Biological evaluation of polyhalo 1,3-diazaheterocycle fused isoquinolin-1(2H)-imine derivatives

Article abstract of DOI:10.1016/j.ejmech.2011.01.036

A series of polyhalo 1,3-diazaheterocycle fused isoquinolin-1(2H)-imines were evaluated in vitro against human tumour cell lines including A431, K562, HL60, HepG2 and Skov-3. As a result, some of the target compounds such as 5b, 5c, 5i, 5o, 6c, 6h and 7f

Full text of DOI:10.1016/j.ejmech.2011.01.036

European Journal of Medicinal Chemistry 46 (2011) 1172e1180
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Biological evaluation of polyhalo 1,3-diazaheterocycle fused
isoquinolin-1(2H)-imine derivatives
,
,
,
**
Chao Huang a 1, Sheng-Jiao Yan a 1, Xiang-Hui Zeng a, Xiao-Yang Dai b, Yin Zhang b, Chen Qing b
,
,
Jun Lin a
*
a Key Laboratory of Medicinal Chemistry for Natural Resource, Ministry of Education, School of Chemical Science and Technology, Yunnan University, Kunming 650091, PR China
b Yunnan Key Laboratory of Pharmacology for Natural Products, Kunming Medical University, Kunming 650031, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of polyhalo 1,3-diazaheterocycle fused isoquinolin-1(2H)-imines were evaluated in vitro against
human tumour cell lines including A431, K562, HL60, HepG2 and Skov-3. As a result, some of the target
compounds such as 5b, 5c, 5i, 5o, 6c, 6h and 7f showed stronger cytotoxicity against K562, H562 and
Skov-3 cells in comparison with cisplatin, and the others displayed moderate cytotoxicity to A431 and
HepG2. Biological investigations using the representative compounds 5c, 6c and 6h were also performed
in mice bearing S180 and H22 tumours. The results indicated that these three compounds inhibit S180 and
H22 growth. In addition, compounds 6c and 6h have very low acute toxicities. The preliminary analysis of
structureeactivity relationships is also discussed.
Received 16 August 2010
Received in revised form
7 October 2010
Accepted 24 January 2011
Available online 31 January 2011
Keywords:
Isoquinolin-1(2H)-imine
Polyhalo heterocyclic ketene aminals
Anti-tumour
Ó 2011 Elsevier Masson SAS. All rights reserved.
Acute toxicity
1. Introduction
anti-anxiety agents [14,15], anti-leishmanial agents [16] and anti-
bacterial drugs [17].
Isoquinolin-1(2H)-imine derivatives exhibit various biological
and medicinal activities, inhibiting acetylcholinesterase (AChE) and
butyryl-cholinester-ase (BChE) [1], and having antimalarial [2],
anti-inflammatory, anti-nociceptive [3], antitumour, antiviral,
antibacterial and antifungal properties [4]. However, the efficient
methods for the production of isoquinolin-1(2H)-imines are very
rare [5,6]. Therefore, the development of effective methodologies
for constructing molecular libraries of isoquinolinimine with great
potential for drug discovery would be desirable.
Polyhalo isophthalonitriles, especially polyfluoro-isophthalonitrile,
have beenwidely usedasanti-cancer[7e9], antiinflammatory [10]and
insecticidal [11] agents, and so on [12] in pharmaceuticals. Fluorine-
containing compounds often show improved biological activity
profiles being their superior metabolic stability and the lipophilicity of
fluorine. In this context, F-containing compounds have been attracting
considerable attention in the field of pharmaceutical chemistry [13].
Heterocyclic ketene aminals (HKAs) are versatile substrates for
the synthesis of fused heterocyclic compounds with a wide variety
of uses, including as anti-cancer agents [7], herbicides, pesticides,
Our previous studies [7] have also shown that the polyhalo
heterocyclic ketene aminals (polyhalo-HKAs), particularly tri-
fluoro-HKAs, possess strong cytotoxicity against human carcinoma
cells in vitro. In this regard, we envisage that the polyhalo iso-
quinolinimines starting from the easily available polyhalo iso-
phthalonitriles and HKAs would also possess potential bioactivity.
In the present study, a series of isoquinolin-1(2H)-imines
bearing multiple halogen atoms in the benzene ring and different
types of substituted acyl groups at position-4 of the isoquinolin-1
(2H)-imine ring were synthesized. The in vitro and in vivo anti-
tumour activities of isoquinolin-1(2H)-imines and their acute
toxicity were further evaluated, and a preliminary study on the
structureeactivity relationship was also carried out.
2. Results and discussion
2.1. Chemistry
The synthetic routes of compounds 5e7 are outlined in Scheme 1.
A mixture composed of a 1:1.1 M ratio of HKAs 1e2 or N,O-acetals 3
to polyhalo isophthalonitrile 4 was treated under microwave irra-
diation (MW) in solvent-free conditions (120 ꢀC for 12 min with
a maximum power of 200 W). Then the mixture was directly mixed
* Corresponding author. Tel./fax: þ86 871 5033215.
** Corresponding author.
E-mail address: linjun@ynu.edu.cn (J. Lin).
1
These authors contributed equally to this paper.
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.01.036

C. Huang et al. / European Journal of Medicinal Chemistry 46 (2011) 1172e1180
1173
O
R
O
N
CN
º
Y
X
NH
Z
R
i) MW: 120 C, 12 min
X
X
Y
n
X
Z
Solvent-free
+
5: Z=NH, n=0
6: Z=NH, n=1
7: Z=O, n=0
H
NC
ii) t-BuOK/dioxane
NC
n
1: Z=NH, n=0
2: Z=NH, n=1
3: Z=O, n=0
X
rt, 30min
NH
X, Y= F or Cl
4
Scheme 1. Synthetic route of novel polyhalo 1,3-diazaheterocycle fused isoquinolin-1(2H)-imine derivatives 5e7.
with 1.1 equiv. of t-BuOK at room temperature for 30 min. Iso-
quinolin-1(2H)-imines 5e7 were isolated in excellent yields of
79e96%.
The structures of the target compounds i.e. 5e7 were charac-
terized by IR, 1H NMR, 13C NMR, 19F NMR and the high resolution
mass spectrometry (HRMS) [18]. Their physical properties are
summarized in Table 2.
However, cytotoxicity against A431, Skov-3 and Hep-2 was similar
to DDP in most cases. Typically, 5b (IC50 0.0008 g/mL), 5c (IC50
0.0005 g/mL) and 7f (IC50 0.0002 g/mL) were more than 1250
times more active than DDP (IC50 g/mL) against K562 cells. The
activity of 5b (IC50 0.003 g/mL), 5i (IC50 0.001 g/mL) and 7f (IC50
0.008 g/mL) was more than 62 times higher than that of DDP (IC50
0.5 g/mL) against HL60 cells. Compounds 5b (IC50 0.05 g/mL), 5c
(IC50 0.04 g/mL) and 5o (IC50 0.02 g/mL) were more than 78
times more active than DDP (IC50 3.9 g/mL) against Skov-3 cells.
m
m
m
1 m
m
m
m
m
m
m
m
m
2.2. X-ray analysis of compound 5c
The different ring-sizes of the isoquinolin-1(2H)-imines (n ¼ 0,1)
appeared to influence bioactivity. For instance, the bioactivity of
five-membered isoquinolin-1(2H)-imines (n ¼ 0) was extremely
different from that of the six-membered isoquinolin-1(2H)-imines
(n ¼ 1) under the experimental conditions (Table 2, entries 5e7 vs.
17e19; 8e10 vs. 23e25; 11e13 vs. 26e28). These results also suggest
that the isostere of N and O (Z]NH, O) does not play an important
role in the modulation of cytotoxic activity to different tumour cell
lines. Consequently, imidazo-[1,2-b]isoquinolin-imines 5ae5o and
oxazo-[1,2-b]isoquinolin-imines 7ae7f showed slight different
activity (Table 2, entries 5e10 vs. 29e34).
The three-dimensional structure of compound 5c was further
determined by X-ray crystallography (as shown in Fig. 1, the Cam-
bridge crystallographic data centre (CCDC) 736001 [19]). The
results showed that there are two kinds of intramolecular hydrogen
bonds: the hydrogen bond between the carbonyl group and the
secondary amine on the imidazole ring N3eH3/O1 (Table 1, entry
4), and the hydrogen bond N4eH4/F1 (Table 1, entry 1). In addi-
tion, intermolecular hydrogen bonds such as N(4)eH(4)/O(2)
(Table 1, entry 2) and N3eH3/O3 (Table 1, entry 3) were also
observed. A summary of hydrogen bonding geometry is given in
Table 1.
2.4. Structureeactivity relationship (SAR)
2.3. In vitro evaluation of cytotoxicity
The cytotoxicity data suggest that the fluorine atom in the title
compounds plays an important role in promoting biological
activity. Moreover, cytotoxic activity increased with an increase in
the number of fluorine atoms (Table 2, entry 4 vs. 3 vs. 2). In
particular, 5a, 5b, and 5c showed a different IC50 range against the
The cytotoxicity of individual compounds against human
myeloid leukaemia cells (HL60 and K562), human epidermoid
carcinoma cells (A431), human ovarian carcinoma cells (Skov-3)
and human laryngeal carcinoma cells (Hep-2) was evaluated
tumour cell lines Hep-2 (IC50 38.2e0.4
mg/mL), Skov-3 (IC50
through
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
50.1e0.04 g/mL) and K562 (IC50 1.6e0.0005
m
mg/mL). This is
bromide (MTT) colourimetric assay [20,21], using cisplatin (DDP) as
the positive control. The IC50 values are presented in Table 2 (IC50
value, defined as the concentration corresponding to 50% growth
inhibition). The majority of compounds 5e7 displayed superior
cytotoxicity compared to DDP against the HL60 and K562 cells.
presumably because the fluorine atoms decrease the Gibbs Energy
(Table 2, entries 3e4 vs. 2; 15e16 vs. 14; 3e4 vs. 5e34). The lower
Gibbs Energy maybe contribute to facile formation of the intra-
molecular and intermolecular hydrogen bonds (Fig. 1 and Table 1).
For example, 5m, 6a and 6d maybe cannot to form intramolecular
hydrogen bonds NeH/Cl and showed little cytotoxicity. In addi-
tion, this kind of intramolecular hydrogen bond i.e. NeH/F would
make a significant positive contribution to cytotoxicity (Table 2,
entries 3e4 vs. 2; 6e7 vs. 5; 9e10 vs. 8; 12e13 vs. 11; 15e16 vs. 14).
Lipophilicity is the most important aspect for drugs to pass
through the bloodebrain barrier. Lipophilicity is described by log P
values. The predicted log P values of the isoquinolin-1(2H)-imine
derivatives 5e7 were within the range of 1.03e4.84. We found that
Table 1
Hydrogen bonding geometry (Å, ꢀ).
No.
DeH/A
d(DeH)
d(H/A)
d(D/A)
<(DHA)
1
2
3
4
5
6
7
N(4)eH(4)/F(1)
N(4)eH(4)/O(2)i
N(3)eH(3)/O(3)ii
N(3)eH(3)/O(1)
N(8)eH(8)/F(4)
N(7)eH(7)/O(1)iii
N(7)eH(7)/O(3)
0.87 (2)
0.87 (2)
0.88 (2)
0.88 (2)
0.88 (2)
0.89 (2)
0.89 (2)
2.19 (2)
2.49 (2)
2.17 (2)
2.22 (2)
2.13 (2)
2.06 (2)
2.28 (2)
2.820 (2)
3.280 (2)
2.977 (2)
2.787 (2)
2.807 (2)
2.885 (2)
2.808 (2)
129.3 (18)
151.5 (18)
152.0 (19)
121.8 (17)
133.5 (18)
153.0 (20)
118.0 (18)
Fig. 1. X-ray crystal structure of 5c.
Symmetry codes: (i) ꢁx þ 1, ꢁy, ꢁz þ 1; (ii) x ꢁ 1, y, z; (iii) x þ 1, y, z.

1174
C. Huang et al. / European Journal of Medicinal Chemistry 46 (2011) 1172e1180
Table 2
Isoquinolin-1(2H)-imines and their physicochemical properties and biological activitiesa (IC50 g/mLb).
, m
No.
1
1e3
4
Compound
Gibbs Energyc
[kJ/mol]
log P
Anti-cancer activity IC50 (mg/mL)
A431
HL60
0.5
HepG2
0.06
Skov-3
3.9
K562
1
DDP
1.1
O
OEt
H
Cl
Cl
CN
Cl
O
H
N
Cl
Cl
Cl
N
OEt
N
N
NC
2
450.2
3.44
40.6
1.7
38.2
50.1
1.6
1a
NC
H
4a
Cl NH
5a yield= 91%
O
OEt
H
Cl
F
CN
F
O
H
N
F
F
F
N
OEt
OEt
N
NH
N
NC
3
4
84.44
2.68
2.04
2.2
3.4
0.003
0.02
1.5
0.4
0.05
0.04
0.0008
0.0005
1a
NC
Cl
H
4b
5b yield= 94%
O
OEt
H
F
CN
F
O
H
N
F
F
F
F
N
N
N
NC
ꢁ98.44
1a
NC
H
4c
F
NH
5c yield= 92%
OMe
CN
Cl
O
O
O
O
H
N
Cl
H
N
Cl
Cl
Cl
OMe
Cl
N
5
6
7
671.53
305.77
122.89
4.02
3.26
2.62
2.1
1.9
7.5
0.3
0.2
0.5
11.3
9.2
4.8
2.4
0.7
0.3
0.4
NC
N
1b
1b
1b
NC
H
4a
Cl NH
5d yield= 79%
OMe
OMe
CN
F
O
H
N
Cl
H
F
F
OMe
OMe
F
N
N
1.2
NC
Cl
N
NH
NC
H
4b
F
5e yield= 84%
CN
F
O
H
N
F
H
F
F
F
F
N
N
0.9
NC
N
NH
NC
H
4c
F
5f yield= 89%
O
CN
Cl
O
Cl
H
N
H
Cl
Cl
Cl
Cl
N
N
N
8
9
777.74
411.98
4.03
3.27
3.4
4.8
0.8
0.1
4.4
0.3
19.7
0.7
0.4
NC
1c
NC
H
4a
Cl NH
5g yield= 89%
O
CN
F
O
Cl
H
H
N
F
F
F
N
N
N
5.8
NC
Cl
1c
NC
H
4b
F
NH
5h yield= 90%
O
CN
F
O
F
H
N
H
F
F
F
F
N
N
N
NH
10
229.1
2.63
0.8
0.001
1.2
0.2
0.08
NC
1c
NC
H
4c
F
5i yield= 94%

C. Huang et al. / European Journal of Medicinal Chemistry 46 (2011) 1172e1180
1175
Table 2 (continued ).
No.
1e3
4
Compound
Gibbs Energyc
[kJ/mol]
log P
Anti-cancer activity IC50 (mg/mL)
A431
HL60
HepG2
Skov-3
31.4
K562
Cl
Cl
Cl
CN
Cl
O
O
O
O
H
Cl
H
N
Cl
Cl
Cl
Cl
Cl
Cl
N
Cl
N
11
756.18
390.42
207.54
4.76
20.3
2.6
37.6
2.4
1.4
1
NC
N
1d
1d
1d
NC
H
4a
Cl NH
5j yield= 80%
CN
F
O
H
N
Cl
H
F
F
F
N
N
12
4.00
5
0.3
12.4
34.9
NC
Cl
N
NH
NC
H
4b
F
5k yield= 84%
CN
F
O
H
N
F
H
F
F
F
F
N
N
13
3.36
4.9
0.5
17.7
34.5
NC
N
NH
NC
H
4c
F
5l yield= 91%
O
Cl
CN
Cl
H
O
H
N
Cl
Cl
Cl
Cl
N
N
N
NC
14
15
16
623.23
257.47
74.59
2.43
1.67
1.03
308
4.7
503
6.4
3.1
78.4
3.6
0.4
0.1
NC
1e
H
4a
Cl NH
yield= 81%
5m
O
Cl
H
CN
F
O
H
N
F
F
F
N
N
N
NC
20.4
0.9
6.2
NC
Cl
1e
H
4b
F
NH
yield= 83%
5n
O
F
CN
F
H
O
H
N
F
F
F
F
N
N
N
NC
4.4
0.02
0.02
NC
1e
H
4c
F
NH
yield= 90%
5o
OMe
CN
Cl
O
Cl
O
H
N
Cl
Cl
Cl
NH
NH
OMe
OMe
Cl
17
667.85
4.10
56.9
4.4
400
67.2
1.4
NC
N
2a
NC
4a
Cl NH
6a yield= 83%
OMe
OMe
CN
F
O
Cl
H
O
O
F
F
NH
NH
F
N
18
302.09
3.34
2
0.5
0.8
26.8
0.07
NC
Cl
N
NH
2a
NC
4b
F
6b yield= 85%
CN
F
O
F
H
F
F
F
NH
NH
OMe
F
N
19
119.21
2.70
1.1
0.02
0.8
2.6
0.01
NC
N
NH
2a
NC
4c
F
6c yield= 92%
(continued on next page)

1176
C. Huang et al. / European Journal of Medicinal Chemistry 46 (2011) 1172e1180
Table 2 (continued ).
No.
1e3
4
Compound
Gibbs Energyc
[kJ/mol]
log P
Anti-cancer activity IC50 (mg/mL)
A431
HL60
HepG2
Skov-3
826
K562
9.3
Me
Me
Me
CN
Cl
O
Cl
O
O
O
H
N
Cl
Cl
Cl
NH
Me
Me
Me
Cl
20
772.85
407.09
224.21
4.61
>1000
58.2
277
NC
N
NH
2b
2b
2b
NC
4a
Cl NH
yield= 84%
6d
CN
F
O
Cl
H
F
F
NH
NH
F
N
N
21
3.85
3.21
27.5
14.6
24.6
76.4
33.9
2.3
0.7
NC
Cl
NC
4b
F
NH
yield= 83%
6e
CN
F
O
F
H
F
F
F
NH
NH
F
N
N
22
2.1
4.7
2.4
NC
NC
4c
F
NH
yield= 90%
6f
O
CN
Cl
Cl
H
N
O
Cl
Cl
Cl
Cl
NC
NH
N
23
24
25
774.06
4.11
3.35
2.71
10.1
10.7
0.01
0.3
9.3
0.3
1.1
51.2
1.2
NC
NH 2c
4a
Cl NH
6g yield= 82%
O
CN
F
Cl
F
H
N
O
F
F
F
NH
N
NH
408.3
0.5
1.2
0.03
0.02
NC
Cl
NH 2c
NC
4b
yield= 85%
6h
O
CN
F
F
F
H
N
O
F
F
F
F
NH
N
NH
225.42
2.4
16.4
NC
NH 2c
NC
4c
yield= 92%
6i
Cl
Cl
Cl
CN
Cl
O
Cl
O
H
N
Cl
Cl
Cl
NH
Cl
Cl
Cl
Cl
NC
26
27
28
752.5
4.84
4.08
3.44
8
4.4
0.2
4.7
30
52.2
17.1
15.1
1.2
0.3
0.5
NC
N
NH
2d
2d
2d
4a
Cl NH
6j yield= 82%
CN
F
O
Cl
H
O
O
F
F
NH
NH
F
N
386.74
203.86
2.8
4.3
2.1
8.4
NC
Cl
N
NH
NC
4b
F
yield= 87%
6k
CN
F
O
F
H
F
F
F
NH
NH
F
N
N
NC
NC
4c
F
NH
yield= 87%
6l

C. Huang et al. / European Journal of Medicinal Chemistry 46 (2011) 1172e1180
1177
Table 2 (continued ).
No.
1e3
4
Compound
Gibbs Energyc
[kJ/mol]
log P
Anti-cancer activity IC50 (mg/mL)
A431
HL60
HepG2
Skov-3
32.2
K562
OMe
OMe
OMe
CN
Cl
O
Cl
O
O
O
H
N
O
Cl
Cl
Cl
OMe
OMe
OMe
Cl
O
29
375.53
131.94
ꢁ50.94
3.39
11
24.2
2.7
0.8
0.5
1.8
NC
N
3a
3a
3a
NC
4a
Cl NH
7a yield= 86%
CN
F
O
Cl
H
N
O
F
F
F
O
30
3.21
2.1
1
2
23.8
NC
Cl
N
NH
NC
4b
F
7b yield= 92%
CN
F
O
F
H
N
O
F
F
F
F
O
31
2.57
1.5
0.1
0.5
0.2
NC
N
NH
NC
4c
F
7c yield= 96%
O
CN
Cl
Cl
O
H
N
O
Cl
Cl
Cl
Cl
O
N
32
603.91
3.97
14.5
6
0.3
17.4
2.2
NC
NC
3b
4a
Cl NH
yield= 84%
7d
O
CN
F
Cl
O
H
N
O
F
F
F
O
N
NH
33
238.15
3.21
2.3
1.5
3.6
29.2
1.7
NC
Cl
NC
3b
4b
F
yield= 88%
7e
O
F
CN
F
O
H
N
O
F
F
F
F
O
N
NH
34
55.27
2.57
0.3
0.008
0.1
0.2
0.0002
NC
NC
3b
4c
F
yield= 94%
7f
a
Cytotoxicity as IC50 for each cell line, refers to the concentration of compound which reduced by 50% the optical density of treated cells with respect to untreated cells
using the MTT assay.
b
Data represent the mean values of three independent determinations.
Gibbs Energy and log P values of the compounds were calculated by ChemDraw Ultra 11.0.
c
the chlorine atom had a stronger impact on the lipophilicity than
the fluorine atom. For example, the lipophilicity of 5a (log P: 3.44)
was higher than that of 5b (log P: 2.68) and 5c (log P: 2.04). As
a result, cytotoxicity decreased as lipophilicity increasing. Indeed,
isoquinolin-1(2H)-imines with log P values of 2.0e2.7, such as 5b,
5c, 5i, 6c, and 7f, showed better activity, and in particular, those
compounds containing three fluorine atoms showed excellent
cytotoxic activity (IC50 < 3.4 mg/mL) against the five human cancer
cell lines. On the contrary, 6a and 6d (log P: 4.10e4.61) with
three chlorine atoms each displayed weaker cytotoxic activity
(IC50 > 1.4 mg/mL).
differences in steric hindrance and Gibbs Energy. Consequently, the
cytotoxicity of various groups at position-4 was ranked as follows:
ester > acetyl > benzoyl > p-methoxy-benzoyl > p-chlorine-
benzoyl > p-methyl-benzoyl (Scheme 2).
In short, the number of fluorine atoms and intramolecular
hydrogen bonds between the imine and the group at position-4 has
a positive influence on the cytotoxic properties of isoquinolin-1
(2H)-imine. For example, the compounds with two or three fluorine
atoms and an ester or benzoyl group at position-4 (5b, 5c, 5i and 7f)
showed better cytotoxicity. In contrast, the compounds contain-
ing three chlorine atoms and an acetyl, p-methoxy-benzoyl or
p-methyl-benzoyl group (5m, 6a and 6d) showed lower cytotox-
icity against human cancer cells. Interestingly, compound 5c
Isoquinolin-1(2H)-imines 5e7 bearing the same substituents
exhibited different levels of cytotoxic activity, probably due to

1178
C. Huang et al. / European Journal of Medicinal Chemistry 46 (2011) 1172e1180
Table 4
Steric hindrance and Gibbs Energy:
The effect of 5c, 6c and 6h on H22 growth.
R= COOEt> MeCO > PhCO>
p-MeOPhCO> p-ClPhCO> p-MePhCO
Compounds
Dose
(mg/kg)
Tumour
weight (g)
(X ꢂ SD)
Inhibitory
rate %
X, Y = F, F >F, Cl >Cl, Cl
Z= NH ≈ O
Y
R
X
Z
Solvent (CMC-Na)
Equal
volume
20
1.34 ꢂ 0.33
e
N
n
NC
Positive
(cyclophosphamide)
0.59 ꢂ 0.30
56.24
X
N
H
5 circle ≈ 6 circle
Best log P: 2.0-2.7
5c
5c
5c
6c
6c
6c
6h
6h
6h
10
30
90
35
70
140
20
80
1.19 ꢂ 0.43
0.97 ꢂ 0.68
0.91 ꢂ 0.42
0.87 ꢂ 0.45
0.73 ꢂ 0.35
0.95 ꢂ 0.47
1.18 ꢂ 0.52
0.96 ꢂ 0.39
1.34 ꢂ 0.69
11.44
27.52
31.90
34.74
45.61
29.29
12.10
28.19
0.00
F> Cl intramolecular hydrogen bonds
Scheme 2. Structureeactivity relationship of isoquinolin-1(2H)-imines.
possessed a broad spectrum of cytotoxicity against the five human
tumour cell lines (IC50 0.0005e3.4 g/mL), being 25e2000-fold
240
m
more active than DDP except against the A431 and HepG2 cell lines.
As a consequence, the compounds containing a framework with
three fluorine atoms on the ring of isophthalonitrile and an electron
withdrawing group (COOEt) at position-4 of isoquinolin-imine
would be promising leads for further structural modifications
according to the valuable information originating from the detailed
SARs.
at dosages of 20, 80 and 240 mg/kg, the inhibitory rates were
12.10%, 28.19% and 0%, respectively (Table 4). These results also
show that compounds 5c, 6c and 6h have an inhibitory effect on
H22 growth, but with lower activity in comparison with that against
S180, and that there is no obvious dose-dependent relationship.
2.5. In vivo anti-tumour activity
2.6. Preliminary acute toxicity test
To evaluate the in vivo anti-tumour activities of the represen-
tative compounds 5c, 6c and 6h, their effects on the tumour
weights of S180 and H22 mice were determined.
Oral administration of compounds 6c and 6h at the dosages of
5000 mg/kg did not induce death in the mice tested, but compound
5c at the dosage of 4500 mg/kg did result in death (4/6). This
indicates that 6c and 6h would be lower acute toxicity than that
of 5c.
2.5.1. The effect of 5c, 6c and 6h on S180 growth
Following daily oral administration of compound 5c at dosages
of 10, 30 and 90 mg/kg for 10 days, S180 growth inhibition rates
were 39.25%, 54.17% and 26.62%, respectively. For compounds 6c
and 6h at dosages of 20, 60 and 180 mg/kg, the S180 growth inhi-
bition rates were 35.85%, 38.04%, 19.13% and 42.76%, 41.09%, 52.79%,
respectively (Table 3). These results indicate that compounds 5c, 6c
and 6h have an inhibitory effect on S180 growth without an obvious
dose-dependent relationship.
3. Conclusions
In summary, a series of polyhalo 1,3-diazaheterocycle fused
isoquinolin-1(2H)-imines were evaluated for their cytotoxicity
against A431, K562, HL60, HepG2 and Skov-3 carcinoma cells in
vitro. Some of the target compounds such as 5b, 5c, 5i, 5o, 6c, 6h
and 7f showed stronger cytotoxicity against K562, H562 and Skov-3
cells than DDP, and the others displayed moderate cytotoxicity to
the tumour cells A431 and HepG2. The preliminary structur-
eeactivity relationship analysis revealed that the presence of
fluorine atoms and intramolecular hydrogen bonds between imine
and fluorine could play an important role in enhancing biological
activity. The representative compounds 5c, 6c and 6h showed good
anti-tumour activity in vivo the S180 and H22 mice models, but the
potency depended on tumour type. In this regard, S180 was more
sensitive than H22. In addition, compounds 6c and 6h exhibited
better levels of safety. These results direct the development of
effective drugs for pharmaceutical application.
2.5.2. The effect of 5c, 6c and 6h on H22 growth
Similarly, following daily oral administration of compound 5c at
dosages of 10, 30 and 90 mg/kg for 10 days, H22 growth inhibition
rates were 11.44%, 27.52% and 31.90%, respectively. For compound
6c at dosages of 35, 70 and 140 mg/kg, the inhibitory rates were
34.74%, 45.61% and 29.29%, respectively (Table 4). For compound 6h
Table 3
The effect of 5c, 6c and 6h on S180 growth.
Compounds
Dose
(mg/kg)
Tumour
Inhibitory
rate %
weight (g)
(X ꢂ SD)
4. Experimental section
Solvent (CMC-Na)
Equal
volume
20
2.89 ꢂ 0.62
e
4.1. Chemistry
Positive
(cyclophosphamide)
0.53 ꢂ 0.52
81.72
4.1.1. General methods
5c
5c
5c
6c
6c
6c
6h
6h
6h
10
30
90
20
60
180
20
60
1.76 ꢂ 0.54
1.33 ꢂ 1.25
2.12 ꢂ 0.60
1.86 ꢂ 1.46
1.79 ꢂ 1.65
2.34 ꢂ 1.38
1.66 ꢂ 1.29
1.70 ꢂ 1.09
1.37 ꢂ 1.09
39.25
54.17
26.62
35.85
38.04
19.13
42.76
41.09
52.79
All compounds were fully characterized by spectroscopic tech-
niques. The NMR spectra were recorded on a Bruker DRX500
spectrometer (1H: 500 MHz, 13C: 125 MHz, 19F: 470 MHz) with
tetramethylsilane (TMS) as the internal standard (
chemical shifts ( ) are expressed in ppm, and J values are given in
Hz. Deuterated DMSO-d6 was used as a solvent. IR spectra were
recorded on a FT-IR Thermo Nicolet Avatar 360 using a KBr pellet.
The reactions were monitored by thin layer chromatography (TLC)
d 0.0 ppm),
d
180

C. Huang et al. / European Journal of Medicinal Chemistry 46 (2011) 1172e1180
1179
using silica gel GF254. The melting points were determined on an
XT-4A melting point apparatus and are uncorrected. HRMS was
performed on an Agilent LC-MSD TOF instrument.
data can be obtained free of charge from The Cambridge Crystal-
lographic Data Center via www.ccdc.cam.ac.uk/datarequest/cif.
All chemicals and solvents were used as received without
further purification unless otherwise stated. Column chromatog-
raphy was performed on silica gel (200e300 mesh).
4.2. In vitro cytotoxic activity [24]
The materials 1be1e and 2 were synthesized according to the
literature [22]. Compounds 1a and 3 were prepared based on
published procedures [23]. 4ae4c were purchased from Aldrich.
Growth inhibition by the sample of tumour cells was measured
by microculture tetrazolium (MTT) assay. Myeloid leukaemia (HL60
and K562), epidermoid carcinoma (A431), ovarian carcinoma
(Skov-3), laryngeal carcinoma (Hep-2) cells were seeded into 96-
well microculture plates, for the adherent cells, the cell were
allowed culture 24 h for adhesion before drug addition, while
suspended cells were seeded just before drug addition. The cell
densities were selected based on the results of preliminary tests, in
order to maintain the control cells in an exponential phase of
growth during the period of the experiment and to obtain a linear
relationship between the optical density and the number of viable
cells. Each tumour cell line was exposed to sample at 0.01, 0.1,1.0,10
4.1.2. General procedure for the synthesis of polyhalo 1,3-
diazaheterocyclic fused [1,2-b]iso-quinolin-1(2H)-imine 5e7
A dry mortar was charged with HKAs 1, 2, 3 (1 mmol) and
polyhalo isophthalonitrile 4 (1.1 mmol). The mixture was mixed at
room temperature by vigorously grinding with a pestle for a few
minutes (ca.1e2 min), then placed in a microwave tube and irra-
diated in a microwave reactor (Discover) with control of power and
temperature by infrared detection at 120 ꢀC for 12 min (maximum
power 200 W). After cooling, the resulting mixture was transferred
to a 50 mL flask, and dissolved in 25 mL 1,4-dioxane, before adding
t-BuOK (1.5 mmol). While stirring at room temperature, the
reaction was monitored by TLC. Upon completion, the resultant was
poured into 60 mL of water and then filtered to obtain the crude
products, which were purified by column chromatography (petro-
l:ethyl acetate ¼ 1:3, v/v) on silica-gel to give the desired products
5e7.
and 100
72 h, suspended cells 48 h) and each concentration was tested in
triplicate. At the end of exposure, 20 L of 5 g per litre MTT was
mg/mL concentrations for different periods (adherent cells
m
added to each well and the plates were incubated for 4 h at 37 ꢀC.
Then triplex solution (10% SDSe5% isobutanole0.012 M HCl) was
added and the plates were incubated for 12e20 h at 37 ꢀC. The
optical density (OD) was read on a plate reader at 570 nm. Media
and DMSO control wells, in which sample was absent, were
included in all the experiments, in order to eliminate the influence
of DMSO. The inhibitory rate of cell proliferation was calculated by
the following formula:
4.1.2.1. Ethyl 5-imino-6,8,9-trichloro-7-cyano-1,2,3,5-tetrahydroimidazo
[1,2-b]isoquinoline-10-carboxylate (5a). Yellow solid (0.351 g, 91%); Mp
234e236 ꢀC; IR (KBr) (nmax, cmꢁ1) 3360, 2223, 1670, 1304, 1158, 1037,
Growth inhibition (%) ¼ (ODcontrol ꢁ ODtreated/ODcontrol) ꢃ 100%.
The cytotoxicity of sample on tumour cells was expressed as IC50
values (the drug concentration reducing by 50% the absorbance in
treated cells, with respect to untreated cells), which were calcu-
lated by LOGIT method.
803, 641; 1H NMR (500 MHz, DMSO-d6)
d 9.30 (br, 1H, NH), 8.30
(br,1H, NH), 4.16 (q, J ¼ 7.1 Hz, 2H, OCH2), 3.99 (t, J ¼ 8.8 Hz, 2H, NCH2),
3.76 (t, J ¼ 8.7 Hz, 2H, NCH2), 1.21 (t, J ¼ 7.0 Hz, 3H, CH3); 13C NMR
(125 MHz, DMSO-d6)
d 165.2, 156.3, 151.6, 141.9, 137.1, 133.8, 125.8,
117.8,114.3,107.5, 80.5, 59.7, 45.2, 43.8,14.1; HRMS (TOF ES-): m/z calcd
for C15H10Cl3N4Oꢁ2 [Mꢁ], 382.9875; found, 382.9876.
4.1.2.2. 6-Imino-7,9,10-trichloro-11-(4-methoxybenz-oyl)-2,3,4,6-
tetrahydro-1H-pyrimido[1,2b]-isoquinol-ine-8-carbonitrile
4.3. In vivo anti-tumour activity [25]
(6a). Yellow solid (0.383 g, 83%); Mp 204e206 ꢀC; IR (KBr) (nmax
,
In vivo anti-tumour activity was evaluated by using a model of
transplant sarcoma (S180) and hepatoma 22 (H22) in KM mice, and
measuring the index of tumour growth inhibitory rate (%). Briefly,
the S180 and H22 tumour cells were harvested at the exponential
growth phase and inoculated subcutaneously into the right flank
axilla, then the inoculated mice were randomly divided into groups
to receive the solvent control, positive control or one of three
different dosages of each compound (5c, 6c, 6h), with 10 mice per
group. The test compounds were dissolved in 0.5% CMC-Na, and the
positive drug cyclophosphamide (CTX) was dissolved in normal
saline (N.S). The positive control was injected intraperitoneally
with CTX (20 mg/kg) and the solvent control was administered
intragastrically with an equal volume of 0.5% CMC-Na daily for 10
days. The different compounds were applied at different dosages
based on the results of preliminary tests and acute toxicity tests
(data not shown). In the S180 model, 20, 60 and 180 mg/kg were
chosen as the dosages of compounds 6c and 6h, and in the H22
model, the dosages of 35, 70,140 mg/kg and 20, 80, 240 mg/kg were
chosen, respectively. The dosage of 6c was 10, 30 and 90 mg/kg in
both models. These three compounds were administered intra-
gastrically daily for 10 days. At the end of the experiment, the mice
were sacrificed and each tumour was weighed. The tumour inhi-
bition rate (%) was calculated using the following formula:
cmꢁ1) 3436, 2226, 1596, 1261, 1162, 1029, 842, 604; 1H NMR
(500 MHz, DMSO-d6) 10.14 (br, 1H, NH), 9.85 (br, 1H, NH), 7.41 (br,
d
2H, ArH), 6.83 (d, J ¼ 7.4 Hz, 2H, ArH), 3.88 (br, 2H, NCH2), 3.77 (s,
3H, OCH3), 3.45 (br s, 2H, NCH2), 2.08 (br s, 2H, CH2); 13C NMR
(125 MHz, DMSO-d6)
d 188.7, 162.1, 154.7, 153.5, 142.2, 137.0, 134.7,
133.5, 130.5, 125.9, 118.7, 114.8, 113.8, 107.5, 89.1, 55.7, 42.9, 38.0,
19.9; HRMS (TOF ES-): m/z calcd for C21H14Cl3N4Oꢁ2 [Mꢁ], 459.0188;
found, 459.0185.
4.1.2.3. 6,8,9-Trichloro-5-imino-10-(4-methoxybenz-oyl)-3,5-dihy-
dro-2H-oxazolo[3,2-b]isoquin-oline-7-carbonitrile (7a). Yellow solid
(0.386 g, 86%); Mp 287e289 ꢀC; IR (KBr) (nmax, cmꢁ1) 3428, 3261,
2237, 1618, 1440, 1251, 974, 780; 1H NMR (500 MHz, DMSO-d6)
d
10.29 (br, 1H, NH), 7.10 (d, J ¼ 8.3 Hz, 2H, ArH), 6.79 (d, J ¼ 8.4 Hz,
2H, ArH), 4.55 (t, J ¼ 8.5 Hz, 2H, OCH2), 3.89 (br s, 2H, NCH2), 3.73
(s, 3H, OCH3); 13C NMR (125 MHz, DMSO-d6)
186.8, 167.7, 161.4,
d
148.9, 140.8, 138.7, 136.3, 133.8, 130.1, 117.9, 114.8, 114.7, 114.1, 113.7,
87.9, 69.3, 56.0, 44.6; HRMS (TOF ES-): m/z calcd for C20H11Cl3N3Oꢁ3
[Mꢁ], 445.9871; found, 445.9870.
4.1.3. X-ray crystallography for compound 5c
Data collection: X-AREA; cell refinement: X-AREA; data reduc-
tion: X-RED32; program used to solve structure: SHELXS97;
program used to refine structure: SHELXL97. CCDC 736001 contains
the supplementary crystallographic data for compound 5c. These
(Average of tumour weight in the control group ꢁ average of
tumour weight in the drug-treated group)/average of tumour
weight in the control group ꢃ 100%.

1180
C. Huang et al. / European Journal of Medicinal Chemistry 46 (2011) 1172e1180
[5] L.A. Burke, R.N. Butler, J. Org. Chem. 74 (2009) 5199e5210.
4.4. Preliminary acute toxicity test
[6] R. Glaser, J. Yin, S. Miller, J. Org. Chem. 75 (2010) 1132e1142.
[7] S.J. Yan, C. Huang, X.H. Zeng, R. Huang, J. Lin, Bioorg. Med. Chem. Lett. 20
(2010) 48e51.
[8] S.J. Yan, H. Zhen, C. Huang, Y.Y. Yan, J. Lin, Bioorg. Med. Chem. Lett. 20 (2010)
4432e4435.
[9] S.J. Yan, Y.J. Liu, Y.L. Chen, L. Liu, J. Lin, Bioorg. Med. Chem. Lett. 20 (2010)
5225e5228.
[10] W.P. Heilman, R.D. Battershell, W.J. Pyne, P.H. Goble, T.A. Magee,
R.J. Matthews, J. Med. Chem. 21 (1978) 906e913.
[11] J. Lin, S.J. Yan, L.J. Yang, J.F. Li, F.C. Liu, Chin. J. Org. Chem. 25 (2005) 304e307.
[12] C. Huang, S.J. Yan, Y.M. Li, R. Huang, J. Lin, Bioorg. Med. Chem. Lett. 20 (2010)
4665e4669.
[13] W.K. Hagmann, J. Med. Chem. 51 (2008) 4359e4369.
[14] B.E. Maryanoff, S.O. Nortey, J.J. McNally, P.J. Sanfilippo, D.F. McComsey,
B. Dubinsky, R.P. Shank, A.B. Reitz, Bioorg. Med. Chem. Lett. 9 (1999) 1547e1552.
[15] H. Kondo, M. Taguchi, Y. Inoue, F. Sakamoto, G. Tsukamoto, J. Med. Chem. 33
(1990) 2012e2015.
Kunming mice (18e22 g) were divided into three groups for
treatment with the three different compounds (5c, 6c and 6h). The
mice were orally administered with 5000 mg/kg in 24 h, and were
then monitored continuously for up to 7 days to observe any
abnormal behaviour or death. If death occurred, the dosage was
decreased and the test repeated.
Acknowledgements
The work was supported by the National Natural Science Foun-
dation of China (Nos. 30860342, 20762013) and the Natural Science
Foundation of Yunnan Province (Nos. 2009CC017, 2008CD063).
[16] S.N. Suryawanshi, S. Pandey, Rashmirathi, B.A. Bhatt, S. Gupta, Eur. J. Med.
Chem. 42 (2007) 511e516.
[17] M.M. Abdelhalim, M.M.T. El-Saidi, S.T. Rabie, G.A. Elmegeed, Steroids 72
(2007) 459e465.
Appendix. Supplementary material
[18] S.J. Yan, C. Huang, C.X. Su, Y.F. Ni, J. Lin, J. Comb. Chem. 12 (2010) 91e94.
[19] CCDC 736001 contains the supplementary crystallographic data for compound
5c. These data can be obtained free of charge from The Cambridge Crystal-
lographic Data Center via www.ccdc.cam.ac.uk/data_request/cif.
[20] J. Carmichael, W.G. DeGraff, A.F. Gazdar, J.D. Minna, J.B. Mitchell, Cancer Res.
47 (1987) 936e942.
Supplementary data related to this article can be found online at
doi:10.1016/j.ejmech.2011.01.036.
References
[21] T. Mossman, J. Immunol. Methods 65 (1983) 55e63.
[22] Z.T. Huang, M.X. Wang, Synthesis 12 (1992) 1273e1276.
[23] Z.T. Huang, P.C. Zhang, Chem. Ber. 122 (1989) 2011e2016.
[24] Y.G. Chen, J.J. Xu, H. Yu, C. Qing, Y.L. Zhang, L.Q. Wang, Y. Liu, J.H. Wang, Food
Chem. 107 (2008) 169e173.
[1] M. Decker, J. Med. Chem. 49 (2006) 5411e5413.
[2] M. Calas, M. Ouattara, G. Piquet, Z. Ziora, Y. Bordat, M.L. Ancelin, R. Escale,
H. Vial, J. Med. Chem. 50 (2007) 6307e6315.
[3] G.A. Elmegeed, A.R. Baiuomy, O.M. Abdel-Salam, Eur. J. Med. Chem. 42 (2007)
1285e1292.
[25] S.Y. Xu, R.L. Bian, X. Chen, Experimental Methods in Pharmacology. People’s
Medical Publishing House, Beijing, 2002, pp. 1757e1784.
[4] A.M.S. El-Sharief, Z. Moussa, Eur. J. Med. Chem. 44 (2009) 4315e4334.