Edson dos A. dos Santos et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4669–4673
4671
Table 1
Compound effects on tubulin polymerization, the binding of colchicine to tubulin, and the growth of MCF-7 human breast cancer cells
Inhibition of binding of
Tubulin polymerization
IC50 (lM) SD
colchicine, % inhibition SD
d
MCF-7 IC50
(lM) SD
c
5
l
M inhibitor
1
lM inhibitor
(CA-4)a 5-G = cis-CH@CH, R1 = R2 = R3 = R5 = OCH3, R4 = H, R6 = OH
(1)b 5-G = S, R1 = R2 = R3 = R5 = OCH3, R4 = R6 = H
(2) 5-G = S, R1 = R2 = R3 = R5 = OCH3, R4 = H, R6 = NO2
(3) 5-G = S, R1 = R2 = R3 = R5 = OCH3, R4 = H, R6 = NH2
(4) 5-G = Se, R1 = R2 = R3 = R5 = OCH3, R4 = R6 = H
(8) 5-G = S2, R1-R6 = OCH3
1.1 0.1
1.2 0.1
2.8 0.3
0.74 0.04
0.62 0.08
5.4 0.4
>20
99 0.06
89
58 0.9
95 0.1
90
56
-
74 0.5
74
—
1
7
0.006 0.003
0.016 0.005
1
0.008 0.003
0.010 0e
>10
1
0e
94
9.0
—
1
1
1
(12) 5-G = Se, R1 = R3 = R4 = R6 = H, R2 = R5 = OCH3
(13) 5-G = Se2, R1 = R3 = R4 = R6 = H, R2 = R5 = OCH3
—
—
>10
>10
1.7 0.06
34 0.2
a
b
c
IC50 for inhibition of tubulin assembly and % inhibition of binding of colchicine for CA-4 were obtained contemporaneously with the values for all compounds except 1.
IC50 for inhibition of tubulin assembly and % inhibition of colchicine binding as reported in the literature.11
IC50 is the concentration inhibiting the extent of tubulin polymerization by 50% after 20 min at 30 °C.
IC50 is the concentration inhibiting 50% of cell growth after a 96 h incubation at 37 °C.
d
e
SD = 0 indicates that there were obtained the same value was obtained in all experiments.
compounds is dominated by interactions with hydrophobic residues
in the receptor pocket. The hydrophobic collapse that occurs between
the small molecules and the receptor is then reinforced by the
formation of key hydrogen bonds. Overall, compounds 1, 3, 4,
and CA-4 occupy steric space in the receptor that is similar to that
of colchicine (i.e., as it binds in co-crystal 1SA0). Specifically: (1) all
of the compounds’ trimethoxphenyl (i.e., A ring) components and
substituents are located in close proximity, with the oxygen atoms
of the para-methoxy substituents of 1, 3, 4, CA-4, and colchicine all
engaging in a hydrogen bond with the side chain thiol of receptor
site residue b-Cys 241 (Fig. 3), and (2) all of the compounds engage
in a hydrogen bond with the backbone amide nitrogen of receptor
Table 2
Cytotoxic activity against human cancer cell lines 786, HT-29 and PC-3
a
Compounds
IC50
(lM
SD)
786 (kidney)
HT-29 (colon)
PC-3 (prostate)
2
3
4
8
1.71 0.29
1.24 0.16
0.68 0.09
111.67 26.30
N.T.b
20.69 4.49
5.0 0.2
0.33 0.03
0.63 0.02
1.31 0.33
0.28 0.08
76.54 19.05
210.42 36.21
59.51 10.29
1.0 0.1
0.20 0.26
0.53 0.13
0.08 0.003
6.75 0.89
8.80 1.70
1.99 0.37
1.0 0d
12
13
CA-4
DOXc
0.44 0.03
0.63 0.15
a
IC50 is the concentration inhibiting 50% of cell growth after a 48 h incubation at
37 °C.
site residue a-Val 181. For 1, 3, and 4, the hydrogen bond is med-
iated by the oxygen atom of the B ring para-methoxy substituent,
for CA-4 it is mediated by the oxygen atom of the ring B meta-hy-
droxyl substituent, and for colchicine it is mediated by the ring B
carbonyl oxygen (Fig. 3).
b
N.T.—not tested.
Doxorubicin—positive control, as a known cytotoxic agent.
SD = 0 indicates that there were obtained the same value was obtained in all
c
d
experiments.
However, there are subtle differences in the compounds binding
modes that provide hypotheses for rationalizing the % colchicine
binding site inhibition data shown in Table 1. For example, com-
pared with the binding modes of 3 and CA-4, the binding mode
of 1 lacks a hydrogen bond donor substituent required to mimic
the hydrogen bond formed between the ring B meta-aniline and -
hydroxyl subsitutents of 3 and CA-4, respectively, and the back-
and the absence of a hydrophilic group at position C30 in ring B.23
The selenium derivative 4, a close analog of the active sulfide 1,
was the most active of all tested samples including CA-4 as an
inhibitor of tubulin polymerization. As an inhibitor of colchicine
binding, however, compound 4 was slightly less active than CA-4
and equivalent to compound 3.
bone amide carbonyl of receptor site residue
a-Thr179. Hence, as
To evaluate atomic-level contacts between the inhibitors and
colchicine binding site residues that might rationalize differences
in the potencies of 1, 3, 4, and CA-4, molecular docking studies
were performed. To initiate the studies, a previously reported
pharmacophore for colchicine site inhibitors24 was used to identify
key features shared by 1, 3, 4, CA-4, and colchicine (Fig. 2).
Following, common pharmacophore features were used to ori-
ent the compounds in the colchicine receptor pocket using their
best pharmacophoric alignments with the conformation of colchi-
cine as it co-crystallizes with tubulin (PDB code 1SAO25). Next, 1, 3,
4, and CA-4 were energy refined in the colchicine binding site using
constrained minimizations. The refined compound binding modes
were then evaluated using the program Hydropathic INTeractions
(HINT),24 which scores both favorable and unfavorable inter-
molecular, atom–atom contacts. The results from the HINT analy-
ses were used to guide manual adjustment (rotational, torsional,
and translational) to provide hydropathically feasible binding
models for all compounds. Figure 3 shows the superimposed bind-
ing modes of 1, 3, and 4, and CA-4; colchicine is also included for
reference purposes. In general, colchicine site binding for all
shown in Table 1, 1 would be expected to be a weaker colchicine
binding site inhibitor than 3 and CA-4. In another example, a
structure-based hypothesis rationalizing the lower potency of
Figure 2. Pharmacophore requirements for colchicine site binding.24 Orange and
pink spheres indicate hydrophobic points; red spheres indicate polar points. Cyan
and green indicate key hydrophobic-aromatic components that form the core
scaffold for colchicine site binding. Aromatic rings are labeled A and B to provide
points of reference to the structures in Table 1 and Figure 1.