Design, synthesis and structure-activity relationships of novel imidazolo[1,2-a]pyrimid-5-ones as potent GnRH receptor antagonists

Article abstract of DOI:10.1016/S0960-894X(02)00371-2

SAR studies of lead GnRH receptor antagonists 2a and 2b reported earlier resulted in the discovery of compound 10b which showed much higher potency (K_i=4.6 nM, compared with 2b, K_i=230 nM) in which the 7-position of the imidazolo[1,2-a]pyrimidone core was substituted with a methyl group, and the ester at the 6-position was replaced by the 3-methoxyphenyl group.

Full text of DOI:10.1016/S0960-894X(02)00371-2

Bioorganic & Medicinal Chemistry Letters 12 (2002) 2185–2187
Design, Synthesis and Structure–Activity Relationships of
Novel Imidazolo[1,2-a]pyrimid-5-ones as Potent GnRH
Receptor Antagonists
Timothy D. Gross,a Yun-Fei Zhu,a,* John Saunders,a Keith M. Wilcoxen,a
a
Yinghong Gao,a Patrick J. Connors, Jr.,a ZhiqiangGuo, R. Scott Struthers,b
b
GregJ. Reinhart and Chen Chena,*
aDepartment of Medicinal Chemistry, Neurocrine Biosciences, Inc., 10555 Science Center Drive, San Diego, CA92121, USA
bDepartment of Exploratory Discovery, Neurocrine Biosciences, Inc., 10555 Science Center Drive, San Diego, CA92121, USA
Received 22 February 2002; accepted 26 April 2002
Abstract—SAR studies of lead GnRH receptor antagonists 2a and 2b reported earlier resulted in the discovery of compound 10b
which showed much higher potency (Ki=4.6 nM, compared with 2b, Ki=230 nM) in which the 7-position of the imidazolo[1,2-
a]pyrimidone core was substituted with a methyl group, and the ester at the 6-position was replaced by the 3-methoxyphenyl group.
# 2002 Elsevier Science Ltd. All rights reserved.
We have previously described the identification of pyrrolo-
pyrimidones 11 and 2-arylimidazolo[1,2-a]pyrimidone
series 22 as potent human GnRH receptor antagonists.
Initial structure–activity relationship studies of 2-aryl-
imidazolo[1,2-a]pyrimidones showed that the bulky ester
Scheme 1. Commercial 2-amino-5-bromo-6-methylpyr-
imid-4-one 4 was treated with a-bromo-40-methoxy-
acetophenone in the presence of sodium hydride in
anhydrous DMF at room temperature, followed by
ammonium hydroxide to give the desired 8H-imida-
zolo[1,2-a]pyrimid-5-one 5, which was purified by crystal-
lization. Compound 5 was alkylated with 2-fluorobenzyl
bromide in the presence of tetrabutylammonium fluoride
in DME at room temperature. The desired compound 7
was separated from the minor O-alkylated by-product
using chromatography on silica gel with ethyl acetate/
hexanes. Palladium catalyzed cross-couplingreaction of
the bromide 7 with a variety of arylboronic acids under
Suzuki conditions gave the corresponding 6-arylimida-
zolo[1,2-a]pyrimidones 8 after a chromatographic pur-
ification. Alternatively, the bromo compound 5 could be
first subjected to the Suzuki reaction with arylboronic
acids to give 6,4 which were then alkylated with alkyl
halides to give compounds 8. The structures 7 and 8
were confirmed by NOE experiments in which the NOE
between the benzylic proton and the methyl group at
the 7-position of the bicyclic system was clearly
observed. The structure of 7 was further confirmed by
single crystal X-ray structure. Compounds 7 and 8 were
then subjected to a Mannich reaction with a variety of
secondary amines and formaldehyde in acetic acid to
afford the desired products 9 and 10.5
substituent at the 6-position and
a
para-iso-
butyroylamidophenyl group at the 2-position of the
bicyclic core are important for high binding affinity, and
resulted in the discovery of a potent analogue (2a) with
a Ki of 7.5 nM. Compounds from this series are much
more stable than pyrrolopyrimidones 1 under acidic
conditions, however, the ester functional group is
hydrolyzed in vivo. For example, 1 had a t1/2 of only
28 min in rat plasma when incubated at 37 ꢀC, and the
major metabolite was the correspondingacid. In order
to replace the labile ester group at the 6-position, some
heterocycles such as oxazole and oxazoline were suc-
cessfully incorporated at this position of the pyrrolo[1,2-
a]pyrimidones.1 In this paper, we describe the design and
synthesis of the 6-arylimidazolo[1,2-a]pyrimidones as
potent antagonists of the human GnRH receptor.
8-Alkylated imidazolo[1,2-a]pyrimid-5-ones 8 were syn-
thesized based on a cyclization reaction3 as outlined in
*Correspondingauthor. Fax: +1-858-658-7619; e-mail: cchen@
neurocrine.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00371-2

2186
T. D. Gross et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2185–2187
The synthesized compounds were evaluated in the
cloned human and rat GnRH receptor assays for their
ability to displace the bindingof [ 125I-Tyr5,DLeu6,
NMeLeu7,Pro9-NEt]GnRH agonist.6 All compounds
with inhibition better than 50% at 10 mM were titrated
on a 6-point curve in a duplicate, and the results are
reported as Ki values.7 The human GnRH receptor
bindingdata for the analogues 10a–t are reported in
Tables 1 and 2.
and 10f, Ki=110 and 370 nM, respectively). Replace-
ment of the bromo group of compound 9, which had a
Ki of 2.5 mM, with a methylenedioxyphenyl group
resulted in over 200-fold increase of activity (10d, Ki=
11 nM). All other phenyl substitutents such as 3-fluoro-,
Table 1. Structure–activity relationships of 5-aryl substitution
Although 2a has good potency, the 3-pentyl ester is
potentially too labile to be used in a drugcandidate.
Since this ester may be functioningboth as a lipophilic
group and a hydrogen bonding acceptor, we designed
the phenyl group bearing a hydrogen bonding acceptor
group to replace it. The prototypical compound in this
series was the 2-(4-methoxyphenyl)-3-{N-[2-(2-pyridyl)
ethyl]-N-methylamino}methyl-6-(3-methoxyphenyl)-7-
methyl-8-(2-fluorobenzyl)imidazolo[1,2-a]pyrimid-5-one
10b, which was found to have high affinity for the
human GnRH receptor (Ki=4.6 nM). This compound
has a 50-fold increase in bindingaffinity over the
homologous 2b (Ki=230 nM). Changing the 3-methoxy
group to the 2- or 4-position of the phenyl group resulted
in less active compounds (10a and 10c, Ki=65 and
18 nM, respectively). Incorporation of an additional
methoxy at the 4- or both 4- and 5-positions of the
phenyl ringdecreased activity almost 25- or 80-fold ( 10e
Compd
R
h-GnRH Ki (nM)
9
Bromo
2-Methoxyphenyl
3-Methoxyphenyl
4-Methoxyphenyl
2500
65
4.6
18
10a
10b
10c
10d
10e
10f
10g
10h
10i
10j
3,4-Methylenedioxyphenyl
3,4-Dimethoxyphenyl
3,4,5-Trimethoxyphenyl
3-Fluorophenyl
3-Trifluoromethylphenyl
3,5-Dichlorophenyl
3-Thiophenyl
11
110
370
79
140
100
56
Scheme 1. Reagents and conditions: (a) a-bromo-40-methoxyacetophenone, NaH, DMF, rt; then NH4OH, rt; (b) ArB(OH)2, Pd(PPh3)4, Na2CO3,
benzene–water–ethanol; (c) TBAF/THF-DME, then 2-fluorobenzyl bromide, rt; (d) R1R2NH, CH2O, AcOH, rt to reflux.

T. D. Gross et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2185–2187
Table 2. Effects of 3-alkylamino substituents
2187
In a conclusion, a series of 2-(4-methoxyphenyl)-7-
methyl-8-(2-fluorobenzyl)imidazolo[1,2-a]pyrimid-5-ones
exemplified by 10a–10t was discovered as havinggood
bindingaffinity for the human GnRH receptor. The
results of the SAR study suggest that the 2-(2-pyr-
idyl)ethyl group on the 3-aminomethyl functionality of
the imidazolo[1,2-a]pyrimidone core structure is
required for optimum human GnRH receptor binding
affinity. To replace the potentially labile ester group of
2, we designed an alkoxy substituted phenyl group
which, possibly in combination with the methyl at the 6-
position, also increased bindingaffinity 50-fold. From
these studies a number of 2-(4-methoxyphenyl)-3-di-
alkylaminomethyl-6-aryl-7-methyl-8-(2-fluorobenzyl)
imidazolo[1,2-a]pyrimid-5-ones havinglow nanomolar
bindingaffinity for the human GnRH receptor has been
identified. More importantly, 6-aryl substituted imid-
azolo[1,2-a]pyrimidones were very stable when they
were incubated with either 0.2 N HCl or rat plasma at
37ꢀC for 2 h. Further studies detailingthe consequences
of their antagonistic effects towards the GnRH receptor8
in vitro and in vivo will be reported elsewhere.
Compd
R1NR2
h-GnRH Ki (nM)
10d
10k
10l
N-Methyl-N-[(2-pyridyl)ethyl]amino
N-Butyl-N-methylamino
N-(Methoxyethyl)-N-methylamino
N-(Diethylaminoethyl)-N-methylamino
N-Benzyl-N-methylamino
11
220
150
130
47
110
67
10m
10n
10o
10p
10q
10r
10s
10t
N-Phenethyl-N-methylamino
N-(2-fluorophenethyl)-N-methylamino
N-Methyl-N-[(4-pyridyl)ethyl]amino
Morpholin-1-yl
4-Methylhomopiperazin-1-yl
N-(1-Methylpiperidin-4-yl)-N-methylamino
1400
28,000
9500
Inactive
3-trifluoromethyl- or 3,5-dichlorophenyl resulted in
some loss of activity compared with the methoxy group
substitution (10g, 10h and 10i, Ki=79, 140 and 100 nM,
respectively). The activity of the heteroaromatic 3-thio-
phenyl substituted analogous was comparable to the
2-methoxyphenyl compound. (10j, Ki=56 nM). These
results indicate the aromatic ringbearinga hydrogen-
bondingacceptor substituent, such as 3-methoxyphenyl,
at the 6-position of the imidazolo[1,2-a]pyrimid-5-one
core, with the possible combination of the 7-methyl
group, is important for high GnRH receptor activity.
Acknowledgements
This work was supported, in part, by National Institutes
of Health grants 1-R43-HD38625–01 and 2-R44-
HD38625–02.
References and Notes
1. (a) Zhu, Y.-F.; Struthers, R. S.; Connors, P. J., Jr.; Gao,
Y.; Gross, T. D.; Saunders, J.; Wilcoxen, K.; Reinhart, G. J.;
Ling, N.; Chen, C. Bioorg. Med. Chem. Lett. 2002, 12, 399. (b)
Zhu, Y.-F.; Wilcoxen, K.; Saunders, J.; Guo, Z.; Gao, Y.;
Connors, P. J., Jr.; Gross, T. D.; Tucci, F. C.; Struthers, R. S.;
Reinhart, G. J.; Chen, C. Bioorg. Med. Chem. Lett. 2002, 12,
403.
2. Wilcoxen, K.; Zhu, Y.-F.; Connors, P. J. Jr.; Saunders, J.;
Gross, T. D.; Gao, Y.; Reinhart, G. J.; Struthers, R. S.; Chen
C. Bioorg. Med. Chem. Lett. 2002, 12, 2179.
The effect of variation in the basic side chain at the
3-position in 10, usingthe methylenedioxyphenyl series
as the template, was examined (Table 2). The analogue,
havingan N-butyl-N-methylaminomethyl group 10k,
was found to have moderate affinity for the human
GnRH receptor (Ki=220 nM). Introduction of a polar
methoxy or diethylamino substituent at the b-position
of basic amine center had similar bindingaffinity ( 10l
and 10m, Ki=150 and 130 nM, respectively). However,
replacement of the butyl group of 10k with a benzyl
group caused almost 5-fold increase of activity (10n,
Ki=47 nM), which was reduced in the phenethyl analogue
(10o, Ki=110 nM). Interestingly, activity was restored
by introduction of a fluorine at the 20-position of the
phenethyl group (10p, Ki=67 nM), when the phenyl
group of 10o was replaced by the 2-pyridyl, the com-
pound gave much better binding affinity (10d,
Ki=11 nM), whereas the 4-pyridyl substitution showed
a dramatic decrease in activity (10q, Ki=1.4 mM).
Finally, cyclic amines, such as morpholin, 1-methylpiper-
idine or 4-methylhomopiperizine showed either low
activity or no activity at all (10r, 10s, and 10t,
Ki=28 mM, 9.5 mM, and inactive, respectively).
3. An example of this cyclization was published recently, see:
Laneri, S.; Sacchi, A.; Abignente, E. J. Heterocycl. Chem.
2000, 37, 1265.
4. This couplingagve low yield of the desired product
Debromonation of 5 was the major by-product isolated.
6.
5. All final compounds were characterized proton NMR and
LC-MS and passed the purity of >85%.
6. Perrin, M. H.; Haas, Y.; Rivier, J. E.; Vale, W. W. Mol.
Pharmacol. 1983, 23, 44.
7. On each assay plate a standard antagonist of comparable
affinity to those beingtested was included as a control for
plate-to-plate variability. Overall, Ki values were highly
reproducible with an average standard deviation of 45% for
replicate Ki determinations.
8. For antagonist activity compounds were tested for their
ability to inhibit GnRH stimulated calcium flux in HEK 293
cells stably expressingthe human GnRH receptor. Repre-
sentative compound from this series was able to fully block
GnRH (10 nM) stimulated Ca++ flux at a concentration of
1 uM antagonist, and did not show any evidence of agonistic
activity.
All compounds showed very poor bindingaffinity
against the rat GnRH receptor. For example, 10d had a
Ki of 11 nM on the human GnRH receptor, but showed
only a Ki of 12 mM on the rat GnRH receptor.