1960
K. J. Kim et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1958–1962
Table 1
Inhibition of PGE2 production, in vitro COX-1 and COX-2 inhibition, COX-2 selectivity index (SI), and Eintermolecular data
O
H2N
O
S
O
N
R1
O
R2
R1
R2
Cell viability (
lM)
IC50 (lM)
COX-2 SIc
Eintermolecular (kcal/mol)
COX-1 (1EQG) COX-2 (3LN1)
a
d
Compound
b
PGE2
COX-1
COX-2
7a
7b
7c
7d
9a
9b
9c
H
H
H
H
CH3
CH3
CH3
CH3
H
>10
>10
>10
>10
>10
>10
>10
>10
0.61
0.87
0.45
4.7
127
9.1
1.5
1.2
26
0.036
0.085
0.053
0.006
1.8
0.072
(0.070)g
30
14
À124
À127
À119
À124
À130
À131
À136
À129
À116
À122
À162
À168
À173
À167
À166
À172
À178
À171
À136
À179
CH3
OCH3
Cl
>100
>100
>100
>1.0
>1.0
>1.0
>1.0
220
>67
>83
>3.8
>28
>12
>19
>168
122
>14
(110)g
0.87
H
0.011
CH3
OCH3
Cl
0.0078
0.0031
0.0087
0.0070
0.0087
9d (MPO-0029)
NS-398e
Celecoxibf
>1.0
(7.7)g
26
Ibuprofenh
0.86
À106
À108
a
b
c
IC50 value is the compound concentration required to produce 50% inhibition of PGE2 production, COX-1 and COX-2 enzymatic activity, respectively.
LPS-induced PGE2 production.
In vitro COX-2 selectivity index (COX-1 IC50/COX-2 IC50).
Calculated energy of intermolecular interactions during docking in the COX-1 (PDB code: 1EQG) and COX-2 (PDB code: 3LN1) active site.
Positive control for PGE2 production inhibition.
d
e
f
Positive control as selective COX-2 inhibitor.
g
h
Reported data (Kaur, J. et al. Bioorg. Med. Chem. Lett. 2012, 22, 2154).16
Positive control as non-selective COX-2 inhibitor.
the COX activity assay. All experiments were carried out at least
three times and the data of bioassays were summarized in Table 1.
As shown in Table 1, all compounds except 7d displayed good
inhibitory potencies against LPS-induced PGE2 production in the
submicromolar into nanomolar range. Specifically, 1-methyl-1H-
pyrrole-2,5-dione series 9a–d (IC50 = 3.1–11 nM range) showed
more potent inhibition against PGE2 production than 1H-pyrrole-
assay, meanwhile, celecoxib showed high COX-2 inhibitory po-
tency with IC50 value of 72 nM and high selectivity for COX-2/1
(>14), which are in agreement with previously reported data
(IC50 = 70 nM, COX-2 SI = 110) in the literature.16 The relative pro-
file with respect to the R2-substituent was Cl (9d: IC50 = 6.0 nM,
COX-2 SI = >168) >> H (9a: IC50 = 36 nM, COX-2 SI = >28) > OCH3
(9c: IC50 = 53 nM, COX-2 SI = >19) > CH3 (9b: IC50 = 85 nM, COX-2
SI = >12) as shown in Figure 2(b and c). This is somewhat in
contrast to PGE2-related data wherein compound 9c (PGE2
IC50 = 3.1 nM, COX-2 IC50 = 53 nM) showed more inhibitory effect
on LPS-induced PGE2 production than compound 9d (PGE2
IC50 = 8.7 nM, COX-2 IC50 = 6.0 nM). As multiple enzymes are in-
volved in PGE2 biosynthesis, including phospholipases, cyclooxy-
genases (COX), 5-lipoxygenase, and the PGE2 synthases (PGES),17
this present results suggest that some of these 1H-pyrrole-2,5-
dione derivatives could inhibit the activity of these enzymes based
on the recent literatures.18–20 Nonetheless, both R1 = CH3 and
R2 = Cl groups were shown to be important pharmacophores to
confer COX-2 selectivity and potency. In the viewpoint of overall
2,5-dione series 7a–d (IC50 = 0.45–4.7 lM range), meaning that
R1 substituent (that is, N-CH3) was a major determinant of PGE2
production inhibitory potency. The relative profile with respect
to R2 substituent was found to be OCH3 (7c) > H (7a) > CH3
(7b) >> Cl (7d) in 7 series (R1 = H), whereas the relative profile of
R2 substituent was OCH3 (9c) > CH3 (9b) P Cl (9d) > H (9a) within
9 series (R1 = CH3). Taken together, 9c (R1 = CH3 and R2 = OCH3;
IC50 = 3.1 nM) was most strong and ca. 2.8-fold more potent than
celecoxib (IC50 = 8.7 nM) for LPS-induced PGE2 production inhibi-
tion in RAW 264.7 cells. Compound 9d (IC50 = 8.7 nM) was found
to be almost comparable to celecoxib as shown in Figure 2(a).
Therefore, this finding implies that N-CH3 (R1) group was shown
to be important pharmacophore to confer PGE2 production inhibi-
tory potency again.
respects (PGE2 IC50
, COX-2 IC50, and COX-2 SI), therefore,
compound 9d (MPO-0029) was identified to be more potent and
selective COX-2 inhibitor than celecoxib.
In vitro COX-1/COX-2 isozyme inhibition studies on compound
7 series (R1 = H) showed that all compounds were more potent
To gain insight into the plausible mode of interaction(s) of syn-
thetic compounds within the COX-1 and COX-2 isozymes, molecu-
lar modeling (docking) experiments were performed using X-ray
crystal structure data for COX-1 (PDB code: 1EQG)21 and COX-2
(PDB code: 3LN1)22 using Molegro Virtual Docker (MVD)
2013.6.0 for Windows.23 The energy associated with intermolecu-
lar interactions (Eintermolecular) obtained upon computational analy-
sis (docking) for compounds within the COX-1 and COX-2 active
site is summarized in Table 1. The docking results of NS-398, cele-
coxib, and ibuprofen were also inserted for comparison. First of all,
all compounds occupy the same binding site as that of celecoxib in
COX-2 structure. The obvious interaction differences between
COX-1 and COX-2 structures appropriately support the experimen-
tal high selectivity indexes for COX-2/COX-1 in Table 1, when
inhibitors of COX-2 (IC50 = 1.2–26
(IC50 = >100
M). The relative profile with respect to the R2-substi-
tuent was OCH3 (7c) > CH3 (7b) > H (7a) >> Cl (7d). Among this ser-
ies, compound 7c (PGE2 IC50 = 0.45 M, COX-2 IC50 = 1.2 M, and
COX-2 selectivity index = >83) was identified as potent and selec-
tive COX-2 inhibitor compared to ibuprofen (COX-2 IC50 = 30 M,
lM
range) than COX-1
l
l
l
l
COX-2 selectivity index = 0.87). On the other hand, compound 9
series (R1 = CH3) displayed very high inhibitory potency towards
COX-2 (IC50 = 6.0–85 nM range) but they exhibited little inhibition
against COX-1 at even higher 1.0 lM concentration, thus the high-
er COX-2 selectivity. This structure–activity data acquired showed
that the N-CH3 (R1) group was a determinant of both COX-2 inhib-
itory potency and COX-2 selectivity. In our enzyme inhibitory