Y. Lai et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7349–7353
7351
Table 1
O
5'
O
Inhibitory activity of the target compounds against COX-1/COX-2 and 5-LOX in vitro
S
H3C
3'
8.48-8.50
H
O
S
6' 5'
7.94-7.96
O
H
H
CH3O2S
H2NO2S
2'
NOE
Br
H
NOE
H
6'
H
CH3
H
4
4
2'
Br
3'
O
O
6
N
N
6
R
O
R
O
H
7
H
7
N
H
N
H
(E)-10f
(Z)-10f
10a-g
9a-g
Chart 2. Determination of 10f configuration by NOE analysis. Data shown are the
chemical shifts of H-20 and H-60 protons of 10f.
COX-2 S. I.b
a
Compound
R
IC50
(
lM)
COX-1
COX-2
5-LOX
DBFc
TNDd
9a
9b
9c
9d
9e
9f
9g
10a
10b
10c
10d
10e
10f
10g
—
—
H
5-F
5-Cl
5-CH3
6-Cl
5-Br
5-NO2
H
37.60
2.56
>50
0.62
0.16
>50
0.11
8.40
>50
0.18
3.31
>50
0.61
36.70
>50
9.87
3.37
>50
341.82
0.31
n.d.
3.44
0.05
n.d.
the E-isomer, consistent with the previous reports.30,31 In this
regard, the chemical shifts of phenyl H-20 and H-60 protons for
other target compounds (9a–g, 10a–e, and 10g) were in the range
of 7.86–8.03 ppm, which were similar to that of 10f E -isomer
(7.94–7.96 ppm), suggesting that the configurations of other target
compounds were the E-isomer.
>50
>50
>50
n.d.
24.63
0.14
9.32
8.99
>50
20.88
3.72
5.52
>50
3.82
>50
2.39
0.10
3.29
10.55
3.06
9.51
9.42
>50
1.18
0.04
1.69
n.d.
n.d.
n.d.
0.10
1.10
0.15
The target compounds 9a–g and 10a–g were preliminarily eval-
uated in vitro for their ability to inhibit 5-LOX and COX activities
by measuring the formation of LTB4 (a product of 5-LOX) and
12-hydroxyheptadecatrienoic acid (12-HHTrE, a product of COX)
in rabbit peripheral venous blood samples stimulated with the cal-
cium ionophore A23187 alone or in combination with lipopolysac-
charide (LPS).32 Two well-known COX/5-LOX dual inhibitors,
darbufelone mesilate (DBF)33 and tenidap sodium (TND), were
used as reference standards. As shown in Table 1, DBF displayed
potent inhibitory activity against COX-2/5-LOX but relatively weak
against COX-1, while TND inhibited COX-1 activity more potently
than COX-2, which were consistent with previous reports.16,33 Six
target compounds (9b, 9c, 9g, and 10e–g) exhibited potent inhibi-
tory action on these three enzymes, and their anti-COX-1 and anti-
5-LOX activities were more potent than DBF and TND, respectively.
Particularly, 10f showed excellent inhibitory activity against COX-
5-F
5-Cl
5-CH3
6-Cl
5-Br
5-NO2
>50
>50
0.24
0.11
0.50
2.99
0.56
3.91
n.d. not determined.
a
The concentration (lM) of test compound causing 50% inhibition of LTB4 (5-
LOX), 12-HHTrE (after 30 min incubation with A23187 alone) for COX-1, or 12-
HHTrE (after 24 h incubation with A23187 plus LPS) for COX-2, determined by
HPLC. Data are expressed as the mean from the dose–response curves of at least
three independent experiments and intra-group variation was less than 12%.
Vehicle control had no inhibitory effect on these enzymes (data not shown).
b
In vitro COX-2 selectivity index (COX-1 IC50/COX-2 IC50).
Darbufelone mesilate.
Tenidap sodium.
c
d
1
(IC50 = 0.11
(IC50 = 0.56 M), respectively, and its inhibitory activities were
well balanced for these enzymes. Compound 9b also had strong
inhibitory activity against COX-1 (IC50 = 0.62 M) and COX-2
(IC50 = 0.18 M), which had the highest COX-2 selectivity among
lM), COX-2 (IC50 = 0.10 lM), and 5-LOX
l
acetic acid induced writhing assay in mice.35 As shown in Table 3,
treatment with each of the target compounds significantly reduced
the numbers of writhes in mice (P <0.01), indicating that these
compounds displayed strong analgesic activity. Especially, 9c,
10f, and 10g had more potent analgesic activity than TND, while
similar to DBF.
To determine the potential ulcerogenic effect, groups of male
Wistar rats were treated orally with vehicle control or individual
compounds daily for consecutive 7 days, respectively. The total
numbers and areas of gastric ulcers in the stomachs of individual
rats were examined (Table 4). Treatment with the test compounds,
but not the control, did induce various sizes of ulcers in the stom-
achs of these rats. All target compounds tested caused fewer gas-
tric ulceration than TND (P <0.01). Notably, the total areas of
gastric ulcers in the groups of rats received some compounds were
similar to that of the DBF-treated rats, while the total areas of gas-
tric ulcers in the 9b, 9c, 9g and 10f treated rats were significantly
smaller than that of the DBF-treated rats (P <0.05). Given that DBF
had been shown to have less gastric adverse effect in human,33 our
data clearly indicated that these new compounds may be better
gastrically tolerized in human.
l
l
the target compounds tested. These novel compounds with potent
inhibitory activity were worthwhile for further bioactivities evalu-
ation in vivo.
Furthermore, the in vivo anti-inflammation activities of
aminosulfonyl derivatives (9b, 9c, and 9g) and methylsufonyl
derivatives (10a, 10c, and 10e–g) were evaluated using
a
well-known rat model of the carrageenan-induced paw oedema.34
Male Wistar rats were randomized for oral treatment with DBF
(13.0 mg/kg body weight), with an equimolar dose of TND, or
individual compounds (n = 10 per group). One hour after treat-
ment, the rats were injected with 0.1 mL of 1% carrageenan in
saline in right footpad and the volumes of the paw for individual
rats were measured using a plethysmometer at 0, 1, 3, and 6 h
post-carrageenan treatment (Table 2). Injection with carra-
geenan-induced inflammation and increased the volume of the
paw at 1 h post-injection. The volume of the paw further increased
with time. In contrast, treatment with most of the target
compounds significantly mitigated the carrageenan-induced
inflammation in rats. Notably, the inhibitory effects of 9b, 10a,
10e, and 10f were stronger than that of DBF and TND, especially
for 10f. These data clearly demonstrated that these compounds
have strong anti-inflammatory activity in vivo.
Analysis of structure–activity relationship (SAR) revealed that
compounds 10a, 10e, and 10f with a methylsulfonyl substituent
had stronger inhibitory activity against COX-1/2 and 5-LOX than
that of the corresponding sulfamoyl analogs 9a, 9e, and 9f. Further-
more, introduction of methyl (an electron-donating group) into the
C-5 position of the indolin-2-one ring reduced their inhibitory
activities (e.g., 9a vs 9d, 10a vs 10d). In contrast, introduction of
Next, the compounds at an equimolar dose of DBF (26.0 mg/kg
body weight) were evaluated for their analgesic activities using the