Synthesis of functionalized 1,8-naphthyridinones and their evaluation as novel, orally active CB1 receptor inverse agonists

Article abstract of DOI:10.1016/j.bmcl.2005.10.028

Synthesis, SAR, and binding affinities are described for a new class of 1,8-naphthyridinone CB1 receptor specific inverse agonists. Food intake, knockout mouse, and pharmacokinetic evaluation of 14 indicate that this compound is an effective orally active

Full text of DOI:10.1016/j.bmcl.2005.10.028

Bioorganic & Medicinal Chemistry Letters 16 (2006) 681–685
Synthesis of functionalized 1,8-naphthyridinones and
their evaluation as novel, orally active CB1 receptor inverse agonists
John S. Debenham,a,* Christina B. Madsen-Duggan,a Thomas F. Walsh,a Junying Wang,a
Xinchun Tong,a George A. Doss,b Julie Lao,c Tung M. Fong,c Marie-Therese Schaeffer,c
Jing Chen Xiao,c Cathy R.-R. C. Huang,c Chun-Pyn Shen,c Yue Feng,c Donald J. Marsh,c
D. Sloan Stribling,d Lauren P. Shearman,d Alison M. Strack,d D. Euan MacIntyre,d
Lex H. T. Van der Ploege and Mark T. Goulete
aDepartment of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
bDepartment of Drug Metabolism, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
cDepartment of Metabolic Disorders, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
dDepartment of Pharmacology, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
eMerck Research Laboratories, 33 Avenue Louis Pasteur, Boston, MA 02115, USA
Received 21 September 2005; revised 5 October 2005; accepted 7 October 2005
Available online 2 November 2005
Abstract—Synthesis, SAR, and binding affinities are described for a new class of 1,8-naphthyridinone CB1 receptor specific inverse
agonists. Food intake, knockout mouse, and pharmacokinetic evaluation of 14 indicate that this compound is an effective orally
active modulator of CB1.
Ó 2005 Elsevier Ltd. All rights reserved.
Modulation of the cannabinoid receptor 1 (CB1; pre-
dominantly found in the central nervous system) has
O
N
N
been demonstrated to have powerful effects on feeding
behavior in both humans and other animals.1 CB1 ago-
nists such as D9-THC stimulate food intake, while CB1
inverse agonists such as SR141716A2 suppress food in-
take demonstrating the utility of CB1 inhibition for
the treatment of obesity.3
H
N
N
SR141716A
hCB-1 IC50 6 nM
Cl
Cl
Cl
Cl
O
N
Our interest in the discovery and development of novel
therapeutics for obesity prompted a high-throughput
screen of the Merck chemical collection. This effort led
to the discovery of compound 1 which exhibited rela-
tively high-affinity binding to cloned human CB1 with
an IC50 value of 526 nM (Fig. 1).
CN
CN
O
N
O
N
F
F
Cl
Cl
2
1
hCB-1 IC50 1.3 nM
hCB-1 IC50 526 nM
We recently described medicinal chemistry optimization
of screening lead 1 to afford the diaryl pyridine deriva-
Figure 1. SR141716A and early Merck leads.
tive 2 which achieved a greater than 400-fold improve-
ment in CB1 binding affinity.4 Unfortunately, the high
in vitro potency of 2 did not equate with high efficacy
in vivo, which was a consequence of its poor CNS
Keywords: Cannabinoid; CB1; Inverse agonist; Anorexigenic; Obesity;
CB2.
*
Corresponding author. Tel.: +1 732 594 5497; fax: +1 732 594
2210; e-mail: john_debenham@merck.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.10.028

682
J. S. Debenham et al. / Bioorg. Med. Chem. Lett. 16 (2006) 681–685
exposure (brain to plasma ratio = 0.03–0.26 at 0.25–4 h
post-dosing).
Exposure of the 2-acyl pyridine to a strong base such
as NaH or LiHMDS allowed annulation of the new ring
to the pyridine core providing the 4-amino-1,8-naphthy-
ridinone 11. The amino group could be further acylated
with acetic anhydride at elevated temperature or by
treatment with various acid chlorides to give 12.
We theorized that improved pharmacokinetic properties
might be achieved by alteration of certain physical prop-
erties for compounds in this series (e.g., the high logD
value).5 One goal therefore became to reduce the lipo-
philicity of the series. From earlier work it appeared that
the biaryl pyridine core was fairly well optimized for
potency, but that there was considerable flexibility to al-
ter the substitution pattern on the b-face of the pyridine
ring to thereby affect the pharmaceutical properties of
the series.
That the 2-aminopyridine was unreactive to acetylation
is clearly shown in Scheme 2. When 9 (R1 = methyl) was
treated to a large excess of Ac2O (co-solvent) in pyridine
at 80 °C for 18 h, no reaction was observed. When
1.2 equiv of DMAP were added and the reaction was
run at 85 °C, the 2-acetyl pyridine 13 was still not
formed in significant quantities. Instead, an unexpected
derivative was formed in an unoptimized 50% isolated
yield. It was determined that the reaction had provided
an unprecedented ring annulation process allowing nov-
el construction of 1,8-naphthyridinone 14 directly from
the aminopyridine.8
The presence of the 3-cyano group on the pyridine
would provide the framework to generate an additional
fused ring to the pyridine. It was hoped that heterobicy-
clic systems would maintain good potency but would
provide more drug-like physical properties. We first
examined the 1,8-naphthyridinone core system.6 Addi-
tional heterobicyclic systems will be the subject matter
of future publications.
The reaction appears to be general for substrates of
general structure 9 allowing the preparation of various
N-substituted analogs quite readily. It is noteworthy
that when the 2-acetyl pyridine 13 (R1 = Me) is treated
to the above reaction conditions, no significant forma-
tion of 14 is observed.
Scheme 1 illustrates the general synthesis of these com-
pounds.7 The dianion generated from acid 3 added to es-
ter 4 with loss of CO2 afforded ketone 5. Condensation
of 5 with dimethylformamide dimethylacetal afforded
vinylogous amide 6, which in turn reacted with cyanoac-
etamide under basic conditions to generate the 2-pyri-
done 7. The 2-chloropyridine 8 was prepared by
chlorination of 7 with POCl3 and then treated with var-
ious amines to generate the 2-aminopyridines 9. Substi-
tuted 2-aminopyridines can often be quite difficult to
acylate and this was indeed the case with compounds
of general structure 9. When 9 was exposed to acid chlo-
rides at room temperature, the reactions failed to pro-
vide the desired 2-acyl pyridines in acceptable yields.
However, deprotonation of the 2-amino group with
MeMgBr first, followed by exposure to various acid
chlorides, allowed rapid (<1 h) formation of the 2-acyl
derivatives at room temperature. When the R2 group
on the newly installed acyl substituent was a leaving
group such as a chlorine, a nucleophilic displacement
could be carried out providing additional diversity.
Cl
Cl
CN
CN
a
H
N
N
N
N
O
R1
R1
9
Cl
Cl
Cl
Cl
13
R1=Me
b
O
Cl
b
NH
O
N
N
O
R1
14
Cl
Cl
Scheme 2. Reagents and conditions: (a) Ac2O, Pyr, 80 °C; (b) Ac2O,
Pyr, DMAP, 85 °C, 50%.
Cl
Cl
Cl
Cl
NMe2
Cl
CN
O
CN
Cl
d
a, b
c
e
CO2H
Cl
CO2CH3
O
O
N
H
N
O
4
3
5
6
7
8
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
R4
Cl
Cl
Cl
Cl
Cl
Cl
NH2
NH
CN
CN R2
R2
O
R2
O
g
j
f
h-i
H
N
N
N
N
R1
O
N
N
R1
N
12
N
R1
R1
9
10
11
Cl
Cl
Cl
Cl
Cl
Cl
Scheme 1. Reagents and conditions: (a) NaN(TMS)2, THF, À20 °C; (b) ester 4, 79%; (c) (CH3)2N(OCH3)2, DMF, 90 °C, 98%; (d) 2-cyanoacetamide,
DMF, NaOCH3, 95 °C, 90%; (e) POCl3, reflux, 81%; (f) R1CH2NH2 (R1 = alkyl or aryl), THF, 25–65 °C, 90–100%; (g) MeMgBr, THF, then
ClC(O)CH2X, 67–93%; (h) optional nucleophilic displacement, DMF, KI, 64–100%; (i) NaH or LiN(TMS)2, THF and/or DMF, 0–25 °C, 73–85%;
(j) Ac2O, 85 °C, or ClC(O)CH2R4, 64–79%.

J. S. Debenham et al. / Bioorg. Med. Chem. Lett. 16 (2006) 681–685
683
The most expeditious way to explore the effect of varied
O
O
substitution at the 4-position on CB1 binding from ana-
log 14 was by cleavage of the N-acetamide group. This
could be readily accomplished via treatment with TFA
in aqueous ethanol at elevated temperature. The 4-
aminonaphthyridinone was readily functionalized by
acylation or alkylation (Scheme 3).
Cl
Cl
Cl
NH
N
O
O
NH
O
a
N
N
N
H
O
Cl
Cl
Cl
20
21
MeO
Alternatively, the nitrogen at this position could be re-
placed with carbon by changing the nature of the ring
annulation step (Scheme 4). Instead of ring annulation
by attack on a cyano group which forms an amino-
substituted ring, aldol condensation on a ketone would
leave a methyl group. Conversion of the 3-cyano to an
acetyl group was carried out via treatment of two equiv-
alents of MeMgBr on intermediate 9 followed by hydro-
lysis in aqueous HCl. Subsequent exposure of ketone 18
to similar reaction conditions to produce 14 afforded 4-
methylnaphthyridinone 19 in 59% yield.
OMe
Scheme 5. Reagents and conditions: (a) TFA, rt, 38%.
agonists. High potency (IC50 < 10 nM) for CB1 binding
could be obtained for a variety of functional groups at
R1 other than hydrogen. Typically as the bulk of the
group at this position increased so did the CB1 potency.
Alkyl and aryl groups had similar potencies as was the
case for either isobutyl 22 or benzyl 23 (both 2.3 nM
CB1). While ethers such as methoxy 24 or tetrahydro-
pyranyl 25 were well tolerated, a sixfold decrease in
potency was observed when R1 contained a hydroxyl
group 26. When R1 was a methyl group, the best CB1
potency was observed when R2 was N-acetyl 14 at
7.5 nM CB1, however potency decreased 15-fold
(120 nM) for the free amino derivative 15. The potency
remained similarly uninspiring (96 nM) when the amino
at R2 was dialkylated 17. Finally, when the polar amine
function at R2 was replaced by methyl 19, we saw a
modest improvement in potency to 55 nM. When R1 is
methyl, an acetyl group at R3 14 was optimal.
In the hope of simplifying the general structure it was of
interest to prepare naphthyridinones that were unsubsti-
tuted on the nitrogen atom. Unfortunately, direct cycli-
zation of 2-N-acyl pyridines (10 R1 = H) was not
possible as deacylation usually occurred after exposure
of these derivatives to strong bases such as NaH.
Scheme 5 shows that by utilizing a 2,4-dimethoxyben-
zyl-protecting group the 1,8-naphthyridinone system
could be generated as above and deprotected in neat tri-
fluoroacetic acid.
Binding affinities were determined using a standard pro-
tocol.9 Table 1 summarizes the SAR for the 1,8-naph-
thyridinone compounds. Most potent compounds
showed high selectivity for CB1 over CB2. All naphthy-
ridinone compounds tested were functional inverse
Several compounds (14 and 22–25) that had been tested
in the CB1 assay in vitro were shown to be quite inter-
esting as they had high potency (CB1 IC50 = 2.3–
7.5 nM) and good selectivity for CB1 over CB2 (between
270- and 2700-fold). In addition to its good CB1 poten-
cy, 14 also had the lowest lipophilicity10 and molecular
weight which we viewed as distinct advantages in com-
pounds that must be brain penetrant. Compound 14
was therefore selected for an evaluation of its effects
on food intake and body weight changes in diet-induced
obese (DIO) rats fed ad libitum on a moderate high fat,
high sucrose diet (Table 2).
O
R1 R2
N
Cl
Cl
Cl
O
O
NH
N
O
O
a
N
N
N
Cl
Cl
The CB1 inverse agonist 14 was tested at 0.3, 1, or 3 mg/
kg, PO.11 Compound 14 inhibited feeding in a dose-de-
pendent manner. At 3 mg/kg it decreased cumulative
food intake from 2 h post-dosing, with an overnight
food intake of 6.9 g (67.9% suppression; p = 0.0009 vs
vehicle treated controls). Compound 14 significantly de-
creased overnight body weight gain compared to vehicle
treatment at all dosing levels with an MED for body
weight effects of 0.3 mg/kg (6 g) relative to vehicle treat-
ed controls (12 g).
14
Cl
15 R1= R2=H
16 R1=H; R2=COCH2R3
17 R1= R2=CH3
b
c
Scheme 3. Reagents and conditions: (a) TFA, EtOH, H2O, 130 °C,
77%; (b) ClC(O)CH2R3, 11–55%; (c) NaH, THF, MeI, 0–25 °C, 16%.
Cl
Cl
O
R
b
In order to verify that the observed anorexigenic effects
were mediated through actions at CB1, an overnight
food intake study12 was conducted with CB1-deficient
(Cnr1À/À) mice. Based upon a dose–response study
conducted in C57BL/6 wild-type mice,13 ad libitum fed
14–16-week-old male mice (Cnr1+/+, n = 10; Cnr1À/À,
n = 9) were dosed orally with 14 at 10 mpk (as per the
N
NH
N
N
O
19
Cl
Cl
Cl
Cl
9 R=CN
18 R=Ac
a
Scheme 4. Reagents and conditions: (a) 2 equiv MeMgBr, THF,
50 °C, 78%; (b) Ac2O, Pyr, DMAP, 90 °C, 59%.

684
J. S. Debenham et al. / Bioorg. Med. Chem. Lett. 16 (2006) 681–685
Table 1. Structures and binding affinities (CB1; CB2) expressed as IC50 (nM)
O
O
Cl
Cl
Cl
NH
O
O
R2
N
O
O
NH
N
R3
O
N
N
N
N
R1
Cl
Cl
Cl
Cl
Cl
Cl
Compound
R1
CB1; CB2
2.3; 6200
Compound
R2
CB1; CB2
7.5; 4100
Compound
R3
CB1; CB2
7.5; 4100
O
O
22
14
27
14
NH
O
23
24
25
14
26
21
2.3; 5100
3.7; 1800
5.2; 1400
7.5; 4100
32; 1500
77; 1400
24; 3200
29
30
31
32
33
34
15; 5400
20; 2500
24; 1100
26; 1500
35; 720
H3CO
NH
N
OCH3
O
O
28
19
17
15
46; 3600
55; 1400
96; 1400
120; 1100
HO
OCH3
NH
CH3
CH3
CH3
N
CN
H
OH
H
NH2
110; 380
wild-type mice study). Compound 14 significantly
(p < 0.001) inhibited 2 h food intake (ꢀ49% reduction),
overnight (18 h) gains in body weight (ꢀ88% reduction;
Fig. 3) in wild-type (Cnr1+/+) mice, and had no signifi-
cant effect on any of these parameters in Cnr1À/À mice.
These data demonstrate that the anorectic actions of 14
are CB1 mechanism-based (Figs. 2 and 3).
5
4
3
2
1
0
*
Cnr1+/+, Vehicle
Cnr1+/+, 14
Cnr1-/-, Vehicle
Cnr1-/-, 14
With these encouraging results the pharmacokinetic
profile of 14 was evaluated in Sprague–Dawley rats,
C57BL/6 mice, beagles, and rhesus macaques. Select
parameters are summarized in Table 3.
Figure 2. Overnight food intake with wild-type and CB1-deficient
mice.
Table 2. Rat food intake/body weight change overnight (18 h) for 14
Dose (mg/kg)
D Body weight (g)
+12
% FI suppression
2
Vehicle
0.3
1.0
1
—
+6 2 (p = 0.049)
+1 2 (p = 0.002)
À4 2 (p = 0.0003)
22 9 (p = 0.077)
39 13 (p = 0.077)
68 14 (p = 0.0009)
1.5
Cnr1+/+, Vehicle
3.0
Cnr1+/+, 14
1
Cnr1-/-, Vehicle
Cnr1-/-, 14
Table 3. Pharmacokinetic profile of 14
0.5
*
Animal
F (%)
Clp (mL/min/kg)
t1/2 (h)
Rat
93
70
12
>8
>8
0
Mouse
Dog
Monkey
13.9
4.4
1.4
100
96
>24
22
Figure 3. Overnight body weight change with wild-type and CB1-
deficient mice.

J. S. Debenham et al. / Bioorg. Med. Chem. Lett. 16 (2006) 681–685
685
M.-T.; Chen, J.; Shen, C.-P.; Stribling, D. S.; Shearman,
L. P.; Strack, A. M.; Van der Ploeg, L. H. T. Bioorg.
Med. Chem. Lett. 2005, 15, 645 and 1755.
As expected from the strong body weight effects ob-
served in rats, 14 had high bioavailability in this species
(93%). Moreover, the rat brain to plasma ratio at 0.25 h
was 0.9 and 2.7 at 4 h post IV dosing, indicating that
this compound was highly brain penetrant. High bio-
availability across all four species was observed. Drug
clearance was low in all species resulting in robust
half-lives.
5. The calculated logD for this compound is 7.9 (Advanced
Chemistry Development Log D software version 6).
6. After this work was completed, Ferrarini et al. described
1,8-naphthyridin-4(1H)-one-3-carboxamide derivatives as
CB2 selective compounds. Ferrarini, P. L.; Calderone, V.;
Cavallini, T.; Manera, C.; Saccomanni, G.; Pani, L.; Ruiu,
S.; Gessa, G. L. Bioorg. Med. Chem. 2004, 12, 1921.
7. For full experimental details, see: Debenham, J. S.; Doss,
G. A.; Madsen-Duggan, C. B.; Walsh, T. F. WO Patent
2005047285 A1, 2005.
In summary, compound 14 is a novel, highly potent, and
specific inverse agonist of the CB1 receptor. Oral admin-
istration of compound 14 potently suppressed food in-
8. Unambiguous structural confirmation of 14 was achieved
by 1H–13C correlation HSQC, HMBC, and NOE NMR
experiments.
9. The reported binding results were an average of 2–8
independent determinations (each done in replicate),
which were normally within 50% of the average value.
For both the binding and functional assay conditions, see:
Felder, C. C.; Joyce, K. E.; Briley, E. M.; Mansouri, J.;
Mackie, K.; Blond, O.; Lai, Y.; Ma, A. L.; Mitchell, R. L.
Mol. Pharmacol. 1995, 48, 443.
10. The calculated logD for this compound is 3.1 (Advanced
Chemistry Development Log D software version 6).
11. Food intake assay protocol: compounds were solubilized
in 10% Tween 80 in water. DIO Sprague–Dawley rats
were dosed PO (orally) 30–60 min before dark onset.
There were six rats in each treatment group. Food cups
for each cage were weighed every 5 min to measure food
consumption (in an automated food intake system)
during the 18 h period following dosing with test
compounds. Overnight body weight changes were also
measured.
12. C57BL/6 mice were purchased from Taconic Farms. Cnr1
knockout mice were generated by the laboratory of Dr.
Andreas Zimmer, National Institute of Mental Health,
NIH (Zimmer, A.; Zimmer, A. M.; Hohmann, A. G.;
Herkenham, M.; Bonner, T. I. Proc. Natl. Acad. Sci. 1999,
96, 5780) and generously provided by him.
take and overnight body weight increases in
a
diet-induced obese rat model of food intake, and the
anorexigenic activity of compound 14 is consistent with
its modulation of the CB1 receptor signaling system.
The potency, oral efficacy, and excellent pharmacokinet-
ic properties observed for compound 14 clearly demon-
strate its utility as a pharmacological tool to evaluate the
potential for inhibition of CB1 receptors for the treat-
ment of obesity. Further SAR studies and pharmacolog-
ical characterization for CB1 inverse agonists derived
from our diaryl pyridine lead will be reported from these
laboratories in due course.
References and notes
1. Cota, D.; Marsicano, G.; Tscho¨p, M.; Grubler, Y.;
¨
Flachskamm, C.; Schubert, M.; Auer, D.; Yassouridis, A.;
Tho¨ne-Reinecke, C.; Ortmann, S.; Tomassoni, F.; Cervino,
C.; Nisoli, E.; Linthorst, A. C. E.; Pasquali, R.; Lutz, B.;
Stalla, G. K.; Pagotto, U. J. Clin. Invest. 2003, 112, 423.
´
2. Rinaldi-Carmona, M.; Barth, F.; Heaulme, M.; Shire, D.;
Calandra, B.; Congy, C.; Martinez, S.; Maruani, J.;
´
Neliat, G.; Caput, D.; Ferrara, P.; Soubrie, P.; Breliere,
J. C.; Le Fur, G. FEBS Lett. 1994, 350, 240.
´
`
13. In support of the knockout mice study, 14 was first
evaluated in C57BL/6 wild-type mice. Ad libitum fed
12-week-old male mice (n = 12 per group) were dosed
3. For extensive reviews in this area, see: (a) Muccioli, G. G.;
Lambert, D. M. Curr. Med. Chem. 2005, 12, 1361;
(b) Hertzog, D. L. Expert Opin. Ther. Pat. 2004, 14,
1435; (c) Smith, R. A.; Fathi, Z. Idrugs 2005, 8, 53;
(d) Lange, J. H. M.; Kruse, C. G. Curr. Opin. Drug Discov.
Dev. 2004, 7, 498; (e) Lange, J. H. M.; Kruse, C. G. Drug
Discov. Today 2005, 10, 693.
4. Meurer, L. C.; Finke, P. E.; Mills, S. G.; Walsh, T. F.;
Toupence, R. B.; Debenham, J. S.; Goulet, M. T.;
Wang, J.; Tong, X.; Fong, T. M.; Lao, J.; Schaeffer,
orally with 14 at 1, 3, or 10 mpk in
a 0.225%
methylcellulose/10% Tween 80 vehicle ꢀ30 min prior
to dark phase of the light cycle. At the highest dose,
14 significantly (P < 0.0001) inhibited 2 h food intake
(ꢀ49% reduction), overnight food intake (ꢀ40% reduc-
tion), and overnight (18 h) gains in body weight
(ꢀ143% reduction).