Structure activity relationship studies of carboxamido-biaryl ethers as opioid receptor antagonists (OpRAs). Part 2

Article abstract of DOI:10.1016/j.bmcl.2007.10.025

A series of 6-bicycloaryloxynicotinamides were identified as opioid receptor antagonists at mu, kappa, and delta receptors. Compounds in the 6-(2,3,4,5-tetrahydro-1H-benzo[c]azepin-7-yloxy)nicotinamide scaffold exhibited potent in vitro functional antagonism at all three receptors.

Full text of DOI:10.1016/j.bmcl.2007.10.025

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 17 (2007) 6841–6846
Structure activity relationship studies of carboxamido-biaryl
ethers as opioid receptor antagonists (OpRAs). Part 2
Kumiko Takeuchi,* William G. Holloway, Charles H. Mitch, Steven J. Quimby,
Jamie H. McKinzie, Todd M. Suter, Michael A. Statnick, Peggy L. Surface,
Paul J. Emmerson, Elizabeth M. Thomas and Miles G. Siegel
Lilly Research Laboratories, A Division of Eli Lilly and Company, Indianapolis, IN 46285, USA
Received 11 September 2007; revised 1 October 2007; accepted 5 October 2007
Available online 17 October 2007
Abstract—A series of 6-bicycloaryloxynicotinamides were identified as opioid receptor antagonists at mu, kappa, and delta recep-
tors. Compounds in the 6-(2,3,4,5-tetrahydro-1H-benzo[c]azepin-7-yloxy)nicotinamide scaffold exhibited potent in vitro functional
antagonism at all three receptors.
Ó 2007 Elsevier Ltd. All rights reserved.
The prevalence of obesity in the industrialized world,
particularly in the United States, has reached epidemic
proportions.1 Today close to 65% of the adult popula-
tion in the US is overweight (BMI P 25) or obese
(BMI P 30). More alarming trend is the increase in
the number of overweight children and teenagers (10–
15% of the youth population). International data indi-
cate that the epidemic is not isolated to the US but is
in fact a growing global health problem. In other words,
the prevalence of obesity is not only rising in other
developed and affluent countries but also is now spread-
ing to less affluent countries. The prevention and treat-
ment of excess weight is critical for the health of both
individuals and our society not only because obesity in-
creases the risk of a number of diseases, such as type 2
diabetes mellitus, dyslipidaemia, hypertension, and cor-
onary heart disease, but also because it disposes an enor-
mous economic impact on our society. Medical
intervention along with lifestyle modification is now
considered a necessity to prevent the growing and mul-
tilateral negative consequences of overweight and
obesity.
removed from the market, and orlistat and sibutramine,
the current approved drugs, have limited clinical use due
to their poor efficacy and significant side effect profiles.2
Most recently, Rimonabant was approved for the treat-
ment of obesity in Europe and a high hope was raised
for this CB1 antagonist as a new therapeutic agent that
held promise not only to treat obesity but also to allevi-
ate other metabolic syndromes.3 But its submission to
FDA in the US was withdrawn just prior to its initially
expected approval when concerns over its side effects
especially on patients with history of depression were
raised and its risk/benefit ratio was questioned by the
FDA Advisory Board, who unanimously voted down
its approval as a result. Continual research and develop-
ment of an antiobesity agent with favorable side effect
profiles for a long-term use is, therefore, warranted
and required. We have reported that nonselective opioid
receptor antagonist (OpRA) LY255582 and its analogs
demonstrated potent anorectic activity, reducing body
fat in obese rats by decreasing food intake and stimulat-
ing lipid utilization.4 LY255582, however, precluded its
clinical development due to its poor oral bioavailability
and unacceptable margin of safety resulting from irrita-
tion at the site of drug administration via other routes.
We then began to search an orally efficacious OpRA
with activity comparable to LY255582 and acceptable
safety profiles for the treatment of obesity and recently
discovered new carboxamido-biaryl ethers that were
structurally unrelated to morphine (Fig. 1). We identi-
fied and reported5 6-(4-(2-benzylaminoethyl)phen-
oxy)nicotinamide (1) as an initial lead having good
Flurries of research activities to discover effective and
safe antiobesity agents have spurred in recent years
as fenfluramine and dexfenfluramine (Fen–Phen) were
Keywords: Opioid receptor antagonists (OpRAs); Carboxamido-biaryl
ethers; Obesity.
*
Corresponding author. Tel.: +1 317 276 6771; fax: +1 317 433
0715; e-mail: ktak@lilly.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.10.025

6842
K. Takeuchi et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6841–6846
OH
OH
O
O
()n
()m
NH2
N
R
OH
H
A2
O
A1
N
N
OH
(A1, A2 = C or N)
H
Morphine
Carboxamido-biaryl ethers
LY255582
Figure 1. Carboxamido-biaryl ethers, novel opioid receptor antagonists structurally unrelated to morphine.
Table 1. Opioid receptor antagonism of carboxamido-biaryl ethers
O
( )
( )m
n N
H
NH2
O
N
Compound
m
n
Receptor binding affinity at high Na, Ki (nM)6
GTPcS functional antagonism, Kb (nM)7
Mu
Kappa
Delta
Mu
2.25 0.47
0.07 0.02
Kappa
Delta
1
2
1
1
2
7.44 2.94
0.17 0.02
0.15 0.01
0.87 0.02
136.26 59.55
8.38 2.30
4.68 0.83
5.28 0.12
70.21
18.41 7.99
1.15 0.24
0.32 0.03
2.99 0.10
23.14 6.88
0.97 0.14
1.19 0.21
11.06 0.32
2
1.16 0.23
4.82 0.49
16.31 0.28
LY255582
Naltrexone
0.043 0.008
0.59 0.02
The data are expressed in mean SEM where the assay run n P 2 and value without SEM where n = 1.
binding affinities at mu, kappa, and delta opioid recep-
tors (Table 1). We also reported how the carbon chain
length and the position of the amine nitrogen between
the two phenyl rings affected the receptor binding affin-
ities and identified a more potent antagonist 2 at all
three opioid receptor subtypes.
Compounds with these modifications were prepared as
shown in Scheme 1.8 3,4-Dimethylphenyl methyl ether
was dibrominated with NBS in the presence of benzoyl
peroxide radical initiator in CCl4 at reflux in 32% (Route
A). Treatment with benzylamine in the presence of ben-
zyltriethylammonium chloride in 50% NaOH/toluene
solution afforded N-benzylisoindolin-5-yl methyl ether
in a 71% yield. Demethylation with 48% HBr at reflux
followed by etherification with 6-chloronicotinamide
and K2CO3 in N,N-dimethylacetamide/toluene solution
at reflux produced the compound 3 in 40%. Debenzyla-
tion by hydrogenolysis followed by alkylation with 2-
phenethylbromide and Et3N in DMF at 70 °C furnished
4. Compounds 5–7 were prepared from 3-methoxyphen-
ethylamine, which was first converted to a tetrahydroiso-
quinoline intermediate with paraformaldehyde and
CH3COCl in 88% HCO2H as shown in Route B.
Demethylation and N-BOC protection set up the inter-
mediate ready for etherification with 6-chloronicotina-
mide (47% for 3 steps). BOC-deprotection with TFA in
One of the concerns raised in our early SAR exploration
of this series of compounds was the stability of unsubsti-
tuted benzylic carbon of the side chain. We therefore
considered two aspects of further SAR exploration: (1)
conformational constraint of the benzylic nitrogen atom
in a ring system with one of the neighboring phenyl
rings, e.g., as isoindoline, tetrahydroisoquinoline, or
benzoazepine; and (2) replacement of the terminal ben-
zylic amine tether with other amine tether. For the first
modification there are two possible sites to constrain the
amine nitrogen in a ring system: (a) tie the nitrogen into
the middle phenyl ring; or (b) attach it to the terminal
phenyl ring (Fig. 2).
O
( )
( )m
R
n N
NH2
O
O
(a)
(b)
( )p1
O
N
( )
( )m
n N
H
NH2
O
N
( )
( )m
n N
NH2
( )p2
O
N
(m, n, p1, p2 = No. of CH2 carbon)
Figure 2. Modification of carboxamido-biaryl ether side chain.

K. Takeuchi et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6841–6846
6843
A
1) 48% HBr, Δ
BzNH2
NBS
(PhCO)O2
Br
Br
2) K2CO3
BzEt3N+Cl-
DMA/tol, Δ
6-Cl-nicotinamide
40%
N
O
50% NaOH/tol
71%
O
CCl4, Δ
32%
O
O
O
NH2
1) H2, 10% Pd/C, EtOH
NH2
2) 2-phenethylbromide
N
Et3N, DMF, 70 o
4%
C
N
O
N
O
N
4
3
B
O
1) K2CO3, DMA/tol, Δ
6-Cl-nicotinamide
2) TFA/CH2Cl2
1) (CH2O)n, 88% HCO2H
CH3COCl
2) 48% HBr, reflux
R
N
O
NH2
NH2
51%
O
N
O
N
3) RX, Et3N, DMF, 70 o
C
3) (BOC)2O, Et3N, THF
O
47%
5-7
OH
O
C
1) Cs2CO3, DMF, 100 oC
Ref. 9
O
6-Cl-nicotinamide
2) TFA/CH2Cl2
3) PhCHO, NaHB(OAc)3
AcOH, (ClCH2)2
NH2
N
OH
OHC
OBz
N
O
N
O
8
D
O
O
H
H
O
1) NaH, DMF
6-Cl-nicotinamide
N
O
NH2
O
1) BH3, THF
N
NaN3,
MsOH
50%
O
2) TFA, CH2Cl2
82%
2) BBr3, CH2Cl2
3) (BOC)2O, TEA
18%
O
N
N
O
O
R
OH
N
NH2
RX
K2CO3
DMF
O
N
10,15-26
E
O
O
N
O
O
Conc. HCl
HOAc,
84%
N
1) ClCO2CHClCH3
1) H2, 5% Pd/C
2) LAH, THF
93%
N
O
N
O
Cl
2) (CF3CO)2O
3) BBr3
72%
NaHCO3
CHCl3
O
O
O
O
O
quant.
O
F
1) NaH
6-Cl-nicotinamide
F
F
NH2
R
N
N
2) 7.0 M NH3in MeOH
O
N
O
OH
34%
3) RX, K2CO3, DMF
11-12
F
O
O
O
( )1-2
N
( )1-2
NH2
NH
Reductive amination
NH2
+
O
N
O
N
13--14
Scheme 1. Synthesis of 6-bicycloaryloxynicotinamides.8
CH2Cl2 after the 6-aryloxynicotinamide formation and
alkylation with an appropriate halide (RX) as in Route
A afforded the target compounds. Compounds 8 and 9
were prepared as exemplified by 8 in Route C. N-BOC-
7-hydroxy-3,4-dihydro-1H-isoquinoline was prepared
according to a literature procedure.9 Etherification with
6-chloronicotinamide and Cs2CO3 in DMF at 100 °C,
BOC-deprotection, and reductive amination of benzal-
dehyde with the resultant isoquinoline in the presence
of NaHB(OAc)3 and acetic acid in dichloroethane
yielded 8. Reductive amination of phenylacetaldehyde
similarly afforded the compound 9. The compound 10

6844
K. Takeuchi et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6841–6846
was prepared as shown in Route D. 6-Methoxytetralone
was treated with NaN3 in methanesulfonic acid to form
7-methoxy-2,3,4,5-tetrahydro-1H-benzo[c]azepin-1-one
in a 50% yield.10 Reduction of the amide to the amine
with BH3 in THF, demethylation with BBr3 in CH2Cl2,
and N-BOC protection afforded a precursor phenol for
etherification with 6-chloronicotinamide. N-BOC depro-
tection after the biaryl ether formation and alkylation
with phenethyl bromide and K2CO3 in DMF as de-
scribed above provided 10. Route E describes the synthe-
sis of compounds 11 and 12. An amide formation from 3-
methoxyphenylacetyl chloride with methylaminoacetal-
dehyde dimethyl acetal and NaHCO3 in CHCl3 followed
by acid-catalyzed cyclization afforded 8-methoxy-3-
methyl-1,3-dihydro-benzo[d]azepin-2-one in an excellent
yield. Sequential treatment of the intermediate by (1)
reduction of the double bond, (2) deoxygenation of the
amide oxygen with LAH, (3) N-demethylation with 1-
chloroethyl chloroformate, (4) trifluoroacetylation of
the resultant amine, and (5) O-demethylation with
BBr3 provided N-trifluoroacetyl-2,3,4,5-tetrahydro-1H-
benzo[d]azepin-7-ol in good yields. Treatment of the
alcohol with NaH, etherification with 6-chloronicotina-
mide, deacetylation with 7.0 M NH3 in MeOH, and
alkylation with an appropriate alkyl halide as before
afforded the target compound 11 or 12.
Table 2 shows the results of modification (a). Isoindo-
line incorporation (3 and 4) was most detrimental to
the receptor antagonism activities. Isoquinolinyl deriv-
atives 5–9 also suffered considerable loss in the binding
affinity, especially at kappa and delta receptors. The
trend that we saw before in the effects of terminal chain
lengths5 was translated into the ring system where 6
(n = 2) exhibited best overall activities among 5–7
(n = 1–3). More pronounced effect of the ring system it-
self was observed with the isoquinolin-7-yl analog 9
(m = 2, p1 = 1), which almost completely lost activities
at all three receptors as compared to the isoquinolin-6-
yl 6 (m = 1, p1 = 2). This may be explained by the fact
that the terminal phenethyl chain was forced to change
its location and orientation due to the ring constraint
in 9. Compounds 5 and 8, however, showed similar
activities at all three receptors. This rationale was also
supported by the results of benzoazepinyl derivatives
10–11. The more flexible benzo[c]azepinyl 10 (m = 1,
p1 = 3) regained activities comparable to 1 and 6,
whereas the benzo[d]azepinyl 11 or 12 (m = 2, p1 = 2)
lost considerable activities presumably due to the
change in the orientation of their terminus. Among
the constrained ring systems studied, the compound
10 showed the best overall profiles at the three
receptors.
Table 2. Effects of incorporation of nitrogen atom into a bicyclic system with the middle phenyl ring
O
( )n
( )m
NH2
N
( )p1
N
O
Compound
No. of CH2 carbon
Receptor binding affinity at high Na, Ki (nM)6
m
n
p1
Mu
1008
Kappa
Delta
3
4
1
1
1
1
1
2
2
1
2
2
1
2
1
2
3
1
2
2
1
2
1
1
2
2
2
1
1
3
2
2
>5000
>5000
>5000
658.8 118.5
179.0
15.95
2274 315
2351 201
179.6 27.8
2139 37
2448.94 29.29
>5000
5
2150
1221
6
7
185.0 18.9
231.59
>3333
555.9 112.5
2428.17
3606
8
9
10
11
12
10.09 1.85
405.8
91.49
360.6 34.6
2873
>4897
169.39 5.80
2708 55
1377 70
The data are expressed in mean SEM where the assay run n P 2 and value without SEM where n = 1.
Table 3. Effects of incorporation of nitrogen atom into a bicyclic system with the terminal phenyl ring
O
( )n
( )m
NH2
N
( )p2
N
O
Compound
No. of CH2 carbon
Receptor binding affinity at high Na, Ki (nM)6
GTPcS functional antagonism, Kb (nM)7
m
n
p2
Mu
Kappa
Delta
Mu
Kappa
Delta
13
14
1
1
2
2
2
1
1
1
2
–
nd
5.05 0.92
7.44 2.94
nd
187.97 34.70
136.26 59.55
nd
144.07 4.71
70.21
27.42 4.69
0.97 0.16
2.25 0.47
38.49 7.05
38.87 0.10
18.41 7.99
146.68 24.24
130.99 7.31
23.14 6.88
The data are expressed in mean SEM where the assay run n P 2 and value without SEM where n = 1. nd = not determined.

K. Takeuchi et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6841–6846
6845
Table 4. Side chain SAR of 6-(2,3,4,5-tetrahydro-1H-benzo[c]azepin-7-yloxy)nicotinamide
O
R
N
NH2
O
N
Compound
R
Receptor binding affinity at high Na, Ki (nM)6
GTPcS functional antagonism, Kb (nM)7
Mu
Kappa
Delta
Mu
Kappa
Delta
15
16
17
18
19
20
21
22
23
24
25
26
n-Butyl
n-Pentyl
n-Hexyl
20.81 0.09
11.86 1.60
12.03 1.96
11.14 1.49
7.55 1.41
nd
503.7 29.1
187.00
163.22
4.22 0.36
5.73 2.99
1.94 0.35
2.36 0.10
1.39 0.27
4.14 2.06
9.12 0.18
15.77
11.06 0.33
9.99 0.93
11.28 0.56
5.49 0.13
8.94 2.16
>13
>43
>28
105.88 21.12
116.41 30.74
78.09 7.79
105.95 4.92
318.95 20.94
71.62 20.57
61.65 12.85
nd
5.91 0.68
3-Methylbutyl
4-Methylpentyl
5-Methylhexyl
3,3-Dimethylbutyl
2-Ethylbutyl
232.46 27.76
107.88 9.38
239.77
48.24 21.24
15.22 5.13
>28
11.93 0.52
18.33 2.55
42.03 0.16
24.42 7.06
41.67 6.69
128.70 3.46
215.87 28.05
409.29 59.44
3888 282
388.15 2.46
466.27
7.58 0.30
25.76
>37
>58
>58
4,4,4-Trifluorobutyl
3-Cyclohexylpropyl
2-Morpholin-4-ylethyl
3-Morpholin-4-ylpropyl
3.87 0.18
5.14
>23
>43
>28
703.78 11.43
359.89 51.63
3254 575
186.75
>13
>13
1909
2576
>28
>28
>23
The data are expressed in mean SEM where the assay run n P 2 and value without SEM where n = 1. nd = not determined.
Another possible incorporation of the nitrogen atom is
to incorporate it with the terminal phenyl ring (route
(b), Fig. 2). Two isoquinolinyl derivatives 13 and 14 in
Table 3 were prepared from tetrahydroisoquinoline
and an appropriate aldehyde intermediate (prepared
by slight modification of methods reported earlier5) via
conventional reductive amination (Route F in Scheme
1). The compound 14 that mimics 1 showed comparable
binding affinities with 1, though less potent in the
in vitro functional antagonism at the delta receptor.
The compound 13 showed much less potent functional
activities as compared to 2 presumably due to the differ-
ences in the orientation of the terminal region between
the two. Though these two compounds showed respect-
able in vitro functional antagonism, the compounds
with the nitrogen incorporated into the ring system with
the middle phenyl ring, shown in Table 2, were chosen
for further SAR exploration, as they provided more ver-
satile and simple side chain modifications.
against three opioid receptor subtypes through the
SAR study of carboxamido-biaryl ethers as opioid
receptor antagonists. The 6-(2,3,4,5-tetrahydro-1H-
benzo[c]azepin-7-yloxy)nicotinamide scaffold (Table 4)
holds potential for further modification and develop-
ment to identify potent and metabolically stable opioid
receptor antagonists.
Acknowledgments
We thank scientists at LOB labs at Lilly Research Lab-
oratories for the SPA binding assay data generation.
References and notes
1. A brief summary of the epidemic of obesity and its health
and economic impact, see Stein, C. J.; Colditz, G. A.
J. Clin. Endocrinol. Metab. 2004, 89, 2522, and references
cited therein.
As stated earlier, one of our initial concerns was the
in vivo stability of the benzylic side chains. We therefore
decided to explore the feasibility of alkyl or cycloalkyl
side chains in the benzo[c]azepinyl series based on 10
(Table 2). Alkyl (15–17) or branched alkyl chains (18–
22) showed reasonable binding affinities at the three opi-
oid receptor subtypes, although more variable and less
potent at the kappa and delta receptors. Cycloalkyl ter-
mini such as cyclohexyl (24) and morpholinyl (25–26)
groups were not favored. Most compounds in Table 4
exhibited excellent in vitro antagonism at the mu and
kappa receptors. The compounds 17–19 showed respect-
able in vitro functional antagonism at all three recep-
tors. This suggests that requirement of the amino side
chain terminus of the carboxamido-biaryl ether series
does not limit to the aryl group but rather seems more
related to the hydrophobic or steric environment in or-
der to antagonize the three opioid receptors.
2. Scheen, A. J.; Ernest, P. H. Diabetes Metab. 2002, 28, 437.
3. (a) Patel, P. N.; Pathak, R. Am. J. Health Syst. Pharm.
2007, 64, 481; (b) Bray, G. A.; Ryan, D. H. Gastroenter-
ology 2007, 132, 2239.
4. (a) Mitch, C. H.; Leander, J. D.; Mendelsohn, L. G.;
Shaw, W. N.; Wong, D. T.; Cantrell, B. E.; Johnson, B.
G.; Reel, J. K.; Snoddy, J. D.; Takemori, A. E.;
Zimmerman, D. M. J. Med. Chem. 1993, 36, 2842; (b)
Statnick, M. A.; Tinsley, F. C.; Eastwood, B. J.; Suter, T.
M.; Mitch, C. H.; Heiman, M. L. Am. J. Physiol. Regul.
Integr. Comp. Physiol. 2003, 284, R1399.
5. Takeuchi, K.; Holloway, W. G.; Mitch, C. H.; Quimby, S.
J.; McKinzie, J. H.; Suter, T. M.; Statnick, M. A.; Surface,
P. L.; Emmerson, P. J.; Thomas, E. M.; Siegel, M. G.
Bioorg. Med. Chem. Lett. 2007, 17, 5349.
6. SPA binding affinity for cloned human mu and kappa
(defined by [3H]-diprenorphine binding), and delta
(defined by [3H]-bremazocine binding) opioid receptors
expressed in CHO cells under high sodium conditions.
Under low sodium binding conditions the affinity was
reduced at the receptor subtypes, consistent with findings
on the prototype Lilly 4-phenylpiperidine opioid antago-
In conclusion, we discovered several biaryl ethers having
a wide range of amine tethers with excellent activity

6846
K. Takeuchi et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6841–6846
nist series: see Emmerson, P. J.; McKinzie, J. H.; Surface,
P.; Suter, T. M.; Mitch, C. H.; Statnick, M. A. Eur. J.
Pharmacol. 2004, 494, 121.
mined. See Rodgers, G.; Hubert, C.; McKinzie, J.; Suter,
T.; Statnick, M.; Emmerson, P.; Stancato, L. Assay Drug
Dev. Technol. 2003, 1, 627, for the assay development
method.
7. In vitro functional assay, inhibiting agonist stimulated G-
protein activation (measured using GTPcS binding) in
CHO membranes expressing the cloned human mu,
kappa, and delta receptors. In the present study, opioid
ligand binding and GTPcS functional assays were con-
verted to a homogeneous SPA permitting the develop-
ment of simpler assays with dramatically increased
throughput. Optimization in the presence of sodium
chloride was designed to bias these assays toward the
detection of opioid antagonists. Kb values for com-
pounds with modest binding affinities were not deter-
8. The reactions have not been optimized. The target
molecules were characterized by H NMR, MS, and one
1
of the following methods for purity verification: HRMS,
EA or HPLC. See WO2004026305 (A1) for details of
compound preparation.
9. Kucznierz, R.; Grams, F.; Leinert, H.; Marzenell, K.;
Engh, R. A.; von der Saal, W. J. Med. Chem. 1998, 41,
4983.
10. Shtacher, G.; Erez, M.; Cohen, S. J. Med. Chem. 1973, 16,
516.