352
N. D. P. Cosford et al. / Bioorg. Med. Chem. Lett. 13 (2003) 351–354
Table 1. In vitro data for mGlu5 receptor antagonists
antagonists with excellent in vitro potency at the mGlu5
receptor and low LogD (low lipophilicity). Thus, in
addition to the determination of LogD for compounds
using an HPLC method,10 the functional potency of
compounds in vitro was assessed using an automated
assay employing Ltk-cells stably expressing human
recombinant mGlu5 receptors (Table 1). This cell-based
assay measures changes in cytosolic Ca2+ concen-
trations ([Ca2+]i) by fluorescence detection using the
Ca2+-sensitive dye fura-2.14À16
Compd
A
B
mGlu5 Ca2+
Flux IC50 (nM)a
LogDb
1
Ph
2
7
3.5
3.0
2.1
2.5
2.4
2.0
2.25
2
3
7
4
6
5
3-MeOPh
3-Py
5
Comparison of MPEP (1) and MTEP (3) indicated that
while both compounds show similar potency at the
mGlu5 receptor in vitro (Table 1), MTEP (3) is much
less lipophilic, with LogD=2.1 versus 3.5 for 1. MTEP
(3) appeared, therefore, to be a promising initial scaffold
and SAR studies revealed that methyl substitution was
tolerated at the 5-pyridyl position, as in compound 7,
giving a slight enhancement of potency (IC50=3 nM).
Further investigation of the SAR demonstrated that a
methoxy substituent on the pyridyl methyl group of 7
was also tolerated as in compound 4 (methoxymethyl-
MTEP) with IC50=7 nM. Interestingly 4, selected as a
candidate for tritium labeling, exhibits similar in vitro
functional potency to 2 (M-MPEP) in the Ca2+ flux
assay (Table 1) however 4 (LogD=2.4) is less lipophilic
than 2 (LogD=3.0), suggesting the potential for lower
non-specific binding for 4. Efforts to reduce lipophilicity
even further led to replacement of the thiazole ring in 4
with a 2-pyridyl moiety to give 6 (LogD=2.0) although
this compound displayed a significant loss of in vitro
potency (IC50=19 nM). Truncation of the methoxy-
methylene unit in 6 to a methoxy moiety gave 5 (methoxy-
PEPy), a compound with excellent functional potency
(IC50=1 nM) and low lipophilicity (LogD=2.25). On
the basis of these results 4 and 5 were selected for tri-
tium labeling and evaluation in binding experiments.
3
7
19
1
aCa2+ flux assay using glutamate (10 mM) as agonist. Concentration–
response curves were performed using 12 concentrations, performed in
duplicate wells in two or more separate experiments.14À16
bSee ref 10.
the present communication reporting our parallel efforts
to develop potent and highly selective tritiated mGlu5
receptor antagonists. This workresulted in the discovery
of [3H]-3-(methoxymethyl)-5-[(2-methyl-1,3-thiazol-4-yl)-
ethynyl]pyridine ([3H]-methoxymethyl-MTEP; 4), and
[3H]-3-methoxy-5-(pyridin-2-ylethynyl)pyridine ([3H]-
methoxy-PEPy; 5). The synthesis and in vitro charac-
terization of these pharmacological tools and described
herein.
Chemistry
The synthesis of compounds 4 to 7 is summarized in
Schemes 1 and 2. Compounds 4 and 6 were each pre-
pared in three steps from methyl 5-bromonicotinate via
Sonogoshira cross-coupling of the appropriate alkyne
The development of a viable radioligand typically
requires the optimization of two parameters: (a) high
specific binding (i.e., high affinity for the receptor in
question), and (b) low non-specific binding to other
endogenous binding sites. Non-specific binding is gen-
erally more pronounced for highly lipophilic molecules.
We therefore sought to identify selective mGlu5 receptor
precursor
with
3-bromo-5-methoxymethylpyridine
(Scheme 1). Compound 5 was prepared by Sonogoshira
cross-coupling of 2-ethynylpyridine with 3-bromo-5-
methoxypyridine, obtained by reaction of sodium
Scheme 1. Reagents and conditions: (a) LiAlH4, THF, À78 ꢀC (63%); (b) NaH, MeI, THF, 0–25 ꢀC, 60 h, (90%); (c) R=Me: PdCl2(PPh3)2, PPh3,
CuI, NEt3, Bu4NI, Bu4NF, DMF, 45 ꢀC, 60 h, (58%); R=H: PdCl2, PPh3, CuI, NEt3, Bu4NF, DME, 75 ꢀC, 60 h, (44%); (d) [3H3]CI, NaH, THF,
0–25 ꢀC; radiochemical purity >99%, specific activity 79.8 Ci/mmol; (e) PdCl2, PPh3, CuI, NEt3, DME, 75 ꢀC, 6 d (58%).
Cosford, Nicholas D. P.
