Synthesis and structure-activity relationships of pyrido[3,2-b]pyrazin- 3(4H)-ones and pteridin-7(8H)-ones as corticotropin-releasing factor-1 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2012.06.034

Pyrido[3,2-b]pyrazin-3(4H)-ones and pteridin-7(8H)-ones were evaluated as corticotropin-releasing factor-1 receptor antagonists. The synthesis, SAR studies and pharmacokinetic evaluation of these analogs are described herein.

Full text of DOI:10.1016/j.bmcl.2012.06.034

Bioorganic & Medicinal Chemistry Letters 22 (2012) 4986–4989
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and structure–activity relationships of pyrido[3,2-b]pyrazin-3(4H)-
ones and pteridin-7(8H)-ones as corticotropin-releasing factor-1 receptor
antagonists
Carolyn D. Dzierba a, , Thais M. Sielecki a, Argyrios G. Arvanitis a, Amy Galka a, Tricia L. Johnson a,
⇑
Amy G. Takvorian a, Maria Rafalski a, Padmaja Kasireddy-Polam a, Shikha Vig a, Bireshwar Dasgupta a,
Ge Zhang b, Thaddeus F. Molski b, Harvey Wong c, Robert C. Zaczek b, Nicholas J. Lodge b,
Andrew P. Combs a, Paul J. Gilligan a, George L. Trainor a, Joanne J. Bronson a, John E. Macor a
a Neuroscience Discovery Chemistry, Research and Development, Bristol-Myers Squibb Co., 5 Research Parkway, Wallingford, CT 06492, United States
b Neuroscience Discovery Biology, Research and Development, Bristol-Myers Squibb Co., 5 Research Parkway, Wallingford, CT 06492, United States
c Pharmaceutical Candidate Optimization, Research and Development, Bristol-Myers Squibb Co., 5 Research Parkway, Wallingford, CT 06492, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Pyrido[3,2-b]pyrazin-3(4H)-ones and pteridin-7(8H)-ones were evaluated as corticotropin-releasing fac-
tor-1 receptor antagonists. The synthesis, SAR studies and pharmacokinetic evaluation of these analogs
are described herein.
Received 4 May 2012
Accepted 11 June 2012
Available online 16 June 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Corticotropin-releasing factor
Depression
Anxiety
Corticotropin Releasing Factor (CRF), a 41 amino acid peptide
first isolated in 1981,1 is considered to be one of the principal reg-
ulators of the hypothalamic–pituitary–adrenal (HPA) axis, which
coordinates the endocrine, behavioral and autonomic responses
to stress. CRF mediates its action through binding to two well char-
acterized, class B subtype G-protein coupled receptors, CRF-R1 and
CRF-R2, which are widely distributed throughout the central and
peripheral nervous systems. In response to stress, CRF is released
from the hypothalamus and binds to CRF receptors in the anterior
pituitary, resulting in release of adrenocorticotropic hormone
(ACTH). Increased ACTH levels stimulate release of cortisol from
the adrenal cortex. Cortisol then mediates a variety of metabolic
and behavioral changes that facilitate adaptation to stressful stim-
uli. Compelling evidence suggests that hypersecretion of CRF, and
the consequent over-stimulation of the stress response, contrib-
utes to development of stress-related disorders such as depression
and anxiety [for recent reviews, see: Refs. 2–4]. To date, CRF-R1 has
been the most extensively studied CRF receptor as a potential ther-
apeutic target, with both preclinical and clinical studies suggesting
that antagonists of CRF-R1 may offer promise in treatment of
stress-related disorders [Refs. 5–9]. Although recent depression
and generalized anxiety clinical trials have not given positive re-
sults for a number of CRF antagonists, interest still remains in this
target, for example, for use in the treatment of alcohol and drug
withdrawal [for recent reviews, see: Refs. 10–12]. Herein, we dis-
close the synthesis and structure–activity relationship (SAR) of a
series of pyrido[3,2-b]pyrazin-3(4H)-ones and pteridin-7(8H)-ones
as selective CRFR1 antagonists.
The synthesis of the pyrido[3,2-b]pyrazin-3(4H)-ones was car-
ried out as shown in Scheme 1 starting with commercially avail-
able 2,4-dihydroxypyridine (1).13 Nitration of the 3-position,
selective chlorination of the 4-position with POCl3 followed by
benzyl protection afforded 2-(benzyloxy)-4-chloro-3-nitropyridine
(4) in good yield. Various aryl groups were installed at the 4-posi-
tion of 4 via Suzuki coupling to give intermediates 5. Deprotection
of the 2-benzyl ether (5), conversion to the chloride (7) and subse-
quent displacement with a variety of amines gave the functional-
ized pyridyl intermediate (8). After reduction of the nitro group,
the second ring was assembled by condensation with methyl 2-
oxopropanoate (10). With an efficient synthesis in hand, a series
of analogs were prepared to develop the SAR for this chemotype.
Initially we examined the effect of varying the alkyl group (R)
on the 4-position of the ring system (Table 1). In accordance with
what is known about the pharmacophore model of the CRFR1
⇑
Corresponding author. Tel.: +1 203 677 7661; fax: +1 203 677 7702.
E-mail address: carolyn.dzierba@bms.com (C.D. Dzierba).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.06.034

C. D. Dzierba et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4986–4989
4987
HO
N
HO
N
HO
N
BnO
O2N
N
between the ethyl and cyclopropyl group extending off of the
branch point. A methoxy group can also be incorporated into this
side chain without a significant loss in potency (compare 10c,
IC50 = 62 nM and 10g, IC50 = 63 nM). In contrast to what was ob-
served with the 1-cyclopropylpropyl enantiomers, there was a sig-
nificant difference in potency between the two 1-methoxy-2-butyl
enantiomers, with an ꢀ18-fold preference for the R-enantiomer
(10h, (R), IC50 = 4.0 nM; 10i, (S), IC50 = 74 nM).
Next we examined the SAR of the appended aryl ring. Based on
studies of related chemotypes,14 we focused mainly on phenyl and
pyridyl groups with 2,4-di- or 2,4,5-trisubstitution (Table 2). In one
example, it was shown that substitution of both meta-positions
can lead to a significant loss in potency (compare 2,3,4,5-tetrasub-
stituted analog 11c, IC50 = 340 nM with 2,4-disubstituted analog
a
b
c
O2N
O2N
OH
1
OH
2
Cl
3
Cl
4
BnO
O2N
N
HO
N
Cl
N
d
f
e
g
O2N
O2N
Aryl
5
Aryl
6
Aryl
7
R
R
R
HN
N
h
HN
N
i
N
N
O
O2N
H2N
N
Aryl
8
Aryl
9
Aryl
10
11a,
IC50 = 8.8 nM
or
2,4,5-trisubstituted
analog
11b,
Scheme 1. Reagents and conditions: (a) HNO3, 80 °C, 10 min, 90% yield; (b)
cyclohexyl amine, MeOH, 15 min; then POCl3, DMF, 78% yield; (c) Ag2NO3, BnBr,
PhH, 5 h, 79% yield; (d) PdCl2(PPh3)2, Na2CO3, ArylB(OH)2, EtOH/tol, 80 °C, 60–90%
yields; (e) TFA, CH2Cl2, 60–90% yields; (f) POCl3, DMF, 95 °C, 60–90% yields; (g)
RNH3Cl, DIEA, n-BuOH, 50–90% yields; (h) Pt/S/H2 50 psi, EtOH, 75–95% yields; (i)
methyl 2-oxopropanoate, EtOH, reflux, 30–75% yields.
IC50 = 5.2 nM). In the 4-position (R2, Table 2) smaller groups were
preferred (compare OMe analog 11h, IC50 = 0.68 nM with OEt ana-
log 11i, IC50 = 2.0 nM and OiPr analog 11j, IC50 = 15 nM). It is likely
that groups in this position fill a relatively small binding pocket.
Both electron withdrawing and donating groups in the 2-position
(R1) provide good potency, however the smaller 2-F group were
not well tolerated (11o, IC50 = 52 nM). It was thought that the role
of the 2-substituent was to induce a twist between this pendant
aryl ring and the bicyclic core, an orientation that seemed to be
necessary for efficient binding. The 5-position (R3) is tolerant of
groups with a variety of steric and electronic properties including
methyl (11h, IC50 = 0.68 nM), cyano (11q, IC50 = 4.8 nM) and F
(11p, IC50 = 0.82 nM). Finally, it was shown that pyridyl groups
could be incorporated in this region. In one case with 2,4-disubsti-
tution, the pyridyl analog (11u, IC50 = 76 nM) lost potency relative
to the phenyl counterpart (11t, IC50 = 11 nM), while in an example
Table 1
SAR of the alkyl group
R
N
O
N
N
Cl
Cl
a,b
Compounds
R
rCRFRI IC50 (nM)
10a
3500 ( 240)
Table 2
10b
10c
10d
380 ( 60)
62 ( 30)
8.2 ( 2.1)
SAR of the aryl group
O
N
N
N
R
S
R1
10e
8.8 ( 4.7)
X
R2
R3
10f
10g
10h
4.3 ( 0.2)
63 ( 12)
4.0 ( 1.3)
a,b
Compounds
ent
R1
X
R2
R3
rCRFR1 IC50 (nM)
O
11a
11b
11c
11d
11e
11f
11g
11h
11i
rac
rac
rac
S
S
S
Me
Me
Me
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Me
Me
Cl
Me
F
CH
CH
Me
Me
Me
CF3
OCHF2
OMe
OMe
OMe
OEt
OiPr
OMe
OCHF2
OMe
OMe
OMe
NMe2
Me
H
8.8 ( 2.3)
5.2 ( 0.33)
340 ( 120)
21 ( 3.9)
R
S
Me
Me
H
H
H
O
O
CMe
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
N
10i
74 ( 23)
7.3 ( 2.4)
4.0 ( 0.92)
4.5 ( 0.96)
0.68 ( 0.31)
2.0 ( 1.0)
15 ( 1.0)
R
H
a
Values are means of two or three experiments, standard deviation is given in
parentheses.
rac
rac
rac
R
rac
R
Me
Me
Me
Cl
Cl
F
F
F
F
CN
Me
Cl
H
H
b
The IC50 of o-CRF is 2.9 1.0 nM in this assay.
11j
11k
11l
4.5 ( 2.0)
9.7 ( 1.1)
1.7 ( 0.37)
9.4 ( 2.1)
52 ( 7.2)
binding site14 relatively small lipophilic groups in this position
tended to be preferred. Branching at the point of attachment was
favored; for example the isobutyl substituent (10a, IC50 = 3500 nM)
was ꢀ10-fold less potent than the 1,2-dimethyl propyl group (10b,
IC50 = 380 nM). Extending the chain in a linear arrangement led to
an additional improvement in potency (10c, IC50 = 62 nM). Cyclo-
propyl substituents in this position were generally very well toler-
ated, with the slightly longer 1-cyclopropylbutyl analogs (10f,
IC50 = 4.3 nM) being modestly more potent than 1-cyclopropylpro-
pyl analogs. Interestingly, there was no potency difference be-
11m
11n
11o
11p
11q
11r
11s
11t
R
rac
rac
rac
rac
rac
R
R
R
R
Cl
Cl
0.82 ( 0.11)
4.8 ( 0.69)
1.1 ( 0.61)
3.4 ( 1.4)
11 ( 4.2)
OMe
OMe
Me
Me
Me
Me
Cl
OMe
OMe
OMe
OMe
OMe
11u
11v
11w
76 ( 11)
130 ( 60)
15 ( 2.8)
CH
N
Me
Me
a
Values are means of three experiments, standard deviation is given in
parentheses.
tween the
R (10d, IC50 = 8.2 nM) and S (10e, IC50 = 7.4 nM)
b
The IC50 of o-CRF is 2.9 1.0 nM in this assay.
enantiomers. This may simply be due to the small steric difference

4988
C. D. Dzierba et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4986–4989
Table 3
a, b
or
Comparison of pyrido[3,2-b]pyrazin-3(4H)-ones vs pteridin-7(8H)-ones analogs
R
R
N
Cl
N
c
b,a
HN
N
O
N
R
N
N
N
N
H2N
O
X
H2N
N
Cl
Aryl
Aryl
Y
N
12
13
14
Aryl
Scheme 2. Reagents and conditions: (a) PdCl2(PPh3)2, Na2CO3, ArylB(OH)2, EtOH/
tol, reflux, 5 h; (b) RNH3Cl, DIEA, AcCN, 40–80% yields for two steps; (c) methyl 2-
oxopropanoate, EtOH, reflux, 20–55% yields.
Compounds Aryl
X,Y
R
rCRFRI
IC50, nM
cLogP R PPB
%free
10e
14a
N,CH
N,N
7.4 ( 2.9) 4.9
24
( 0.80)
55 ( 7.6)
0.2
0.4
Cl
Cl
4.1
HO
Cl
Cl
20a
CH,N
5.1
-
Cl
Cl
b
e
a
d
e
10h
14b
20b
N,CH
N,N
CH,N
4.0 ( 1.3) 4.1
97 ( 13) 3.4
150 ( 60) 4.4
0.4
2.2
-
N
OH
N
N
Aryl
17
O
O2N
O2N
O2N
Cl
15
16
11l
14c
N,CH
N,N
9.7 ( 1.1) 4.9
38 ( 1.4)
0.3
0.40
R
R
R
N
4.2
HN
HN
O
Cl
N
Aryl
18
N
Aryl
19
N
O
F
O2N
H2N
N
Aryl
20
F
21
14d
N,CH
N,N
3.0 ( 1.4) 3.8
37 ( 1.4)
3.0
1.4
3.1
Scheme 3. Reagents and conditions: (a) POCl3, reflux, 85% yield; (b) PdCl2(PPh3)2,
Na2CO3, ArylB(OH)2, EtOH/tol, 80 °C, 50–75% yields; (c) RNH3Cl, DIEA, n-BuOH, 60–
80% yields; (d) Na2S2O4, EtOH, H2O, NH4OH 60–80% yields; (e) methyl 2-oxopro-
panoate, EtOH, reflux, 30–50% yields.
O
2.1
( 0.32)
16
O
O
Cl
22
N,CH
N,N
3.5
2.7
3.0
1.4
14e
F
O
( 0.50)
1.5
( 0.33)
7.5
with 2,4,5-trisubstitution, the pyridyl analog (11w, IC50 = 15 nM)
showed improved potency over the phenyl counterpart (11v,
IC50 = 130 nM).
Cl
23
N,CH
N,N
3.4
2.6
<0.2
2.80
14f
F
O
( 0.05)
Although compounds such as 11h (IC50 = 0.68 nM) and 11p
(IC50 = 0.82 nM) showed excellent binding potency, they were both
very highly protein bound to plasma proteins (rat fu <0.2%) and we
wanted to make structural modifications to try to increase the free
fraction. To this end, we investigated a related series of compounds
in which a second nitrogen was incorporated into the core. Calcu-
lated logP values for the pteridin-7(8H)-ones were on average 0.5–
1 log units lower than for the pyrido[3,2-b]pyrazin-3(4H)-ones,
suggesting that incorporation of a second N into the ring system
would reduce the liophilicity and lead to a decrease in plasma pro-
tein binding.
O
O
O
O
N
N
N
N
N
N
N
Cl
Me
F
Cl
OMe
14f
OMe
24
The synthesis of the pteridin-7(8H)-ones was carried out as
shown in Scheme 2. Starting from 5-amino-4,6-dichloropyrimidine
(12), the aryl group was introduced via a Suzuki coupling reaction
to give 13. The amino group was subsequently incorporated by dis-
placement of the chlorine to provide 14. Owing to the symmetry of
the starting material, the two side chains could be introduced in
either order with similar yields, allowing for more efficient analog
synthesis. The pyrazinone ring was then assembled by condensa-
tion with methyl 2-oxopropanoate.
In addition to evaluating the pteridin-7(8H)-one analogs, a
subset of pyrido[4,3-b]pyrazin-2(1H)-ones was prepared for evalu-
ation as shown in Scheme 3. Starting from 2,4-dihydroxy-3-nitro-
pyridine, bis-chlorination followed by displacement of the 2-Cl
gave intermediate 17. The 4-Cl was displaced with amines,
followed by reduction of the nitro group to afford diamines 19.
Condensation with methyl 2-oxopropanoate provided pyrido[4,3-
b]pyrazin-2(1H)-one analogs 20.
A comparison of the 5-N-pyridopyrazinone and 7-N-pyridopyr-
azinone vs pteridinone analogs is shown in Table 3. Comparing the
2,4-dichlorophenyl substituted compounds, the 5-N-pyridyl analog
(10e, IC50 = 7.4 nM) was ꢀ3-fold more potent than the pyrimidyl
analog (14a, IC50 = 24 nM). Additionally, the 5-N-pyridyl analog
(10e) was ꢀ7-fold more potent than the 7-N-pyridyl analog (20a,
IC50 = 55 nM). Although 14a was shown to be less potent than
10e, the lower cLogP of 14a did translate into an improvement
Figure 1. Comparison of a pteridin-7(8H)-one and a pyrido[3,2-b]pyrazin-3(4H)-
one.
in rat plasma protein binding (14a, fu = 0.4% vs 10e, fu = 0.2%). An-
other cohort of three 2,4-dichlorophenyl substituted compounds
with an (R)-1-methoxybutan-2-yl side chain showed a similar
trend with the 5-N-pyridyl analog (10h, IC50 = 4.0 nM, fu = 0.4%)
showing the best potency, followed by the pyrimidyl analog
(14b, IC50 = 97 nM; fu = 2.2%), with the 7-N pyridyl analog (20b,
IC50 = 150 nM) showing the poorest potency. Again, the pyrimidyl
analog showed a higher plasma free fraction relative to the pyridyl
analog. Analogs with 2,4,5-trisubstitution were also examined. The
5-N-pyridyl analog (11l, IC50 = 9.7 nM) was ꢀ4-fold more potent
than pyrimidyl analog (14c, IC50 = 38 nM). Comparison of addi-
tional related cohorts showed that the 5-N pyrido[3,2-b]pyrazin-
3(4H)-ones were generally 3–10-fold more potent than the corre-
sponding pteridin-7(8H)-ones and >7-fold more potent than the
corresponding 7-N-pyrido[4,3-b]pyrazin-2(1H)-ones.
Compounds 14f and 24 (Fig. 1) were examined in pharmacoki-
netic studies to further characterize the in vivo properties of these
CRF1 antagonists. Compound 14f (IC50 = 7.6 nM; fu = 2.2%) was
administered to rats at an intravenous dose of 0.5 mg/kg (n = 2)
and at an oral dose of 1.5 mg/kg (n = 2). The following pharmacoki-
netic parameters were derived non-compartmentally from the

C. D. Dzierba et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4986–4989
4989
study: Cl = 2.02 L/hr/kg, Vss = 2.05 L/kg, AUC = 300 nM Â h/L (po),
Supplementary data
t1/2 = 2.0 h (iv). Compound 14f appeared to be a moderate/high
clearance compound in rats. Using the same conditions, compound
24 (IC50 = 5.5 nM; fu = 0.4%) was administered to rats at an intrave-
nous dose of 0.5 mg/kg (n = 2) and at an oral dose of 1.5 mg/kg
(n = 2). The following pharmacokinetic parameters were derived
non-compartmentally from the study: Cl = 0.57 L/h/kg, Vss = 1.8 L/
kg, AUC = 8500 nM Â h/L (po), t1/2 = 5.8 h (iv). Compound 24 ap-
peared to be a low clearance compound in rats. These data are typ-
ical of what was observed for other sets of related compounds,
where the pteridine-based analogs tended to show less attractive
in vivo metabolic stability compared to their pyridopyrazinone-
based counterparts. Based on the data presented herein, the pyr-
ido[3,2-b]pyrazin-3(4H)-ones appeared to be superior in both po-
tency and pharmacokinetic parameters as compared to the
pteridin-7(8H)-ones analogs, although the pteridin-7(8H)-ones
had better plasma free fractions.
In summary, novel classes of pyrido[3,2-b]pyrazin-3(4H)-ones
and pteridin-7(8H)-ones were synthesized and investigated as po-
tential CRFR1 ligands. SAR investigation has shown that these
compounds are potent ligands for CRFR1. While the pyrido[3,2-
b]pyrazin-3(4H)-one analogs generally showed improved potency
over pteridin-7(8H)-ones, the pteridin-7(8H)-ones had improved
plasma protein binding. By balancing the potency, plasma protein
binding and pharmacokinetic parameters through modification of
the side chains, attractive CRFR1 ligands such as compound 24
have been discovered. Further optimization and characterization
of these series will be reported in due course.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.
06.034. These data include MOL files and InChiKeys of the most
important compounds described in this article.
References and notes
1. Vale, W.; Spiess, J.; Rivier, C.; Rivier, J. Science 1981, 213, 1394.
2. Paezpereda, M.; Hausch, F.; Holsboer, F. Expert Opin. Investig. Drugs 2011, 20,
519.
3. Holsboer, F.; Ising, M. Eur. J. Pharmacol. 2008, 583, 350.
4. Kehne, J. H. CNS Neurolog. Disord. Drug Targets 2007, 6, 163.
5. Hemley, C. F.; McCluskey, A.; Keller, P. A. Curr. Drug Targets 2007, 8, 105.
6. Dzierba, C. D.; Hartz, R. A.; Bronson, J. J. Ann. Rep. Med. Chem. 2008, 43, 3.
7. Ising, M.; Holsboer, F. Exp. Clin. Psychopharmacol. 2007, 15, 519.
8. Valdez, G. R. Curr. Pharm. Design 2009, 15, 1587.
9. Zorrilla, E. P.; Koob, G. F. Drug Disc. Today 2010, 15, 371.
10. Lowery, E. G.; Thiele, T. E. CNS Neurol. Disord. Drug Targets 2010, 9, 77.
11. Logrip, M. L.; Koob, G. F.; Zorrilla, E. P. CNS Drugs 2011, 25, 271.
12. Shaleva, U.; Erbb, S.; Shaham, Y. Brain Res. 2010, 1314, 15.
13. T.M. Sielecki-Dzurdz, A.G. Arvanitis, C.D. Dzierba, Patent WO 031,189 A1 2004.
14. (a) Dzierba, C. D.; Takvorian, A. G.; Rafalski, M.; Kasireddy-Polam, P.; Wong, H.;
Molski, T. F.; Zhang, G.; Li, Y.-W.; Lelas, S.; Peng, Y.; McElroy, J. F.; Zaczek, R. C.;
Taub, R. A.; Combs, A. P.; Gilligan, P. J.; Trainor, G. L. J. Med. Chem. 2004, 47,
5783–5790; (b) Hartz, R. A.; Ahuja, V. T.; Arvanitis, A. G.; Rafalski, M.; Yue, E.
W.; Denhart, D. J.; Schmitz, W. D.; Ditta, J. L.; Deskus, J. A.; Brenner, A. B.;
Hobbs, F. W.; Payne, J.; Lelas, S.; Li, Y.-W.; Molski, T. F.; Mattson, G. K.; Peng, Y.;
Wong, H.; Grace, J. E.; Lentz, K. A.; Qian-Cutrone, J.; Zhuo, X.; Shu, Y.-Z.; Lodge,
N. J.; Zaczek, R.; Combs, A. P.; Olson, R. E.; Bronson, J. J.; Mattson, R. J.; Macor, J.
E. J. Med. Chem. 2009, 52, 4173–4191.