Synthesis and evaluation of imidazo[1,5-a]pyrazines as corticotropin releasing hormone receptor ligands

Article abstract of DOI:10.1016/S0960-894X(01)00741-7

A novel series of imidazo[1,5-a]pyrazines was synthesized and evaluated as corticotropin releasing hormone (CRH) receptor ligands. SAR studies focused primarily on dialkylamino side chain optimization. SAR of the aryl and small alkyl substituents was also explored.

Full text of DOI:10.1016/S0960-894X(01)00741-7

Bioorganic & Medicinal Chemistry Letters 12 (2002) 291–294
Synthesis and Evaluation of Imidazo[1,5-a]pyrazines as
Corticotropin Releasing Hormone Receptor Ligands
Richard A. Hartz,a,* Paul J. Gilligan,a,* Kausik K. Nanda,a Andrew J. Tebben,a
Larry W. Fitzgeraldb,y and Keith Millerb
aDuPontPharmaceuticalsCompany, Chemical andPhysicalSciences, ExperimentalStation, POBox80500, Wilmington, DE19880, USA
bDuPont Pharmaceuticals Company, CNS Diseases Research, 500 S Ridgeway Ave., Glenolden, PA 19036, USA
Received 12 September 2001; accepted 26 October 2001
Abstract—A novel series of imidazo[1,5-a]pyrazines was synthesized and evaluated as corticotropin releasing hormone (CRH)
receptor ligands. SAR studies focused primarily on dialkylamino side chain optimization. SAR of the aryl and small alkyl sub-
stituents was also explored. # 2002 Bristol-Myers Squibb Company. Published by Elsevier Science Ltd. All rights reserved.
Corticotropin releasing hormone (or factor) is a 41
amino acid neuropeptide that serves as the primary
regulator of the hypothalamic–pituitary–adrenal axis.1,2
Hypersecretion of CRH is associated with a variety of
endocrine and psychiatric disorders, including depres-
sion,3 anxiety,1,3 anorexia nervosa,3 and post traumatic
stress disorder.3 CRH levels were found to be elevated
in the cerebrospinal fluid of depressed patients relative
to that of a control group.4 In addition, elevated plasma
adrenocorticotropin (ACTH) and cortisol levels have
also been found in a large subset of depressed
patients.1,5 CRH receptors are distributed throughout
the central and peripheral nervous systems. Two receptor
subtypes have been identified and designated as CRH1
and CRH2, each with distinct anatomical localization.
Within the CRH2 receptor subtype class three splice
variants—a, b, and g—have been cloned and sequenced.
however, no control patients were involved in this
study. The data also suggested that R-121,919 does not
impair corticotropin and cortisol release with or without
an exogenous CRH challenge, indicating that its mode
of action does not compromise the stress–hormone sys-
tem. Thus, it appears that small molecule CRH1 recep-
tor antagonists have considerable therapeutic potential
for the treatment of anxiety and depression in humans.
A series of triazolopyridines represented by 2 (Ki=3.6
nM, human) was reported to be potent CRH1 receptor
antagonists.8 Given the good potency of 2, it was of
interest to explore a series of closely related compounds
in our continuing search for novel CRH antagonists. An
imidazo[1,5-a]pyrazine scaffold with a novel substitu-
tion pattern (N3-aryl-1-alkyl-N8,N8-dialkyl-3,8-diamine;
e.g., 3f) was chosen for exploratory studies (Fig. 1). The
choice of substituents in the imidazo[1,5-a]pyrazine
series was based on previous SAR results in the tri-
azolopyridine series, where compound 2 was one of the
most potent compounds in that series.
Preclinical and clinical data suggest that CRH antago-
nists may be useful for the treatment of depression.
Pioneering studies with CP-154,526 (1) indicated that
this small molecule CRH antagonist had activity in rat
models for anxiety or depression.6 More recently, the
results of the first human open-label study of a small
molecule CRH1 receptor antagonist, R-121,919, were
reported.7 Reductions in depression and anxiety scores
were noted using both patient and clinician ratings;
It was envisioned that the imidazo[1,5-a]pyrazine het-
erocyclic ring system could be used in place of the tria-
zolopyridine scaffold with the five- and six-membered
rings of the core heterocycle transposed relative to 2. A
proposed overlay is shown in Figure 2.9 At the outset of
this study, it was not known what impact the altered
electronic properties of the core heterocycle would have
on the biological activity of these compounds.
*Corresponding authors. Tel.:+1-302-695-1236; fax.:+1-302-695-
1173 (R.A.H.); e-mail: richard.a.hartz@dupontpharma.com;
tel.: +1-302-695-1439; fax: +1-302-695-1173 (P.J.G.); e-mail: paul.j.
gilligan@dupontpharma.com
yPresent address: 301 Henrietta Street, Kalamazoo, MI 49007-4940,
USA.
The general route employed for the preparation of this
series of N3-aryl-1-alkyl-N8,N8-dialkyl-imidazo[1,5-a]pyr-
0960-894X/02/$ - see front matter # 2002 Bristol-Myers Squibb Company. Published by Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00741-7

292
R. A. Hartz et al. / Bioorg. Med. Chem. Lett. 12 (2002) 291–294
azine-3,8-diamines is described in Scheme 1. The syn-
thesis begins with 2-chloropyrazine. The aryl ring was
deprotonated and acylated according to a procedure
developed by Queguiner to afford desired alcohol 5.10
The secondary alcohol was treated with methanesulfo-
nyl chloride followed by displacement of the resulting
mesylate with sodium azide to give 6. Reduction of the
azide with triphenylphosphine afforded the correspond-
ing amine, which was subsequently treated with an aryl
isocyanate to furnish urea 7. Urea 7 was readily cyclized
upon treatment with POCl3 to form the imidazole
ring.11 The yield for this reaction was quite variable
depending on the workup conditions employed. The
optimized workup procedure involved removal of the
POCl3 by evaporation under reduced pressure. The
residue was then taken up in ethyl acetate, cooled to
0 ꢀC, and was treated with a solution of 1 N NaOH to
afford the desired product (8) in good yield (>70%).12
In some cases, the aniline was alkylated at this point.
Finally, the chloride was displaced with a variety of
alkyl amines to furnish the desired product (3).
The CRH receptor binding affinities for this series of
imidazo[1,5-a]pyrazines are shown in Table 1. Binding
affinities were determined by displacement of [125I]Tyr-
o-CRF from rat frontal cortex homogenates by our test
compounds.13 a-Helical CRF9-41 was found to have a
Ki=7.6Æ0.8 nM (n=3) in this assay.
Three different alkyl substituents were investigated at
the R1 position, which is believed to fit into a small
hydrophobic pocket of the pharmacophore model.3 An
ethyl group was found to be optimal (cf. 3f vs 3a and
3j). The SAR at R2 was probed with hydrogen, methyl,
and ethyl groups. There was no clear trend in binding
affinity among these three substituents.
A more detailed investigation of the SAR at the R3
position was undertaken in an effort to optimize the
binding affinity and protein binding properties. The R3
substituent is believed to fit into a large hydrophobic
pocket of the receptor. Structure–activity relationship
studies on previous series found that the large hydro-
phobic pocket is capable of accommodating a variety of
alkyl substituents, thus providing the opportunity to
optimize binding affinity and physical properties.3
Inspection of the results shown in Table 1 reveals that,
in general, the optimal substituent at R3 is diethylamine,
with the most potent compound in the series being 3f
(Ki=127 nM). Compounds with a dipropylamino sub-
stituent at R3 were slightly less potent, all other sub-
stituents being equal. However, analogues with smaller
(e.g., 3h) or larger (e.g., 3c) alkyl substituents were sig-
nificantly less potent.
Figure 1. Development of the imidazo[1,5-a]pyrazines as CRH1
receptor ligands.
Two different substituents were investigated at the aryl
position. The 2,4-dichlorophenyl and 2,4,6-trimethyl-
phenyl substituents were chosen based on optimized
Scheme 1. Reagents and conditions: (a) n-BuLi, TMPH, THF then
R1CHO, 49–68%; (b) MsCl, Et3N, CH2Cl2, 0 ꢀC to rt, 88–99%; (c)
NaN3, DMF, rt, 79–86%; (d) PPh3, H2O, THF, 50 ꢀC; (e) ArNCO,
Et3N, EtOH, 67–89%, two steps; (f) POCl3, 75 ꢀC, 43–97%; (g) NaH,
R2I, DMF, 90%; (h) R2NH, pyridine, 85 ꢀC, 50–90%.
Figure 2. Overlay of compounds 2 and 3f.

R. A. Hartz et al. / Bioorg. Med. Chem. Lett. 12 (2002) 291–294
293
Table 1. Rat receptor binding affinities of imidazo[1,5-a]pyrazines
pyridine moiety. Calculations14 reveal significant
differences in the electrostatic potential about the core
heterocycles. (Fig. 3). Future studies are planned to
further explore the correlation between the electronic
properties of the core heterocycle with CRH receptor
binding affinity.
Compd R1 R2
R3
Aryl
Mean Ki
(nM, rat)a
n
Acknowledgements
The authors gratefully acknowledge Anne Marshall and
Susan Keim for in vitro binding studies.
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
3o
3p
3q
3r
Me
Me
Me
Me Me
Me Et
Et
Et
Et
Et
Pr
Pr
H
H
H
NEt2
NPr2
NH(3-pentyl)
NPr2
NPr2
2,4-Cl2–Phb
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4,6-Me3–Ph
2,4,6-Me3–Ph
485Æ250
597Æ228
>3000
1361Æ830
671Æ126
127Æ45
3
3
3
2
2
5
3
3
3
3
3
3
3
4
3
2
3
3
3
H
H
Et
Et
H
H
NEt2
NPr2
NMe2
NPr2
NEt2
NPr2
NEt2
NPr2
NMe2
NEt2
NEt2
References and Notes
223Æ110
2884Æ720
1540Æ30
825Æ274
492Æ260
324Æ3.8
623Æ207
2420Æ1445
2220Æ947
143Æ1
1. (a) Gilligan, P. J.; Hartig, P. R.; Robertson, D. W.; Zaczek,
R. In Annual Reports in Medicinal Chemistry; Bristol, J. A.,
Ed.; Academic: San Diego, 1997; Vol. 32, p 41. (b) Owens,
M. J.; Nemeroff, C. B. Pharmacol. Rev. 1991, 43, 425.
2. DeSouza, E. B.; Grigoriadis, D. E. In Psychopharmacology;
The Fourth Generation of Progress; Bloom, F. E., Kupfer, D.
J., Eds.; Raven: New York, 1995; p 505.
3. Gilligan, P. J.; Robertson, D. W.; Zaczek, R. J. Med.
Chem. 2000, 43, 1641, and references cited therein.
4. (a) Banki, C. M.; Bisette, G.; Arato, M.; O’Conner, L.;
Nemeroff, C. B. Am. J. Psychiatry 1987, 144, 873. (b) Zobel,
Pr Me
Pr Me
Pr
Pr
Et
Et
Pr
Et
Et
H
H
H
NPr2
NEt2
173Æ49
2,4,6-Me3–Ph 1140Æ61
a-Helical CRF9-41
7.6Æ0.8
aBinding affinities were determined by displacement of [125I]Tyr-o-
CRF from rat frontal cortex homogenates by our test compounds.
The standard deviation is also reported.
A. W.; Nickel, T.; Kunzel, H. E.; Ackl, N.; Sonntag, A.; Ising,
¨
M.; Holsboer, F. J. Psych. Res. 2000, 34, 171, and references
cited therein. (c) Nemeroff, C. B.; Widerlov, E.; Bissette, G.;
Walleus, H.; Karlsson, E.; Eklund, K.; Kilts, D. C.; Loosen,
P. T.; Vale, W. Science 1984, 226, 1342.
bPh, phenyl.
5. Banki, C. M.; Karmasci, L.; Bissette, G.; Nemeroff, C. B. J.
Affect. Disord. 1992, 25, 39.
6. (a) Chen, Y. L.; Mansbach, R. S.; Winter, S. M.; Brooks,
E.; Collins, J.; Corman, M. L.; Dunaiskis, A. R.; Faraci, W. S.;
Gallaschun, R. J.; Schmidt, A.; Schulz, D. W. J. Med. Chem.
1997, 40, 1749. (b) Mansbach, R. S.; Brooks, E. N.; Chen,
Y. L. Eur. J. Pharmacol. 1997, 323, 21. (c) Arborelius, L.;
Skelton, K. H.; Thrivikraman, K. V.; Plotsky, P. M.; Schulz,
D. W. J. Pharmacol. Exp. Ther. 2000, 294, 588.
7. Zobel, A. W.; Nickel, T.; Kunzel, H. E.; Ackl, N.; Sonntag,
¨
A.; Ising, M.; Holsboer, F. J. Psych. Res. 2000, 34, 171.
8. Bakthavatcahalam, R.; Arvanitis, A. G.; Gilligan, P. J.;
Olsen, R. E.; Robertson, D. W.; Trainor, G. L.; Smith, S. C.;
Fitzgerald, L. W.; Zaczek, R.; Shen, H.; Christ, D. D.; ACS
National Meeting, Boston, MA, Aug 23–27, 1998; MEDI 134.
9. The overlay was created using the drug discovery module
of Cerius 2, Version 4.5.
Figure 3. A comparison of the electrostatic potentials of 3f and 2.
10. Ple, N.; Turck, A.; Heynderickx, A.; Queguiner, G. Tetra-
hedron 1998, 54, 9701.
11. Abushanab, E.; Bindra, A. P.; Lee, D.-Y.; Goodman, L.
J. Heterocycl. Chem. 1975, 12, 211.
SAR from previous work.8 In this series, analogues with
a 2,4,6-trimethylphenyl substituent were similar in
potency to the corresponding analogues with a 2,4-
dichlorosubstituted phenyl substituent (e.g., 3f and 3p).
12. In some cases, the chloride on the pyrizine ring was
replaced with an iodide in the starting material. The reaction
worked equally well and gave the corrosponding chloride as
the product due to displacement of the iodide by chloride. A
representative experimental procedure follows: A solution of
N-(2,4-dichlorophenyl)-N0 -[1-(3-iodo-2-pyrazinyl)ethyl]urea
(568 mg, 1.30 mmol) in POCl3 (5 mL) was heated at 75 ꢀC for
1.5 h. The mixture was then cooled to rt and concentrated.
The residue was treated with cold H2O (10 mL) followed by
the addition of cold 1 N NaOH until basic. The mixture was
transferred to a separatory funnel and the aqueous layer was
extracted with ethyl acetate (3ꢁ50 mL). The combined organic
layers were washed with brine, dried over MgSO4, filtered and
concentrated. The residue was purified by column chromato-
In conclusion, a novel series of imidazo[1,5-a]pyrazines
was synthesized and investigated as potential CRH1
receptor antagonists. Several members of this series
were found to have modestly potent CRH1 receptor
binding affinity. The most potent member within this
series is 3f. In addition, this study illustrates that the
electronic properties of the core heterocycle may be cri-
tical to CRH1 receptor binding affinity. Based on
these results, it appears that the imidazo[1,5-a]pyra-
zine core is not a good bioisostere for the triazolo-

294
R. A. Hartz et al. / Bioorg. Med. Chem. Lett. 12 (2002) 291–294
graphy on neutral silica gel (Davisil Grade 643, 30% ethyl
acetate in hexanes with 0.5% methanol>50% ethyl acetate in
hexanes with 0.5% methanol) to afford 8-chloro-N-(2,4-
dichlorophenyl)-1-methyl-imidazo[1,5-a]pyrazine-3-amine (305
mg, 72% yield) as a yellow solid: mp 184–185 ꢀC; 1H NMR
(300 MHz, CDCl3) d 7.41 (d, J=2.6 Hz, 1H), 7.30 (d, J=5.1
Hz, 1H), 7.19 (d, J=5.1 Hz, 1H), 7.14 (dd, J=8.8, 2.2 Hz,
1H), 6.97 (d, J=8.8 Hz, 1H), 6.43 (s, 1H), 2.80 (s, 3H); LRMS
(APCI) m/e 327.0 [(M + H)+, calcd for C13H10N4Cl3 327.0].
13. De Souza, E. B. J. Neurosci. 1987, 7, 88.
14. Electron density was calculated in Gaussian94 using the
HF/3-21G basis set. Density was rendered using Molden 3.6.