Oxidative DNA damage by radicals generated in the thermolysis of hydroxymethyl-substituted 1,2-dioxetanes through the α cleavage of chemiexcited ketones

Article abstract of DOI:10.1021/ja9726318

The 3-(hydroxymethyl)-3,4,4-trimethyl-1,2-dioxetane (HTMD) highly efficiently damages DNA compared to the merely alkyl-substituted derivative 3,3,4,4-tetramethyl-1,2-dioxetane (TMD). To elucidate this difference in oxidative reactivity, two additional hydroxymethyl-substituted 1,2- dioxetanes, namely cis/trans-3-(hydroxymethyl)-3,4-dimethyl-4-(phenylmethyl)- (lα/1β) and 3-(hydroxymethyl)-4,4-dimethyl-3(phenylmethyl)-1,2-dioxetane (2), were investigated in regard to their photochemical and photobiological properties. The high genotoxic effects of the hydroxymethyl-substituted 1,2- dioxetanes, which are reflected in the significant formation of single- strand breaks in plasmid pBR 322 DNA and the efficient oxidation of guanine in calf thymus DNA and the nucleoside 2'-deoxyguanosine (dGuo), are for the first time understood in terms of radical chemistry. The reactivity order of the dioxetanes 1α/1β > HTMD > 2 >> TMD to damage DNA parallels the propensity of these dioxetanes to generate radicals. These reactive species are formed in the thermolysis of the dioxetanes through α cleavage of the intermediary triplet-excited α-hydroxy- and α-phenyl-substituted carbonyl products. The presence of radicals was confirmed by spin-trapping experiments with 5,5-dimethyl-1-pyrroline N-oxide and by laser-flash photolysis. These carbon-centered radicals are efficiently scavenged by molecular oxygen to produce peroxyl radicals, which are proposed as the active DNA damaging species in the thermal decomposition of the hydroxymethyl-substituted 1,2- dioxetanes HTMD, 1α/1β, and 2.

Full text of DOI:10.1021/ja9726318

J. Am. Chem. Soc. 1998, 120, 3549-3559
3549
Oxidative DNA Damage by Radicals Generated in the Thermolysis of
Hydroxymethyl-Substituted 1,2-Dioxetanes through the R Cleavage of
Chemiexcited Ketones
Waldemar Adam,† Simone Andler,*,† Werner M. Nau,‡ and Chantu R. Saha-Mo1ller†
Contribution from the Institute of Organic Chemistry, UniVersity of Wu¨rzburg, Am Hubland,
D-97074 Wu¨rzburg, Germany, and Institute of Physical Chemistry, UniVersity of Basel,
Klingelbergstrasse 80, CH-4056 Basel, Switzerland
ReceiVed July 31, 1997
Abstract: The 3-(hydroxymethyl)-3,4,4-trimethyl-1,2-dioxetane (HTMD) highly efficiently damages DNA
compared to the merely alkyl-substituted derivative 3,3,4,4-tetramethyl-1,2-dioxetane (TMD). To elucidate
this difference in oxidative reactivity, two additional hydroxymethyl-substituted 1,2-dioxetanes, namely cis/
trans-3-(hydroxymethyl)-3,4-dimethyl-4-(phenylmethyl)- (1R/1â) and 3-(hydroxymethyl)-4,4-dimethyl-3-
(phenylmethyl)-1,2-dioxetane (2), were investigated in regard to their photochemical and photobiological
properties. The high genotoxic effects of the hydroxymethyl-substituted 1,2-dioxetanes, which are reflected
in the significant formation of single-strand breaks in plasmid pBR 322 DNA and the efficient oxidation of
guanine in calf thymus DNA and the nucleoside 2’-deoxyguanosine (dGuo), are for the first time understood
in terms of radical chemistry. The reactivity order of the dioxetanes 1R/1â > HTMD > 2 . TMD to damage
DNA parallels the propensity of these dioxetanes to generate radicals. These reactive species are formed in
the thermolysis of the dioxetanes through R cleavage of the intermediary triplet-excited R-hydroxy- and R-phenyl-
substituted carbonyl products. The presence of radicals was confirmed by spin-trapping experiments with
5,5-dimethyl-1-pyrroline N-oxide and by laser-flash photolysis. These carbon-centered radicals are efficiently
scavenged by molecular oxygen to produce peroxyl radicals, which are proposed as the active DNA-damaging
species in the thermal decomposition of the hydroxymethyl-substituted 1,2-dioxetanes HTMD, 1R/1â, and 2.
Introduction
Scheme 1. Generation of Electronically Excited Ketones in
the Thermolysis of 1,2-Dioxetanes
1,2-Dioxetanes have been postulated as labile intermediates
in a variety of biologically relevant processes1 such as in the
enzymatic generation of electronically excited states in oxidative
stress and in the induction of spontaneous mutations.2 The most
characteristic property of such high-energy cyclic peroxides is
their efficient generation of electronically excited carbonyl
fragments on thermolysis (Scheme 1).1,3 Therefore, 1,2-
dioxetanes have been extensively used in the past to study the
genotoxicity of excited carbonyl compounds in the dark.1 The
thermal generation of excited states from dioxetanes has the
advantageous feature that it circumvents the direct exposure of
biological systems, in particular cells, to UV radiation. Thereby,
valuable mechanistic insight into the complex process of DNA
photogenotoxicity and its mutagenic consequences may be
assessed. This has paved the way for extensive studies on an
important bioorganic topic, namely the photobiology without
light,4-11 a subject primarily propagated by the late Professor
G. Cilento (Sa˜o Paulo).
The use of dioxetanes as thermal source of excited ketones
has demonstrated their efficiency to damage DNA in cell-free
(4) Adam, W.; Beinhauer, A.; Mosandl, T.; Saha-Mo¨ller, C. R.; Vargas,
F.; Epe, B.; Mu¨ller, E.; Schiffmann, D.; Wild, D. EnViron. Health Perspect.
1990, 88, 89-97.
(5) Adam, W.; Ahrweiler, M.; Saha-Mo¨ller, C. R.; Sauter, M.; Scho¨n-
berger, A.; Epe, B.; Mu¨ller, E.; Schiffmann, D.; Stopper, H.; Wild, D.
Toxicol. Lett. 1993, 67, 41-55.
(6) (a) Lamola, A. A. Biochem. Biophys. Res. Commun. 1971, 43, 893-
898. (b) Adam, W.; Beinhauer, A.; Epe, B.; Fuchs, R.; Griesbeck, A.; Hauer,
H.; Mu¨tzel, P.; Nassi, L.; Schiffmann, D.; Wild, D. In Primary Changes
and Control Factors in Carcinogenesis; Friedberg, T., Oesch, F., Eds.;
Deutscher Fachzeitschriften Verlag: Wiesbaden, Germany, 1986; pp 64-
67.
† University of Wu¨rzburg.
‡ University of Basel.
* Fax: +49(0931)888 4756. E-mail: Adam@chemie.uni-wuerzburg.de.
(1) (a) Cilento, G.; Adam, W. Photochem. Photobiol. 1988, 48, 361-
368. (b) Gibson, D. T.; Cardini, G. E.; Maseles, F. C.; Kallio, R. E.; Reino,
E. Biochemistry 1970, 9, 1631-1635. (c) Boyd, D. R.; Sharma, N. D.; Boyle,
R.; McMordie, R. A. S.; Chima, J.; Dalton, H. Tetrahedron Lett. 1992, 33,
1241-1244.
(2) (a) Sies, H. Angew. Chem., Int. Ed. Engl. 1986, 25, 1058-1071. (b)
Smith, K. C. Mutat. Res. 1992, 277, 139-162.
(3) (a) Baumstark, A. L. In AdVances in Oxygenated Processes; Baum-
stark, A. L., Ed.; JAI Press Inc.: Greenwich, CT, 1988; Vol. 1, pp 31-84.
(b) Adam, W.; Heil, M.; Mosandl, T.; Saha-Mo¨ller, C. R. In Organic
Peroxides; Ando, W., Ed.; Wiley & Sons: Chichester, U.K., 1992; pp 221-
254.
(7) Epe, B.; Mu¨ller, E.; Adam, W.; Saha-Mo¨ller, C. R. Chem.-Biol.
Interact. 1992, 85, 265-281.
(8) Adam, W.; Treiber, A. J. Am. Chem. Soc. 1995, 117, 2686-2693.
(9) Adam, W.; Saha-Mo¨ller, C. R.; Scho¨nberger, A.; Berger, M.; Cadet,
J. Photochem. Photobiol. 1995, 62, 231-238.
(10) Adam, W.; Saha-Mo¨ller, C. R.; Scho¨nberger, A. J. Am. Chem. Soc.
1996, 118, 9233-9238.
(11) Adam, W.; Saha-Mo¨ller, C. R.; Scho¨nberger, A. J. Am. Chem. Soc.
1997, 119, 719-723.
S0002-7863(97)02631-0 CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/04/1998

3550 J. Am. Chem. Soc., Vol. 120, No. 15, 1998
Adam et al.
and cellular systems.4 Besides the formation of pyrimidine
dimers,6 we have shown by means of the formamidopyrimidine
DNA-glycosylase (FPG protein) that the majority of the induced
DNA lesions constitute oxidative guanine modifications [e.g.,
7,8-dihydro-8-oxo-2’-deoxyguanosine (8-oxodGuo), formami-
dopyrimidines (Fapy)] and AP sites.7
Scheme 2. DNA Damage Induced by Radicals Generated in
the Thermal Decomposition of Hydroxymethyl-Substituted
1,2-Dioxetanes
nucleoside 2’-deoxyguanosine (dGuo)11 demonstrated that HTMD
generates, besides the characteristic type-I photooxidation
products oxazolone16 and 2-(S)-2,5’-anhydro-1-(2-deoxy-â-D-
erythro-pentofuranosyl)-5-guanidinylidene-2-hydroxy-4-oxoimi-
dazolidine17 (oxoimidazolidine), also the type-II photooxidation
products 4R* and 4S* diastereomers of 4-hydroxy-8-oxo-4,8-
dihydro-2’-deoxyguanosine18,19
(4-HO-8-oxodGuo)
and
8-oxodGuo.20 The type-I/type-II product ratio of approximately
unity was explained in terms of the simultaneous involvement
of a direct electron-transfer process between the triplet-excited
ketone and guanine (type-I photooxidation)21 and energy transfer
to molecular oxygen with the formation of singlet oxygen (type-
II photooxidation)21 as oxidizing species.
Recently, the chemical nature of the dioxetane-induced DNA
modifications has been evaluated.8-11 By using acetylated
guanine as model compound, we have shown that the nucleo-
philic attack of the guanine base on the peroxide bond of 3,3-
disubstituted 1,2-dioxetanes leads to the oxidation of guanine.8
Investigations on calf thymus DNA have established that 1,2-
dioxetanes, in particular alkyl-substituted ones, oxidize ef-
ficiently and almost exclusively the guanine base to form the
mutagenic 7,8-dihydro-8-oxoguanine (8-oxoGua)12-15 and the
2,2-diamino-[(2-deoxy-â-D-erythro-pentofuranosyl)-4-amino]-
5(2H)-oxazolone (oxazolone)16 in high yields.9 The generation
of these well-known DNA-photooxidation products was ex-
plained by the involvement of triplet-excited carbonyl com-
pounds as reactive species, formed in the thermolysis of the
dioxetanes.
The remarkable reactivity of the hydroxymethyl-substituted
1,2-dioxetane HTMD to induce efficient oxidative DNA damage
was tentatively attributed to its possible association to DNA or
dGuo through hydrogen bonding. Our present study with
HTMD and the two additional hydroxy-substituted dioxetanes
1R/1â and 2 now demonstrates that the enhanced reactivity and
distinct selectivity of the hydroxymethyl-substituted dioxetanes
in the oxidative damage of DNA are due to radicals. The latter
derive from the efficient and fast R cleavage of the electronically
excited R-hydroxy-substituted ketone intermediates22,23 (Scheme
2). The efficiency of radical formation and their identity was
established by spin-trapping experiments with 5,5-dimethyl-1-
pyrroline N-oxide (DMPO) coupled with EPR spectroscopy, by
time-resolved UV spectroscopy, and by the inhibitory effect of
radical scavengers. Our results reveal for the first time that
hydroxymethyl-substituted dioxetanes serve as effective thermal
sources of radicals, which provides a new, valuable tool for the
investigation of radical-induced oxidative DNA damage (strand
breaks and guanine oxidation) under nonphotolytic conditions.
3-Hydroxymethyl-3,4,4-trimethyl-1,2-dioxetane (HTMD) was
hitherto found to be the most reactive 1,2-dioxetane compared
to various derivatives without hydroxymethyl substitutents, e.g.
Results
Preparation of the Hydroxymethyl-Substituted 1,2-Diox-
etanes. The hydroxymethyl-substituted 1,2-dioxetanes 1R/1â,24
2,24 and HTMD25 were prepared according to the literature
procedures. The relative configurations of the cis and trans
isomers of dioxetane 1R/1â could not be determined by NOE
(17) Buchko, G. W.; Cadet, J.; Ravanat, J.-L.; Labataille, P. Int. J. Radiat.
Biol. 1993, 63, 669-676.
(18) (a) Ravanat, J.-L.; Berger, M.; Benard, F.; Langlois, R.; Ouellet,
R.; van Lier, J. E.; Cadet, J. Photochem. Photobiol. 1992, 55, 809-814.
(b) Ravanat, J.-L.; Cadet, J. Chem. Res. Toxicol. 1995, 8, 379-388.
(19) (a) Sheu, C.; Foote, C. S. J. Am. Chem. Soc. 1993, 115, 10446-
10447. (b) Sheu, C.; Foote, C. S. J. Am. Chem. Soc. 1995, 117, 474-477.
(20) Schneider, J. E.; Price, S.; Maidt, L.; Gutteridge, J. M. C.; Floyd,
R. A. Nucleic Acid Res. 1990, 18, 631-635.
3,3,4,4-tetramethyl- (TMD) or 3-(acetoxymethyl)-3,4,4-tri-
methyl-1,2-dioxetane (AcTMD).9 Our previous studies with the
(12) Kasai, H.; Nishimura, S. In OxidatiVe Stress, Oxidants and Anti-
oxidants; Sies, H., Ed.; Academic Press: New York, 1991; pp 99-116.
(13) Ames, B. N. Science 1983, 221, 1256-1264.
(21) Foote, C. S. Photochem. Photobiol. 1991, 54, 659.
(22) Steenken, S.; Jaenicke-Zauner, W.; Schulte-Frohlinde, D. Photo-
chem. Photobiol. 1975, 21, 21-26.
(14) Floyd, R. A. Carcinogenesis 1990, 11, 1447-1450.
(15) Shibutani, S.; Takeshita, M.; Grollman, A. P. Nature 1991, 349,
431-434.
(23) Palm, W. U.; Dreeskamp, H. J. Photochem. Photobiol. A: Chem.
1990, 52, 439-450.
(16) (a) Kasai, H.; Yamaizumi, Z.; Berger, M.; Cadet, J. J. Am. Chem.
Soc. 1992, 114, 9692-9694. (b) Cadet, J.; Berger, M.; Buchko, G. W.;
Joshi, P. C.; Raoul, S.; Ravanat, J.-L. J. Am. Chem. Soc. 1994, 116, 7403-
7404. (c) 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.
(24) Leclercq, D.; Bats, J.-P.; Picard, P., Moulines, P. Synthesis 1982,
778-779.
(25) Adam, W.; Bushan, V.; Dirnberger, T.; Fuchs, R. Synthesis 1986,
330-332.

OxidatiVe DNA Damage
J. Am. Chem. Soc., Vol. 120, No. 15, 1998 3551
Table 1. Activation Parameters and Excitation Yields (ΦS, ΦTacetone) of the 1,2-Hydroxymethyl-Substituted 1,2-Dioxetanes HTMD, 1R/1â, and
a
2 and of TMD for Comparison
t1/2 (h)
dioxetane
37 °C
50 °C
log A
Ea (kcal/mol)
∆Hq (kcal/mol)
∆Sq (eu)
ΦT
9.0 ( 0.9d
d
b,c (%)
ΦS × 102 b (%)
acetone
HTMDc
1R
39 ( 11
44 ( 4
37 ( 1
46 ( 5
45 ( 2
7.8 ( 0.5
8.2 ( 0.2
7.7 ( 0.2
8.1 ( 0.4
11.9 ( 0.4
12.3 ( 0.1
12.1 ( 0.2
14.2 ( 0.2
f
24.1 ( 0.5
24.7 ( 0.4
24.5 ( 0.5
27.5 ( 0.8
25.0 ( 1.0
23.4 ( 0.5
23.6 ( 0.3
24.1 ( 0.6
27.1 ( 0.8
24.9 ( 0.7
-6.3 ( 0.3
-6.2 ( 0.1
-4.7 ( 0.2
+4.5 ( 0.1
-2.8 ( 2.3
1.0 ( 0.2
1.8 ( 0.2
3.0 ( 0.3
2.2 ( 0.2
25 ( 5
1â
d
2
14.6 ( 2.1
35 ( 4
TMDe
a CH3CN as solvent, temperature constant within ( 0.1 °C, mean value of at least three independent runs, experimental data are listed in the
Supplementary Information in Table S-1. b Determined at 50 °C in CH3CN/toluene (80/20), experimental data are listed in the Supplementary
Information in Table S-2. c Values referring to triplet-excited acetone only (cf. text). d Triplet quantum yields for the short-lived ketones 1-phenyl-
2-propanone and 1-hydroxy-2-propanone were determined to be ca. 1% (cf. text). e Literature values27 in toluene. f Not known.
experiments nor by X-ray crystallography since for the latter
no suitable crystals were obtained.
(ΦTacetone) of HTMD and 2 (ca. 15%) only reflect the fraction
of triplet acetone (the hydroxy-substituted ketones are too short-
lived, see below) and are about one-half that for the TMD (35%),
which in view of its symmetrical nature releases exclusively
the rather long-lived triplet-excited acetone.27 The much smaller
ΦT values determined for the dioxetanes 1R and 1â (ca. 1%)
represent, of course, a lower limit since these dioxetanes
generate only the quite short-lived (efficient R cleavage) excited
ketones 1-phenyl-32 or 1-hydroxy-2-propanone.22
Activation Parameters and Excitation Yields of the Di-
oxetanes. Although the dioxetanes 1R/1â and 2 are known for
some time,24 their photophysical data have not been reported.
Therefore, the activation parameters and excitation yields of
these dioxetanes were determined. The rates of the dioxetane
thermolysis were monitored by the decay of the chemilumi-
nescence intensity in acetonitrile over a temperature range of
30 °C and were found to follow clean first-order kinetics. The
activation parameters of 1R/1â and 2 were determined according
to standard isothermal kinetics.26 The results are summarized
in Table 1 together with the literature results for HTMD and
TMD27 for comparison. The activation enthalpy (∆Hq) of the
dioxetane 2 was found to be by ca. 3 kcal/mol higher than those
for HTMD and the two isomers of dioxetane 1R/1â. The ∆Sq
terms are negative and close to zero, as is usually the case for
simple dioxetanes.28 An exception is dioxetane 2, for which
∆Sq is positive and offsets its higher ∆Hq value compared to
the other dioxetanes.
Without added fluorescers, the thermolyses of the dioxetanes
1R, 1â, and 2 were only weakly chemiluminescent. Addition
of 9,10-diphenylanthracene (DPA) and 9,10-dibromoanthracene
(DBA) resulted in an increase in the chemiluminescence
intensity without affecting the rate of dioxetane decomposition.
The singlet and triplet excitation yields were determined by
established chemiluminescence methods29 with DPA for sin-
glet30 and DBA30,31 for triplet counting (variation of the
concentration of added fluorescers at constant dioxetane con-
centration and temperature). As expected,3,28 the thermal
decomposition of the dioxetanes HTMD, 1R, 1â, and 2 produced
greater yields of triplet-excited carbonyl products than of excited
singlets (Table 1). The singlet yields (ΦS) for all hydroxy-
methyl-substituted dioxetanes were e0.1%.
EPR Studies of the Dioxetanes and Their Ketone Prod-
ucts. The R cleavage of triplet-excited R-hydroxy- and R-phen-
yl-substituted ketones to generate radicals is a well-known
photochemical reaction.22,23,32 Since such excited ketones are
formed in the thermolysis of the hydroxymethyl- and phenyl-
methyl-substituted 1,2-dioxetanes HTMD, 1R/1â, and 2, spin-
trapping experiments with DMPO were performed to provide
spectroscopic evidence for the generation of radicals in the
thermal decomposition of such reactive cyclic peroxides. The
results of the EPR investigations are shown in Table 2.
On thermolysis of HTMD in the presence of DMPO in
aqueous acetonitrile, two characteristic doublet-of-triplets EPR
signals were observed (Table 2, entry 1) for the DMPO adducts
of the acetyl33 and the hydroxymethyl radicals.34 A control
experiment revealed that no EPR signal was detected in the
thermolysis of DMPO at 37 °C without dioxetane. For
comparison, 1-hydroxy-2-propanone (1 mM) was irradiated at
300 nm for 30 min in the presence of DMPO (45 mM). The
EPR spectrum was a superposition of the two DMPO-radical
adducts (Table 2, entry 7), described above for the thermolysis
of HTMD, which confirms that a pair of R-hydroxymethyl and
acyl radicals is formed in the thermolysis of HTMD from triplet-
excited 1-hydroxy-2-propanone through its R cleavage.22
The thermolysis of the simple alkyl-substituted dioxetane
TMD, which only releases triplet-excited acetone as decomposi-
tion product, afforded a doublet-of-triplets signal pattern in its
EPR spectrum (Table 2, entry 2). This EPR pattern is assigned
to the known DMPO-DMPO-yl radical adduct, generated by
hydrogen-atom abstraction from the hetero-allylic position of
the nitrone by triplet-excited acetone and subsequent addition
of the resulting DMPO-yl radical to a further molecule of
DMPO.35 The same EPR spectrum (Table 2, entry 8) was
The measured triplet yields must be interpreted with some
caution since the experimental method, which relies on efficient
bimolecular energy transfer between the triplet-excited ketone
and the fluorescer, works reliably only for relatively long-lived
(>10 ns) triplet states. Therefore, the triplet excitation yields
(26) Adam, W.; Zinner, K. In Chemical and Biological Generation of
Electronic Excited States; Adam, W., Cilento, G., Eds.; Academic Press:
New York, 1982; Chapter 5.
(27) Adam, W.; Beinhauer, A.; Hauer, H. In Handbook of Organic
Photochemistry; Scaiano, J. C., Ed.; CRC Press: Boca Raton, FL, 1989;
Vol. 2, pp 271-327.
(32) (a) Houk, K. N. Chem. ReV. 1976, 76, 1-74. (b) Baer, R.; Paul, H.
Chem. Phys. 1984, 87, 73-80.
(33) Nukaya, H.; Inaoka, Y.; Ishida, H.; Tsuji, K.; Suwa, Y.; Wakaba-
yashi, K.; Kosuge, T. Chem. Pharm. Bull. 1993, 41, 649-653.
(34) Haire, D. L.; Kotake, Y.; Janzen, E. G. Can. J. Chem. 1988, 66,
1901-1911.
(28) Baumstark, A. L. In Singlet Oxygen; Frimer, A. A., Ed.; Uniscience,
CRC Press: Cleveland, OH, 1985; Vol. 2, p 1.
(29) Adam, W. Reference 26, Chapter 4.
(30) (a) Wilson, T.; Schaap, A. P. J. Am. Chem. Soc. 1971, 93, 4126-
4136. (b) Turro, N. J.; Lechtken, P.; Schuster, G.; Orell, J.; Steinmetzer,
H.-C.; Adam, W. J. Am. Chem. Soc. 1974, 96, 1627-1629.
(31) Belyakov, V. A.; VasilÅev, R. F. Photochem. Photobiol. 1970, 11,
179-192.
(35) (a) Finkelstein, E.; Rauckman, E. J.; Rosen, G. M. Arch. Biochem.
Biophys. 1980, 200, 1-16. (b) Chignell, C. F.; Motten, A. G.; Sik, R. H.;
Parker, C. E.; Reszka, K. Photochem. Photobiol. 1994, 59, 5-11. (c)
Kalyanaraman, B.; Janzen, E. G.; Mason, R. P. J. Biol. Chem. 1985, 260,
4003-4006.

3552 J. Am. Chem. Soc., Vol. 120, No. 15, 1998
Adam et al.
Table 2. EPR-Spectral Data of the Nitroxyl-Radical Adducts 3
Generated in the Thermolysis of Hydroxymethyl- and
Phenylmethyl-Substituted Dioxetanes and the Photolysis of the
Corresponding Ketone Fragmentation Products in the Presence of
the DMPO as Radical Trapa
propensity to produce radicals through R cleavage of the triplet-
excited 1-hydroxy-2-propanone.
In the thermal decomposition of the acetoxymethyl-substituted
1,2-dioxetane AcTMD in the presence of DMPO, no EPR signal
was detected even at a 5-fold higher concentration of the
dioxetane (Table 2, entry 3). This was further supported by a
control experiment in which 1-acetoxy-2-propanone was irradi-
ated at 300 nm for 30 min, but again no EPR signal of a DMPO
adduct was obtained (Table 2, entry 9). Thus, acylation of the
hydroxy group in 1-hydroxy-2-propanone reduces the ease of
R cleavage36 (otherwise the acetyl radical should have been
trapped with DMPO).
For the thermolysis of the dioxetanes 1R and 1â (20 mM) in
the presence of DMPO (45 mM) also two characteristic doublet-
of-triplets EPR signals were observed (Table 2, entry 4). The
main paramagnetic species was attributed to the acetyl-radical
product of DMPO, but the expected hydroxymethyl- and
phenylmethyl-radical adducts of DMPO could not be definitively
assigned because their hyperfine coupling constants and the
Lande´ factor are quite similar34,37 to the DMPO-yl radical adduct
(cf. Table 2, entries 2 and 8). A control experiment, in which
1-phenyl-2-propanone (1 mM) was irradiated at 300 nm to
generate the authentic pair of phenylmethyl and acetyl radicals
through R cleavage,32 yielded only the trapping product of the
acetyl radical (Table 2, entry 10). The DMPO adduct of the
phenylmethyl radicals was not observed, presumably due to its
lower reactivity to add to double bonds compared to acyl
radicals.38,39
In the thermolysis of the dioxetane 2, hydroxymethyl as well
as phenylmethyl radicals may be generated from the R cleavage
of the electronically excited 1-hydroxy-3-phenyl-2-propanone
and subsequent decarbonylation. In the presence of DMPO,
only a very weak and noisy EPR signal with a doublet-of-triplets
pattern was observed at a relatively high dioxetane concentration
(Table 2, entry 6), the structure of the spin adduct could not be
definitively assigned. Besides alkyl-radical adducts, the main
EPR-active species was the DMPO-OH adduct,35a,40 which was
also observed in traces in the thermolysis of all the other
dioxetanes. Its formation is explained in terms of the direct
oxidation of DMPO by the dioxetanes to generate the DMPO
radical cation and subsequent addition of water, analogous to
what is described in the literature for other oxidizing agents.41
A control experiment with authentic 1-hydroxy-3-phenyl-2-
propanone showed that the direct irradiation of this ketone at
0 °C in the presence of DMPO leads to an acyl radical adduct
of DMPO (Table 2, entry 11), which was not found in the
thermolysis (37 °C) of the dioxetane 2. The structure of the
acyl radical (hydroxyacetyl or phenylacetyl) could not be
unequivocally assigned. Although the photolysis of 1-hydroxy-
3-phenyl-2-propanone affords the phenylacetyl radical in nearly
quantitative yield (proven by laser flash studies, see below), its
fast decarbonylation (kCO ca. 5 × 106 s-1, this work)42,43 renders
a [DMPO] ) 45 mM, H2O/CH3CN (95:5). b A: 37 °C, 40 min, B:
hν (300 nm, Rayonet photoreactor), 30 min. c Lande´ g factor, error (
0.0001. d The literature values33-35 are given in parentheses. e Hyperfine
coupling constants aH (with R-H) and aN (with N) of the nitroxyl radical
8 of DMPO, error ( 0.0001 G. f In all EPR spectra, traces of the
DMPO-OH trapping product [aH ) aN ) 14.8 G, g ) 2.0054, literature
values:35a,40 aH ) aN ) 14.9 G, g ) 2.0053] were detected. g Mean
values since the EPR signals of the DMPO-CH2OH, DMPO-DMPO-
yl, and DMPO-CH2Ph adducts overlap. h X ) CH2OH versus CH2Ph
are not clearly distinguishable, no literature data available.
obtained when acetone (10 mM) was irradiated at 300 nm for
40 min in the presence of DMPO (45 mM). Since HTMD leads
to both triplet-excited acetone and 1-hydroxy-2-propanone on
thermolysis, besides spin trapping of the released hydroxymethyl
radicals from triplet-excited 1-hydroxy-2-propanone, also the
DMPO-DMPO-yl radical adduct from the triplet-excited acetone
is expected. Unfortunately, the similar hyperfine coupling
constants of the DMPO-CH2OH and DMPO-DMPO-yl radical
adducts (cf. Table 2, entries 1 and 2) do not allow a definitive
EPR-spectral confirmation. Nevertheless, the formation of the
adduct between DMPO and the acyl radical in the thermolysis
of HTMD (Table 2, entry 1) demonstrates unequivocally its
(36) Lewis, F. D.; Lauterbach, R. T.; Heine, H.-G., Hartmann, W.;
Rudolph, H. J. Am. Chem. Soc. 1975, 97, 1519-1525.
(37) Gilbert, B. C.; Smith, J. R. L.; MacFaul, P.; Taylor, P. J. Chem.
Soc., Perkin Trans. 2 1993, 2033-2037.
(38) Ryu, I.; Sonoda, N. Angew. Chem., Int. Ed. Engl. 1996, 35, 1150-
1166.
(39) Walbiner, M.; Wu, J. Q.; Fischer, H. HelV. Chim. Acta 1995, 78,
910-924.
(40) Buettner, G. R. Free Radical Biol. Med. 1987, 3, 259-303.
(41) Finkelstein, E.; Rosen, G. M.; Rauckman, E. J. J. Am. Chem. Soc.
1980, 102, 2, 4994-4999.
(42) La¨ufer, M.; Dreeskamp, H. J. Magn. Res. 1984, 60, 357-365.
(43) (a) Maouf, A.; Lemmetyinen, H.; Koskikallio, J. Acta Chem. Scand.
1990, 44, 336-338. (b) Tsentalovich, Y. P.; Fischer H. J. Chem. Soc., Perkin
Trans. 2 1994, 729-733.

OxidatiVe DNA Damage
J. Am. Chem. Soc., Vol. 120, No. 15, 1998 3553
spin trapping of this shorter-lived acyl radical unlikely. In
contrast, the less efficient produced hydroxyacetyl radical may
be trapped, since its rate of decarbonylation is slow enough (kCO
) 1 × 105 s-1).44
Photochemical Studies of the Dioxetane-Derived Ketone
Products. Conventional Photolysis of 1-Hydroxy-3-phenyl-
2-propanone. This R,R’-disubstituted ketone was photolyzed
at 300 nm in a Rayonet photoreactor, and the product distribu-
tion and the photolysis rate were monitored by 1H NMR
spectroscopy. The photolysis led to 2-phenylethanol (43%) and
1,2-diphenylethane (49%) as major products; furthermore, 1,2-
dihydroxyethane (6%) and traces of methanol (<2%) were
detected, the total mass balance was 78%. The photolytic
consumption of 1-hydroxy-3-phenyl-2-propanone was slightly
higher compared to the well-investigated32,42 1-phenyl-2-pro-
panone, i.e. 85 ( 4% versus 78 ( 4% in 2 h.
Figure 1. Transient-absorption traces (λmon ) 317 nm) produced upon
laser-flash photolysis (λexc ) 308 nm, corrected for concomitant
fluorescence) of optical-density-matched (OD ) 0.40) solutions of 1,3-
diphenyl-2-propanone (upper trace) and 1-hydroxy-3-phenyl-2-pro-
panone (lower trace).
Laser-Flash Photolysis of the Ketones. The direct laser-
flash photolyses (λ ) 351 nm) of the dioxetanes HTMD and
TMD gave no observable radical transients. Therefore, the
kinetic fate of the electronically excited ketones acetone,
1-hydroxy-3-phenyl-2-propanone, 1-hydroxy-2-propanone, and
1-phenyl-2-propanone was investigated by time-resolved (nano-
second) UV spectroscopy.45,46 Since the singlet-excited states
of aliphatic ketones are very short-lived (<5 ns) due to quite
fast intersystem crossing to their triplet states,47 only the lifetimes
and reactions of the latter may be determined with this technique.
Moreover, since the simple dioxetanes produce quite selectively
triplet states upon thermolysis (cf. Table 1), only the triplet
reactivity is relevant in the context of the present study.
Acetone. The triplet state of acetone is long-lived48 in
acetonitrile (50 µs) and water (20 µs), the solvents which were
employed for the photobiological investigations, and does not
undergo significant R cleavage.47 To demonstrate the ability
of triplet acetone to engage efficiently in hydrogen abstraction,
the quenching rate constant with diphenylmethanol was deter-
mined through the kinetics of the time-resolved growth of the
resulting diphenylhydroxymethyl radical as a function of
hydrogen-donor concentration. A value of kq ) 4.7 × 107 M-1
s-1 (cf. Table S-3) was found for hydrogen abstraction by triplet
acetone from diphenylmethanol.
Table 3. Solvent Effects on the Decarbonylation Rate of Phenyl-
acetyl Radicals and the Relative Quantum Yields of Their Forma-
tion in the Laser-Flash Photolysis of 1-Hydroxy-3-phenyl-2-
propanone Compared to 1,3-Diphenyl-2-propanonea
a Optical-density-matched (OD ) 0.4) solutions of 1-hydroxy-3-
phenyl- and 1,3-diphenyl-2-propanone; direct excitation with XeCl-
excimer laser pulse (λ ) 308 nm, 25-ns laser pulse, 75 mJ). b Maximum
absorbance of the time-resolved rise is equated to the amount of
generated phenylacetyl radicals; ratio defined as quotient of OD(PhCH2•)
for 1-hydroxy-3-phenyl-2-propanone and 1,3-diphenyl-2-propanone,
error ( 10%. c Error ( 10%. d Degassed by two freeze-pump-thaw
cycles. e Determined by monoexponential fitting of the time-resolved
increase in the absorbance at λmon ) 317 nm due to phenylmethyl
radical, error ( 10%. f Reference 49, error ( 0.06.
1-Hydroxy-3-phenyl-2-propanone Compared to 1,3-Diphen-
yl-2-propanone. The triplet state of 1,3-diphenyl-2-propanone
is too short-lived (<1 ns)43,49 to allow direct detection in the
nanosecond time regime, since it undergoes very fast and
efficient (ΦR ) 0.84 ( 0.06)49c R cleavage to produce a pair of
phenylmethyl and phenylacetyl radicals. The latter undergoes
fast decarbonylation (kCO ca. 5 × 106 M-1 s-1) to produce a
second phenylmethyl radical. In accordance with previous
studies,43,49-51 laser-flash photolysis produced a transient ab-
sorption at 317 nm characteristic for phenylmethyl radicals
(Figure 1). The absorption is characterized by a step-and-rise
pattern, in which the initial sudden rise corresponds to the fast
R cleavage to produce the first phenylmethyl radical. The time-
resolved slow growth pertains to the decarbonylation of the
phenylacetyl radical to produce the second phenylmethyl radical.
The ratio of the absorbances of the time-resolved and sudden
rise is approximately unity. The rate of decarbonylation is
solvent dependent, and the measured values (Table 3) are in
good agreement with the absolute values reported by Turro50
and Scaiano51 and with the trend in solvent effects observed by
Fischer,43b namely, kCO (acetonitrile) < kCO (methanol) < kCO
(hexane).
For the flash photolysis of the 1-hydroxy-3-phenyl-2-pro-
panone, we also have observed a transient step-and-rise absorp-
tion trace similar to that of 1,3-diphenyl-2-propanone, except
that the ratio of the absorbances for the slow and sudden rise
was not unity but rather about 2-fold higher for the time-resolved
growth (Figure 1). The latter could be unambiguously assigned
to the decarbonylation of a phenylacetyl radical to produce
phenylmethyl radicals on the following basis: (a) the absorption
(44) Vollenweider, J.-K., Paul, K. Int. J. Chem. Kinet. 1986, 18, 791-
800.
(45) For example: (a) Leyva, E.; Platz, M. S.; Persy, G.; Wirz, J. J.
Am. Chem. Soc. 1986, 108, 3783- 3790. (b) Burnett, M. N.; Boothe, R.;
Clark, E.; Gisin, M.; Hassaneen, H. M.; Pagni, R. M.; Persy, G.; Smith, R.
J.; Wirz, J. J. Am. Chem. Soc. 1988, 110, 2527-2538.
(46) For example: (a) Paul, H.; Small Jr., R. D.; Scaiano, J. C. J. Am.
Chem. Soc. 1978, 100, 4520- 4527. (b) Baigne´e, A.; Howard, J. A.; Scaiano,
J. C.; Stewart, L. C. J. Am. Chem. Soc. 1983, 105, 6120-6123.
(47) Turro, N. J. In Modern Molecular Photochemistry; Benjamin/
Cummings Publishing Co.: Menlo Park, CA, 1978; pp 528-539.
(48) (a) Porter, G.; Dogra, S. K.; Loutfy, R. O.; Sugamori, S. E.; Yip,
R. W. J. Chem. Soc., Faraday Trans. 1 1973, 69, 1462-1474. (b) Encinas,
M. V.; Lissi, E. A.; Scaiano, J. C. J. Phys. Chem. 1980, 84, 948-951.
(49) (a) Engel, P. S. J. Am. Chem. Soc. 1970, 92, 6074-6076. (b)
Robbins, W. K.; Eastman, R. H. J. Am. Chem. Soc. 1970, 92, 6076-6077.
(c) Noh, T. H.; Step, E.; Turro, N. J. J. Photochem. Photobiol. A: Chem.
1993, 72, 133-145.
(50) Turro, N. J.; Gould, I. R.; Baretz, B. H. J. Phys. Chem. 1983, 87,
531-532.
(51) Lunazzi, L.; Ingold, K. U.; Scaiano, J. C. J. Phys. Chem. 1983, 87,
529-530.

3554 J. Am. Chem. Soc., Vol. 120, No. 15, 1998
Adam et al.
spectrum matched that of the phenylmethyl radicals, (b) the
kinetics of the growth was, within the error limit, the same as
for 1,3-diphenyl-2-propanone, and (c) the solvent effects on the
growth rate (cf. Table 3).
Optical-density-matched solutions of 1-hydroxy-3-phenyl-2-
propanone and 1,3-diphenyl-2-propanone were employed (at
constant pulse energy of the incident laser flash) to determine
the relative efficiencies (ΦR) of phenylacetyl formation of these
ketones (Table 3). For this purpose, the absorbance due to the
time-resolved rise was set equal to the amount of the phenyl-
acetyl precursor. In general, the relative quantum efficiency
of phenylacetyl-radical formation from 1-hydroxy-3-phenyl-2-
propanone was higher (ca. 10-25%); the actual ratios are listed
in Table 3. From these data and the known R-cleavage quantum
yield (ΦR ) 0.84 ( 0.06) for 1,3-diphenyl-2-propanone,49c the
quantum efficiency of phenylacetyl-radical formation from the
triplet-excited 1-hydroxy-3-phenyl-2-propanone may be esti-
mated to be unity (within a 15% error) and its R cleavage
slightly more efficient than for 1,3-diphenyl-2-propanone.49c
Since the triplet lifetime of 1,3-diphenyl-2-propanone has been
estimated to be <1 ns,49 we may conclude that the lifetime of
1-hydroxy-3-phenyl-2-propanone is even less and, hence, the
R cleavage for the hydroxymethyl is faster than for the
phenylmethyl group.
Figure 2. Gel-electrophoretic detection of single-strand breaks in pBR
322 DNA [c ) 10 mg/L in 0.5 mM KH2PO4 buffer (pH 7.4) with 10%
CH3CN as cosolvent] induced in the thermolysis of the hydroxymethyl-
substituted dioxetanes HTMD, 1R/1â, and 2 (2 mM) at 37 °C for 5 h
in the dark. Data shown are estimated from the light intensities of the
spots for supercoiled (SC) and open-circular (OC) forms of pBR 322
DNA and are mean values of at least three independent runs, error (
10% of the stated value.
Scheme 3. Mechanistic Pathways for the Formation of
Radicals in the R Cleavage of Electronically Excited Ketones
Generated in the Thermolysis of Hydroxymethyl- and
Phenylmethyl-Substituted Dioxetanes and the Photolysis of
Their Carbonyl Productsa
1-Hydroxy- and 1-Phenyl-2-propanone. No time-resolved
growth of triplet absorption signals could be observed in the
laser-flash photolysis of 1-hydroxy- and 1-phenyl-2-propanone.
This is a consequence of the very fast R cleavage of the
phenylmethyl and hydroxymethyl groups, which is corroborated
by the above experiments with 1,3-diphenyl-2-propanone and
1-hydroxy-3-phenyl-2-propanone. In addition, no hydrogen
abstraction was found for these triplet-excited ketones to afford
the diphenylhydroxymethyl radical from diphenylmethanol (0.3
M), which implies a triplet lifetime of less than 20 ns. The
flash photolysis of 1-hydroxy-2-propanone afforded immediately
(within the 25-ns pulse time of the laser pulse) a long-lived (µs
range) transient at ca. 320 nm (trace not shown), presumably
due to the acetyl radical. In the flash photolysis of 1-phenyl-
2-propanone, an initial fast rise in the absorption was obtained
within the laser pulse at the detection wavelength of 317 nm.
The latter is ascribed to the formation of phenylmethyl and
acetyl radicals, which disappear by biexponential kinetics due
to recombination. By taking the estimated lifetimes of the
triplet-excited 1,3-diphenyl-2-propanone (<1 ns)49 and 1-hy-
droxy-3-phenyl-2-propanone (<1 ns, this work) as a measure
of the ease of R cleavage to produce phenylmethyl and
hydroxymethyl radicals, we may conclude that the lifetimes of
triplet-excited 1-hydroxy- and 1-phenyl-2-propanone are also
within this range (<1 ns).
Photobiological Studies with DNA and dGuo. Formation
of Strand Breaks in pBR 322 DNA. The DNA-cleaving
properties of these dioxetanes were assessed in supercoiled pBR
322 DNA by gel electrophoresis. The thermolysis (37 °C) of
the dioxetanes 1R and 1â (2 mM) for 5 h in the presence of
supercoiled pBR 322 DNA (c ) 10 mg/L) resulted in the
significant formation (80%) of open-circular (OC) DNA (Figure
2, lanes 3 and 4). This should be compared with about 15%
OC DNA (blank value) detected in the thermolysis of pBR 322
DNA under the same conditions without dioxetane (Figure 2,
lane 1). Only 40% open-circular form was observed for the
thermal decomposition of dioxetane 2 (Figure 2, lane 5) in the
presence of pBR 322 DNA. For comparison, HTMD was
thermolyzed with pBR 322 DNA and 69% of open-circular
DNA was found. The efficiency of the hydroxymethyl-
a Also the R3COR4 product may be triplet-excited, but for mechanistic
simplicity of the scheme, only the electronically excited R1COR2
fragment and its subsequent transformations are shown.
substituted 1,2-dioxetanes in inducing DNA strand breaks
follows the order 1R/1â > HTMD > 2, which qualitatively
parallel their propensity to generate active radicals, namely 3:2:1
per decomposed dioxetane molecule (see Scheme 4 below).
Oxidation of Calf Thymus DNA. Isolated calf thymus DNA
at a concentration of 0.1 mg/mL (62.5 µM guanine concentra-
tion) was treated at 37 °C in the presence of the hydroxymethyl-
substituted 1,2-dioxetanes 1R/1â and 2 (concentration range
1-10 mM) for 21 h in the dark to establish their efficacy to
generate oxidative DNA-base modifications. The well-exam-
ined HTMD9-11 was also applied for comparison. Besides the
established oxidation product 8-oxoGua,12-15 substantial amounts
of oxazolone16 were detected by the fluorescence-labeling HPLC
assay with 1,2-naphthoquinone-4-sulfonic acid.16-18a,52 Former
studies9,10 have established that the oxazolone yield rises with
increasing dioxetane concentration, whereas the guanine oxida-
tion product 8-oxoGua is prone to further oxididation by excess
dioxetane and goes through a maximum.10 Thus, under these
(52) Kobayashi, Y.; Kinoshita, T.; Kubo, H. Anal. Biochem. 1987, 160,
392-398.

OxidatiVe DNA Damage
J. Am. Chem. Soc., Vol. 120, No. 15, 1998 3555
Scheme 4. Reaction Pathways for the Electronically Excited Carbonyl Products in the Thermolysis of the Hydroxymethyl-
Substituted 1,2-Dioxetanes HTMD, 1R/1â, and 2 as a Mechanistic Rationale for Their Reactivity Profile in the Type-I
Photooxidation of Guanosine in DNA and DGuoa
a Phenylmethyl radicals are ineffective in the damage of DNA.
total yield of guanine oxidation products. Guanine was
selectively oxidized due to its low oxidation potential,53 whereas
the other bases remained mainly intact (data not shown).
The concentration profile for the DNA oxidation by dioxetane
2 clearly shows (Figure 3) that this hydroxymethyl-substituted
1,2-dioxetane is the least reactive in regard to guanine conver-
sion and oxazolone formation, which parallels the findings with
plasmid pBR 322 DNA. This is demonstrated by a lower
consumption of guanine together with a lower yield of oxazolone
(Figure 3). The formation of 8-oxoGua dominates with increas-
ing concentration of the dioxetane, presumably because under
these conditions no significant further oxidation takes place and
it accumulates. Therefore, the qualitative reactivity order for
the dioxetanes 1R/1â > HTMD > 2 . TMD applies.
Oxidation of dGuo. To gain mechanistic insight into which
processes may be responsible for the high reactivity of hy-
droxymethyl-substituted dioxetanes 1R, 1â, 2, and HTMD in
the formation of oxidative DNA-base modifications, further
studies with the nucleoside 2’-deoxyguanosine (dGuo) were
performed. The dGuo (0.5 mM) was treated with the respective
dioxetanes at various concentrations and 50 °C for 15 h. The
four major oxidation products of the guanine base, namely the
diastereomeric mixture of (4R*)- and (4S*)-4-HO-8-oxodGuo
(type-II photooxidation),18,19 8-oxodGuo20 (formed in dGuo only
by type-II process), and the guanidine-releasing products
oxazolone16 and oxoimidazolidine17 (type-I photooxidation),
were observed and quantified as previously described.9-11 For
all dioxetanes an almost linearly dependent degradation of dGuo
was observed with increasing dioxetane concentration (Figure
4). The product balance was 50 ( 8% and nearly the same for
all dioxetanes.
Figure 3. Concentration profiles for the oxidation of calf thymus DNA
(0.1 mg/mL, 62.5 µM guanine) by the hydroxymethyl-substituted
dioxetanes HTMD, 1R/1â, and 2 in the dark and in acetonitrile (10 vol
%) at 37 °C for 20 h in 5 mM phosphate buffer (pH 7.0, 90 vol %).
The ordinate values are averages of at least three independent runs,
error ( 10% of the stated value.
reaction conditions, the yield of oxazolone constitutes a more
reliable measure of the oxidative reactivity of the dioxetanes.
The concentration profiles are displayed in Figure 3. For
the dioxetanes 1R and 1â, which were used as a 40:60
diastereomeric mixture (the isomerically pure dioxetanes 1R and
1â gave exactly the same results, data not shown), an increase
in the dioxetane concentration from 1 to 10 mM enhanced the
oxazolone yield ca. 4-fold. As a second marker for the oxidative
reactivity of the dioxetanes, the conversion of guanine also rises
at higher dioxetane concentration (Figure 3). In contrast, the
yield of 8-oxoGua decreased ca. 3-fold when the dioxetane
concentration was increased from 1 to 10 mM. Compared to
the well-investigated HTMD, the dioxetanes 1R/1â are more
effective, which is clearly reflected in the conversion of guanine
as well as in the yield of oxazolone (Figure 3), but not in the
Mechanistically more significant, the concentration profiles
for the dGuo oxidation in Figure 4 reflect the reactivity trend
of the hydroxymethyl-substituted dioxetanes obtained in the
(53) (a) Steenken, S. Chem. ReV. 1989, 89, 503-520. (b) Candeias, L.
P.; Steenken, S. J. Am. Chem. Soc. 1989, 111, 1094-1099. (c) Steenken,
S.; Jovanovic, S. V. J. Am. Chem Soc. 1997, 119, 617-618.

3556 J. Am. Chem. Soc., Vol. 120, No. 15, 1998
Adam et al.
Figure 4. Concentration profiles for the thermal photooxidation of dGuo by hydroxymethyl-substituted 1,2-dioxetanes in the dark with 0.5 mM
dGuo at 50 °C for 15 h in 5 mM phosphate buffer (pH 7.0) and acetonitrile (10 vol %) as cosolvent. The ordinate values are averages of at least
three independent runs, error ( 10% of the stated value.
Table 4. Inhibitory Effect of Additives on the Oxidation of dGuo
studies with DNA (strand cleavage and oxidative damage). A
Induced in the Thermolysis of the Hydroxymethyl-Substituted
1,2-Dioxetanes HTMD, 1R/1â, and 2a or in the Irradiation of
control experiment under the same reaction conditions confirmed
the low reactivity of the merely alkyl-substituted derivative
TMD (20 mM, only 2-3% conversion of dGuo).
Benzophenone (BP)a
inhibitory effectb (%)
In this context and for comparison purposes, it was of interest
to determine the efficacy of the photochemically excited ketones,
namely 1-phenyl-2-propanone, 1-hydroxy-3-phenyl-2-propanone,
and 1,3-diphenyl-2-propanone (the thermolysis products of the
various dioxetanes), to cause dGuo oxidation. Therefore, 0.25
mM dGuo was irradiated at 300 nm and 0 °C for 30 min in the
presence of 10 mM of these ketones54 and the conversion of
dGuo was measured. For the 1,3-diphenyl-2-propanone, an
efficient source of phenylmethyl radicals,43,47,49-51 only 15%
of dGuo conversion compared to 85% in the case of 1-phenyl-
2-propanone,54 which releases acetyl as well as phenylmethyl
radicals, was observed.32,42 This reflects the higher propensity
of acetyl versus phenylmethyl radicals to degrade dGuo oxi-
datively. The hydroxymethyl radical, which is generated in the
photolysis of 1-hydroxy-3-phenyl-2-propanone along with the
phenylmethyl radical (after decarbonylation), shows a moderate
tendency to oxidize dGuo (50% conversion of dGuo). The low
reactivity of the phenylmethyl radicals was independently
confirmed by the rather ineffective oxidation of dGuo (4%
conversion of dGuo compared to 40% for HTMD) when 3,3-
bis(phenylmethyl)-1,2-dioxetane (DBD, 20 mM) was employed
as efficient thermal source of such species.55
Significant differences in the product balances of type-I and
type-II photooxidation products were found for the dioxetanes
1R/1â and 2. The predominant oxidative modifications in the
thermolysis of dioxetane 1R and 1â are the type-I products
oxazolone and oxoimidazolidine, which were generated in ca.
2-fold higher amounts compared to the characteristic type-II
product 4-HO-8-oxodGuo (Figure 4). The other type-II pho-
tooxidation product, namely the easily oxidized 8-oxodGuo, was
generated in negligible yields (<0.1%). A completely different
additivec
concentration
HTMD
1R/1â
2
BP
cresold
20 mM
8 vol %
20 mM
5 mM
34
70
58
37
90
37
80
27
45
90
33
70
60
42
90
47
88
iPrOH
DBH
nde
87
galvinoxyl
DMPOf
2 mM
94
a 0.5 mM dGuo, 5 mM phosphate buffer (pH 7.0), 10% acetonitrile
as cosolvent, 10 mM dioxetane (thermolyzed at 50 °C for 14 h) or 1
mM benzophenone (photolyzed at 350 nm in the Rayonet photoreactor
at 0 °C for 30 min). b Mean values determined from the conversion of
dGuo with and without additives according to the following: inhibitory
effect (%) ) [1 - (conv. dGuo with additive/conv. dGuo without
additive)] × 100, mean values of three independent experiments,
standard deviation ( 10% of the stated value. c Cresol: 2,6-di-tert-
butyl-4-methylphenol. DBH: 2,3-diazabicyclo[2.2.1]hept-2-ene. d 0.1
mM dGuo, 5 mM phosphate buffer (pH 7.0), 10% acetonitrile as
cosolvent, 2 mM dioxetane (thermolyzed at 50 °C for 7 h) or 1 mM
benzophenone (photolyzed at 350 nm in the Rayonet photoreactor at 0
°C for 30 min). e nd: not determined. 0.1 mM dGuo, 5 mM phosphate
buffer (pH 7.0), 10% acetonitrile as cosolvent, 2 mM dioxetane
(thermolyzed at 50 °C for 14 h) or 0.5 mM benzophenone (photolyzed
at 350 nm in the Rayonet photoreactor at 0 °C for 30 min).
f
situation applies in the oxidation of dGuo by dioxetane 2, for
which the ratio of type-I to type-II products is even smaller
than unity and significant amounts of 8-oxodGuo (up to 1.8%)
were detected. A quite similar trend was found for HTMD,
for which maximally 0.4% 8-oxodGuo and a type-I to type-II
ratio of about unity were obtained.9,11
To assess the nature of the oxidizing species derived from
the dioxetanes 1R/1â, 2, and HTMD upon thermolysis, the
inhibitory effect of a number of additives was tested in the
dioxetane-mediated photooxidation of dGuo (Table 4) and
compared to photoexcited benzophenone (BP) triplets.
DMPO,33-35 cresol,55a and galvinoxyl55a were representatively
examined as well-established radical scavengers. Such radical
scavengers do not only intercept radicals, but also influence the
triplet-excited states. Triplet states may be deactivated through
hydrogen abstraction (DMPO and cresol) or by assisted
intersystem crossing (paramagnetic galvinoxyl). Indeed, a
strong inhibitory effect was observed for these additives with
(54) Carried out in 5 mM phosphate buffer (pH 7.0) with acetonitrile
(10 vol %) as cosolvent. The values for the conversion of dGuo are averages
of at least three independent runs, error ( 10%. The least effective ketone,
namely 1,3-diphenyl-2-propanone, has the highest absorption coefficient
[ꢀ (308 nm, CH3CN) ) 120] compared to 1-phenyl-2-propanone and
1-hydroxy-3-phenyl-2-propanone [ꢀ (308 nm, CH3CN) ) 40].
(55) (a) Adam, W.; Heil, M. J. Am. Chem. Soc. 1992, 114, 8807-8809.
(b) Adam, W.; Andler, S. J. Am. Chem. Soc. 1994, 116, 5674-5678.

OxidatiVe DNA Damage
J. Am. Chem. Soc., Vol. 120, No. 15, 1998 3557
Figure 6. Effect of oxygen on the photooxidation of dGuo (0.5 mM)
in the thermal decomposition of hydroxymethyl-substituted 1,2-
dioxetanes at 50 °C for 15 h in 5 mM phosphate buffer (pH 7.0) and
acetonitrile (10 vol %) as cosolvent. The ordinate values are averages
of at least three independent runs, error ( 10% of the stated value.
For the saturation with molecular oxygen, the solutions were purged
with oxygen gas for 1 min. Deaerated solutions were obtained by two
freeze-pump-thaw cycles and saturated with argon gas.
Figure 5. Effect of D2O on the 8-oxodGuo yield in the photooxidation
of dGuo induced by the thermolysis of the hydroxymethyl-substituted
1,2-dioxetanes HTMD, 1R/1â, and 2 in the dark at 50 °C for 15 h with
0.25 mM dGuo and 1.25 mM dioxetane in 5 mM phosphate buffer
[pH(D) 7.0] and acetonitrile as cosolvent (10 vol %). Absolute yields
are derived from the mean value of at least three independent runs,
error ( 10% of the stated value.
the long-lived (ca. 25 µs) triplet state of benzophenone,56 which
is known to cause dGuo photooxidation.16c Therefore, at least
one part of the inhibitory effect in the cases of HTMD and 2
may also derive from the deactivation of the relatively long-
lived (20-50 µs)48 triplet-excited acetone as decomposition
product of the dioxetanes. The dioxetanes 1R and 1â, however,
do not yield long-lived triplet ketones upon thermal decomposi-
tion, and thus, the observed inhibitory effect in the dioxetane-
induced photooxidation of dGuo (Table 4) is attributed to the
scavenging of reactive radicals. Analogously, the substantial
reduction of the dGuo oxidation by the dioxetanes 1R and 1â
with 2-propanol as hydrogen donor (Table 4) also implicates a
decrease of the active radical concentration. For the dioxetanes
HTMD and 2 and for benzophenone 2-propanol may also
intercept the triplet-excited ketones and, thus, account for the
observed inhibition.
Also mechanistically instructive is the azoalkene DBH, which
was employed as an efficient triplet-excited ketone quench-
er.55b,57-59 As expected, DBH efficiently suppressed the
conversion of dGuo for all dioxetanes which release long-lived
ketone triplet states (HTMD and 2) and, of course, for
benzophenone (Table 4). In marked contrast, in the case of
the dioxetanes 1R and 1â, which do not generate long-lived
triplet-excited carbonyl products to allow quenching by a triplet-
energy acceptor, the inhibitory effect was quite small and, in
fact, may be due to some (less efficient) interception of radicals
by DBH through scavenging.
for dGuo, the yield of 8-oxodGuo was significantly enhanced
in D2O compared to H2O. The D2O effect for the dioxetanes 2
and HTMD (300-320% increase in 8-oxodGuo formation) was
much more pronounced than for the dioxetanes 1R and 1â
(∼150% increase), for which only a very small, if at all
significant D2O effect was observed. This contrast is readily
accounted for since HTMD and 2, but not 1R and 1â, release
long-lived triplet ketones upon thermolysis, which are capable
of undergoing energy transfer to produce singlet oxygen (type-
II photooxidation).62 Note in Figure 5 that the total amount of
8-oxodGuo, the product derived from singlet oxygen with dGuo,
is about the same for HTMD and 2, which release relatively
long-lived triplet acetone, but very small for the dioxetanes 1R
and 1â, which produce exclusively short-lived ketone triplets.
These results are in accord with the type-I versus type-II product
ratios observed in the oxidation of dGuo by the dioxetanes 1R/
1â, 2, and HTMD.
Influence of Molecular Oxygen on the Photooxidation of
dGuo. To assess whether peroxyl radicals, produced by
dioxygen trapping of the carbon radicals63 derived from the R
cleavage of the electronically excited R-hydroxy ketones, are
responsible for the formation of oxidative DNA base modifica-
tions by these highly reactive hydroxymethyl-substituted diox-
etanes, the effect of molecular oxygen was examined. For this
purpose, the dioxetane-mediated conversion of dGuo was
investigated in the presence (saturated with O2) and absence
(degassed with Ar) of molecular oxygen (Figure 6). For the
dioxetanes HTMD, 1R/1â, and 2, the conversion of dGuo in
the presence of O2 was within the error limit the same as under
normal aerated conditions (Figure 6). This reveals that under
regular conditions the oxygen concentration in water (0.5-1
mM)64 is sufficient for scavenging the carbon radicals gradually
formed during the R cleavage. In the absence of O2 (degassed
with argon), however, a significant reduction (ca. 70-80%) of
the dGuo oxidation was observed.
To confirm the participation of singlet oxygen, which is also
known to cause oxidative DNA-base modifications, the diox-
etane-mediated oxidation of dGuo was examined in deuterium
oxide, in which the lifetime of singlet oxygen is prolonged by
about 10-fold.60,61 The results for the dioxetane-induced forma-
tion of 8-oxodGuo in H2O and D2O are shown in Figure 5. At
concentrations of 1.25 mM for the dioxetanes and 0.25 mM
(56) Benasson, R. V.; Gramain, J.-C. J. Chem. Soc., Faraday Trans. 1
1980, 76, 1801-1810.
(62) Wu, K. C.; Trozzolo, A. M. J. Photochem. 1979, 10, 407-410.
(63) For example: (a) Tappel, A. L. Fed. Proc. 1973, 32, 1870-1874.
(b) von Sonntag, C.; Schuchmann, H.-P. Angew. Chem., Int. Ed. Engl. 1991,
30, 1229-1253. (c) Marchaj, A.; Kelley, D. G.; Bakac, A.; Espenson, J.
H. J. Phys. Chem. 1991, 95, 4440-4441. (d) Dix, T. A.; Aikens, J. Chem.
Res. Toxicol. 1993, 6, 2-18. (e) Step, E. N.; Turro, N. J.; Gande, M. E.;
Klemchuk, P. P. Macromolecules 1994, 27, 2529-2539. (f) Brown, C. E.;
Neville, A. G.; Rayner, D. M.; Ingold, K. U.; Lusztyk, J. Aust. J. Chem.
1995, 48, 363-379.
(57) Catalani, L. H.; Bechara, E. J. H. Photochem. Photobiol. 1984, 39,
823-830.
(58) Kavarnos, G. J.; Turro, N. J. Chem. ReV. 1986, 86, 401-449.
(59) Catalani, L. H.; Wilson, T.; Bechara, E. J. H. Photochem. Photobiol.
1987, 45, 273-281.
(60) Rodgers, M. A. J.; Snowden, P. T. J. Am. Chem. Soc. 1982, 104,
5541-5543.
(61) Kearns, D. R. In Singlet Oxygen; Wasserman, H. H., Murray, R.
W., Eds.; Academic Press: New York, San Francisco, London, 1979; pp
115-137.
(64) Weast, R. C. Handbook of Chemistry and Physics, 67th ed.; CRC
Press: Cleveland, OH, 1987.

3558 J. Am. Chem. Soc., Vol. 120, No. 15, 1998
Adam et al.
Mechanistic Discussion
source, namely 1-hydroxy-2-propanone (for HTMD) or 1-hy-
droxy-3-phenyl-2-propanone (for dioxetane 2), is formed by the
less reactive dioxetanes HTMD and 2 (triplet acetone R-cleaves
inefficiently47), whereas the most reactive dioxetanes 1R and
1â release two radical precursors, i.e. 1-phenyl-2-propanone and
1-hydroxy-2-propanone.22,32,42 Therefore, the number of po-
tential radicals per dioxetane molecule decomposed is 4:2:2 for
1R/1â, HTMD, and 2 and it is tempting to attribute the higher
reactivity of the dioxetanes 1R and 1â to this enhanced radical
formation. However, the mere number of potential radicals per
dioxetane molecule decomposed as a criterion for the reactivity
of the dioxetanes does not account adequately the difference in
the DNA-damaging efficiency of the dioxetanes HTMD and 2
Analogous to HTMD, also the hydroxy-functionalized diox-
etanes 1R, 1â, and 2 cause DNA damage, principally strand
breaks (Figure 2) and guanine oxidation (Figures 3 and 4). Their
efficacy follows the reactivity sequence 1R/1â > HTMD > 2
. TMD. While for some time the high efficacy of HTMD to
promote DNA modifications was a mystery,9-11 through the
present study we know now that radicals, in particular acetyl
radicals, are responsible for this oxidative reactivity and
presumably also for the formation of DNA strand breaks in the
case of the hydroxymethyl-substituted dioxetanes HTMD, 1R/
1â and 2 versus TMD; the latter does not release radicals! In
view of the large volume of results given here, it should be
helpful for the mechanistic discussion to summarize briefly the
salient experimental findings and reactivity trends, which are
(a) the thermal decomposition of the hydroxymethyl-substituted
dioxetanes HTMD, 1R/1â, and 2 leads to acyl and alkyl radicals
as shown by spin-trapping experiments with DMPO (Table 2),
(b) the laser-flash photolysis of the R-hydroxy- and R-phenyl-
functionalized ketones (decomposition products of the dioxet-
anes HTMD, 1R/1â, and 2) reveal short lifetimes for their triplet-
excited states due to fast and efficient R cleavage, (c) the
significant reactivity differences of the dioxetanes in the
formation of DNA strand breaks and the oxidation of guanosine
are found to correlate qualitatively with a reactivity order of
the released radicals
•
(Scheme 4). Also the capacity of the radicals RC(O)•, HOCH2 ,
•
and PhCH2 to induce DNA damage needs to be considered.
Indeed, the reactivity profile obtained in photooxidation studies
of dGuo with various R-phenyl-substituted ketones reveals the
qualitative order RC(O)• > HOCH2 . PhCH2 , with the
phenylmethyl radical by far the least reactive. Thus, the number
of actiVe radicals per dioxetane molecule decomposed capable
to induce DNA damage is for HTMD twice that for 2 (Scheme
4). For HTMD a pair of acetyl (active) and hydroxymethyl
(active) radicals are obtained from the triplet-excited 1-hydroxy-
2-propanone, whereas for the dioxetane 2 the R cleavage of
1-hydroxy-3-phenyl-2-propanone leads to a pair of phenylmethyl
(inactive) and hydroxymethyl (active) radicals. Adjustment for
the number of actiVe radicals formed per dioxetane molecule
decomposed, the reactivity order 1R/1â > HTMD > 2 of the
dioxetanes to damage DNA is nicely accounted for in terms of
the R cleavage of the triplet-excited ketones produced in the
dioxetane thermolysis. Consequently, the much lower reactivity
of TMD to induce DNA damage than the hydroxy-substituted
derivatives is due to the lack of R cleavage of triplet acetone;
the latter operates through less effective direct interaction with
•
•
•
•
H3CC(O)• > HOCH2 . PhCH2
and (d) the involvement of reactive peroxyl radicals as the
proposed ultimate oxidizing species in the dioxetane-mediated
oxidation of DNA in the presence of molecular oxygen, as
suggested in the photochemically induced R cleavage of
R-substituted ketones in oxygen-saturated solutions.63e
1
the DNA by sensitization66 or/and through O2 generation by
3
energy transfer to O2.62
The generation of radicals from the hydroxymethyl-substituted
dioxetanes may be rationalized in terms of the well-established
R cleavage of their R-functionalized, n,π*-triplet-excited car-
bonyl products formed upon thermal decomposition (Scheme
3).22,23,32,42 Such R cleavage of the hydroxy- and phenyl-
substituted carbonyl compounds leads to radicals, which have
been trapped by DMPO and identified by EPR spectroscopy.
Since laser-flash-photolysis studies establish a short triplet
lifetime (τ < 1 ns) for these R-hydroxy- and R-phenyl-
functionalized ketones, the direct interaction of such reactive,
triplet-excited species with the DNA to induce its damage is
negligible, in contrast to what has been reported for the longer-
lived, triplet-excited benzophenone65 or acetone.66 Thus, the
efficient R cleavage of the triplet-excited, R-substituted ketones
qualifies them as good radical sources in the thermally initiated,
dioxetane-induced DNA damage. Indeed, the fact that the
addition of radical scavengers significantly inhibits the oxidation
of guanosine by the dioxetanes 1R and 1â (Table 4) supports
that trappable radicals are involved in the formation of oxidative
DNA-base modifications by the hydroxymethyl-substituted
dioxetanes HTMD, 1R/1â, and 2.
The carbon radicals produced as the primary reactive species
in the R cleavage of R-substituted, triplet-excited ketones are
efficiently scavenged by molecular oxygen to form peroxyl
radicals (k > 2 × 109 s-1).63 This was previously corroborated
in the photolysis of 2,4-diphenylpentan-3-one in oxygen-
saturated hexane solution, which provides a clean method for
the production of peroxyl radicals.63e Since hydroxymethyl-
substituted dioxetanes are effective in the oxidation of dGuo
under normal aerated but not under argon-deaerated conditions
(Figure 6), we suggest that mainly peroxyl radicals are the
ultimate oxidizing species;67 however, the direct participation
of carbon-centered radicals in the oxidation of guanine cannot
be excluded. Unfortunately, to date, little is known on the
oxidative modification of DNA bases by peroxyl radicals.68
A closer look at the product distribution in the dioxetane-
mediated oxidation of the nucleoside dGuo provides information
on the mode of oxidative action of the peroxyl and/or carbon-
centered radicals. For this purpose, the ratio of the radical-
1
derived type-I (guanidine-releasing products) versus the O2-
derived type-II (4-HO-8-oxodGuo, 8-oxodGuo) oxidation products
of guanine offers a valuable mechanistic probe.11 The observed
type-I/type-II ratio of approximately 2 (Figure 4) for the dGuo
oxidation by the dioxetanes 1R and 1â suggests that the
oxidation of guanine to the characteristic type-I products
oxazolone and oxoimidazolidine may be caused by peroxyl
radicals. Carbon-centered radicals are prone to add to the C8
To rationalize the DNA-damaging efficiency of these diox-
etanes, i.e. the reactivity trend 1R/1â > HTMD > 2 . TMD,
we shall now assess their capacity to release active radicals upon
thermal decomposition, which is depicted in Scheme 4. Com-
parison of the electronically excited carbonyl products generated
on thermolysis of these dioxetanes, only one potential radical
(65) Morin, B.; Cadet, J. Photochem. Photobiol. 1994, 60, 102-109.
(66) Epe, B.; Henzl, H.; Adam, W.; Saha-Mo¨ller, C. R. Nucleic Acid
Res. 1993, 21, 863-869.
(67) Marnett, L. J. Carcinogenesis 1987, 8, 1365-1373.
(68) Martini, M.; Termini, J. Chem. Res. Toxicol. 1997, 10, 234-241.

OxidatiVe DNA Damage
J. Am. Chem. Soc., Vol. 120, No. 15, 1998 3559
Scheme 5. Proposed Mechanism for the Formation of Guanidine-Releasing Products in the Thermolysis of Hydroxymethyl-
Substituted 1,2-Dioxetanes by Radicals70, 71
position of guanine;69 nevertheless, the direct participation of
such radicals in the formation of oxazolone and oxoimidazo-
lidine cannot be excluded (cf. Scheme 5). The formation of
these guanine oxidation products may be explained in terms of
the well-established mechanism70 with the guanine radical as
intermediate (Scheme 5). Therefore, we propose that the active
radicals formed in the thermal decomposition of the hydroxy-
methyl-substituted 1,2-dioxetanes trap molecular oxygen to
afford peroxyl radicals, which generate guanine radicals by
hydrogen abstraction (Scheme 5).71
Conclusion
The present in-depth study of the DNA damage caused by
hydroxymethyl-substituted dioxetanes has allowed us to ratio-
nalize their high efficiency compared to merely alkyl-substituted
derivatives in terms of radical chemistry. The radicals, which
are formed through the efficient R cleavage of the R-substituted,
triplet-excited ketones that are generated in the dioxetane
thermolysis, have been identified as the reactive species by EPR
spin trapping and time-resolved UV spectroscopy. Due to the
established fast and efficient O2 trapping of the initially formed
carbon-centered radicals to afford the peroxyl radicals,63 we
suggest that the latter act as the ultimate active oxidizing species.
Therefore, we propose that peroxyl radicals cause the oxidative
guanine modifications and DNA strand breaks and it seems
likely that such oxygen-centered radicals might be involved in
the DNA damage in vivo,67,68 a novel aspect of oxidatiVe stress
that merits more intensive work. Clearly, the hydroxymethyl-
substituted dioxetanes should serve this purpose well, since they
constitute efficient sources for carbon-centered radicals as
potential precursors for peroxyl radicals at 37 °C in the dark,
certainly advantageous mild conditions for biological studies.
In contrast, the type-I/type-II ratio of about unity for HTMD
and 0.7 for the dioxetane 2 (Figure 4) implies that reactive
species other than radicals are involved in the dioxetane-induced
guanosine oxidation. The significant increase in the relative
amount of the characteristic singlet-oxygen-derived oxidation
products 8-oxodGuo72 (in DNA it may be also formed by a
type-I process)70 and 4-HO-8-oxodGuo18 for these dioxetanes
compared to 1R and 1â speaks definitively for the involvement
1
of O2 as reactive oxygen species. This was confirmed for
HTMD10 and the dioxetane 2 (Figure 5) by the substantial D2O
effect in the formation of 8-oxodGuo. This is readily accounted
for since only HTMD and 2 release triplet-excited acetone,
which lives long enough to be quenched by molecular oxygen
and to produce significant amounts of singlet oxygen.62 In the
case of the dioxetane 2, the extent of dGuo oxidation by 1O2 is
even more pronounced than for HTMD (Figure 4). Although
both dioxetanes afford triplet-excited acetone, 2 produces only
the hydroxymethyl radical as active DNA oxidant, whereas
HTMD leads also to the acetyl radical as reactive species and,
consequently, more singlet oxygen but less radical activity is
expected for dioxetane 2 versus HTMD.
Acknowledgment. This work was supported by the Deutsche
Forschungsgemeinschaft (Sonderforschungsbereich 172 “Mole-
kulare Mechanismen kanzerogener Prima¨rvera¨nderungen“) and
the Fonds der Chemischen Industrie. W.M.N. thanks the latter
institution for a Liebig fellowship and the Schweizer Nation-
alfond. We thank G. N. Grimm for help with the EPR
spectroscopy.
(69) Cf., for example: (a) Salomon, J.; Elad, D. J. Org. Chem. 1973,
38, 3420-3421. (b) Maeda, M.; Nushi, K.; Kawazoe, Y. Tetrahedron 1974,
30, 2677-2682. (c) Zady, M. F.; Wong, J. L. J. Am. Chem. Soc. 1977, 99,
5096-5101. (d) Zady, M. F.; Wong, J. L. J. Org. Chem. 1979, 44, 1450-
1454. (e) Hiramoto, K.; Kaku, M.; Sueyoshi, A.; Fujise, M.; Kikugawa, K.
Chem. Res. Toxicol. 1995, 8, 356-362.
Supporting Information Available: Text describing ex-
perimental details for the preparation and spectral data of the
key compounds, tables with the decomposition rate constants,
the chemiluminescence intensities for the determination of the
singlet (φS) and triplet (ΦT) quantum yields of the dioxetanes
1R/1â and 2, and the rate constants of hydrogen abstraction by
triplet-excited acetone from diphenylmethanol, and selected EPR
spectra (20 pages, print/PDF). See any current masthead page
for ordering information and Web access instructions.
(70) Cadet, J.; Berger, M.; Morin, B.; Ravanat, J.-L.; Raoul, S.; Buchko,
G. W.; Weinfeld, M. Spectrum 1994, 7, 21-24.
(71) The hydrogen abstraction from O-H and/or N-H bonds to generate
the resonance-stabilized guanine radical by peroxyl radicals should be in
principle feasible, cf.: (a) Ingold, K. U. Acc. Chem. Res. 1969, 2, 1-9. (b)
Nangia, P. S.; Benson, S. W. Int. J. Chem. Kinet. 1980, 12, 29-42.
(72) Devasagayam, T. P. A.; Steenken, S.; Obendorf, M. S. W.; Schulz,
W. A.; Sies, H. Biochemistry 1991, 30, 6283-6289.
JA9726318