N-Aroyloxynaphthalimides as novel highly efficient DNA photocleavers: Substituent effects

Article abstract of DOI:10.1016/S0040-4039(01)01214-X

A novel DNA photocleaver family of N-benzoyloxythioxonaphthalimides, which show improved DNA intercalation and photocleaving activity compared with their oxo- counterparts, were synthesized and evaluated. The substituents on the benzene ring of the benzoyloxy moiety have an important and unusual influence on the DNA photocleaving activity. N-(p-Methoxylbenzoyloxy)thioxonaphthalimide is able to photocleave DNA at low concentration (5 μM) with 'time-resolved' and pH independent properties.

Full text of DOI:10.1016/S0040-4039(01)01214-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6175–6178
N-Aroyloxynaphthalimides as novel highly efficient DNA
photocleavers: substituent effects
Xuhong Qian,a,* Wei Yao,b Gang Chen,b Xiayu Huangb and Ping Maob
aState Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116012, China
bInstitute of Pesticides & Pharmaceuticals, East China University of Science and Technology, Shanghai 200237, China
Received 30 April 2001; accepted 5 July 2001
Abstract—A novel DNA photocleaver family of N-benzoyloxythioxonaphthalimides, which show improved DNA intercalation
and photocleaving activity compared with their oxo- counterparts, were synthesized and evaluated. The substituents on the
benzene ring of the benzoyloxy moiety have an important and unusual influence on the DNA photocleaving activity.
N-(p-Methoxylbenzoyloxy)thioxonaphthalimide is able to photocleave DNA at low concentration (5 mM) with ‘time-resolved’ and
pH independent properties. © 2001 Elsevier Science Ltd. All rights reserved.
There are a variety of studies concerning DNA photo-
cleavage by photosensitizers which either initiate a sin-
gle electron transfer from a base to the triplet state of
the chromophore or generate an active oxygen species
upon photoirradiation. Oxygen radicals are very impor-
tant in chemistry, biology and medicine.1–3 In recent
years, the design and study of DNA ‘artificial photo-
nucleases’ of the oxygen radical type with no need of a
metal or external oxidant, have attracted much atten-
tion and have led to the development of both specific
gene-targeted drugs and artificial restriction enzymes.4,5
However, there have been few reports about substituent
effects on DNA photocleaving activity. We have
reported the effects of thioxo- and oxo- groups in
aromatic heterocycles as DNA photocleavers of the
electron-transfer type, and that of trifluoromethyl- and
methyl groups in hydroperoxides as DNA photo-
cleavers of the hydroxyl radical type. Here we report
important and unusual substituent effects in N-aroyl-
oxynaphthalimides as DNA photocleavers of the aroyl-
oxy radical type.
It is known that N-hydroxypyridinethiones (A1, A2)
and N-aroyloxypyridine-2-thiones (B) are DNA photo-
cleavers through release of hydroxy or aroyloxy
radicals6,7 (Scheme 1). On the other hand, naphthal-
imides usually exhibit strong, long-wavelength absorp-
tion and good DNA intercalative binding abilities,
Scheme 1.
* Corresponding author. E-mail: xhqian@dlut.edu.cn, xhqian@ecust.edu.cn
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01214-X

6176
X. Qian et al. / Tetrahedron Letters 42 (2001) 6175–6178
whilst some hydroperoxides and other derivatives have
shown DNA photocleaving activity.8–10 Therefore, by
integrating the N-aroyloxy-2-thiopyridone photocleav-
ing moiety with the naphthalimide intercalative moiety,
a novel highly efficient DNA photocleaving family of
reagents of N-aroyloxythionaphthalimides (C1ꢀC5)
became our target.
Figure 2. ESR spectroscopy of the radical released by C1.
ESR with PBN (N-tert-butyl-a-phenylnitrone) trapping in
benzene under light. Microwave frequency 9.782 GHz, mid
range 3430 G, scan range 100 G.
The target N-benzoyloxythionaphthalimides C1ꢀC5
were synthesized by condensation of N-hydroxy–thio-
naphthalimide with substituted benzoyl chlorides. The
oxo-compound C6 was synthesized for comparison and
1
all the compounds were identified by using H NMR,
system by internal conversion processes. Meanwhile, a
study of the intercalation of C1 and C6 to calf thymus
DNA was carried out using an electronic absorption
spectra technique instead of the fluorescence quenching
method,13 as C1ꢀC5 have only very weak fluorescence.
During addition of calf thymus DNA the absorption of
C1 became weaker in intensity with a red shift for the
maxima with an increase in the concentration of DNA
(Fig. 3). C1 had obvious DNA intercalating activity
EI-MS, IR and elemental analysis (Scheme 2).
Interestingly, thioxo-compound C1 behaved differently
from oxo-compound C6 in DNA photocleavage. Fig.
1a shows that C1 could cleave the DNA plasmid effec-
tively under irradiation, while its oxo-analog had little
effect. It is widely accepted that the DNA photocleav-
ing activities of N-hydroxy- and N-aroyloxypyridine-2-
thione are attributable to the formation of persistent
hydroxyl or aroyloxyl radicals during the photolysis,
and that they are able to give ESR signals through spin
trapping methods, whilst a thiyl radical signal is not
detected under similar conditions.6,11,12 In our case,
PBN (N-tert-butyl-a-phenylnitrone) trapping ESR
spectroscopy identified that C119 in benzene could gen-
erate a free radical signal under irradiation (Fig. 2), but
not in the case of C6.21 This suggested that free benzoyl-
oxy radicals might be responsible for the DNA damage
caused by C1.
It was found that compared with C6 all thioxo-com-
pounds (C1ꢀC5) had a long-wavelength of absorption
and a very weak fluorescence (Table 1). This implies
that for the latter, their excitation energy is more easily
transferred from the singlet excited state to cleave the
NꢀO bond,11 besides being dissipated as heat in the
Figure 3. Interaction of C5 and DNA. Absorption changes
from 300–500 nm of compound C5 (20 mM) during addition
of calf thymus DNA (0, 50, 100, 200 mM) in THF, 10 mM
Tris·HCl (pH 7.4) solution (3:10, v/v).
Scheme 2.
Figure 1. Photocleavage on pBR 322 DNA. The cleavage activities were evaluated using supercoiled circular pBR322 (form I)
DNA (50 M/bp) with the compound in the buffer Tris·HCl (pH 7.6) under photoirradiation (2300 W/cm2) with a transluminator
(366 nm) at a distance of 20 cm at 0oC and then analyzed on a 1% agarose gel. DNA photocleavage efficiency was defined as the
degree of relaxation of the supercoiled DNA, relaxed circular DNA (single-stranded cleavage) as form II. (a) Comparison of
photocleavage on pBR 322 DNA by C1 and C6 (100 mM; hw, 1 h). Lane 1: DNA alone (no hw); lane 2: DNA and C6 (no hw); lane
3: DNA and C6 (hw); lane 4: DNA and C1 (no hw); lane 5: DNA and C1 (hw); lane 6: DNA alone (hw). (b) Effects of different
substituents on photocleavage (100 mM; hw, 45 min). Lane 1: DNA alone (no hw); lane 2: DNA and C1; lane 3: DNA and C 2;
lane 4: DNA and C3; lane 5: DNA and C4; lane 6: DNA andC5; lane 7: DNA alone (hw).

X. Qian et al. / Tetrahedron Letters 42 (2001) 6175–6178
Table 1. UV–vis and FL spectra data of N-benzoyloxynaphthalimidesa,b
6177
Compound
UV umax/nm (log m)
FL umax/nm (ƒ)
Compound
UV umax/nm (log m)
FL umax/nm (ƒ)
C1
C2
C3
C4
392 (4.29)
400 (4.37)
390 (4.34)
390 (4.23)
491 (B0.001)
454 (B0.001)
466 (B0.001)
466 (B0.001)
C5
C6
B1
390 (4.21)
350 (4.04)
365 (3.69)
454 (B0.001)
378 (0.10)
409.6 (0.047)
a In absolute ethanol.
b With quinine sulfate in sulfuric acid as quantum yield standard.
Fig. 4a indicates that the benzoyloxyl radicals were
generated from C520 at a relatively linear rate. This
continuous generation of reactive oxygen-centered radi-
cals provides an alternative to Fenton-based chemistry,
where oxygen radicals are formed in a rapid burst. This
kind of reagent is attractive for ‘time-resolved’ DNA
cleavage studies.18 Fig. 4b demonstrates that C5 could
damage DNA effectively at a concentration of 5 mM.
While it has been reported and has also been found in
our experiments that B1 photocleaves DNA at a con-
centration as high as 500 mM.10 This further confirmed
the rational design of the new molecules through inte-
grating the photocleaving moiety and the intercalative
moiety through a covalent linkage. Fig. 4c shows that
pH had little influence on the photocleaving activity of
C5, which means that the benzoyloxyl radicals are quite
stable over a wide range of pH values. In addition,
preliminary experimentals in polyacrylamide gel elec-
trophoresis showed that C5 could give different photo-
cleaving results to double strands of 5%-GCGCGCGC
and 5%-ATATATAT and a detailed study of its specific
photocleavage of duplex DNA will be carried out.
while C6 did not, which would also make them differ in
their photobiology as DNA binding is helpful for the
photocleavage. In addition, we found that thioxo-com-
pounds as DNA photocleavers of the electron transfer
type, had better photocleaving and intercalative activity
than the corresponding oxo-compounds,14,15 but the
difference was not great.
The photocleaving ability for DNA double strand-scis-
sion was evaluated for all the novel N-benzoyloxy-
thionaphthalimides (Fig. 4b). It is known that in radical
chemistry, electron donating substituents can improve
the stability of oxygen- and heteroatom-centered radi-
cals, while electron withdrawing substituents do not.16
The more stable the oxygen-centered radicals, the
stronger the radical DNA photocleaving ability should
be. In our case, it was found that the DNA photocleav-
ing abilities of these substituents were in the order:
CH3O>H>CH3>Cl>F; the compound with the methoxy
group (C6) showed the highest activity, while that with
fluoro- group was the worst. This observation in the
DNA system under light irradiation was very different
from the case of O-type radicals in a chemical system,16
and the photocleaving abilities of C1 and C5 were at
the same level (54.8 and 56.8%), so the electron donat-
ing effect was not so important. This at least means
that there might be some other unknown factor playing
a very important role rather than an electron effect, and
that the substituent effects in DNA photocleavage are
unusual and very complicated. Similarly, we never
observed that the trifluoromethyl group had a negative
effect on the DNA intercalative and DNA photocleav-
ing activity for DNA photocleavers of the hydroxyl
radical type in comparison with a methyl group.17
Thus, our study provides a novel ‘time-resolved’ and
highly efficient DNA photocleaving family of N-aroyl-
oxythionaphthalimides, which have more efficient
DNA binding and photocleaving activity than their
oxo-counterparts, and which generate benzoyloxyl radi-
cals upon visible light illumination and which are pH
independent (pH range: 7.0–8.5). It was found that
substitutents on the benzene ring of the benzoyloxy
group had an important and unusual influence on
DNA photocleaving activity, the methoxy compound
C5 showed optimum photocleaving activity.
Figure 4. (a) Time-dependant photocleavage of DNA using C5 (100 mM). Lane 1: DNA alone (no hw); lane 2: C5 and DNA (hw,
90 min); lane 3: C5 and DNA (hw, 75 min); lane 4: C5 and DNA (hw, 60 min); lane 5: C5 and DNA (hw, 30 min); lane 6: DNA
alone (hw, 1.5 h). (b) Concentration-dependant photocleavage of DNA using C5 (hw, 1 h). Lane 1: DNA alone (no hw); lane 2–8:
DNA and C5 at concentrations of 200, 100, 50, 20, 10, 5 and 1 mM, respectively. Lane 9: DNA alone (hw). (c) pH effect on DNA
photocleavage of C5 (100 mM; hw, 1 h). Lane 1, 3, 5, 7: DNA alone, pH 8.5, 8.0, 7.5, 7.0, respectively; lane 2, 4, 6, 8: DNA and
C5, pH 8.5, 8.0, 7.5, 7.0, respectively.

6178
X. Qian et al. / Tetrahedron Letters 42 (2001) 6175–6178
12. Boivin, J.; Crepon, E.; Zard, S. Z. Bull. Soc. Chim. Fr.
Acknowledgements
1992, 129, 145–150.
13. Gupta, M.; Ali, R. J. Biochem. 1984, 95, 1253–1257.
14. Qian, X.; Huang, T.; Wei, D.; Zhu, D.; Fan, M.; Yao, W.
J. Chem. Soc., Perkin Trans. 2 2000, 715–718.
15. Yao, W.; Qian, X.; Hu, Q. Tetrahedron Lett. 2000, 41,
7711–7715.
This work was supported by the National Natural
Science Foundation of China, Fuk Ying Tung Founda-
tion and the Ministry of Education of China.
16. Cheng, J. P.; Liu, B.; Zhao, Y.; Wen, Z.; Sun, Y. J. Am.
Chem. Soc. 2000, 122, 9987–9992.
References
17. Yao, W.; Qian, X. J. Fluor. Chem. 2000, 106, 69–72.
18. Singer, P.; Wu, C. W. J. Biol. Chem. 1987, 262, 14478.
19. C1: mp: 189–190°C EI–MS (m/z, %): 333 (M+, 1.9), 213
1. Absalon, M. J.; Kozarich, J. W.; Stubbe, J. Biochemistry
1995, 34, 2065–2075.
2. Sigman, D. S.; Chen, C.-H. B.; Gorin, M. B. Nature
+
+
’
([M+H −PhCOO ], 2.8), 105 (PhCO , 100); IR (KBr):
1993, 364, 474.
3030, 1780, 1715, 1610, 1580, 1270, 780, 740 cm−1 1H
;
3. Burdon, R. H. Free Rad. Biol. Med. 1995, 18, 775–794.
4. Nicolaou, K. C.; Dai, W. M.; Tsay, S. C.; Estevez, V. A.;
Wrasidlo, W. Science 1992, 256, 1172–1178.
5. Ikemoto, N.; Kumar, R. A.; Ling, T. T.; Ellestad, G. A.;
Danishefsky, S. J.; Patel, D. J. Proc. Natl. Acad. Sci.
USA 1995, 92, 10506–10510.
NMR (CDCl3): l 9.01 (dd, J1=7.6 Hz, J2=0.9 Hz, 1H,
7-H), 8.67 (dd, J1=7.2 Hz, J2=1.1 Hz, 1H, 2-H), 8.27
(m, 4H, 3ꢀ6H), 7.72 (m, 3H, 2%-,4%-,6%-H), 7.57 (m, 2H,
3%-,5%-H); C19H11NO3S requires: C, 68.46, H, 3.32, N,
4.20; Found: C, 68.51, H, 3.49, N, 4.22%.
20. C5: mp: 200–201°C. EI–MS (m/z, %): 363 (M+, 1.6), 213
6. Theodorakis, E. A.; Xiang, X.; Lee, M.; Gibson, T.
(1.5), 135 (p-MeOPhCO+, 100); IR (KBr): 3040, 1780.
Tetrahedron Lett. 1998, 39, 3383–3386.
;
1710, 1600, 1580, 1510, 1270, 845 cm−1 1H NMR
7. Adam, W.; Ballmaier, D.; Epe, B.; Grimm, G. N.; Saha-
Moeller, C. R. Angew. Chem., Int. Ed. Engl. 1995, 34,
2156–2158.
8. Tao, Z.; Qian, X.; Wei, D. Dyes Pigment. 1996, 31,
245–251.
(DMSO-d6): l 8.92 (dd, J1=7.6 Hz, J2=1.0 Hz, 1H,
2-H), 8.63 (m, 3H, 7-,4-,5-H), 8.16 (m, 2H, 3-,6-H), 7.96
(m, 2H, 2%-,6%-H), 7.22 (m, 2H, 3%-,5%-H), 3.92 (s, 3H,
-OCH3). C20H13NO4S requires: C, 66.10; H, 3.61; N,
3.85. Found: C, 66.57; H, 3.78; N, 3.65%.
9. Saito, I.; Takayama, M.; Sugiyama, H.; Nakatani, K. J.
21. C6: mp 260ꢀ262°C. IR (KBr): 3030, 1780, 1610, 1600,
Am. Chem. Soc. 1995, 117, 6406–6407.
10. Theodorakis, E. A.; Xiang, X.; Blom, P. J. Chem. Soc.,
Chem. Commun. 1997, 1463–1464.
1
1580, 790, 710 cm−1; H NMR (CDCl3): l 8.65 (d, J=7.3
Hz, 2H, 2-, 7-H), 8.29 (m, 4H, 3-ꢀ6-H), 7.52–7.83 (m,
5H, Ph). EIMS (m/z, %): 317 (M+, 4.0), 105 (PhCO+,
100).
11. Aveline, B. M.; Kochevar, I. E.; Redmond, R. W. J. Am.
Chem. Soc. 1995, 117, 9699–9708.
.