A potent, nonpeptidyl 1H-quinolone antagonist for the gonadotropin-releasing hormone receptor

Article abstract of DOI:10.1021/jm000275p

Extensive development of the structure - activity relationships of a screening lead determined three important pharmacophores for gonadotropin-releasing hormone (GnRH) receptor antagonist activity. Incorporation of the 3,4,5-trimethylphenyl group at the 3

Full text of DOI:10.1021/jm000275p

J . Med. Chem. 2001, 44, 917-922
917
A P oten t, Non p ep tid yl 1H-Qu in olon e An ta gon ist for th e
Gon a d otr op in -Relea sin g Hor m on e Recep tor
Robert J . DeVita,* Thomas F. Walsh, J onathan R. Young, J inlong J iang, Feroze Ujjainwalla,
Richard B. Toupence, Mamta Parikh, Song X. Huang, J ason A. Fair, Mark T. Goulet, Matthew J . Wyvratt,
J ane-L. Lo,† Ning Ren,† J oel B. Yudkovitz,† Yi T. Yang,† Kang Cheng,† J isong Cui,† George Mount,†
Susan P. Rohrer,† J ames M. Schaeffer,† Linda Rhodes,#,§ J ennifer E. Drisko,# Erin McGowan,#
D. Euan MacIntyre,# Styliani Vincent, J osephine R. Carlin, J udith Cameron,‡ and Roy G. Smith†,|
Departments of Medicinal Chemistry, Biochemistry & Physiology, Pharmacology, and Drug Metabolism, Merck Research
Laboratories, P.O. Box 2000, Rahway, New J ersey 07065-0900, and Oregon Regional Primate Research Center, Oregon Health
Sciences University, 505 NW 185th Avenue, Beaverton, Oregon 97006
Received J une 26, 2000
Extensive development of the structure-activity relationships of a screening lead determined
three important pharmacophores for gonadotropin-releasing hormone (GnRH) receptor an-
tagonist activity. Incorporation of the 3,4,5-trimethylphenyl group at the 3-position, 2-(2(S)-
azetidinyl)ethoxy group at the 4-position, and N-4-pyrimidinylcarboxamide at the 6-position
of the quinolone core resulted in the identification of 4-(2-(azetidin-2(S)-yl)ethoxy)-7-chloro-2-
oxo-3-(3,4,5-trimethylphenyl)-1,2-dihydroquinoline-6-carboxylic acid pyrimidin-4-ylamide (1) as
a potent antagonist of the GnRH receptor. A 104-fold increase in in vitro binding affinity is
observed for the GnRH receptor as compared to the initial screening lead. Compound 1 exhibits
nanomolar binding activity and functional antagonism at the human receptor and is 7-fold
less active at the rhesus receptor. Intravenous administration of compound 1 to rhesus monkeys
results in a significant decrease of the serum levels of downstream hormones, luteinizing
hormone (79% decrease in area under the curve) and testosterone (92% decrease in area under
the curve), at a dose of 3 mg/kg. Quinolone 1 is a potent nonpeptidyl antagonist for the human
GnRH receptor that is efficacious for the suppression of luteinizing hormone and testosterone
in primates.
In tr od u ction
Gonadotropin-releasing hormone (GnRH) is a deca-
peptide released by the hypothalamus which binds to
receptors on the pituitary.1 The activation of this
G-protein coupled receptor2 causes the release of lutein-
izing hormone and follicle-stimulating hormone which
regulate gonadal steroid hormone production. A variety
of disease conditions such as prostate cancer, breast
cancer, and endometriosis may be treated by suppres-
sion of the hypothalamic-pituitary-gonadal hormonal
axis. There is recent clinical evidence that peptidic
GnRH antagonists directly lower sex hormone levels
alleviating disease symptoms with a superior side effect
profile as compared to other therapies.3 Therefore, the
development of small molecule antagonists of the GnRH
receptor may be clinically useful without the usual
liabilities associated with large peptidyl therapeutics.4
In this paper, we report the identification of a nonpep-
tidyl 3-arylquinolone antagonist of the GnRH receptor
including in vivo efficacy in a primate model for
hormone suppression.
F igu r e 1. Merck lead compound.
4-alkoxy-7-chloroquinolone lead structure (Figure 1).5
Extensive development of the structure-activity rela-
tionships (SARs) of our lead identified three important
pharmacophores for GnRH receptor antagonist activity.
Incorporation of the 3,4,5-trimethylphenyl group6 at the
3-position, 2-(2(S)-azetidinyl)ethoxy group7 at the 4-po-
sition, and N-4-pyrimidinylcarboxamide8 at the 6-posi-
tion of the quinolone core resulted in the identification
of compound 1 as a potent antagonist of the human
GnRH receptor.
Earlier publications from this laboratory disclosed the
initial screening efforts and the identification of a
Resu lts a n d Discu ssion
* To whom correspondence should be addressed. Phone: (732)-594-
7039. Fax: (732)-594-3220. E-mail: robert_devita@merck.com.
† Department of Biochemistry & Physiology.
Ch em istr y. The synthetic route to quinolone 1 is
similar to those reported in the earlier publications from
our group (Scheme 1).5-8 Treatment of methyl 4-chlo-
roanthranilate (2) with iodine and silver sulfate pro-
vided a quantitative crude yield of the 5-iodinated
derivative. Acetylation of the amino group with acetyl
#
Department of Pharmacology.
Department of Drug Metabolism.
‡ Oregon Health Sciences University.
§ Current address: Merial Inc., Woodbridge, NJ .
| Current address: Baylor University Medical School, Institute for
Aging, Houston, TX.
10.1021/jm000275p CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/08/2001

918 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 6
Sch em e 1. Synthesis of Compound 1a
DeVita et al.
a
Reagents and conditions: (a) I2, Ag2SO4, EtOH, rt, 1 h, 100% crude yield; (b) AcCl, 1,2-dichloroethane, 80 °C, 3 h, 93%; (c) (PPh3)2PdCl2,
CO (1 atm), Et3N, DMF/MeOH (4:1), 95 °C, 16 h, 74%; (d) H2SO4, MeOH, reflux, 1 h, 80%; (e) 3,4,5-trimethylphenylacetyl chloride,
1,2-dichloroethane, 80 °C, 3 h, 78%; (f) LiN(TMS)2, THF, 0 °C, 2 h, 82%; (g) alcohol 6, PPh3, DEAD, THF, 0 °C to rt, 24 h; (h) LiOH,
THF/H2O/EtOH, 80 °C, 5 h, 86%, 2 steps; (i) 4-aminopyrimidine, EDAC, DMAP, CH2Cl2, rt, 16 h, 84%; (j) CF3CO2H-CH2Cl2 (1:1), rt, 3
h, 87%.
F igu r e 2. In vitro characterization of compound 1.
chloride was necessary to prevent side reactions in the
ensuing carbonylation. Reaction of the aryl iodide with
a Pd-catalyst under carbon monoxide atmosphere in
N,N-dimethylformamide/methanol solution provided the
isophthalate diester. Filtration through a silica gel pad
to remove the palladium byproducts followed by acet-
amide hydrolysis in sulfuric acid/methanol afforded the
crystalline diester 3.
ate free of any impurities in 86% yield for the two steps.
Amide formation under anhydrous conditions with
4-aminopyrimidine gave the desired amide 8 in 84%
yield. Removal of the BOC group with trifluoroacetic
acid provided compound 1 in 87% yield.
In Vitr o Ch a r a cter iza tion . The initial SARs of the
screening lead were developed based on data from a rat
pituitary membrane binding assay.5 Later, a radioligand
binding assay using cloned CHO cells expressing the
human GnRH receptor6-8 supplemented the data pro-
vided by the initial rat pituitary membrane radioligand
binding assay. In addition, an assay to determine
functional antagonism of GnRH-stimulated phospha-
tidyl inositol (PI) hydrolysis in CHO cells expressing
cloned human GnRH receptors was developed (PI
turnover assay). All in vitro data are expressed as the
IC50 for inhibition of binding (vs [125I]buserelin) or
functional antagonism (vs GnRH) at the GnRH receptor
of the particular species for which the assay is being
conducted.
Compound 1 possesses an IC50 ) 0.44 ( 0.39 nM in
the cloned human GnRH receptor membrane binding
assay and excellent functional antagonism at the human
GnRH receptor, IC50 ) 1.0 ( 0.6 nM (Figure 2). A
significant decrease in binding and functional activity
is observed for the rat and dog receptors (10- and 150-
fold shifts in binding, respectively) despite the high
degree of receptor identity between the four species.11
Acylation of intermediate 3 with the acid chloride of
3,4,5-trimethylphenylacetic acid9 gave the crystalline
amide 4 in 78% yield. Quinolone formation under basic
conditions followed by aqueous acid quench, filtration,
and trituration with cold acetonitrile provided the
4-hydroxyquinolone 5 in 82% yield. This key intermedi-
ate was dried in vacuo to remove any residual water
which would interfere with the subsequent coupling
reaction. The protected (S)-azetidineethanol derivative
6 was prepared in several steps from (S)-azetidine-2-
carboxylic acid with the key transformation being the
Arndt-Eistert homologation.10 Coupling of the two
fragments 5 and 6 was accomplished by utilizing our
previously reported Mitsunobu protocol.7 Silica gel
chromatography of the crude reaction mixture provided
the desired 4-alkylated quinolone 7 contaminated with
a minor amount of diethyl N,N′-hydrazinedicarboxylate
byproduct which is removed in the next step. Hydrolysis
of the methyl ester under basic conditions, followed by
acidic workup, afforded the 6-carboxylic acid intermedi-

Nonpeptidyl Quinolone Antagonist for GnRH Receptor
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 6 919
Compound 1 shows equal binding affinity to the rhesus
GnRH receptor (IC50 ) 0.5 nM) and only a 7-fold shift
in functional antagonist activity (7.0 ( 2.2 nM) as
compared to the human receptor. This result dictated
that any proof of concept study in animal models be
performed in primates such as rhesus monkey.
In Vivo Stu d ies w ith Com p ou n d 1. Compound 1
was studied in rhesus macaques to determine the
efficacy for hormone suppression in vivo. Effective
antagonism of the GnRH receptor would be expected to
block production of the downstream hormones lutein-
izing hormone (LH) and testosterone (T). Earlier at-
tempts to explore efficacy of quinolone 1 in a rat LH-
lowering model were only moderately successful due to
the 100-fold decrease in intrinsic potency in vitro
relative to the human receptor along with extremely
high plasma clearance of the compound. A more infor-
mative study was performed in rhesus monkeys, a
species in which compound 1 has only a 7-fold shift in
functional antagonism in vitro.
F igu r e 3. Rhesus LH levels.
Rhesus iv pharmacokinetics were determined in a
separate study to ensure proper drug exposure prior to
utilization of this species as an animal model. Com-
pound 1 gave, after a 0.5 mg/kg iv dose, a plasma drug
level area under the curve (AUC) of 303 ( 11 ng‚h/mL,
a plasma clearance (Clp) of 30 ( 0.7 mL/min/kg, and a
terminal half-life (t1/2) of 9.5 h. Similarly at the higher
3 mg/kg dose, AUC of 1324 ( 134 ng‚h/mL, Clp of 33 (
3.4 mL/min/kg, and t1/2 of 5 ( 0.7 h were observed. These
results confirmed that adequate exposure of compound
1 could be obtained in the rhesus macaques to warrant
further study.
Rhesus macaques were dosed iv with vehicle (ethanol/
PEG400/saline, 10/40/50) or compound 1 (in the same
vehicle) at doses of 0.5 and 3 mg/kg, and serum LH and
T levels were determined by RIA from blood samples
taken at 20-min intervals. In these experiments, each
monkey served as its own control due to the variability
and pulsatile nature of the blood levels of both hor-
mones, with vehicle and each drug dose administration
occurring on separate days. Previously conditioned
rhesus monkeys were chronically catheterized in order
to allow dosing and blood sampling without disturbing
the animals. The animals were fed on a daily schedule
with a coordinated light-dark cycle. Sampling occurred
for 130 min prior to dosing of the vehicle or antagonist
to establish basal levels and to determine that the
sampling system was functioning properly. Dark cycle
for these animals began 30-50 min after the adminis-
tration of the compound, since in male rhesus there is
a diurnal rhythm of LH and T, with highest levels
observed at night. The time course data for one repre-
sentative animal are presented below. (See Supporting
Information for data for all monkeys included in this
study.)
F igu r e 4. Rhesus T levels.
Serum T levels were also determined from the blood
samples removed at each time point (Figure 4). A
significant decrease in serum T was observed at the
lower dose, but rebound to control levels was observed
at 6 h postdose. At the higher dose, suppression of T to
near castrate levels was observed for the duration of
the study.
The time course data for the four monkeys in the
study were combined to calculate hormone AUC for LH
and T levels for the duration of the experiment (t )
-130 to 470 min). The average calculated AUC for LH
in the vehicle arm of the study was determined to be
320 ( 96 ng‚h/mL (Figure 5). In comparison, LH AUC
was reduced by an average of 41% (176 ( 64 ng‚h/mL,
p ) 0.02) at the 0.5 mg/kg dose and 79% (62 ( 32 ng‚
h/mL, p ) 0.0002) at the 3 mg/kg dose, respectively.
Mean T AUC for the vehicle-treated leg of the study was
325 ( 118 ng‚h/mL with an average reduction of 54%
(145 ( 21 ng‚h/mL, p ) 0.006) at the lower dose and a
92% reduction (28 ( 25 ng‚h/mL, p ) 0.0004) in T AUC
at the higher dose of compound 1 (Figure 5).
Figure 3 shows LH levels after vehicle dosing and
demonstrates the pulsatile nature of serum LH levels
with a large increase in LH after the beginning of the
dark cycle (t > 50 min). Intravenous administration of
the 0.5 mg/kg dose of compound 1 resulted in a signifi-
cant decrease in the amplitude of the LH peaks.
Exposure to compound 1 at a higher 3 mg/kg iv dose
results in complete suppression of LH peaks for the
duration of the study.

920 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 6
DeVita et al.
(400 MHz, CDCl3): δ 3.85 (s, 3H), 5.80 (s, 2H), 6.80 (s, 1H),
8.24 (s, 1H). FAB-MS: calcd for C8H7ClINO2 311; found 312
(M + H, 100).
Meth yl 2-N-Acetyla m in o-4-ch lor o-5-iod oben zoa te. To
a solution of 2 g (6.42 mmol) methyl 2-amino-4-chloro-5-
iodobenzoate in 15 mL 1,2-dichloroethane was added 0.5 mL
(7.06 mmol, 1.1 equiv) acetyl chloride. The resulting reaction
mixture was heated at reflux until all solids had dissolved.
The reaction mixture was cooled to room temperature and the
solvent was removed under vacuum. The resulting off-white
solid was triturated in 25 mL of hot methanol. The mixture
was cooled in an ice bath. The solids were filtered, washed
with ice-cold methanol and dried to afford 2.12 g (93%) of the
1
product as a white solid. H NMR (400 MHz, CDCl3): δ 2.23
(s, 3H), 3.93 (s, 3H), 8.45 (s, 1H), 8.49 (s, 1H). FAB-MS: calcd
for C10H9ClINO3 353; found 354 (M + H, 100).
Dim eth yl 6-N-Acetyla m in o-4-ch lor o-1,3-isop h th a la te.
To a solution of 2.12 g (6 mmol) methyl 2-N-acetylamino-4-
chloro-5-iodobenzoate in 50 mL N,N-dimethylformamide and
12 mL methanol were added 1.7 mL (12 mmol, 2 equiv.)
triethylamine and 0.21 g (0.30 mmol, 0.05 equiv) dichlorobis-
(triphenylphosphine)palladium (II). The resulting reaction
F igu r e 5. Effect of compound 1 on hormone AUC in rhesus
macaques.
mixture was degassed and then stirred under
a carbon
monoxide atmosphere (1 atm) at 80 °C for 16 h. The reaction
mixture was cooled to room temperature and poured into 150
mL ice/water mixture. The resulting solids were filtered,
washed with ice-cold water and air-dried. Filtration of the
crude product through a pad of silica gel eluting with CH2Cl2/
EtOAc gradient (100% to 90%) gave 1.74 g (74%) of product
as a white solid. H NMR (400 MHz, CDCl3): δ 2.26 (s, 3H),
3.93 (s, 3H), 3.96 (s, 3H), 8.59 (s, 1H), 8.93 (s, 1H). FAB-MS:
calcd for C12H12ClNO5 285; found 286 (M + H, 100).
Con clu sion
We have reported the design and synthesis of com-
pound 1 as a potent antagonist of the human GnRH
receptor. Significant improvements in antagonist activ-
ity for this lead class were achieved by incorporation of
a 3,4,5-trimethylphenyl group at the 3-position, 2-((S)-
2-azetidinyl)ethoxy group at the 4-position, and N-(4-
pyrimidinyl)carboxamide at the 6-position of the qui-
nolone core contained in the initial lead structure. These
modifications resulted in a 104-fold improvement of the
binding and functional activity in vitro. Compound 1
exhibits nanomolar functional antagonism at the human
GnRH receptor and is 7-fold less active at the rhesus
receptor. Intravenous administration of compound 1 to
rhesus macaque at a dose of 3 mg/kg resulted in a
significant decrease in the blood levels of downstream
hormones: LH (79% decrease in AUC) and T (92%
decrease in AUC). Quinolone 1 is a potent nonpeptidyl
antagonist of the human GnRH receptor that is effica-
cious for the suppression of LH and T in primates.
1
Dim eth yl 6-Am in o-4-ch lor o-1,3-isop h th a la te (3). To a
suspension of 1.27 g (4.4 mmol) dimethyl 6-N-acetylamino-4-
chloro-1,3-isophthalate in 15 mL methanol was added 2 mL
concentrated sulfuric acid. The resulting reaction mixture was
heated at reflux 1 h. The reaction mixture was cooled to room
temperature and the solvent was removed under vacuum. The
residue was dissolved in 150 mL ethyl acetate and washed
with 10% aqueous sodium bicarbonate (3 × 75 mL), water (75
mL) and saturated aqueous sodium chloride (75 mL). The
organic layer was dried over sodium sulfate, filtered and the
filtrate concentrated under vacuum. The resulting off-white
solid was recrystallized in methanol to afford 0.86 g (80%) of
1
product as a white solid. H NMR (400 MHz, CDCl3): δ 3.87
(s, 3H), 3.89 (s, 3H), 6.18 (br. S, 2H), 6.72 (s, 1H), 8.54 (s, 1H).
FAB-MS: calcd for C10H10ClNO4 243; found 244 (M + H, 100).
3,4,5-Tr im eth ylp h en yla cetyl Ch lor id e. To a solution of
9.4 g (52.7 mmol) 3,4,5-trimethylphenylacetic acid7 in 60 mL
ethylene chloride under nitrogen atmosphere at 0 °C was
added via syringe 5.06 mL (58 mmol) oxalyl chloride followed
by 1 drop N,N-dimethylformamide. The resulting mixture was
stirred at room temperature for several hours until gas
evolution ceased. Use test (reaction with methanol) of a small
aliquot showed all starting material was consumed by thin-
layer chromatography (hexanes/ethyl acetate, 1/1). The solvent
of the reaction mixture was removed under vacuum and the
resulting oil was used in the next reaction without further
purification.
Exp er im en ta l Section
Ch em istr y. Gen er a l Meth od s. 1H NMR spectra were
recorded on Varian XL series spectrometers at the indicated
field strengths. Low-resolution mass spectral analyses were
obtained with a LKB 9000 at an ionizing voltage of 70 eV.
High-resolution mass spectral analysis was obtained using a
Finnigan New Star FT/ICR with electrospray ionization.
Reagents, solvents and drying agents were obtained from
commercial sources and used without further purification or
drying with the exception of tetrahydrofuran and methylene
chloride which were distilled from the appropriate drying
agents before use. Normal-phase column chromatography was
carried out utilizing silica gel 60 (E. Merck).
Meth yl 2-Am in o-4-ch lor o-5-iod oben zoa te. To a mixture
of 6.84 g (27 mmol) iodine and 8.4 g (27 mmol) silver sulfate
in 270 mL absolute ethanol was added 5.0 g (27 mmol) methyl
2-amino-4-chlorobenzoate (2). The resulting reaction mixture
was stirred at room temperature for 45 min. The reaction
mixture was filtered through a pad of Celite and the solvent
was removed under vacuum. The residue was dissolved in 400
mL ethyl acetate and washed with saturated aqueous sodium
bicarbonate (3 × 50 mL), water (3 × 50 mL) and once with
brine. The organic layer was dried over magnesium sulfate,
filtered and the filtrate concentrated under vacuum to afford
8.5 g (∼100%) of the product as an off-white solid. 1H NMR
Dim eth yl 6-N-(3,4,5-Tr im eth ylp h en yl)a cetyla m in o-4-
ch lor o-1,3-isop h th a la te (4). To a solution of 12.8 g (52.7
mmol) dimethyl 6-amino-4-chloro-1,3-isophthalate (3) in 80 mL
1,2-dichloroethane was added 10.4 g (52.7 mmol) 3,4,5-tri-
methylphenylacetyl chloride. The resulting reaction mixture
was heated at reflux for 12 h. The reaction mixture was cooled
to room temperature and the solvent was removed under
vacuum. The resulting off-white solid was triturated in hot
methanol. The mixture was then cooled in an ice bath, solids
were filtered, washed with ice-cold methanol and dried to
1
afford 16.6 g (78%) of product as a white solid. H NMR (400
MHz, CD3OD): δ 2.16 (s, 3H), 2.30 (s, 6H), 3.67 (s, 2H), 3.89
(s, 3H), 3.93 (s, 3H), 7.01 (s, 2H), 8.55 (s, 1H), 8.95 (s, 1H).
FAB-MS: calcd for C21H22ClNO5 403; found 404 (M + H, 100).

Nonpeptidyl Quinolone Antagonist for GnRH Receptor
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 6 921
6-Ca r bom eth oxy-7-ch lor o-4-h yd r oxy-3-(3,4,5-tr im eth -
ylp h en yl)-1H-qu in olin -2-on e (5). To a solution of 2.04 g (5
mmol) dimethyl 6-N-(3,4,5-trimethylphenyl)acetylamino-4-
chloro-1,3-isophthalate (4) in 20 mL dry tetrahydrofuran under
nitrogen atmosphere at 0 °C was added dropwise via syringe
12.6 mL (12.6 mmol, 2.5 equiv) of a solution of sodium bis-
(trimethylsilyl)amide (1.0 M in tetrahydrofuran). The resulting
reaction mixture was stirred at 0 °C for 1 h then quenched
with 60 mL 6 N aqueous HCl/ice (1:1). The resulting solids
were stirred vigorously, filtered, washed with ice-cold water
followed by ice-cold acetonitrile. The resulting off-white solid
was dried in a vacuum oven at 50 °C for 16 h to afford 1.6 g
(3×) and the combined organic layers were dried over sodium
sulfate, filtered and the solvent removed under vacuum. The
residue was purified by column chromatography on silica gel
eluting with ethyl acetate/hexanes (70/30 to 80/20 gradient)
to provide 236 mg (84%) of the product as a white solid. 1H
NMR (400 MHz, CDCl3): δ 1.30 (s, 9H), 1.69 (m, 1H), 1.82
(m,1H), 2.09 (s, 3H), 2.17 (m, 1H), 2.25 (s, 6H), 3.69 (m, 3H),
3.78 (m, 1H), 4.17 (m, 1H), 7.02 (s, 2H), 7.38 (s, 1H), 8.22 (s,
1H), 8.25 (d, J ) 4 Hz, 1H), 8.65 (d, J ) 4 Hz, 1H), 8.90 (s,
1H), 9.47 (s, 1H). FAB-MS: calcd for C33H36ClN5O5 617; found
618 (M + H, 100).
4-(2-(Azet id in -2(S)-yl)et h oxy)-7-ch lor o-2-oxo-3-(3,4,5-
tr im eth ylph en yl)-1,2-dih ydr oqu in olin e-6-car boxylic Acid
P yr im id in -4-yla m id e Dih yd r och lor id e Sa lt (1). To a solu-
tion of 698 mg (1.13 mmol) 4-(2-(N-tert-butoxycarbonylazetidin-
2(S)-yl)ethoxy)-7-chloro-2-oxo-3-(3,4,5-trimethylphenyl)-1,2-
dihydroquinoline-6-carboxylic acid pyrimidin-4-ylamide in 10
mL dry methylene chloride under nitrogen atmosphere was
added 3 drops anisole followed by 10 mL trifluoroacetic acid.
The reaction mixture was stirred at ambient temperature for
4 h at which time the volatiles were removed under vacuum.
The resulting material was purified by column chromatogra-
phy on silica gel eluting with methylene chloride/2 N ammonia
in methanol (95/5 to 93/7 gradient) to provide 510 mg (87%)
4-(2-(azetidin-2(S)-yl)ethoxy)-7-chloro-2-oxo-3-(3,4,5-trimeth-
ylphenyl)-1,2-dihydroquinolin-6-carboxylic acid pyrimidin-4-
ylamide. 1H NMR (400 MHz, CD3OD): δ 2.05 (m, 1H), 2.19
(m, 3H), 2.25 (s, 3H), 2.34 (s, 6H), 3.72 (m, 1H), 3.78 (m, 2H),
3.98 (m, 1H), 4.42 (m, 1H), 7.09 (s, 2H), 7.49 (s, 1H), 8.11 (s,
1H), 8.35 (d, J ) 4 Hz, 1H), 8.71 (d, J ) 4 Hz, 1H), 8.90 (s,
1H). FAB-MS: calcd for C28H28ClN5O3 517; found 518 (M +
H, 100). The free base was converted to the hydrochloride salt
by dissolution in methanol and addition of a 2 N HCl solution
in ether followed by removal of the solvent under vacuum to
provide 1 as an off-white amorphous solid. 1H NMR (400 MHz,
CD3OD): δ 2.05 (m, 1H), 2.20 (m, 3H), 2.25 (s, 3H), 2.35 (s,
6H), 3.74 (m, 1H), 3.79 (m, 2H), 3.98 (m, 1H), 4.43 (m, 1H),
7.09 (s, 2H), 7.51 (s, 1H), 8.25 (s, 1H), 8.75 (d, J ) 6.8 Hz,
1H), 8.93 (d, J ) 6.8 Hz, 1H), 9.25 (s, 1H). High-resolution
ESI MS: calcd C28H28ClN5O3 MW ) 518.1953; found 518.1942.
In Vitr o Bin d in g Assa ys.5-8 Crude membranes prepared
from rat pituitary glands or Chinese hamster ovary K1 cells
stably expressing human GnRH receptors were used as the
sources for GnRH receptors. 5-[125I-Tyr]Buserelin (a peptidyl
GnRH agonist obtained from Woods Assays) having specific
activity of 1000 Ci/mmol was used as the radiolabeled ligand.
Competitive binding was measured in a 50 mM Tris-HCl based
buffer (pH 7.5) containing 2 mM MgCl2 and 0.1% bovine serum
albumin. The binding activity is reported as an IC50 value
which is the antagonist concentration required to inhibit the
specific binding of [125I]buserelin to GnRH receptors by 50%.
In Vitr o F u n ction a l Assa y.11 Chinese hamster ovary cells
stably expressing human GnRH receptors functionally coupled
to phospholipase C were used to evaluate the functional GnRH
antagonism of test compounds. Clones were seeded at a
concentration of 60 000 cells/mL/well in inositol-free F12
medium containing 10% dialyzed fetal bovine serum, 1% Pen/
Strep, 2 mM glutamine, 500 µg/mL G418 and 1 µCi [3H]inositol
in 24-well tray. 48 h after seeding, cells were washed with 3
× 1 mL of PBS containing 10 mM LiCl and treated with
various concentrations of test compounds for 2 h at 37 °C
before addition of 0.5 nM GnRH (Sigma Chemical Co.). After
incubation at 37 °C for an additional 60 min, the medium was
removed and the cells were lyzed with 1 mL of 0.1 M formic
acid. The trays were freeze-thawed once and the cell extract
was applied to a Dowex AG1-X8 column. The column was
washed with 2 × 1 mL H2O to remove free [3H]inositol and
[3H]inositol phosphates were eluted with 3 × 1 mL 2 M
ammonium formate in 1 M formic acid. The eluate was counted
in a scintillation counter. The results are reported as an IC50
value which is the antagonist concentration required to inhibit
the GnRH-stimulated PI hydrolysis by 50%.
1
(82%) of the product. H NMR (400 MHz, DMSO-d6): δ 2.14
(s, 3H), 2.24 (s,6H), 3.85 (s, 3H), 6.93 (s, 2H), 7.37 (s, 1H), 8.43
(s, 1H). FAB-MS: calcd for C20H18ClNO4 371; found 372 (M +
H, 100).
4-(2-(N-ter t-Bu toxyca r bon yla zetid in -2(S)-yl)eth oxy)-6-
car bom eth oxy-7-ch lor o-3-(3,4,5-tr im eth ylph en yl)-1H-qu in -
olin -2-on e (7). To a vigorously stirred solution of 1.35 g (6.72
mmol) N-tert-butoxycarbonylazetidine-2(S)-ethanol (6) in 67
mL dry tetrahydrofuran under nitrogen was added 2.50 g (6.72
mol) finely powdered 6-carbomethoxy-7-chloro-4-hydroxy-3-
(3,4,5-trimethylphenyl)-1H-quinolin-2-one (5). To the resulting
mixture was added 1.94 g (7.40 mmol) triphenylphosphine
then dropwise by syringe 1.2 mL (7.4 mmol) diethyl azodicar-
boxylate. The resulting mixture was stirred at ambient tem-
perature for 16 h at which time 50 mL silica gel was added to
the reaction mixture. The excess solvent was removed under
vacuum to provide a free flowing powder which was applied
to the top of a prepacked silica gel column. The column was
eluted with hexanes/ethyl acetate (65/35) to provide 3.2 g of
the product contaminated with a minor amount of diethyl
1
N,N′-hydrazinedicarboxylate. H NMR (400 MHz, CDCl3): δ
1.37 (s, 9H), 1.71 (m, 1H), 1.83 (m,1H), 2.15 (m, 2H), 2.20 (s,
3H), 2.32 (s, 6H), 3.70 (m, 4H), 3.94 (s, 3H), 4.14 (m, 1H), 6.50
(br. S, 1H), 7.10 (s, 2H), 7.32 (s, 1H), 8.42 (s, 1H). Resonances
for diethyl N,N′-hydrazinedicarboxylate: δ 1.25 (t), 4.19 (q).
FAB-MS: calcd for C30H35ClN2O6 554; found 555 (M + H, 100).
4-(2-(N-ter t-Bu toxyca r bon yla zetid in -2(S)-yl)eth oxy)-7-
ch lor o-2-oxo-3-(3,4,5-tr im eth ylp h en yl)-1,2-d ih yd r oqu in o-
lin e-6-ca r boxylic Acid . To a mixture of 3.2 g (∼5.8 mmol)
4-(2-(N-tert-butoxycarbonylazetidin-2(S)-yl)ethoxy)-6-carbo-
methoxy-7-chloro-3-(3,4,5-trimethylphenyl)-1H-quinolin-2-
one (7) in 65 mL 95% aqueous ethanol was added 2.42 g (57.6
mmol) lithium hydroxide monohydrate. The resulting mixture
was heated at 80 °C for 6.5 h. The reaction mixture was cooled
to room temperature and the solvent was removed under
vacuum. The residue was dissolved in a minimal amount of
water then cooled to 0 °C in an ice bath. The solution was
acidified to pH 2 with a saturated solution of potassium
hydrogen sulfate. The aqueous solution was extracted thor-
oughly with ethyl acetate (3×). The resulting organic layers
were combined, dried over sodium sulfate, filtered and the
solvent removed under vacuum to afford 1.79 g (86%, 2 steps)
of the product as a white solid. 1H NMR (400 MHz, CDCl3
containing 1 drop CD3OD): δ 1.31 (s, 9H), 1.68 (m, 1H), 1.78
(m,1H), 2.12 (m, 2H), 2.25 (s, 3H), 2.36 (s, 6H), 3.64 (m, 3H),
3.70 (m, 1H), 4.12 (m, 1H), 7.01 (s, 2H), 7.27 (s, 1H), 8.40 (s,
1H). FAB-MS: calcd for C29H33ClN2O6 540; found 541 (M +
H), 358 (M - 182, 100).
4-(2-(N-ter t-Bu toxyca r bon yla zetid in -2(S)-yl)eth oxy)-7-
ch lor o-2-oxo-3-(3,4,5-tr im eth ylp h en yl)-1,2-d ih yd r oqu in o-
lin e-6-ca r boxylic Acid P yr im id in -4-yla m id e (8). To a
solution of 245 mg (0.45 mmol) 4-(2-(N-tert-butoxycarbonyl-
azetidin-2(S)-yl)ethoxy)-7-chloro-2-oxo-3-(3,4,5-trimethylphen-
yl)-1,2-dihydroquinoline-6-carboxylic acid in 4 mL dry meth-
ylene chloride under nitrogen atmosphere were added 55 mg
(0.45 mmol) 4-(dimethylamino)pyridine, 215 mg (2.26 mmol)
4-aminopyrimidine, and 260 mg (1.36 mmol) ethyl(dimethyl-
aminopropyl)carbodiimide. The resulting reaction mixture was
stirred at ambient temperature for 24 h, diluted with meth-
ylene chloride and transferred to a separatory funnel. The
organic layer was washed with water followed by brine. The
aqueous layers were back extracted with methylene chloride
In tr a ven ou s P h a r m a cok in etics of Com p ou n d 1 in
Rh esu s Mon k ey. Compound 1 in ethanol/PEG400/saline (10/

922 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 6
DeVita et al.
40/50, v/v/v) was administered by the brachiocephalic vein to
male rhesus monkeys (average dose, 2.64 mg/kg). Blood was
collected from the femoral vein at 5, 15, 30 min and at 1, 2, 4,
6, 8, 24 h after dosing. From the blood collected at each
sampling, plasma was obtained by centrifugation. Concentra-
tions of compound 1 in plasma were determined by LC/MS/
MS after liquid/liquid extraction (ethyl acetate) versus an
internal standard. The limit of quantitation was 2.0 ng/mL.
The HPLC system consisted of two Shimadzu LC-600
pumps, an SCL-6B controller and an SIL-6B autoinjector.
Chromatography was carried out on a Spheresorb C8 (5 µm,
4.6 × 50 mm) column using isocratic mobile phase consisting
of 80% acetonitrile with 20% water with 10 mM ammonium
acetate and 0.1% trifluoroacetic acid. The flow rate was 1.0
mL/min.
Ack n ow led gm en t. We thank Amy Bernick, Ziqiang
Guan, Debra Zink, Nathan Yates, and Pat Griffin for
providing mass spectrometry services; Phil Eskola, Glen
Reynolds, J oe Leone, J udith Pisano, and Steven Fabian
for preparation of several synthetic intermediates;
Susan Iliff for assistance with the rhesus pharmacoki-
netic study.
Su p p or tin g In for m a tion Ava ila ble: Time course graphs
of hormone levels (LH and T) for the other three subjects in
the rhesus monkey study. This material is available free of
charge via the Internet at http://pubs.acs.org.
Refer en ces
LC/MS/MS assays were performed on a SCIEX API III
tandem mass spectrometer using the heated nebulizer inter-
face. Mass spectra and production spectra were obtained using
an ionspray interface and positive ion detection was used with
argon as the collision gas.
(1) (a) Matsuo, H.; Baba, Y.; Nair, R. M. G.; Arimura, A.; Schally,
A. V. Structure of the Porcine LH- and FSH-Releasing Hormone.
I. The Proposed Amino Acid Sequence. Biochem. Biophys. Res.
Commun. 1971, 43, 1334-1339. (b) Baba, Y.; Matsuo, H.;
Schally, A. V. Structure of the Porcine LH- and FSH-Releasing
Hormone. II. Confirmation of the Proposed Structure by Con-
ventional Sequential Analyses. Biochem. Biophys. Res. Commun.
1971, 44, 459-463.
(2) Conn, P. M.; J anovick, J . A.; Stanislaus, D.; Kuphal, D.; J ennes,
L. In Vitamins and Hormones; Litwack, G., Ed.; Academic
Press: New York, 1995; Vol. 50, pp 151-214.
(3) Kardamakis, E.; Tzigounis, V. GnRH Analogues in Reproductive
Medicine and Gynaecologic Cancer. In Biomedical Health Re-
search: Bioactive Peptides in Drug Discovery and Design:
Medical Aspects; Marsoukas, J ., Mavromoustakos, T., Eds.; IOS
Press, 1999; Vol. 22, pp 275-285.
(4) Goulet, M. T. Gonadotropin Releasing Hormone Antagonists. In
Annual Reports in Medicinal Chemistry; Bristol, J . A., Ed.;
Academic Press: New York, 1995; Vol. 30, pp 169-178.
(5) DeVita, R. J .; Hollings, D. D.; Goulet, M. T.; Wyvratt J r., M. J .;
Fisher, M. H.; Lo, J .-L.; Yang, Y. T.; Cheng, K.; Smith, R. G.
Identification and Initial Structure-Activity Relationships of a
Novel Non-Peptide Quinolone GnRH Antagonist. Bioorg. Med.
Chem. Lett. 1999, 9, 2615-2620.
Areas under the plasma concentration versus time curve
(AUC) were determined using linear trapezoidal interpolation
in the ascending slope and logarithmic trapezoidal interpola-
tion in the descending slope. Concentrations lower than the
limit of quantification were treated as zero for the purpose of
calculation of AUC, AUMC, and means. The portion of the
AUC from the last measurable plasma concentration to infinity
was estimated by Ct/λ, where Ct represents the last measurable
plasma concentration and λ is the terminal rate constant
determined from the plasma concentration versus time curve
by the linear regression of the elimination phase of the
semilogarithmic plot. The portion of the AUMC from the last
measurable plasma concentration to infinity was estimated by
t × Ct/λ + Ct/λ2. Elimination half-life (t1/2) and plasma clearance
(Clp) were calculated by the following equations: t1/2 ) 0.693/
λ; Clp ) dose/AUC(0-∞)
.
In Vivo Effica cy in Rh esu s Ma ca qu es. Four adult male
rhesus macaques were previously administered with chronic
intravenous catheters maintained by constant infusion of
saline containing heparin. Catheters were tunneled sc to exit
via a stainless steel tether and passed through a wall to allow
for blood sampling from another room. This allowed the
animals to be dosed and have blood samples collected without
disturbing the animals or turning on the lights in the room.
Lights were set to a 0700:20:00 h light-dark cycle. Animals
were completely conditioned to this arrangement over the
course of several months. They were fed on a daily schedule
of 08:30 and 15:00.
All animals received compound 1 at 0.5 or 3.0 mg/kg as a
single iv dose dissolved in ethanol, PEG400 and saline (10/
40/50). Two hours before dosing, blood sampling began, and
samples of 1 mL were collected into heparinized syringes every
20 min throughout the study. Initiation of blood sampling was
at 16:00, compound or vehicle administration was at 18:00,
and sampling finished at 02:00. Approximately every hour, red
cells were returned to the animal via the iv catheter, using
sterile technique. A total blood volume of approximately 40
mL was collected during the study. Compound 1 was dosed
using a 10-12 min iv infusion. RIA was used to determine
serum LH and T levels from the samples obtained at each time
point. Hormone AUC was calculated using a standard graph-
ing program for all hormone datapoints recorded during the
duration of the study (t ) -130 to 470 min).
(6) Young, J . R.; Chen, I.; Walsh, T. F.; DeVita, R. J .; Wyvratt J r.,
M. J .; Fisher, M. H.; Goulet, M. T.; Ren, N.; Lo, J .-L.; Yang, Y.
T.; Yudkovitz, J . B.; Cheng, K.; Smith, R. G. Quinolones as
Gonadotropin Releasing Hormone (GnRH) Antagonists: Simul-
taneous Optimization of the C(3)-Aryl and C(6)-Substituents.
Bioorg. Med. Chem. Lett. 2000, 10, 4435-4439.
(7) DeVita, R. J .; Goulet, M. T.; Wyvratt J r., M. J .; Fisher, M. H.;
Lo, J .-L.; Yang, Y. T.; Cheng, K.; Smith, R. G. Investigation of
the 4-O-Alkylamine Substituent of Non-Peptide Quinolone
GnRH Antagonists. Bioorg. Med. Chem. Lett. 1999, 9, 2621-
2624.
(8) Walsh, T. F.; Toupence, R. B.; Young, J . R.; Huang, S. X.;
Ujjainwalla, F.; DeVita, R. J .; Goulet, M. T.; Wyvratt, M. J .;
Fisher, M. H.; Lo, J .-L.; Ren, N.; Yudkovitz, J . B.; Yang, Y. T.;
Cheng, K.; Smith, R. G. Potent Antagonists of Gonadotropin
Releasing Hormone Receptors Derived from Quinolone-6-Car-
boxamides. Bioorg. Med. Chem. Lett. 2000, 10, 443-447.
(9) Bennington, F.; Morin, R. D.; Clark, J r., L. C. Mescaline
Analogues. VII. 3,4,5-Trimethyl-â-phenethylamine. J . Org. Chem.
1957, 22, 332-333.
(10) (a) Cassal, J .-M.; Furst, A.; Meier, W. 197. Synthese der
enatiomeren 2-Pyrrolidinessigsauren. Helv. Chim. Acta 1976,
59, 1917-1924. (b) Ye, J .; McKervey, M. A. Synthesis of Chiral
N-Protected R-Amino-â-Diketones From R-Diazoketones Derived
from Natural Amino Acids. Tetrahedron 1992, 48, 8007-8022.
(11) Cui, J .; Smith, R. G.; Mount, G. R.; Lo, J . L.; Yu, J .; Walsh, T.
F.; Singh, S. B.; DeVita, R. J .; Goulet, M. T.; Schaeffer, J . M.;
Cheng, K. Identification of Phe313 of the GnRH receptor as a site
critical for the binding of non-peptide GnRH antagonists. Mol.
Endocrinol. 2000, 14, 671-681.
J M000275P