N-propargyl-2-alkynylbenzothiazolium aza-enediynes: Role of the 2-alkynylbenzothiazolium functionality in DNA cleavage

Article abstract of DOI:10.1016/S0960-894X(01)00606-0

The 2-alkynylbenzothiazolium salts 3a-d incorporating an N-propargyl moiety have been prepared as aza-enediyne analogues. While these aza-enediynes are shown to be modest DNA cleavage agents, DNA cleavage was also observed with the N-methyl-2-alkynylbenzothiazolium salt 4, which lacks the aza-enediyne moiety. The structural requirements for DNA cleavage, and the correlation of DNA cleavage efficiency with the propensity of these compounds to undergo nucleophilic addition by methanol support a proposed DNA cleavage mechanism involving DNA alkylation by appropriate 2-alkynyl-substituted benzothiazolium salts.

Full text of DOI:10.1016/S0960-894X(01)00606-0

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2971–2974
N-Propargyl-2-alkynylbenzothiazolium Aza-enediynes: Role of the
2
-Alkynylbenzothiazolium Functionality in DNA Cleavage
Dalip Kumar, Wendi M. David and Sean M. Kerwin*
Division of Medicinal Chemistry, College of Pharmacy, University of Texas at Austin, Austin, TX 78712, USA
Received 10 October 2000; accepted 5 September 2001
Abstract—The 2-alkynylbenzothiazolium salts 3a–d incorporating an N-propargyl moiety have been prepared as aza-enediyne
analogues. While these aza-enediynes are shown to be modest DNA cleavage agents, DNA cleavage was also observed with the N-
methyl-2-alkynylbenzothiazolium salt 4, which lacks the aza-enediyne moiety. The structural requirements for DNA cleavage, and
the correlation of DNA cleavage efficiency with the propensity of these compounds to undergo nucleophilic addition by methanol
support a proposed DNA cleavage mechanism involving DNA alkylation by appropriate 2-alkynyl-substituted benzothiazolium
salts. # 2001 Elsevier Science Ltd. All rights reserved.
The design of new DNA cleavage agents has been an
area of great interest recently. This interest has been
fueled by the discovery of several potent antitumor,
These aza-enediynes may undergo isomerization to
generate an aza-enyne allene (Scheme 1, path a). The
aza-enyne allene may be capable of an aza-Myers type
cyclization, thereby generating a diradical persistent
enough to perform DNA cleavage chemistry under
physiological conditions. Alternatively, these aza-ene-
diynes may alkylate DNA via the addition of a DNA
nucleophile at the 2-alkyne position (Scheme 1, path b).
In the case of a DNA base serving as nucleophile, the
resulting DNA adduct might undergo spontaneous loss
of the base (e.g., depurination) and decomposition of
1
antibiotic natural products such as calichemicin, dyne-
2
3
micin, and neocarzinostatin, all of which cleave DNA
upon suitable triggering of core diradical-generating
structures. While these natural products have proven
too toxic for use as anticancer drugs, an antibody–cali-
cheamicin–analogue conjugate has recently been
approved for cancer therapy, demonstrating the feasi-
bility of using diradical-based DNA cleavage agents as
4
,5
10
warheads for selective cytotoxic agents.
There has
been much effort directed towards the design of syn-
the abasic site to yield a strand break. In preliminary
6
thetic enediyne analogues capable of cleaving DNA.
There are, however, very few examples of aza-sub-
stituted analogues. Several groups have investigated
the ability of aza-substituted enyne allenes to undergo
radical-generating cyclizations, but none have
7
8
demonstrated DNA cleavage activity.
We have been exploring the utilization of mild, physio-
logically relevant ‘triggering’ mechanisms and the effect
of heteroatom substitution in enediynes in the design of
novel DNA cleavage warheads with the goal of increas-
6
ing antitumor selectivity. We recently reported the
design of a series of N-propargyl-2-alkynyl heterocycles,
exemplified by the benzimidazolium compound 1
(
Scheme 1), which incorporate an aza-enediyne moiety.9
Scheme 1. Two possible mechanisms for DNA cleavage arising from
an N-propargyl-2-alkynyl heterocyclic system. Path a: isomerization,
followed by Myers cyclization to generate a diradical. Path b: nucleo-
philic attack.
*
2
Corresponding author. Tel.: +1-512-471-5074; fax: +1-512-232-
606; e-mail: skerwin@mail.utexas.edu
0
960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00606-0

2972
D. Kumar et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2971–2974
examination of the DNA cleavage chemistry of 1, evi-
dence for both diradical and electrophillic DNA clea-
vage pathways was found. Here, we report the
synthesis and DNA cleavage ability of a series of the
stable in methanolic solution. Similarly, the
benzothiazolium salts 5 and 6 are stable in methanol.
9
,11
Isolation of the product formed from methanolic solu-
tions of 3b and characterization by COSEY, NOSEY,
and HMQC NMR demonstrate that it is the methanol
addition product (Z)-2-(2-methoxy-2-p-tolylethylene)-3-
pro-2-ynyl benzothiazolium triflate (7, Scheme 3). The
benzimidazolium salt 1 was also observed to form a
structurally related methanol addition product; how-
ever, in the case of 1 the methanol addition product was
corresponding benzothiazolium aza-enediynes 3a–d
(
Scheme 2). The benzothiazolium compounds reported
here appear to cleave DNA by a pathway that is inde-
pendent of the ability to form diradical intermediates.
A series of N-propargyl-2-alkynyl substituted ben-
zothiazolium salts 3a–d in which the alkynyl substituent
was varied were prepared by alkylation of the corre-
sponding 2-alkynyl substituted benzothiazoles 2a–d,
which were prepared by palladium-catalyzed coupling
9
only formed in the presence of base. Apparently, the
benzothiazolium salts 3a–c are more susceptible to
nucleophilic attack at the 2-alkynyl substitutent than
the corresponding benzimidazolium salts, and as a
result would be expected to serve as pH-independent
electrophillic DNA alkylating/cleavage agents.
1
2
using the method of Schlegel and Maas (Scheme 2).
Alkylation of these benzothiazoles with propargyl tri-
flate1 afforded the desired benzothiazolium salts 3a–d.
3
14
Less reactive propargylating reagents did not afford
good yields of the desired salts. A model N-methyl ben-
zothiazolium salt was prepared by alkylation of 2-(2-
phenylethynyl)benzothiazole 2a with methyl triflate to
afford methyl triflate salt 4 in good yield. The 2-unsub-
stituted N-methyl benzothiazolium triflate 5 and the
corresponding N-propargyl salt 6 were also prepared by
the alkylation of benzothiazole with methyl triflate and
propargyl triflate, respectively.
We have studied the DNA cleavage ability of the ben-
zothiazolium salts 3a–d, 4, 5, and 6 by measuring the
amount of circular relaxed (Form II) and linear (Form
III) DNA produced when fixed concentrations (100 mM
and 1 mM) of these compounds were incubated with
supercoiled ÈX174 DNA in 20 mM TRIS buffer (pH 8
ꢀ
or 9) for 20 h at 37 C. In no case was any significant
amount of linear DNA produced, indicating that the
DNA cleavage reactions afforded only single-strand
DNA breaks (Fig. 1). The DNA cleavage products were
also formed from reactions run in the dark (data not
shown), precluding the role of photochemical reactions
in the DNA cleavage process.
Some of the benzothiazolium salts are unstable in sol-
vents like methanol and DMSO. In the case of metha-
nolic solutions of salts 3a and 3b, standing at room
temperature for 30 min results in complete dis-
appearance of the starting material along with the for-
mation of one major product. Benzothiazolium salts 3c
and 4 also undergo conversion in methanolic solution. In
contrast, the benzothiazolium 3d, in which the alkynyl
substituent is the sterically demanding TIPS group, is
Scheme 3. Facile addition of methanol to benzothiazolium salt 3b.
Figure 1. DNA cleavage by compounds 3a–d, 4, and 2a. Supercoiled
ÈX174 phage DNA (50 mM base pairs) was incubated with 100 mM of
compounds 4, 3a–d, and 2a or alone (C) (pH 8, 20 mM Tris buffer,
o
13% v/v DMSO, 20 h at 37 C) and analyzed by gel electrophoresis
(0.8% agarose, ethidium bromide stain).
Scheme 2. Synthesis of benzothiazolium salts.

D. Kumar et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2971–2974
2973
Table 1. Cleavage of supercoiled DNA by some benzothiazolium
salts
The DNA cleavage results for the 2-alkynylbenzothia-
zolium salts 3a–b and 4 may be explained by a cleavage
mechanism involving alkylation of DNA by the elec-
trophilic 2-alkynyl benzothiazolium substituent, fol-
lowed by DNA scission at the alkylated site (Scheme 1,
path b). Similar Michael-type addition of DNA bases to
electrophiles have been reported to provide labile DNA
adducts, particularly those involving the N7 and C8 site
Compd
Normalized percent DNA cleavagea
pH 8.0
pH 9.0
3
3
3
3
4
5
6
2
a
b
c
30ꢁ19 (6)
33ꢁ18 (6)
20ꢁ4 (6)
7ꢁ4 (6)
40ꢁ13 (4)
42ꢁ1 (2)
33ꢁ5 (2)
7ꢁ1 (2)
1
5
d
of purine bases. Support for this proposed mechanism
comes from DNA cleavage reactions performed on sin-
39ꢁ15 (6)
2.7ꢁ0.7 (2)
16ꢁ0.2 (2)
0ꢁ0 (2)
48ꢁ15 (2)
b
nd
nd
0
vage reactions reveal that the DNA cleavage due to 4 is
32
b
gly 5 -[ P]-labeled oligodeoxynucleotides. These clea-
a
0ꢁ0 (2)
1
1
specific for guanosine residues. The observation of
low-mobility products from both the oligodeoxynucleo-
tide and supercoiled DNA cleavage reactions is com-
mensurate with DNA alkylation or crosslinking by
these 2-alkynyl benzothiazolium salts. Additional sup-
port for this electrophilic DNA cleavage mechanism
comes from studies on the effect of inhibitors on the
DNA cleavage reaction by 4. Addition of 10 mM KBr
as a sacrificial nucleophile to DNA cleavage reactions
involving 4 results in complete inhibition of the DNA
cleavage (data not shown).
aNormalized percent cleavage of DNA from cleavage reactions (20 h)
containing 100 mM compound and 50 mM (bp) supercoiled DNA at
the indicated pH in 20 mM TRIS buffer. The values reported are
meanꢁstandard deviation for the number of separate determinations
indicated in parentheses.
b
Not determined.
Agarose gels of DNA cleavage reaction mixtures con-
taining the benzothiazolium salts 3a–c and 4 showed
evidence of low mobility DNA products that remained
in or near the loading wells of the gels (Fig. 1). These
low mobility products made quantification of the DNA
cleavage somewhat imprecise, as evidenced by the large
standard deviations reported in Table 1. In the case of
reaction mixtures containing 1 mM concentration of
these compounds, all of the DNA remained in the
loading wells. This abnormally low mobility of the DNA
was not prevented by chloroform/phenol extraction of
the reaction mixtures prior to gel electrophoresis.
For the benzothiazolium salts examined here that cleave
DNA but lack the 2-alkynyl substitutent (6), or have a
sterically and electronically blocked 2-alkynyl sub-
stituent (3c–d), a different DNA cleavage mechanism
must be operating. We propose that the DNA cleavage
observed with these compounds is due to the N-pro-
pargyl substituent. Benzothiazolium salt 5, which lacks
both the N-propargyl and 2-alkynyl substituents, does
not cleave DNA to any significant extent. However, the
simple N-propargyl benzothiazolium salt 6 cleaves
DNA about as well as compounds 3c–d, in which
nucleophilic attack at the 2-alkynyl substituent is both
sterically and electronically disfavored. The mechanism
of this alternative DNA cleavage reaction is currently
under investigation.
As shown in Table 1, aza-enediyne benzothiazolium
salts 3a–b cleave DNA to approximately the same
extent at pH 8. However, both the simple N-methyl-2-
alkynylbenzothiazolium salt
pargylbenzothiazolium salt 6 cleave DNA as well. Thus,
4
and the N-pro-
in contrast to the aza-enediyne benzimidazolium analo-
9
gues, the DNA cleavage by these aza-enediyne ben-
zothiazolium salts does not require the presence of the
aza-enediyne functionality. Neither the N-methyl-2-
unsubstituted benzothiazolium salt 5 nor the aza-ene-
diyne benzothiazole 2a cleave DNA to any appreciable
extent. Based on these results, it appears that either 2-
alkynyl substituted benzothiazolium salts (e.g., 4) or N-
propargyl benzothiazolium salts (e.g., 6) can give rise to
DNA cleavage. In the aza-enediyne benzothiazolium
salts containing both of these functionalities, the
importance of the 2-alkynyl functionality is demon-
strated by the manner in which the 2-alkynyl substituent
affects the DNA cleavage potential. The aryl-substituted
compounds 3a–b cleave DNA much better than the
silyl-substituted compounds 3c–d, which cleave DNA
with about the same efficiency as the N-propargyl-2-
unsubstituted derivative 6.
We have prepared a series of N-propargyl-2-alkynyl
substituted benzothiazolium aza-enediynes. These com-
pounds are moderately potent DNA cleavage agents.
We present a DNA cleavage structure–activity relation-
ship that demonstrates the aza-enediyne functionality is
not required for DNA cleavage activity. We find that
these 2-alkynyl substituted benzothiazolium salts
undergo facile nucleophilic addition of methanol under
neutral conditions. These results support a mechanism
for DNA cleavage involving DNA alkylation by the
electrophilic 2-alkynyl substituent. Thus, simple 2-alky-
nyl-substituted benzothiazolium salts may serve as a
new class of DNA-directed electrophiles. We are cur-
rently in the process of further defining the mechanism
of covalent DNA modification and evaluating the bio-
logical activity of these novel DNA cleavage agents.
The difference in DNA cleavage ability of the ben-
zothiaolium salts 3a–d and 4 at pH 8 and 9 was not
statistically significant, indicating that these com-
pounds, unlike the benzimidazolium salt 1, do not
undergo base-promoted isomerization to aza-enyne
allene species in order to cleave DNA.
Acknowledgements
This research was funded by the Foundation for
Research. We gratefully acknowledge the technical
assistance of Stephanie Jones.

2974
D. Kumar et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2971–2974
References and Notes
76.32, 78.11, 117.70, 118.08, 119.27, 124.61, 129.70, 130.11,
30.58, 131.61, 133.88, 139.51, 149.45, 153.76. Anal.
1
1
(
3
C
19
H
12
F
H, CH ), 2.80 (t, 1H, CH, J=2.5 Hz, acetylenic proton), 5.77
3
NO
3 2 3
S ) C, H, N, S. 3b: H NMR (CDCl ) d 2.50 (s,
1. Lee, M. D.; Dunne, T. S.; Chang, C. C.; Ellestad, G. A.;
Siegel, M. M.; Morton, G. O.; McGahren, W. J.; Borders,
D. B. J. Am. Chem. Soc. 1987, 109, 3464.
2. Konishi, M.; Ohkuma, H.; Matsumoto, K.; Tsuno, T.;
Kamei, H.; Miyaki, T.; Oki, T.; Kawaguchi, H.; VanDuyne,
G. D.; Clardy, J. J. Antibiot. 1989, 42, 1449.
. (a) Ishida, N.; Miyazaki, K.; Kumagai, K.; Rikimaru, M. J.
Antibiot. 1965, 18, 68. (b) Edo, K.; Mizugaki, M.; Koide, Y.;
Seto, H.; Furihata, K.; Otake, N.; Ishida, N. Tetrahedron Lett.
985, 26, 331.
. Dowell, J. A.; King, S. P.; Lui, H.; Berger, M. S.; Korth-
Bradley, J. M. Clin. Pharm. Ther. 2000, 67, PII92.
. Naito, K.; Takeshita, A.; Shigeno, K.; Nakamura, S.;
Fujisawa, S.; Shinjo, K.; Yoshida, H.; Ohnosho, K.; Mori, M.;
3
(
d, 2H, J=2.5 Hz, CH ), 7.40 (d, 2H, J=7.9 Hz, C –H and
0
2
3
C –H), 7.77 (d, 2H, C –H and C –H J=8.2 Hz), 7.80–8.01
0
0
0
6
5
2
,
(
5 6
m, 2H, C –H and C –H), 8.02–8.31 (dd, 2H, J=8.2, 8.6 Hz,
13
4 7 3
C –H and C –H); C NMR (CDCl ) d 22.24, 41.48, 73.20,
7
1
6.49, 78.02, 114.90, 117.47, 120.24, 124.70, 129.97, 130.32,
30.47, 131.44, 133.86, 139.35, 145.69, 153.78. Anal.
3
1
(
(
C
20
H
14
F
s, 9H, TMS), 2.64 (t, 1H, J=2.5 Hz, acetylenic proton), 5.71
3
NO
3
S
2
) C, H, N, S. 3c: H NMR (CDCl
3
) d 0.413
1
4
(
d, 2H, J=2.3 Hz, CH ), 7.78–7.89 (m, 2H, C –H and C –H),
2
5
6
7
8
.18 (d, 1H, J=8.4 Hz, C
4
–H), 8.27 (d, 1H, J=8.2 Hz, C
–H);
13
C NMR (CDCl
24.67, 129.50, 129.65, 130.16, 131.16, 138.84, 152.37. 3d: H
NMR (CD Cl ) d 1.02–1.49 (m, 21H, TIPS), 2.76 (t, 1H,
3
) d ꢂ1.23, 41.26, 72.85, 77.40, 87.99, 117.49,
5
1
1
2
2
Terakawa, S.; Ohno, R. Leukemia 2000, 14, 1436.
6
J=2.5 Hz), 5.74 (d, 2H, J=2.5 Hz, CH
–H and C –H), 8.39 (dd, 2H, J=8.2, 8.6 Hz, C
H); C NMR (CD Cl ) d 11.42, 18.63, 41.63, 72.86, 78.00,
2
), 7.90–8.05 (m, 2H,
. For recent references see: Basak, A.; Bdour, H. M.; Shain,
C
5
6
4
–H and C –
7
J. C.; Mandal, S.; Rudra, K. R.; Nag, S. Bioorg. Med. Chem.
Lett. 2000, 10, 1321. Hay, M. P.; Wilson, W. R.; Denny, W. A.
Bioorg. Med. Chem. Lett. 1999, 9, 3417.
. David, W. M.; Kerwin, S. M. J. Am. Chem. Soc. 1997, 119,
464.
. Shi, C. S.; Zhang, Q.; Wang, K. K. J. Org. Chem. 1999, 64,
25. Shi, C. S.; Wang, K. K. J. Org. Chem. 1998, 63, 4413.
Wu, M.-J.; Lin, C.-F.; Chen, S.-H.; Lee, F.-C. J. Chem. Soc.,
Perkin Trans. 1 1999, 20, 2875.
13
2
2
9
0.37, 117.79, 124.84, 129.21, 129.91, 130.84, 131.83,
1
1
NMR (CDCl
39.29,152.60. Anal. (C22
+DMSO-d
H
28
F
3
NO
3
S
2
Si) C, H, N, S. 4:
H
), 7.21–7.69 (m,
7
1
8
9
3
6
) d 4.24 (s, 3H, CH
3
7
H); C NMR (CDCl
H, Ar–H), 7.82–7.95 (m, 1H, C –H), 8.04–8.18 (m, 1H, C –
4
7
13
3 6
+DMSO-d ) d 37.26, 75.02, 116.08,
1
16.66, 123.61, 128.16, 128.73 (2C), 129.34, 129.82, 132.15
2C), 139.25, 151.82. Anal. (C17 NO ) C, H, N, S. 7:
H NMR (CD Cl ) d 2.49 (s, 3H, CH ), 2.65 (t, 1H, J=2.6,
(
H
12
F
3
3 2
S
1
2
2
3
9
. David, W. M.; Kumar, D.; Kerwin, S. M. Bioorg. Med.
Chem. Lett. 2000, 10, 2509.
0. Kushida, T.; Uesugi, M.; Sugiura, Y.; Kigoshi, H.;
3
acetylenic proton), 4.71 (s, 3H, OCH ), 5.50 (d, 2H, J=2.6
Hz), 6.69 (s, 1H, olefinic proton), 7.43 (d, 2H, J=8.4 Hz, C 0
–
H and C –H), 7.60 (d, 2H, J=8.2 Hz, C –H and C –H),
3
1
0
0
5
2
6
Tanaka, H.; Hirokawa, J.; Ojika, M.; Yamada, K. J. Am.
Chem. Soc. 1994, 116, 479.
11. Tuntiwechapikul, W.; David, W. M.; Kumar, D.; Kerwin,
S. M.; Salazar, M. Submitted for publication.
7
1
.68–7.71 (m, 1H, C –H), 7.80–7.84 (m, 1H, C –H), 7.95 (dd,
5
6
H, J=1.8, 8.6 Hz, C
13
7 4
–H), 8.10 (dd, 1H, J=1.7, 8.6 Hz, C –
2 2
H); C NMR (CD Cl ) d 21.80, 38.55, 61.26, 73.87, 76.83,
9
1
7.23, 115.33, 123.82, 128.31, 128.42, 128.59, 129.11, 130.10,
30.44, 139.20, 144.49, 166.92, 177.62. HRMS (CI) calcd for
1
1
1
2
2. Schlegel, J.; Maas, G. Synthesis 1999, 100.
3. Vedejs, E.; Mullins, M. J. J. Org. Chem. 1977, 42, 3109.
+
20
C H18NOS (M ) m/z 320.1118, found 320.1109.
1
4. Selected data for compounds: 3a: H NMR (CD Cl ) d
2
2
15. (a) Gates, K. G. In Comprehensive Natural Products
Chemistry; Barton, D., Nakanishi, K., Meth-Cohn, O., Eds.;
Pergamon: Oxford, 1999; Vol 7, p 491. (b) Neagu, I.; Koi-
visto, P.; Neagu, C.; Kostianen, R.; Stenby, K.; Peltonen, K.
Carcinogenesis 1995, 16, 1809. (c) Mueller, N.; Eisenbrand, G.
Chem-Biol. Interact. 1985, 53, 173.
.75 (t, 1H, J=2.6 Hz, acetylenic proton), 5.78 (d, 2H, J=2.6
), 7.56–7.59 (m, 2H, C –H and C –H), 7.69 (dd, 1H,
J=1.3, 8.1 Hz, C –H), 7.85–7.92 (m, 3H, C –H, C –H and
–H), 7.99 (dd, 1H, J=1.2, 8.4 Hz, C –H), 8.25 (dd, 2H, J=7.9,
Hz, CH
2
3
0
0
5
0
0
0
6
4
2
C7
0
7
1
3
8
5 6 2 2
.4 Hz, C –H and C –H); C NMR (CD Cl ) d 41.74, 73.18,