Spiroiminodihydantoin is a major product in the photooxidation of 2′-deoxyguanosine by the triplet states and oxyl radicals generated from hydroxyacetophenone photolysis and dioxetane thermolysis

Article abstract of DOI:10.1021/ol017138m

(formula presented) Photolysis of hydroxyacetophenone and thermolysis of the corresponding dioxetane afford spiroiminodihydantoin rather than 4,8-dihydro-4-hydroxy-8-oxo-2′-deoxyguanosine (4-HO-8-oxodG) through the oxidation of 2′-deoxyguanosine (dG) by triplet-excited hydroxyacetophenone and the peroxyl radicals derived thereof by α cleavage and subsequent oxygen trapping. The structure of the spiroiminodihydantoin is assigned by the SELINQUATE NMR technique, which unequivocally establishes the spirocyclic connectivity.

Full text of DOI:10.1021/ol017138m

ORGANIC
LETTERS
2002
Vol. 4, No. 4
537-540
Spiroiminodihydantoin Is a Major
Product in the Photooxidation of
2′-Deoxyguanosine by the Triplet States
and Oxyl Radicals Generated from
Hydroxyacetophenone Photolysis and
Dioxetane Thermolysis
,†
Waldemar Adam,* Markus A. Arnold,† Matthias Gru1ne,† Werner M. Nau,‡
Uwe Pischel,‡ and Chantu R. Saha-Mo1ller†
Institut fu¨r Organische Chemie, UniVersita¨t Wu¨rzburg, Am Hubland,
D-97074 Wu¨rzburg, Germany, and Institut fu¨r Physikalische Chemie,
UniVersita¨t Basel, Klingelbergstrasse 80, CH-4056 Basel, Switzerland
adam@chemie.uni-wuerzburg.de
Received November 28, 2001
ABSTRACT
Photolysis of hydroxyacetophenone and thermolysis of the corresponding dioxetane afford spiroiminodihydantoin rather than 4,8-dihydro-4-
hydroxy-8-oxo-2′-deoxyguanosine (4-HO-8-oxodG) through the oxidation of 2′-deoxyguanosine (dG) by triplet-excited hydroxyacetophenone
and the peroxyl radicals derived thereof by r cleavage and subsequent oxygen trapping. The structure of the spiroiminodihydantoin is assigned
by the SELINQUATE NMR technique, which unequivocally establishes the spirocyclic connectivity.
The major photooxygenation product of 2′-deoxyguanosine
(dG) has previously been assigned the 4-HO-8-oxodG
structure,1 which for almost the last two decades has served
as a characteristic probe for singlet oxygen (type II photo-
oxidation)2 in photobiological studies.1c Recently, the 4-HO-
8-oxodG structure has been questioned,3 since it was shown
that this compound is identical (HPLC analysis) to spiro-
iminodihydantoin, formed in the one-electron oxidation of
8-oxo-7,8-dihydro-2′-deoxyguanosine (8-oxodG) and char-
acterized on the basis of mass spectral analysis (MS/MS
spectrum similar to that of the independently synthesized
spiro compound but without the deoxyribose residue) and
NMR spectroscopy.3,4 These spectral data would allow a
definitive differentiation between the spiroiminodihydantoin
and 4-HO-8-oxodG structures if both were available as
authentic materials. Since this is not the case, an unequivocal
assignment requires the establishment of the connectivity of
† University of Wu¨rzburg.
‡ University of Basel.
(1) (a) Cadet, J.; Decarroz, C.; Wang, S. Y.; Midden, W. R. Isr. J. Chem.
1983, 23, 420-429. (b) Cadet, J.; Berger, M.; Decarroz, C.; Wagner, J. R.;
van Lier, J. E.; Ginot, Y. M.; Vigny, P. Biochimie 1986, 68, 813-834. (c)
Ravanat, J.-L.; Berger, M.; Benard, F.; Langlois, R.; Ouellet, R.; van Lier,
J. E.; Cadet, J. Photochem. Photobiol. 1992, 55, 809-814. (d) Ravanat,
J.-L.; Cadet, J. Chem. Res. Toxicol. 1995, 8, 379-388.
(2) Foote, C. S. Photochem. Photobiol. 1991, 54, 659-659.
(3) Niles, J. C.; Wishnok, J. S.; Tannenbaum, S. R. Org. Lett. 2001, 3,
963-966.
(4) (a) Luo, W.; Muller, J. G.; Rachlin, E. M.; Burrows, C. J. Org. Lett.
2000, 2, 613-616. (b) Luo, W.; Muller, J. G.; Burrows, C. J. Org. Lett.
2001, 3, 2801-2804. (c) Luo, W.; Muller, J. G.; Rachlin, E. M.; Burrows,
C. J. Chem. Res. Toxicol. 2001, 14, 927-938.
10.1021/ol017138m CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/25/2002

the marked carbon atoms in the 4-HO-8-oxodG and spiro-
iminodihydantoin structures given below.
fragments into carbon-centered radicals. For comparison, the
photolysis of acetophenone (AP) was examined, a ketone
substrate that is known to be photoresistant toward R
cleavage into radicals.10 We conclude that the direct photo-
oxidation of the guanine in dG takes place by AP-OH triplets
(presumably electron transfer) as well as radicals generated
thereof; the latter dominate at low dG concentrations in the
presence of molecular oxygen.
The SELINQUATE NMR technique distinguished un-
equivocally between the 4-HO-8-oxodG and spiroimino-
dihydantoin, since the two structures possess different
carbon-carbon connectivities. The necessary amount of this
oxidation product for NMR analysis was obtained by
photooxygenation of dG with rose bengal.11 The 13C NMR
spectrum showed five pairs of peaks (two diastereomers),
which are attributed to the base part of the molecule, one at
δ ca. 80 ppm and the other four between δ 155 and 182
ppm (see Figure 1). The structurally definitive resonance at
This disputed oxidation product has also been observed
in the photooxidation of the nucleosides dG and 8-oxodG
by triplet-excited ketones and the radicals derived therefrom,
generated either by the thermolysis of appropriate dioxetanes
or by the photolysis of ketones.5 The fact that this potentially
photobiologically significant oxidation product6 is also
formed under the conditions of type I photooxidation
(electron transfer, hydrogen abstraction) and oxyl-radical-
mediated dG oxidation7 urges a rigorous confirmation of the
spiroiminodihydantoin structure.
For an unequivocal structural elucidation of the disputed
dG oxidation product, we have chosen the SELINQUATE8
NMR technique (selective INADEQUATE: selective incred-
ible natural abundance double quantum transfer experiment).
This technique permits the definitive assignment of the atom
connectivities through characteristic coupling patterns of
nearest neighbors in the spirocyclic and annelated structures.
As a type I photooxidant of dG, we chose triplet-excited
2-hydroxyacetophenone (AP-OH). The AP-OH triplets were
to be generated photochemically by direct excitation of the
ketone and thermally by the decomposition of the hitherto
unknown dioxetane 1 (for its preparation, see Supporting
Information). The advantage of AP-OH triplets is that the
involvement of singlet oxygen should be minimal and, thus,
type II photooxidation of no concern. Moreover, it is known
that AP-OH triplets cleave exclusively into benzoyl and
hydroxymethyl radicals,9 which on trapping with O2 afford
peroxyl radicals for the oxidation of dG. Photomechanisti-
cally relevant, the benzoyl chromophore in the AP-OH
triplets, in contrast to the acetyl chromophore in previously
used acetone derivatives,5c enables transient spectroscopy.
This advantage should allow assessment of the relative
importance of the direct photooxidation of the nucleoside
dG by the triplet-excited AP-OH (electron transfer, hydrogen
abstraction) versus oxidative damage by the radicals gener-
ated through R cleavage of the AP-OH triplets.
Figure 1. Top: 151 MHz 13C-SELINQUATE spectrum of
spiroiminodihydantoin 2′-deoxyribonucleosides in DMSO-d6 ob-
tained upon selective coherence transfer carried out by a selec-
tive 270° Gauss pulse (1 ms pulse length) at 80 ppm in the
SELINQUATE8 pulse sequence; only the region between 150 and
185 ppm of the base signals is shown. Bottom: 13C NMR spectrum
for comparison.
δ 80 ppm belongs either to the C-4 atom of 4-HO-8-oxodG1d
or to the C-5 atom of spiroiminodihydantoin4 (marked as
squares; see above structures). Selective 13C-13C coherence
transfer of this resonance resulted in distinct coupling patterns
in the SELINQUATE spectra of the two structures: while
the C-4 atom in 4-HO-8-oxodG possesses only one direct
carbon neighbor (the imino functionality), the C-5 atom in
spiroiminodihydantoin is flanked by two carbon atoms (the
carbonyl groups, marked with triangles). The SELINQUATE
spectrum (Figure 1) clearly shows two pairs of signals at δ
169.4 (52 Hz) and 169.5 ppm (52 Hz), as well as at δ 180.9
(42 Hz) and 181.5 ppm (42 Hz); the observed coupling
constants (42 and 52 Hz) reveal that these carbon atoms
Our present results disclose the efficacious photooxidation
of dG to spiroiminodihydantoin in the thermolysis of
dioxetane 1 and in the photolysis of AP-OH; both modes of
operation generate triplet-excited AP-OH and the latter
(5) (a) Adam, W.; Saha-Mo¨ller, C. R.; Scho¨nberger, A. J. Am. Chem.
Soc. 1997, 119, 719-723. (b) Adam, W.; Andler, S.; Nau, W. M.; Saha-
Mo¨ller, C. R. J. Am. Chem. Soc. 1998, 120, 3549-3559. (c) Adam, W.;
Arnold, M. A.; Saha-Mo¨ller, C. R. J. Org. Chem. 2001, 66, 597-604.
(6) Leipold, M. D.; Muller, J. G.; Burrows, C. J.; David, S. S.
Biochemistry 2000, 39, 14984-14992.
(7) Adam, W.; Kurz, A.; Saha-Mo¨ller, C. R. Chem. Res. Toxicol. 2000,
13, 1199-1207.
(8) Berger, S. Angew. Chem., Int. Ed. Engl. 1988, 27, 1196-1197.
(9) (a) Palm, W. U.; Dreeskamp, H. J. Photochem. Photobiol., A 1990,
52, 439-450. (b) Grimme, S. Chem. Phys. 1992, 163, 313-330.
(10) Xu, Y.; Langford, C. H. J. AdV. Oxid. Technol. 1997, 2, 408-414.
(11) Ravanat, J.-L.; Douki, T.; Incardona, M. F.; Cadet, J. J. Liq.
Chromatogr. 1993, 16, 3185-3202.
538
Org. Lett., Vol. 4, No. 4, 2002

1
possess J(C-C) couplings. Thus, this NMR spectral evi-
and t1/2. The results (Table 1) show that below 65 °C, the
amount of emitted light decreases, which indicates that dark
decomposition of the dioxetane dominates at low tempera-
tures. To minimize such dark decomposition of the dioxetane
1, decomposition temperatures >65 °C were employed.
The thermolysis of dioxetane 1, as well as the photolysis
of AP-OH, afforded the spiroiminodihydantoin as the major
photooxidation product (Figure 2). Besides spiroiminodi-
dence speaks against the 4-HO-8-oxodG and supports the
spiroiminodihydantoin structure. With the structure of the
2′-deoxyguanosine photooxidation product established as
spiroiminodihydantoin, it was pertinent to confirm whether
this photooxidation product is formed with triplet AP-OH,
generated by photolysis of the ketone and thermolysis of the
dioxetane 1. For this purpose, the synthesis of the dioxetane
1 was achieved from the known 2-methyl-3-phenylbut-3-
ene-2-yl hydroperoxide,12 which was obtained as a mixture
with the 2-methyl-3-phenylbut-1-ene-3-yl regioisomer by the
ene reaction of 1,2-dimethyl-1-propylbenzene with singlet
oxygen (Scheme 1). The corresponding epoxides were
Scheme 1. Synthesis of Dioxetane 1
Figure 2. Oxidation of dG (0.200 mM) in the thermolysis of
dioxetane 1 (5 equiv, 80 °C, 3 h) and the photolysis of AP-OH (2
equiv, 0 °C, 45 min) in 2.00 mM phosphate buffer:CH3CN (9:1).
prepared from the crude mixture by epoxidation with
peracetic acid, of which the 1-methyl-1-(2-phenyloxiranyl)-
ethyl hydroperoxide was isolated and purified by silica gel
chromatography. Subsequent base-catalyzed ring closure with
dilute NaOH in methylene chloride gave the dioxetane 1 in
17% yield, which was fully characterized (see Supporting
Information).
The rate of the dioxetane (1.00 mM) thermolysis was
determined by monitoring the decay of the chemilumines-
cence intensity at various temperatures in 9:1 H2O:CH3CN
(Table 1). The half-lives (t1/2) were determined by fitting
hydantoin, also 8-oxodG and guanidine-releasing products
(GRP), e.g., oxazolone,13 were formed in comparable amounts
for both cases. Thus, irrespective of the mode of triplet AP-
OH generation, a similar photooxidative reactivity was
observed in both the dioxetane thermolysis and ketone
photolysis. However, note in Figure 2 that in the photolysis
of AP-OH, the relative yields of 8-oxodG and GRP are
significantly smaller (ca. 63% for the dioxetane thermolysis
versus ca. 34% for the AP-OH photolysis) compared to the
conversion of dG. Evidently, due to the higher oxidative
reactivity (dG conversion) in the AP-OH photolysis, these
photoproducts (8-oxodG14 and GRP) are subject to further
oxidation. The similar product distributions in both cases
imply common oxidizing species, presumably radicals formed
by R cleavage of the triplet-excited state of AP-OH. The
participation of singlet oxygen in this photooxidation is
unlikely in view of the absence of a D2O effect in the
photolysis of AP-OH (data not shown).
Table 1. Half-Lives and Chemiluminescence Emission
Intensities of the Thermal Decomposition of Dioxetane 1
temp (°C)a
Imax (mV)
t1/2 (h)
Imax‚t1/2 (mV h)
37
65
80
80b
4.6 ( 0.2
310 ( 10
630 ( 30
550 ( 30
42 ( 10
4.1 ( 0.3
0.82 ( 0.04
0.86 ( 0.09
190
500
510
480
a 1.00 mM in H2O:CH3CN (9:1). b In the presence of 0.50 mM dG and
5.00 mM phosphate buffer (pH 7.0).
To assess the relative importance of the direct interaction
of dG with the triplet-excited AP-OH in the dG photo-
oxidation, half of the dG concentration was employed. No
decrease of the photooxidative reactivity (dG conversion)
at the lower dG concentration was observed for the ketone
photolysis, as well as for the dioxetane thermolysis (data
not shown). Since the direct oxidation of dG by triplet-excited
AP-OH, a bimolecular process, should diminish due to its
the data to a monoexponential function (first-order kinetics).
The maximum light intensity Imax that is reached within a
few minutes of thermal decomposition was determined
by monitoring the chemiluminescence emission with the
Mitchell-Hastings photometer. From these values, the total
light emission, which is proportional to the area under the
chemiluminescence curve, was calculated by multiplying Imax
(13) Raoul, S.; Berger, M.; Buchko, G. W.; Joshi, P. C.; Morin, B.;
Weinfeld, M.; Cadet, J. J. Chem. Soc., Perkin Trans. 2 1996, 371-381.
(14) (a) Steenken, S.; Jovanovic, S. V.; Bietti, M.; Bernhard, K. J. Am.
Chem. Soc. 2000, 122, 2373-2374. (b) Adam, W.; Saha-Mo¨ller, C. R.;
Scho¨nberger, A. J. Am. Chem. Soc. 1996, 118, 9233-9238.
(12) Richardson, W. H.; Stiggal-Estberg, D. L.; Chen, Z.; Baker, J. C.;
Burns, D. M.; Sherman, D. G. J. Org. Chem. 1987, 52, 3143-3150.
Org. Lett., Vol. 4, No. 4, 2002
539

more efficient deactivation at lower dG concentrations, less
dG conversion is expected. Such behavior was found for AP
but not for AP-OH. Consequently, at the dG concentrations
employed, the radical species formed from the AP-OH
triplets by R cleavage must play the dominant role in the
photooxidation process.
Table 2. EPR Data of the DMPO Adducts Obtained in the
Photolysisa of AP-OH and the Thermolysisb of Dioxetane 1
The kinetics of triplet-excited AP and AP-OH quenching
by dG was determined through transient absorption spec-
troscopy. The triplet ketones were generated by excitation
of degassed solutions at 308 nm and detected at 340 nm.
The generation of AP-OH triplets by means of dioxetane 1
thermolysis for spectral monitoring is not feasible. Mono-
exponential fits of the decay curves gave the triplet lifetimes
(in the absence of dG) of 6.6 ( 1 µs for AP and 1.1 ( 0.2
µs for AP-OH. The fact that the triplet lifetime of AP-OH is
six times shorter than that of AP is attributed to the efficient
Norrish-type I cleavage of AP-OH; the rate constant for R
cleavage is estimated to be 5 ((2) × 105 s-1. For the
determination of the quenching rate constants of the triplet
ketones by dG, the time dependence of the absorption decay
curve obtained by time-resolved laser-flash spectroscopy was
monitored at various dG concentrations. For both triplet-
excited AP-OH and AP, quenching rate constants with dG
of 3.3 ((0.2) × 109 M-1 s-1 were obtained. Thus, at the dG
concentrations needed for the analysis of its photooxidation
products (0.200 mM), direct photodeactivation of the AP-
OH triplet by dG competes with R cleavage.
To confirm the formation of radicals through the R
cleavage of AP-OH triplets, spin-trapping experiments for
dioxetane 1 with DMPO have been conducted. When the
dioxetane/DMPO solution was heated at 80 °C for 15 min,
one radical product was observed, which was assigned to
the hydroxymethyl radical5b adduct of DMPO. No benzoyl-
radical product with DMPO was detected (Table 2), which
is presumably due to the thermal instability of the benzoyl-
DMPO radical adduct at 80 °C. Indeed, in a control
experiment, the authentic radical adduct decomposed exten-
sively already at 20 °C within 1 h. For the photolysis of
AP-OH, a hydroxymethyl as well as a benzoyl radical adduct
was observed (Table 2).
a 2.00 mM in a 9:1 H2O:CH3CN mixture at 300 nm for 30 min and 20
°C in a Rayonet photoreactor [sixteen 300-nm lamps (24 W)], [DMPO] )
50.0 mM. b 100 mM in a 9:1 H2O:CH3CN mixture, 15 min, 80 °C, [DMPO]
) 50.0 mM. c In parentheses are given the data for an analogous acyl-
radical adduct in H2O:CH3CN (9:1); see ref 5b. d In parentheses are given
the literature data; see ref 5b.
photolysis, a pair of benzoyl and hydroxymethyl radicals is
formed from the triplet-excited AP-OH by R scission.9
We affirm that the major dG oxidation product in the
dioxetane thermolysis and ketone photolysis (type I oxida-
tion) is not 4-HO-8-oxodG but spiroiminodihydantoin,
unequivocally characterized by means of the SELINQUATE
NMR technique. Evidently, the spiroiminodihydantoin is the
ubiquitous oxidation product of dG by metal-based one-
electron oxidants, triplet ketones, singlet oxygen, and oxyl
radicals. The advantage of the dioxetane 1 for photobiological
studies is that oxidizing triplet ketones and radical species
may be generated thermally without the direct exposure of
biological materials to light.
Acknowledgment. Generous financial support by the
Deutsche Forschungsgemeinschaft (SFB 172 “Molekulare
Mechanismen kanzerogener Prima¨rvera¨nderungen”) and by
the Fonds der Chemischen Industrie is gratefully appreciated.
Supporting Information Available: Complete experi-
mental procedures (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
In the photolysis of AP, which does not undergo R
cleavage,10 no radical adducts were detected. These results
confirm that during the dioxetane thermolysis and the ketone
OL017138M
540
Org. Lett., Vol. 4, No. 4, 2002