Synthesis and reactivity of azobenzene-based bispropargyl sulfones: Interesting comparison between cyclic and acyclic systems

Article abstract of DOI:10.1016/j.bmcl.2007.06.001

Azobenzene-based bispropargyl bissulfone 3 containing stable E-azo moiety has been synthesized. Upon irradiation with long wavelength UV it isomerized to the Z-form 4, which can be thermally reisomerized to the E-isomer. Reactivity towards isomerization t

Full text of DOI:10.1016/j.bmcl.2007.06.001

Bioorganic & Medicinal Chemistry Letters 17 (2007) 4514–4517
Synthesis and reactivity of azobenzene-based bispropargyl
sulfones: Interesting comparison between cyclic and acyclic systems
Debarati Mitra, Moumita Kar, Rhitankar Pal and Amit Basak*
Department of Chemistry, Indian Institute of Technology, Kharagpur 721 302, India
Received 17 February 2007; revised 14 May 2007; accepted 1 June 2007
Available online 6 June 2007
Abstract—Azobenzene-based bispropargyl bissulfone 3 containing stable E-azo moiety has been synthesized. Upon irradiation with
long wavelength UV it isomerized to the Z-form 4, which can be thermally reisomerized to the E-isomer. Reactivity towards isom-
erization to the allenic system as well as DNA-cleaving efficiency under basic conditions was found to be significantly lower as com-
pared to the previously synthesized cyclic sulfones 1 and 2. This lowering of reactivity can be explained in terms of low conversion to
the allenic form and hence the lower extent of alkylation of DNA-bases, the only possible DNA-cleavage pathway for 3 and 4.
ꢀ 2007 Elsevier Ltd. All rights reserved.
An important aspect of development of anticancer drugs
that target DNA and bring about its destruction is to
enhance the efficiency of cleavage of the genetic mate-
rial.1 There can be various mechanisms by which
DNA can be cleaved; amongst these the radical medi-
ated H-abstraction from the sugar moiety followed by
oxidative cleavage2 as well as alkylation of DNA bases3
cover a wide range of molecules. Recently,4 we have
developed a novel photochemical way to enhance the
DNA-cleaving efficiency of bispropargyl cyclic sulfones.
The process involved photoconversion of the thermally
stable E-form 1 into the Z-form 2, which by virtue of
higher chemical reactivity towards isomerization into
bisallenic sulfone brings about greater extent of DNA-
cleavage (Fig. 1). Interestingly, bisallenic sulfone5 offers
two pathways for DNA-cleavage namely, the Garratt–
Braverman rearrangement6 as well as alkylation (Mi-
chael pathway).7
N N
N N
O
O
light
heat
O
O
S
O
O
S
2
1
O
O
Figure 1. Cyclic sulfones 1 and 2.
Garratt–Braverman rearrangement in this case is not
possible. In this communication, we describe our results.
The synthesis of the target sulfone 3 was straight for-
ward and was accomplished in four steps starting from
2,20-dihydroxyazobenzene (5). Bis O-alkylation with bu-
tyn-1,4-diol mono tosylate (2 equiv) in the presence of
K2CO3/DMF followed by treatment with mesyl chloride
and triethylamine gave the bismesylate 7. Displacement
with thiophenol followed by oxidation with m-CPBA
produced the target sulfone 3 (Scheme 1). No isomeriza-
In order to strengthen the above rationality, we under-
took a project which involved the synthesis of azo
benzene-based bispropargyl bissulfone 3. We were curi-
ous to know its reactivity, both in terms of isomerization
to the allene and subsequent DNA-cleavage. We ex-
pected lower DNA-cleavage efficiency as the cleavage
can take place only via alkylation; cleavage via
1
tion to the allene could be seen as judged from the H
NMR spectrum (Scheme 2).
The sulfone 3 was in the thermally more stable E-form
as determined from the kmax which was at 360 nm, char-
acteristic of the E-configuration.8 Photo irradiation
from a high pressure Hg-lamp for 5 h produced a 3:1
mixture of Z- and E-isomers with the Z-isomer showing
Keywords: Azobenzene; Bispropargyl; Sulfone; DNA-cleavage; Photo-
isomerization.
1
*
a new kmax at 410 nm. In the H NMR spectrum, new
Corresponding author. Tel.: +91 322 228 3300; fax: +91 322 228
2252; e-mail: absk@chem.iitkgp.ernet.in
peaks at d 6.6 (2,20-aromatic Hs) and d 4.6 (OCH2)
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.06.001

D. Mitra et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4514–4517
OMs
4515
OH
O
N=N
O
O
HO
N=N
OH
b
N=N
a
O
OMs
7
5
6
OH
c
SPh
SO2Ph
O
O
N=N
O
d
N=N
O
SPh
3
SO2Ph
8
Scheme 1. Synthesis of bispropargyl bissulfone. Reagents and conditions: (a) 2-butyn-1,4-diol mono tosylate, K2CO3, DMF, rt, 12 h, 90%; (b) MsCl,
Et3N, CH2Cl2, 0 ꢁC, 15 min, 95%; (c) PhSH, Et3N, CH2Cl2, 0 ꢁC, 30 min, 20%; (d) m-CPBA, CH2Cl2, 0 ꢁC, 1.5 h, 90%.
appeared besides signals for aromatic hydrogens which
were assigned to the Z-isomer (Fig. 2). The kinetics of
thermal reisomerization was followed by monitoring
lene. Similarity of environment for the two propargylic
arms may be the reason for the failure to observe the
monoallene which is fast converted into the bisallene.
Regarding the inversion of allene–propargyl ratio for
3/3a or 4/4a pair as compared to the 9/10 pair, it is pos-
sible that the steric strain as imposed by the peri-hydro-
gen (Fig. 3) to the methylene is less in the propargyl form
than that in the allene form for 3/3a and 4/4a pair. For 9/
10 pair the greater conformational flexibility allows the
formation of more stable allene form in excess. The
aryl-O bond length is possibly more in the propargyl
form because of higher electronegativity of sp-carbon
which reduces the participation of oxygen lone pair for
resonance with the aromatic p-cloud.
1
the H NMR at different time points at a temperature
of 25 ꢁC. Interestingly, the thermal reisomerization
was significantly slower as compared to the cyclic sul-
fone 2 where strain as well as p–p repulsion9 obviously
play major roles in speeding up the reisomerization.
The kinetics of isomerization of the propargyl to allene
system was then studied. As described before, the isom-
erization was followed by recording the H NMR in
1
presence of triethylamine (3 equiv) at 25 ꢁC in CDCl3.
Within 10 min, the equilibrium was reached between
the bispropargyl and the bisallenic bissulfone; however,
the equilibrium favoured the bispropargyl system (6:1).
Similar result was also obtained in case of isomerization
for the cis sulfone 4. The formation of allene moiety was
indicated by the appearance of new peaks at the olefin re-
gion at d 6.28 and d 6.09. The ratio of propargyl to allene
did not change even after keeping the solution for 12 h at
25 ꢁC. This result is in sharp contrast to the isomeriza-
tion of the monopropargyl sulfone 9. In this case also,
the equilibrium was reached within 15 min, however,
the allene 10 was favoured under equilibrium condition
(ratio of allene to propargyl was 2:1). The allene could
be trapped with methanol10 to provide the mono meth-
oxy derivative. In case of isomerization of sulfone 3 or
4, the formation of bisallene was confirmed by trapping
with methanol to produce the dimethoxy derivative de-
tected by mass spectrometry (631, MH+ peak). Peak cor-
responding to the monoallene could not be observed.
The 1H NMR spectrum is also consistent with the bisal-
Thus, it can be concluded that although the better
thermal stability of the Z-isomer 4 in comparison to
the corresponding cyclic sulphone analogue 2 would
allow us to study the DNA-cleavage by carrying out
the incubation for a longer time, the low extent of allene
formation should reduce the efficiency of DNA-cleav-
age. As a matter of fact, both the E-isomer 3 and the
3:1 mixture of Z- and E-isomers (4 and 3) are both poor
DNA-cleaving agents11 even at 100 lM concentrations
(Fig. 4). The mixture however, showed ꢀ20% better
cleavage efficiency as compared to the pure E-isomer.
This can probably be attributed to the greater cleavage
efficiency of Z-isomer, possibly because of better bind-
ing with the DNA due to greater complimentarily12 in
shape between the two. The monopropargyl sulfone 9
also showed slightly higher cleavage activity (ꢀ15%) as
compared to the sulfone 3. The greater extent of allene
formation is perhaps the reason for its better efficiency.

4516
D. Mitra et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4514–4517
SO2Ph
N=N
O
light
heat
N=N
O
O
O
SO2Ph
PhO2S
SO2Ph
3
4
SO2Ph
C
Et3N
Et3N
O
N=N
N=N
O O
O
C
C
C
SO2Ph
PhO2S
SO2Ph
4a
3a
C
SO2Ph
SO2Ph
O
O
Et3N
9
10
Scheme 2. Base mediated isomerization to allene.
However, both these acyclic sulfones were much inferior
as compared to the cyclic sulfone 2.
Figure 2. 1H NMR spectra for thermal Z- to E-isomerization in
CDCl3.
These results clearly indicated the importance of having
a sulfone in a cyclic network in order to develop highly
efficient DNA-cleaving agent. The inherent strain, the
greater acidity of the Hs a- to the sulfone moiety and
the possibility of occurrence of Garratt–Braverman
rearrangement can be exploited in the cyclic system as
compared to an acyclic counterpart where alkylation
remains as the only possible pathway leading to DNA-
cleavage. Moreover, the greater strain in the cyclic
sulfone makes it more reactive towards alkylation via
conversion of sp2 carbons to sp3 state. These parameters
need to be considered while designing sulfone-based
DNA-cleaving agents.
PhO2S
SO2Ph
C
O
N=N
O
N=N
O
O
H H2C
H
H2C
C
Selected spectral data. For 6: dH (200 MHz, CDCl3) 7.7–
7.6 (2H, m), 7.5–7.4 (2H, m), 7.2–7.1 (4H, m), 4.96 (4H,
t, J = 2 Hz,), 4.28 (4H, t, J = 2 Hz); dC (50 MHz,
CDCl3) 154.4, 142.7, 131.6, 121.2, 116.9), 114.9, 86.9,
78.4, 57.2, 49.3; Mass (ES+) m/z 351 (MH+).
SO2Ph
PhO2S
3a
3
Figure 3. The steric interaction with the peri-hydrogens.
For 7: dH (200 MHz, CDCl3) 7.64–7.60 (2H, m), 7.44–
7.40 (2H, m), 7.19–7.06 (4H, m), 5.0 (4H, t,
J = 1.7 Hz), 4.87 (4H, t, J = 1.7 Hz), 2.99 (6H, s); Mass
(ES+) m/z 507 (MH+).
For 8: dH (400 MHz, CDCl3) 7.6 (2H, m), 7.26 (6H, m),
7.18 (6H, m), 7.04 (2H, d, J = 4 Hz), 6.96 (2H,
J = 7.6 Hz), 4.83 (2 · CH2, s), 3.56 (2 · CH2, s); dC
(100 MHz, CDCl3) 155.0, 143.4, 132.0, 130.0, 128.9,

D. Mitra et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4514–4517
4517
References and notes
1. (a) Thurston, D. E. In Introduction to the Principles of
Drug Design and Action, 4th ed.; Smith and Williams,
2006; p 411; (b) Christensen, L. A.; Finch, R. A.; Booker,
A. J.; Vasquez, K. M. Cancer Res. 2006, 66, 4089.
2. Richter, C.; Park, J. W.; Ames, B. N. Proc. Natl. Acad.
Sci. U.S.A. 1988, 85, 6465.
3. Luczak, M. W.; Jagodzinski, P.; Folia, P. Histochem.
Cytobiol. 2006, 44, 143.
4. Kar, M.; Basak, A. Chem. Commun. 2006, 3818.
5. Nicolaou group first reported the DNA-cleaving ability of
propargyl sulfones via isomerization to the allenic coun-
terpart: see Nicolaou, K. C.; Wendeborn, S.; Maligres, P.;
Isshiki, K.; Zein, N.; Ellestad, G. Angew. Chem. Int. Ed.
Engl. 1991, 30, 418.
6. (a) Garratt, P. J.; Neoh, S. B. J. Org. Chem. 1979, 44,
2667; (b) Cheng, Y. S. P.; Garratt, P. J.; Neoh, S. B.;
Rumjanek, V. H. Isr. J. Chem. 1985, 26, 101; (c)
Braverman, S.; Duar, Y.; Segev, D. Tetrahedron Lett.
1979, 3181.
7. Maxam, A. M.; Gilbert, W. Methods Enzymol. 1980, 65,
499; Maxam, A. M.; Gilbert, W. Proc. Natl. Acad. Sci.
U.S.A. 1977, 74, 560.
8. Bose, M.; Groff, D.; Xie, J.; Brustad, E.; Schultz, P. G.
J. Am. Chem. Soc. 2006, 128, 388.
Figure 4. (a) pBR322 DNA-cleavage experiment of compounds 3 and
4 after 2.5-h incubation at 37 ꢁC; lane 1: control DNA in TAE buffer
(pH 8.5, 7 lL) + CH3CN (5 lL); lane 2: DNA in TAE buffer (pH 8.5,
7 lL) + Z-sulfone 4 (0.16 mM) in CH3CN (5 lL); lane 3: DNA in TAE
buffer (pH 8.5, 7 lL) + E-sulfone 3 (0.16 mM) in CH3CN (5 lL); (b)
DNA-cleavage experiment of compounds 2 and 1 after 1.5-h incuba-
tion at 37 ꢁC; lane 1: control DNA in TAE buffer (pH 8.5,
7 lL) + CH3CN (5 lL); lane 2: DNA in TAE buffer (pH 8.5,
7 lL) + Z-sulfone 2 (0.02 mM) in CH3CN (5 lL); lane 3: DNA in
TAE buffer (pH 8.5, 7 lL) + E-sulfone 1 (0.02 mM) in CH3CN (5 lL);
(c) DNA-cleavage experiment of compound 9 after 2.5-h incubation at
37 ꢁC lane 1: control DNA in TAE buffer (pH 8.5, 7 lL) + CH3CN
(5 lL); lane 2: DNA in TAE buffer (pH 8.5, 7 lL) + sulfone 9
(0.16 mM) in CH3CN (5 lL).
9. Koga, N.; Morokuma, K. J. Am. Chem. Soc. 1991, 113,
1907; Pickard, F. C., IV; Shepherd, R. L.; Gillis, A. E.;
Dunn, M.; Feldgus, S.; Kirschner, K. N.; Shields, G. C.;
Manoharan, M.; Alabugin, I. V. J. Phys. Chem. A 2006,
110, 2517.
126.8, 121.8, 117.6, 115.4, 83.9, 78.1, 57.8, 22.8; Mass
(ES+) m/z 535 (MH+).
For 3: dH (400 MHz, CDCl3) 7.85 (4H, d, J = 7.6 Hz),
7.59 (4H, m), 7.43 (6H, m), 7.08 (4H, m), 4.91 (2 · CH2,
m), 3.99 (2 · CH2, m); dC (100 MHz, CDCl3) 134.3,
134.2, 132.2, 129.2, 129.1, 128.6, 122.1 (CH), 117.6,
115.2, 82.6, 75.6, 57.4, 42.6; Mass (ES+) m/z 599 (MH+).
10. Kerwin, S. M. Tetrahedron Lett. 1994, 35, 1023.
11. The DNA used for the present study is pBR322 plasmid
DNA in mainly the supercoiled form. The identity of the
bands has been ascertained from the control DNA which
has Form I as a major band. The cleavage efficiency was
determined by checking the relative UV-absorbance of the
bands at 260 nm. The cleavage efficiency was also
measured by densitometry using image processing soft-
ware (Kodak 1D version V.3.6.3) and similar results were
obtained. It is to be noted that the control DNA specimen
is usually contaminated with some nicked form (form II).
The reason for having an alkaline pH of 8.5 was to aid the
propargyl to allene isomerization.
For 9: dH (200 MHz, CDCl3) 7.86–7.72 (5H, m), 7.50–
7.26 (5H, m), 7.17–7.10 (2H, m), 4.79 (2H, t, J = 2 Hz),
4.01 (2H, t, J = 2 Hz). Mass (ES+) m/z 337 (MH+).
Acknowledgments
M.K. and D.M. thank Council of Scientific and Indus-
trial Research (CSIR, Govt. of India) for research fel-
lowships. A.B. thanks the CSIR for financial support.
12. Caamano, A. M.; Vazquez, M. E.; Martinez-Costas, J.;
Castedo, L.; Mascarenas, J. L. Angew. Chem. Int. Ed.
2000, 39, 3104.