S. Sharma et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4729–4734
4731
M)
Table 1
IC50 values of different compounds against human cancer cell lines
Table 2
HDAC inhibition by different molecules
Compd
Cell type
HCT-116 MIAPaCa-2
IC50 values ( M)
7.2
Molecule
Concn (
lM)
% HDAC inhibition
IC50
34
(l
HL-60
A549
MCF-7
fR-2
7a
50
30
54
46
1
0
l
7c
7d
8a
50
30
56
50
1
2
30
3
4
10
6.1
13.2
6
8.7
15.3
19
6.9
14
8.7
13.8
>20
17
>20
15.1
<0.25
2
3.6
4.4
>20
6.4
7.1
13.9
7.1
15.1
11.4
>20
>20
>20
>20
15.1
8
50
30
45
34
1
0
>50
6*
5a
5b
5c
5d
6a
6b
6c
6d
7a
7b
7c
7d
8a
8b
8c
8d
9a
9b
9c
9d
10a
10b
10c
10d
17.1
>20
>20
>20
17.6
>20
>20
10
11.7
>20
>20
12.4
1.6
>20
>20
>20
>20
7.4
15.1
2
>20
>20
>20
>20
>20
>20
>20
17.2
>20
>20
>20
>20
1.9
>20
>20
>20
>20
15.1
15.7
18
>20
>20
>20
11.8
15.1
>20
5.2
50
30
20
10
1
88
86
70
65
23
0
1
1
0
2
10
6.1
8.8
8b
8c
8d
50
30
57
37
1
2
42
>20
>20
>20
15
11.8
>20
11.9
15.1
13.1
8.7
>20
11
>20
>20
15.9
6
50
30
39
12
0
0
>50
48
13
50
30
52
31
0
0
6.6
3
3.1
6.9
>20
>20
>20
>20
>20
6.6
19.1
2.9
>20
>20
4
7.2
4.3
>20
13.4
>20
>20
>20
>20
>20
>20
7.5
12
11
>20
>20
>20
>20
>20
>20
>20
>20
16.6
*
For determination of IC50 values, lower concentrations were used.
>20
>20
>20
>20
7.1
>20
>20
>20
14.3
>20
>20
16.5
11.1
18
that of during G2/M and during S-phase DNA content is intermedi-
ate between G0/G1 and G2/M phases. Using fluorescent DNA inter-
calating dye propidium iodide as a probe, the percentage of cells in
different phases of cell cycle can be detected on flow cytometer.
14.1
8.6
>20
>20
15.1
Molecule 8a which showed least IC50 of 6
causes 50% HL-60 cell death at 1.6 M and prominent G1 arrest
at 1 M concentration in HL-60 cell line. Interestingly, at concen-
trations of above 1 M, that is, 5 and 10 M more prominent cell
death was observed (Figs. 2a and b).
lM against HDACs and
Note: HL-60 = leukemia; A549 = lung; HCT-116 = pancreas; MIAPaCa-2 = pancreas;
MCF-7 = breast; fR-2 = normal epithelial cells.
l
l
l
l
most of the cell lines in the panel. Among the carboxylic acid based
molecules, 9a–d having b-boswellic acid based cap were again
found to be inactive against most of the cell lines in the panel
whereas, 10a–d with 11-keto-b-boswellic acid based cap were
found to be significantly cytotoxic against some of the cell lines
in the panel (please see Table S1 Supporting info.) but all molecules
of 10a–d series did not exhibit significant cytotoxicity against nor-
mal cell line fR-2.
It is well known that the mitochondria plays a dual role in cel-
lular physiology; it helps in keeping cell alive by playing pivotal
role in aerobic metabolism of the cell and is also involved in apop-
totic cell death under such demanding conditions.13 During apop-
totic cell-death, loss of membrane potential across the inner
mitochondrial membrane may occur and Rh-123 dye is usually
used to measure this.14 The fluorescence level of Rh-123 decreases
with increase in loss of membrane potential. We observed that 8a
induced significant loss in mitochondrial membrane potential at 5
Based on preliminary screening lead molecules were chosen for
determination of IC50 values against the cell lines. Molecule 8a was
and 10 lM concentrations (Fig. 3). In order to find, whether HDAC
found to be most cytotoxic against HL-60 (IC50 = 1.6
was found to be significantly toxic against normal cancer cell line,
fR-2 (IC50 = 1.9 M) and had less cytotoxicity against A549, HCT-
116, MIAPaCa-2 and MCF-7 cell lines having IC50 values of 13,
15.1, 15.1 and 7.1 M, respectively.
The most specific and potent molecules, that is, 7a, 7c–d and
8a–d having hydroxamic acids attached to their linker group were
selected for enzyme based HDAC inhibition evaluation. The 7a,
l
M). Also, 8a
specific inhibition by 8a has any direct role in G1 arrest, cell death
and loss in mitochondrial membrane potential as observed in the
present study, another molecule, that is, 7c was used. Molecule
7c which was found to be inactive against HL-60 cells and showed
l
l
IC50 of 34
age of population in G1 phase and caused only 4.5% and 36% apop-
tosis at 5 and 10 M concentrations as against 92.2% and 97.5% at
lM against HDACs did not cause any change in percent-
l
similar concentrations (Figs. 2a and b). 7c also causes significantly
lesser loss of mitochondrial membrane potential as compared to 8a
at the corresponding concentrations (Fig. 3). Altogether, this sug-
gests that the potent HDAC inhibition as induced by 8a is respon-
sible for cancer cell death.
In order to predict the binding conformation and mode of inhi-
bition of most potent analogs 7c and 8a, molecular modeling stud-
ies of b-boswellic acid derivatives carried out with HDAC8–SAHA
complex retrieved from Protein data bank (PDB ID: 1T69)15 where
in this complex, protein is homotrimeric, consisting three identical
chains (377 residues) with zinc as cofactor and SAHA molecule as
ligand. Therefore, grid file for docking was constructed considering
SAHA ligand as centroid of grid box which has been replaced by
compound 7c and 8a during the docking studies. Considering both
ligand and protein flexible in nature, induce fit docking method has
been applied under the default conditions16 (for details please see
Supporting info.). Results demonstrates that zinc binding group,
7c–d exhibited IC50 values at concentrations 34, 30 and >50
Molecules 8a–d showed IC50 values at concentrations 6, 42, >50
and 48 M, respectively (Table 2). From the structure and activity
lM.
l
data (Table S2, Supporting info.), it becomes clear that glutaric acid
based 3C long chain (present in 7a and 8a) is the linker that can
enhance the anticancer potency of these molecules; however the
molecule with 11-keto-b-boswellic acid derived cap (present in
8a) exhibit better in vitro activity. Compound 3 which does not
contain enzyme inhibitor group as well as characteristic linker
did not show significant HDAC inhibition even at 50 lM, also the
compound 6a and 10c having carboxylic ester and acid respec-
tively as a substitute of hydroxamic acid results in significant loss
of activity (IC50 > 50 lM).
Cancer cells are characterized by high growth rate owing to
deregulations in the cell cycle. DNA content of cells varies in differ-
ent phases of cell cycle. DNA content during G0/G1 phase is half