.1)”109 mꢀ1 sꢀ1 (inset b). An average value of k
CHTREUNG
CTHREUNG
The reactivity of the transient species towards the weaker
oxidant methyl viologen (MV2+), for which EA(MV2+/
Figure 5. Dependence of kobs for decay at l=350 nm on [K3[Fe(CN)6]]
obtained from the pulse radiolysis of Ar-purged solutions containing 8
(1 mm) and tBuOH (0.25m) at pHꢁ7. Inset: Time-dependence of the ab-
sorption at 350 nm in the presence of K3[Fe(CN)6] (25 mm); optical
path=2.0 cm, dose=23.9 Gy. The solid line represents the first-order ki-
netic fit to the data.
Figure 6. Dependence of DA at l=350 nm on [O2] obtained from the
pulse radiolysis of solutions containing 8 (1 mm) and tBuOH (0.25m) at
pHꢁ7 saturated with Ar/O2 mixtures; optical path=2.0 cm, dose=
22.5 Gy. The solid line represents an exponential fit to the data. Insets:
a) Dependence of DAꢀ1 at 350 nm on [O2]; b) Dependence of kobs for
growth at 350 nm on [O2].
7688
www.chemeurj.org
ꢁ 2006Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7684 – 7693

Chemical Radiation Studies
FULL PAPER
(2Æ1)”109 mꢀ1 sꢀ1 is suggested for the reaction of the C5’
radical 10 with molecular oxygen.
The presence of O2 also influenced the disappearance of
radical 11. The second-order (SO) kinetics became faster as
the concentration of O2 increased. Figure 7 shows the de-
Table 1. Rate constants for the oxidation of C5’ radicals.
Oxidizing agent kox of radical 3[a,b] [mꢀ1 sꢀ1
kox of radical 10
(2Æ1)”109
(2.0Æ0.3)”109
(2.0Æ0.6)”108
]
ACHTREUNG ]
[a] [mꢀ1 sꢀ1
oxygen
N
ACHTREUNG
[Fe(CN)6]3ꢀ
MV2+
ACHTREUNG
ACHTREUNG
[a] Reactions in neutral water at 22Æ28C. [b] Data from ref. [2].
probably involve an inner-sphere electron-transfer proc-
ess.[31]
Product studies from continuous radiolysis of 8 and 12: Dea-
erated samples of an aqueous solution of 8 (ꢁ1.5 mm) and
tBuOH (0.25m) at pHꢁ7, were irradiated under steady-
state conditions with a total dose of 2.0 kGy at a dose rate
of ꢁ13 Gyminꢀ1. HPLC analysis of the reaction showed
complex unresolved mixtures. Repetition of the experiment
in the presence of K4[Fe(CN)6] (4 mm) gave different re-
sults.[32] HPLC analysis showed that 34% of the starting bro-
mide was consumed and that a few products were formed.
The compounds (5’R)-5’,8-cyclo-2’-deoxyinosine (21) and
(5’S)-5’,8-cyclo-2’-deoxyinosine (23) were identified by com-
parison with authentic samples (see below), the major com-
pound being the 5’R isomer in a 64% yield (based on the re-
acted bromide). Hypoxanthine (25) and 2’-deoxyinosine (26)
were also identified as minor products by comparison with
commercially available compounds, whereas hydrated 5’-car-
boxaldehyde-2’-deoxyinosine (24) is tentatively assigned on
Figure 7. Dependence of kobs for second-order decay at l=350 nm on
[O2] obtained from the pulse radiolysis of solutions containing 8 (1 mm)
and tBuOH (0.25m) at pHꢁ7 saturated with Ar/O2 mixtures. Inset:
Time-dependence of absorption at 350 nm in the presence of 50 mm O2;
optical path=2.0 cm, dose=20.9 Gy. The solid line represents a second-
order kinetic fit to the data.
the basis of LC–MS analysis. Interestingly, the ratio (5’R)/ACHTUNG-
pendence of k(SO)obs on [O2], and its inset shows the
second-order kinetic fit to the data for 50 mm of oxygen.
Such behavior is consistent with the reversible reaction of
aminyl radical 11 with O2 to give the corresponding peroxyl
adduct [Eq. (3/ꢀ3)], followed by cross-termination between
this radical and a second radical 11 [Eq. (4)]. Under these
conditions, the second-order
R
gous experiments with 8-bromo-2’-deoxyadenosine.[2]
rate constant, k(SO)obs, will be
a complex function of [O2], k3,
kꢀ3, and k4.[28]
11C þ O2 Ð 11-OOC
ð3= ꢀ 3Þ
11-OOC þ 11C ! products
ð4Þ
Table 1 summarizes the results
obtained for the oxidation of
C5’ radicals 3 and 10. The reac-
tivities of the two species are
very similar. The rate constants
for the reactions with oxygen
are typical for a-heteroatom-
substituted alkyl radicals.[16,29,30]
The rate constants for the reac-
tions with MV2+ are at least
one order of magnitude slower
than those with the stronger
oxidant [Fe(CN)6]3ꢀ, which
Scheme 4. Proposed mechanism for the formation of nucleosides derived from C5’ radical 10.
Chem. Eur. J. 2006, 12, 7684 – 7693
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7689

C. Chatgilialoglu et al.
conformation. Radicals 20 and 22 are readily oxidized by
[Fe(CN)6]3ꢀ to products 21 and 23. Hydrated aldehyde 24
should be derived from the oxidation of radical 10. The
minor amounts of 26 could be explained by reduction of
For one isomer, the coupling constant JACHTREUN(G H4’,H5’) is 0.6Hz
(in DMSO), which is indicative of a dihedral angle of ꢁ908
between these two protons and is compatible with 5’R ster-
eoconfiguration. The 5’S isomer has JACHTRE(UNG H4’,H5’)=6.9 Hz.
radicals 9 or 10; alternatively, a H might add to the C8=N7
These findings are in very good agreement with 1H NMR
data of other 5’,6-cyclopurine derivatives.[2,33]
C
double bond of 8. The release mechanism of base 25 may in-
volve minor paths of radical translocation of intermediate 9
to positions other than the C5’ of the sugar moiety
(Scheme 2).
Entry 1 in Table 2 shows the reaction in an aqueous
medium and at natural pH. After 3 h of photolysis, the
cyclic products (21 and 23) were formed in a 38% overall
yield and in a ratio (5’R)/ACHTREU(NG 5’S)=3.9, together with a 13%
yield of hydrated 5’-carboxaldehyde (24). The reduced prod-
uct (26) was also formed. As was reported for the analogous
adenine derivative 1, N-glycosidic bond hydrolysis is favored
at acidic pH values[7] and leads to 27. This pathway was in-
hibited in a buffered solution at pH 7 (entry 2).
ꢀ
C
To test the ability of halide anions (I ) to trap Br atoms
in the photolysis mixture, the reaction was carried out in
ꢀC [7]
water in the presence of NaI to generate IBr . Entry 3 in
Table 2 shows the results of photolysis of a deaerated aque-
ous solution containing 1 mm of 8 in the presence of 2 equiv
of NaI. A much higher conversion rate (91% relative to
35% in water) was observed, although the diastereomeric
Deaerated aqueous solutions of 12 (ꢁ1.5 mm) and
tBuOH (0.25m) at pHꢁ7 were irradiated under stationary-
state conditions with a total dose of up to 2.0 kGy at a dose
rate of ꢁ12 Gyminꢀ1, followed by HPLC and LC–MS analy-
ses. The main reaction product was hypoxanthine (25),
which accounts for ꢁ20% of the reacted bromide. These re-
sults are in good agreement with the aforementioned pulse
radiolysis observations. It is worth mentioning that 25 was a
minor reaction product in the analogous experiment with 8,
which further supports our conclusion that the free base
may be derived by C2’ radical chemistry.
ratio between the cyclic photoproducts 21 and 23 was main-
ꢀ
C
Cꢀ
tained. Therefore, the equilibrium Br +I ÐBrI plays an
important role in this synthetically useful radical cascade by
regulating the relative concentrations of the two reactive ox-
idizing species.[7]
All these results support the mechanism shown in
Scheme 4 in which the formation of radical C5’ is followed
by different reaction pathways. Among the oxidized prod-
ucts, formation of 5’,8-cyclopurines 21 and 23 competes with
that of the hydrated aldehyde 24 as the main processes.
Steady-state photolysis studies of 8: Bromide 8 was irradiat-
ed with UV light under a variety of experimental conditions.
The crude reaction mixtures were analyzed by means of
HPLC coupled with UV (diode array) and MS (ion trap)
detection, using authentic samples as reference compounds
for identification. Table 2 and Scheme 4 summarize the ex-
perimental findings and the reaction products. The yields
are based on the consumption of the starting bromide for a
better comparison. Workup of the reaction in entry 3 al-
lowed the isolation and characterization of the two diaster-
eoisomers 21 and 23. Support for the correct configuration
assignment at the C5’ position was provided by 1H NMR
The origin of differences in behavior between 8-bromopur-
ine nucleosides: The radical anions of 8-bromohypoxanthine
derivatives are not stable and lose Brꢀ to form the neutral
s-type radical at C8, as found experimentally for the radical
anion of 8-bromoadenine derivatives. On the other hand,
radical anions of 8-bromoguanine derivatives have been
found to be stable and undergo very fast protonation.
BB1K-HMDFT calculations were carried out on 8-bromo-9-
methyladenine (28), 8-bromo-9-methylguanine (29), and 8-
bromo-9-methylhypoxanthine (30) to understand this differ-
ent behavior.
analyses and, in particular, by the magnitude of JACTHRE(UNG H4’,H5’).
At this level of theory, the radical anion of 28 is computed
to be unstable, as found previously for 8-bromo-2’-deoxya-
denosine with B3LYP-HDFT calculations.[3] On the contra-
Table 2. Yields of products obtained by UV photolysis of deaerated solu-
tions of 8.[a]
ꢀ
ry, radical anions of 29 and 30 are stable. The C8 Br bond
Entry Solvent
Conditions
Conv.
[%]
Product yields [%]
length increases only slightly on going from the neutral mol-
ecule to the radical anion, namely, from 1.845 to 1.864 Š for
29 and from 1.844 to 1.864 Š for 30, as expected for a delo-
t [h]
pH
21 23
24 25
26 27
1
2
3
H2O
H2O
3
3
7!4[b]
35
25
91
30
8
13 <5 11 28
7[c]
31 <1 17
38 10 2610
13 31
3 8
–
ꢀ
calized p-radical anion. As the C Br bond elongates, these
H2O/NaI[d] 1.5
7!4[b]
radical anions remain planar and their energy increases.
However, the energy gap decreases between the p*-radical
anion and the s*-radical anion, where the unpaired electron
[a] Irradiation of 8 (10 mL per tube, 1 mm) under N2 was performed with
a multilamp photoreactor at l=254 nm. Yields are based on the conver-
sion of bromide 8. Estimated errors are less than 10% of the stated
values. [b] pH changed during the course of the reaction. [c] Buffer solu-
tion with KH2PO4/Na2HPO4 (10 mm). [d] Solution of 8 with 2 mm NaI.
ꢀ
is located at the s*C8-Br antibonding MO, until, at a large C
Br distance, the s*-anionic state becomes more stable than
7690
www.chemeurj.org
ꢁ 2006Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7684 – 7693

Chemical Radiation Studies
FULL PAPER
thus favoring the loss of Brꢀ. The electron densities at the
C8 (7.5%) and Br (1.2%) atoms decrease significantly in
the LUMO of 30 (C) so that the p-radical anion state is
slightly more stable than the s-radical anion state. This ex-
ꢀ
plains the small value computed for the C Br dissociation
barrier. In the LUMO of 29 (B), the electron density be-
comes negligible (2.3% at C8 and 0.3% at Br) so that the
p-radical anion state is significantly more stable than the s-
ꢀ
radical anion state. The C Br dissociation barrier increases
and the lifetime of the p-radical anions is now sufficient to
be protonated at C8, as occurs in the unsubstituted 2’-deoxy-
adenosine and 2’-deoxyguanosine.[34] Figure 8 shows that
there is a shift of electron density from the five-membered
ring to the six-membered ring on going from 28 to 30 and
from 30 to 29. The electronic distribution in the LUMO of
28 is delocalized over the two rings. Substitution of a p-elec-
tron-releasing amino group in the six-membered ring with a
p-electron-withdrawing carbonyl group in the LUMO of 30
tends to enhance the extra electron delocalization in the six-
membered ring. In the LUMO of 29, the extra electron is
mainly localized on the six-membered ring owing to the ad-
junctive presence of a carbon atom (C2) that is extremely
electron deficient in the neutral molecule, because it is sur-
rounded by three electronegative atoms.
ꢀ
the p*-anionic state. At a larger C8 Br distance, the C8 site
becomes strongly pyramidal and the energy of the radical
anions decreases rapidly so that they tend to lose Brꢀ. Inter-
estingly, crossover between the two states occurs at a shorter
ꢀ
C8 Br bond length for the radical anion of 30 (2.01 Š) than
for the radical anion of 29 (2.095 Š). This indicates that
cleavage of the C Br bond should be easier in the 8-bromo-
hypoxanthine derivatives than in 8-bromoguanine analogues
during pulse radiolysis experiments. Indeed, the energy in-
ꢀ
ꢀ
crease caused by the elongation of the C Br bond at the
crossing point is small (15.9 kJmolꢀ1) for the radical anion
of 30, whereas it is significant (35.2 kJmolꢀ1) for the radical
anion of 29. This different behavior can be rationalized on
the basis of the electron distribution in the LUMO of 28, 29,
and 30, namely, in the MO where the extra electron is cap-
tured.
Conclusion
The results described herein demonstrate that the reaction
ꢀ
Figure 8 shows that in the LUMO of 28 (A) the electron
density is largely localized at C8 (22.1%), which is the
carbon linked to the Br atom. A significant electron density
is also computed at the Br atom (4.1%). Hence the un-
paired electron in the anion tends to occupy the antibonding
s*C8-Br MO upon relaxation of the structural parameters
of 8 with eaq at pHꢁ7 leads to a C5’ radical 10, whereas
ꢀ
the analogous reaction with 8-bromoinosine 12 with eaq at
pHꢁ7 leads to C5’ radical 14 and C2’ radical 16 in a ratio of
ꢁ60:40. We found that the C5’ radical adds intramolecularly
to the C8=N7 double bond in the hypoxanthine moiety with
rate constants of 1.4”105 and 1.3”104 sꢀ1 for the 2’-deoxyri-
bo and ribo forms, respectively, affording the 5’R and 5’S
isomers. The C2’ radical liberates hypoxanthine with a rate
constant of 7”104 sꢀ1. Redox properties of radicals 10 and
11 have also been obtained. The similarities between 8-bro-
mohypoxanthine and 8-bromoadenine[2,3] derivatives to-
ꢀ
wards eaq and the differences between 8-bromohypoxan-
ꢀ
thine and 8-bromoguanine[4,5] derivatives towards eaq can
be rationalized in terms of the energy gap between the p*-
radical anion and the s*-radical anion by means of BB1K-
HMDFT calculations. We took advantage of the observed
chemical behavior to develop a synthetically useful photore-
action that converts 8 into (5’R)-5’,8-cyclo-2’-deoxyinosine
(21) and (5’S)-5’,8-cyclo-2’-deoxyinosine (23) in a ratio of
4:1 and in a one-pot procedure. Our findings extend the ob-
servations of a different radical reactivity between 2’-deoxy-
ribo and ribo nucleosides, and provide a molecular basis for
forthcoming experiments with model DNA or RNA radicals
to elucidate biological damage. Moreover, the radical reac-
tivity of hypoxanthine derivatives may be useful for studies
concerning nucleoside-type natural products, which are in-
teresting for developing drugs, and in other pharmacological
studies.[35]
Figure 8. LUMO of derivatives 28 (A), 29 (B), and 30 (C) computed at
the BB1K-HMDFT level.
Chem. Eur. J. 2006, 12, 7684 – 7693
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7691

C. Chatgilialoglu et al.
(333) [M+H]+; HRMS (FAB): calcd for C10H11BrN4O4: 331.0042; found:
331.0038 [M+H]+.
Experimental Section
AHCTREUNG
Chemicals: 8-Bromo-2’-deoxyadenosine was provided by Berry & Associ-
ates (Ann Arbor, MI, USA). 8-Bromoinosine, sodium iodide,
K4[Fe(CN)6], and the phosphate buffer were purchased from Sigma-Al-
drich. Acetonitrile and methanol, both HPLC grade, were from Scharlau.
Water was purified through a Millipore Milli-RO plus 30 system.
AHCTREUNG
A
ACHTREUNG
A
ACHTREUNG
A
G
ACHTREUNG
Instrumentation: Reverse-phase HPLC analysis was performed on a
Waters apparatus equipped with a Agilent column (SB-Zorbax, 150”
4.6mm, 5 mm), a Waters 2996photodiode array detector at fixed wave-
length of 254 nm and a Waters 600 controller. The chromatographic
system used for analytical experiments consisted of acetonitrile and
water as eluents [linear gradient from 0 to 5% of acetonitrile (30 min),
from 5% to 15% of acetonitrile (50 min), 5 min at 15:85 (CH3CN/H2O)
4.8 Hz, 1H; H2’’); 13C NMR (75.5 MHz, [D7]DMSO): d=44.8 (C2’), 64.9
(C5’), 69.7 (C3’), 84.7 (C1’), 90.1 (C4’), 123.8 (C5), 145.1 (C8), 145.6(C4),
146.2 (C2), 156.9 ppm (C=O); MS (ESI): m/z: 251 [M+H]+; HRMS
(FAB): calcd for C10H10N4O4: 251.0780; found: 251.0764 [M+H]+.
AHCTREUNG
AHCTREUNG
and from 15% to 0% of acetonitrile (60 min)] at
0.7 mLminꢀ1
a flow rate of
AHCTREUNG
.
A
Pulse radiolysis: Pulse radiolysis with optical absorption detection was
performed on the 12 MeV linear accelerator, which delivered 20–200 ns
C
ACHTREUNG
AHCTRE(GNU H2’’,H1’)=4.8, JAHCTRE(UNG H2’’,H3’)=4.1 Hz, 1H; H2’’); MS (ESI): m/z: 251
C
C
[M+H]+; HRMS (FAB): calcd for C10H10N4O4: 251.0780; found:
251.0769 [M+H]+.
electron pulses with doses between 5 and 50 Gy. This generated HO , H ,
ꢀ
and eaq in concentrations of 1–20 mm. The pulse irradiations were per-
formed at room temperature (22Æ28C) on samples contained in Spectro-
sil quartz cells of 2 cm optical path length. Solutions were protected from
the analyzing light by means of a shutter and appropriate cutoff filters.
The bandwidth used throughout the pulse radiolysis experiments was
5 nm. The radiation dose per pulse was monitored by means of a charge
collector placed behind the irradiation cell and calibrated with a N2O-sa-
Computational details: Hybrid meta-DFT calculations with the BB1K
(Becke 88[38]-Becke 95[39] 1-parameter model for kinetics) functional[40]
were carried out using the Gaussian 03 system of programs.[41] This
HMDFT model was very recently tailored to give good reaction barrier
heights.[40] An unrestricted wave function was used for radical species.
Total energies were obtained employing the valence double-z basis set
supplemented with polarization functions.[42,43] Addition of standard dif-
fuse functions on heavy atoms[44,45] to better describe the anion states
were found to give unreliable results as expected for radical anions that
are unstable in the gas phase.[46] For example, the extra electron in the
radical anion of 3 is localized at the diffuse s atomic orbitals of the
carbon and nitrogen atoms of the C2-N10-H2 group. The use of moder-
ate diffuse functions with a scaling factor l of 0.7 provides reliable results
according to the criterions previously outlined.[47] Hence the BB1K calcu-
lations were carried out with the 6-31+(l=0.7)G** basis set. The geom-
etry and the energy of the anions were calculated at fixed values of the
ꢀ
turated solution containing 0.1m HCO2 and 0.5 mm methyl viologen
with Ge=9.66”10ꢀ4 m2 Jꢀ1 at 602 nm.[36] G(X) represents the number of
moles of species X formed or consumed per joule of energy absorbed by
the system.
Continuous radiolysis: Continuous radiolysis was performed at room
temperature (22Æ28C) on 4 mL samples in a 60Co-Gammacell with dose
rates of 12–13 Gyminꢀ1. The absorbed radiation dose was determined
[37]
with the Fricke chemical dosimeter by taking G
A
.
ꢀ
The reactions of 8 with eaq were investigated in deaerated aqueous solu-
tions containing 1.5 mm substrate and 0.25m tBuOH in the presence or
absence of K4[Fe(CN)6] (4 mm) at pHꢁ7. The crude reaction mixture
was passed through an ion-exchange resin (Amberlite IRA-400), to elimi-
nate the iron salts, and monitored by HPLC on a C18-reverse-phase
column as described above. Products were identified and quantified by
comparison with authentic samples.
ꢀ
C8 Br bond length. An increment step of 0.005 Š was used near the
crossing points.
Steady-state photolysis: Aqueous solutions (10 mL) containing bromide 8
(ꢁ1 mm) were irradiated under N2 through quartz inside a Luzchem mul-
tilamp photoreactor, with light from twelve 8 W lamps emitting mainly at
l=254 nm. The two cyclic diastereoisomers 21 and 23 were obtained as
pure materials from the crude reaction mixture (by irradiation of 8 in the
presence of iodide ion) employing preparative reverse-phase HPLC
(Waters column, Spherisorb, 250”10 mm, 10 mm) and were spectroscopi-
cally characterized.
Acknowledgements
This work was supported in part by the European Communityꢀs Marie
Curie Research Training Network under contract MRTN-CT-2003-
505086(CLUSTOXDNA). We also thank the financial support given by
the Spanish MCYT (CTQ 2004-03811), the Generalitat Valenciana
(Grupos 03/082, CTBPRB/2003/68, and Project GV04 A-349), the UPV
(Project PPI-06-03), and Laboratori Fitocosmesi
(Milan). We thank A. Monti and A. Martelli for assistance with pulse ra-
diolysis experiments.
Synthesis of 8-Bromo-2’-deoxyinosine (8): Sodium nitrite (ꢁ20 equiv)
was added to a suspension of 1 (0.303 mmol) in glacial acetic acid
(5 mL), and the mixture was stirred for 1.5 h. At the end of the reaction,
the solution was neutralized with 1m NaOH (ꢁ82 mL) and evaporated
to dryness in vacuo. The residue was purified by using reverse-phase
chromatography (silica gel, aqueous acetonitrile (0 to 5%)). Two frac-
tions were obtained: 8-bromohypoxanthine (27) and partially impure 8.
The latter was recrystallized from acetonitrile/water to obtain the pure
compound. The overall yield for 8 was ꢁ30%. 1H NMR (300 MHz,
& Farmaceutici srl
[1] R. Flyunt, R. Bazzanini, C. Chatgilialoglu, Q. G. Mulazzani, J. Am.
Chem. Soc. 2000, 122, 4225–4226.
[2] C. Chatgilialoglu, M. Guerra, Q. G. Mulazzani, J. Am. Chem. Soc.
2003, 125, 3839–3848.
[3] C. Chatgilialoglu, M. Duca, C. Ferreri, M. Guerra, M. Ioele, Q. G.
Mulazzani, H. Strittmatter, B. Giese, Chem. Eur. J. 2004, 10, 1249–
1255.
[4] M. Ioele, R. Bazzanini, C. Chatgilialoglu, Q. G. Mulazzani, J. Am.
Chem. Soc. 2000, 122, 1900–1907.
[D7]DMSO): d=8.08 (s, 1H; H2), 6.28 (t, J
5.40 (d, J=4.5 Hz, 1H; OH3’), 4.90 (t, J=5.9 Hz, 1H; OH5’), 4.45 (m,
1H; H3’), 3.84 (td, J(H4’,H5’)=5.2 Hz, J(H4’,H3’)=3.3 Hz, 1H; H4’),
3.62 (dt, J (H5’,H4’)=5.2 Hz, 1H; H5’), 3.5 (m,
(H5’,H5’’)=ꢀ12.0 Hz, J
1H; H5’’), 3.17 (m, 1H; H2’), 2.23 ppm (ddd, J
(H2’’,H2’)=ꢀ13.3 Hz, J-
(H2’’,H3’)=3.3 Hz, J
(H2’’,H1’)=7.1 Hz, 1H; H2’’); 13C NMR (75.5 MHz,
ACHTRE(UNG H’,H2’’)=7.1 Hz, 1H; H1’),
A
ACHTREUNG
A
ACHTREUNG
AHCTREUNG
A
ACHTREUNG
[D7]DMSO): d=37.0 (C2’), 61.9 (C5’), 70.9 (C3’), 86.0 (C1’), 88.1 (C4’),
125.4 (C5), 125.6(C8), 146.0 (C4), 149.1 (C2), 155.3 ppm (C =O); UV/
Vis: lmax (e)=255 nm (1.35”104 mꢀ1 cmꢀ1) at pH 7; MS (ESI): m/z: 331
[5] C. Chatgilialoglu, C. Caminal, M. Guerra, Q. G. Mulazzani, Angew.
Chem. 2005, 117, 6184–6186; Angew. Chem. Int. Ed. 2005, 44, 6030–
6032.
7692
www.chemeurj.org
ꢁ 2006Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7684 – 7693

Chemical Radiation Studies
FULL PAPER
[6] C. Chatgilialoglu, C. Caminal, A. Altieri, G. C. Vougioukalakis,
Q. G. Mulazzani, T. Gimisis, M. Guerra, J. Am. Chem. Soc. 2006,
128, in press.
[7] L. B. Jimenez, S. Encinas, M. A. Miranda, M. L. Navacchia, C. Chat-
gilialoglu, Photochem. Photobiol. Sci. 2004, 3, 1042–1046.
[8] T. Kimura, K. Kawai, S. Tojo, T. Majima, J. Org. Chem. 2004, 69,
1169–1173.
[9] M. de ChampdorØ, L. De Napoli, D. Montesarchio, G. Piccialli, C.
Caminal, Q. G. Mulazzani, M. L. Navacchia, C. Chatgilialoglu,
Chem. Commun. 2004, 1756–1757.
[27] T. Watanabe, K. Honda, J. Phys. Chem. 1982, 86, 2617–2619.
[28] For kinetic details of a similar mechanistic scheme, see: C. A. Kelly,
E. L. Blinn, N. Camaioni, M. DꢀAngelantonio, Q. G. Mulazzani,
Inorg. Chem. 1999, 38, 1579–1584; erratum: C. A. Kelly, E. L.
Blinn, N. Camaioni, M. DꢀAngelantonio, Q. G. Mulazzani, Inorg.
Chem. 1999, 38, 2756.
[29] G. E. Adams, R. L. Willson, Trans. Faraday Soc. 1969, 65, 2981–
2987.
[30] S. Steenken, M. J. Davies, B. C. Gilbert, J. Chem. Soc. Perkin Trans.
2 1986, 1003–1010.
[10] A. Manetto, S. Breeger, C. Chatgilialoglu, T. Carell, Angew. Chem.
2006, 118, 325–328; Angew. Chem. Int. Ed. 2006, 45, 318–321.
[11] For selected recent papers on the transfer of excess electrons in
DNA using 5-bromouracil moieties as the detection system, see: T.
Ito, S. E. Rokita, J. Am. Chem. Soc. 2003, 125, 11480–11481; T. Ito,
S. E. Rokita, Angew. Chem. 2004, 116, 1875–1878; Angew. Chem.
Int. Ed. 2004, 43, 1839–1842; T. Ito, S. E. Rokita, J. Am. Chem. Soc.
2004, 126, 15552–15559; C. Wagner, H.-A. Wagenknecht, Chem.
Eur. J. 2005, 11, 1871–1876; P. Kaden, E. Mayer-Enthart, A. Trifo-
nov, T. Fiebing, H.-A. Wagenknecht, Angew. Chem. 2005, 117,
1620–1623; Angew. Chem. Int. Ed. 2005, 44, 1636–1639; E. Mayer-
Enthart, P. Kaden, H.-A. Wagenknecht, Biochemistry 2005, 44,
11749–11757.
[31] J. Grodkowski, P. Neta, C. J. Schlesener, J. K. Kochi, J. Phys. Chem.
1985, 89, 4373–4378.
[32] g-Irradiation of aqueous solutions of K4[Fe(CN)6] (4 mm) (reduc-
tant) continuously generates micromolar levels of the oxidant
[Fe(CN)6]3ꢀ. Such a low concentration of oxidant should allow radi-
cal 10 to cyclize to radical 11 before being oxidized, thus improving
the overall reaction performance.[2]
[33] A. Romieu, D. Gasparutto, J. Cadet, Chem. Res. Toxicol. 1999, 12,
412–421.
[34] a) L. P. Candeias, S. Steenken, J. Phys. Chem. 1992, 96, 937–944;
b) L. P. Candeias, P. Wolf, P. OꢀNeill, S. Steenken, J. Phys. Chem.
1992, 96, 10302–10307.
[35] D. Enders, I. Breuer, E. Drosdow, Synthesis 2005, 3239–3244.
[36] G. V. Buxton, Q. G. Mulazzani in Electron Transfer in Chemistry.
Vol. 1: Principles, Theories, Methods and Techniques (Ed.: V. Balza-
ni), Wiley-VCH, Weinheim, 2001, pp. 503–557.
[37] J. W. T. Spinks, R. J. Woods, An Introduction to Radiation Chemistry,
3rd ed., Wiley, New York, 1990, p. 100.
[12] R. E. Holmes, R. K. Robins, J. Am. Chem. Soc. 1964, 86, 1242–1245.
[13] F. Seela, C. Mittelbach, Nucleosides Nucleotides 1999, 18, 425–441.
[14] C. T. Aravindakumar, H. Mohan, M. Mudaliar, B. S. M. Rao, J. P.
Mittal, M. N. Schuchmann, C. von Sonntag, Int. J. Radiat. Biol. 1994,
66, 351–365.
[15] a) L. P. Candeias, S. Steenken, J. Am. Chem. Soc. 1989, 111, 1094–
1099; b) S. V. Jovanovic, M. G. Simic, J. Phys. Chem. 1986, 90, 974–
978.
[16] a) G. V. Buxton, C. L. Greenstock, W. P. Helman, A. B. Ross, J.
Phys. Chem. Ref. Data 1988, 17, 513–886; b) A. B. Ross, W. G. Mal-
lard, W. P. Helman, G. V. Buxton, R. E. Huie, P. Neta, NDRL-NIST
Solution Kinetic Database—Version 3, Notre Dame Radiation Labo-
ratory, Notre Dame, IN and NIST Standard Reference Data, Gai-
thersburg, MD, 1998.
[38] A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100.
[39] A. D. Becke, J. Chem. Phys. 1996, 104, 1040–1046.
[40] Y. Zhao, B. J. Lynch, D. G. Truhlar, J. Phys. Chem. A 2004, 108,
2715–2719.
[41] Gaussian 03, Revision B.5, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr.,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V.
G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V.
Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefa-
nov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong,
C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford CT, 2004.
[42] P. C. Hariharan, J. A Pople, Theor. Chim. Acta 1973, 28, 213–222.
[43] M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon,
D. J. DeFrees, J. A. Pople, J. Chem. Phys. 1982, 77, 3654–3665.
[44] T. Clark, J. Chandrasekhar, G. W. Spitznagel, P. von R. Schleyer, J.
Comput. Chem. 1983, 4, 294–301.
[17] L. Wojnꢂrovits, E. Takꢂcs, K. Dajka, S. S. Emmi, M. Russo, M.
DꢀAngelantonio, Radiat. Phys. Chem. 2004, 69, 217–219.
C
[18] The rate constants for the reaction of the HO radical with inosine is
[16]
4.4”109 mꢀ1 sꢀ1
.
A similar value can be assumed for the analogous
reaction of 8-bromo derivatives 8 or 12. Therefore, more than 97%
C
of HO radicals are expected to be scavenged by tBuOH when [8] or
[12]ꢂ1”10ꢀ3 m.
[19] G. L. Hug, Natl. Stand. Ref. Data Ser. (US Natl. Bur. Stand.) 1981,
69, 1–159.
[20] For example, see: G. V. Buxton, Q. G. Mulazzani, A. B. Ross, J.
Phys. Chem. Ref. Data 1995, 24, 1055–1349.
[21] a) All redox potentials are versus NHE; b) P. Wardman, J. Phys.
Chem. Ref. Data 1989, 18, 1637–1755.
[22] S. Fujita, S. Steenken, J. Am. Chem. Soc. 1981, 103, 2540–2545; S.
Steenken, A. J. S. C. Vieira, Angew. Chem. 2001, 113, 581–583;
Angew. Chem. Int. Ed. 2001, 40, 571–573.
[23] P. S. Rao, E. Hayon, J. Phys. Chem. 1975, 79, 1063–1066.
[24] Generally, electrophilic-type alkyl radicals, such as a-acyl deriva-
[45] P. M. W. Gill, B. G. Johnson, J. A. Pople, M. J. Frisch, Chem. Phys.
Lett. 1992, 197, 499–505.
[46] M. Guerra, Chem. Phys. Lett. 1990, 167, 315–319.
[47] M. Guerra, J. Phys. Chem. A 1999, 103, 5983–5988.
tives, react with TMPD with a rate constant of ꢁ108 mꢀ1 sꢀ1. The rate
8
constant of H with TMPD is reported to be 2.2”10 mꢀ1 sꢀ1 at pH 1
C
where TMPD is doubly protonated.[16]
[25] Under the conditions employed: 1) the competition between 12 and
ꢀ
TMPD for eaq is well in favor of 12 because k(eaqꢀ+TMPD)=9.1”
107 mꢀ1 sꢀ1, (ref. [23]), and [12]ꢃ10”[TMPD], and 2) although the
[16]
rate constant of HO with TMPD is 1.0”1010 mꢀ1 sꢀ1
,
Received: January 10, 2006
Published online: July 5, 2006
C
C
the HO radi-
cals are still scavenged efficiently by tBuOH.[18]
[26] K. E. Hausler, W. J. Lorenz in Standard Potentials in Aqueous Solu-
tion (Eds.: A. J. Bard, R. Parsons, J. Jordan), Marcel Dekker, New
York, 1985, p. 408.
Please note: Minor changes have been made to this manuscript since its
publication in Chemistry—A European Journal Early View. The Editor.
Chem. Eur. J. 2006, 12, 7684 – 7693
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7693