5440
I. Borza et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5439–5441
H
N
iv, v, vi
i, ii, iii
R1
R1
R1
R2
OH
N
H
N
H
CN
N
H
O
NH2
Cl
Scheme 1. Reagents and conditions: (i) SOCl2, CHCl3, reflux, 2 h; (ii) NH4OH; (iii) POCl3, CHCl3, reflux 2 h; (iv) HCl, EtOH, rt, 2 h;
(v) benzylamine derivative, rt, 6 h; (vi) HCl.
Solution-phase parallel synthesis was utilized to prepare
a range of amidines (Scheme 1). Approximately 200
compounds were prepared from unsubstituted- or 5-
substituted-indole-2-nitriles and 48 benzyl amines via
the Pinner synthesis. The indol-2-carbonitriles can be
synthesized by standard procedures from commercially
available indol-2-carboxylic acids. The nitriles were
transformed to the corresponding imidates by Pinner
chemistry.8 Amidines were prepared by treatment of in-
dol-2-carboximidates with the appropriate benzyl
amines, and purified by column chromatography. Their
purity was determined by the HPLC–MS method. The
potent compounds had been resynthesized and fully
NMDA-evoked changes of intracellular Ca2+ were mon-
itored with fluorimetry using a Ca2+-selective fluorescent
dye (Fluo-4/AM) and a plate reader fluorimeter.9 The
results of the functional assay for selected compounds
are summarized in Table 1.
Selectivity toward NR2A subunit containing NMDA
receptors was tested by the same functional assay using
cells expressing recombinant NR1/NR2A receptors and
none of the compounds exhibited significant activity up
to 15 lM concentration. In vivo analgesic activity was
tested in the mouse formalin test,10,11 a model of persis-
tent pain.
1
characterized (IR, H and 13C NMR, and high resolu-
tion MS).
Initial investigation of the indole-2-carboxamidines fo-
cussed on substitution of some of the phenyl rings. Sub-
stitution of the benzyl moiety (R2) showed that
substitution at either 3 or 4-position was poorly tolerat-
ed; for example, 4-methoxy (5n), and 3-methoxy (5m)
analogues were practically inactive in the functional test.
On the other hand, the 2-methoxy substituent (5a)
showed high activity, while other groups at this position
(5k or 5l) were less active. Disubstitution by MeO
groups at the 2 and 6 positions, as in 5e–h, gave en-
hanced activity. Other disubstitution seemed to give
inferior compounds (5i and j). A possible reason why
ortho substitution, and to an even greater extent 2,6-
disubstitution, leads to potent compounds is that the
ortho substituents force the aromatic B-ring out of pla-
narity with the indolecarboxamide moiety. Holding
2-methoxy R2 group as constant, significant effects on
affinity were observed by the substitution of the indol
ring. The activity depended on the electron-donating
ability of R1 substituents, increased from 5-Cl toward
5-F and 5-MeO.
Biological activity of the prepared compounds was mea-
sured in a functional assay where the inhibition of
NMDA-evoked increase of intracellular Ca2+ level was
determined on rat cortical cell culture. Baseline and
Table 1. Functional assay results for compounds 5a–o
H
R1
R2
N
N
H
NH2
Cl
5a-5o
Compound R1
R2
NMDA-evoked D[Ca2+
]
n
a,b
i
Inhib (%)c IC50 (nM)
5ad
5bd
5cd
5dd
5ed
5fd
5gd
5h
5i
H 2-OCH3
5-MeO 2-OCH3
35
19
33
5
3
6
5
3
3
3
3
5
3
4
4
3
4
3
2
1
1
9
2
5-F
5-Cl
H
2-OCH3
2-OCH3
2,6-(OCH3)2 5.4 0.8
64 12
5-MeO 2,6-(OCH3)2 26
5
3
5-F
5-C1
H
2,6-(OCH3)2 17
The most potent compound in this series is 5e which is a
2,6-dimethoxybenzyl derivative and its indole portion is
unsubstituted. Compound 5e showed good subtype
selectivity too (NR1A/2A inhibition at 15 lM was
32.1%). The unsubstituted analogue (5o) was almost
inactive, indicating that the common scaffold in itself
was not enough for the activity.
2,6-(OCH3)2 69 13
2,6-Di-F
3,5-Di-Cl
2-CH3
2-F
143 27
159 17
170 34
232 36
57.0%
5j
H
5k
5l
H
H
5m
5n
5o
1
H
3-OCH3
4-OCH3
H
H
10.5%
32.8%
H
470 51
131 10
Compound 5a had the best oral efficacy in formalin test
(ED50 15 mg/kg po).
2
3
18
6.6 1.1
4
13
3
4
In summary, a series of new indole-2-carboxamidine
derivatives were prepared and found to be potent and
selective antagonists of the NR2B subtype of NMDA
receptors. The conclusion of this study was that the
activity of this type of compounds unusually strongly
depended on the character and position of the substitu-
ents on the aromatic rings. Results of detailed biological
investigations will be published under separate cover.
a Values represent mean SEM. The number of experiments (n) is
indicated.
b NMDA-evoked changes of intracellular Ca2+
.
c Inhib (%) were obtained using 1 lM concentration of compound.
d The salt form was proved by potentiometric titration. Melting points
for the compounds are as follows: 5a, mp 186–187 °C; 5b, mp 195–
196 °C; 5c, mp 164–165 °C; 5d, mp 190–191 °C; 5e, mp 243–244 °C;
5f, mp 242–244 °C; 5g, mp 248–250 °C.