Small molecule antagonists of the gonadotropin-releasing hormone (GnRH) receptor: Structure-activity relationships of small heterocyclic groups appended to the 2-phenyl-4-piperazinyl-benzimidazole template

Article abstract of DOI:10.1016/j.bmcl.2009.02.043

A previous report described the serum LH suppression pharmacology of the 2-phenyl-4-piperazinyl-benzimidazole N-ethyluracil GnRH receptor antagonist 1 following oral administration in rats. A series of small heterocycles were appended to the 2-(4-tert-but

Full text of DOI:10.1016/j.bmcl.2009.02.043

Bioorganic & Medicinal Chemistry Letters 19 (2009) 1986–1990
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Small molecule antagonists of the gonadotropin-releasing hormone (GnRH)
receptor: Structure–activity relationships of small heterocyclic groups
appended to the 2-phenyl-4-piperazinyl-benzimidazole template
b
b
c
a
Diane B. Hauze a, , Murty V. Chengalvala , Joshua E. Cottom , Irene B. Feingold , Lloyd Garrick ,
Daniel M. Green a, Christine Huselton c, Wenling Kao a, Kenneth Kees a, Joseph T. Lundquist IV a,
Charles W. Mann a, John F. Mehlmann a, John F. Rogers a, Linda Shanno b, Jay Wrobel a, Jeffrey C. Pelletier a
*
a Departments of Chemical & Screening Sciences, Wyeth Research, 500 Arcola Rd., Collegeville, PA 19426, USA
b Musculoskeletal Biology, Wyeth Research, 500 Arcola Rd., Collegeville, PA 19426, USA
c Drug Safety and Metabolism, Wyeth Research, 500 Arcola Rd., Collegeville, PA 19426, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A previous report described the serum LH suppression pharmacology of the 2-phenyl-4-piperazinyl-
benzimidazole N-ethyluracil GnRH receptor antagonist 1 following oral administration in rats. A series
of small heterocycles were appended to the 2-(4-tert-butylphenyl)-4-piperazinyl-benzimidazole tem-
plate in place of the N-ethyluracil. Two imidazole analogues, 32 and 41, were shown to possess substan-
tial in vitro potency at the target receptor (hGnRH IC50 = 7 and 18 nM, respectively) and aqueous
Received 12 December 2008
Revised 9 February 2009
Accepted 10 February 2009
Available online 13 February 2009
solubility (55 and 100 lg/mL at pH 7.4, respectively). Both compounds had high oral bioavailability in
Keywords:
rats and 32 was further examined in an orchidectomized rat model for serum LH suppression based on
increased volume of distribution over 41. Serum LH levels trended lower in orchidectomized rats follow-
ing oral administration of 32.
Gonadotropin releasing hormone (GnRH)
receptor
2-Phenyl-4-piperazinyl-benzimidazole
Ó 2009 Elsevier Ltd. All rights reserved.
Antagonists of the gonadotropin releasing hormone (GnRH)
Receptor have shown positive clinical results in numerous repro-
ductive tissue disorders, such as endometriosis and prostate
cancer, due to their ability to inhibit the release of the gonadotro-
pins leuteinizing hormone (LH) and follicle stimulating hormone
(FSH).1 Abarelix and Cetrorelix are peptide antagonists of GnRH
currently sold in injectable form due to their poor pharmacokinetic
properties and low bioavailability when taken orally.2 Abarelix is
also administered in extended release formulations which lower
sex hormones to castration levels in patients, triggering numerous
mechanistic side effects. An orally active small molecule GnRH
antagonist could offer benefits such as improved compliance, the
ability to titrate drug and hormone levels, as well as the flexibility
to withdraw the drug relatively quickly when adverse symptoms
are seen. Numerous small molecule GnRH antagonists have been
reported in the literature.3 Many of these examples have poor bio-
availability and exhibit species selectivity precluding the use of
simple in vivo models such as the rat.
rats.4 However, compound 1 has poor solubility, CYP 3A4 inhibi-
tion and poor liver microsome stability (Fig. 1) which were ex-
pected to be significant liabilities in drug development. As a
result, we set out to mitigate the molecular property issues via
structural modification. As shown earlier,4 the heterocycle linked
to the piperazine via a methylene can withstand significant struc-
tural variability and retain biological activity. We chose empirical
modification of this portion of the molecule with the intention of
retaining human and rat GnRH activity, then improving pharma-
ceutical properties on a suitable lead molecule. In addition, small
heterocycles were particularly attractive since molecular weight
would not deviate significantly from the lead and may even be re-
duced, which is an attractive feature for enhancing the pharmaceu-
tical profile.5
In an attempt to quickly determine GnRH receptor affinity for
new analogues, a diverse array of commercially available heterocy-
clic aldehydes was reacted in parallel with the 2-phenyl-4-piperaz-
inyl-benzimidazole template
2
under reductive amination
We recently reported a potent small molecule antagonist of the
GnRH receptor (1, Fig. 1) with a 2-phenyl-4-(1-piperazinyl)benz-
imidazole template that displayed excellent oral bioavailability
and elicited plasma suppression of LH after oral administration in
conditions (Scheme 1). As shown in Table 1, simple thiophenes,
furans, pyrroles and a 2-thiazole analogue showed only weak affin-
ity (Table 1, entries 3–10) for the receptor at 50 nM. The most po-
tent compounds were found to be the 4-imidazoles (11–13) with
substantial binding affinity for the human GnRH receptor. The sole
example of a weakly active imidazole analogue in our initial set
was the 2-imidazole 14. Affinity for the rat GnRH receptor, which
* Corresponding author. Tel.: +1 484 865 2139.
E-mail address: hauzed@wyeth.com (D.B. Hauze).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.02.043

D. B. Hauze et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1986–1990
1987
hGnRH Binding IC50 (nM)
1.7
18
rGnRH Binding IC50 (nM)
O
CYP 3A4 inhib. @ 3 µM (%)
74 (<20)
N
NH
N
HN
N
Solubility @ pH 7.4 (µg/ mL)
11 (>100)
12 (>80)
N
O
rat liver microsome stability
(% remaining after 15 mins @ 1.0 µM)
1
Figure 1. Structure, binding data and in vitro pharmaceutical profiling data (with ideal values in parentheses) for an orally active GnRH antagonist lead.
Ar-CHO, NaBH(OAc)3
NH
N
N
Ar
N
NMP or DCE
45 - 90 %
HN
HN
N
N
2
3 - 45
Scheme 1. Preparation of GnRH antagonists.
was necessary for the evaluation of compounds in early in vivo
models, was also substantial in the 4-imidazoles (IC50 = 28–
290 nM for 11–13, respectively). From this data it was clear that
a nitrogen in the heterocyclic ring was required but not necessarily
sufficient for affinity to both rat and human receptors.
functional assays as well as in vitro pharmaceutical profiling (Table
4). Two compounds were chosen for PK evaluation based on their
overall profile. Compounds 32 and 41 showed reasonable potency
in functional assays as well as good solubility in buffer (pH 7.4).
Their increased solubility over the lead, 1, is most likely due to de-
creased crystallinity since the uracil is more polar than the imida-
zoles. In addition, both compounds were stable in rat liver
microsomes (t1/2 > 15 min.). Although both compounds appeared
to inhibit CYP 3A4 at a level comparable to 1, they did not inhibit
liver enzymes CYP 2D6 and CYP 2C9.
Both compounds 41 and 32 were evaluated in one-day pharma-
cokinetic assays in orchidectomized rats. Although oral bioavail-
ability was high in both instances, 41 had low volume of
distribution. In contrast, 32 had moderate volume of distribution
making it likely that it would quickly diffuse to the target tissue
(Table 5). As a result, compound 32 was examined in vivo for the
ability to lower serum LH levels.
Antagonism of GnRH in vivo leads to a drop in serum LH levels
which in turn leads to lower levels of sex hormones. This decrease
in serum LH has shown clinical significance in the treatment of sex
hormone sensitive conditions. Intact animals have pulsatile pat-
terns of serum LH, making accurate measurements difficult. Orchi-
dectomized rats, however, display elevated, stable levels of serum
LH due to the removal of the testosterone mediated feedback loop
that controls the central release of GnRH. Two cohorts of orchidec-
tomized rats were treated with compound 32 and vehicle, respec-
tively, via oral gavage. Serum levels of LH were then measured
periodically over 24 h. The results of this experiment are shown
in Table 6. Hormone levels trended lower in the drug treated
group. LH levels at 4 and 6 h, however, were considerably higher
than expected for an agent with significant in vitro GnRH antago-
nist properties and failed to drop LH levels as significantly as com-
pound 1.4 We attribute this to residual LH level variability in
several animals of the drug treated group which may be due to
ineffective castration in some animals.
Further investigation of the SAR surrounding the pendant het-
erocycle included substituted azoles. A series of alkylated 5-substi-
tuted oxazoles showed an increase in GnRH affinity with the
introduction of a 2-alkyl group (15 vs 16, 17). Alkylated 4-substi-
tuted oxazoles (18) were equipotent with the 5-substituted ver-
sion (17). Thiazole compounds substituted in the 5 position were
in the same range of activity as the oxazole analogs (19 vs. 16).
Increasing the 4-alkyl chain length from methyl to ethyl (19, 21)
caused a loss in activity, while the 2 position tolerated the change
to ethyl (19, 20). A 2-n-butyl group in this position (22) also caused
a loss of activity. In general, oxazole and thiazole affinity for the rat
receptor was 10–30 times weaker than it was for the human pro-
tein (Table 2). This level of potency was too modest for in vivo con-
sideration so attention was turned to pyrazoles and, as suggested
by the data in Table 1, imidazoles as well.
Pyrazoles were more sensitive to ring substitution than the
oxazoles and thiazoles. N1-Methyl analogs substituted at the 5-po-
sition (Table 3, 25 and 26) gave the most potent compounds in this
series. As in the case of the oxazoles and thiazoles, however, rat
GnRH affinity was considerably weaker than affinity for the human
receptor. The substituted imidazoles, on the other hand, provided
the most potent analogs overall. Alkyl disubstitution at the 1,5-
and 1,2-imidazole positions provided potent human GnRH binding
agents (IC50’s = 5–46 nM). Affinity for the rat receptor for most of
these compounds was only 3–5 times weaker (IC50’s = 24–78 nM)
than they were for the human receptor. Alkylation of the nitrogen
adjacent to the point of attachment (36, 45) provided the weakest
analogs in the series.
The imidazole analogs that had both human and rat GnRH
receptor affinity of 50 nM or less were evaluated in secondary

1988
D. B. Hauze et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1986–1990
Table 1
Table 2
GnRH receptor affinity for compounds prepared from commercially available
aldehydes
GnRH receptor affinity for oxazole and thiazole analogs
N
Ar
N
HN
N
Ar
N
N
HN
N
Compd
number
Ar
hGnRH IC50 (nM) or %
inhibition at 50 nMa
rGnRH IC50
(nM)a
Cmpd number
Ar=
hGnRH IC50 (nM) or
rGnRH IC50
(nM)a
% inhibition at 50 nMa
N
15
16
17
18
19
20
21
22
90
850
740
320
550
—
O
3
24%
6%
—
—
—
—
—
—
—
—
S
N
31
4
O
S
N
37
5
12%
705
20%
7%
O
O
O
27
6
N
O
N
27
7
S
O
N
8
20
120
—
N
H
S
N
9
6%
N
26%
4%
21
S
N
10
26%
N
S
N
—
S
S
11
12
21
33
290
98
23
—
N
H
N
a
Binding to overexpressed human or rat GnRH receptors in competition with
-Trp6)-GnRH (Ref. 6). Results are given as an average of two independent
experiments run in triplicate. The standard deviations for these assays were typi-
cally within 50% of the IC50
N
125I-(
D
N
.
N
13
7.5
28
N
H
In conclusion, we used empirical techniques to diversify the ap-
pended uracil of lead compound 1. Systematic replacement of the
uracil with smaller heterocycles led to the eventual discovery of
imidazoles 32 and 41 with nanomolar affinity for the human and
rat receptors. Both compounds had improved liver microsome sta-
bility and solubility over the uracil. Inhibition of the CYP 3A4 iso-
zyme, however, remained unchanged from the parent molecule.8
Oral and iv pharmacokinetics on compound 32 supported an
in vivo efficacy trial. Although this compound showed excellent
N
14
40%
—
N
H
a
Binding to overexpressed human or rat GnRH receptors in competition with
-Trp6)-GnRH (Ref. 6). Results are given as an average of two independent
experiments run in triplicate. The standard deviations for these assays were typi-
cally within 50% of the IC50
125I-(
D
.

D. B. Hauze et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1986–1990
1989
Table 3 (continued)
Table 3
GnRH receptor affinity for pyrazole and imidazole analogs
Compd
Ar
hGnRH IC50 (nM) or
rGnRH IC50
(nM)a
number
% inhibition at 50 nMa
N
N
35
46
49
N
Ar
N
HN
N
N
36
37
67%
10
—
Compd
number
Ar
hGnRH IC50 (nM) or
rGnRH IC50
(nM)a
% inhibition at 50 nMa
N
N
N
N
N
N
24
25
25%
23
—
34
N
N
N
N
N
310
38
39
18
26
78
58
N
N
26
27
28
22
68
38
250
—
N
N
40
41
10
18
51
41
N
H
N
450
N
N
N
N
29
30
31
83
43
—
N
N
N
42
43
11
69
49
350
—
N
N
N
N
N
N
32%
8
N
N
32
33
7
24
47
N
N
N
44
29
57
57
N
N
12
N
45
120
N
N
a
Binding to overexpressed human or rat GnRH receptors in competition with
-Trp6)-GnRH (Ref. 6). Results are given as an average of two independent
experiments run in triplicate. The standard deviations for these assays were typi-
cally 50% of the average or less.
34
19
61
125I-(
D
N

1990
D. B. Hauze et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1986–1990
Table 4
In vitro profile and pharmaceutical properties of lead imidazoles
Cmpd number
1
13
7.5
28
44
250
46
>30
85
14
32
35
46
49
71
570
3
>30
84
37
10
34
23
300
51
9
41
18
41
43
240
55
25
43
hGnRH IC50 (nM)a
1.7
18
4
59
13
5
74
0
7
24
14
200
>100
21
73
14
27
8
rGnRH IC50 (nM)a
49
13
296
25
12
—
Human inositol phosphate IC50 (nM)b
Rat LH release IC50 (nM)c
Solubility at pH 7.4 (
l
g/mL)d
Rat liver microsome t1/2 (min)d
CYP 3A4 (% inhibition at 3
CYP 2D6 (% inhibition at 3
CYP 2C9 (% inhibition at 3
l
M)d
66
3
12
85
13
19
l
M)d
M)d
21
53
—
—
l
34
23
a
See footnote a in Tables 1–3.
b
c
Compound driven IP reduction in whole cells following stimulation with (
D
-Trp6)-GnRH (Ref. 7).
Compound driven reduction in LH release from primary rat pituitary cells stimulated with (D
-Trp6)-GnRH (Ref. 4).
d
Details for pharmaceutical profiling assays can be found in Ref. 9.
Table 5
Single dose pharmacokinetics of compounds 32 and 41 in orchidectomized SD rats
Cmpd number
iv
po
) ng h/mL
Bioavailability (% F)
Clearance (mL/min/kg)
Vss (L/kg)
t1/2 (h)
AUC (0–
a
t1/2 (h)
41a
32b
10
134
0.2
2.7
0.9
0.3
3830
3444
2.9
3.7
100
93
a
iv dose of 1 mg/kg in 20% DMSO/PEG 200; oral dose of 20 mg/kg in 0.4 N HCl/2% TWEEN 80.
iv dose of 1 mg/kg in 20% DMSO/PEG 200; oral dose of 30 mg/kg in PEG 400.
b
Table 6
Acknowledgements
Serum LH levels of rats treated with compound 32 (30 mg/kg po, PEG 400)
We thank Magid Abou-Gharbia, Ron Magolda, Richard Lyttle,
Greg Kopf and Len Freedman for their support. We also thank
our colleagues in Analytical Chemistry and Pharmaceutical Profil-
ing for compound characterization.
Time (h)
Compound 32a
Average LHb
Vehiclea
Average LHb
Change
SD
SD
0.5
1
2
4
6
70.3
50.0
24.0
15.9
16.9
50.5
42.5
34.8
70.5
68.0
99.3
86.3
125.8
93.5
28.0
24.1
48.8
18.9
52.2
24.2
À0.2
À18
46.5
100.8
94.5
90.0
À52.8
+14.5
À32.3
À3.5
References and notes
24
1. Moreau, J.-P.; Delavault, P.; Blumberg, J. Clin. Ther. 2006, 28, 1485.
2. Herbst, K. Curr. Opin. Pharmacol. 2003, 3, 660–666.
a
n = 6 animals per group.
Relative LH levels at t = 0 adjusted to 100.
3. Betz, S. F.; Zhu, Y.-F.; Chen, C.; Struthers, R. S. J. Med. Chem. 2008, 51, 3331.
4. Pelletier, J. C.; Chengalvala, M.; Cottom, J.; Feingold, I.; Garrick, L.; Green, D.;
Hauze, D.; Huselton, C.; Jetter, J.; Kao, W.; Kopf, G. S.; Lundquist, J. T., IV; Mann, C.;
Mehlmann, J.; Rogers, J.; Shanno, L.; Wrobel, J. Bioorg. Med. Chem. 2008, 16, 6617.
5. Gleeson, M. P. J. Med. Chem. 2008, 51, 817.
b
6. Millar, R. P.; Davidson, J.; Flanagan, C.; Wakefield, I. Methods Neurosci. 1995, 25,
145.
7. Chengalvala, M.; Kostek, B.; Frail, D. E. J. Biochem. Biophys. Methods 1999, 38, 163.
8. Hu, Q.; Negri, M.; Jahn-Hoffmann, K.; Zhuang, Y.; Olgen, S.; Bartels, M.; Mueller-
Vieira, U.; Lauterbach, T.; Hartmann, R. W. Bioorg. Med. Chem. 2008, 16, 7715.
9. Di, L.; Kerns, E. H. Curr. Opin. Drug Disc. Dev. 2005, 8, 495.
bioavailability and a trend toward pharmacological activity in a rat
in vivo assay, it was not as efficacious as 1 in lowering serum LH
levels. We attribute this to variable GnRH levels in several animals
that may have resulted from inadequate orchidectomy.