736
L. Li et al. / Bioorg. Med. Chem. Lett. 21 (2011) 734–737
Table 2
The binding activities of 6a against a full panel of prostanoid receptors in absence of
HSA
N
O
S
S
N
O
Receptors EP4
EP1
EP2
EP3 CRTH2 DP FP
IP
TP
Kia (nM)
0.48 >9500 3330 530 >7230 73 >8700 >7700 61
c, d, e
a, b
a
Average of two or more assays.
O
S
O
S
O
Na
O
O
HO
and 17b, were prepared. We found that they both were beneficial
in maintaining potency on the EP4 receptor and lowering protein
shift. A dramatic improvement in plasma protein shift was ob-
served with p-toluenesulfonylurea by comparing compounds 4a
with 9 (from 632-fold to 40-fold) and 4b with 6a (from 22-fold
to 2.6-fold), respectively. The improvement in intrinsic potency
brought by this acid surrogate is approximately a twofold. SAR
modifications on the 2-position of the benzothiophene moiety of
compound 6a were also explored. Based on assay results obtained
with 6a, 11a–c, we concluded that (1) a small alkyl substituent on
the amide nitrogen was beneficial for intrinsic potency; (2) a ter-
tiary amide is optimal for minimizing the protein shift; (3) the
N,N-dimethylaminocarbonyl group in 6a is the optimized
substituent.
Our further SAR exploration was focused on the substituents on
the phenyl ring of the benzothiophene pharmacophore. It was
found that, in comparison to the 5-Me/7-Me in 6a, other substitu-
ent combination such as of 5-Cl/7-Cl in 12a, 5-F/7-F in 12b, 5-CF3/
7-Cl in compounds 12a–c had little effect on intrinsic potency but
all had detrimental effect on protein shift. The same phenomena
was observed with the change of substituent pattern from 5,7-di-
methyl in 6a to 4,7-dimethyl in 12d. One interesting discovery dur-
ing our SAR study is that intermediate 3f was active in binding
assay which suggests that an acidic group is not a requirement
for the EP4 antagonizing activity in the series. Thus we devoted
some limited efforts to further investigate this and identified that
amides 7a–d and compound 8 showed reasonable intrinsic affini-
ties. The best amide analog was 7a, which had a Ki of 0.2 and
3.4 nM in the absence and presence of HSA, respectively.
All the EP4 ligands reported herein were determined to be full
EP4 antagonist when evaluated in human HEK-293 cell based func-
tional assay. For example, compounds 6a and 7a has an IC50 = 1.2
and 2.4 nM, respectively, in the absence of HSA. They were also
evaluated against other prostanoid receptors and showed similar
selectivity profile. To illustrate this, the binding activities of 6a
against all prostanoid receptors are summarized in Table 2.20 The
selectivities of compound 6a for EP4 over the DP (152 folds) and
TP (127 folds) receptors are moderate but acceptable, good over
EP3 (>1000 folds) receptor and excellent over the remaining
prostanoid receptors (>6900 folds).
In summary, we have disclosed our preliminary SAR study on
the identification of substituted benzothiophene derivatives as
PGE2 subtype 4 receptor antagonists. Highly potent EP4 antagonist
6a (Ki = 1.4 nM with 10% HSA) was identified. Furthermore, it was
also found that an acidic group is not essential for the EP4 binding
activity in the series and non-acidic analogs were identified. This
opens a new direction for future EP4 antagonist design.
O
13
O
3l
14
O
N
N
O
S
S
O
f
H
N
Ar
Ar
H2N
O
OH
S
S
O
O
O
O
O
15
16a
16b
17a: Ar = 2-MeOC6H4
17b: Ar = 2,6-Cl2C6H3
Scheme 6. Reagents and conditions: (a) PPh3, Br2, NEt3, DCM; (b) 13, DMF, 85 °C;
(c) MeONa, MeOH, reflux.; (d) NCS, DMF; (e) NH3 (gas), THF; (f) EDCI, DMAP, 16a
and 16b, THF.
Table 1
EP4 receptor binding affinitiesb
Entry
Compound
EP4 Kia (nM)
0% HSA
10% HSA
Shift (folds)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
4a
4b
4c
4d
4e
9
6a
6b
6c
17a
17b
11a
11b
11c
12a
12b
12c
12d
3f
7a
7b
7c
7d
8
1.9
1.1
160
25
10
0.97
0.54
0.4
0.42
0.33
0.44
3.8
1200
25
2300
543
270
39
632
22
14
22
27
40
1.4
2.2
2.7
5.2
5.5
650
134
11
17
40
8.7
19
2.6
5.5
6.4
16
13
171
268
27
29
75
16
25
29
17
17
5.7
11
42
0.5
0.41
0.59
0.53
0.55
0.77
68
1991
3.4
40
0.2
2.3
3
17
17
42
1.6
0.98
a
Average of three or more assays.
A table with structure available as Supplementary data.
b
References
the intrinsic affinity was slightly improved to 1.1 nM but the pro-
tein shift was reduced to 22-fold. Modification of the linker length
between the tetrazole and the phenyl ring in compound 4b led to
compounds 4c and 4d, resulted in loss of potency while maintain-
ing a similar shift to 4b. Replacing the tetrazole group in 4d with a
carboxylic acid (4e) resulted in a slight improvement in potency
but none in reducing protein shift. Compounds containing two
other types of acidic surrogates, benzenesulfonylurea as in com-
pounds 9 and 6a–c and acylsulfonamides as in compounds 17a
1. Rao, P. P. N.; Knaus, E. E. J. Pharm. Pharm. Sci. 2008, 11, 81s–110s.
2. Meek, I. L.; van de Laar, M. A. F. J.; Vonkeman, H. E. Pharmaceuticals 2010, 3,
2146–2162.
3. Allegaert, K. L.; de Hoon, J.; Debeer, A.; Gewillig, M. Expert Opin. Drug Safety
2009, 8, 669–681.
4. Abdel-Tawab, M.; Zettl, H.; Schubert-Zsilavecz, M. Curr. Med. Chem. 2009, 16,
2042–2063.
5. Jones, R. L.; Giembycz, M. A.; Woodward, D. F. Br. J. Pharmacol. 2009, 158, 104–145.
6. For a recent review, see: Sakata, D.; Yao, C.; Narumiya, S. J. Pharm. Sci. 2010,
112, 1–5.