Potent and orally bioavailable zwitterion GnRH antagonists with low CYP3A4 inhibitory activity

Article abstract of DOI:10.1016/j.bmcl.2008.04.036

Incorporation of a carboxylic acid into a series of uracil derivatives as hGnRH-R antagonists resulted in a significant reduction of CYP3A4 inhibitory activity. Highly potent hGnRH antagonists with low CYP3A4 inhibitory liability, such as 8a and 8d, were identified. Thus, 8a had a K_i of 2.2 nM at GnRH-R and an IC₅₀ of 36 μM at CYP3A4.

Full text of DOI:10.1016/j.bmcl.2008.04.036

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 3301–3305
Potent and orally bioavailable zwitterion GnRH antagonists
with low CYP3A4 inhibitory activity
Chen Chen,a,* Yongsheng Chen,a Joseph Pontillo,a Zhiqiang Guo,a Charles Q. Huang,a
Dongpei Wu,a Ajay Madan,c Takung Chen,c Jenny Wen,c Qiu Xie,b Fabio C. Tucci,a
Martin Rowbottom,a Yun-Fei Zhu,a Warren Wade,a John Saunders,a
Haig Bozigianc and R. Scott Struthersb
aDepartment of Medicinal Chemistry, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
bDepartment of Endocrinology, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
cDepartment of Preclinical Development, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
Received 18 March 2008; revised 10 April 2008; accepted 11 April 2008
Available online 26 April 2008
Abstract—Incorporation of a carboxylic acid into a series of uracil derivatives as hGnRH-R antagonists resulted in a significant
reduction of CYP3A4 inhibitory activity. Highly potent hGnRH antagonists with low CYP3A4 inhibitory liability, such as 8a
and 8d, were identified. Thus, 8a had a Ki of 2.2 nM at GnRH-R and an IC50 of 36 lM at CYP3A4.
Ó 2008 Elsevier Ltd. All rights reserved.
The relationship between cytochrome P450 (CYP) inhi-
bition and drug–drug interactions is well understood,
therefore, modern drug discovery has incorporated
CYP inhibition detection at an early stage.1 One of the
major liver enzymes responsible for the metabolism of
many drugs is CYP3A4.2 Previously, we identified a ser-
ies of uracils as potent antagonists of the human gona-
dotropin-releasing hormone receptor (hGnRH-R), and
found that a trifluoromethyl substitution of the 1-benzyl
group of the uracil results in a substantial increase in po-
tency.3 Thus, compound 1a (Fig. 1) possesses a Ki value
of 0.64 nM, which is about ninefold better than its flu-
oro analog (1b, Ki = 5.3 nM).3 However, 1a potently
inhibited the CYP3A4 enzyme with an IC50 of 0.1 lM,
showing potential drug–drug interaction liability. We
then embarked on a study to address this issue by incor-
porating a polar group into the molecule. We reasoned
that acid groups in the ligand may be tolerated because
of the presence of several basic residues (e.g., Lys-121,
Arg-299, Lys-115) in the ligand binding pocket.4 In this
paper, we describe a series of butyrate derivatives of 1,
exemplified by 8a and 8d, which possesses high GnRH
antagonist potency and good oral bioavailability, but
low CYP3A4 inhibitory activity.
The synthesis of compounds 4–6 started from the 5-bro-
mouracil 23 (Scheme 1). Suzuki coupling reactions of 2
with 4-substituted phenylboronic acids and palladium
(0) afforded the desired products 3a–c. Coupling reac-
tions of the benzyl alcohol 3a with carboxylic acids or
acid chlorides gave, after Boc-deprotection with trifluo-
roacetic acid, esters 4a–c in good yields. Ether 6a was
prepared from 3a by alkylation and Boc-removal, while
thioether 6b was made by a conversion of 3a to the iso-
propylsulfonate, followed by displacement with mercap-
toacetic acid, and Boc-deprotection. Alkylation and
deprotection of benzylamine 3b gave ester 5, and the
corresponding carboxylic acid 6c. Finally, 6d was
formed by reductive amination and deprotection of the
benzaldehyde 3c.
Compounds 8 were synthesized according to Scheme 2.
Alkylations of 1a–f3 with methyl 4-bromobutyrate pro-
vided the secondary amines 7, which were then hydro-
lyzed under basic conditions to give the desired
carboxylic acids 8a–f.
Keywords: Zwitterion; Gonadotropin-releasing hormone; Receptor;
Antagonist; CYP3A4; Inhibition; Structure–activity relationship.
The synthesized compounds were screened in a GnRH
receptor binding assay as previously described.5 These
*
Corresponding author. Tel.: +1 858 617 7600; fax: +1 858 617
7967; e-mail: cchen@neurocrine.com
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.04.036

3302
C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3301–3305
X
R = 2-F,3-MeO:
1a: Y = CF3
1b: Y = F
1c: Y = SO2Me
R = 2-Cl:
1d: Y = CF3
1e: Y = F
1f: Y = SO2Me
O
N
O
N
Y
N
N
R
H2N
H2N
O
F
O
F
1g: X = H
1h: X = CH2OH
1i: X = CH2NH2
F3C
Figure 1. 2R-Aminophenethyl uracil GnRH antagonists.
O
O
O
N
O
O
N
O
R'
N
H2N
O
H2N
OH
N
F
O
F
4a-c
F3C
6a
F3C
3a
3a
b.c
d,c
X
O
O
N
S
O
N
O
Br
F
N
N
e,c
a
HN
Boc
H2N
HN
Boc
OH
3a
O
N
F
O
N
F
O
3a: X = CH2OH
3b: X = CH2NH2
3c: X = CHO
6b
F3C
F3C
F3C
2
f,c or g,c
3c h,c
3b
O
NH
O
N
NH
O
N
H2N
H2N
O
O
N
F
R"
O
OH
O
N
F
5: R" = Et
6c: R" = H
F3C
6d F3C
Scheme 1. Reagents: (a) 4-XC6H4B(OH)2/Pd(PPh3)4/aq Na2CO3; (b) R0COCl/Et3N/DCM or R0COOH/CDI/DMF; (c) TFA/DCM; (d) KOtBu/
BrCH2COOtBu/THF; (e) i— iPrSO2Cl/Et3N/DCM; ii— HSCH2COOH; (f) BrCH2COOEt/ACN; (g) BrCH2COOtBu/ACN; (h) NH2(CH2)2COOtBu/
NaBH3CN/DCM.
O
N
Y
O
N
Y
O
N
Y
b
N
a
N
N
R
R
R
HN
HN
H2N
O
F
O
F
O
F
O
O
OH
OMe
1
7
8
Scheme 2. Reagents: (a) BrCH2CH2CH2COOMe/Na2CO3/THF–H2O; (b) NaOH/EtOH–H2O.
were also screened for their ability to inhibit ligand bind-
ing to the CYP3A4 enzyme in an in vitro assay.6 These
data are summarized in Tables 1 and 2. Mean, standard
error, and number of replicates associated with these
data for key compounds are given in the text.
aminoacetate 4c (Ki = 77 nM) was substantially less po-
tent compared to 1g. These results suggest that this re-
gion can tolerate a large and non-polar group. Like 1g
(IC50 = 0.3 lM), these compounds showed high inhibi-
tory activity toward the CYP3A4 enzyme (IC50 0.048–
0.80 lM). The highly hydrophilic benzylamine 1i (K i =
1100 nM) possessed poor binding affinity at hGnRH-R,
but also low inhibition at CYP3A4 (IC50 = 9.1 lM).
Compared to 1i, the more lipophilic glycine ester 5 (Ki =
110 nM) had 10-fold improvement, while zwitterion 6c
(K i = 3900 nM) was fourfold less potent at hGnRH-R
The lipophilic cyclopropanecarboxylate 4a (K i =
3.3 nM) and hydroxyacetate 4b (Ki = 10 nM) exhibited
potencies similar to the unsubstituted 5-phenyluracil 1g
(Ki = 7.8 nM), but higher than the hydro-xymethyl ana-
log 1h (K i = 69 nM) at hGnRH-R, The weakly basic

C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3301–3305
3303
Table 1. SAR of the substituted 5-phenyl group of uracils 1 and 4–6 at
hGnRH-R and CYP3A4
while the carboxylic acid derivative 6a was over 100-fold
less potent than alcohol 1h at hGnRH-R.
O
N
Z
Attaching a butyric acid to the basic nitrogen of the po-
N
tent hGnRH-R antagonist 1a (Ki = 0.64 nM, pKi
9.2 0.1, N = 3) resulted in zwitterion 8a (Ki
=
=
H2N
O
F
2.2 nM, pK i = 8.7 0.1, N = 8) with comparable po-
tency. More importantly, 8a had an IC50 value of
36 lM at CYP3A4 (pIC50 = 4.4 0.3, N = 3), represent-
ing a 360-fold improvement compared to 1a. In direct
comparison with its methyl ester 7a, which had a Ki of
32 nM at the GnRH receptor and an IC50 = 0.51 lM
at CYP3A4, the acid functionality of 8a significantly
changed the profile of this class of compounds, which
could be associated with the unique physicochemical
property of an ampholyte.7 Compound 8a had pKa val-
ues of 4.16 for the acid and 7.65 for the amine. Its
molecular structure might allow the carboxylic group
to interact with the basic nitrogen via folded conform-
ers, therefore, it possesses relatively high lipophilicity
at physiological pH (logD = 1.8).
4-6
F3C
Compound
Z
GnRH
Ki (nM)
CYP3A4
IC50 (lM)
1g
1h
1i
7.8
69
0.30
0.80
9.1
OH
NH2
1100
3.3
4a
4b
4c
5
OCOPr-c
OCOCH2OH
0.048
0.12
0.67
0.35
20
10
77
OCOCH2NH2
NHCH2COOEt
OCH2COOH
110
6a
6b
6c
6d
>10,000
3700
3900
>10,000
SCH2COOH
NHCH2COOH
NHCH2CH2COOH
16
36
31
Similar results were also observed for other analogs.
Thus, the butyric acid derivative of 1d (hGnRH-R
Ki = 1.1 nM, pK i = 9 0.1, N = 3; CYP3A4 IC50
=
0.16 lM, pIC50 = 6.8 0.1, N = 3) possessed similar
GnRH potency (8d, Ki = 3.4 nM, pK i = 8.5 0.1,
N = 9) but much improved CYP3A4 inhibitory activity
(IC50 = 17 lM, pIC50 = 4.8 0.2, N = 5), while its
methyl ester 7d exhibited low affinity at hGnRH-R
and high CYP3A4 activity.
and the propionic acid 6d showed no binding. However,
zwitterions 6c and 6d had insignificant inhibition at
CYP3A4 (IC50 > 30 lM), while ester 5 was more potent
CYP3A4 inhibitor (IC50 = 0.35 lM) than its parent 1i.
The other two carboxylic acids (6a–b) also showed a
weak inhibitory activity at CYP3A4 (Table 1). These re-
sults demonstrated that a polar group was intolerable in
this region of compound 1g as an hGnRH-R antagonist,
but an acidic functionality diminished its interaction
with the CYP3A4 enzyme. It is worth noting that zwit-
terion 6c was only fourfold less potent than amine 1i,
Several close analogs were also synthesized and studied.
The butyric acids 8b–c and 8e–f displayed a range of
potencies (Ki 3.4–42 nM) at hGnRH-R, however, these
all showed
a
low CYP3A4 inhibitory activity
Table 2. SAR of uracil derivatives with or without a butyric acid
O
N
Y
N
R
HN
R'
O
F
1, 7 and 8
Compound
Y
R
R0
H
GnRH Ki (nM)
0.64
32
CYP3A4 IC50 (lM)
1a
7a
8a
1b
8b
1c
8c
1d
7d
8d
1e
8e
1f
CF3
2-F,3-MeO
0.10
0.51
36
(CH2)3COOMe
(CH2)3COOH
H
2.2
5.3
F
2-F,3-MeO
2-F,3-MeO
2-Cl
0.60
33
(CH2)3COOH
H
22
MeSO2
CF3
0.9
3.4
1.1
0.38
>50
0.16
0.13
17
(CH2)3COOH
H
(CH2)3COOMe
(CH2)3COOH
H
66
3.4
6.0
F
2-Cl
2-Cl
0.44
20
(CH2)3COOH
H
(CH2)3COOH
42
MeSO2
1.3
3.6
0.18
48
8f

3304
C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3301–3305
(IC50 P 20 lM, Table 2). These results indicate that the
incorporation of a butyric acid into amines 1 reduces
the binding affinity of these compounds by two- to four-
fold at hGnRH-R, and CYP3A4 inhibition by >50-fold.
1.9 lg/mL h, resulting in an oral bioavailability of
22.6% in monkeys.
Both 8a and 8d are highly selective for hGnRH-R, com-
pared to rat and macaque GnRH receptors, as has been
described for several other classes of non-peptide GnRH
antagonists. The affinity for the rat receptor was 11.2
and 12.5 lM, respectively. Thus, pharmacodynamic re-
sponses in the rat were not evaluated. Affinity for the ma-
caque receptor was better (81 and 100 nM, respectively)
than for the rat, but greatly reduced compared to the hu-
man receptor (ꢀ60-fold). This species selectivity is great-
er than that reported for another uracil GnRH-R
antagonist, NBI-42902 (ꢀ12- to 17-fold depending on
the assay).10 Accordingly, when compared in the castrate
macaque, despite improved plasma exposure of 8b, the
degree of LH suppression observed was substantially less
than that observed for an equivalent dose of NBI-
42902.11 These data will be discussed in detail elsewhere.
The functional hGnRH-R antagonism of select potent
compounds was demonstrated in a calcium flux assay.8
Thus, 8a and 8d possessed IC50 values of 4.9 and
11 nM, respectively. In contrast, 6a with a low binding
affinity was much less active in this assay
(IC50 = 7.9 lM).
In addition to their low CYP3A4 inhibitory activity,
compounds 8 also exhibited good metabolic stability
based on in vitro incubation with human liver micro-
somes.9 The predicted scaled systemic clearance was
10, 9.5, and 7.1 mL/min kg, respectively, for 8a, 8d,
and 8f.
Due to their desirable physicochemical properties (mea-
sured logD at pH of 7.4 was 1.8 and 2.1, respectively,
for 8a and 8d) and in vitro pharmacological and meta-
bolic profiles, 8a and 8d were further studied in Spra-
gue–Dawley rats and Cynomolgus monkeys for their
pharmacokinetic profiles (Table 3). After a 10 mg/kg
intravenous injection to rats, 8d displayed a moderate
plasma clearance (CL = 37.6 mL/min kg) and a low vol-
ume of distribution (Vd = 1.3 L/kg), resulting in a short
half-life (t1/2) of 0.4 h. The oral exposure of this com-
pound in rats was low, and the area under curve
(AUC) was only 132 ng/mL h after a 10 mg/kg oral
administration, which gave an oral bioavailability of
3%. In monkeys, 8d had a low CL of 7.9 mL/min h
and a moderate Vd of 4.9 L/kg, which resulted in a long
t1/2 of 7.5 h. After the oral administration of 10 mg/kg
8d to monkeys, a maximal concentration (C max) of
7.9 lg/mL was attained at 0.63 h (T max) and the AUC
was 11.1 lg/mL h. The monkey oral bioavailability
was 52.7%.
The precise nature of the receptor–ligand interactions
which enable the mutual satisfaction of both CYP3A4
and GnRH-R affinity design objectives is less clear.
Based on our previous mutagenesis and molecular mod-
eling of uracil based GnRH-R antagonists,12 several ba-
sic residues in the receptor are potential counter-ions for
the acid moiety in the non-peptide ligands discussed
here. In addition, polar residues from the extracellular
domains which have not been modeled may also be in
proximity to the ligand based on the involvement of this
region in trapping insurmountable GnRH-R antago-
nists.4 However, initial mutagenesis experiments failed
to identify a potential basic residue acting as a coun-
ter-ion to the acids in compounds such as 8a and 8d.
Further, because basic groups attached by alkyl linkers
analogous to the butyric acid group here are also consis-
tent with high hGnRH-R binding affinity (unpublished),
we hypothesize that these charged groups may not be
engaged in specific receptor interactions, but rather sim-
ply remain solvent accessible.
Similarly, 8a exhibited a high CL of 64.1 mL/min kg and
poor oral exposure in rats (Table 3). In monkeys, 8a had
a moderate CL of 20.9 mL/min kg and a low Vd of
1.8 L/kg. Its t1/2 of 1 h was much shorter than that of
8d. The higher plasma clearance of 8a than that of 8d
could be associated with O-demethylation of 8a, which
was identified as a major metabolite from in vitro stud-
ies. An oral dose of 8a at 10 mg/kg gave an AUC of
In conclusion, a series of uracils were synthesized as
hGnRH-R antagonists. Attempts to incorporate an
acidic moiety at the 5-phenyl ring of uracil 1g failed to
generate potent compounds at hGnRH-R, but provided
valuable information for lowering CYP3A4 inhibition.
Incorporating a butyric acid into the amino group of 1
resulted in several potent hGnRH-R antagonists, such
as 8a and 8d, with much improved CYP3A4 profiles.
8d was also found to have a long half-life and good oral
bioavailability in monkeys. Thus, we have demonstrated
that incorporating a carboxylic acid at an appropriate
position of potent hGnRH-R antagonists greatly re-
duces their CYP3A4 inhibitory activity while maintains
their potency at the target receptor.
Table 3. Pharmacokinetic parameters of compounds 8a and 8d after
an intravenous or oral dose to rats and monkeys (10 mg/kg, N = 3 for
each point)
Compound
8a
Monkey
8d
Monkey
Rat
Rat
CL (mL/min kg)
Vd (L/kg)
t1/2 (h)
64.1
1.7
0.3
77
20.9
1.8
37.6
1.3
7.9
4.9
References and notes
1
0.4
7.5
Cmax (ng/mL)
Tmax (h)
2039
0.5
149
0.25
131.8
3
7874
0.63
11,097
52.7
1. Zlokarnik, G.; Grootenhuis, P. D.; Watson, J. B. Drug
Discovery Today 2005, 10, 1443.
2. Riley, R. J.; parker, A. J.; Trigg, S.; Manners, C. N.
Pharm. Res. 2001, 18, 652.
0.25
po AUC (ng/mL h)
F (%)
63.3
2.4
1904
22.6

C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3301–3305
3305
3. Guo, Z.; Chen, Y.; Huang, C. Q.; Gross, T. D.; Pontillo,
J.; Rowbottom, M. W.; Saunders, J.; Struthers, S.; Tucci,
F. C.; Xie, Q.; Wade, W.; Zhu, Y. F.; Wu, D.; Chen, C.
Bioorg. Med. Chem. Lett. 2005, 15, 2519.
4. Kohout, T. A.; Xie, Q.; Reijmers, S.; Finn, K. J.; Guo, Z.;
Zhu, Y. F.; Struthers, R. S. Mol. Pharmacol. 2007, 72, 238.
5. Guo, Z.; Zhu, Y. F.; Gross, T. D.; Tucci, F. C.; Gao, Y.;
Moorjani, M.; Connors, P. J., Jr.; Rowbottom, M. W.;
Chen, Y.; Struthers, R. S.; Xie, Q.; Saunders, J.; Reinhart,
G.; Chen, T. K.; Bonneville, A. L.; Chen, C. J. Med.
Chem. 2004, 47, 1259.
6. Emoto, C.; Iwasaki, K. Xenobiotica 2006, 36, 219.
7. Pagliara, A.; Carrupt, P-A. P.-A.; Caron, G.; Gaillard, P.;
Testa, B. Chem. Rev. 1997, 97, 3385.
8. Calcium flux assay: Zhou, W.; Rodic, V.; Kitanovic, S.;
Flanagan, C. A.; Chi, L.; Weinstein, H.; Maayani, S.;
Millar, R. P.; Sealfon, S. C. J. Biol. Chem. 1995, 270,
18853, Also see Ref. 11.
9. For assay conditions, see Ref. 5.
10. Tucci, F. C.; Zhu, Y. F.; Struthers, R. S.; Guo, Z.;
Gross, T. D.; Rowbottom, M. W.; Acevedo, O.; Gao,
Y.; Saunders, J.; Xie, Q.; Reinhart, G. J.; Liu, X. J.;
Ling, N.; Bonneville, A. K.; Chen, T.; Bozigian, H.;
Chen, C. J. Med. Chem. 2005, 48, 1169.
11. Struthers, R. S.; Xie, Q.; Sullivan, S. K.; Reinhart, G.
J.; Kohout, T. A.; Zhu, Y. F.; Chen, C.; Liu, X. J.;
Ling, N.; Yang, W.; Maki, R. A.; Bonneville, A. K.;
Chen, T. K.; Bozigian, H. P. Endocrinology 2007, 148,
857.
12. Sullivan, S. K.; Hoare, S. R.; Fleck, B. A.; Zhu, Y. F.;
Heise, C. E.; Struthers, R. S.; Crowe, P. D. Biochem.
Pharmacol. 2006, 72, 838.