Design and synthesis of tricyclic imidazo[4,5-b]pyridin-2-ones as corticotropin-releasing factor-1 antagonists

Article abstract of DOI:10.1021/jm050384+

The synthesis and SAR studies of tricyclic imidazo[4,5-b]pyridin-2-ones as human corticotropin-releasing factor receptor (CRF₁) antagonists are discussed herein. Compound 16g was identified as a functional antagonist that inhibited CRF-stimulat

Full text of DOI:10.1021/jm050384+

5104
J. Med. Chem. 2005, 48, 5104-5107
Design and Synthesis of Tricyclic
Imidazo[4,5-b]pyridin-2-ones as
Corticotropin-Releasing Factor-1
Antagonists
Zhiqiang Guo,*,† John E. Tellew,† Raymond S. Gross,†
Brian Dyck,† Jonathan Grey,† Mustapha Haddach,†
Mehrak Kiankarimi,† Marion Lanier,† Bin-Feng Li,†
Zhiyong Luo,† James R. McCarthy,†
Manisha Moorjani,† John Saunders,†
Robert Sullivan,† Xiaohu Zhang,† Said Zamani-Kord,†
Dimitri E. Grigoriadis,‡ Paul D. Crowe,‡
Ta Kung Chen,§ and John P. Williams†
Departments of Medicinal Chemistry, Pharmacology, and
Preclinical Development, Neurocrine Biosciences, Inc., 12790
El Camino Real, San Diego, California 92130
Received April 22, 2005
Figure 1. Some small-molecule CRF1 receptor antagonists.
Abstract: The synthesis and SAR studies of tricyclic imidazo-
[4,5-b]pyridin-2-ones as human corticotropin-releasing factor
receptor (CRF1) antagonists are discussed herein. Compound
16g was identified as a functional antagonist that inhibited
CRF-stimulated cyclic adenosine monophosphate production
and CRF-induced adrenocorticotrophic hormone release. Phar-
macokinetics studies in rats showed that 16g was orally
bioavailable, had good brain penetration, and had a moderate
half-life. In our effort to identify CRF1 antagonists with
improved pharmacokinetics properties, 16g exhibited a favor-
ably lower volume of distribution.
Figure 2. Design of tricyclic CRF1 receptor antagonists.
and several of them exhibited in vivo efficacy in the
anxiety models when dosed orally, most of these com-
pounds reported earlier suffer from high lipophilicity
and poor water solubility.8 For example, 1 has a very
long half-life (51 h) and high volume of distribution
(105 L/kg) in rats.9 More recent efforts on the discovery
of more hydrophilic CRF1 antagonists generated 5
(NBI-30775/R121919),10 which has antidepressant ef-
ficacy in an open label phase IIA clinical trial in major
depressive disorder patients.
Compounds 1-5 are based on monocyclic or bicyclic
cores. We envisioned that synthesizing compounds with
a third ring on the core structure would restrict the
conformational freedom of their flexible “top region”
(Figure 2) and thus would result in compounds that
would populate a smaller subset of the biologically
relevant conformations required for optimal biological
activity. We recently reported on a series of conforma-
tionally constrained tricyclic pyrazolo[4,3-b]pyridines
such as 6, which was found to be a potent functional
antagonist of CRF1.11 In our effort to prepare potent
CRF1 antagonists with reduced lipophilicity, we have
made tricyclic analogues of the polar 1,3-dihydroimi-
dazo[4,5-b]pyridin-2-one series12 (Figure 2), and the
initial SAR and pharmacokinetics profiles of represen-
tative compounds from this series of CRF1 antagonists
are reported herein.
Corticotropin-releasing factor (CRF, also known as
corticotropin-releasing hormone), a 41 amino acid neuro-
peptide isolated from mammalian brain, is the prime
regulator of the hypothalamic-pituitary-adrenal (HPA)
stress response.1 CRF mediates its actions through high-
affinity binding to its receptors, CRF1 and CRF2, both
of which are members of the class B subfamily of
G-protein-coupled receptors. Prolonged activation of the
central CRF system may play a fundamental role in the
etiology of major depression.2 Hypersecretion of hypo-
thalamic CRF manifests itself in a down-regulation of
CRF receptors in the anterior pituitary as demonstrated
by the blunted adrenocorticotrophic hormone (ACTH)
response to peripherally administered CRF in severely
depressed patients. Because of the key role that CRF
plays in stress response, centrally acting CRF1 antago-
nists are expected to have utility in treatment of stress-
related psychiatric disorders such as anxiety and de-
pression.3
Many non-peptide CRF1 antagonists from different
chemical classes have been reported in recent years, and
several representative examples are illustrated in Fig-
ure 1. Potent CRF1 antagonists such as 1 (CP-154,526),4
2 (antalarmin),5 3 (NBI-27914),6 and 4 (DMP696)7 have
been widely studied in many different animal models
of CRF-related behavior. Although these compounds
possess excellent in vitro activities as CRF1 antagonists,
For the synthesis of tricyclic analogues we employed
two general methods. In the first method (Scheme 1), a
substituted pyridine ring was prepared (13), followed
by construction of the imidazolone ring (15), and finally
closure of the tetrahydropyrazine ring. The starting
material 2,4-dichloro-6-methyl-3-nitropyridine (9) was
readily prepared from commercially available 4-hy-
* To whom correspondence should be addressed. Phone:
1-858-617-7657. Fax: 1-858-617-7619. E-mail: zguo@neurocrine.com.
† Department of Medicinal Chemistry.
‡ Department of Pharmacology.
§ Department of Preclinical Development.
10.1021/jm050384+ CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/09/2005

Letters
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 16 5105
Scheme 1a
Scheme 2a
a Reagents and conditions: (a) concentrated H2SO4, 65 °C, 50%;
t
(b) R1Br or R1I, NaH or BuOK, DMF, 20-50 °C, 15-70%.
Table 1. Binding Affinities of 16a-l to the Human CRF1
Receptor
a Reagents and conditions: (a) POCl3, ACN, 82 °C, 85%;
(b) 4-aminoheptane, Et3N, ACN, -10 °C, 40%; (c) PhSO2Cl, Et3N,
DMAP, 60 °C, 91%; (d) 4-aminoheptane, TsOH, acetonitrile,
68 °C, 80%; (e) POCl3, acetonitrile, DIPEA, 90 °C, 87%; (f) ArNH2,
ACN, 65 °C, 95%; (g) Na2S2O4, THF/water, room temp, 91%;
(h) triphosgene, NEt3, 0 °C, 80%; or carbonyldiimidazole, NEt3,
DMF, 76%; (i) 1,2-dibromoethane, DMF, NaH, room temp, 30-
80%.
droxy-6-methyl-3-nitro-2(1H)-pyridone (8) and POCl3.12
9 was then treated with 4-aminoheptane in acetonitrile
at -10 °C to give a 3:1 mixture of 10 and its regioisomer
(10a). In an improved version of this method, 8 was
selectively monosulfonylated with benzenesulfonyl chlo-
ride, and the resulting sulfonate intermediate 11 was
directly treated with 4-aminoheptane. This one-pot
procedure yielded 12 in 73% yield with no purification
issues. Chlorination of 12, followed by reaction with
substituted anilines at elevated temperatures (∼140 °C),
afforded the condensation products 13. The 5-nitro
group of 13 was reduced with sodium hydrosulfite to
provide the corresponding diaminopyridines 14, which
were cyclized with triphosgene to form the imidazolone
ring. The third ring was installed by treatment of 15
with 1,2-dibromoethane using sodium hydride as base
in DMF or sodium hydroxide as base under phase-
transfer conditions.
With the above synthetic routes, the upper amino
substituent R1 was fixed at the beginning of the
synthesis. We developed an alternative approach (Scheme
2) to explore the SAR at this position while keeping the
bottom Ar substituent constant. Dealkylation of 16g13
was carried out in neat concentrated sulfuric acid to
yield the secondary amine 17a. Top region R1 groups
were then installed via base-mediated alkylation of
intermediate 17a to yield the final compounds 17b-e.
Compounds 18a-c and 19a-c were also prepared
according to the synthetic route in Scheme 1 with the
3-pentyl or 1-methoxy-2-butyl group as the top amino
substituent.
a Receptor binding was conducted as described previously. Data
are the average of three or more independent determinations.
Typical standard errors were less than 30%.
expressed in leukocyte tyrosine kinase cells with sau-
vagine as the radiolabeled ligand, and the Ki values
were determined from dose-response curves using
concentrations from 1 nM to 10 µM as described.14
The binding activity data of substitutions at the
bottom aryl region of the tricyclic imidazolone series are
summarized in Table 1. The top amino substituent (R1)
was initially kept constant as 4-heptyl, since this
substituent was potent in our other tricyclic series (i.e.,
6).11 The SAR pattern that emerged was in part similar
to that seen for other CRF1 antagonist scaffolds in that
2,4-disubstituted aryl groups gave good potency, par-
ticularly with lipophilic and relatively small substitu-
ents (16a,b). Analogue 16a with a 2,4-dichlorophenyl
group demonstrated a potent Ki of 5.1 nM, while the
slightly bigger 2-bromo-4-isopropyl analogue 16b was
even more potent (Ki ) 1.1 nM). Our goal, however, was
to identify polar substituents to help offset the contribu-
tion of the amino substituent (R1) to the overall lipo-
The compounds thus synthesized were tested for
binding affinity at the cloned human CRF1 receptor

5106 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 16
Letters
Table 2. Binding Affinities of 17b-e, 18a-c, and 19a-c to
the Human CRF1 Receptor
analogue had very good affinity (17d, 8 nM), even
though it was still about 4-fold less potent than 16g.
Insertion of an oxygen atom into the five-carbon chain
of 17d (as shown in 17e) decreased affinity further.
Further studies were carried out to determine whether
the SAR of the bottom aryl region would be different if
the heptyl substituent of 16 was replaced by a smaller,
less lipophilic group. Binding data for representative
3-pentyl compounds 18a,b and methoxybutyl com-
pounds 19a,b are also presented in Table 2. The results
in general paralleled those seen in the 4-heptyl series,
and the uniformally lower affinity of these compounds
served to confirm the requirement for a lipophilic upper
amino substituent for optimal potency. Adding a 2-aryl
substitution to reimburse the loss of affinity from the
top region (as shown in 18c and 19c) failed to improve
the binding affinity.
Compound 16g was very weakly basic with a pKa of
4.9, and it possessed moderate lipophilicity with a log P
value of 4.5, which was comparable to the reported
value10 for 5 (log P ) 4.9). Even though the free base
had limited solubility in water (0.34 mg/mL), its L-malic
acid salt was readily soluble in 1 N HCl (38 mg/mL).
This compound had essentially no affinity for the CRF2
receptor expressed in Chinese hamster ovary (CHO)
cells at 10 µM.15 16g showed no significant binding at
10 µM to over 60 different receptors, ion channels, and
transporters. As a consequence, 16g was further evalu-
ated for functional CRF1 antagonism in vitro and in
vivo. Compound 16g dose-dependently inhibited CRF-
stimulated cyclic adenosine monophosphate (cAMP)
production in CHO cells expressing the CRF1 receptor16
with an IC50 of 71 nM, which was comparable to that of
5 (IC50 ) 50 nM). In a second functional assay, 16g was
also found to functionally antagonize CRF-stimulated
ACTH release from rat primary anterior pituitary cell
cultures.14 The result showed that 16g was a potent
antagonist in this assay with an IC50 of 150 nM.
We further determined the pharmacokinetics profile
of 16g in male Sprague-Dawley rats, following intra-
venous and oral administration at 10 mg/kg. After an
iv injection, 16g had a high plasma clearance (CL )
53.4 mL min-1 kg-1) and a small volume of distribution
(Vd ) 7.2 L/kg) in rats, which resulted in a terminal
half-life of 1.6 h. After oral administration of 16g, the
Cmax in plasma and brain occurred simultaneously at
1 h after dosing. The mean maximal concentrations
(Cmax) in plasma and brain tissue via oral administra-
tion were 356 ng/mL and 1180 ng/g, respectively. Oral
bioavailability was estimated to be 30%, and on the
basis of the AUC (0-6 h) ratio, the brain tissue was
exposed to approximately 280% of the plasma concen-
tration of 16g at this dosage. 16g had a notably smaller
volume of distribution and shorter terminal half-life in
rats than the previous leading CRF antagonists such
as 1 (Vd ) 105 L/kg, t1/2 ) 51 h)9 and 4 (DMP696, Vd )
22 L/kg, t1/2 ) 16 h).7 In light of these results, 16g was
further evaluated in vivo.
a Receptor binding was conducted as described previously. Data
are the average of three or more independent determinations.
Typical standard errors were less than 30%.
philicity of the compounds. Even though removal of the
2-chloro substitution in 16a resulted in a 2-fold loss of
potency (16c), it demonstrated that the 2-aryl substi-
tuent is not required for good activity in this series.
Accordingly, a series of 4-monosubstituted aryl ana-
logues (16d-g) were prepared to look for simple and
less lipophilic groups as the bottom. Although a strong
electron-withdrawing trifluoromethyl group at the 4-po-
sition was tolerated (16d, 17 nM), the polar, electron-
withdrawing methylsulfone and carboethoxy substitu-
ents dramatically reduced the binding affinity as seen
in 16e and 16f (Ki ) 465 and 920 nM, respectively).
Incorporation of a less lipophilic, electron-donating
methoxy group at the different positions on the bottom
phenyl ring suggested that the 4-position (16g, 2 nM)
was preferred over the 2- or 3-position analogues in
potency. Moreover, extending the 4-methoxy to 4-ethoxy
led to a decrease in potency by 2 orders of magnitude.
Incorporation of a substituted pyridine ring on the
bottom (16k,l) lowered the lipophilicity further but
provided less active analogues than the corresponding
phenyl compounds.
Binding results from variation of the upper region
amino substituent R1 are depicted in Table 2. In our
attempt to reduce overall lipophilicity, we investigated
smaller alkyl replacements for the 4-heptyl group of
16g. From this study, it is clear that the top region
nitrogen substituents of tricyclic 17 participitate in a
key binding interaction with the CRF receptor. Alkyl
substituents smaller than the isopropyl group resulted
in a total loss of binding activity, while n-butyl analogue
17c exhibited only very weak binding. The 3-pentyl
The in vivo functional antagonism of 16g was deter-
mined using the CRF-induced ACTH release assay in
normal rats. As shown in Figure 3, intravenous CRF
(0.3 nmol/kg) administration in rats produced a robust
increase in plasma levels of ACTH (p < 0.0007).

Letters
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 16 5107
(3) O’Bien, D.; Skelton, K. H.; Owen, M. J.; Nemeroff, C. B. Are CRF
receptor antagonists potential antidepressants? Hum. Psycho-
pharmacol. Clin. Exp. 2001, 16, 81-87.
(4) Schulz, D. W.; Mansbach, R. S.; Sprouse, J.; Braselton, J. P.;
Collins, J.; Corman, M.; Dunaiskis, A.; Faraci, S.; Schmidt, A.
W.; Seeger, T.; Seymour, P.; Tingley, F. D., III.; Winston, E. N.;
Chen, Y.; Heym, J. CP-154,526: a potent and selective non-
peptide antagonist of corticotropin releasing factor receptors.
Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 10477-10482.
(5) Webster, E. L.; Lewis, D. B.; Torpy, D. J.; Zachman, E. K.; Rice,
K. C.; Chrousos, G. P. In vivo and in vitro characterization of
antalarmin, a nonpeptide corticotropin-releasing hormone (CRH)
receptor antagonist: suppression of pituitary ACTH release and
peripheral inflammation. Endocrinology 1996, 137, 5747-5750.
(6) Chen, C.; Dagnino, R., Jr.; De Souza, E. B.; Grigoriadis, D. E.;
Huang, C. Q.; Kim, K. I.; Liu, Z.; Moran, T.; Webb, T. R.;
Whitten, J. P.; Xie, Y. F.; McCarthy, J. R. Design and synthesis
of a series of non-peptide high-affinity human corticotropin
releasing factor 1 receptor antagonists. J. Med. Chem. 1996, 39,
4358-4360.
(7) He, L.; Gilligan, P. J.; Zaczek, R.; Fitzgerald, L. W.; McElroy,
J.; Shen, H.-S. L.; Saye, J. A.; Kalin, N. H.; Shelton, S.; Christ,
D.; Trainor, G.; Hartig, P. 4-(1,3-Dimethoxyprop-2-ylamino)-2,
7-dimethyl-8-(2,4-dichlorophenyl)-pyrazolo[1,5-a]-1,3,5-triazine:
a potent, orally bioavailable CRF1 receptor antagonist. J. Med.
Chem. 2000, 43, 449-456.
(8) Hsin, L.-W.; Tian, X.; Webster, E. L.; Coop, A.; Caldwell, T. M.;
Jacobson, A. E.; Chrousos, G. P.; Gold, P. W.; Habib, K. E.; Ayala,
A.; Eckelman, W. C.; Contoreggi, C.; Rice, K. C. CRHR1 receptor
binding and lipophilicity of pyrrolopyrimidines, potential non-
peptide corticotropin-releasing hormone type 1 receptor antago-
nists. Bioorg. Med. Chem. 2002, 10, 175-183.
Figure 3. Dose-dependent effects of 16g on CRF-induced
ACTH release 10 min following CRF administration: (/) p <
0.0007 vs 10 mL/kg 5% D-mannitol (w/v) in water (Veh); (//)
p < 0.002 vs CRF.
Compound 16g significantly attenuated this effect in a
dose-dependent manner with statistically significant
reductions observed at the 10 and 30 mg/kg doses (p <
0.002 vs CRF). The observed attenuation of ACTH was
86% at the 30 mg/kg dose.
In conclusion, we have shown that tricyclic imidazo-
[4,5-b]pyridin-2-one compounds are potent CRF1 an-
tagonists. The antagonistic activity of these compounds
was demonstrated by their high affinities in binding to
the CRF1 receptor and inhibition of CRF-stimulated
cAMP production. A representative from this series,
16g, possessed acceptable physicochemical properties
for use as a central nervous system agent, showed a
good overall in vitro profile, and had appropriate phar-
macokinetics (especially favorably small volume of
distribution) in rats. 16g not only inhibited CRF-
stimulated ACTH release in vitro but also dose-
dependently attenuated the elevation in plasma ACTH
levels induced by CRF in rats in vivo. These results
confirmed that 16g is a functional CRF1 antagonist with
improved physicochemical properties and a good phar-
macokinetics profile. Further evaluation of 16g will be
reported in due course.
(9) Keller, C.; Bruelisauer, A.; Lemaire, M.; Enz, A. Brain pharma-
cokinetics of a nonpeptidic corticotropin-releasing factor receptor
antagonist. Drug Metab. Dispos. 2002, 30, 173-176.
(10) Chen, C.; Wilcoxen, K. M.; Huang, C. Q.; Xie, Y. F.; McCarthy,
J. R.; Webb, T. R.; Zhu, Y.-F.; Saunders, J.; Liu, X. J.; Chen,
T.-K.; Bozigian, H.; Grigoriadis, D. E. Design of 2,5-dimethyl-
3-(6-dimethyl-4-methylpyridin-3-yl)-7-dipropylaminopyrazolo-
[1,5-a]pyrimidine (NBI 30775/R121919) and structure-activity
relationships of a series of potent and orally active corticotropin-
releasing factor receptor antagonists. J. Med. Chem. 2004, 47,
4787-4798.
(11) Gross, R. S.; Guo, Z.; Dyck, B.; Coon, T.; Huang, C. Q.; Lowe, R.
F.; Marinkovic, D.; Moorjani, M.; Nelson, J.; Zamani-Kord, S.;
Grigoriadis, D. E.; Hoare, S. R.; Crowe, P. D.; Bu, J. H.; Haddach,
M.; McCarthy, J.; Saunders, J.; Sullivan, R.; Williams, J. P.
Design and synthesis of tricyclic corticotropin-releasing factor-1
(CRF1) antagonists. J. Med. Chem., in press.
(12) Beck, J. P.; Arvanitis, A. G.; Curry, M. A.; Rescinito, J. T.; F.,
Fitzgerald, L. W.; Gilligan, P. J.; Zaczek, R.; Trainor, G. L. Purin-
8-ones as corticotropin-releasing hormone (CRH-R1) receptor
antagonists. Bioorg. Med. Chem. Lett. 1999, 9, 967-972.
(13) The crystal structure of 16g was determined and is shown in
Figure 4 of Supporting Information.
(14) De Souza, E. B. Corticotropin-releasing factor receptors in the
rat central nervous system: characterization and regional
distribution. J. Neurosci. 1987, 7, 88-100.
Supporting Information Available: Experimental pro-
cedures for the synthesis and characterization of representa-
tive compounds, experimental details of the binding and
functional assays, and X-ray structure for 16g. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(15) Grigoriadis, D. E.; Liu, X. J.; Vaughn, J.; Palmer, S. F.; True,
C. D.; Vale, W. W.; Ling, N.; De Souza, E. B. 125I-Tyro-
sauvagine: a novel high affinity radioligand for the pharmaco-
logical and biochemical study of human corticotropin-releasing
factor2R receptors. Mol. Pharmacol. 1996, 50, 679-586.
(16) Battaglia, G.; Webster, E. L.; De Souza, E. B. Characterization
of corticotropin-releasing factor receptor-mediated adenylate
cyclase activity in the rat central nervous system. Synapse 1987,
1, 572-581.
References
(1) De Souza, E. B.; Grigoriadis, D. E. In Psychopharmacology: The
Fourth Generation of Progress; Bloom, F. E., Kupfer, D. J., Eds.;
Raven: New York, 1995; pp 505-517.
(2) Holsboer, F. The rationale for corticotropin-releasing hormone
receptor (CRH-R) antagonists to treat depression and anxiety.
J. Psychiatr. Res. 1999, 33, 181-214.
JM050384+