Antitumor benzothiazoles. 3. Synthesis of 2-(4- aminophenyl)benzothiazoles and evaluation of their activities against breast cancer cell lines in vitro and in vivo

Article abstract of DOI:10.1021/jm9600959

A new series of 2-(4-aminophenyl)benzothiazoles substituted in the phenyl ring and benzothiazole moiety has been synthesized by simple, high- yielding routes. The parent molecule 5a shows potent inhibitory activity in vitro in the nanomolar range against a panel of human breast cancer cell lines, but is inactive (IC₅₀ > 30 μM) against other cell types: activity against the sensitive breast lines MCF-7 and MDA 468 is characterized by a biphasic dose-response relationship. Structure-activity relationships derived using these cell types has revealed that activity follows the heterocyclic sequence benzothiazole > benzoxazole >> benzimidazole and that 2-(4-aminophenyl)benzothiazoles bearing a 3'-methyl- 9a, 3'-bromo- 9c, 3'- iodo- 9f, and 3'-chloro-substituent 9i are especially potent and their activity extends to ovarian, lung, and renal cell lines. Four compounds have been evaluated in vivo against human mammary carcinoma models in nude mice. Compound 9a showed the most potent growth inhibition against the ER⁺ (MCF- 7 and BO) and ER^- (MT-1 and MT-3) tumors. Our efforts to identify a pharmacological mechanism of action for these intriguing compounds have not, as yet, been successful.

Full text of DOI:10.1021/jm9600959

J . Med. Chem. 1996, 39, 3375-3384
3375
An titu m or Ben zoth ia zoles. 3.1 Syn th esis of 2-(4-Am in op h en yl)ben zoth ia zoles
a n d Eva lu a tion of Th eir Activities a ga in st Br ea st Ca n cer Cell Lin es in Vitr o
a n d in Vivo
Dong-Fang Shi,† Tracey D. Bradshaw,† Samantha Wrigley,† Carol J . McCall,‡ Peter Lelieveld,§
Iduna Fichtner,| and Malcolm F. G. Stevens*,†
Cancer Research Laboratories, Department of Pharmaceutical Sciences, University of Nottingham,
Nottingham NG7 2RD, U.K., Pharmaceutical Sciences Institute, Aston University, Birmingham B4 7ET, U.K.,
EORTC Screening and Pharmacology Group, New Drug Development Office, Free University,
1081 HV Amsterdam, The Netherlands, and Max-Delbru¨ck-Centrum fu¨r Molekulare Medizin, 13122 Berlin
Received February 1, 1996X
A new series of 2-(4-aminophenyl)benzothiazoles substituted in the phenyl ring and benzothi-
azole moiety has been synthesized by simple, high-yielding routes. The parent molecule 5a
shows potent inhibitory activity in vitro in the nanomolar range against a panel of human
breast cancer cell lines, but is inactive (IC50 > 30 µM) against other cell types: activity against
the sensitive breast lines MCF-7 and MDA 468 is characterized by a biphasic dose-response
relationship. Structure-activity relationships derived using these cell types has revealed that
activity follows the heterocyclic sequence benzothiazole > benzoxazole . benzimidazole and
that 2-(4-aminophenyl)benzothiazoles bearing a 3′-methyl- 9a , 3′-bromo- 9c, 3′- iodo- 9f, and
3′-chloro-substituent 9i are especially potent and their activity extends to ovarian, lung, and
renal cell lines. Four compounds have been evaluated in vivo against human mammary
carcinoma models in nude mice. Compound 9a showed the most potent growth inhibition
against the ER+ (MCF-7 and BO) and ER- (MT-1 and MT-3) tumors. Our efforts to identify
a pharmacological mechanism of action for these intriguing compounds have not, as yet, been
successful.
In tr od u ction
retained in the benzoxazole analog 5b but abolished in
the corresponding benzimidazole 5c. In this paper we
describe our extensive studies to explore the SAR in this
simple new antitumor benzothiazole pharmacophore4
and our efforts to identify how the subtle and selective
effects might be explained.
Previously we have described the cytotoxicities of
polyhydroxylated 2-phenylbenzothiazoles against a range
of human tumor cell lines2 and compared their phar-
macological properties with those of the flavone quer-
cetin (1) and the isoflavone genistein (2). These ben-
zothiazoles were designed as potential tyrosine kinase
inhibitors modeled on structural comparisons with the
natural products which compete at the ATP-binding
sites of tyrosine kinases.3 The most interesting bioac-
tive compounds were 4,6-dihydroxy-2-(4-hydroxyphen-
yl)benzothiazole (3) and the related acyclic structure
2,4,4′-trihydroxyazobenzene (4) which have the same
overall substitution pattern as genistein. During the
course of this work, we prepared 2-(4-aminophenyl)-
benzothiazole (5a ) and some substituted arylamines as
synthetic intermediates. These compounds were evalu-
ated against murine fibroblast 3T3 cells and their
counterparts transformed by the Abelson murine
leukemia virus which express the tyrosine kinase
pp120gag-abl: the compounds were more potent against
the transformed cell lines which lent support to the
possibility that they might inhibit the encoded tyrosine
kinase. Surprisingly, we now report that compound 5a
elicits pronounced inhibitory effects against certain
breast cancer cell lines with an unusual biphasic dose-
response relationship against MCF-7 mammary carci-
noma cells in vitro (see later). Activity was partially
Ch em istr y
The most versatile route to 2-(4-aminophenyl)ben-
zothiazoles bearing substituents in both phenyl and
benzothiazolyl rings started with nitrobenzanilides 6,
prepared by interaction of nitrobenzoyl chlorides and
arylamines in pyridine (method A). The benzanilides
were converted to thiobenzanilides 7 with Lawesson’s
reagent in hexamethylphosphoramide (HMPA) (method
B) or chlorobenzene (method C) and cyclized to ben-
zothiazoles 8 by the J acobson synthesis5 with alkaline
* Corresponding author: Dr. Malcolm F. G. Stevens at Cancer
Research Laboratories, University of Nottingham, Nottingham NG7
2RD, U.K.; e-mail, Malcolm.Stevens@nottingham.ac.uk.
†University of Nottingham.
‡Aston University.
§EORTC Screening and Pharmacology Group.
| Max-Delbru¨ch-Centrum, Berlin.
X Abstract published in Advance ACS Abstracts, J uly 15, 1996.
S0022-2623(96)00095-7 CCC: $12.00 © 1996 American Chemical Society

3376 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17
Shi et al.
Ta ble 1. Structure, Yields, and Physical Characteristics of Nitro-Substituted Benzanilides 6, Nitro-Substituted Thiobenzanilides 7,
and 2-(4-Nitrophenyl)benzothiazoles 8
compd
R1
R2
methodj
yield (%)
mp (°C)
formulaa
6a
6b
6c
6d
6e
6f
6g
6h
6i
7a
7b
7c
7d
7e
7f
7g
7h
7i
8a
8b
8c
8d
8e
8f
8g
8h
8i
4-OMe
H
H
H
H
Cl
Cl
Cl
Cl
Cl
H
H
H
H
Cl
Cl
Cl
Cl
Cl
H
H
H
H
H
Cl
Cl
Cl
Cl
Cl
Cl
A
A
A
A
A
A
A
A
A
B
B
B
B
C
C
C
C
B
Bf
D
D
D
D
D
D
D
D
D
D
85
85
83
88
83
82
90
91
97
84
89
85
90
57
71
70
63
83
85
62
29
38
60
46
12
24i
58
32
46
199-202b
173-175c
180-182d
211-213c
144-146b
138-140b
181-183b
190-191b
208-210b
173-175b
184-185e
171-172b
138-140e
145-147b
119-121b
177-179b
111-113b
205-207e
229-231b
216-217g
215-217g
235-236g
238-239h
187-188b
225-227g
177-178g
180-182g
156-159g
216-217g
-
2,4-di-OMe
3,4-di-OMe
3,5-di-OMe
2-OMe
3-OMe
4-OMe
3-benzyloxy
4-benzyloxy
4-OMe
2,4-di-OMe
3,4-di-OMe
3,5-di-OMe
2-OMe
C15H14N2O5
C15H14N2O5
C15H14N2O5
C14H11 ClN2O4
C14H11 ClN2O4
C14H11 ClN2O4
C20H15 ClN2O4
C20H15 ClN2O4
-
C15H14N2O4S
C15H14N2O4S
C15H14N2O4S
C14H11 ClN2O3S
C14H11 ClN2O3S
C14H11 ClN2O3S
C20H15ClN2O3S
C20H15ClN2O3S
-
C14H10N2O3S
C15H12N2O4S
C15H12N2O4S
C15H12N2O4S
C14H9ClN2O3S
C14H9ClN2O3S
C14H9ClN2O3S
C14H9ClN2O3S
C20H13ClN2O3S
C20H13ClN2O3S
3-OMe
4-OMe
3-benzyloxy
4-benzyloxy
H
6-OMe
4,6-di-OMe
5,6-di-OMe
5,7-di-OMe
4-OMe
5-OMe
6-OMe
7-OMe
5-benzyloxy
7-benzyloxy
8j
8k
a
b
Analyses of all new compounds for C, H and N agree within ( 0.4% of calculated values for indicated atoms. Crystallized from
methanol. c Crystallized from ethanol. Crystallized from aqueous methanol. e Crystallized from aqueous ethanol. f Also prepared (74%)
d
from 2-aminothiophenol and 4-nitrobenzoyl chloride in pyridine. g Purified by flash chromatography (ethyl acetate-hexane). Crystallized
h
i
j
from ethanol-ethyl acetate. Also formed (72%) was the product 10. Synthetic methods (see Experimental Section for details): A, from
the aniline and nitrobenzoyl chloride in pyridine; B, from the nitro-substituted benzanilides and Lawesson’s reagent in HMPA; C, from
the nitro-substituted benzanilides and Lawesson’s reagent in chlorobenzene: D, from the thiobenzanilides and potassium ferricyanide in
aqueous sodium hydroxide.
1
potassium ferricyanide (method D). In the case of the
3-methoxythiobenzanilide (7f) a mixture of 5- and
7-substituted benzothiazoles 8g and 8i, respectively,
was isolated from the J acobson cyclization in the ratio
1:5. Similarly, the 3-(benzyloxy)thiobenzanilide 7h gave
a mixture of isomers 8j and 8k with cyclization pre-
ferred at the sterically crowded, but more electron-rich
center ortho to the alkoxy group. In one case an
unexpected product was isolated. J acobson cyclization
of the 4-methoxythiobenzanilide 7g gave the expected
6-methoxybenzothiazole 8h in only 24% yield but the
major product (72%) had MW 643 (FAB MS) corre-
sponding to a “dimer-2H” of the starting material.
Originally a disulfide structure was assigned to this
product, logically formed by oxidation of the thiol
tautomer of 7g in the alkaline ferricyanide:2 however,
this product was surprisingly stable with no cyclization
to the benzothiazole 8h being observed at 160 °C in
diphenyl ether, behavior incompatible with a disulfide
structure. Also, the 1H and 13C NMR spectra were more
complex than those expected for a “dimeric” dithiol.
Thus there were two distinct OMe absorptions at δ 3.15
in 10 rather than the thioamide tautomer since the H
NMR spectrum in CHCl3 showed an exchangeable SH
proton at δ 3.29 whereas other thiobenzanilides 7
typically show (exchangeable) NH absorptions at δ 9.5-
12.5. This anomalous behavior is restricted to thiobenz-
anilides derived from 4-methoxyanilines and explains
the low yields of benzothiazoles from these starting
materials by J acobson cyclization. (Structures, yields
and physical properties of all nitrobenzanilides, ni-
trothiobenzanilides and nitrophenylbenzothiazoles are
listed in Table 1). In consistent manner, no cyclized-
benzothiazole was isolated from the attempted cycliza-
tion of the 4-(benzyloxy)thiobenzanilide 7i.
1
and 3.31 in the H spectrum. We propose, tentatively,
structure 10 for this product, formed by a substitution
ortho to the methoxy group in the Π-excessive ring of
the starting substrate by its oxidatively derived thiol
radical.2 We prefer the imidoylthiol tautomer as shown

Antitumor Benzothiazoles
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17 3377
Sch em e 1a
a
Reagents: (i) Lawesson’s reagent, HMPA or chlorobenzene; (ii) K3Fe(CN)6, aqueous NaOH; (iii) SnCl2‚H2O, EtOH.
Sch em e 2
Reduction of (nitrophenyl)benzothiazoles with tin(II)
chloride dihydrate in ethanol6 (method E) gave the
required (aminophenyl)benzothiazoles 9 (Scheme 1).
Structures, yields, and physical characteristics of all
aminophenylbenzothiazoles prepared by this, and other,
routes are listed in Table 2.
Certain 2-(4-aminophenyl)benzothiazoles with no sub-
stituents in the benzothiazolyl moiety can be prepared
by interaction of 2-aminothiophenol 11 (R1 ) H) and a
4-aminobenzoic acid 12 (method F)7 or benzonitrile 13
(method G) in polyphosphoric acid (PPA) at 220 °C.
Thus, treatment of 2-aminothiophenol with the corre-
sponding substituted 4-aminobenzoic acids (12, R2
)
Me; 3,5-di-Cl; OH) in PPA gave compounds 9a , 9k , and
9w , respectively. This route also afforded the 2-(2-
aminopyridin-5-yl)benzothiazole 19a from the reaction
of 6-aminonicotinic acid and 2-aminothiophenol in PPA.
Similarly, the monochloroamine 9i and chlorotoluidine
9j can be prepared from condensation of 4-amino-3-
chlorobenzonitrile (13; R2 ) 3-Cl) and 4-amino-3-chloro-
5-methylbenzonitrile (13, R2 ) 3-Cl, 5-Me) with 2-ami-
nothiophenol in PPA, respectively (Scheme 2). However,
a variation of this route using 2-aminothiophenol and
the ester, methyl 4-amino-3-iodobenzoate (14, R2 ) 3-I),
in PPA did not afford the expected iodoarylamine 9f:
instead, the deiodinated product 5a was isolated in 14%
yield. Similarly, interaction of 2-aminothiophenol and
4-amino-3,5-diiodobenzoic acid in PPA led only to the
detection of 5a by TLC. Using 4-amino-3-methoxyben-
zoic acid (12, R2 ) 3-OMe) as the acid component gave
a mixture of five products. However, this route can be
applied for the synthesis of the 2-(4-aminophenyl)-
benzoxazole 5b and -benzimidazole 5c by employing
2-aminophenol and o-phenylenediamine, respectively,
to construct the hetero fragment. 2-(4-Nitrophenyl)-
benzothiazole (8a ) is most conveniently formed from
2-aminothiophenol and 4-nitrobenzoyl chloride in pyri-
dine.
2-Phenylbenzothiazoles bearing tertiary amino groups
were prepared by two routes. The pyrrolidine 15a and
the pyridine 18 were prepared by oxidative condensation
of the thiophenol and 4-(pyrrolidin-1-yl)benzaldehyde or
4-formylpyridine, respectively, in refluxing DMSO,8 and
the amines 15b and 15c by displacement of the fluoro
group in 15d by piperidine or morpholine in aqueous
sodium hydroxide-DMSO.
Controlled electrophilic substitution of amine 5a and
its 6-methyl analog with bromine in dichloromethane
at -5 °C gave monobromo-products 9c and 9d (method
H). This method was also extended to mono-bromina-
tion of 19a to afford the (aminobromopyridyl)benzothi-
azole 19b in 82% yield. More forcing bromination of
5a in acetic acid gave the (dibromophenyl)benzothiazole
9e (method I). Monoiodination of 5a , and its 6-methyl-
and 6-methoxy-analog 9m was achieved with iodine
monochloride9 in acetic acid (method J ) to afford 9f, 9g,
and 9u , respectively. The facility for iodine loss from
9f, especially in the presence of copper reagents, was

3378 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17
Shi et al.
Ta ble 2. Structure, Yields, and Physical Characteristics of Alkyl-, Halo, Cyano-, Alkoxy-, and Hydroxy-substituted
2-(4-Aminophenyl)benzothiazoles 9
synthetic
compd
R1
R2
methodf
yield, %
mp,°C
formulaa
5a
9a
9b
9c
9d
9e
9f
9g
9h
9i
9j
9k
9l
9m
9n
9o
9p
9q
9r
H
H
H
H
6-Me
H
H
6-Me
H
H
H
E (F)
F
G
H
H
I
J
J
E
G (K)
G
F
L
E
E
E
E
E
E
E
E
J
E
F
M
M
M
N
O
90 (57)
58
30
79
85
78
70
67
93
54 (63)
62
64
37
92
83
89
90
77
83
90
87
55
92
60
89
62
155-157b
193-195b
118-120c
160-161.5d
188-189.5d
201-203d
143-144d
176-178d
100-101c
159-161d
159-160d
205.5-207b
209-211d
191-193d
181-183d
221-222d
150-152d
148-150d
260 decd
-
3-Me
3-Et
3-Br
3-Br
3,5-di-Br
3-I
C14H12N2S
C15H14N2S
C13H9BrN2S
C14H11BrN2S
C13H8Br2N2S
C13H9IN2S
C14H11IN2S
-
C13H9ClN2S
C14H11ClN2S
C13H8Cl2N2S
C14H9N3S
C14H12N2OS
C15H14N2O2S
C15H14N2O2S
C15H14N2O2S
C14H11ClN2OS
C14H11ClN2OS
C20H15ClN2OS
C14H11ClN2OS
C14H11IN2OS
C14H11ClN2OS
C13H10N2OS
C13H10N2OS
C13H10N2O2S
C14H12N2O2S
C13H9ClN2OS
C13H9ClN2OS
3-I
2-Cl
3-Cl
3-Cl,5-Me
3,5-di-Cl
3-CN
H
H
H
H
2-Cl
2-Cl
2-Cl
2-Cl
3-I
2-Cl
3-OH
H
H
H
2-Cl
2-Cl
H
H
H
6-OMe
4,6-di-OMe
5,6-di-OMe
5,7-di-OMe
4-OMe
5-OMe
5-benzyloxy
6-OMe
6-OMe
7-OMe
H
6-OH
9s
9t
119-121d
183-185e
179-181d
246-249d
214-216d
262-263d
294-295d
282-283d
276-279d
184-186d
9u
9v
9w
9x
9y
9z
9a a
9bb
5,7-di-OH
5-OH, 7-OMe
5-OH
23
45
46
6-OH
a
b
Analyses of all new compounds for C, H, and N agree within (0.4% of calculated values for indicated atoms. Crystallized from
aqueous methanol. c Crystallized from aqueous ethanol. Purified by flash chromatography (ethyl acetate-hexane). e Crystallized from
methanol. f Synthetic methods (see Experimental Section for details): E, reduction of the precursor nitro compound with tin(II) chloride
dihydrate in ethanol; F, from 2-aminothiophenol and a 4-aminobenzoic acid in polyphosphoric acid; G, from 2-aminothiophenol and a
4-aminobenzonitrile in polyphosphoric acid; H, from the unbrominated precursor with bromine (1 mol equiv) in dichloromethane; I, from
the unbrominated precursor with bromine (2.5 mol equiv) in acetic acid; J , from the uniodinated precursor with iodine monochloride in
acetic acid; K, from the iodoamine with copper(I) chloride in DMF; L, from the iodoamine with copper(I) cyanide in DMF; M, demethylation
of the precursor methoxybenzothiazole with boron tribromide in dichloromethane; N, hydrogenation of 8j over 10% palladium-charcoal;
O, demethylation of 9t with borane-NN-diethylaniline and iodine in methanol.
d
Sch em e 3a
attempted Ullmann coupling of 9f to form the 2,2′-
biphenyl 16 employing activated copper powder led only
to deiodination and the isolation of 5a in low yield
(Scheme 3). Diaminobiphenyls of type 16 are major
products from the decomposition of 2-(4-azidophenyl)-
benzothiazoles in trifluoromethanesulfonic acid.1 The
iodinated benzoxazole 17 was prepared (66%) from 5b
and iodine monochloride in acetic acid.
Demethylation of the 6-methoxybenzothiazole 9m
with excess boron tribromide in dichloromethane at -70
°C to yield the phenol 9x proceeded smoothly (89% yield)
even in the presence of the potentially complicating 2-(4-
aminophenyl) substituent (method M). However, de-
methylation of the 5,7-dimethoxybenzothiazole 9p under
a
Reagents: (i) Br2, CH2Cl2; (ii) ICl, AcOH; (iii) CuX (X ) Cl,
CN), DMF; (iv) Cu, DMF.
exploited to achieve the synthesis of chloroaniline 9i and
cyanoaniline 9l using copper(I) chloride (method K) or
cyanide (method L) in boiling DMF. However, an

Antitumor Benzothiazoles
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17 3379
Ta ble 3. In Vitro Cytotoxicity of
2-(4-Aminophenyl)benzothiazole 5a against Human Breast and
Other Cancer Cell Lines
cell line
type
receptor status IC50 (µM)a
MCF-7b
breast
breast
ER+
0.0003
0.0008
MCF-7c
ER+
MCF-7/Adr breast
-
>10
MDA 468
MDA 231
SKBR 3
ZR 75
breast
breast
breast
breast
breast
ER- EGFR+
0.0016
0.017
EGFR+ erbB3+
ER-
ER+
ER-
-
0.0081
0.028
>10
T47D
DU-145
PC-3
WiDr
prostate
prostate
colon
>30
>30
>30
>30
>30
>30
>30
>30
-
-
HT29
colon
-
A204
rhabdomyosarcoma
ovarian
melanoma
bladder
-
A2780
IGR-37
T 24
-
-
-
a
All IC50 values are the mean of at least three determinations.
Determined at the University of Nottingham. c Determined at
b
TNO, Rijswijk.
F igu r e 1. In vitro activity of 2-(4-aminophenyl)benzothiazole
(5a ) against MCF-7 (s) and MDA 468 (bsb) cells.
Ta ble 4. In Vitro Activity of 2-(4-Aminophenyl)benzothiazoles
and Related Compounds against MCF-7 and MDA 468
Mammary Carcinoma Cell Lines
similar conditions was only partial, giving a mixture of
the 5,7-dihydroxy- (9y) and 5-hydroxy-7-methoxyben-
zothiazole 9z. The isomeric 7-hydroxy-5-methoxyben-
zothiazole structure for the demethylation product was
excluded by an NOE difference experiment in DMSO-
d6 degassed with helium. Irradiation of the hydroxyl
signal of 9z resulted in enhancement of H-4 (NOE )
20%) and H-6 (10%); irradiation of the methoxyl singlet
resulted in enhancement only at the H-6 signal (7%).
Catalytic hydrogenation of 5-(benzyloxy)-2-(2-chloro-4-
nitrophenyl)benzothiazole (8j) effected both removal of
the benzyl group and reduction of nitro to amino
affording the aminophenol 9a a (method N). Removal
of the 6-methoxy group of 9t to yield the 6-hydroxyben-
zothiazole 9bb was best accomplished with borane-
N,N-diethylaniline complex and iodine in methanol
(method O).
IC50 range (µM)a
<0.0001-0.001
MCF-7
MDA 468
5a * 9a 9a b,c 9c
9d 9f 9g 9i 9u
9b 9h * 9k 9m *
5b* 9e 9j 9l 19b
15a *
9a 9a b,c 9c 9d
9f 9g 9i 9u
>0.001-0.01
>0.01-0.1
>0.1-1
5a * 5a b * 9b 9k
9e 9l 18 19b
5b* 5c 9h * 9j 9x
9n 9o 9w 15b 15c
18 19a
>1-10
5c 9x 15b 15c 18
>10
9n 9o 9p 9q 9s
9w 9y 19a
9m * 9p 9q 9s
a
b
Initial cytotoxic event. Methanesulfonic acid salt. c Ethane-
sulfonic acid salt. * Biphasic dose-response relationship.
between 1 and 30 µM). Removal of drug by aspiration
of well contents and subsequent washing of cells with
sterile PBS before replenishment with medium alone
led to rapid cell death at these concentrations but did
not affect the growth inhibitory profile at concentrations
less than 1 µM. Aspiration of well contents, subsequent
washing, and replenishment with medium containing
5a did not affect the biphasic dose response.
Inhibition of growth appears to be selective for certain
mammary carcinoma lines (Table 3) as 5a was inactive
(IC50 > 30 µM) against a range of colon, rhabdomyo-
sarcoma, ovarian, melanoma, prostatic, and bladder cell
lines (Table 3). MCF-7/Adr (an MCF-7 variant cell line
with acquired resistance to adriamycin by expression
of the MDR phenotype) appeared resistant to the growth
inhibitory properties of 5a (IC50 > 10 µM).
In order to determine whether compound 5a was
cytotoxic or merely cytostatic, assays were performed
to measure lactate dehydrogenase leakage from treated
cells. Following a drug exposure period of 96 h a
biphasic cytotoxic response was observed. Maximum
toxicity occurred between 10 and 300 nM in both MCF-7
and MDA 468 cell lines. In MCF-7 cells LC50 values of
0.001 µM and 24 µM were estimated for the initial and
second phase cytotoxic events. Estimated LC50 values
of 0.001 and 38 µM were obtained in MDA 468 cells.
Str u ctu r e-Activity Rela tion sh ip s. In MTT as-
says the cytotoxicities (IC50 values) of 2-(4-aminophen-
yl)benzothiazoles and related compounds against MCF-7
and MDA 468 cells extend over a 6 log span (Table 4).
There is good overall consistency in response between
Biologica l Resu lts a n d Discu ssion
In Vitr o Stu d ies. The three isomeric 2-(aminophen-
yl)benzothiazoles were originally prepared as precursors
to the corresponding 2-(hydroxyphenyl)benzothiazoles.2
Evaluation of the prototype benzothiazole 5a against
murine 3T3 cells in vitro revealed an IC50 value of 55
µM compared with an IC50 of 16 µM against the Abelson
leukemia virus transformed, tyrosine kinase pp120gag-abl
encoded, counterpart (ANN-1). The comparable values
for 2-(4-amino-2-chlorophenyl)-6-methoxybenzothiazole
(9t) were 25 and 2 µM, respectively. However, we have
been unable to demonstrate that these compounds elicit
pharmacological effects by tyrosine kinase inhibition
(see later).
In vitro evaluation of 2-(4-aminophenyl)benzothiazole
(5a ) against MCF-7 mammary carcinoma cells in two
laboratories confirmed that the potency of the agent
(IC50 0.0003 and 0.0008 µM) was associated with an
unusual biphasic dose-response relationship (Figure 1).
Maximum growth arrest followed exposure to 100-300
nM; at concentrations between 1 and 30 µM proliferat-
ing colonies were observed (the “second proliferative
phase”) with cytotoxicity at >30 µM. A similar biphasic
dose-response profile was seen for 5a against MDA 468
cells. Continuous exposure of cells to amine 5a was
essential to maintain the second phase (concentrations

3380 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17
Shi et al.
Ta ble 5. In Vivo Evaluation of 2-(4-Aminophenyl)benzothiazole (5a ) and 2-(3-Aminophenyl)benzothiazole against Human Breast
Carcinoma BO
compound
schedule
dosea (mg/kg/inj)
BWCb (%)
RTVc (T/C %)
evaluationd
5a
qd 27, 34, 41
qd 27, 34, 41
qd 27, 34, 41
qd 27, 34, 41
qd 27, 34, 41
qd 27, 34, 41
100
10
1
200
20
2
-7
-4
-5
-6
-3
2
52*
52*
41*
32*
48*
78
(+)
(+)
+
2-(3-aminophenyl)benzothiazole
++
+
-
a
b
d
By ip route. Body weight change. c Relative tumor volume. (+), T/C % g51%; +, T/C % ) 36-50%; ++, T/C ) 21-35%. * Significant
versus controls (p < 0.05).
Ta ble 6. In Vivo Evaluation of 2-(4-Amino-3-methylphenyl)benzothiazole (9a ) and 2-(4-Amino-3-iodophenyl)benzothiazole (9f) against
Human Breast Carcinoma Xenografts
optimum doseb
(mg/kg/inj)
WBCc
(% of controls)
platelets
(% of controls)
BWCd
(%)
RTVe
(T/C%)
compound
tumora
schedule
evaluationf
9a
BO
25
12.5
6.25
12.5
200
200
qd 27, 34, 41
qd 12, 19, 26
qd 7, 14, 21
qd 7, 14, 21
qd 27, 34, 41
qd 12, 19, 26
n.t.
100
n.t.
n.t.
n.t.
112
n.t.
108
n.t.
n.t.
n.t.
106
-9
-12
-3
35*
31*
34*
22*
97
++
++
++
++
-
MCF-7
MT-1
MT-3
BO
-3
9f
-14
-14
MCF-7
68*
(+)
a
b
d
Implanted sc. By ip route. c White blood cells. Body weight change. e Relative tumor volume. f See footnote d, Table 7. n.t. Not
tested. * Significant versus controls (p < 0.05).
the two cell lines. Discrepancies arise when compounds
eliciting a biphasic dose response (indicated with an
asterix) induce an initial phase IC50 value in one cell
line only. To aid evaluation of structure-activity
relationships the data presented in Table 4 refers to IC50
ranges calculated for the initial cytotoxic event (see
Figure 1). Further analysis of the biphasic nature of
the response to selected compounds, and the influence
of exposure time and drug biotransformation in vitro,
will be published separately.
the enhanced potency of the benzothiazole series of
compounds over the benzoxazoles.
Replacement of the 2-(4-aminophenyl) group of 5a
with a 2-(pyridin-4-yl) (18) or 2-(2-aminopyridin-5-yl)
residue (19a ) gave compounds cytotoxic only at concen-
trations >1 µM. Introduction of a bromo group to give
2-(2-amino-3-bromopyridin-5-yl)benzothiazole (19b) en-
hanced activity >100-fold over the unbrominated ami-
nopyridine 19a , but potency was still considerably less
than that of the bromoarylamine 9c.
The lead benzothiazole 5a is 10-fold more cytotoxic
against the MCF-7 and MDA 468 lines than the benz-
oxazole 5b but approximately 10000-fold more active
than the benzimidazole 5c congener. In vitro potency
is retained in the methane- and ethanesulfonic acid salts
of 5a . The tertiary amine analogs 15a -c have IC50
values in the 0.1-10 µM range. Introduction of alkoxy
or hydroxy groups into the benzothiazole nucleus, in the
absence of additional substituents in the arylamine
residue, has a dyschemotherapeutic effect. Particularly
striking is the inactivity of 2-(4-aminophenyl)-5,7-dihy-
droxybenzothiazole (9y) (IC50 80 µM)sactually, the least
active of all the compounds evaluatedsagainst the
MCF-7 cell line, despite this molecule having the same
overall disposition of hydroxy groups in the heterocyclic
nucleus as quercetin (1), genistein (2), and the trihy-
droxybenzothiazole (3).2
In Vivo Stu d ies. As a preliminary to the in vivo
tests the maximum tolerated doses (MTDs) of four test
compounds administered as single doses (ip) in female
BDF1 mice were assessed. Whereas the MTDs of the
parent amine 5a , a reference isomeric compound 2-(3-
aminophenyl)benzothiazole, and 2-(4-amino-3-iodophen-
yl)benzothiazole (9f) were about 200 mg/kg, the corre-
sponding value for the 2-(4-amino-3-methylphenyl) analog
9a was only 25 mg/kg. Possibly these discrepancies
reflect differences in solubility and/or routes of metabo-
lism of the agents. Both compound 5a and the 2-(3-
aminophenyl) isomer elicited inhibitory effects against
the breast carcinoma BO (Table 5), but in the former
case both tumor growth and body weight were influ-
enced relatively independent of dose. To some extent
this mimics the unusual biphasic dose-response seen
in in vitro tests on this compound (eg Figure 1). The
isomer 2-(3-aminophenyl)benzothiazole gave a more
classical in vivo dose-response relationship (Table 5).
The in vivo activity of the amines 9a and 9f in a panel
of four or two, respectively, xenotransplanted breast
carcinomas is recorded in Table 6. Whereas compound
9f displayed only borderline activity in one of the
models, compound 9a induced a consistent tumor growth
inhibition in all four tumors. One representative test
against the MCF-7 carcinoma is documented in Figure
2. Although compound 9a is toxic at the top dose of 25
mg/kg with only one survivor, the surviving animal was
tumor free and no overall change in white blood cell or
platelet counts were measured, indicating that bone
marrow toxicity is not dose-limiting. The activity of 9a
against the ER- tumors MT-1 and MT-3 is notable
because these tumors are predictive for the clinical
Introduction of a Cl substituent into the 2′-position
of the aminophenyl group of 5a gave compound 9h with
reduced activity compared to the parent amine espe-
cially against the MDA 468 cell line (IC50 0.6 µM).
However, a rich seam of activity extends through
compounds with a 3′-methyl- (9a ), 3′-ethyl- (9b), 3′-
bromo- (9c), 3′-iodo- (9f)-, and 3′-chloro- (9i) substitu-
ents. Particularly effective against both cell lines (IC50
< 0.0001 µM) are 2-(4-amino-3-bromophenyl)-6-meth-
ylbenzothiazole (9d ) and the iodo analog 9g. In con-
trast, 3′-cyano- (9l) and 3′-hydroxy- (9w ) substituents
reduced activity markedly compared to the parent
amine 5a . A similar dyschemotherapeutic effect was
observed in 3′,5′-disubstituted compounds 9e, 9j, and
9k . The 3′-iodo derivative in the benzoxazole series 17
is approximately 10-fold less active than 9f, confirming

Antitumor Benzothiazoles
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17 3381
F igu r e 2. In vivo activity of 2-(4-amino-3-methylphenyl)benzothiazole (9a ) against MCF-7 in nude mice administered on days
12, 19, and 26. Doses: 9s9, saline control; )s), 6.25 mg/kg/inj; (s(, 12.5 mg/kg/ inj; 0-0, 25 mg/kg/ inj. *Significant versus
controls (p < 0.05).
activity of cyclophosphamide, adriamycin, and mito-
xantrone and are exquisitively sensitive to hexade-
cylphosphocholine and other ether lipids.10 In contrast
these tumors are completely unresponsive to metho-
trexate and vincristine, and only modestly sensitive to
cisplatin.
Mech a n ism Stu d ies. Our search for a pharmaco-
logical mechanism to explain the unusual activity of this
new series of compounds has not been illuminating.
Thus (data not shown) compound 5a has no effect on
the assembly of microtubules in P388 cells; it does not
inhibit human topoisomerase II (IC50 > 100 µM). At
concentrations of 2, 5, and 10 µM, 5a had no effect on
aromatase and lyase obtained from human placental
microsomes; also 5a did not modulate protein kinase C
(PKC) activity. The cell cycle regulators of G2/M
transition, cdc 2 kinase, and cdc 25 phosphatase ob-
tained from starfish oocytes were not affected by 5a , 9a ,
9c, or 9f at concentrations of <50 µM.
Western blot assays performed using antiphosphoty-
rosine primary antibody demonstrated no inhibition of
tyrosine phosphorylation following treatment with com-
pounds 5a and 9f in unstimulated and EGF-stimulated
MCF-7 or MDA 468 cells. 5a (1, 10 and 100 µM) did
not compete with EGF for EGF receptor in MCF-7 and
MDA 468 cell lines. In the HBL-316 cell line, which is
genetically engineered to overexpress a point mutation-
activated erbB2 kinase, no inhibition of total phospho-
tyrosine levels measured by immunoassay, was de-
tected. EGF (5, 50, and 100 ng/mL), diethylstilboestrol,
R- and â-oestradiol (0.1 nM-1 µM) did not interfere with
the biphasic nature of the dose response in MCF-7 cells
to compound 5a in phenol red free medium supple-
mented with charcoal stripped FCS, or medium supple-
mented with 1% FCS. Moreover, 5a did not compete
with oestradiol for estrogen receptors (ER). Thus it
appears unlikely that the benzothiazoles subvert ER or
EGFR dependent pathways to elicit their response.
Concentrations of insulin (10 ng/mL), IGF I (10 nM),
and IGF II (5 nM) which have been reported to stimu-
late the growth of breast cancer cell lines11,12 did not
affect the dose-response profiles nor the estimated IC50
values for growth arrest in MCF-7 cells induced by 5a .
To aid future studies to elucidate the mode(s) of action
of this series of compounds, MCF-7 cell lines have been
established with acquired resistance to 5a by long term
exposure to 10 nM or 10 µM drug. These cells yielded
estimated IC50 values of 48 µM and 57 µM, respectively,
when rechallenged with drug. In addition, certain
compounds have been evaluated in the NCI tumor cell
panel. Compounds 9a , 9c, 9f, and 9i, four of the most
active compounds described in the present work (Table
4), also show potent activities and disease selectivity,
particularly in the NCI breast and ovarian subpanels
but also against certain non-small cell lung (NCI-H226
and NCI-H460), colon (HCC-2998), and renal cell lines
(A498 and TK-10), with biphasic dose-response rela-
tionships in the sensitive lines. In these evaluations,
cells were exposed to drug for only 48 h compared with
the longer drug exposures detailed in this paper. The
cytotoxicity fingerprints of these four compounds are
remarkably consistent. When 2-(4-amino-3-iodophenyl)-
benzothiazole (9f) was used as a seed in a COMPARE
analysis13 the three closest compounds in a database of
48 000 profiles were the structurally related benzothi-
azoles 9a , 9c, and 9f (Pearson correlation coefficient >
0.75). Moreover, as the compounds do not COMPARE
with any clinical agents of known pharmacological type
they may be operating by a new mechanism. Further
details of these NCI results and the activities of 2-(4-
aminophenyl)benzothiazoles in lung and ovarian cell
panels will be published separately.
Con clu sion
The new 2-(4-aminophenyl)benzothiazoles reported in
this paper are simple to prepare, the molecules are
chemically robust and display potent inhibitory proper-
ties in a range of cell types in vitro. No pharmacological
mechanism has yet been adduced to explain these subtle
and selective effects, particularly against breast cancer
cell lines. For one example, 2-(4-amino-3-methylphen-
yl)benzothiazole (9a ), this effectiveness extends to useful
inhibition of a panel of human breast cancer models in
nude mice. The possibility that the compounds might
act by a novel mechanism makes the search for a
pharmaceutically suitable candidate for clinical trial an
urgent one.
Exp er im en ta l Section
All new compounds listed in the tables were characterized
by elemental microanalysis (C, H, and N values (0.4% of
theoretical values). Mass spectra were recorded on an AEI
MS-902 or a VG Micromass 7070E spectrometer. IR spectra
were determined in KBr on a Mattson 2020 GALAXY series

3382 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17
Shi et al.
FT-IR spectrophotometer. 1H NMR and 13C NMR spectra were
recorded on Bruker AC250 or ARX250 spectrometers. TLC
systems for routine monitoring of reaction mixtures and
confirming the homogeneity of analytical samples employed
Kieselgel 60F254 (0.25 mm) with either CHCl3 or CHCl3-2%
ethanol as developing solvents. Sorbsil silica gel C 60-H (40-
60 µm) and an ethyl acetate-hexane solvent mixture was used
for flash chromatographic separations.
Gen er a l Meth od for th e Syn th esis of Su bstitu ted
Nitr oben za n ilid es 6. Meth od A. A mixture of the substi-
tuted aniline (0.02 M) and the appropriate nitrobenzoyl
chloride (0.02 M) in pyridine (50 mL) was stirred under reflux
(2 h) and poured into water. The precipitate was collected,
washed with water, and purified as indicated in Table 1.
Gen er a l Meth od s for th e Syn th esis of Su bstitu ted
Nitr oth ioben za n ilid es 7. Meth od B. A mixture of the
substituted nitrobenzanilide (0.01 M) and Lawesson’s reagent
(0.6 mol equiv) in HMPA (30 mL) was stirred at 100 °C for 6
h and poured into water. The precipitate was collected and
washed with water. Yields, physical characteristics, and
purification methods are listed in Table 1.
Meth od C. As Method B, but using refluxing chlorobenzene
(5-10 mL) for 3 h as solvent. Products precipitated slowly
from the cooled reaction mixtures.
Gen er a l Meth od for th e J a cobson Syn th esis of Su b-
stitu ted 2-(4-Nitr op h en yl)ben zoth ia zoles 8. Meth od D.
Substituted nitrothiobenzanilides 7 (0.05 M) were wetted with
a little ethanol, and 30% aqueous sodium hydroxide solution
(8 mol equiv) was added. The mixture was diluted with water
to provide a final solution/suspension of 10% aqueous sodium
hydroxide. Aliquots of this mixture (5 mL) were added at 1
min intervals to a stirred solution of potassium ferricyanide
(4 mol equiv) in water at 80-90 °C. The reaction mixture was
heated for a further 0.5 h and then allowed to cool. Products
8 were collected and washed with water. Details of yields,
physical characteristics, and purification methods are given
in Table 1.
Meth od I. A solution of 2-(4-aminophenyl)benzothiazole
(5a ) (0.6 g) in acetic acid (15 mL) was mixed with a solution
of bromine (0.98 g, 2.5 mol equiv) in acetic acid at 25 °C and
then warmed to 80 °C for 2 h. The acetic acid was removed
by vacuum evaporation and the residue partitioned between
10% aqueous sodium bicarbonate (150 mL) and dichlo-
romethane (150 mL). The separated organic layer was purified
as in method H to yield 2-(4-amino-3,5-dibromophenyl)ben-
zothiazole 9e as a yellow solid (78%). Physical characteristic
are recorded in Table 2.
Meth od J . To a solution of a 2-(4-aminophenyl)benzothi-
azole (0.013 M) in acetic acid (35 mL) was added dropwise over
10 min at 25 °C a solution of iodine monochloride (0.017 M)
in acetic acid (35 mL). The mixture was stirred at 25 °C for
a further 1.5 h and excess acetic acid removed by vacuum
evaporation. The residue was partitioned between aqueous
sodium bicarbonate-dichloromethane and the separated or-
ganic layer purified as in method H.
Meth od K. A mixture of 2-(4-amino-3-iodophenyl)ben-
zothiazole (9f) (0.3 g) and copper(I) chloride (0.16 g) was boiled
in DMF (20 mL) for 4 h. The reaction mixture was concen-
trated by vacuum evaporation and the residue stirred with
water. Organic products were extracted into ethyl acetate,
dried (MgSO4), and chromatographed on silica gel using ethyl
acetate-hexane (2:5) as eluting solvent. 2-(4-Amino-3-chlo-
rophenyl)benzothiazole (9i) was isolated (63%) as a cream solid
(see Table 2).
Meth od L. Interaction of the benzothiazole 9f with copper-
(I) cyanide in boiling DMF as in method K gave 2-(4-amino-
3-cyanophenyl)benzothiazole (9l) (37%) (see Table 2).
Gen er a l Met h od s for t h e Syn t h esis of H yd r oxy-
Su bstitu ted 2-(4-Am in op h en yl)ben zoth ia zoles. Meth od
M. To a stirred suspension of the methoxy-substituted 2-(4-
aminophenyl)benzothiazole (0.01 M) in dry dichloromethane
(25 mL) under nitrogen at -70 °C was added a 1 M solution
of boron tribromide (2 mol equiv for each methoxy group plus
a further 3 mol equiv)2 dropwise over 30-45 min. The mixture
was maintained at -70 °C for 1 h and then allowed to warm
slowly to 20 °C and stirred overnight. The reaction mixture
was recooled to -70 °C and quenched by the dropwise addition
of methanol until no further reaction occurred. The mixture
was poured into 8% aqueous sodium hydroxide solution (50
mL). The aqueous phase was separated, neutralized to pH 7
with 5 M hydrochloric acid, and extracted three times with a
mixture of dichloromethane-methanol (4:1). The combined
organic layers were dried (MgSO4) and evaporated to yield
hydroxy-substituted 2-phenylbenzothiazoles. The crude prod-
ucts were purified by flash column chromatography employing
ethyl acetate-hexane (3:2) as solvent. Yields and physical
characteristics of the products are listed in Table 2.
Meth od N. A solution of 5-(benzyloxy)-2-(2-chloro-4-nitro-
phenyl)benzothiazole 8j (0.064 g) in acetone was hydrogenated
at 30 psi over a 10% palladium-charcoal catalyst (0.025 g).
After removal of catalyst through Celite and evaporation of
solvent, the residue was purified by flash chromatography,
eluting with ethyl acetate-hexane (1:2), to yield 2-(4-amino-
2-chlorophenyl)-5-hydroxybenzothiazole (9a a ) (45%): see Table
2 for physical characteristics.
Meth od O. Compound 9t (0.068 g) was reacted with
borane-N,N-diethylaniline complex (0.057 g) and iodine (0.8
g) in methanol at room temperature overnight. After removal
of solvent the residue was dissolved in ethyl acetate (50 mL)
and shaken successively with 5% aqueous sodium thiosulfate
(2 × 50 mL) and water (2 × 100 mL). The organic layer was
concentrated and the residue chromatographed using ethyl
acetate-hexane (1:1) as eluting solvent. The 6-hydroxyben-
zothiazole 9bb was isolated in 46% yield (see Table 2 for
physical characteristics).
J a cobson Cycliza tion of Th ioben za n ilid e 7g. Cycliza-
tion of this thiobenzanilide by method D gave a mixture which
was separated by flash chromatography using ethyl acetate-
hexane (1: 4) as eluant. The minor product was 2-(2-chloro-
4-nitrophenyl)-6-methoxybenzothiazole 8h (24%). (For details
of physical characteristics see Table 1). The major product
1
(72%) was the imidoyl sulfide 10; mp 136-137 °C; H NMR
(CHCl3) δ 3.15 (s, 3 H, OMe), 3.29 (s, 1 H, SH, exchangeable
with CD3OD), 3.31 (s, 3 H, OMe). Anal. (C28H20Cl2N4O6S2)
C, H, N.
Gen er a l Meth od s for th e Syn th esis of Su bstitu ted 2-(4-
Am in op h en yl)ben zoth ia zoles 9. Meth od E. A mixture of
the appropriate 2-(4-nitrophenyl)benzothiazoles 8 (0.015 M)
and tin(II) chloride dihydrate (0.075 M) in boiling ethanol (50
mL) was stirred under nitrogen for 4 h. Ethanol was removed
by vacuum evaporation, the residue was extracted into ethyl
acetate (3 × 100 mL), and the combined organic fraction
shaken with 2M aqueous sodium hydroxide solution (3 × 100
mL) followed by water (2 × 100 mL). The organic layer was
evaporated to leave a residue of the amine. Yields, physical
characteristics, and purification methods are listed in Table
2.
Meth od F . 2-Aminothiophenol 11 (R1 ) H) (0.053 M) and
the appropriate substituted benzoic acids 12 (0.05 M) were
heated in polyphosphoric acid (PPA; 85 g) at 220 °C for 4 h,
cooled, and poured into ice-cold 10% aqueous sodium carbon-
ate. The solid product was collected, washed (water), and
purified as indicated in Table 2.
Meth od G. As Method F but reacting together 2-ami-
nothiophenol and a substituted benzonitrile 13 in PPA.
Meth od H. A solution of bromine (0.32 g) in dichlo-
romethane (10 mL) was added at -5 °C to a solution of a 2-(4-
aminophenyl)benzothiazole (1 mol equiv) in dichloromethane
(50 mL). After being stirred at -5 °C for 2 min the mixture
was shaken with ice-water (400 mL) for 1 h. The separated
organic layer was washed with 10% aqueous sodium thiosul-
fate (2 × 50 mL) and water (2 × 60 mL), dried (MgSO4), and
vacuum evaporated to furnish a residue which was purified
by flash chromatography.
2-(4-Am in op h en yl)ben zoth ia zole (5a ). This compound
was prepared from reduction of 2-(4-nitrophenyl)benzothiazole1
with tin(II) chloride dihydrate in ethanol: mp 155-157 °C
1
(lit.,14 157 °C); H NMR (DMSO-d6) δ 8.00 (d, 1H, J 7.1 Hz,
H-7), 7.88 (d, 1H, J 7.7 Hz, H-4), 7.75 (d, 2H, J 8.6 Hz, H-2′,
H-6′), 7.45 (dt, 1H, J 1.3, 7.6 Hz, H-5), 7.33 (dt, 1H, J 1.2, 7.5
Hz, H-6), 6.66 (d, 2H, J 8.7 Hz, H-3′, H-5′), 5.90 (brs, 2H, NH2);
IR 3458, 3296, 3179, 1637, 1604, 1479, 1432, 1312, 1239, 1186,
963, 826, 760, 730, 715 cm-1
.

Antitumor Benzothiazoles
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17 3383
2-(4-Am in oph en yl)ben zoxazole (5b). 2-Aminophenol (1.5
g) and 4-aminobenzoic acid (1.89 g) were reacted in PPA at
190 °C according to General Method F. The benzoxazole 5b
(62%) had mp 176-178 °C (MeOH-H2O) (lit.,15 mp 185 °C).
2-(4-Am in op h en yl)ben zim id a zole (5c). Similarly pre-
pared, from 2-aminoaniline and 4-aminobenzoic acid, the
benzimidazole 5c (70%) had mp 244.5-246 °C (EtOH-H2O)
(lit.,16 mp 246-247 °C).
2-(4-Am in o-3-m eth ylp h en yl)ben zoth ia zole (9a ). Pre-
pared from 4-amino-3-methylbenzoic acid and 2-aminothiophe-
nol at 220 °C in PPA according to General Method F. 1H NMR
(CDCl3) δ 8.02 (d, 1H, J 8.0 Hz, H-7), 7.90-7.86 (m, 2H, H-4,
H-2′), 7.79 (dd, 1H, J 2.1, 8.2 Hz, 6′-H), 7.48 (dt, 1H, J 1.3, 7.7
Hz, 5-H), 7.37 (dt, 1H, J 1.2, 7.6 Hz, 6-H), 6.76 (d, 1H, J 8.2
Hz, H-5’), 3.98 (br s, 2H, NH2), 2.28 (s, 3H, CH3); 13C NMR
(CDCl3) δ 169.2 (C), 154.7 (C), 148.0 (C), 135.0 (C), 130.2 (CH),
127.4 (CH), 126.4 (CH), 124.8 (CH), 124.3 (C), 122.9 (CH),
122.6 (C), 121.8 (CH), 115.0 (CH), 17.6 (CH3); IR 3469, 3323,
3198, 1625, 1481, 1438, 1403, 1310, 1237, 753 cm-1; MS (EI)
m/ z 240 (M+), 223.
2-(2-Am in op yr id in -5-yl)ben zoth ia zole (19a ). Prepared
from 2-aminothiophenol and 6-aminonicotinic acid according
to General Method F, the benzothiazole 19a (59%) had mp
188-189 °C; 1H NMR (DMSO-d6) δ 8.65 (d, 1 H, J 2.3 Hz,
H-2′), 8.08-8.01 (m, 2 H, H-6′,4), 7.95 (d, 1 H, J 8.1 Hz, H-7),
7.49 (t, 1 H, J 7.3 Hz, H-5), 7.38 (t, 1 H, J 7.4 Hz, H-6), 6.77
(brs, 2 H, NH2), 6.59 (d, 1 H, J 8.8 Hz, H-5′); IR (KBr) 3370,
3150, 1610, 1470, 1400, 955, 840, 750 cm-1; MS (EI) m/ z 228
(M+ + 1). Anal. (C12H9N3S) C, H, N.
2-(2-Am in o-3-br om op yr id in -5-yl)ben zoth ia zole (19b).
The amine 19a was brominated in dichloromethane according
to General Method H. The crude product was chromato-
graphed, eluting with hexane-ethyl acetate (2:1), to give title
compound (82%): mp 232-233.5 °C, 1H NMR (DMSO-d6) δ
8.66 (d, 1 H, J 2.0 Hz, H-2′), 8.33 (d, 1 H, J 2.0 Hz, H-6), 8.11
(d, 1 H, J 7.9 Hz, H-4), 7.98 (d, 1 H, J 8.0 Hz, H-7), 7.53 (t, 1
H, J 7.5 Hz, H-5), 7.42 (t, 1 H, J 7.5 Hz, H-6), 7.06 (brs, 2 H,
NH2); IR (KBr) 3420, 3275, 3150, 1650, 1465, 1400, 1220, 755
cm-1; MS (EI) m/ z 308 (M+ + 2), 305 (M+ - 1). Anal. (C12H8-
BrN3S) C, H, N.
Syn t h esis of Sa lt s of 2-(4-Am in op h en yl)b en zot h i-
azoles. 2-(4-Am in oph en yl)ben zoth iazole Meth an esu lfon -
ic Acid Sa lt. To a solution of the aminophenylbenzothiazole
5a (0.5 g) in ethyl acetate (65 mL) was added, dropwise,
2-(4-Am in o-3-br om op h en yl)ben zoth ia zole (9c). Com-
pound 5a was brominated with bromine in dichloromethane
at -5 °C according to General Method H to give 9c. 1H NMR
(DMSO-d6) δ 8.08-8.05 (m, 2H, H-7, H-2′), 7.95 (d, 1H, J 8.0
Hz, H-4), 7.78 (dd, 1H, J 1.6, 8.5 Hz, H-6′), 7.50 (t, 1H, J 7.6
Hz, H-5), 7.39 (t, 1H, J 7.5 Hz, H-6), 6.91 (d, 1H, J 8.5 Hz,
H-5′), 6.13 (brs, 2H, NH2); 13C NMR (DMSO-d6) δ 167.3 (C),
154.5 (C), 149.7 (C), 134.8 (C), 131.8 (CH), 128.8 (CH), 127.3
(CH), 125.6 (CH), 122.9 (2CH), 122.6 (C), 115.9 (CH), 107.9
(C); IR 3440, 3291, 3174, 1625, 1600, 1476, 1404, 1320, 1225,
754, 724 cm-1; MS (EI) m/ z 306 (M+ + 1), 225 (M+ + 1, - Br),
198.
o
methanesulfonic acid (0.215 g) at 25 C. The salt was collected
and washed with ethyl acetate to give a yellow powder (0.67
1
g, 94%): mp 261-262 °C; H NMR (DMSO-d6) 2.41 (s, 3 H,
CH3). Anal. (C14H14N2S2O3) C, H, N. Similarly prepared were
the following:
2-(4-Am in op h en yl)ben zoth ia zole eth a n esu lfon ic a cid
o
1
sa lt: a yellow solid (90%), mp 211-213 C; H NMR (DMSO-
d6) δ 2.52 (q, 2 H, CH2), 1.10 (t, 3 H, CH3). Anal. (C15H16
N2S2O3) C, H, N.
-
2-(4-Am in o-3-iod op h en yl)ben zoth ia zole (9f). Prepared
from iodination of the amine 5a with iodine monochloride
according to General Method J ; 1H NMR (DMSO-d6) δ 8.28
(d, 1H, J 2.1 Hz, 2′-H), 8.05 (d, 1H, J ) 7.6 Hz, H-7), 7.94 (d,
1H, J 7.7 Hz, H-4), 7.79 (dd, 1H, J 2.1, 8.5 Hz, H-6′), 7.49 (dt,
1H, J 1.2, 7.7 Hz, H-5), 7.38 (dt, 1H, J 1.2, 7.6 Hz, H-6), 6.86
(d, 1H, J 8.5 Hz, H-5′), 5.98 (brs, 2H, NH2); 13C NMR (DMSO-
d6) 166.4 (C), 153.6 (C), 151.8 (C), 137.5 (CH), 134.0 (C), 128.8
(CH), 126.6 (CH), 124.9 (CH), 123.0 (C), 122.20 (CH), 122.18
(CH), 114.0 (CH), 82.6 (C, C-3′); IR 3426, 3286, 3173, 1623,
2-(4-Am in o-3-m eth ylp h en yl)ben zoth ia zole m eth a n e-
1
su lfon ic a cid sa lt: a yellow solid (94%), mp 199-201 °C; H
NMR (DMSO-d6) δ 2.42 (s, 3 H, CH3SO3), 3.30 (s, 3 H, CH3).
Anal. (C15H14N2S2O3) C, H, N.
2-(4-Am in o-3-m et h ylp h en yl)b en zot h ia zole et h a n e-
su lfon ic a cid sa lt: as a yellow solid (93%), mp 206-209 °C;
1H NMR (DMSO-d6) 2.54 (q, 2 H, CH2), 2.33 (s, 3 H, CH3),
1.11 (t, 3 H, CH3) Anal. (C16H16N2S2O3) C, H, N.
In Vitr o Cytotoxicity Assa ys. Stock solutions of drugs
(10 mM) were prepared in DMSO and stored protected from
light at 4 °C for 4 weeks. Compounds tested were stable up
to 50 °C in a pH range of 1-10 (>28 days; data not shown).
Serial dilutions were prepared in medium prior to assay so
that the final concentration of DMSO did not exceed 0.25%.
3T3 a n d ANN-1 Cells. Cells (2 × 104) were grown in
Dulbecco's modified Eagles medium, supplemented with 10%
fetal calf serum. Assays were conducted according to a
published method.18
MCF -7, MDA 468, MCF -7/Ad r , SKBR 3, ZR 75, a n d MDA
231, T47D Hu m a n Br ea st Ca r cin om a Cell Lin es, DU145
a n d P C3 Hu m a n P r osta te Ca r cin om a Cell Lin es. Cells
were cultured in an atmosphere of 5% CO2 in RPMI 1640
medium containing 2 µM L-glutamine supplemented with 10%
foetal bovine serum, 100 IU/mL penicillin, and 100 mg/mL
streptomycin. Cells were routinely subcultured twice weekly
to maintain logarithmic growth. Cells were seeded into 96-
well plates at a density of 2.5-3 × 102 per well and allowed
to adhere for 4 h before drug was introduced. A final
concentration range between 1 pM and 100 µM was achieved
(n ) 8). Cultures were incubated for 10 days (MDA 468) or 7
days (all other cell lines). 3-(4,5-Dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT, final concentration 400
mg/mL) was added to each well. The following 4 h incubation
period allows metabolism of MTT by mitochondrial dehydro-
genases of viable cells to form an insoluble formazan product.
Medium was aspirated and formazan solubilized by addition
of DMSO (100 µL) and glycine buffer (25 µL). Absorbance, as
a measure of viable cell number, was read at 550 nm on an
Anthos Labtec systems plate reader.
1587, 1473, 1435, 1398, 1310, 1225, 973, 874, 753, 723 cm-1
;
MS (EI) m/ z 352 (M+), 260, 225 (M+ - I).
2-(4-Am in o-3-ch lor op h en yl)ben zoth ia zole (9i). Com-
pound 9i was prepared from condensation of 4-amino-3-
chlorobenzonitrile with 2-aminothiophenol in PPA at 230 °C
according to General Method G: 1H NMR (CDCl3) δ 8.07 (d,
1H, J 2.0 Hz, H-2′), 8.05 (d, 1H, J 8.0 Hz, H-7), 7.89 (d, 1H, J
7.9 Hz, H-4), 7.81 (dd, 1H, J 2.0, 8.4 Hz, H-6′), 7.50 (dt, 1H, J
1.3, 7.7 Hz, H-5), 7.37 (dt, 1H, J 1.2, 7.6 Hz, H-6), 6.85 (d, 1H,
J 8.4 Hz, H-5′), 4.43 (brs, 2H, NH2); 13C NMR δ (CDCl3) 167.4
(C), 154.5 (C), 145.8 (C), 135.1 (C), 129.0 (CH), 127.7 (CH),
126.7 (CH), 125.2 (CH), 125.1 (C), 123.1 (CH), 121.9 (CH),
119.7 (C), 115.8 (CH); IR 3451, 3300, 1626, 1475, 1406, 1384,
1224, 754, 723 cm-1; MS (EI) m/ z 260 (M+), 225 (M+ - Cl),
108.
2-[(4-P yr r olid in -1-yl)p h en yl]ben zoth ia zole (15a ). In-
teraction of 2-aminothiophenol (1.5 g, 12.5 mmol) and 4-(pyr-
rolidin-1-yl)benzaldehyde (2.24 g, 12.5 mmol) in DMSO (50
mL) at 180 °C for 15 min gave the pyrrolidinobenzothiazole
15a (2.8 g, 80%) when the mixture was diluted with water:
mp 240-241 °C (EtOH-DMSO) (lit.,8 mp 241-243 °C).
2-[4-(P iper idin -1-yl)ph en yl]ben zoth iazole (15b). A mix-
ture of 2-(4-fluorophenyl)benzothiazole17 (0.8 g, 3.49 mmol),
piperidine (3.0 mL), 50% aqueous KOH (1.5 mL), and DMSO
(5 mL) was heated at 100 °C for 24 h and cooled, and the melt
was poured into water. An ethyl acetate extract furnished 15b
(0.43 g, 42%): mp 171-173 °C (MeOH) (lit.,8 mp 175-176 °C).
2-[4-(Mor p h olin -4-yl)p h en yl]ben zoth ia zole (15c). Simi-
larly prepared, from 2-(4-fluorophenyl)benzothiazole and mor-
pholine, this compound (50% yield) had mp 272-273 °C
(MeOH) (lit.,8 mp 277-278 °C).
MCF -7 Br ea st, WiDr a n d HT29 Colon , A204 Rh a b-
d om yosa r com a , A2780 Ova r ia n , IGR-37 Mela n om a , a n d
T 24 Bla d d er Tu m or Cell Lin es. Cells were seeded at a
density of 104 in Dulbecco’s medium and exposed to drug for
120 h. The end-point was determined by the addition of 100
2-(P yr idin -4-yl)ben zoth iazole (18). From 2-aminothiophe-
nol and 4-formylpyridine, according to the method used to
prepare compound 15a , the pyridylbenzothiazole 18 (85%) had
mp 131-133.5 °C (lit.,8 mp 135-136 °C).

3384 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17
Shi et al.
µL of saline containing 0.002% w/v propidium bromide (Sigma),
0.3% drawing ink (Staedtler marsmatic 745), and Triton
X-100.2 The plates were kept at 4 °C overnight before reading
with an automated scanning stage. Fluorescence intensity was
measured in arbitrary units by a photomultiplier. An HP-87
computer controlled the movement of the stage and collected
and processed the data from the multiplier. For each com-
pound, a dose-response curve was obtained and the IC50 value
(the drug concentration producing 50% inhibition of cell
growth) was calculated.
to the NCI for access to the in vitro cell panel and to
Dr. Kenneth D. Paull for conducting the COMPARE
analyses.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR chemical
shifts, coupling contants, and 13C NMR chemical shifts for 6a -
i, 7a -i, 8b-8k , 9b, 9d -e, 9g, 9j-z, 9a a , and 9bb (4 pages).
Ordering information is given on any current masthead page.
Refer en ces
Agen t Cytotoxicity. This was estimated by measuring the
leakage of lactate dehydrogenase (LDH) from cells damaged
by toxic insult. Cells were seeded into 24-well plates in
medium supplemented with 1% FCS at a density of 5 × 104/
well and allowed 4 h to attach before drug was administered
(final concentration 1 nM-100 µM, n ) 3/control n ) 6).
Following 96 h exposure, medium was collected, centrifuged
to pellet any debris and assayed for LDH activity. Concur-
rently, cells were counted using a haemocytometer. The
oxidation of NADH to NAD+ by LDH was measured spectro-
photometrically by following the decrease in absorption at 340
nM (Leathwood and Plummer, 1969). An amount of 2.4 mL
of PBS, pH 7.4, 0.1 mL of NADH (3.5 mM), and 0.4 mL of
medium sample were added to a cuvette. The assay mixture
was allowed to equilibrate at 37 °C before initiating the
reaction by addition of 0.1 mL of sodium pyruvate solution
(32 mM). The rate of change of absorbance over 5 min was
monitored on a Pye Unicam SP8-400 UV/vis spectrophotom-
eter. Maximal release of LDH, representing 100% cell death,
was determined following lysis of untreated cells in 1%
Triton-X 100. LDH release was measured in untreated cells
to obtain a value representing natural cell death. Agent
cytotoxicity was expressed as % Triton- releasable LDH
activity/105 cells, and the drug concentration which elicited
50% toxicity (LC50 value) calculated.
In Vivo An titu m or Tests. Four human breast carcinomas
xenotransplanted into female Ncr: nu/nu (Taconic, German-
town, USA) or male Bom: NMRI-nu/nu mice (Bomholtgaard,
Ry, Denmark) were used for the evaluation of antitumor
activity. Mice weighing 20-25 g at the start of experiments
were held under sterile conditions at 24-26 °C, 50% relative
humidity, and 12 h light-dark rhythm in laminar flow shelves.
The animals received autoclaved food and bedding (Sniff,
Soest, Germany); the drinking water was filtered and acidified
(pH 4.0).
The following breast carcinoma cell lines were used: BO
(T61, gift of Prof. Fortmeyer, Frankfurt, Germany); MCF-7
(NCI, USA); MT-1 and MT-3.10 The tumors BO and MCF-7
are ER+ models, the carcinomas MT-1 and MT-3 ER- ones.
Tumors were transplanted subcutaneously (sc) as pieces (2 ×
2 mm) into the left flank of 5-8 nude mice/experimental group.
Drug treatment was initiated when the tumors reached a
diameter of 4-5 mm. Compounds were solubilized with Tween
80 (maximum 10% of final volume) and suspended in saline.
Suspensions were prepared freshly for each drug administra-
tion and injected in a volume of 0.2 mL/20 g body weight
employing a once-weekly schedule (× 3). Tumor size was
measured twice weekly with a caliper. Median tumor volume/
group was related to the first treatment day and expressed as
relative tumor volume (RTV). For the estimation of toxicity,
body weight was determined twice weekly and the mean
percentage body weight change (BWC) was calculated. In one
experiment blood was obtained from the retroorbital venous
plexus of mice and blood cells were determined with a Coulter
counter (Model T41).
(1) Part 2 of this series: Stevens, M. F. G.; Shi, D.-F.; Castro, A.
Antitumour benzothiazoles. Part 2. Formation of 2,2′-diamino-
biphenyls from the decomposition of 2-(4-azidophenyl)benzazoles
in trifluoromethanesulfonic acid. J . Chem. Soc., Perkin Trans.
1 1996, 83-93.
(2) Stevens, M. F. G.; McCall, C. J .; Lelieveld, P.; Alexander, P.;
Richter, A.; Davies, D. E. Structural studies on bioactive
compounds. 23. Synthesis of polyhydroxylated 2-phenylben-
zothiazoles and a comparison of their cytotoxicities and phar-
macological properties with genistein and quercetin. J . Med.
Chem. 1994, 37, 1689-1695.
(3) Yates, P. C.; McCall, C. J .; Stevens, M. F. G. Structural studies
on benzothiazoles. Crystal and molecular structure of 5,6-
dimethoxy-2-(4-methoxyphenyl)benzothiazole and molecular or-
bital calculations on related compounds. Tetrahedron 1991, 47,
6493-6502.
(4) Stevens, M. F. G.; McCall, C. J .; Lelieveld, P. Benzazole
compounds for use in therapy. International Publication Number
WO 95/06469; Stevens, M. F. G.; Shi, D-F.; Bradshaw, T. D.;
Wrigley, S. Benzazole compounds. Br. Patent 9503946.7.
(5) Metzger, J . V. In Comprehensive Heterocyclic Chemistry; Katritz-
ky, A. R., Rees, C. W., Eds.; Pergamon Press: Oxford, 1984; Vol.
6, p 322.
(6) Bellamy, F. D.; Ou, K. Selective reduction of aromatic nitro
compounds with stannous chloride in nonacidic and nonaqueous
medium. Tetrahedron Lett. 1984, 25, 839-842.
(7) Hein, D. W.; Alheim, R. J .; Leavitt, J . J . The use of polyphos-
phoric acid in the synthesis of 2-aryl- and 2-aryl-substituted
benzimidazoles, benzoxazoles, and benzothiazoles. J . Am. Chem.
Soc. 1957, 79, 427-429.
(8) Deligeorgiev, T. G. An improved method for the preparation of
2-aryl-, 2-hetaryl-, and 2-styrylbenzothiazoles. Dyes Pigm. 1990,
12, 243-248; Chem. Abstr. 1990, 113, 40539r.
(9) Muathen, H. A. Mild and efficient iodination of aromatic
compounds with pyridine-iodine monochloride complex (PylCl).
J . Chem. Res. (S) 1994, 405.
(10) Naundorf, E. C.; Rewasowa, E. C.; Fichtner, I.; Bu¨ttner, B.;
Becker, M.; Go¨rlich, M. Characterisation of two human mam-
mary carcinomas, MT-1 and MT-3, suitable for in vivo testing
of ether lipids and their derivatives. Breast Cancer Res. Treat-
ment 1992, 23, 87-95.
(11) Cullen, K. J .; Yee, D.; Sly, W. S.; Perdue, J .; Hampton, B.;
Lippman, M. E.; Rosen, N. Insulin-like growth factor receptor
expression and function in human breast cancer. Cancer Res.
1990, 50, 48-53.
(12) Pratt, S. E.; Pollak, M. N. Estrogen and antiestrogen modulation
of MCF-7 human breast cancer cell proliferation is associated
with specific alterations in accumulation of insulin-like growth
factor binding proteins in conditioned media. Cancer Res. 1993,
53, 5193-5198.
(13) Bates, S. E.; Fojo, A. T.; Weinstein, J . N.; Myers, T. G.; Alvarez,
M.; Paull, K. D.; Chabner, B. A. Molecular targets in the
National Cancer Institute drug screen. J . Cancer Res. Clin.
Oncol. 1995, 121, 495-500.
(14) Varma R. S.; Chauhan S. Synthesis of substituted 2-phenylben-
zothiazoles and 5(6)-nitro-1,3-disubstituted-benzimidazoline-2-
thiones as CNS active agents. Indian J . Chem. 1988, 27B, 438-
442.
(15) Tripathy, P. B.; J ena, E. Azomethine dyes. I. J . Inst. Chem.
(Calcutta) 1970, 42, 65-69; Chem. Abstr. 1970, 73, 676459.
(16) Haskell, T. H.; Peterson, F. E.; Watson, D.; Plessas, N. R.;
Culbertson, T. Neuraminidase inhibition and viral chemo-
therapy. J . Med. Chem. 1970, 13, 697-704.
(17) Wattenberg, L. W.; Page, M. A.; Leons, J . L. Induction of
increased benzopyrene hydroxylase activity by 2-phenylben-
zothiazoles and related compounds. Cancer Res. 1968, 28, 2539-
2544.
(18) Cunningham, B. D. M.; Threadgill, M. D.; Groundwater, P. W.;
Dale, I.; Hickman, J . A. Synthesis and biological evaluation of
a series of flavones designed as inhibitors of protein tyrosine
kinases. Anti-Cancer Drug Des. 1992, 7, 365-384.
Ack n ow led gm en t. We thank the Cancer Research
Campaign, U.K., for financial support of this work. We
also thank the Overseas Research Students Awards
Scheme (U.K.) and Aston University for Studentships
(to D.-F. S. and C. J . M., respectively). We are grateful
J M9600959