D. A. Sabbah et al. / Bioorg. Med. Chem. Lett. 25 (2015) 3120–3124
3121
R
O
O
Cl
O
O
O
a
+
2
OH
OH
R
3
R
4
R
(3a-f)
O
1
1, R = H
2, R = OCH3
4, R = H, 65%
7, R = (m-F), 3-F, 52%
8, R = (p-F), 4-F, 76%
9, R = (p-Cl), 4-Cl, 71%
5, R = (p-NO2), 4-NO2, 73%
6, R = (o-F), 2-F, 60%
Figure 1. Chemical structures of 1 and 2.
H3CO
O
Cl
OCH3
O
O
O
a
+
OH
H3CO
2
H3CO
O
F
F
(3e)
10, 66%
Scheme 1. Condition: (a) THF, Et3N, 50 °C, 24 h.
Table 1
Growth inhibition IC50 (mM) of HCT116 cell line after
24 h exposure time
Compounds
HCT-116
1
2
4
5
6
7
8
9
10
2.12
1.17
2.66
0.88
0.91
0.97
1.29
0.73
2.82
Figure 2. PI3K
a
pharmacophore model with 1 (C atoms in yellow) and 2 (C atoms
Standard deviation (SD) never exceeded 5%.
in blue). Aro stands for aromatic rings; Acc for H-bond acceptor; and Hyd for
hydrophobic groups. Picture made by MOE.38
harboring two methoxy groups at p-position is not preferred due
to steric clash as well as electron-donating effect. Surprisingly,
the parent compound 2 showed similar potency to that of 8. This
implies that the steric effect (size, shape, and volume) of the veri-
fied molecules determines their potency. Also, the activity of 1, 4,
and 10 confirms that both the steric and substituent effects should
be balanced to attain higher potency. The fact that HCT116 con-
The target products 4–9 were prepared by reacting benzoin (1)
with substituted benzoyl chlorides (3a–f), in a one pot reaction, in
a dry aprotic polar solvent (tetrahydrofuran) under heating condi-
tion via
a nucleophilic substitution mechanism (Scheme 1).
Triethylamine (Et3N) was added to the reaction mixture as a cata-
lyst and an acid scavenger; in a similar manner, 4-fluorobenzoyl
chloride (3e) was reacted with p-anisoin (2) to produce 10. The
chemical structure of the prepared compounds was determined
using spectroscopic approaches; FT-IR and NMR (1H and 13C).
In order to examine their antiproliferative activities, we
assessed the growth inhibition in the colon cancer cell line
(HCT-116) induced by 1, 2 and 4–10 (Table 1).
tains both alleles (wild-type and mutant H1047R PI3Ka) has hin-
dered our efforts to extrapolate more reasons for our results. Our
future research will aim to knockout either allele to harvest only
one specific gene in HCT116 cell line.
In order to determine possible pathways responsible for the
observed growth inhibition, caspase-3 enzyme activity was mea-
sured via the caspase-3 colorimetric assay. Activation of the cas-
HCT116 is a highly malignant colon carcinoma cell line harbor-
ing both wild-type and mutant (H1047R) PI3K
a
that was created
pase-3 pathway is a
hallmark of apoptosis40,41 and can be
from a primary tumor tissue culture.39
utilized in cellular assays to appraise activators and inhibitors of
the ‘death cascade’. To determine the apoptotic effects of the
examined compounds on HCT116 cells, cells were incubated with
The tested compounds inhibited PI3Ka activity in HCT-116 cell
line. Interestingly, 5, 6, 7, and 9 exerted a relatively high potency in
HCT116 cell line (IC50 <1 mM). Additionally, 2 and 8 with IC50s 1.17
and 1.29 mM showed comparable potency. A two-fold increase in
IC50s suggests a weak inhibitor such as 1, 4, and 10.
700 lM of each compound for either 24 or 48 h and changes in cas-
pase-3 enzyme activities were analyzed. The results demonstrate
that there is a significant increase in caspase-3 activity in response
to 24 h treatment for each compound. Indeed, lengthening of the
treatment period results in further increase in caspase-3 activity
relative to the untreated cells control (Fig. 3).
Interestingly, similar pattern of efficacy was exhibited in the
apoptosis assay for p- (5, 8, and 9) as well as o- and m-substituents
(6 and 7), and the parent structure 2. The consistency of the results
of both bioassays provides a clue for the significance of this scaffold
as a potential antitumor agent.
This result suggests that derivative with p-substituent is more
favored for –Cl and –NO2 functionalities. This might infer that
the size and electron-withdrawing effects of the p-substituent
are significant to attain potent inhibition. The size of –Cl and
–NO2 is bigger compared to that of –F (H isostere). The electron-
withdrawing effect of –NO2 and –F is higher than that of –Cl.
Substitution is favored for –F at o- or m-position. This implies that
the size and electron-withdrawing effects of substituents at
p-position should be considered to induce higher potency.
Compound 10 having p-methoxy groups on benzoin scaffold
showed to be the least potent compound. This indicates that
In order to investigate the structural basis of binding of the
verified inhibitors in the active domain of PI3K
a, we carried out
docking studies employing Glide42,43 dock approach against the