Cyclohexenyl- and dehydropiperidinyl-alkynyl pyridines as potent metabotropic glutamate subtype 5 (mGlu5) receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2005.06.099

Structure-activity relationship studies leading to the discovery of novel mGlu5 receptor antagonists are described. These compounds show high in vitro potency, have good in vivo receptor occupancy, and a reasonable intravenous pharmacokinetic profile.

Full text of DOI:10.1016/j.bmcl.2005.06.099

Bioorganic & Medicinal Chemistry Letters 15 (2005) 4589–4593
Cyclohexenyl- and dehydropiperidinyl-alkynyl pyridines
as potent metabotropic glutamate subtype 5 (mGlu5)
receptor antagonists
Peter C. Chua,a,* Johnny Y. Nagasawa,a Leo S. Bleicher,a Benito Munoz,a
Edwin J. Schweiger,a Lida Tehrani,a Jeffrey J. Anderson,b Merryl Cramer,a Janice Chung,c
Mitchell D. Green,a Chris D. King,a Grace Reyes-Manaloc and Nicholas D. P. Cosforda
aDepartment of Medicinal Chemistry, Merck Research Laboratories, MRLSDB2 3535 General Atomics Court,
San Diego, CA 92121, USA
bDepartment of Pharmacology, Merck Research Laboratories, MRLSDB1 3535 General Atomics Court, San Diego, CA 92121, USA
cDepartment of Molecular Profiling, Merck Research Laboratories, MRLSDB2, 3535 General Atomics Court,
San Diego, CA 92121, USA
Received 28 January 2005; accepted 29 June 2005
Available online 22 August 2005
Abstract—Structure–activity relationship studies leading to the discovery of novel mGlu5 receptor antagonists are described. These
compounds show high in vitro potency, have good in vivo receptor occupancy, and a reasonable intravenous pharmacokinetic
profile.
Ó 2005 Elsevier Ltd. All rights reserved.
S
Metabotropic glutamate (mGlu) receptors are a family
of G-protein coupled receptors in the mammalian ner-
N
N
vous system that are activated by L-glutamate.1,2 Eight
mGlu receptor subtypes have been identified to date
and these are organized into 3 groups (Groups I, II,
and III) based on sequence homology. Group I mGlu
receptors (mGlu1 and mGlu5) are localized primarily
in the postsynaptic region, but are also widely distribut-
ed in the hippocampus, thalamic nuclei, and spinal cord.
Stimulation of mGlu1 and mGlu5 receptors leads to
phosphoinositide (PI) hydrolysis and elevation of intra-
cellular Ca2+ levels ([Ca2+]i) via G-protein coupling to
phospholipase C.3,4 Excessive activation of mGlu5
receptors has been implicated in a number of CNS dis-
orders including pain,5 anxiety and depression,6–12 drug
dependence13, and mental retardation.14 It is for these
reasons that efforts have been directed toward the devel-
opment of potent and selective mGlu5 receptor
antagonists.
N
1
MPEP
2 MTEP
The discovery of a number of potent and highly selective
ligands for the mGlu5 receptor, such as 2-methyl-6-
(phenylethynyl)-pyridine (MPEP, 1)15 and 3-[(2-meth-
yl-1,3-thiazol-4-yl)-ethynyl] pyridine (MTEP, 2)16,17
,
has been recently reported. Our colleagues have previ-
ously reported SAR studies on the pyridyl ring of
MPEP,18 on replacing the alkyne moiety with various
heterocycles,19 and on the pyridyl ring of MTEP.20 In
this communication, SAR studies focused on replacing
the aryl group appended to the pyridine–alkyne motif
are described. The direction was to explore novel satu-
rated moieties, which were not previously investigated.
These efforts led to the discovery of new derivatives of
1 that display good in vitro potency, receptor occupan-
cy, and pharmacokinetics in rat. The functional
Keywords: mGlu5; Antagonist; CNS.
*
Corresponding author. Tel.:+85 88 75 51 18; fax:+ 85 88 75 5101;
e-mail: pchua@cylenepharma.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.06.099

4590
P. C. Chua et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4589–4593
potencies of these compounds in vitro were assessed
using an automated assay employing Ltk cells that sta-
bly express human recombinant mGlu5 receptors (Table
1). This cell-based assay measures changes in cytosolic
Ca2+ concentrations ([Ca2+]i) by fluorescence detection
using the Ca2+-sensitive dye fura-2.21–23 Binding to na-
tive mGlu5 receptors in vitro was determined by mea-
suring the displacement by test compounds of [3H]-3-
methoxy-5-(pyridin-2-ylethynyl)pyridine from rat corti-
cal membranes.17 Research from this laboratory has
demonstrated that most alkyne derivatives, such as 1
and 2, are inhibitors of CYP1A2;24 therefore, all com-
pounds were assayed for potency as CYP1A2 inhibitors
using recombinant CYP450 under Gentest-based proto-
cols.25 Furthermore, the hydrophobic nature of many of
these compounds could limit their solubility in CSF and
very likely reduce in vivo efficacy, so the logD was mea-
sured to assist in developing a compound with better
physical properties.26
ethynylcyclohexane to furnish the product directly
(Scheme 1).
Compounds 6–12 were synthesized by the addition of
lithium 2-pyridylacetylide to 3-ethoxy cyclohexenone,
which gave the corresponding a,b-unsaturated ketone
after work-up. Reduction with sodium borohydride in
methanol gave the corresponding allylic alcohol, which
could be derivatized using Mitsunobu conditions
(Scheme 2).
The synthesis of compounds 13–16 also utilized lith-
ium 2-pyridylacetylide which was added to 1,4-
cyclohexanedione monoethylene ketal to afford the
corresponding tertiary alcohol. The ketal was then
removed and the ketone was subjected to reductive
amination conditions using sodium triacetoxyborohy-
dride. Subsequent treatment with phosphorus oxy-
chloride furnished
(Scheme 3).
the
desired
compounds
The general scheme for the synthesis of alkyne deriva-
tives 3 and 4 involved the formation of lithium 2-pyr-
idylacetylide and subsequent trapping of this anion
with a carbonyl compound. The corresponding tertiary
alcohol was then dehydrated with phosphorus oxy-
chloride. Compound 5 was readily prepared using a
Sonogashira reaction between 2-bromopyridine and
For compounds 17–24, lithium 2-pyridylacetylide was
added to N-(t-butoxycarbonyl)-4-piperidone and then
dehydrated with phosphorus oxychloride. Deprotection
with TFA gave the free amine, which was derivatized
with the corresponding carbonyl or sulfonyl chloride
(Scheme 4).
Table 1. In vitro potency data for mGlu5 receptor antagonistsa
N
N
N
N
N
R
N
Cmpds 6-12b
Cmpds 13-16b
Cmpds 17-24
R
R
Cmpds 3-4
Cmpds 5
n
Compound
R
hmGlu5 Ca2+ Flux IC50 (nM)c
mGlu5 Ki (nM)d
CYP1A2 IC50 (nM)
logDe
3
4
n = 1
n = 2
—
3.1 (1.0, 4)
2.4 (0.7, 2)
106 (33, 3)
213 (160, 5)
27 (4.0, 2)
22 (1.0, 4)
16 (14, 5)
10
0.65
163
75
2184
676
3.5
3.8
3.9
5
2779
5308
6
S-2-naphthyl
N-phthalyl
S-4-pyridyl
O-4-pyridyl
N-3-pyridyl
O-3-pyridyl
O-3-(5-chloro)-pyridyl
N-2-naphthyl
N-c-pentyl
N-morpholino
N-3-pyridyl
t-Boc
Fluorescent
3.9
4.0
7
5.8
43
5249
505
8
9
21
1021
3418
3.5
3.3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
8.6 (0.4, 3)
2.2 (1.0, 2)
0.90 (0.3, 4)
306 (68, 2)
987 (193, 2)
164 (37, 5)
12.6 (3.0, 6)
2.6 (0.9, 4)
580 (221; 3)
35 (13, 2)
7.3
1.8
1.1
304
7200
683
35
1038
608
3.4
3.9
>20000
>20000
>20000
>20000
>20000
>20000
>20000
>20000
>20000
>20000
>20000
>20000
5.1
2.1
2.8
3.2
20
—
1.6
H
3400
171
0.83
389
13
Phenoxy carbonyl
cBz
3.7
—
12 (12, 3)
58 (12, 2)
Benzylsulfonyl
m-tosyl
p-tosyl
3.4
4.6
8.2 (2, 2)
7.9 (10; 4)
2.2 (0.4, 3)
12
3.3
3.7
p-brosyl
3.1
a Data are presented as the geometric mean followed in parentheses by the standard deviation and the number of replicates.
bCompounds were tested as racemates.
c Ca2+ flux assay using glutamate (10 lM) as agonist.18–20
d Displacement by test compounds of [3H]-3-methoxy-5-(pyridin-2-ylethynyl)pyridine from rat cortical membranes.16
e See Ref. 16.

P. C. Chua et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4589–4593
4591
a, b
(50-85%)
3
n = 1
n = 2
N
N
N
4
H
n
H
c
N
5
Br
(75%)
Scheme 1. Reagents and conditions: (a) n-BuLi/THF, À78 °C, then cycloalkanone. (b) POCl3, pyridine, 70 °C. (c) Pd(PPh3)4, CuI, pyrrolidine, 80 °C.
6
7
8
9
R = S-2-naphthyl
R = N-phthalyl
R = S-4-pyridyl
R = O-4-pyridyl
a, b, c
(60%)
d
N
N
R
OH
N
(30-80%)
10 R = N-3-pyridyl
11 R = O-3-pyridyl
12 R = O-3-(5-chloro)-pyridyl
Scheme 2. Reagents and conditions: (a) n-BuLi/THF, À78 °C, then cycloalkanone. (b) Pyridine, POCl3, 70 °C. (c) NaBH4, MeOH, rt. (d)
Diethylazidodicarboxylate, PPh3, THF, phenol, thiol, or phthalimide.
14 R = N-2-naphthyl
a
b, c, d
13 R = N-c-pentyl
15 R = N-morpholino
16 R = N-3-pyridyl
N
N
OH
(73%)
N
(35-70%)
O
R
O
Scheme 3. Reagents and conditions: (a) n-BuLi/THF, À78 °C, then cycloalkanone. (b) THF, 10% H2SO4 (aq), 70 °C. (c) NaBH3CN, CH2Cl2, AcOH.
(d) POCl3, pyridine, 70 °C.
19 R = phenyl carbonyl
20 R = cBz
d
a, b
c
21 R = benzylsulfonyl
22 R = m-tosyl
23 R = p-tosyl
24 R = p-brosyl
N
N
N
(65-85%)
(65%)
(89%)
boc
N
N
N
N
H
R
17
18
Scheme 4. Reagents and conditions: (a) n-BuLi/THF, À78 °C, then cycloalkanone. (b) POCl3, pyridine, 70 °C. (c) TFA, CH2Cl2. (d) Triethylamine,
CH2Cl2, acid chloride.
In replacing the phenyl ring of MPEP, these efforts ini-
tially gave rise to compounds 3 and 4 that contained
cycloalkenes. Interestingly, the fully saturated cyclohex-
ane 5 resulted in a 50-fold loss in potency, suggesting the
importance of the double bond for potency. Incorporat-
ing a large non-polar group, such as the sulfur-linked
2-naphthyl group in compound 6, resulted in a 100-fold
loss in potency. The direction turned towards smaller
and more polar groups to improve the logD and poten-
cy of these compounds. By directly linking a phthalyl
group, compound 7 was found to be 10-fold less potent.
Further reduction in size, but maintaining the presence
of basic moieties to decrease the logD gave rise to 8
and 9 which were slightly more potent. Changing the
4-pyridyl to a 3-pyridyl gave rise to 10 and 11, which
were equipotent with the original hits. The 5-chloro-3-
pyridyl analog (12) was found to be 2-fold more potent
in both the functional assay, as well as the binding assay.
Unfortunately, with increasing potency as mGlu5
inhibitors, these compounds were also found to be
increasingly potent as inhibitors of cytochrome P450
1A2 (CYP1A2).
Concurrently, substitution at the homo-allylic position
was investigated (compounds 13-16). Similar to substi-
tution at the allylic position (compounds 6-12), a large
naphthyl group (13) was found to be detrimental to
potency. Smaller substituents in 14 and 15 had improved
potency and even the N-3-pyridyl 16 was found to be
5-fold less potent. Although these compounds were not
quite as potent as their allylic counterparts, they were
found to be completely non-inhibitory toward CYP1A2,
which led to further investigation at this position.
For additional studies at the homo-allylic position, we
chose to replace the carbon with a more polar atom,
such as nitrogen. Compound 17 was surprisingly
equipotent with our original hit and not inhibitory to-
ward CYP1A2. Removal of the Boc-group gave second-
ary amine (18), which had a much lower logD.

4592
P. C. Chua et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4589–4593
Table 2. Selected rat pharmacokinetic dataa
compounds tend to have moderate logD. Furthermore,
substitution at the homo-allylic position allowed us to
generate potent compounds, such as 23 and 24, which
were non-inhibitory toward CYP1A2. In vivo experi-
ments revealed that these compounds have moderate
half-lives and moreover, a lower receptor occupancy in
the rat brain model than their CYP-inhibiting counter-
parts. It is for this reason that this series of compounds
was excluded from further development.
Compound
R
T1/2 Cl Vd
(h) (mL/min/kg) (L/kg)
4
9
n = 2
O-4-pyridyl
O-3-pyridyl
3.9 45
0.8 41
0.3 94
9.1
1.2
11
12
16
23
24
2.2
7.2
O-(5-chloro)-3-pyridyl 4.8 32
N-3-pyridyl
N-p-tosyl
N-p-brosyl
0.5 34
1.5 29
1.1 64
1.4
14.7
3.98
a i.v. dosing at 2 mg/kg.
Acknowledgments
The authors thank Bill Bray and Joyce James for expert
technical assistance.
Table 3. Selected rat receptor occupancy data
Compound
R
Occupancy (%)
4
12
23
24
n = 2
94
97
25
57
O-(5-chloro)-3-pyridyl
N-p-tosyl
N-p-brosyl
References and notes
1. Ozawa, S.; Kamiya, H.; Tsuzuki, K. Prog. Neurobiol.
(Oxford) 1998, 54, 581.
2. Brauener-Osborne, H.; Egebjerg, J.; Nielsen, E. O.;
Madsen, U.; Krogsgaard-Larsen, P. J. Med. Chem. 2000,
43, 2609.
3. Pin, J.-P.; Acher, F. Curr. Drug Targets: CNS Neurol.
Disord. 2002, 1, 297.
4. Conn, P. J.; Pin, J.-P. Annu. Rev. Pharmacol. Toxicol.
1997, 37, 205.
5. Varney, M. A.; Gereau, R. W. I. Curr. Drug Targets: CNS
Neurol. Disord. 2002, 1, 283.
6. Brodkin, J.; Busse, C.; Sukoff, S. J.; Varney, M. A.
Pharmacol. Biochem. Behav. 2002, 73, 359.
7. Spooren, W. P. J. M.; Vassout, A.; Neijt, H. C.; Kuhn, R.;
Gasparini, F., et al. J. Pharmacol. Exp. Ther. 2000, 295,
1267.
8. Spooren, W. P. J. M.; Schoeffter, P.; Gasparini, F.; Kuhn,
R.; Gentsch, C. Eur. J. Pharmacol. 2002, 435, 161.
9. Tatarczynska, E.; Klodzinska, A.; Chojnacka-Wojcik, E.;
Palucha, A.; Gasparini, F., et al. Br. J. Pharmacol. 2001,
132, 1423.
Unfortunately, compound 18 lost a significant amount
of potency, suggesting that positively charged atoms
are not well-tolerated in this SAR. In an effort to make
this nitrogen less basic, derivatization as a phenoxycar-
bamate (19) and the benzyloxycarbamate (20) was
found to be less potent than the t-butoxycarbamate
(17). Derivatization as sulfonamides proved to be more
fruitful. The corresponding benzylsulfonamide 21 was
almost equipotent with carbamate 17. More rigid sul-
fonamides, such as 22, 23, and 24, had increased poten-
cy, with 24 being equipotent with the orginal hit 4. It
was gratifying to find that these sulfonamides were not
inhibitors of CYP1A2.
Ongoing pharmacokinetic evaluation revealed that some
compounds had reasonable profiles (Table 2). Com-
pounds 9, 11 and 16 were abandoned due to poor
half-lives, as well as CYP1A2 issues. Compounds 4
and 12 had better half-lives, but were not pursued fur-
ther due to CYP1A2 concerns. Compounds 23 and 24
had somewhat shorter half-lives, but warranted further
investigation. Compound 24 was dosed orally at
10 mg/kg to reveal a bioavailability of 10% and Cmax
of 0.02 lM (which is 10-fold the IC50).
10. Klodzinska, A.; Tatarczynska, E.; Chojnacka-Wojcik, E.;
Pilc, A. Pol. J. Pharmacol. 2000, 52, 463.
11. Schulz, B.; Fendt, M.; Gasparini, F.; Lingenhohl, K.;
Kuhn, R., et al. Neuropharmacology 2001, 41, 1.
12. Busse, C. S.; Brodkin, J.; Tattersall, D.; Anderson, J. J.;
Warren, N. A.; Tehrani, L.; Cosford, N. D. P.; Bristow, L.
J.; Varney, M. A. Unpublished results.
13. Chiamulera, C.; Epping-Jordan, M. P.; Zocchi, A.;
Marcon, C.; Cottiny, C., et al. Nat. Neurosci. 2001, 4, 873.
14. Huber, K. M.; Gallagher, S. M.; Warren, S. T.; Bear, M.
F. Proc. Nat. Acad. Sci. U.S.A 2002, 99, 7746.
15. Gasparini, F.; Lingenhohl, K.; Stoehr, N.; Flor, P. J.;
Heinrich, M. Neuropharmacology 1999, 38, 1493.
16. Cosford, N. D. P.; Roppe, J.; Tehrani, L.; Seiders, T. J.;
Schweiger, E. J., et al. Bioorg. Med. Chem. Lett. 2003, 13,
351.
17. Cosford, N. D. P.; Tehrani, L.; Roppe, J.; Schweiger, E.;
Smith, N. D., et al. J. Med. Chem. 2003, 46, 204.
18. Smith, N. D.; Poon, S. F.; Huang, D.; Green, M.; King,
C.; Tehrani, L.; Roppe, J. R.; Chung, J.; Chapman, D. P.;
Cramer, M.; Cosford, N. D. P. Bioorg. Med. Chem. Lett.
2004, 14, 5481.
Concurrently with intravenous pharmacokinetic analy-
sis, compounds were also evaluated for in vivo receptor
occupancy (Table 3). The test compound was adminis-
tered intraperitoneally at 10 mg/kg and at 59 min, while
[3H]-3-methoxy-5-(pyridin-2-ylethynyl)pyridine
was
injected via the tail vein. One minute later, the rats were
euthanized and binding was measured from the brain
homogenate.27 Compounds 4 and 12 were found to be
highly occupant at 10 mg/kg (IP dosing). Unfortunately,
compounds 23 and 24, which were not CYP1A2 inhibi-
tory, were less potent in the occupancy assay.
19. (a) Roppe, J.; Smith, N. D.; Huang, D.; Tehrani, L.;
Wang, B.; Anderson, J.; Brodkin, J.; Chung, J.; Jiang, X.;
King, C.; Munoz, B.; Varney, M. A.; Prasit, P.; Cosford,
N. D. P. J. Med. Chem. 2004, 47, 4645; (b) Huang, D.;
Poon, S. F.; Chapman, D. F.; Chung, J.; Cramer, M.;
Replacing the aromatic or hetero-aromatic moieties of
MPEP (1) or MTEP (2) with semi-saturated rings
allowed us to identify several novel compounds that
are highly potent antagonists of mGlu5 receptors. These

P. C. Chua et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4589–4593
4593
Reger, T. S.; Roppe, J. R.; Tehrani, L.; Cosford, N. D. P.;
Smith, N. D. Bioorg. Med. Chem. Lett. 2004, 14, 5473.
20. Roppe, J. R.; Wang, B.; Huang, D.; Tehrani, L.; Kam-
enecka, T.; Schweiger, E. J.; Anderson, J. J.; Brodkin, J.;
Jiang, X.; Cramer, M.; Chung, J.; Reyes-Manalo, G.;
Munoz, B.; Cosford, N. D. P. Bioorg. Med. Chem. Lett.
2004, 14, 3993.
23. Daggett, L. P.; Sacaan, A. I.; Akong, M.; Rao, S.
P.; Hess, S. D., et al. Neuropharmacology 1995, 34,
871.
24. Schweiger, E. J.; Tehrani, L.; King, C. D.; Green, M. D.
Unpublished results.
25. Green, M. D.; Jiang, X.; King, C. D. Life Sci. 2004, 75,
947.
21. Varney, M. A.; Cosford, N. D. P.; Jachec, C.; Rao, S. P.;
Sacaan, A., et al. J. Pharmacol. Exp. Ther. 1999, 290, 170.
22. Varney, M. A.; Suto, C. M. Drug Discovery Today: HTS
Suppl. 2000, 1, 20.
26. See Ref. 17, supporting information.
27. Anderson, J. J.; Bradbury, M. J.; Giracello, D. R.;
Chapman, D. F.; Holtz, G.; Roppe, J., et al. Eur. J.
Pharmacol. 2003, 473, 35.