Thiol-dependent DNA cleavage by 3H-1,2-benzodithiol-3-one 1,1-dioxide

Article abstract of DOI:10.1016/S0960-894X(00)00125-6

Hydrodisulfides (RSSH) have previously been implicated as key intermediates in thiol-triggered oxidative DNA damage by the antitumor agent leinamycin. In an effort to better understand DNA damage by RSSH and to expand on the number and type of chemical systems that produce this reactive intermediate, the ability of 3H-1,2-benzodithiol-3-one 1,1-dioxide (11) to serve as a thio-dependent DNA cleaving agent has been investigated. The findings reported here indicate that reaction of 11 with thiols results in release of RSSH and subsequent oxidative DNA strand cleavage. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00125-6

Bioorganic & Medicinal Chemistry Letters 10 (2000) 885±889
Thiol-Dependent DNA Cleavage by 3H-1,2-Benzodithiol-3-one
1,1-dioxide
Leonid Breydo and Kent S. Gates a,b,*
a
aDepartment of Chemistry, University of Missouri-Columbia, Columbia, MO 65211, USA
bDepartment of Biochemistry, University of Missouri-Columbia, Columbia, MO 65211, USA
Received 10 December 1999; accepted 14 February 2000
AbstractÐHydrodisul®des (RSSH) have previously been implicated as key intermediates in thiol-triggered oxidative DNA damage
by the antitumor agent leinamycin. In an e€ort to better understand DNA damage by RSSH and to expand on the number and
type of chemical systems that produce this reactive intermediate, the ability of 3H-1,2-benzodithiol-3-one 1,1-dioxide (11) to serve
as a thiol-dependent DNA cleaving agent has been investigated. The ®ndings reported here indicate that reaction of 11 with thiols
results in release of RSSH and subsequent oxidative DNA strand cleavage. # 2000 Elsevier Science Ltd. All rights reserved.
Reaction of thiols with the antibiotic leinamycin (1)
triggers oxidative DNA cleavage and DNA alkylation
gical activities of the antitumor agent leinamycin and
7
various cytotoxic polysul®des such as varacin (9) and
8
via two chemically distinct pathways (Scheme 1).1
Thiol-triggered oxidative DNA cleavage by leinamycin
�4
lissoclinotoxin A, and bis(2-hydroxyethyl)trisul®de (10).
9
and by synthetic leinamycin analogues such as 5, 6, and
7
In an e€ort to better understand the DNA-cleaving
properties of hydrodisul®des and to expand on the
number and type of chemical systems capable of gen-
erating this reactive species, we have examined the abil-
ity of 3H-1,2-benzodithiol-3-one 1,1-dioxide (11) to
serve as a thiol-activated DNA-cleaving agent. We
anticipated that attack of thiol on this sulfur heterocycle
would result in the release of RSSH by the mechanism
shown in Scheme 2. Although the reaction of thiols with
occurs by a general mechanism involving the conversion
of molecular oxygen to DNA-cleaving oxygen radicals via
superoxide, hydrogen peroxide, and a trace metal-catalyzed
2
,4
Fenton reaction.
generated in this process by O -mediated degradation of
Superoxide radical is presumably
2
unstable hydrodisul®de intermediates (RSSH, 3;
Scheme 1) that are produced during the reaction of the
4,5
1,2-dithiolan-3-one 1-oxide heterocycle with thiols.
1
1 has not previously been reported, the mechanism
Hydropolysul®des (RS SH) are also suspected as key
x
proposed in Scheme 2 is directly analogous to that elu-
cidated for the reaction of phosphites with this sulfur-
10
transfer agent.
intermediates in thiol-triggered oxidative DNA cleavage
6
by polysul®des such as 7-methylbenzopentathiepin (8)
4
and bis(2-hydroxyethyl)trisul®de (10). DNA damage and/
or general oxidative stress caused by the production of
oxygen radicals in the decomposition of hydrodisul®de
intermediates may play an important role in the biolo-
We ®nd that, in the presence of excess thiol, micromolar
concentrations of 11 eciently cause single-strand
breaks in duplex DNA, as measured by the conversion
*
0
Corresponding author. Tel.: +1-573-882-6763; fax: +1-573-882-2754; e-mail: gates@missouri.edu
960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00125-6

8
86
L. Breydo, K. S. Gates / Bioorg. Med. Chem. Lett. 10 (2000) 885±889
Scheme 1. Mechanisms of DNA damage by leinamycin.
Scheme 2. Proposed reaction of thiols with 3H-1,2-benzodithiol-3-one 1,1-dioxide (11).
of supercoiled (Form I) plasmid DNA to the open cir-
cular form (Form II) (Fig. 1). In accord with the
mechanism proposed in Scheme 2, ecient DNA clea-
vage by 11 is completely dependent upon added thiol
used11 oxygen radical scavengers such as methanol and
mannitol signi®cantly inhibit thiol-dependent DNA
cleavage by 11. Chelating agents such as diethylene-
triaminepentaacetic acid (DETAPAC) inhibit strand
scission in this system, presumably by sequestering
(
Fig. 2). As expected for the involvement of a hydro-
1
1,12
disul®de intermediate (3), mechanistic experiments indi-
cate that 11, in conjunction with added thiols, mediates
the conversion of molecular oxygen to DNA-cleaving
oxygen radicals (Fig. 3). The involvement of reduced
adventitious trace metals in a non-redox-active form
that cannot participate in the Fenton reaction. Addition
of the hydrogen peroxide-destroying enzyme catalase to
the assay mixture almost completely inhibits DNA
cleavage. Addition of superoxide dismutase (SOD)
increases the eciency of strand scission by 11. The
marked e€ect of SOD is consistent with the presence of
superoxide radical in these reactions. SOD catalyzes the
disproportionation of superoxide to hydrogen peroxide
and O2,13 and the increase in cleavage eciency can be
rationalized by the fact that this enzyme facilitates the
ꢀ
�
ꢀ
oxygen species (O2 !H O !HO ) and a trace metal-
2
2
catalyzed, Fenton-type conversion of hydrogen peroxide
to DNA-cleaving radicals in thiol-mediated DNA clea-
vage by 11 is supported by several ®ndings. Commonly
Figure 1. Thiol-dependent DNA cleavage by various concentrations
of 11. Supercoiled pBR322 DNA (38 mM bp) was incubated for 6 h at
ꢁ
2
4 C with various concentrations of 11 and 5 equiv of 2-mercapto-
ethanol in sodium phosphate bu€er (50 mM, pH 7.0), followed by
agarose gel electrophoretic analysis as previously described.2 Lane 1,
DNA alone; lane 2, 11 (1 mM) (0.02‹0.01); lane 3, thiol (5 mM)
(
0.06‹0.01); lane 4, 11 (10 mM)+thiol (0.13‹0.02); lane 5, 11 (20
mM)+thiol (0.26‹0.05); lane 6, 11 (50 mM)+thiol (0.63‹0.11); lane
(100 mM)+thiol (1.53‹0.64); lane 8, (250 mM)+thiol
7
,
1
1
(
(
2.77‹0.69); lane 9, 11 (500 mM)+thiol (3.08‹0.61); lane 10, 11
1 mM)+thiol (3.47‹0.78). The value in parentheses following each
Figure 2. E€ect of thiol concentration on DNA cleavage by 11.
ꢁ
lane description represents the mean number of strand breaks per
where f
Supercoiled pBR322 DNA (38 mM bp) was incubated for 6 h at 24 C
plasmid molecule (S) calculated using the equation S=� ln f
I
I
with 11 (100 mM) in sodium phosphate bu€er (50 mM, pH 7.0) with
various concentrations of 2-mercaptoethanol, followed by agarose gel
electrophoretic analysis as previously described.2
15
is the fraction of plasmid present as form I. The values are corrected
for background nicks present in control DNA.

L. Breydo, K. S. Gates / Bioorg. Med. Chem. Lett. 10 (2000) 885±889
887
ꢁ
Figure 3. E€ect of various additives on thiol-activated DNA cleavage by 11. Supercoiled pBR322 DNA (38 mM bp) was incubated for 6 h at 24 C
with 11 (100 mM) and 2-mercaptoethanol (500 mM) in sodium phosphate bu€er (50 mM, pH 7.0) in the presence of various additives, followed by
agarose gel electrophoretic analysis as previously described.2
production of hydrogen peroxide.14 The observation
that >2 equiv of thiol are required for signi®cant DNA
cleavage (Fig. 2) suggests that secondary reactions of
thiol with 3 may be important.
exchange23 reactions with excess 2-mercaptoethanol in
the mixture eciently convert them to thiosalicylic acid
18. Importantly, the production of polysul®des (14,
X=1,2) in this reaction is indicative of the formation of
a hydrodisul®de intermediate (3). Hydrodisul®des read-
2
4,25
In order to characterize the chemical events underlying
thiol-dependent oxidative DNA cleavage by 11, we
examined the products and intermediates resulting from
reaction of this sulfur heterocycle with thiols. The reaction
of 11 (40 mM) with excess thiol (10 equiv of 2-mer-
captoethanol) in a 4:1 water:acetonitrile mixture is rapid
ily decompose to yield mixtures of polysul®des.
Finally, in support of the proposed mechanism (Scheme
2), under conditions designed to favor its detection (0.5
equiv n-propanethiol, 0.5 equiv triethylamine, in dry
CH Cl ), we have directly observed the unstable 1,2-
2
2
benzoxathiolan-3-one 1-oxide (13) as the major product
26
stemming from the reaction of 11 with thiol.
(
complete in seconds).16 The resulting mixture of pro-
ducts was separated and characterized using standard
methods, and the identity of the products was con®rmed
by comparison of NMR spectra with that of authentic
samples prepared by independent routes. The aromatic
carboxylic acids 16, 17, and 18 were isolated and char-
acterized as their methyl ester derivatives following work
This work establishes the 3H-1,2-benzodithiol-3-one
1,1-dioxide heterocycle as a new type of thiol-activated
DNA-cleaving agent. Examination of products and
intermediates resulting from the reaction of 3H-1,2-
benzodithiol-3-one 1,1-dioxide with thiol supports the
notion that a DNA-cleaving hydrodisul®de species
up of the reaction mixture with diazomethane.1
7�21
(
tem. The eciency and mechanism of thiol-triggered
DNA cleavage by 11 and 5 are very similar. This
RSSH) is released by attack of thiol on this ring sys-
The products isolated from the reaction of 11 with
thiols are consistent with those expected from the
breakdown of the hydrodisul®de and 1,2-benzox-
athiolan-3-one 1-oxide intermediates (3 and 13) shown
in Scheme 2. Initial attack of thiol on the 1,2-benzox-
athiolan-3-one 1-oxide intermediate 13 is expected to
yield the thiosul®nate 15. In accord with literature pre-
2
observation is meaningful because RSSH is an inter-
mediate generated by the attack of thiol on both of these
heterocyclic systems and, therefore, thiol-triggered oxi-
dative DNA damage by 11 and 5 is most reasonably
attributed to this common reactive intermediate. Fur-
ther studies designed to characterize the mechanism(s)
by which hydrodisul®des generate oxygen radicals are
currently underway in this laboratory.
2
2
cedent, this thiosul®nate reacts further with thiol to
a€ord a mixture of disul®des 16 and 17. The disul®des
16 and 17 are isolated in low yield because thiol-disul®de
Scheme 3. Products resulting from attack of thiol on 3H-1,2-benzodithiol-3-one 1,1-dioxide (11).

8
88
L. Breydo, K. S. Gates / Bioorg. Med. Chem. Lett. 10 (2000) 885±889
Previous work has shown that 5 serves as an excellent
model for the chemical reactivity of the 1,2-dithiolan-3-
13. Fridovich, I. Acc. Chem. Res. 1972, 5, 321.
14. In systems where addition of SOD inhibits DNA damage it
is likely due to the fact that superoxide radical serves as a
necessary reducing agent for trace metals involved in the Fen-
ton reaction. In systems where other reducing agents (e.g.,
thiols) are present, addition of SOD can actually increase
DNA-damage eciency. See, for example: Eliot, H.; Gianni,
L.; Myers, C. Biochemistry 1984, 23, 928. Parraga, A.; Orozco,
M.; Portugal, J. Eur. J. Biochem. 1992, 208, 227. Weglarz, L.;
Bartosz, G. Int. J. Biochem. 1991, 23, 663. Nagai, K.; Hecht,
S. M. J. Biol. Chem. 1991, 266, 23994.
one 1-oxide heterocycle found in leinamycin (1).1
,3,4
Thus, our studies with 11 suggest that a 1,1-dioxide
analogue of leinamycin (19) would retain the ability to
produce RSSH upon reaction with thiols. On the other
hand, such a leinamycin analogue would likely be
devoid of DNA-alkylating properties because the elec-
trophilic oxathiolanone intermediate (2, Scheme 1)
required for the generation of the leinamycin-derived,
DNA-alkylating episulfonium ion (4) is not produced
by attack of thiol on the 1,1-dioxide system. Rather,
attack of thiol on the 1,1-dioxide system produces a 1,2-
oxathiolan-3-one 1-oxide (e.g., 13) which, in the context
of leinamycin, is not expected to lead to formation of a
15. Hertzberg, R. P.; Dervan, P. B. Biochemistry 1984, 23,
3934.
1
6. Compound 11 undergoes hydrolysis slowly compared to
thiolysis. In the DNA-cleavage reactions reported here, 11 was
the ®nal component added to thiol-containing reaction mix-
tures to ensure that thiol-triggered chemistry predominates.
2
7
DNA-alkylating electrophile. Consequently, the 1,1-
dioxide analogue of leinamycin 19 could serve as a
useful research tool to examine the biological e€ects
stemming from hydrodisul®de production by leinamy-
cin, in the absence of DNA alkylation by the natural
product.
1
7. In a typical reaction, 2-mercaptoethanol (1.4 mL, 20
10
mmol) in water (40 mL) was added to a stirred solution of 11
400 mg, 2 mmol) in acetonitrile (10 mL) to produce a cloudy
(
mixture. Under these conditions, all 11 is consumed (TLC)
within 1 min. After 10 min, 2 M HCl (2 mL, 4 mmol) was
added, and resulting suspension was extracted with ether
(
5 Â 50 mL). The combined organic layers were dried (Na SO )
2
4
and concentrated under reduced pressure to yield a white
solid. To this solid was added diazomethane (50 mL of 0.3 M
solution in ether), and the resulting yellow solution stirred for
ꢁ
5
min at 24 C, concentrated under reduced pressure, and
subjected to column chromatography (12:1 hexane:EtOAc!
1
(
00% EtOAc) to yield 2-(methylthio)carboxymethylbenzene
20, the dimethyl derivative of 18; 320 mg, 88%), dimethyl
0
2
2
,2 -dithiodibenzoate (21, the methyl ester of 17; 11 mg, 3%),
0
-(2 -hydroxyethyldithio)carboxymethylbenzene (22, the methyl
ester of 16; 32 mg, 7%) and 2-(hydroxyethyl)di-, tri-, and
tetrasul®des (14 where n=0, 1 and 2; as 90:9:1 mixture, 558
mg, 3.6 mmol). A trace amount (2 mg, 0.5%) of the dimethyl
ester of o-carboxybenzenesul®nic acid, resulting either from
direct hydrolysis of 11 or hydrolysis of intermediates 13 or 15,
was also detected. 2-(Methylthio)carboxymethylbenzene (20):
Acknowledgements
We thank the National Institutes of Health (GM51565)
for support of this work. The NMR facilities at the
University of Missouri-Columbia are partially sup-
ported by grants from the National Science Foundation
ꢁ
white solid (320 mg, 88% from 11), mp 62±64 C (lit: 63±
ꢁ
18 1
19
6
1
5 C); H NMR (250 MHz, CDCl
H), 7.45 (t, J=7.5 Hz, 1H), 7.25 (d, J=7.5 Hz, 1H), 7.13 (t,
3
): d 7.98 (d, J=7.5 Hz,
(
Grants 9221835 and 8908304).
13
J=7.5 Hz, 1H), 3.89 (s, 3H), 2.43 (s, 3H); C NMR (62.5
MHz, CDCl ): d 166.6, 143.1, 132.3, 131.1, 126.5, 124.1, 123.2,
1.8, 15.3; HRMS (EI) m/z calcd for C S 182.0418,
3
References and Notes
. Behroozi, S. B.; Kim, W.; Gates, K. S. J. Org. Chem. 1995,
0, 3964.
5
9 10 2
H O
found 182.0401. The identity of 20 was further con®rmed by
comparison (NMR, TLC) to an authentic sample prepared via
reaction of 2-mercaptobenzoic acid (Aldrich) with excess dia-
1
6
2
0
. Behroozi, S. J.; Kim, W.; Dannaldson, J.; Gates, K. S. Bio-
zomethane (CAUTION: explosion hazard!). Dimethyl 2,2 -
dithiodibenzoate (21): Initially isolated as a mixture contain-
chemistry 1996, 35, 1768.
. Asai, A.; Hara, M.; Kakita, S.; Kanda, Y.; Yoshida, M.;
Saito, H.; Saitoh, Y. J. Am. Chem. Soc. 1996, 118, 6802.
. Mitra, K.; Kim, W.; Daniels, J. S.; Gates, K. S. J. Am.
Chem. Soc. 1997, 119, 11691.
. Mitra, K.; Gates, K. S. Recent Res. Devel. Org. Chem. 1999,
, 311.
. Chatterji, T.; Gates, K. S. Bioorg. Med. Chem. Lett. 1998, 8,
35.
. Davidson, B. S.; Molinski, T. F.; Barrows, L. R.; Ireland,
0
3
ingꢂ10% dimethyl 2,2 -tri and tetrathiodibenzoate. An analy-
tical sample of 21 was prepared by treating the mixture with
triphenylphosphine to decompose the polysul®de impurities.
This a€orded 21 as white crystals (11 mg, 3% from 11), mp
4
ꢁ
ꢁ
20
1
21
5
3
6
5
7
128±129 C (lit: 131.5±133.5 C);
CDCl ): d 8.06 (dd, J=7.5, 1 Hz, 2H), 7.76 (dd, J=7.5, 1 Hz,
2H), 7.41 (td, J=7.5, 1 Hz, 2H), 7.23 (td, J=7.5, 1 Hz, 2H),
3.99 (s, 6H); 13C NMR (62.5 MHz, CDCl
): d 166.9, 140.3,
133.0, 131.4, 127.3, 125.8, 125.4, 52.4; HRMS (EI) m/z calcd
for C16 334.0333, found 334.0328. The identity of 21
H NMR (250 MHz,
3
3
C. M. J. Am. Chem. Soc. 1991, 113, 4709.
14 4 2
H O S
8
9
. Searle, P. A.; Molinski, T. F. J. Org. Chem. 1994, 59, 6600.
. Kohama, Y.; Iida, K.; Semba, T.; Mimura, T.; Inada, A.;
was further con®rmed by comparison (NMR, TLC) with an
0
authentic sample prepared by esteri®cation of 2,2 -dithiodi-
0
Tanaka, K.; Nakanishi, T. Chem. Pharm. Bull. 1992, 40, 2210.
0. Iyer, R. P.; Phillips, L. R.; Egan, W.; Regan, J. B.;
Beaucage, S. L. J. Org. Chem. 1990, 55, 4693.
benzoic acid (Aldrich) with diazomethane. 2-(2 -Hydroxy-
ethyldithio)carboxymethylbenzene (22): clear oil (32 mg, 7%
1
1
from 11): H NMR (250 MHz, CDCl ): d 8.20 (d, J=7.5 Hz,
3
1
1
1
1. Halliwell, B.; Gutteridge, J. M. C. Methods Enzymol. 1990,
86, 1.
2. Graf, E.; Mahoney, J. R.; Bryant, R. G.; Eaton, J. W. J.
1H), 8.02 (dd, J=7.5 Hz, 1 Hz, 1H), 7.58 (td, J=7.5 Hz, 1 Hz,
1H), 7.26 (td, J=7.5 Hz, 1 Hz, 1H), 3.94 (s, 3H), 3.86 (m, 2H),
2.88 (t, J=6 Hz, 2H); 13C NMR (62.5 MHz, CDCl
): d 166.7,
141.3, 132.8, 131.4, 127.1, 125.6, 125.2, 60.2, 52.2, 40.5;
3
Biol. Chem. 1984, 259, 3620.

L. Breydo, K. S. Gates / Bioorg. Med. Chem. Lett. 10 (2000) 885±889
889
HRMS m/z calcd for C10
H
12
O
Bis(2-hydroxyethyl)di-,tri- and tetrasul®des (14): As a clear oil.
3
S
2
244.0228, found 244.0232.
mL, 0.25 mmol) and 1-propanethiol (23 mL, 0.25 mmol). After
10 min, the reaction mixture was concentrated under reduced
1
Analytical data for 14 ( H NMR and HPLC retention time)
,9
pressure and redissolved in CDCl
3
. Analysis of the reaction
mixture by C NMR revealed that 1,2-benzoxathiolan-3-one
1-oxide (13) was the major product of the reaction. The 13
4
13
matches that reported previously. Obtained as a mixture
where n=0, 1, and 2 (90:9:1, 558 mg, 3.6 mmol).
C
1
0
1
8. Lu
tus Liebig Ann. Chem. 1992, 1977
9. Ozaki, S.; Yang, H.-J.; Matsui, T.; Goto, Y.; Watanabe,
Y. Tetrahedron: Asymmetry 1999, 10, 183.
0. Overman, L. E.; Matzinger, D.; O'Connor, E. M.; Overman,
J. D. J. Am. Chem. Soc. 1974, 96, 6081.
È
dersdorf, R.; Martens, J.; Pakzad, B.; Praefcke, K. Jus-
NMR spectrum of 13 in CDCl has previously been reported.
3
27. Literature precedent suggests that a leinamycin-derived
1,2-oxathiolan-3-one 1-oxide (analogous to 13) will not react
with the C6-C7 alkene of the natural product. For example,
other activated sul®nic acid derivatives, such as sul®nyl chlor-
ides, react with alkenes but Lewis acid catalysis and relatively
harsh conditions are required. Alternatively, thiol-triggered
DNA alkylation by 19 could, in principle, occur via attack of
thiol on a leinamycin-derived intermediate analogous to 15 to
yield a carboxy-sulfenic acid which might cyclize to the acti-
vated form of the antibiotic (2). However, unless the cycliza-
tion reaction is very rapid, the sulfenic acid intermediate
would decompose by reaction with excess thiol (to a€ord a
mixed disul®de analogueueous to 16),22 and would not yield
`activated' leinamycin (2).
1
2
2
5
2
2
2
2
3
2
1. Wentrup, C.; Bender, H.; Gross, G. J. Org. Chem. 1987,
2, 3838.
2. Kice, J. L.; Rogers, T. E. J. Am. Chem. Soc. 1974, 96, 8015.
3. Lees, W. J.; Whitesides, G. M. J. Org. Chem. 1993, 58, 642.
4. Derbesy, G.; Harpp, D. N. Sulfur Lett. 1992, 14, 199.
5. Kawamura, S.; Abe, Y.; Tsurugi, J. J. Org. Chem. 1969,
4, 3633.
6. To a stirred solution of 11 (100 mg, 0.5 mmol) in freshly
distilled dichloromethane (15 mL) was added triethylamine (35