Singlet oxygen oxidation of 2′-deoxyguanosine. Formation and mechanistic insights

Article abstract of DOI:10.1016/j.tet.2006.08.097

Emphasis was placed in this work on the delineation of mechanistic aspects of the singlet oxygen-mediated oxidation reactions of 2′-deoxyguanosine 1 used as a DNA model compound in aerated aqueous solution. For this purpose a thermolabile naphthalene endoperoxide derivative was used allowing the generation of [¹⁸O]-labeled singlet oxygen for dedicated mechanistic studies. The analysis and characterization of the oxidized nucleosides of the ¹O₂ reactions were achieved on the basis of accurate HPLC-tandem mass spectrometry measurements. Thus it was found that primary oxidation products include, in addition to the previously identified 8-oxo-7,8-dihydro-2′-deoxyguanosine 5 and the two diastereomers of spiroiminodihydantoin 8, two relatively minor nucleosides, namely the two diastereomers of 4-hydroxy-8-oxo-4,8-dihydro-2′-deoxyguanosine 9.

Full text of DOI:10.1016/j.tet.2006.08.097

Tetrahedron 62 (2006) 10709–10715
Singlet oxygen oxidation of 20-deoxyguanosine. Formation
and mechanistic insights
Jean-Luc Ravanat,a Glaucia R. Martinez,b,c Marisa H. G. Medeiros,c
Paolo Di Mascioc and Jean Cadeta,
*
a
ꢀ
´
Laboratoire ‘Lesions des Acides Nucleiques’, DRFMC/LCIB UMR-E n 3 CEA/UJF, CEA/Grenoble,
´
F-38054 Grenoble cedex 9, France
Departamento de Bioquımica e Biologia Molecular, Setor de Cie^ncias Bioloꢀgicas, Universidade Federal do Paranaꢀ,
Curitiba-PR, Brazil
b
´
c
´
Departamento de Bioquımica, Instituto de Quımica, Universidade de Sao Paulo, CP 26077,
´
~
CEP 05513-970 Sa~o Paulo, SP, Brazil
Received 7 July 2006; revised 9 August 2006; accepted 16 August 2006
Available online 25 September 2006
Abstract—Emphasis was placed in this work on the delineation of mechanistic aspects of the singlet oxygen-mediated oxidation reactions of
20-deoxyguanosine 1 used as a DNA model compound in aerated aqueous solution. For this purpose a thermolabile naphthalene endoperoxide
derivative was used allowing the generation of [18O]-labeled singlet oxygen for dedicated mechanistic studies. The analysis and char-
1
acterization of the oxidized nucleosides of the O2 reactions were achieved on the basis of accurate HPLC–tandem mass spectrometry
measurements. Thus it was found that primary oxidation products include, in addition to the previously identified 8-oxo-7,8-dihydro-20-
deoxyguanosine 5 and the two diastereomers of spiroiminodihydantoin 8, two relatively minor nucleosides, namely the two diastereomers
of 4-hydroxy-8-oxo-4,8-dihydro-20-deoxyguanosine 9.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
pointed out that the contribution of the chemical quenching
to the overall kT value is rather low, in fact less than 2%
as inferred from the measurement of the kr rate constant of
guanosine with 1O2 in acetone-d6 that was found to be
1.36ꢁ105 Mꢂ1 sꢂ1.5 Earliest experiments have shown that
guanine components including nucleosides and nucleotides
Guanine is the only normal nucleic acid base that signifi-
cantly reacts with singlet oxygen (1O2) in the Dg state
1
(E¼22.4 kcal) at neutral pH.1 This is illustrated by the com-
1
parison of the values of the total rate constant (kT) of O2
with nucleic acid components that is the sum of chemical
1
were highly susceptible to O2 generated by type II photo-
quenching (kr) and physical quenching (kq). The reactivity
of the main DNA and RNA nucleosides toward O2 in or-
sensitizers, hydrogen peroxide–sodium hypochlorite reac-
tion, peroxochromate or microwave electric discharge.6–14
Several degradation products including 2-deoxy-D-erythro-
pentose and the 20-deoxyribonucleosides of urea, cyanuric
acid, and oxamide whose formation is not very informative
from a mechanistic point of view were isolated and charac-
terized.13,14 Almost 15 years were necessary to gain relevant
1
ganic solvents was found to decrease in the following order:
guanine[cytosine>adenine>uracil>thymine.2,3 Similar
observations were made in water in which the measured
value of kT for guanosine4 was established to be 5.3ꢁ
106 Mꢂ1 sꢂ1, which is close to the value determined
for 20,30,50-O-tert-butyldimethylsilylguanosine in 1,1,2-tri-
chlorotrifluoroethane (kT¼3.0ꢁ106 Mꢂ1 sꢂ1).2 It may be
1
insights into the mechanism of O2 oxidation of guanine
components. The isolation and identification of the main
degradation products of guanine nucleosides have been
facilitated by the availability of highly resolutive HPLC
analytical tools. This has involved the use of suitable LC col-
umns such as those packed with an amino substituted silica
gel stationary phase that have been shown to be appropriate
to achieve the separation of the rather polar spiroiminodihy-
dantoin 20-deoxyribonucleoside diastereomers 8 (dSp).15–17
Major progress has been made in the measurement of analy-
tes at the output of the HPLC columns with the advent of
Keywords: Singlet oxygen; Cycloaddition; Endoperoxide; Dioxetane;
8-Oxo-7,8-dihydroguanine; Spiroiminodihydantoin.
Abbreviations: dGuo, 20-deoxyguanosine; 8-oxodGuo, 8-oxo-7,8-dihydro-
20-deoxyguanosine; dSp, spiroiminodihydantoin 20-deoxyribonucleoside;
4-OH-8-oxodGuo,
4-hydroxy-8-oxo-4,8-dihydro-20-deoxyguanosine;
DHPNO2, N,N0-di(2,3-dihydroxypropyl)-1,4-naphthalenedipropanamide
endoperoxide.
* Corresponding author. Tel.: +33 4 38 78 49 87; fax: +33 4 38 78 50 90;
e-mail: jean.cadet@cea.fr
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.08.097

10710
J.-L. Ravanat et al. / Tetrahedron 62 (2006) 10709–10715
O
O
N
O
O
N
1O2
N
N
N
HN
H2N
HN
H2N
N
H
HN
H2N
red
H2N
N
N
HN
O
O
N
dR
OH
N
O
N
dR
N
OH
dR
dR
1
2
3
4
-H2O
O
O
O
N
H
N
N
N
N
HN
H2N
HN
HN
HN
O
O
O
N
OH
N
N
N
H2N
dR
dR
dR
5
6
9
+H2O
O
OO
OH
H
N
HN
NH
N
O
N
N
HN
O
N
dR
H2N
N
H
dR
8
7
Scheme 1. Degradation pathway of 20-deoxyguanosine (1) mediated by singlet oxygen.
the tandem mass spectrometry detection technique involv-
ing an electrospray ionization source (EIS). This allows
accurate and sensitive quantitation of the stable primary
oxidation products of 20-deoxyguanosine (1) (dGuo) under
conditions where the formation of secondary decomposi-
tion compounds of 8-oxo-7,8-dihydro-20-deoxyguanosine
was found in earlier studies that UVA excited photosensi-
tizers including phthalocyanine and methylene blue, two
predominant type II (1O2) photosensitizers, are able to effi-
ciently degrade 20-deoxyguanosine (1) giving rise to two
main oxidized nucleosides.15–17 The structures of the two
compounds that were tentatively assigned as a pair of
(4R*)- and (4S*)-diastereomers of 4-hydroxy-8-oxo-4,8-
dihydro-20-deoxyguanosine (9)15–17 have been subsequently
revisited. Thus, it was reported that the base moiety of the
1
(5) (8-oxodGuo), which is a much better target for O2
than the precursor 1, is minimized.17 As discussed below
in more detail it was found that in addition to relatively
1
1
minor 8-oxodGuo 5, the two main O2 oxidation products
main O2 oxidation products of guanosine in aerated aque-
of dGuo 1 were the two diastereomers of dSp 818
(Scheme 1) and not the two diastereomeric forms of 4-hy-
droxy-8-oxo-4,8-dihydro-20-deoxyguanosine (9) as proposed
initially.15–19
ous solution has the same spiroiminodihydantoin structure18
that the overwhelming modified nucleosides that arise from
one-electron oxidation of 8-oxo-7,8-dihydroguanosine19–21
(structure 8, Scheme 1). Further support for the spirocyclic
connectivity of related dSp 20-deoxyribonucleosides 8 was
provided from SELINQUATE 13C NMR measurements,
a method allowing assessment of carbon connectivity
through the consideration of 13C scalar couplings.22 It may
be also pointed out that relevant chemical and conforma-
tional features on dSp diastereomers 8 either as free nucleo-
sides or when inserted into DNA duplexes as inferred from
theoretical calculations are now available.23–25
In the present work, evidence is provided from tandem mass
spectrometry measurement that the pair of diastereomers 9 is
in fact generated as primary 1O2 oxidation products of dGuo
1. Efforts were also made, using the chemical precursor of
labeled singlet oxygen, to gain further insights into the
mechanism of formation of the two diastereomers of dSp 8
that involves either a direct oxidation of dGuo 1 or a second-
1
ary efficient reaction of O2 with 8-oxodGuo 5 initially
generated.
Further investigation on the singlet oxygen oxidation
reactions of dGuo 1 was recently performed using N,N0-
di(2,3-dihydroxypropyl)-1,4-naphthalenedipropanamide
1
26
2. Results and discussion
endoperoxide (DHPNO2) as a clean source of O2 and
HPLC–MS/MS as the analytical tool for the accurate detec-
tion of the degradation products. Thus, the presence of two
major peaks in the elution profile provided by HPLC analy-
sis on a graphite column was observed using the neutral loss
scan mode detection. This is based on the search of frag-
ments corresponding to a constant loss of 116 amu that is
characteristic of the fragmentation of most 20-deoxyribo-
nucleosides.27 It may be noted that the MS spectrum of the
targeted nucleosides exhibits a signal corresponding to the
2.1. Formation of the 4R* and 4S* diastereomers of
4-hydroxy-8-oxo-4,8-dihydro-20-deoxyguanosine (9)
Numerous investigations have been aimed at elucidating the
mechanism of 1O2 oxidation reactions of the guanine moiety
of nucleosides and other nucleic acid components in aqueous
solutions during the last twenty years in relation with the
known genotoxic effects of this reactive oxygen species. It

J.-L. Ravanat et al. / Tetrahedron 62 (2006) 10709–10715
10711
molecular ion [M+H]+ at m/z¼300 in agreement with the
dSp structure 8. As relevant fragments that were acquired
with the cone potential at 30 V we may note those of the pro-
tonated base [BH+H]+ at m/z¼184 and the 2-deoxyribose
residue [dR]+ at m/z¼117. Interestingly, two new oxidized
nucleosides that exhibit similar ESI-MS/MS fragmentation
than the predominant dSp diastereomers 8 were found to
be present in the reaction mixture (Fig. 1).
1O2. This is strongly indicative that they are primary oxida-
tion products of singlet oxygen oxidation of dGuo 1. An-
other relevant piece of information was gained from the
experiments that have involved the reaction of dGuo 1
1
with isotopically labeled O2 generated from thermolabile
DHPN18O2. HPLC–MS/MS analysis indicates that both
nucleosides have incorporated two labeled oxygen atoms
(m/z¼304/188). On the basis of these observations and
taking into consideration the fact that only two diastereomers
of 8 are possible, the two newly isolated dGuo oxidation
products were assigned as the (4R*)- and (4S*)-diastereo-
mers of 4-hydroxy-8-oxo-4,8-dihydro-20-deoxyguanosine
(9). It may also be added that the third possible structure
that could exhibit similar MS features, namely 5-hydroxy-
8-oxo-7,8-dihydro-20-deoxyguanosine regioisomer 7, the
precursor of dSp 8, was ruled out due to its high instability
at room temperature.21 The ESI-MS patterns of 8 and 9
that are reported in Figure 2 show similar fragmentation pat-
terns that are dominated by the cleavage of the N-glycosidic
bond and further splitting of the 2-deoxyribose moiety as is
usually observed for 20-deoxyribonucleosides. However, we
note the presence of a specific fragment at m/z¼210.2 in the
massspectrumofdSp8, which hasnotyetbeencharacterized.
The detection of the two latter nucleosides was achieved by
using the specific transition 300/184 that was previously
applied to detect 8 by HPLC–MS/MS in the multiple reac-
tion monitoring (MRM) mode.28 The two new oxidized
nucleosides 9, which are produced in a relatively low yield
compared to dSp 8 were found to be less rapidly eluted on
the highly hydrophobic graphite HPLC column than the
two diastereomers of dSp (8) (Fig. 1). It may be noted that
these two compounds were not previously detected when
the amino substituted silica gel column was used for the sep-
arations of the 1O2 oxidation products of dGuo 1. This may
be explained by the fact that under these analytical condi-
tions the minor oxidation products co-elute with the over-
whelming 20-deoxyguanosine (1). It may be added that the
two latter products 9, together with dSp diastereomers 8
are not retained on the usual octadecylsilyl silica gel col-
umns even when water was used as the mobile phase. This
most likely explains why the two oxidized dGuo decompo-
sition compounds 9 were not detected so far despite exten-
sive investigations. Several pieces of information that were
inferred from complementary studies were considered for
the assignment of the nucleosides. First, it was found in con-
trast to what was observed for diastereomers 8 (vide infra)
that the two products are not formed upon incubation of
8-oxodGuo 5 in the presence of the chemical generator of
The formation of 9 as already pointed out is a relatively mi-
nor process of the 1O2-mediated oxidation of dGuo 1. Inter-
estingly, this provides further support for the occurrence of
a second mode of degradation pathway of 1,4-endoperoxides
2 that was initially proposed and then discarded. It would be
of interest to check in forthcoming studies whether nucleo-
sides 9 are produced in DNA and oligonucleotides.
2.2. Mechanistic aspects on the formation of spiro-
iminodihydantoin nucleosides as the result of
primary and secondary oxidation reactions
The use of a clean source of 1O2 that can be isotopically la-
beled with 18O atoms has been shown to be particularly suit-
able for mechanistic studies dedicated to oxidation of 1 and 5
in aqueous solution.29 As already discussed it is well docu-
O O
HN
NH
N
H
N
HN
O
1
dR
mented that O2 oxidation of dGuo 1 in water at neutral
pH gives rise to the predominant formation of a diastereo-
meric pair of 8.29–31 The HPLC–MS/MS analysis of the
two dSp nucleosides generated upon the incubation of
dGuo 1 with DHPN18O2, as a generator of 18[1O2], showed
the presence of two sets of molecules (Fig. 3).
dSp 8
One is labeled with two [18O]-labeled atoms whereas only
one 18O has been incorporated in the second molecule during
the 18[1O2] oxidation of dGuo 1. The formation of the latter
nucleosides 8 whose ESI-MS spectra exhibit a molecular ion
at m/z¼302 may be rationalized in terms of initial 1O2 cyclo-
addition across the 4,8-bond leading to the transient forma-
tion of an oxidized quinonoid 6 that upon reaction with
a water molecule gives rise to 8 as shown in Scheme 1. There
is another route leading to dSp 8 through the intermediacy of
5, which has been found to be a primary 1O2 oxidation prod-
uct of dGuo 1. The observation of a doubly [18O]-labeled
dSp molecule 8 as inferred from the presence of a pseudo-
molecular ion at m/z¼304 in the HPLC–mass spectrum is
fully supportive of the significant involvement of a secondary
oxidation process in the formation of 8. This is in agreement
with the fact that 8-oxodGuo 5, which is more susceptible
4-OH-8-oxodGuo 9
O
N
HN
O
N
H2N
N
OH
dR
0
2
4
6
8
10
12
14
Time, min
Figure 1. HPLC–MS/MS chromatogram obtained by analysis of oxidation
products of 20-deoxyguanosine (1) upon incubation with DHPNO2. The MS/
MS detection was achieved using the specific transition 300/184 for dSp
(8). Separation was performed using the Hypercarb column (see Section 4).

10712
J.-L. Ravanat et al. / Tetrahedron 62 (2006) 10709–10715
184.2
OO
117.2
HN
NH
N
N
dR
HN
O
H
dSp 9
300.2
99.1
100
210.2
80.7
60
140
18
0
2
20
26
0
300
m/z, amu
184.2
O
N
N
N
HN
O
H2N
OH
dR
4-OH-8-oxodGuo 8
117.2
300.2
99.2
100
80.7
60
140
18
0
2
20
26
0
300
m/z, amu
Figure 2. Collision-induced dissociation mass spectra of ion m/z¼300 (positive mode) of either dSp 8 (upper panel) or 4-OH-8-oxodGuo 9 (lower panel).
than dGuo 1 to 18O oxidation almost by two orders of mag-
5-hydroxy-8-oxo-7,8-dihydro-20-deoxyguanosine (7), the
unstable regioisomer of 9. Subsequent acyl shift rearrange-
ment, which is favored at neutral and alkaline pHs, is likely
to explain the formation of 8. Strong support for this
pathway was provided by the observation of key intermedi-
ates including related ribonucleoside derivatives of dioxe-
tane 10 and 5-hydroperoxide 11 in the photosensitized
organic solution of the 20,30,50-tri-O-tert-butyl dimethylsilyl
nitude,4 is gradually consumed as soon it is produced in a
1
reaction with O2.15 The secondary formation of 8 may be
accounted for by the initial formation of a diastereomeric
mixture of dioxetane 10 as the result of a [2+2] cycloaddition
of O2 across the 4,5-ethylenic bond of 5. In a subsequent
1
step, 10 is likely to rearrange giving rise to the 5-hydroper-
oxide 11, which upon reduction leads to the formation of
O
O
N
O
N
18[1O2]
18[1O2]
N
N
N
N
N
HN
HN
18O
HN
18OH
N
dR
HN
N
H2N
H2N
dR
dR
H216O
18[1O2]
302
304
18O
O
16O
O
100
HN
NH
HN
NH
N
H
N
dR
N
H
N
dR
18O
18O
HN
HN
0
245 250 255 260 265 270 275 280 285 290 295 300 305 310 315 320
m/z
Figure 3. Mass spectra of dSp 8 generated upon incubation of dGuo 1 with DHPN18O2, a chemical generator of [18O]-labeled singlet oxygen.

J.-L. Ravanat et al. / Tetrahedron 62 (2006) 10709–10715
10713
derivative of 8-oxo-7,8-dihydroguanosine.31,32 This was
2.3. Mechanism of 1O2 oxidation of 20-deoxyguanosine
achieved by both 13C NMR and 2-D HMBC measurements
performed at ꢂ60 ꢀC. It may be added that support for
the assignment of the 13C chemical shifts was provided
by hybrid Hartree–Fock density functional theory cal-
culations.31 Evidence for the dimethylsulfide-mediated
reduction of 5-hydroperoxy-8-oxo-7,8-dihydro-20-deoxy-
guanosine (11) into the alcohol precursor 7 of spiroiminodi-
hydantoin nucleosides 8 was also gained from NMR
measurements carried out at ꢂ40 ꢀC.31
1
The formation of the main primary O2 oxidation products
of dGuo 1 including the two diastereomers of dSp 8, 8-
oxodGuo 5, and the recently characterized (4R*)- and (4S*)-
4-OH-8-oxodGuo 9 may be rationalized in terms of a
Diels–Alder [4+2] cycloaddition of 1O2 across the 4,8-bond
of the imidazole ring of the guanine moiety of 1. This gives
rise to unstable 4,8-endoperoxides 2 whose characterization
has not been so far achieved even at temperature lower than
ꢂ80 ꢀC. However, related endoperoxides that were gener-
ated by type II photosensitization of the 20,30,50-O-tert-butyl-
dimethylsilyl derivative of 8-methylguanosine in organic
solvents were found to be produced and their structures
assigned on the basis of NMR measurements at low tem-
perature.34,35 The endoperoxides 2 are likely to undergo in
aqueous solution a predominant rearrangement that leads
to putative 8-hydroperoxy-20-deoxyguanosine 3 for which
direct structural proofs are also lacking. In a subsequent
step, 3 may be reduced into 8-hydroxy-20-deoxyguanosine
(4), which is known to be in a dynamic equilibrium with
its 6,8-diketo tautomer namely 8-oxo-7,8-dihydro-20-deoxy-
guanosine (5) (Scheme 1). It may be mentioned that 2,6-
diamino-4-hydroxy-5-formamidopyrimidine (FapyGua), a
degradation product that may be formed by hydration of gua-
nine radical cation followed by the opening of the imidazole
ring according to a reductive pathway,36,37 is not generated
at least in detectable amount within isolated DNA.38 This
was achieved using a thermolabile naphthalene endoperox-
The delineation of mechanistic aspects concerning the for-
mation of other O2 oxidation products of 8-oxodGuo 5
1
has been the subject of recent studies that have involved
the use of either unlabeled DHPNO2 or DHPN18O2.29 It ap-
pears that the oxidized nucleosides so far identified arise
from the initial formation of dioxetane precursors 10. In ad-
dition to the conversion of 10 into the linear 5-hydroperoxide
11, a second major degradation pathway was found to occur.
Thus the cleavage of the 1,2-bond of 10 was proposed to
generate the nine-membered ring intermediate 14, which
is converted into 1,3,5-triazine-1(2H)-carboximidamide,
3-(2-deoxy-b-D-erythro-pentofuranosyl)-tetrahydro-2,4,6-tri-
oxo- (15) upon intramolecular cyclization.33 Subsequent
slow hydrolysis of the six-membered ring nucleoside 15
leads to the formation of 1-(2-deoxy-b-D-erythro-pentofura-
nosyl)-cyanuric (16) that is accompanied by the release of
urea (Scheme 2).
1
Another degradation pathway whose mechanism remains to
be established gives rise to the imidazolone nucleoside 12,
the precursor of oxazolone nucleoside 13 that has been
shown previously to be the main one-electron oxidation
product of dGuo 1 in aqueous aerated solutions.34
ide derivative as a clean chemical source of O2.26 This
has led to rule out the possibility for 1O2 to act as a one-elec-
tron oxidant in contrast to a previous proposal.39 The main
degradation pathway of 3 involves dehydration, which leads
to the formation of oxidized quinonoid 6. In a subsequent
O
N
O
OH
NH
O
NH
acyl shift
HN
H2N
O
HN
HN
O
N
red
NH
N
O
OOH
NH
HN
7
8
O
N
H2N
N
O
O
11
O
N
N
N
H2O
H2N
H2N
H2N
O
N
O
N
H
H
N
N
NH
1O2
NH
N
HN
H2N
HN
H2N
O
O
O
N
13
12
O
5
10
O
O
N
NH2
O
O
HN
HN
HN
NH
HN
N
NH
NH
HN
O
O
O
N
O
O
N
O
16
14
15
Scheme 2. Singlet oxygen oxidation of 8-oxo-7,8-dihydro-20-deoxyguanosine (5).

10714
J.-L. Ravanat et al. / Tetrahedron 62 (2006) 10709–10715
step hydration would give rise to 5-hydroxy-8-oxo-7,8-di-
hydro-20-deoxyguanosine (7) that through an acyl shift,
which is favored at neutral pH, leads to the formation of
the two diastereomers of 8. It should be added that support
for a common precursor to both 8-oxodGuo 5 and dSp 8 is
provided by the fact that the presence of reducing agent
such as thiols or Fe2+ ions leads to a significant increase
in the yield of 8-oxodGuo 5 at the expense of 8.18,40 The
second degradation pathway is likely to involve the cleavage
of the peroxidic bond of 2 with subsequent formation of
4-hydroxy-8-oxo-4,8-dihydro-20-deoxyguanosine (9).
using a mobile phase composed of 80% of acetonitrile and
20% of ammonium formate (25 mM) at a flow rate of
0.8 ml/min or using a 150ꢁ2.1 mm Hypercarb column
(Thermo Electron corporation, Cheshire, UK) using a linear
gradient (0–50% in 30 min) of acetonitrile in 2 mM ammo-
nium formate. Detection of the products was achieved using
a Shimadzu SPD-10Avp UV–visible detector (Shimadzu,
Tokyo, Japan) set at 230 nm. ESI-MS experiments were car-
ried out on a Quattro II instrument (Micromass, Manchester,
UK) or using an API3000 (Applied Biosystems, Toronto,
Canada) as described in detail elsewhere.28 The spectra
were obtained in the positive ion mode. The source temper-
ature was maintained at 100 ꢀC. The optimal flow-rates of
the drying and nebulizing gas (nitrogen) were found to be
300 and 30 L/h, respectively. The capillary and HVelectrode
potentials were set at 3.50 and 0.5 kV, respectively. The cone
voltage was set at 15 and 30 V to the ESI-MS-scan mode and
60 V for the ESI-MS/MS analysis; besides, a pressure of
5.5ꢁ10ꢂ4 mbar was used in the gas cell and the collision
energy was set at 26 eV. The data were processed and
transformed into values in a mass/charge scale by means
of the Mass Lynx NTÔ data system 3.20 version (Micro-
mass, Manchester, UK).
3. Conclusion
The use of a pure source of singlet oxygen, together with
the possibility of producing [18O]-labeled singlet oxygen, in
association with HPLC–MS/MS analysis allowed us to pro-
pose a coherent mechanism for the singlet oxygen-mediated
decomposition of dGuo 1 as a free nucleoside. Interestingly,
the formation of the two diastereomers of 4-hydroxy-8-
oxo-4,8-dihydro-20-deoxyguanosine (9) may be rationalized
in terms of cleavage of the peroxidic bond of 2, which con-
stitutes a minor reaction with respect to the predominant
rearrangement into the hydroperoxide 3. It may be noted
Acknowledgements
1
that the chemical reactions of O2 with the guanine moiety
of both isolated and cellular DNAs are partly different since
so far only 8-oxodGuo 5 has been shown to be produced in
detectable amount.41 Further efforts should be made to
The following Brazilian research funding institutions
~
including FAPESP (Fundac¸ao de Amparo a Pesquisa do
Estado de Sao Paulo), CNPq (Conselho Nacional para o
ꢁ
~
1
´
ꢀ
Desenvolvimento Cientıfico e Tecnologico), CNPq-Instituto
check for the putative O2-mediated formation in DNA of
^
other oxidized nucleosides including 8 and 9 by taking
advantage of the recent availability of more sensitive HPLC–
MS/MS apparatus and optimized conditions of enzymatic
release of the targeted lesions.
do Milenio Redoxoma, and John Simon Guggenheim Me-
morial Foundation (P.D.M. Fellowship) are acknowledged
for financial support. G.R.M. was a post-doctorate recipient
of an FAPESP fellowship. Partial support was also provided
to J.C. from the European EU Marie Curie Training
and Mobility program (project nꢀ MRTN-CT2003 ‘CLUS-
TOXDNA’).
4. Experimental
4.1. Chemicals
References and notes
DHPNO2 and DHPN18O2 were prepared as previously de-
scribed.26 Nucleosides including dGuo 1 and 8-oxodGuo 5
were obtained from Sigma (St Quentin-Fallavier, France)
and used without further purification.
1. Cadet, J.; Sage, E.; Douki, T. Mutat. Res. 2005, 571, 3.
2. Prat, F.; Houk, N.; Foote, C. S. J. Am. Chem. Soc. 1997, 119,
3951.
3. Lee, P. C. C.; Rodgers, M. A. J. Photochem. Photobiol. 1987,
45, 79.
4. Sheu, C.; Foote, C. S. J. Am. Chem. Soc. 1995, 117, 6439.
5. Sheu, C.; Kang, P.; Khan, S.; Foote, C. S. J. Am. Chem. Soc.
2002, 124, 3905.
4.2. Incubation of 20-deoxyguanosine with the
chemical source of singlet oxygen
Incubation of dGuo 1 was performed using a 5 mM solution
of the nucleoside in a deuterated aqueous solution of either
DHPNO2 or DHPN18O2 at a 20 mM concentration. The
resulting solution was heated for 2.5 h at 37 ꢀC to allow
the total decomposition of the endoperoxide.
6. Simon, M. I.; Van Vunakis, H. J. Mol. Biol. 1962, 4, 488.
7. Simon, M. I.; Van Vunakis, H. Arch. Biochem. Biophys. 1964,
105, 197.
8. Waskell, L. A.; Sastry, K. S.; Gordon, M. P. Biochim. Biophys.
Acta 1966, 129, 42.
9. Simon, M. I.; Van Vunakis, H. Biochim. Biophys. Acta 1966,
129, 49.
4.3. HPLC–MS/MS analysis
10. Hallett, P. R.; Hallett, B. P.; Snipes, W. Biophys. J. 1970, 10,
305.
11. Rosenthal, I.; Pitts, J. N., Jr. Biophys. J. 1971, 11, 963.
12. Saito, I.; Inoue, K.; Matsuura, T. Photochem. Photobiol. 1975,
21, 27.
The diastereomeric mixture of spiroiminodihydantoin 20-de-
oxyribonucleosides 8 (dSp) was analyzed using an HPLC
system that consisted of two Shimadzu LC-10AD/VP pumps
(Shimadzu, Tokyo, Japan) connected to a 7125 Rheodyne
injector valve (Rheodyne, Cotati, CA, USA). The separation
was performed either on a SupelcosilÔ (Supelco, Belle-
fonte, PA, USA) LC-NH2 column (5 mm, 250ꢁ4.6 mm),
ꢀ
13. Cadet, J.; Teoule, R. Photochem. Photobiol. 1978, 28, 661.
14. Cadet, J.; Balland, A.; Voituriez, L.; Hahn, B.-S.; Wang, S. Y.
Oxygen and Oxy-Radicals in Chemistry and Biology;

J.-L. Ravanat et al. / Tetrahedron 62 (2006) 10709–10715
10715
Rodgers, M. A. J., Powers, E. L., Eds.; Academic: New York,
NY, 1981; p 610.
28. Ravanat, J.-L.; Remaud, G.; Cadet, J. Arch. Biochem. Biophys.
2000, 374, 118.
15. Ravanat, J.-L.; Berger, M.; Benard, F.; Langlois, R.; Ouellet, R.;
van Lier, J. E.; Cadet, J. Photochem. Photobiol. 1992, 55, 809.
16. Buchko, G. W.; Cadet, J.; Berger, M.; Ravanat, J.-L. Nucleic
Acids Res. 1992, 20, 4847.
17. Ravanat, J.-L.; Cadet, J. Chem. Res. Toxicol. 1995, 8, 379.
18. Niles, J. C.; Wishnok, J. S.; Tannenbaum, S. R. Org. Lett. 2001,
3, 963.
29. Martinez, G. R.; Medeiros, M. H. G.; Ravanat, J.-L.; Cadet, J.;
Di Mascio, P. Biol. Chem. 2002, 383, 607.
30. Ye, Y.; Muller, J. G.; Luo, W.; Mayne, C. L.; Shalopp, A. J.;
Jones, R. A.; Burrows, C. J. J. Am. Chem. Soc. 2003, 125,
13926.
31. McCallum, J. E. B.; Kuniyoshi, C. Y.; Foote, C. S. J. Am. Chem.
Soc. 2004, 126, 16777.
19. Luo, W.; Muller, J. G.; Rachlin, E. M.; Burrows, C. J. Org. Lett.
2000, 2, 613.
20. Luo, W.; Muller, J. G.; Burrows, C. J. Org. Lett. 2001, 3,
2801.
21. Luo, W.; Muller, J. G.; Rachlin, E. M.; Burrows, C. J. Chem.
Res. Toxicol. 2001, 14, 927.
22. Adam, W.; Arnold, M. A.; Grune, M.; Nau, W. M.; Pischel, U.;
32. Sheu, C.; Foote, C. S. J. Am. Chem. Soc. 1995, 117, 474.
33. Raoul, S.; Cadet, J. J. Am. Chem. Soc. 1996, 118, 1892.
34. Cadet, J.; Berger, M.; Buchko, G. W.; Joshi, P. C.; Raoul, S.;
Ravanat, J.-L. J. Am. Chem. Soc. 1994, 116, 7403.
35. Kang, P.; Foote, C. S. J. Am. Chem. Soc. 2002, 124, 4865.
36. Cadet, J.; Douki, T.; Gasparutto, D.; Ravanat, J.-L. Mutat. Res.
2003, 531, 5.
€
Saha-Moller, C. R. Org. Lett. 2002, 4, 537.
37. Douki, T.; Ravanat, J.-L.; Angelov, D.; Wagner, J. R.; Cadet, J.
Top. Curr. Chem. 2004, 236, 1.
38. Ravanat, J.-L.; Saint-Pierre, C.; Di Mascio, P.; Martinez,
G. R.; Medeiros, M. H. G.; Cadet, J. Helv. Chim. Acta 2001,
84, 3702.
23. Jia, L.; Shafirovich, V.; Shapiro, R.; Geacintov, N. E.; Broyde,
S. Biochemistry 2005, 44, 6043.
24. Jia, L.; Shafirovich, V.; Shapiro, R.; Geacintov, N. E.; Broyde,
S. Biochemistry 2005, 44, 13342.
25. Karwowski, B.; Dupeyrat, F.; Bardet, M.; Ravanat, J.-L.;
Krajewski, P.; Cadet, J. Chem. Res. Toxicol., in press.
26. Martinez, G. R.; Ravanat, J.-L.; Medeiros, M. H. G.; Cadet, J.;
Di Mascio, P. J. Am. Chem. Soc. 2000, 122, 10212.
27. Regulus, P.; Spessotto, S.; Gateau, M.; Cadet, J.; Favier, A.;
Ravanat, J.-L. Rapid Commun. Mass Spectrom. 2004, 16, 2223.
39. Boiteux, S.; Gajewski, E.; Laval, J.; Dizdaroglu, M.
Biochemistry 1992, 31, 106.
40. Buchko, G. W.; Wagner, J. R.; Cadet, J.; Raoul, S.; Weinfeld,
M. Biochim. Biophys. Acta 1995, 1263, 17.
41. Ravanat, J.-L.; Di Mascio, P.; Martinez, G. R.; Medeiros,
M. H. G.; Cadet, J. J. Biol. Chem. 2000, 275, 40601.