Discovery and structure-activity relationships of (2-(arylthio)benzylideneamino)guanidines as a novel series of potent apoptosis inducers

Article abstract of DOI:10.1016/j.bmc.2009.02.029

1-(2-(2,5-Dimethoxyphenylthio)benzylidene)semicarbazide (2a) was discovered as a potent apoptosis inducer through our cell based HTS assay. SAR study led to the discovery of a more aqueous soluble analog (2-(2,5-dimethoxyphenylthio)-6-methoxybenzylideneam

Full text of DOI:10.1016/j.bmc.2009.02.029

Bioorganic & Medicinal Chemistry 17 (2009) 2852–2858
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Discovery and structure–activity relationships of (2-(arylthio)-
benzylideneamino)guanidines as a novel series of potent apoptosis inducers
*
Han-Zhong Zhang, Candace Crogan-Grundy, Chris May, John Drewe, Ben Tseng, Sui Xiong Cai
EpiCept Corporation, 6650 Nancy Ridge Drive, San Diego, CA 92121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
1-(2-(2,5-Dimethoxyphenylthio)benzylidene)semicarbazide (2a) was discovered as a potent apoptosis
inducer through our cell based HTS assay. SAR study led to the discovery of a more aqueous soluble ana-
log (2-(2,5-dimethoxyphenylthio)-6-methoxybenzylideneamino)guanidine (5e) with EC50 value of 60 nM
in the caspase activation assay and GI50 value of 62 nM in the growth inhibition assay in T47D cells. Com-
pound 5e was found to be an inhibitor of tubulin polymerization and efficacious in a MX-1 breast tumor
model.
Received 6 January 2009
Revised 11 February 2009
Accepted 14 February 2009
Available online 21 February 2009
Keywords:
Ó 2009 Elsevier Ltd. All rights reserved.
Apoptosis inducers
Cell growth inhibition
Anticancer agents
Tubulin inhibitors
In vivo activity
HTS
1. Introduction
Applying this ASAP HTS assay, we have discovered several clas-
ses of small molecules as novel apoptosis inducers. 4-Aryl-4H-
Many traditional anticancer agents have been discovered
through tumor cell growth inhibition assays in neoplastic cells.
Agents active in the screening fall into two categories, cytostatic
and cytotoxic, with the latter classified into two modes of action,
necrotic and apoptotic. Apoptotic cell death is controlled and exe-
cuted by a group of precisely regulated proteins that are frequently
altered in tumor cells.1 The mechanism of apoptosis involves a cas-
pase cascade that is activated sequentially.2,3 Within the caspase
family, caspase-3 has been identified as one of the key effector
caspases that cleave multiple protein substrates in cells, and lead-
ing to irreversible cell death.4 In addition, several clinically used
cytotoxic agents, including paclitaxel, docetaxel and vinca alka-
loids, are known to primarily act by inducing apoptosis in cancer
cells.5
In our effort to discover and develop apoptosis inducers as new
anticancer agents, we have developed a cell-based, anticancer
screening apoptosis platform (ASAP) HTS technology to identify
apoptosis inducers using the activation of caspase-3 as a read-
out.6 One of the major advantages of this technology is that by
monitoring the activation of downstream caspase-3 in a cell-based
assay we can discover compounds that initiate the apoptosis signal
pathway through either a known mechanism or molecular target,
or a novel mechanism or molecular target.7
chromenes, such as compound 1a, were found to be potent apop-
tosis inducers8 and several analogs have been found to have vascu-
lar targeting activity.9 Gambogic acid (1b) was discovered as a fast
apoptosis inducer10 and its molecular target has been identified as
transferrin receptor.11 The 3-aryl-5-aryl-1,2,4-oxadiazoles such as
1c were found to induce apoptosis selectively in certain tumor
cells,12 and through interaction with TIP47, an insulin growth
O
O
O
Br
OH
O
O
O
O
O
CN
O
NH2
N
OH
1b
1a
O
Cl
N
CF3
N
N
S
O
N
N
Cl
1d
1c
* Corresponding author. Tel.: +1 858 202 4006; fax: +1 858 202 4000.
E-mail address: scai@epicept.com (S.X. Cai).
Chart 1.
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.02.029

H.-Z. Zhang et al. / Bioorg. Med. Chem. 17 (2009) 2852–2858
2853
factor II (IGF II) receptor binding protein.13 4-Anilinoquinazilines
O
O
such as 1d were discovered as highly potent apoptosis inducers
with excellent in vivo anticancer activity (Chart 1).14 Herein, we re-
port the discovery of (2-(arylthio)benzylideneamino)guanidines as
a novel series of apoptosis inducers as well as their structure–
activity relationship (SAR) studies using our ASAP program.
O
O
S
N
S
H2, 50 psi
PtO2/ MeOH
HN
NH
NH
2. Results and discussion
2.1. Chemistry
H2N
O
H2N
O
2a
3
Scheme 1.
Substituted (benzylidene)semicarbazides 2a and 2b (Table 1)
were purchased from a commercial library and the structures were
confirmed by 1H NMR and MS. Semicarbazide 3 was prepared from
hydrogenation of 2a (Scheme 1).15 Compounds 5a–i were prepared
as illustrated in Scheme 2. Reaction of substituted ortho-fluoro- or
chloro-benzaldehyde with substituted benzenethiol in the pres-
ence of sodium carbonate in heated DMF produced various substi-
tuted 2-(phenylthio)benzaldehyde 4a–i,16 which was refluxed
with aminoguanidine and sodium acetate to produce 5a–i. The
N-methyl substituted analog 5j was prepared by via condensation
of N-Me-aminoguanidine with 4a (Table 1). Substituted (2-phen-
oxybenzylideneamino)guanidines 7a and 7b were prepared simi-
sulfinylbenzaldehyde 8, followed by condensation of 8 with ami-
noguanidine (Scheme 3).
2.2. HTS and SAR studies
1-(2-(2,5-Dimethoxyphenylthio)benzylidene)semicarbazide 2a
was identified as an inducer of apoptosis using our cell- and cas-
pase 3-based ASAP assays as described previously.8 Compound
2a was then tested in the same ASAP assay over a range of 5–
10,000 nM to provide the dose–response curves for the determina-
tion of EC50. The ASAP assay was also used for the testing of all
compounds for SAR studies. Compound 2a was found to have an
EC50 of 52 nM for T47D breast cancer cells. It also demonstrated
potent caspase activation activity for HCT116 colon cancer cells
larly to
5 via replacing the substituted benzenethiol with
substituted phenol to generate substituted 2-phenoxybenzaldehy-
des 6a and 6b, followed by condensation with aminoguanidine
(Scheme 3). Similarly, compound 9 was prepared through oxida-
tion of the thiobenzaldehyde 4a with sodium periodate to produce
Table 1
Caspase activation activity of [2-(arylthio)benzylideneamino]guanidines and related compounds
4
3
5
6
R2
R2
R2
2
3
6
S
S
X
4
5
R1
N
HN
N
NH
NH
NH
NH
H2N
O
H2N
O
RHN
2a-2b
3
5a-5j, 7a-7b and 9
a
Entry
R1
R2
R
X
EC50
(lM)
T47D
HCT116
SNU398
2a
2b
2,5-OMe
2-OMe
0.052 0.003
>10
0.073 0.004
>10
0.062 0.002
>10
3
5a
5b
5c
2,5-OMe
2,5-OMe
3-OMe
0.15 0.01
0.063 0.003
0.29 0.01
>10
0.29 0.01
0.14 0.02
0.41 0.05
>10
0.21 0.02
0.060 0.006
0.31 0.01
>10
H
H
H
H
H
H
S
S
S
4-OMe
5d
5e
5f
5g
5h
5i
5j
7a
H
3,4-OMe
2,5-OMe
2,5-OMe
2,5-OMe
2,5-OMe
2,5-OMe
2,5-OMe
2,5-OMe
3,5-OMe
2,5-OMe
NA
H
H
H
H
H
H
Me
H
H
S
S
S
S
S
S
S
O
O
SO
NA
NA
0.62 0.06
0.060 0.003
6.9 0.5
0.20 0.03
0.13 0.01
>10
0.070 0.002
0.35 0.02
0.23 0.01
>10
0.57 0.023
0.035 0.003
0.86 0.12
0.11 0.01
>10
0.56 0.11
0.27 0.03
>10
0.16 0.04
0.59 0.02
0.36 0.04
>10
0.29 0.042
0.023 0.006
0.61 0.02
0.051 0.018
4.2 1.4
0.18 0.03
0.12 0.04
>10
0.071 0.025
0.40 0.07
0.27 0.01
>10
0.39 0.064
0.010 0.001
6-OMe
5-OMe
5-NO2
4-OMe
3-OMe
H
H
H
H
7b
9
H
NA
NA
Doxorubicin
Taxol
NAb
NA
NA
a
Data are the mean of three or more independent dose–response experiments and are reported as mean standard error of the mean (SEM).
NA, not applied.
b

2854
H.-Z. Zhang et al. / Bioorg. Med. Chem. 17 (2009) 2852–2858
a) R1 = H, R2 = 2,5-OMe
b) R1 = H, R2 = 4-OMe
c) R1 = H, R2 = 3-OMe
d) R1 = H, R2 = 3,4-OMe
e) R1 = 6-OMe, R2 = 2,5-OMe
f) R1 = 5-OMe, R2 = 2,5-OMe
g) R1 = 5-NO2, R2 = 2,5-OMe
h) R1 = 4-OMe, R2 = 2,5-OMe
i) R1 = 3-OMe, R2 = 2,5-OMe
R2
X
R2
Na2CO3/ DMF
100oC
H
NH2
NH
NaOAc/EtOH/reflux
R2
+
N
R1
S
H2N
CHO
R1
S
SH
R1
X = F or Cl
N
CHO
NH
NH
4a-4i
H2N
5a-5i
Scheme 2.
4a
NaIO4/H2O/MeOH
R2
F
R2
Na2CO3/ DMF
100oC
H
NH2
NH
NaOAc/EtOH/reflux
R2
+
N
7a) X = O, R2 = 2,5-OMe
7b) X = O, R2 = 3,5-OMe
9) X = SO, R2 = 2,5-OMe
X
H2N
CHO
X
OH
N
CHO
NH
NH
6a-6b, X = O
8, X = SO
H2N
Scheme 3.
and SNU398 liver cancer cells (Table 1). In comparison, the EC50
values of taxol and doxorubicin for T47D cells were 35 and
570 nM, respectively. Compound 2b was found to be inactive in
indicating a substitution at this position is well tolerated. The 4-
OMe analog 5h was about twofold less active than 5a, indicating
a substitution at this position is also well tolerated. The 5-OMe
analog 5f was about 100-fold less active than 5a, while the 5-
NO2 analog 5g was only threefold less active than 5a, suggesting
that an electron withdrawing group at the 5-postion is much more
preferred than an electron donating group. The 3-OMe analog 5i
the assay up to 10 lM, >200-fold less active than that of 2a, indi-
cating that an OMe group at the meta-position of the top ring is
critical for activity. Reduction of the C@N double bond resulted
in compound 3, which was threefold less potent than compound
2a, indicating that the rigid C@N double bond is preferred over
the flexible single C–N bond. Since compounds 2a and 2b had lim-
ited aqueous solubility due to the semicarbazide moiety, we
decided to explore the replacement of the semicarbazide group
with a basic aminoguanidine group to increase solubility.
Compound 5a was found to have activity similar to compound
2a in the three cell lines tested, indicating that the semicarbazide
group can be replaced by an aminoguanidine group. In addition,
compound 5a can be prepared as a hydrochloride salt and the salt
showed good aqueous solubility (10 mg/mL in Peg 400:Tween
80:0.03 M tartaric acid water = 7:9:84), for a formulation amenable
to in vivo studies. Similarly, compound 5j, with the semicarbazide
moiety replaced by a N-methyl-aminoguanidine group, also had
activity equivalent to 2a.
was not active up to 10 lM, which is >150-fold less active than 5a.
The caspase activation activity of these (2-(arylthio)benzylide-
neamino)guanidines and related compounds in the human colo-
rectal cancer cell line HCT116 and liver cancer cell line SNU398
roughly paralleled the activity with T47D cells. The differences in
EC50 values between the different cell lines in general are within
twofold. These data suggested that these compounds most proba-
bly should have similar activity against many different cancer cell
lines.
Representative compounds were also tested by the traditional
cell growth inhibition (GI50) assay to confirm that the active com-
pounds can inhibit tumor cell growth. The growth inhibition assays
in T47D, HCT116 and SNU398 cells were run in a 96-well microti-
ter plate as described previously.9,10 Compound 2a was found to be
a potent inhibitor of cancer cell growth with GI50 value of 45 nM in
T47D cells. Compounds 5a and 5e also were highly active with GI50
values of 82 and 62 nM in T47D cells, respectively. In comparison,
the GI50 values of vinblastine and colchicine for T47D cells were
both 7 nM.
The SAR of substitutions in the top ring was explored first. The
3-OMe analog 5b was about fivefold less active than the 2,5-di-
OMe analog 5a. The 4-OMe analog 5c was found to be inactive
up to 10 lM, which was >150-fold and >30-fold less active than
5a and 5b, respectively, confirming the importance of a meta-
OMe group for activity. The 3,4-di-OMe analog 5d was slightly less
active than 5b, indicating that the 4-OMe group is not contributing
to activity.
2.3. Additional characterization of compound 5e and related
compounds
We then explored the replacement of the sulfur linker by other
groups. Compound 7a, with an oxy linker replacing the thio linker
in 5a, was fivefold less active than 5a, indicating that the thio lin-
ker is preferred over the oxy linker. The 3,5-di-OMe oxy linker ana-
log 7b was about as active as 7a, confirming the importance of the
OMe group at the meta-position of top ring for activity. Interest-
ingly, oxidation of the thiol linker to a sulfinyl linker led to com-
The apoptosis-inducing activity of compound 5e was also char-
acterized by flow cytometry. T47D cells were treated with 50 nM of
compound 5e for 48 h, then stained with propidium iodide and
analyzed by flow cytometry. As shown in Figure 1, control cells
(Fig 1A) have a normal cell cycle profile. Cells treated with com-
pound 5e for 48 h showed a shift of the population into G2/M phase
as well as the apoptotic sub-G1 phase (Fig 1B).
pound 9 which was not active up to 10 lM.
We also explored substitutions at various positions of the center
benzylidene ring. The 6-OMe analog 5e was about as active as 5a,
Since many compounds that arrest cells at G2/M followed by
apoptosis are tubulin inhibitors, we suspected that this series of

H.-Z. Zhang et al. / Bioorg. Med. Chem. 17 (2009) 2852–2858
2855
3000
G1 Vehicle; q5dx2; i.v.
25 mg/kg; q5dx2; i.v.
2000
1000
0
0
2
4
6
8
10
12
14
16
18
20
22
Day After Initial Dose
Figure 2. Compound 5e inhibited the growth of established (ꢀ100 mm3) MX-1
tumor xenografts in Crl:Nu/Nu-nuBR mice. Compound 5e was dosed iv at 25 mg/kg
on days 1–5 and 8–12. P-value as calculated by Student’s t-test is 0.03.
pounds such as 5a and 5e, as potent apoptosis inducers with good
aqueous solubility. These compounds also were found to have good
activity in the cell growth inhibition assay. The ability of com-
pound 5e to induce apoptosis was confirmed in a flow cytometry
assay in T47D cells. Compounds 2a, 5a and 5e were found to inhibit
the polymerization of tubulin, which most probably is the mecha-
nism of action for the apoptosis inducing activity of these com-
pounds. Compound 5e also was found to have good activity in a
MX-1 breast cancer in vivo model, suggesting that 5e or related
compounds could be developed as potential anticancer agents.
Figure 1. Drug-induced apoptosis in T47D cells as measured by flow cytometric
analysis. The x-axis is the fluorescence intensity and the y-axis is the number of
cells with that fluorescence intensity. (A) Control cells showing most of the cells in
G1 phase of the cell cycle. (B) Cells treated with 50 nM of compound 5e for 48 h
showing a shift to G2/M as well as cells with sub-diploid DNA content, which are
apoptotic cells with fragmented nuclei.
4. Experimental
apoptosis inducers might also be tubulin inhibitors. The original hit
4.1. General methods and materials
2a was found to be active in the tubulin polymerization assay,19
with an IC50 value of 8
5h were found to inhibit tubulin polymerization with IC50 values
of 3–9 M. In comparison, the IC50 values for vinblastine and col-
chicine were 0.5 M. While not definitive these data suggest that
lM. Similarly, compounds 3, 5a, 5e and
Commercial-grade reagents and solvents were obtained from
Acros, Aldrich, Lancaster, or Butt Park and were used without fur-
ther purification. Thin-layer chromatography (TLC), usually using
ethyl acetate/hexane as the solvent system, was used to monitor
reactions. The 1H NMR spectra were recorded at 300 MHz. Elemen-
tal analyses were performed by Numega Resonance Labs, Inc. (San
Diego, CA). Human breast cancer cells T47D, human colorectal can-
cer cells HCT116 and liver cancer cells SNU398 were obtained from
American Type Culture Collection (Manassas, VA). Compounds 4a–
i were synthesized according to reported procedure via reaction of
the corresponding substituted 2-chlorobenzaldehyde or 2-fluoro-
benzaldehyde with substituted benzenethiol.16
l
l
inhibition of tubulin polymerization probably is the mechanism
of action for the apoptosis inducing activity of these compounds.
Since the original hit semicarbazide 2a and the aminoguanidine
analogs such as 5a and 5e have similar activity profiles in both
the caspase activation assay and growth inhibition assay, it is rea-
sonable that they should have the same mechanism of action.
Due to its good solubility profile, the HCl salt of compound 5e
was selected for in vivo studies. Compound 5e was tested in the
MX-1 breast cancer model, a model that is not specific for certain
agents and is widely used for testing common chemotherapeutics
such as camptothecin derivatives and paclitaxel derivatives.17 The
MX-1 in vivo experiment was performed as described previously.18
Compound 5e inhibited tumor growth dose dependently and pro-
duced 50% tumor growth inhibition when dosed once a day for
5 days for two weeks at a dose of 25 mg/kg (Figure 2), and is well
tolerated with maximum body weight decrease of <10%. In com-
parison, taxol dosed daily IP for 5 days at 8 mg/kg, or a single intra-
venous dose of doxorubicin at 10 mg/kg, both the MTD, resulted in
50–80% reduction of tumor growth.
4.1.1. 2-(3,5-Dimethoxyphenoxy)benzaldehyde (6b)
A mixture of 2-fluorobenzaldehyde (2.48 g, 20 mmol), 3,5-
dimethoxyphenol (3.08 g, 20 mmol) and potassium carbonate
(2.76 g, 20 mmol) in DMF (20 ml) was heated to 150 °C under ar-
gon for 8 h. The reaction mixture was then allowed to cool to room
temperature and partitioned between ethyl acetate and water. The
organic layer was washed with saline, dried over Na2SO4 and con-
centrated to give solid, which was purified by column chromatog-
raphy (hexane/EtOAc 10:1) to give 2.9 g (56%) of 6b as white solid.
1H NMR (CDCl3): 10.47 (s, 1H), 7.92 (q, J1 = 7.5 Hz, J2 = 2.1 Hz, 1H),
7.53 (t, J = 7.5 Hz, 1H), 7.20 (t, J = 7.8 Hz, 1H), 6.99 (d, J = 8.1 Hz,
1H), 6.28 (t, J = 1.8 Hz, 1H), 6.21 (d, J = 2.1 Hz, 2H), 3.77 (s, 6H).
3. Conclusion
In conclusion, 1-(2-(2,5-dimethoxyphenylthio)benzylidene)
semicarbazide (2a) was identified as a potent apoptosis inducer
through our ASAP assay. Exploration of the SAR led to the discovery
of a series of (2-(arylthio)benzylideneamino)guanidines, com-
4.1.2. 2-(2,5-Dimethoxyphenoxy)benzaldehyde (6a)
47%, solid. 1H NMR (CDCl3): 10.62 (s, 1H), 7.91 (q, J1 = 7.5 Hz,
J2 = 2.1 Hz, 1H), 7.45 (t, J = 6.9 Hz, 1H), 7.13 (t, J = 7.8 Hz, 1H),
6.95 (d, J = 9.0 Hz, 1H), 6.78–6.65 (m, 3H), 3.76 (s, 6H).

2856
H.-Z. Zhang et al. / Bioorg. Med. Chem. 17 (2009) 2852–2858
4.1.3. 2-(2,5-Dimethoxyphenylsulfinyl)benzaldehyde (8)
To a 5 °C solution of sodium periodate (449 mg, 2.1 mmol) in
water (20 mL) and methanol (10 mL) was added 2-(2,5-dimeth-
oxyphenylthio)benzaldehyde (4a, 0.55 g, 2 mmol). It was then
warmed to room temperature and stirred for 2 days. The mixture
was poured into water (30 mL) and extracted with ethyl acetate
(3 Â 10 mL). The organic layer was dried over Na2SO4, concen-
trated to give crude product, which was purified by column chro-
matography to give 380 mg (66%) of 8 as solid. 1H NMR (CDCl3):
10.65 (s, 1H), 8.06 (d, J = 8.1 Hz, 1H), 7.97–7.93 (dd, J = 7.8,
1.5 Hz, 1H), 7.76–7.71 (dt, J = 7.8, 1.2 Hz, 1H), 7.60–7.55 (t,
J = 6.6 Hz, 1H), 7.39 (d, J = 3.0 Hz, 1H), 6.95–6.91 (m, 1H), 6.77–
6.74 (d, J = 8.7 Hz, 1H), 3.83 (s, 6H), 3.69 (s, 3H).
(s, 2H), 3.91 (s, 3H), 3.83 (s, 3H). Anal. Calcd for C16H18N4O2S: C,
58.16; H, 5.49; N, 16.96. Found: C, 58.07; H, 5.81; N, 16.91.
4.1.9. (2-(2,5-Dimethoxyphenylthio)-6-methoxy
benzylideneamino)guanidine (5e)
Yield: 98%, solid. 1H NMR (CDCl3): 8.80 (s, 1H), 7.12–6.90 (m,
4H), 6.67 (d, J = 8.4 Hz, 1H), 6.33 (d, J = 8.1 Hz, 1H), 3.91 (s, 3H),
3.77 (s, 3H), 3.75 (s, 3H). Anal. Calcd for C17H20N4O3S: C, 56.65;
H, 5.59; N, 15.54. Found: C, 56.26; H, 5.34; N, 15.48.
4.1.10. (2-(2,5-Dimethoxyphenylthio)-5-methoxy
benzylideneamino)guanidine (5f)
Yield: 72%, solid. 1H NMR (CDCl3): 8.62 (s, 1H), 7.61 (d,
J = 3.0 Hz, 1H), 7.41 (d, J = 8.4 Hz, 1H), 6.90 (q, J1 = 8.4 Hz,
J2 = 2.7 Hz, 1H), 6.74 (d, J = 9.0 Hz, 1H), 6.57 (q, J1 = 8.4 Hz,
J2 = 2.7 Hz, 1H), 6.17 (d, J = 3.0 Hz, 1H), 3.87 (s, 3H), 3.85 (s, 3H),
3.59 (s, 3H). Anal. Calcd for C17H20N4O3S: C, 56.65; H, 5.59; N,
15.54. Found: C, 56.53; H, 5.72; N, 15.54.
4.1.4. 1-(2-(2,5-Dimethoxyphenylthio)benzyl)semicarbazide (3)
A
mixture of 1-(2-(2,5-dimethoxyphenylthio)benzylidene)-
semicarbazide (2a, 120 mg, 0.36 mmol), and platinum(IV) oxide
in methanol (100 mL) and 1.5 N hydrochloride in ethanol (five
drops) was hydrogenated under 50 psi for 20 h. It was filtered
and evaporated and the residue was purified by column chroma-
tography (EtOAc/MeOH/Et3N 10:1:1) to give 30 mg (25%) of com-
pound 3 as white solid. 1H NMR (DMSO-d6): 7.62 (d, J = 7.2 Hz,
1H), 7.48–7.219 (m, 3H), 7.21 (br s, 1H), 7.07 (d, J = 8.7 Hz, 1H),
6.91–6.87 (m, 1H), 6.37 (d, J = 2.7 Hz, 1H), 5.90 (br s, 2H), 5.07
(br s, 1H), 4.00 (s, 2H), 3.84 (s, 3H), 3.60 (s, 3H). Anal. Calcd for
C16H19N3O3SÁ0.5H2O: C, 56.12; H, 5.89; N, 12.27. Found: C, 56.79;
H, 6.09; N, 12.08.
4.1.11. (2-(2,5-Dimethoxyphenylthio)-5-nitrobenzy-
lideneamino)guanidine (5g)
Yield: 93%, solid. 1H NMR (CDCl3): 8.56 (s, 1H), 8.49 (d,
J = 2.7 Hz, 1H), 7.88–7.84 (m, 1H), 7.06–6.88 (m, 4H), 5.35 (br s,
2H), 4.32(br s, 2H), 3.78 (s, 3H), 3.76 (s, 3H). Anal. Calcd for
C16H17N5O4S: C, 51.19; H, 4.56; N, 18.66. Found: C, 51.44; H,
4.87; N, 18.35.
4.1.12. (2-(2,5-Dimethoxyphenylthio)-4-methoxybenzylidene-
amino)guanidine (5h)
4.1.5. (2-(2,5-Dimethoxyphenylthio)benzylideneamino)-
guanidine (5a)
Yield: 54%, solid. 1H NMR (CDCl3): 8.80 (s, 1H), 7.77 (d,
J = 7.8 Hz, 1H), 7.40 (t, J = 8.1 Hz, 1H), 6.95 (d, J = 8.1 Hz, 1H), 6.71
(d, J = 9.0 Hz, 1H), 6.54–6.50 (m, 1H), 6.03 (d, J = 2.7 Hz, 1H), 5.15
(br s, 2H), 4.11 (br s, 2H), 3.85 (s, 3H), 3.80 (s, 3H), 3.56 (s, 3H).
Anal. Calcd for C17H20N4O3S: C, 56.65; H, 5.59; N, 15.54. Found:
C, 56.58; H, 5.86; N, 14.90.
A mixture of 2-(2,5-dimethoxyphenylthio)benzaldehyde (4a,
1.09 g, 4 mmol), aminoguanidine bicarbonate (0.55 g, 4 mmol)
and sodium acetate (0.33 g, 4 mmol) in ethanol (30 mL) was re-
fluxed overnight. The reaction mixture was then allowed to cool
to room temperature and partitioned between ethyl acetate and
water. The organic layer was washed with saline, dried over Na2SO4
and concentrated to give 1.31 g (99%) of 5a as white solid. 1H NMR
(CDCl3): 8.60 (s, 1H), 7.87 (d, J = 7.20 Hz, 1H), 7.31–7.18 (m, 3H),
6.84–6.70 (m, 2H), 6.55 (d, J = 3.0 Hz, 1H), 5.25 (br s, 2H), 4.25 (br
s, 1H), 3.82 (s, 3H), 3.65 (s, 3H). Anal. Calcd for C16H18N4O2S: C,
58.16; H, 5.49; N, 16.96. Found: C, 58.19; H, 5.68; N, 16.93.
The following compounds were prepared from the correspond-
ing substituted 2-phenylthiobenzaldehyde 4b–i, 2-phenoxybenz-
aldehyde 6a–b, or 2-(phenylsulfinyl)benzaldehyde 8 and amino-
guanidine, or N1-methylaminoguanidine by a procedure similar
to that described for the preparation of compound 5a.
4.1.13. (2-(2,5-Dimethoxyphenylthio)-3-
methoxybenzylideneamino)guanidine (5i)
Yield: 36%, solid. 1H NMR (CDCl3): 8.55 (s, 1H), 7.82 (d,
J = 9.0 Hz, 1H), 6.85–6.72 (m, 4H), 6.59 (d, J = 3.0 Hz, 1H), 5.16 (br
s, 2H), 4.06 (br s, 2H), 3.82 (s, 3H), 3.72 (s, 3H), 3.66 (s, 3H). Anal.
Calcd for C17H20N4O3S: C, 56.65; H, 5.59; N, 15.54. Found: C,
56.70; H, 5.76; N, 15.54.
4.1.14. (2-(2,5-Dimethoxyphenylthio)benzylideneamino)-N-3-
methylguanidine (5j)
Yield: 46%, solid. 1H NMR (CDCl3): 8.62 and 8.55 (2br s, 1H),
7.88 (d, J = 8.1 Hz, 1H), 7.30–7.15 (m, 3H), 6.83–6.70 (m, 2H),
6.55 (br s, 1H), 6.30 (br s, 1H), 3.81 (s, 3H), 3.65 (s, 3H), 3.56 (s,
3H), 2.92 and 2.86 (2 br s, 3H). Anal. Calcd for C17H20N4O2S: C,
59.28; H, 5.85; N, 16.27. Found: C, 59.43; H, 5.76; N, 15.96.
4.1.6. (2-(3-Methoxyphenylthio)benzylideneamino)guanidine
(5b)
Yield: 74%, solid. 1H NMR (CDCl3): 8.62 (s, 1H), 7.85 (d,
J = 7.8 Hz, 1H), 7.29–7.19 (m, 5H), 6.91–6.77 (m, 2H), 5.25 (s, 2H),
4.25 (s, 2H), 3.76 (s, 3H). Anal. Calcd for C15H16N4OS: C, 59.98; H,
5.37; N, 18.65. Found: C, 59.83; H, 5.66; N, 18.90.
4.1.15. (2-(2,5-Dimethoxyphenoxy)benzylideneamino)-
guanidine (7a)
4.1.7. (2-(4-Methoxyphenylthio)benzylideneamino)guanidine
(5c)
Yield 84%, white solid. 1H NMR (CDCl3): 8.53 (s, 1H), 7.96 (q,
J1 = 7.8 Hz, J2 = 1.8 Hz, 1H), 7.24 (t, J = 7.2 Hz, 1H), 7.08 (t,
J = 7.8 Hz, 1H), 6.90 (d, J = 9.0 Hz, 1H), 6.81 (d, J = 7.8 Hz, 1H),
6.57 (q, J1 = 8.7 Hz, J2 = 3.0 Hz, 1H), 6.38 (d, J = 3.0 Hz, 1H), 4.80
(br s, 4H), 3.79 (s, 3H), 3.68 (s, 3H). Anal. Calcd for C16H18N4O3ÁH2O:
C, 57.82; H, 6.07; N, 16.86. Found: C, 57.83; H, 6.07; N, 16.86.
Yield: 34%, solid. 1H NMR (CDCl3): 8.58 (s, 1H), 7.66 (q, J1=
7.5 Hz, J2= 1.8 Hz, 1H), 7.42 (d, J = 6.6 Hz, 2H), 7.17–7.07 (m, 2H),
6.94–6.89 (m, 3H), 5.38 (s, 2H), 4.38 (s, 2H), 3.83 (s, 3H). Anal. Calcd
for C15H16N4OS: C, 59.98; H, 5.37; N, 18.65. Found: C, 59.77; H,
5.59; N, 18.91.
4.1.16. (2-(3,5-Dimethoxyphenoxy)benzylideneamino)-
guanidine (7b)
4.1.8. (2-(3,4-Dimethoxyphenylthio)benzylideneamino)
guanidine (5d)
Yield 86%, solid. 1H NMR (CDCl3): 8.44 (s, 1H), 8.02 (q,
J1 = 7.8 Hz, J2 = 1.5 Hz, 1H), 7.30 (t, J = 7.2 Hz, 1H), 7.17 (t,
J = 7.8 Hz, 1H), 6.99 (d, J = 8.1 Hz, 1H), 6.15 (t, J = 2.1 Hz, 1H), 6.08
Yield: 46%, solid. 1H NMR (CDCl3): 8.56 (s, 1H), 7.66 (q,
J1 = 7.5 Hz, J2 = 1.8 Hz, 1H), 7.20–6.90 (m, 6H), 5.34 (s, 2H), 4.35

H.-Z. Zhang et al. / Bioorg. Med. Chem. 17 (2009) 2852–2858
2857
(d, J = 3.0 Hz, 2H), 5.20 (br s, 2H), 3.72 (s, 6H), 2.00 (br s, 2H). Anal.
4.4. Measurement of apoptosis by flow cytometry
Calcd for C16H18N4O3ÁH2O: C, 57.82; H, 6.07; N, 16.86. Found: C,
57.81; H, 6.20; N, 17.01.
Cell cycle analysis was performed as previously described.12
Briefly, T47D cells were treated with 50 nM of compound 5e, and
incubated for 48 h at 37 °C. Control cells were treated with the sol-
vent (DMSO). After the 48 h incubation, cells were treated with
propidium iodide, and analyzed on a flow cytometer. All flow
cytometry analyses were performed on FACScalibur (Becton Dick-
inson) using Cell Quest analysis software. Fluorescence intensity is
plotted on the x-axis and the number of cells with that fluores-
cence intensity is plotted on the y-axis. The sub-diploid amount
of DNA is indicative of apoptotic cells that have undergone DNA
degradation or fragmentation.
4.1.17. (2-(2,5-Dimethoxyphenylsulfinyl)benzylideneamino)
guanidine (9)
Yield 32%, white solid. 1H NMR (CD3OD): 8.43 (s, 1H), 7.89 (q,
J1 = 7.5 Hz, J2 = 1.8 Hz, 1H), 7.74 (q, J1 = 7.8 Hz, J2 = 1.8 Hz, 1H),
7.57–7.44 (m, 2H), 7.10–6.98 (m, 3H), 3.73 (s, 3H), 3.31 (s, 3H).
Anal. Calcd for C16H18N4O3SÁH2O: C, 51.46; H, 5.67; N, 15.00.
Found: C, 51.57; H, 5.27; N, 15.07.
4.2. Caspase activation assay (EC50
)
The potency of compound 2a and related compounds as inducer
of apoptosis was measured by our caspase-based cell assay as pre-
viously described.12 Human breast cancer cells T47D, colon cancer
cells HCT116 and liver cancer cells SNU398 were grown according
to media component mixtures designated by American Type Cul-
ture Collection in RPMI-1640 + 10% FCS in a 5% CO2–95% humidity
incubator at 37 °C. Cells were harvested using trypsin and washed
at 600 g and resuspended at 0.65 Â 106 cells/mL into RPMI med-
4.5. Tubulin inhibition assay
The tubulin inhibition assay was run as previously reported.19
In brief, lyophilized tubulin (Cytoskeleton #ML113, 1 mg, 3 MAP-
rich) was assayed for the effect of the test compound on tubulin
polymerization as measured by change in fluorescence of 4,6-
diamidino-2-phenylindole (DAPI). Polymerized tubulin (sample
treated with DMSO and sample treated with the tubulin stabilizer
Taxol) gives a higher DAPI fluorescence as compared to non-poly-
merized tubulin (vinblastine used to determine baseline). The IC50
for tubulin inhibition was the concentration found to decrease the
fluorescence of DAPI by 50%.
ia + 10% FCS. The cells were treated with 0.005–10 lM concentra-
tions of compound 2a or related compounds in RPMI-1640 and
incubated at 37 °C for 48 h. The samples were then treated with
9
the fluorogenic substrate N-(Ac-DEVD)-N’-ethoxycarbonyl-R110,
and incubated at 37 °C for 3 h. The fluorescent signal was mea-
sured using a fluorescent plate reader (Model Spectrafour Plus Te-
can). The EC50 (lM) was determined by a sigmoidal dose–response
4.6. MX-1 tumor model
calculation (XLFit3, IDBS), as the concentration of compound that
produces the 50% maximum response. The caspase activation
activity (EC50) of the compounds tested in the three cancer cell
lines, T47D, HCT116 and SNU398 are summarized in Table 1.
MX-1 tumors were passaged serially in the mammary fat pad of
female Crl:Nu/Nu-nuBR mice (Charles River Labs). For this study,
tumors from several animals were excised. The viable portion of
the tumor was cut into pieces (ꢀ30 mm3) and a single piece was
implanted into the right lateral mammary pad of each animal using
a trocar. The tumor was allowed to grow to an average volume of
100 mm3 and mice randomized into test groups (N = 8). The ani-
4.3. Cell growth inhibition assays (GI50
)
The potency of compound 2a and related compounds as inhib-
itors of cell proliferation was measured as previously described.12
Briefly, T47D, HCT116 and SNU398 cells were treated with various
concentrations of 2a or related compounds. The samples were
mals were dosed intravenous with 25 lL of 6% ETOH/5% Tween
80/89% D5 W formulation of compound 5e (HCl salt) at dose of
25 mg/kg/day on days 1–5 and 8–12. The mice were observed daily
for morbitity, mortality and signs of toxicity. Tumors and body
weights were measured from day 1 to the end of study with the
frequency measurements determined by the rate of growth of
the tumor. Tumor growth was monitored using external measure-
ments with a caliper and tumor volumes were calculated using the
formula/6(width2 Â length), where width represents the smaller
tumor diameter. The results are shown in Figure 2. These studies
conformed to the recommendations set forth in the USPHS Policy
on Humane Care and Use of Laboratory Animals.
TM
incubated at 37 °C for 48 h, then treated with CellTiter-Glo re-
agent (Promega). The samples were mixed by agitation and incu-
bated at room temperature for 10–15 min. Plates were then read
using a luminescent plate reader (Model Spectrafluor Plus Tecan
Instrument). GI50 values were calculated from dose–response
curves using XLFit3 (IDBS) software to determine the concentra-
tion for 50% reduction in proliferation, the baseline being the initial
cell number. The GI50 for compound 2a and related compounds are
summarized in Table 2.
References and notes
Table 2
1. Henson, P. M.; Bratton, D. L.; Fadok, V. A. Curr. Biol. 2001, 11, R795.
2. Kasibhatla, S.; Tseng, B. Mol. Cancer. Ther. 2003, 2, 573.
3. Green, D. R.; Reed, J. C. Science 1998, 281, 1309.
Cell growth inhibition of [2-(arylthio)benzylideneamino]guanidines and related
compounds
a
4. Salvesen, G. S.; Dixit, V. M. Cell 1997, 91, 443.
Entry
GI50
(lM)
5. Kolenko, V. M.; Uzzo, R. G.; Bukowski, R.; Finke, J. H. Apoptosis 2000, 5, 17.
6. Cai, S. X.; Zhang, H. Z.; Guastella, J.; Drewe, J.; Yang, W.; Weber, E. Bioorg. Med.
Chem. Lett. 2001, 11, 39.
7. Cai, S. X.; Drewe, J.; Kasibhatla, S. Curr. Med. Chem. 2006, 13, 2627.
8. Kemnitzer, W.; Drewe, J.; Jiang, S.; Zhang, H.; Crogan-Grundy, C.; Labreque, D.;
Bubenick, M.; Attardo, G.; Denis, R.; Lamothe, S.; Gourdeau, H.; Tseng, B.;
Kasibhatla, S.; Cai, S. X. J. Med. Chem. 2008, 51, 417.
9. Kasibhatla, S.; Gourdeau, H.; Meerovitch, K.; Drewe, J.; Reddy, S.; Qiu, L.; Zhang,
H.; Bergeron, F.; Bouffard, D.; Yang, Q.; Herich, J.; Lamothe, S.; Cai, S. X.; Tseng,
B. Mol. Cancer Ther. 2004, 3, 1365.
10. Zhang, H. Z.; Kasibhatla, S.; Wang, Y.; Herich, J.; Guastella, J.; Tseng, B.; Drewe,
J.; Cai, S. X. Bioorg. Med. Chem. 2004, 12, 309.
T47D
HCT116
SNU398
2a
3
5a
5e
5h
7a
0.045 0.007
0.31 0.04
0.082 0.003
0.062 0.013
0.037 0.003
0.32 0.02
0.29 0.14
0.90 0.02
0.061 0.014
0.048 0.015
0.097 0.010
0.46 0.02
NDb
0.055 0.008
0.18 0.02
0.021 0.005
0.028 0.01
0.034 0.009
0.12 0.01
ND
Vinblastine
Colchicine
0.007 0.002
0.007 0.001
ND
ND
a
Data are the mean of three or more independent dose–response experiments
and are reported as mean standard error of the mean (SEM).
11. Kasibhatla, S.; Jessen, K.; Maliartchouk, S.; Wang, J.; English, N.; Drewe, J.; Qui,
L.; Archer, S.; Ponce, A.; Sirisoma, N.; Jiang, S.; Zhang, H.-Z.; Gehlsen, K.; Cai, S.
X.; Green, D. R.; Tseng, B. Proc. Natl. Acad. Sci. 2005, 102, 12095.
b
ND, not determined.

2858
H.-Z. Zhang et al. / Bioorg. Med. Chem. 17 (2009) 2852–2858
12. Zhang, H.-Z.; Kasibhatla, S.; Kuemmerle, J.; Kemnitzer, W.; Oliis-Mason, K.; Qui,
L.; Crogran-Grundy, C.; Tseng, B.; Drewe, J.; Cai, S. X. J. Med. Chem. 2005, 48, 5215.
13. Jessen, K.; English, N.; Wang, J.; Qui, L.; Brand, R.; Maliartchouk, S.; Drewe, J.;
Kuemmerle, J.; Zhang, H.-Z.; Gehlsen, K.; Tseng, B.; Cai, S. X.; Kasibhatla, S. Mol.
Cancer Ther. 2005, 4, 761.
16. (a) Jilek, J.; Sindelar, K.; Pomykacek, J.; Kmonicek, V.; Sedivy, Z.; Hrubantova,
M.; Holubek, J.; Svatek, E.; Ryska, M. Collect. Czech., Chem. Commun. 1989, 54,
3294; (b) Urban, J.; Sedivy, Z.; Holubek, J.; Svatek, E.; Ryska, M.; Koruna, I.;
Dlabac, A.; Valchar, M.; Metysova, J. Collect. Czech., Chem. Commun. 1990, 55,
1077.
14. (a) Sirisoma, N.; Kasibhatla, S.; Pervin, A.; Zhang, H.; Jiang, S.; Willardsen, J. A.;
Anderson, M.; Baichwal, V.; Mather, G. G.; Jessing, K.; Hussain, R.; Hoang, K.;
Pleiman, C. M.; Tseng, B.; Drewe, J.; Cai, S. X. J. Med. Chem. 2008, 51, 4771; (b)
Kasibhatla, S.; Baichwal, V.; Cai, S. X.; Roth, B.; Skvortsova, I.; Skvortsov, S.;
Lukas, P.; English, N. M.; Sirisoma, N.; Drewe, J.; Pervin, A.; Tseng, B.; Carlson, R.
O.; Pleiman, C. M. Cancer Res. 2007, 67, 5865.
17. Sapra, P.; Zhao, H.; Mehlig, M.; Malaby, J.; Kraft, P.; Longley, C.; Greenberger, L.
M.; Horak, I. D. Clin. Cancer Res. 2008, 14, 1888.
18. Gourdeau, H.; Leblond, L.; Hamelin, B.; Desputeau, C.; Dong, K.; Kianicka, I.;
Custeau, D.; Bourdeau, C.; Geerts, L.; Cai, S. X.; Drewe, J.; Labrecque, D.;
Kasibhatla, S.; Tseng, B. Mol. Cancer Ther. 2004, 3, 1375.
19. Zhang, H.-Z.; Drewe, J.; Tseng, B.; Kasibhatla, S.; Cai, S. X. Bioorg. Med. Chem.
2004, 12, 3649.
15. Thomas, T.L.;Fedorchuk,M.;Shetty,B. V.;Anderson, F.E. J.Med. Chem.1970, 13,196.