Synthesis and anti-tumor evaluation of new trisulfide derivatives

Article abstract of DOI:10.1016/j.bmcl.2006.06.070

New bis-aromatic and heterocyclic trisulfide derivatives 5, 7-10 were synthesized by optimizing lead dibenzyl trisulfide natural product (4) to evaluate their anti-tumor activities. Five compounds 5-7, 9, and 10 exhibited potent anti-tumor activities against eight different tumor cell lines with low cytotoxicity against HepG2. Initial SAR was discussed, and MOA of these anti-microtubule agents was suggested based on cell kinetic response patterns observed on RT-CES system.

Full text of DOI:10.1016/j.bmcl.2006.06.070

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4826–4829
Synthesis and anti-tumor evaluation of new trisulfide derivatives
Haoyun An,* Jenny Zhu, Xiaobo Wang and Xiao Xu
ACEA Biosciences, Inc., 11585 Sorrento Valley Road, Suite 103, San Diego, CA 92121, USA
Received 22 May 2006; revised 18 June 2006; accepted 21 June 2006
Available online 7 July 2006
Abstract—New bis-aromatic and heterocyclic trisulfide derivatives 5, 7–10 were synthesized by optimizing lead dibenzyl trisulfide
natural product (4) to evaluate their anti-tumor activities. Five compounds 5–7, 9, and 10 exhibited potent anti-tumor activities
against eight different tumor cell lines with low cytotoxicity against HepG2. Initial SAR was discussed, and MOA of these
anti-microtubule agents was suggested based on cell kinetic response patterns observed on RT-CES system.
Ó 2006 Elsevier Ltd. All rights reserved.
Cancer, including over 200 diseases, is the second big-
gest cause of death in the developed countries. A variety
of cancer chemotherapeutic drugs have been used in
clinics and greatly improved survival rates of different
human cancers.1 However, cancer chemotherapy has
generally not been curative, and the tumor cells often
develop multi-drug resistance (MDR) to various chemo-
therapeutic agents.2 The limited efficacy and serious side
effects of available cytotoxic agents often resulted in the
termination of the chemotherapy for some patients.
Therefore, cancer is still the most important unmet med-
ical challenge, and there is an urgent need for new and
safe drugs for cancer chemotherapy.
Calicheamicin,7 and esperamicin8 derivatives have cyclic
or acyclic polysulfide moieties, which are critical for bio-
logical activities of these natural products. To explore
new classes of anti-tumor agents, we discovered anti-tu-
mor activity of natural compound dibenzyl trisulfide (4)
through cell-based assay screening using RT-CES (real-
time cell electronic sensing) cellular screening technolo-
gy (a label-free and non-invasive cell-based assay tech-
nology to quantitatively monitor the dynamic cell
response in real time instead of traditional single point
cellular assays).9 Compound 4 is a biologically active tri-
sulfide that was isolated from the sub-tropical shrub
Petiveria alliacea L.10 The immunomodulatory activity,
molecular mechanism, and some other biological activi-
Over 50% of the anti-cancer drugs currently used are
either natural products or derived from natural prod-
ucts.3 The clinical successes of anti-mitotic drugs paclit-
axel and epithilone stimulated the search for new
natural products. Clinically used anti-microtubule
agents of taxanes, alkaloids, and other types of natural
products have complicated chemical structures and
restricted access to the natural resources.4 Therefore,
derivatization of the biologically active natural
compounds and exploration of new small-molecule
anti-microtubule agents5 are efficient strategies to
accelerate the discovery and development of novel
cancer chemotherapeutic agents.
O
R1
R2
S
S
S
S
S
H2N
1, Varacin; R1 = COCH3, R2 = H
2, Lessoclinotoxin A: R1 = OH, R2 = H
3, 5-(Methylthio)varacin: R1 = COCH3, R2 = SCH3
The naturally-occurring antibiotics varacin (1), lesso-
clinotoxin A (2), 5-(methylthio)varacin (3)6 (Fig. 1)
S
S
S
Keywords: Trisulfide; Anti-tumor activity; Synthesis; SAR.
*
4, Dibenzyl Trisulfide
Corresponding author. Tel.: +1 858 724 0928; fax: +1 858 724
0927; e-mail: han@aceabio.com
Figure 1. Organo sulfur natural products.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.06.070

H. An et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4826–4829
4827
ties of natural compound 4 were reported.11 However,
SCl2
r.t.
N
N
N
2
N
SiMe3
its anti-tumor, anti-inflammatory, and/or anti-infective
activities as well as its potential pharmaceutical applica-
tion have not been studied. Based on the real-time inhib-
itory results against various tumor cell lines (see below)
and the initial mechanistic studies of compound 4, we
decided to optimize lead 4 and further study in vitro
and in vivo anti-tumor activity as well as the mechanistic
property of this new class of anti-tumor agents.
N
N
S
2 ArCH2SH
or 2 ArSH
S
S
S
˚
0-25 C
R
R
4 - 13
Ar
S
Ar
or
S
S
Dibenzyl trisulfide (4) was synthesized from benzyl
sulfenylthiocarbonate in low yield and from labile
benzyl hydrodisulfide intermediate.12 It was also syn-
thesized by the reaction of labile benzyl sulfur chloride
with bis(tributyltin)sulfide.13 The direct coupling of
halides catalyzed by copper (II) and tin (II) reagents
resulted in a mixture of di-/tri-/tetra-sulfides.14 Banerji
and Kalena15 synthesized compound 4 in 66% yield
through diimidazolylsulfide derivative. This would be
an efficient method for the synthesis of new trisulfide
compounds if the distillation of the toxic and easily
decomposed sulfur dichloride reagent can be avoided.
Therefore, we simplified the synthetic procedures by
directly utilizing commercially available sulfur dichlo-
ride solution. The new substituted benzyl trisulfide
derivatives 5–13 were then synthesized in good to
excellent isolated yields using the modified procedure
(Table 1 and Scheme 1). In order to explore different
heterocyclic and flexibility effect of trisulfide deriva-
tives on their anti-tumor activity, we also utilized
the modified protocol and synthesized new heterocy-
clic or flexible trisulfide derivatives 14–20 (Table 2).
The bis-aryl methyl type of trisulfides 5–12 and 14
as well as bis(phenyl-ethylene)trisulfide (16) were syn-
thesized in higher yields, while the sterically hindered
bis(2,4,6-trimethyl-benzyl)trisulfide (13) was obtained
in 68% yield. The bis-heterocyclic trisulfides 18–20
were also obtained in relatively lower yields because
of the low nucleophilicity of the corresponding hetero-
cyclic mercaptans. Bis(4-fluorobenzyl)trisulfide (5) was
fully characterized by 1H NMR, 13C NMR, 19F
NMR, MS, FT-IR, and UV–vis spectroscopic as well
as elemental analyses.16 Other derivatives were also
14 - 20
Scheme 1. Synthesis of trisulfide anti-tumor agents.
Typical procedure for the synthesis ofbis-aryl trisulfides is
as follows: To a stirred solution of N-trimethylsilylimi-
dazole (1.42 mmol) in 1.5 mL of anhydrous hexanes was
added slowly sulfur dichloride solution in dichlorometh-
ane (0.7 mL, 1.0 M, 0.7 mmol) at room temperature
under a nitrogen atmosphere. The reaction mixture hav-
ing white precipitate was stirred for 40 min and then
cooled to 0 °C. A solution of a mercaptan (1.41 mmol)
in 2 mL of anhydrous hexanes was added dropwise under
stirring and nitrogen atmosphere. The resulting reaction
mixture was stirred at 0 °C for 1 h and then at room
temperature for 3 h. The white solid was filtered off
through a pad of Celite and washed with small amount
of hexanes. The filtrate was washed with water and then
brine. The organic phase was dried over anhydrous
sodium sulfate and concentrated under reduced pressure.
The residue was purified by flash chromatography on a
silica gel column using hexanes–ethyl acetate (40–60:1)
as an eluent. The fractions were monitored with silica
gel TLC using hexanes–ethyl acetate (40:1) as a develop-
ing solvent. The resulting white solid product (except
for oily products 10, 14, and 16–18) was re-crystallized
from hexanes to give desired white crystalline products.
Compounds 4–20 were screened against Ovarian (A2780
and OVCAR4), fibrosarcoma (HT1080), non-small cell
lung (H460), breast (MCF7 and M231), and adenocarci-
noma (HeLa) tumor cell lines as well as toxicity-indicat-
ing cell line HepG2 using RT-CES cell-based assay.9
1
characterized by H NMR spectroscopic analysis.17
Table 1. Anti-tumor cellular activity of trisulfides (IC50, lM)
S
S
S
R
R
Compound
R
Yielda (%)
Jurkatb
A2780
OVCAR4
HT1080
H460
MCF7
M231
HeLa
2.5
HepG2
4
5
H15
p-F
p-Cl18
o-Cl
p-Br
0.35
0.15
0.47
0.51
—
0.40
0.5
1.4
0.85
2.25
0.50
1.31
0.9
1.9
1.0
5.1
2.1
6.6
2.1
2.4
3.1
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
87
90
0.29
6
0.7
1.2
2.2
9.6
4.4
7.8
2.3
8.5
2.0
1.1
0.42
6.1
2.7
7
77
84
0.75
1.50
1.43
0.75
11.8
>50
>50
23.2
4.8
1.06
1.33
0.78
1.06
41
8
1.39
0.82
1.9
16
9
p-Me
m-Me
m-CF3
p-tBu
2,4,6-tri-Me
97
99
—
0.6
2.3
2.75
49
10
11
12
13
0.34
8.0
0.60
33.5
>50
>50
12.5
50
100
96
27.5
>50
>50
—
—
>50
>50
>50
>50
>50
>50
>50
>50
68
a Isolated yields.
b Result from MTT assay.

4828
Table 2. Anti-tumor cellular activity of bis-aromatic/heterocyclic trisulfides (IC50, lM)
Ar Ar
H. An et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4826–4829
S
S
S
Compound
Ar
Yieldsa (%)
100
Jurkatb
2.6
A2780
1.3
OVCAR4
12.2
HT1080
3.6
H460
33.5
MCF7
6.25
M231
17.5
HeLa
7.0
HepG2
>100
14
S
15
16
17
18
19
61
96
79
47
53
51
1.2
4.68
7.27
>25
>50
>50
2.6
6.25
4.4
5.1
10.5
11.5
34.8
>50
5.3
9.4
20
42.2
18.2
31.5
>50
>50
5.6
8.8
4.8
19.1
16.0
>50
>50
>50
2.2
4.3
—
—
—
—
>100
>100
>100
>100
>100
>100
S
9.0
11.9
31.6
>50
>50
N
N
17.4
>50
>50
36
N
F3C
>50
>50
N
20
>50
S
a Isolated yields.
b Result from MTT assay.
The inhibitory activities, expressed as IC50 (Tables 1 and
2), were obtained at the 48 h time point-although the
real time inhibitory effect for any given time point was
recorded. The inhibitory activity against leukemia (Jur-
kat) tumor cell line was obtained by MTT assay.19 Bis-
aryl trisulfide derivatives 4–10 substituted with small
functional groups exhibited potent cellular anti-tumor
activity. Bis(4-fluoro-benzyl)trisulfide (5) and bis(4-
methylbenzyl)trisulfide (9) showed the most potent
activity against most of the tested tumor cell lines.
Compounds 5 and 9 have slightly different inhibitory
preferences against different cell lines. However,
compounds 12 and 13 with bulky or more alkyl substit-
uents are inactive. Bis(thiophen-2-yl-methyl)trisulfide
(14) and bis-(benzo[B]thiophen-2-yl-methyl)trisulfide
(15) showed similar anti-tumor activities. Inserting a
CH2 group on both sides of the trisulfide 4 gave more
flexible compound 16, which was less active than 4.
Compound 17 having pyridinyl moiety has equivalent
activity with that of compound 16. However, pyrizinyl
derivative 18 having two nitrogen atoms in aromatic
rings was much less active. The less flexible compounds
19 and 20 completely lost activity. The tested trisulfide
derivatives have more potent inhibitory effect against
Jurkat, A2780, OVCAR4, HT1080, and HeLa tumor
cells than against H460, MCF7, and M231 tumor cells.
The non-inhibitory effect of these compounds against
HepG2 cell line indicated low cytotoxicity of these com-
pounds, which also shows the potential for this type of
compound to be used as anti-tumor agents.
the kinetic responses. It provides important insights in
to various cellular processes and useful mechanistic
information for the action of drugs. Therefore, high-
quality lead compounds and the mechanism of action
can be easily discovered from the screening results.
Some of the potent trisulfide derivatives such as 4, 5,
7, and 9 showed the same kinetic responding pattern
as that of paclitaxel. Therefore, these compounds were
predicted and then verified to have the same mode of
action as anti-cancer drug paclitaxel.20 These com-
pounds also inhibit tumor cells by disturbing tubulin-
microtubule dynamic equilibrium, therefore, arrest the
cells to M phase to block cell division, eventually
causing cell apoptosis.20 This new class of anti-microtu-
bule agents has potential to become useful cancer
chemotherapeutic agents. The detail mechanism of
action, drug target, and cell cycle specific inhibition
as well as in vivo anti-tumor activities of these agents
will be discussed in due course.
References and notes
1. Nygren, P. Acta Oncol. 2001, 40, 166.
2. (a) Liscovitch, M.; Lavie, Y. IDrugs 2002, 5, 349; (b)
Drukman, S.; Kavallaris, M. Int. J. Oncol. 2002, 21, 621.
3. (a) Livett, B. G.; Gayler, K. R.; Khalil, Z. Curr. Med.
Chem. 2004, 11, 1715; (b) Wojcikowski, K.; Johnson, D.
W.; Gobe, G. Nephrology 2004, 9, 400; (c) Topliss, J. G.;
Clark, A. M.; Ernst, E.; Hufford, C. D.; Johnston, G. A.
R.; Rimoldi, J. M.; Weimann, B. J. Pure Appl. Chem.
2002, 74, 1957; (d) Lee, K.-H. J. Biomed. Sci. 1999, 6, 236;
(e) Hung, D. T.; Jamison, T. F.; Schreiber, S. L. Chem.
Biol. 1996, 3, 623.
The real-time cell electronic sensing (RT-CESTM) system
is a label-free and cell-based screening assay.9 The key
advantage of this assay is the real-time measurement of
4. Kingston, D. G. Chem. Commun. 2001, 867.

H. An et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4826–4829
4829
5. (a) Bacher, G.; Beckers, T.; Emig, P.; Klenner, T.;
Kutscher, B.; Nickel, B. Pure Appl. Chem. 2001, 73,
1459; (b) Jordan, A.; Hadfield, J. A.; Lawrence, N. J.;
McGown, A. T. Med. Res. Rev. 1998, 18, 259.
16. Characterization of 5: mp 61.0–62.0 °C; UV–vis (EtOH)
k = 205 nm (x, 23,339), k = 219 nm (x, 22,324),
k = 273 nm (x, 2,382); FT-IR (KBr, cmÀ1): 3047, 3010
(w, mCH), 2950, 2917 (w, mCH ), 1603 (m, mC@C), 1511 (s,
2
6. (a) Davidson, B. S.; Molinski, T. F.; Barrows, L. R.;
Ireland, C. M. J. Am. Chem. Soc. 1991, 113, 4709; (b)
Compagnone, R. S.; Faulkner, D. J.; Carte, B. K.; Chan,
G.; Freyer, A.; Hemling, M. E.; Hofmann, G. A.;
Mattern, M. R. Tetrahedron 1994, 50, 12785; (c) Searle,
P. A.; Molinski, T. F. J. Org. Chem. 1994, 59, 6600.
7. (a) Lam, K. S.; Veitch, J. A.; Golik, J.; Krishnan, B.;
Klohr, S. E.; Volk, K. J.; Forenza, S.; Doyle, T. W. J. Am.
Chem. Soc. 1993, 115, 12340; (b) Lee, M. D.; Dunne, T. S.;
Chang, C. C.; Siegel, M. M.; Morton, G. O.; Ellestad, G.
A.; McGahren, W. J.; Borders, D. B. J. Am. Chem. Soc.
1992, 114, 985.
mC@C), 1415 (m, ArCH2), 1233 (s, mC–F), 834 (s, dCH,
2ortho-H), 652 (m, mCH –S), 527 (s, UC–C, para-substitu-
2
1
tion), 462 (m, mS–S); H NMR (499.1 MHz, CDCl3) d 4.00
(s, 4H), 7.01 (t, 4H, J = 8.8 Hz), 7.27 (dd, 4H, J = 8.8,
5.4 Hz); 13C NMR (125.7 MHz, CDCl3) d 42.4, 115.6,
115.8, 131.2, 131,3, 132.4, 162.5 (C–F, J = 250 Hz); 19F
NMR (376.5 MHz, CDCl3) d À114.2; MS (EI) m/z 314
(M)+. Anal. Calcd for C14H12F2S3: C, 53.48; H, 3.85; F,
12.08; S, 30.59. Found: C, 53.54; H, 3.84; F, 12.35; S,
30.73.
17. 1H NMR (499.1 MHz, DMSO-d6 for 21 and CDCl3 for
other compounds) spectra of compounds: 6: d 3.98 (s, 4H),
7.22 (d, 4H, J = 8.4 Hz), 7.29 (d, 4H, J = 8.4 Hz); 7: d 4.17
(s, 4H), 7.23–7.28 (m, 4H), 7.35–7.43 (m, 4H); 8: d 3.96 (s,
4H), 7.17 (d, 4H, J = 8.3 Hz), 7.45 (d, 4H, J = 8.3 Hz); 9: d
2.33 (s, 6H), 4.01 (s, 4H), 7.14 (d, 4H, J = 8.0 Hz), 7.21 (d,
4H, J = 8.0 Hz); 10: d 2.39 (s, 6H), 4.05 (s, 4H), 7.08–7.20
(m, 6H), 7.23–7.29 (m, 2H); 11: d 4.04 (s, 4H), 7.41–7.49
(m, 4H), 7.51–7.58 (m, 4H); 12: d 1.30 (s, 18H), 4.02 (s,
4H), 7.25 (d, 4H, J = 8.3 Hz), 7.35 (d, 4H, J = 8.3 Hz); 13:
d 2.27 (s, 6H), 2.42 (s, 12 H), 4.23 (s, 4H), 6.87 (s, 4H); 14:
d 4.24 (s, 4H), 6.92–6.96 (m, 2H), 6.99–7.04 (m, 2H), 7.24–
7.28 (m, 2H); 15: d 3.74 (s, 4H), 7.01 (s, 2H), 7.34–7.45 (m,
4H), 7.75 (d, 2H, J = 7.4 Hz), 7.85 (dd, 2H, J = 7.8,
1.1 Hz); 16: d 3.05–3.11 (m, 4H), 3.13–3.18 (m, 4H), 7.19–
7.27 (m, 6H), 7.30–7.35 (m, 4H); 19: d 7.70 (d, 4H,
J = 8.4 Hz), 7.84 (dd, 4H, J = 8.4, 2.4 Hz), 8.73 (s, 2H); 20:
d 7.50–7.62 (m, 4H), 8.34 (d, 2H, J = 6.6 Hz), 8.58 (d, 4H,
J = 4.4 Hz).
8. Beutler, J. A.; Clark, P.; Alvarado, A. B. J. Nat. Prod.
1994, 57, 629.
9. Solly, K.; Wang, X.-B.; Xu, X.; Strulovici, B.; Zheng, W.
Assay Drug Dev. Technol. 2004, 2, 363.
10. (a) De Sousa, J. R.; Demuner, A. J.; Pinheiro, J. A.;
Breitmaier, E.; Cassels, B. K. Phytochemistry 1990, 29,
3653; (b) Johnson, L.; Williams, L. A. D.; Roberst, E. V.
Pestic. Sci. 1997, 50, 228.
11. (a) Williams, L. A. D.; Gardner, T. L.; Fletcher, C. K.;
Naravane, A.; Gibbs, N.; Fleishhacher, R. Phytother. Res.
1997, 11, 251; (b) Rosner, H.; Williams, L. A. D.; Jung, A.;
Kraus, W. Biochim. Biophys. Acta 2001, 1540, 166.
12. (a) Harpp, D. N.; Granata, A. Tetrahedron Lett. 1976,
3001; (b) Harpa, D. N.; Ash, D. K.; Back, T. G.; Gleason,
J. G.; Orwig, B. A.; Vanhorn, W. F.; Snyder, J. P.
Tetrahedron Lett. 1970, 3551.
13. Harpp, D. N.; Gingras, M.; Aida, T.; Chan, T. H.
Synthesis 1987, 1122.
14. Sinha, P.; Kundu, A.; Roy, S.; Prabhakar, S.; Vairamani,
M.; Sankar, A. R.; Kunwar, A. C. Organometallics 2001,
20, 157.
18. Berbesy, G.; Harpp, D. N. Tetrahedron Lett. 1994, 35,
5381.
19. Technical Bull. No. 112, 2004, Promega.
20. Xu, X.; An, H., Zhu, J.; Wang, X. B., et al. Anticancer
Res., in preparation.
15. Banerji, A.; Kalena, G. P. Tetrahedron Lett. 1980, 21, 3003.