Download the pharmacokinetics of a thiazole benzenesulfonamide

0090-9556/02/3007-771–777$7.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2002 by The American Society for Pharmacology and Experimental Therapeutics
DMD 30:771–777, 2002
Vol. 30, No. 7
630/988017
Printed in U.S.A.
THE PHARMACOKINETICS OF A THIAZOLE BENZENESULFONAMIDE ␤3-ADRENERGIC
RECEPTOR AGONIST AND ITS ANALOGS IN RATS, DOGS, AND MONKEYS: IMPROVING
ORAL BIOAVAILABILITY
RALPH A. STEARNS, RANDY R. MILLER, WEI TANG, GLORIA Y. KWEI, FRANK S. TANG, ROBERT J. MATHVINK,
ELIZABETH M. NAYLOR, DAWN CHITTY, VINCENT J. COLANDREA, ANN E. WEBER, ADRIA E. COLLETTI,
JOHN R. STRAUSS, CAROL ANN KEOHANE, WILLIAM P. FEENEY, SUSAN A. ILIFF, AND SHUET-HING LEE CHIU
Departments of Drug Metabolism (R.A.S., R.R.M., W.T., A.E.C., F.S.T., C.A.K., S-H.L.C.), Medicinal Chemistry (R.J.M., E.M.N., D.C., V.J.C.,
A.E.W.), Comparative Medicine (W.P.F., J.R.S., S.A.I.), and Pharmaceutical Research and Development (G.Y.K.), Merck Research Laboratories,
Rahway, New Jersey
(Received November 2, 2001; accepted March 18, 2002)
ABSTRACT:
The pharmacokinetics and oral bioavailability of (R)-N-[4-[2-[[2hydroxy-2-(pyridin-3-yl)ethyl]amino]ethyl]phenyl]-4-[4-[4-(trifluoromethylphenyl]thiazol-2-yl]benzenesulfonamide (1), a 3-pyridyl thiazole
benzenesulfonamide ␤3-adrenergic receptor agonist, were investigated
in rats, dogs, and monkeys. Systemic clearance was higher in rats (⬃30
ml/min/kg) than in dogs and monkeys (both ⬃10 ml/min/kg), and oral
bioavailability was 17, 27, and 4%, respectively. Since systemic clearance was 25 to 40% of hepatic blood flow in these species, hepatic
extraction was expected to be low, and it was likely that oral bioavailability was limited either by absorption or a large first-pass effect in the
gut. The absorption and excretion of 3H-labeled 1 were investigated in
rats, and only 28% of the administered radioactivity was orally absorbed. Subsequently, the hepatic extraction of 1 was evaluated in rats
(30%) and monkeys (47%). The low oral bioavailability in rats could be
explained completely by poor oral absorption and hepatic first-pass
metabolism; in monkeys, oral absorption was either less than in rats or
first-pass extraction in the gut was greater. In an attempt to increase
oral exposure, the pharmacokinetics and oral bioavailability of two
potential prodrugs of 1, an N-ethyl [(R)-N-[4-[2-[ethyl[2-hydroxy-2-(3pyridinyl)ethyl]amino]ethyl]phenyl]-4-[4-[4-(trifluoromethyl)phenyl]thiazol-2-yl]benzenesulfonamide; 2] and a morpholine derivative [(R)-N-[4[2-[2-(3-pyridinyl)morpholin-4-yl]ethyl]phenyl]-4-[4-[4-(trifluoromethyl)phenyl]thiazol-2-yl]benzenesulfonamide; 3], were evaluated in monkeys. Conversion to 1 was low (⬍3%) with both derivatives, and neither
entity was an effective prodrug, but the oral bioavailability of 3 (56%)
compared with 1 (4%) was significantly improved. The hypothesis that
the increased oral bioavailability of 3 was due to a reduction in hydrogen bonding sites in the molecule led to the design of (R)-N-[4-[2-[[2hydroxy-2-(pyridin-2-yl)ethyl]amino]ethyl]phenyl]-4-[4-(4-trifluoromethylphenyl)thiazol-2-yl]benzenesulfonamide (4), a 2-pyridyl ␤3-adrenergic
receptor agonist with improved oral bioavailability in rats and monkeys.
The prevalence of obesity is increasing throughout the world and
especially in the United States (Kuczmarski et al., 1994, Wickelgren,
1998). This health risk is associated with the onset of other conditions
such as noninsulin-dependent diabetes and cardiovascular disease
(Pi-Sunyer, 1993). Basically, obesity is the result of an imbalance
between caloric intake and energy expenditure. The challenge of
reducing caloric intake in an obese population has led to a focus on
pharmaceutical intervention to achieve weight loss. A number of
therapeutic strategies have been explored for the pharmacological
treatment of obesity (Kordik and Reitz, 1999). One potential mechanism for increasing energy expenditure is via activation of a unique
␤-adrenergic receptor subtype, the ␤3-adrenergic receptor (Arch and
Wilson, 1996; Dow, 1997; Weber, 1998). This receptor has been
identified on the surface of rat adipocytes, and its activation results in
the stimulation of lipolysis (Arch et al., 1984). In brown adipose
tissue, ␤3-adrenergic receptor activation leads to up-regulation of
UCP1, a mitochondrial uncoupling protein, which uncouples oxidative phosphorylation from fatty acid ␤-oxidation, resulting in an
increase in energy expenditure (Lowell and Flier, 1997). In rats, this
increase is reflected in thermogenesis and weight loss due to a
decrease in body lipid with no loss in muscle mass (Arch et al., 1990).
Recently, it was demonstrated that chronic treatment of monkeys with
selective human ␤3-adrenergic receptor agonists elicits lipolysis and
metabolic rate elevation (Fisher et al., 1998; Hom et al., 2001). Thus,
␤3-adrenergic receptor agonists might be useful for the treatment of
obesity in humans. A major barrier to the development of this class of
compounds has been low oral bioavailability. Much effort has focused
in the design of more orally bioavailable ␤3-adrenergic receptor
agonists (Naylor et al., 1998, 1999; Weber et al., 1998; Parmee et al.,
1999; Shih et al., 1999; Biftu et al., 2000; Brockunier et al., 2000;
Feng et al., 2000; Ok et al., 2000). In this paper, we report the
pharmacokinetics of the ␤3-adrenergic receptor agonist (R)-N-[4-[2[[2-hydroxy-2-(pyridin-3-yl)ethyl]amino]ethyl]phenyl]-4-[4-[4-
Address correspondence to: Ralph A. Stearns, Ph.D., Department of Drug
Metabolism, Merck and Co., P.O. Box 2000, RY800B211, Rahway, NJ, 07065.
E-mail: [email protected]
771
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 11, 2017
This article is available online at http://dmd.aspetjournals.org
772
STEARNS ET AL.
(trifluoromethylphenyl]thiazol-2-yl]benzenesulfonamide (11, Fig. 1)
and its analogs (Mathvink et al., 2000b) in rats, dogs, and monkeys.
We describe efforts to identify the factors that contribute to the poor
oral bioavailability of 1. Based on that information, we proposed a
rationale for redesign that led to (R)-N-[4-[2-[[2-hydroxy-2-(pyridin2-yl)ethyl]amino]ethyl]phenyl]-4-[4-(4-trifluoromethylphenyl)thiazol-2-yl]benzenesulfonamide (4, Figure 1), a ␤3-adrenergic receptor
agonist with improved oral bioavailability in rats and monkeys.
Materials and Methods
Chemicals. 1, 2, 3, and 4 were obtained from Merck Research Laboratories.
H-labeled 1 (226.9 mCi/mg) was synthesized by the Labeled Compound
Synthesis Group, Merck Research Laboratories. The purity of the labeled and
unlabeled compounds was ⬎97%. HPLC grade solvents were from Fisher
Scientific (Fair Lawn, NJ). Ammonium acetate, 85% lactic acid solution, PEG
400, and TFA were purchased from Aldrich Chemical Co. (Milwaukee, WI).
Isotonic saline solution was from Radix Laboratories (Eau Claire, WI).
Animal Studies. All procedures were approved by the Merck Research
Laboratories Institutional Animal Care and Use Committee. All dose levels are
reported as the free base.
Studies in Rats. Male Sprague-Dawley rats (Charles River Laboratories
Inc., Wilmington, MA) weighing 300 to 400 g were used in all studies. The rats
were housed under standard conditions and had ad libitum access to water and
a standard laboratory diet. All rats were dosed following an overnight fast;
3
1
Abbreviations used are: 1, (R)-N-[4-[2-[[2-hydroxy-2-(pyridin-3-yl)ethyl]amino]ethyl]phenyl]-4-[4-[4-(trifluoromethylphenyl]thiazol-2-yl]benzenesulfonamide; 2,
(R)-N-[4-[2-[ethyl[2-hydroxy-2-(3-pyridinyl)ethyl]amino]ethyl]phenyl]-4-[4-4(trifluoromethyl)phenyl]thiazol-2-yl]benzenesulfonamide; 3, (R)-N-[4-[2-[2-(3-pyridinyl)morpholin-4-yl]ethyl]phenyl]-4-[4-[4-(trifluoromethyl)phenyl]thiazol-2-yl]benzenesulfonamide; 4, (R)-N-[4-[2-[[2-hydroxy-2-(pyridin-2-yl)ethyl]amino]ethyl]phenyl]-4-[4(4-trifluoromethylphenyl)thiazol-2-yl]benzenesulfonamide; 5, (R)-4-[5-(3,4-difluorophenylmethyl)-1,2,4-oxadiazol-3-yl]-N-[4-[2-[[2-hydroxy-2-(3-pyridinyl)ethyl]amino]
ethyl]phenyl]benzenesulfonamide; HPLC, high pressure liquid chromatography; TFA,
trifluoroacetic acid; PEG 400, polyethylene glycol 400; LC-MS/MS, liquid chromatography-tandem mass spectrometry; AUC, area under the plasma concentrationtime curve; Clp, plasma clearance; Vdss, volume of distribution at steady state.
food was returned 2 to 4 h after dosing. Water was available ad libitum
throughout the experiments.
Pharmacokinetics. Polyethylene and/or microrenethane cannulas were implanted in the femoral artery and vein 2 days before the experiment while the
animals were anesthetized with pentobarbital (50 mg/kg i.p.). The cannulas
were externalized at the back of the neck and filled with heparinized saline (20
units/ml). Three rats were dosed with 1 i.v. at 3 mg/kg (1 ml/kg; 3 mg/ml) in
0.3 M lactic acid. Three rats were dosed by oral gavage at 10 mg/kg (2 ml/kg;
5 mg/ml) in a vehicle consisting of 0.5% methylcellulose (Methocel:Tween 80
(90:10, v/v). Serial blood samples (0.3 ml) were collected via the arterial
cannula at 2 (i.v. only), 5, 15, and 30 min and at 1, 2, 4, 6, 8, 10, and 24 h post
dose. Blood was transferred to heparinized microcentrifuge tubes; plasma was
collected by centrifugation and stored at ⫺20°C. Rats were euthanized after the
last blood sample. The same oral dose formulation was used for compound 4.
For i.v. dosing, 4 was administered at the same concentration as 1 but in a
vehicle that consisted of ethanol/polyethylene glycol 400/0.9% saline (20:60:
20, v/v/v).
Hepatic extraction. Nine rats (N ⫽ 3/dose) had cannulas surgically implanted
in the femoral artery, femoral vein, and portal vein. Rats were dosed with 1 by
30 min infusion via the femoral vein (N ⫽ 3) or via the portal vein (N ⫽ 3) at
0.2 mg/kg (1 ml/kg; 0.2 mg/ml) in a vehicle consisting of ethanol/polyethylene
glycol 400/0.9% saline (20:60:20, v/v/v). Three rats were dosed with 1 by oral
gavage at 10 mg/kg (2 ml/kg; 5 mg/ml) in 0.5% Methocel containing 0.03 M
HCl. Serial blood samples (0.3 ml) were collected from the portal vein and
femoral artery at 10, 20, and 30 min after the onset of infusion and at 5, 15, and
30 min and at 1, 2, 4, 10, and 24 h postinfusion. Only arterial blood was
sampled from the rats dosed via the femoral vein. Blood was transferred to
heparinized microcentrifuge tubes; plasma was collected by centrifugation and
stored at ⫺20°C. The same oral dose formulation was used for compound 4.
For the femoral and portal vein routes of administration for the infusion
studies; however, 4 was administered as described above for 1, but the vehicle
was fresh rat plasma.
Absorption and excretion in rats. To determine route of excretion, silastic
cannulas were implanted in the bile duct and the proximal duodenum 1 day
before the experiment while the rats were under pentobarbital anesthesia. The
tubing was externalized at the center of the back and protected using a Harvard
Apparatus tethering system (Harvard Apparatus, Holliston, MA). The cannulas
were reconnected to permit recirculation of bile. All rats had polyethylene
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 11, 2017
FIG. 1. Structures of thiazole benzenesulfonamide ␤3-adrenergic receptor agonists.
PHARMACOKINETICS OF ␤3-ADRENERGIC RECEPTOR AGONISTS
trifugation. Concentrations of 1 in blood and plasma were quantified using an
LC/MS/MS method.
In vitro. Stock solutions of 1 were made at 1, 5, 20, 50, and 100 ␮g/ml in
0.02 M lactic acid. A stock solution of 3H-labeled 1 (0.02mCi/ml) was made
in the same vehicle. Aliquots of rat, dog, and monkey blood (1 ml) were
preincubated at 37°C for 15 min and then spiked with 10 ␮l of the radiolabel
solution and 10 ␮l of stock solution to give final concentrations that ranged
from 10 to 1000 ng/ml (each containing 0.2 ␮Ci). The pH of the resulting
samples was 7.4. Samples were incubated for an additional 20 min at 37°C, and
then plasma was prepared by centrifugation. Radioactivity content in 100-␮l
plasma aliquots was determined by direct liquid scintillation counting and
compared with the recovery of radioactivity from an aliquot of plasma spiked
with the same amount of radioactivity.
Analytical Assays. HPLC. HPLC system components included a Shimadzu
SCL 10A system controller, two Shimadzu LC-10AD pumps (Shimadzu
Scientific Instruments, Kyoto, Japan), and a PerkinElmer series 200 autoinjector (PerkinElmer Instruments, Norwalk, CT).
Radioactivity measurement. Radioactivity in plasma was determined by
counting 0.1 to 0.2 ml aliquots in glass vials containing 6 ml of liquid
scintillation cocktail (Insta-Gel XF, Packard BioScience, Meriden, CT). Blood
and feces homogenates (in triplicate) were transferred to combustion cups,
dried and combusted on a Packard model 307 sample oxidizer. Radioactivity
measurements were made in a Beckman LS5000TD liquid scintillation spectrometer (Beckman Coulter, Inc., Fullerton, CA). Quench corrections were
automatic and by the H-number method.
LC/MS/MS. Plasma or blood concentrations of 1 (or 4) were quantified by
using an LC/MS/MS method. A volume from 0.05 to 0.2-ml plasma or blood
was spiked with 20 ng 5, diluted with 0.5 ml 0.1M sodium carbonate (pH 10),
and extracted with 4 ml of ethyl acetate. The ethyl acetate extract was
evaporated under nitrogen, reconstituted in a mobile phase consisting of 80%
acetonitrile, 20% 10 mM NH4OAc, 0.1% TFA, and injected onto a Spherisorb
C8 HPLC column (4.6 ⫻ 50 mm; Waters Corp., Milford, MA) equilibrated
with mobile phase at a flow rate of 1 ml/min. HPLC column effluent was
analyzed directly on a Sciex API III mass spectrometer (PE Sciex, Ontario,
Canada) using the heated nebulizer interface (positive ion detection). The
parent/product ion combinations 625-487 and 592-574 were used for multiple
reaction monitoring of 1 (or 4) and 5, respectively. A standard curve was
generated from the mean of three replicates, which were made by spiking an
equal volume of plasma from untreated animals with increasing amounts
(0.1–1000 ng; 11 concentrations and a blank control) of 1 (or 4) and 20 ng of
the internal standard (5). The peak area ratio of the 1 (or 4) response to that of
the internal standard was plotted versus the amount of drug spiked to obtain the
standard curve. The amount of drug in plasma was quantified by comparison
of its peak area ratio in a given sample to the standard curve. The lower limit
of quantification was typically between 0.5 and 2 ng/ml. Precision was assessed every time the assay was performed by using a minimum of three
replicates at each concentration. At each standard concentration (0, 0.1, 0.2,
0.5, 1, 5, 10, 50, 100, 500, and 1000 ng), the percent relative standard deviation
was typically less than 10%. Accuracy of the regression fit also was assessed
by calculating the read back concentrations of the standards, and these typically were within 10% of their nominal values. Prodrugs 2 and 3 were
quantified as described above with the exception that the parent/product ion
combinations m/z 653-635 and 651-558, respectively, were used for multiple
reaction monitoring.
Pharmacokinetic Calculations. Pharmacokinetic parameters after i.v. and
p.o. dosing were analyzed by noncompartmental methods (Gibaldi and Perrier,
1982). Plasma AUCo-x values (x being the time of the last plasma concentration measured) were estimated by using the log-linear trapezoidal method. The
terminal phase was determined by visual inspection of the log-transformed
concentration-time data. The terminal phase rate constant (kel) was obtained
from linear regression analysis of the log-transformed terminal phase concentration-time data. AUCx-⬁ was estimated by dividing the last plasma concentration value measured by the terminal phase rate constant; this residual area
was added to AUCo-x to obtain AUCo-⬁. In all cases, extrapolated areas
(AUCx-⬁) accounted for less than 30% and typically less than 10% of the total
area under the plasma concentration-time curves (AUCo-⬁). Plasma clearance
(Clp), terminal phase half-life (t1/2), volume of distribution at steady state
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 11, 2017
cannulas implanted in the femoral artery to facilitate collection of blood;
animals to be dosed intravenously had a microrenethane cannula implanted in
the femoral vein. The vascular cannulas were externalized at the same site as
the biliary cannulas. The rats were placed in Nalge metabolism cages (Nalge
Nunc International, Naperville, IL) where they were given access to a solution
of 280 mM dextrose:150 mM NaCl:7 mM KCl ad libitum. The next day, the
cannulas were attached to a dual channel swivel for collection of bile and
infusion of 10 mM sodium taurocholate:280 mM dextrose:150 mM NaCl:7
mM KCl (2 ml/h). Rats were dosed with 1 by oral gavage at 10 mg/kg (N ⫽
3) or by bolus injection into the venous cannula at 3 mg/kg (N ⫽ 3). Bile, urine,
and feces were collected for 5 days; blood samples (0.3 ml) were collected at
predetermined time points as described previously. The volume of bile was
determined gravimetrically, assuming a density of 1.0 g/ml.
Studies in Dogs. Pharmacokinetics. Six male beagle dogs, weighing 13 to
16 kg were used for the experiment. The dogs were fed a standard lab diet and
were fasted overnight before being dosed; food was returned 4 h after dosing.
Water was available ad libitum throughout the experiment.
Three dogs were dosed by gastric intubation with 1 (10 mg/kg; 2 mg/ml)
prepared in a vehicle of 0.5% Methocel containing 0.03 M HCl. Three dogs were
dosed intravenously via the cephalic vein with 1 (3 mg/kg; 0.5 ml/kg), which had
been formulated as a solution in ethanol/polyethylene glycol 400/0.9% saline
(20:60:20, v/v/v). Serial blood samples (5 ml) were collected by venipuncture from
the jugular vein into Vacutainer tubes containing heparin and stored on ice (BD
Vacutainer Systems, Franklin Lakes, NJ). Collection was made predose and at 2
(i.v. only), 5, 15, and 30 min and at 1, 2, 4, 6, 8, 10, and 24 h postdose. Plasma was
collected by centrifugation and stored at ⫺20°C.
Studies in Monkeys. Pharmacokinetics. Three male rhesus monkeys (Macaca mulatta), weighing 5 to 8 kg, were used for the experiment. The monkeys
were fed a standard lab diet and were fasted overnight before being dosed; food
was returned 4 h after dosing. Water was available ad libitum throughout the
experiment. The experiment was conducted as a two-period crossover study. In
the first period, the monkeys received via nasogastric intubation a solution of
1 (10 mg/kg; 5 mg/ml) in 0.5% Methocel containing 0.02 M lactic acid. One
of the monkeys did not receive the complete oral dose, and data obtained from
this animal was not included in pharmacokinetic calculations. After a 3-week
washout period, the same monkeys were dosed i.v. via the cephalic vein with
1 (3 mg/kg; 6 mg/ml) formulated as a solution in 0.3 M lactic acid. Serial blood
samples (2 ml) were collected from the saphenous vein at 2 (i.v. only), 5, 15,
and 30 min and at 1, 2, 4, 6, 8, 10, and 24 h postdose. At 10 and 24 h postdose,
samples were collected by venipuncture from the femoral vein while the
monkeys were under Telazol anesthesia. The same i.v. and oral dose formulations were used for compounds 2, 3, and 4.
Hepatic extraction. Four male rhesus monkeys (6 – 8 kg) with vascular
access ports chronically implanted in the portal vein and iliac artery were used
for the experiment. Temporary vascular cannulas were implanted in the saphenous vein to facilitate systemic blood collection. The monkeys were dosed
after an overnight fast. Water was available ad libitum; food was returned 2 h
after dosing. The experiment was conducted in three periods, with at least a
3-week washout between periods. In period 1, the monkeys were dosed by
nasogastric intubation with 10 mg/kg (3 mg/ml) of 1 formulated in 0.5%
Methocel containing 0.03 M lactic acid. Systemic venous blood samples (3 ml)
were collected from each animal predose, at 15 and 30 min and 1, 2, 4, 6, 8,
10, and 24 h postdose. In period 2, the monkeys were dosed via intravenous
infusion into the cephalic vein for 1 h with 0.3 mg/kg (1 ml/kg) of 1 in
ethanol/PEG 400/0.9% saline (20:60:20, v/v/v). Systemic venous blood samples (3 ml) were collected from each animal predose, at 5, 15, 30, and 45 min
after the onset of infusion and at 2, 5, 15, and 30 min and 1, 2, 4, 6, and 8 h
postinfusion. In period 3, the animals were dosed via portal vein infusion for
1 h with 0.3 mg/kg (1 ml/kg) of 1 in ethanol/PEG 400/0.9% saline (20:60:20,
v/v/v). Systemic venous blood samples (3 ml) were collected from each animal
as described above for the i.v. infusion.
Determination of Blood-to-Plasma Ratio. In vivo. Three male SpragueDawley rats with cannulas implanted in the femoral artery and vein were dosed
by i.v. bolus with 1 at 3 mg/kg (0.5 ml/kg) prepared as a solution in ethanol/
PEG 400/0.9% saline (20:60:20, v/v/v). Serial blood samples (300 ␮l) were
collected prior to dosing and at 2, 5, 15, and 30 min and at 1, 2, 4, 6, 8, 10, and
24 h postdose. Aliquots of the blood samples (150 ␮l) were transferred to
microcentrifuge tubes containing heparin, and plasma was collected by cen-
773
774
STEARNS ET AL.
TABLE 1
Summary of pharmacokinetic parameters for 1 in preclinical species
AUCnorm
Clp
Clbb
t1/2
VdSS
␮g 䡠 min 䡠 kg/mg 䡠 ml
20.0 ⫾ 4.5a
16.2 ⫾ 0.9
24.5 ⫾ 6.0
51.5 ⫾ 5.8
44.3 ⫾ 4.7
54.5 ⫾ 11.8
ml/min/kg
52.0 ⫾ 13.7
61.7 ⫾ 3.4
42.8 ⫾ 12.0
19.6 ⫾ 2.1
22.8 ⫾ 2.3
19.0 ⫾ 4.6
ml/min/kg
26
31
22
10
11
10
h
NC
8.1 ⫾ 0.2
11.5 ⫾ 3.2
14.0 ⫾ 3.0
NC
9.0 ⫾ 0.9
l/kg
NC
22.7 ⫾ 2.3
16.9 ⫾ 6.6
12.0 ⫾ 1.1
NC
6.2 ⫾ 1.8
AUCnorm
Eh
Tmax
F
h
2.0 ⫾ 0.0
4.0 ⫾ 0.0
3.4 ⫾ 2.4
3.0 ⫾ 1.4
4.0 ⫾ 2.0
%
17 ⫾ 6
19 ⫾ 7
27 ⫾ 8
4⫾1
5⫾2
i.v. Dose
Rats
Dogs
Monkeys
mg/kg
0.2 (N ⫽ 3)
3 (N ⫽ 3)
3 (N ⫽ 3)c
3 (N ⫽ 3)
0.3 (N ⫽ 3)d
3 (N ⫽ 3)e
p.v. Dose
Rats
Monkeys
Rats
Oral Dose
AUCnorm
mg/kg
10 (N ⫽ 3)
10 (N ⫽ 3)c
10 (N ⫽ 5)
10 (N ⫽ 2)e
10 (N ⫽ 3)d
␮g 䡠 min 䡠 kg/mg 䡠 ml
2.7 ⫾ 0.9
4.8 ⫾ 1.7
10.8 ⫾ 3.0
2.0 ⫾ 0.1
2.0 ⫾ 0.1
%
30
47
Cmax
␮g/ml
0.126 ⫾ 0.082
0.091 ⫾ 0.023
0.150 ⫾ 0.032
0.032 ⫾ 0.005
0.035 ⫾ 0.018
Clb, blood (systemic) clearance; p.v., portal vein; NC, not calculated.
a
Values represent mean ⫾ S.D.
b
Blood clearance [calculated by dividing Clp by the blood-to-plasma ratio (Cb/Cp ⫽ 2)].
c
Data from study in bile duct-cannulated rats.
d
Data from 3-period crossover study to evaluate hepatic extraction in monkeys.
e
Results from initial 2-period crossover study in monkeys; one animal did not receive complete oral dose.
(Vdss), oral bioavailability (F), and hepatic extraction (Eh) were estimated
based on eqs. 1 through 6 shown below:
Cl p ⫽ Dose/AUC0-⬁
(1)
t 1/2 ⫽ ln2/kel
(2)
Vd ss ⫽ (Dose)(AUMC0-⬁)/(AUC0-⬁)
(3)
F ⫽ AUCoral 䡠 Doseiv/AUCiv 䡠 Doseoral ⫽ fh 䡠 fg 䡠 fabs
(4)
f h ⫽ AUCpv/AUCiv
(5)
E h ⫽ 1 ⫺ fh
(6)
2
(AUCpv is the systemic AUC value obtained after portal vein infusion;
AUMC is the area under the first moment of the plasma concentration-time
curve; fh, fg, and fabs are the hepatic, gut, and absorbed fractions). For infusion
studies, clearance values (Clp) were calculated as described above, and AUCo-x
was calculated from the onset of infusion until 8 to 24 h postinfusion. AUC
values were estimated by using the log-linear trapezoidal method. Residual
values were not added in these studies since plasma drug concentrations
postinfusion in rats and monkeys at the last time point were near the lower
limit of quantification of the LC/MS/MS assay. When i.v. and p.o. pharmacokinetic data were obtained in different animals, values for oral bioavailability were calculated by taking AUCpo of each individual animal and dividing by
the mean AUCiv value.
Results
The pharmacokinetics and oral bioavailability of 1 (Fig. 1) were
investigated in Sprague-Dawley rats, beagle dogs, and rhesus monkeys dosed orally at 10 mg/kg and intravenously at doses that ranged
from 0.2 to 3 mg/kg. Pharmacokinetic data are summarized in Table
1. Following i.v. dosing, the total plasma clearance (Clp) was higher
in rats (62 ml/min/kg) than in dogs (20 ml/min/kg) and monkeys (19
ml/min/kg). In rats and monkeys, the increase in plasma AUC value
was proportional to the increase in dose over the range from 0.2(0.3)
to 3 mg/kg. A single i.v. dose was evaluated in dogs. The volume of
distribution at steady state (Vdss) was generally large and ranged from
23 l/kg in rats and 12 l/kg in dogs to 6 l/kg in monkeys. The high
plasma clearance values obtained with 1 prompted examination of the
blood-to-plasma partition ratio.
The blood-to-plasma concentration ratios of 3H-labeled 1, determined in vitro in rat, dog, and monkey blood, were approximately 1.6
to 2 over a drug concentration range of 10 to 1000 ng/ml. This
estimate was supported by a time course study of the distribution of 1
in blood collected from i.v.-dosed rats over a 24-h period, in which the
blood-to-plasma concentration ratio was ⬃2 over a range from 13 to
5900 ng/ml.
After oral dosing, 1 was absorbed with a Tmax that ranged from 2
to 3 h in rats, dogs, and monkeys. Dose-normalized plasma AUC
values for the parent drug were higher in dogs (11 ␮g 䡠 min 䡠 kg/mg 䡠
ml) than in rats (3 ␮g 䡠 min 䡠 kg/mg 䡠 ml) or in monkeys (2 ␮g 䡠 min 䡠
kg/mg 䡠 ml). The oral bioavailability in rats and dogs was 17 and 27%,
respectively, and was lower in monkeys (4%).
The absorption and excretion of 1 were investigated in rats. 3Hlabeled 1 was given orally and intravenously to bile duct-cannulated
rats, and the excretion of radioactivity in bile, urine, and feces was
monitored. Data are presented in Table 2. Five days after i.v. dosing,
63 ⫾ 13% of the administered dose was recovered in bile, 11 ⫾ 2%
in urine, and 9 ⫾ 3% in feces. After oral dosing, 22 ⫾ 9% of the dose
was recovered in bile, 6 ⫾ 3% in urine, and 75 ⫾ 20% in feces. Most
of the radioactivity was excreted within 24 h post i.v. and p.o. dosing.
Pharmacokinetic data from both studies are shown in Table 1.
Pharmacokinetic studies also were carried out in portal and femoral
vein-cannulated rats and rhesus monkeys to evaluate hepatic extraction in these species. Animals were dosed via infusion into the portal
or femoral vein, and systemic circulation was sampled at specified
intervals postdose. Data are presented in Table 1. Hepatic extraction
was calculated by comparing plasma AUC values obtained in rats and
monkeys after i.v. (20.0 and 44.3 ␮g 䡠 min 䡠 kg/mg 䡠 ml) and portal
vein (14.0 and 23.7 ␮g 䡠 min 䡠 kg/mg 䡠 ml) infusions. Hepatic
extraction of 1 was 30 and 47%, respectively.
To determine whether N-alkyl derivatives of 1 could serve as
prodrugs and improve oral absorption, particularly in monkeys, the
pharmacokinetics of the N-ethyl (2) and morpholine (3) derivatives of
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 11, 2017
Dogs
Monkeys
␮g 䡠 min 䡠 kg/mg 䡠 ml
14.0 ⫾ 1.0
23.7 ⫾ 3.3
mg/kg
0.2 (N ⫽ 2)
0.3 (N ⫽ 4)e
PHARMACOKINETICS OF ␤3-ADRENERGIC RECEPTOR AGONISTS
TABLE 2
Disposition of 3H-labeled 1 in bile duct-cannulated rats
Values represent mean ⫾ S.D.
i.v. Dose
(3 mg/kg)a
Bile
Urine
Feces
Total:
a
Oral Dose
(10 mg/kg)
% of dose recovered
63 ⫾ 13
22 ⫾ 9
11 ⫾ 2
6⫾3
9⫾3
75 ⫾ 20
83 ⫾ 8
103 ⫾ 26
0–120 h:
0–120 h:
0–120 h:
0–120 h:
N ⫽ 3 rats/dose
Discussion
The pharmacokinetics and oral bioavailability of 1 were investigated in
rats, dogs, and monkeys. Moderate to high plasma clearance values (Clp)
in these species were attributed in part to partitioning into red blood cells
since blood-to-plasma ratios were estimated to be ⬃2. Therefore, blood
clearance was 2-fold lower than plasma clearance and approximately 25
to 40% of hepatic blood flow in each species. On this basis, if 1 were well
absorbed and cleared mainly in the liver, the oral bioavailability would be
expected to exceed 60%. After oral dosing, however, 1 was absorbed
relatively slowly with Tmax ranging from 2 h in rats to 3 h in dogs and
monkeys. Plasma AUC values for the parent drug were higher in dogs
(10.8 ␮g 䡠 min 䡠 kg/mg 䡠 ml) than rats (2.7 ␮g 䡠 min 䡠 kg/mg 䡠 ml) or
monkeys (2.0 ␮g 䡠 min 䡠 kg/mg 䡠 ml), and, consequently, oral bioavailabilities were 27, 17, and 4%, respectively. This suggested that 1 was
either not well absorbed or underwent extraction in the gut. Low oral
bioavailability was a major issue in rats and monkeys but not in dogs;
therefore, studies focused on identifying the factors that contributed to
low oral bioavailability in these two species.
To verify that the generally low oral bioavailability was due to poor
oral absorption, the absorption and excretion of 1 were investigated in
rats. 3H-labeled 1 was administered orally and intravenously to bile
duct-cannulated rats, and the excretion of radioactivity in bile, urine,
and feces was monitored. Five days after i.v. dosing, 63 ⫾ 13% of the
administered radioactivity was recovered in bile and 11 ⫾ 2% in
urine, indicating that biliary excretion of compound-related material
was a primary route of elimination for this compound. The excretion
of 9 ⫾ 3% of the i.v. dose into the feces of bile duct-cannulated rats
suggested that 1 underwent efflux from the gastrointestinal tract.
Since 1 has been identified as a substrate for P-glycoprotein (Tang et
al., 2002), it is likely that efflux into the gastrointestinal tract was
mediated by this transporter.
After oral dosing, 22 ⫾ 9% of the radioactivity was recovered in
bile and 6 ⫾ 3% in urine, indicating that only 28% of the administered
radioactivity was absorbed. The remainder of the oral dose (75 ⫾
20%) was recovered in feces. These results supported the conclusion
that 1 was not well absorbed and suggested that P-glycoproteinmediated efflux might contribute, at least in part, to the low oral
bioavailability. The metabolite profile was determined in bile collected from rats dosed orally with 1. The metabolites that were
identified were primarily hepatic oxidation products and included the
pyridine N-oxide derivative, a primary amine resulting from N-dealkylation and loss of the pyridinyl-2-hydroxyethyl group; a carboxylic
acid derived from N-dealkylation and loss of pyridinyl-2-hydroxyethylamine; and the corresponding taurine and isoethionic acid conjugates (Tang et al., 2002).
To evaluate whether gut extraction contributed to the poor absorption
of 1, it was necessary to determine hepatic extraction. Since F ⫽ fh 䡠 fg 䡠
fabs, and F and fabs have been determined experimentally, at least in the
rat, determination of fh would allow us to assess fg indirectly (fg ⫽ F/(fh 䡠
fabs)). The hepatic first-pass effect was evaluated in rats and monkeys
following portal or systemic infusion of drug. The increase in plasma
TABLE 3
Pharmacokinetic data in monkeys dosed with 2 and 3
Compound
2
3
i.v. Dose
mg/kg
3 (N ⫽ 2)
3 (N ⫽ 2)
Oral Dose
mg/kg
2
3
10 (N ⫽ 2)
10 (N ⫽ 2)
AUCprodrug
AUCl
␮g 䡠 min 䡠 kg/mg 䡠 ml
8
0.4
16
0.3
AUCprodrug
AUC1
Clp
t1/2
Vdss
ml/min/kg
124
64
h
3.3
2.4
l/kg
22.3
3.5
Cmax
Tmax
F
␮g 䡠 min 䡠 kg/mg 䡠 ml
␮g/ml
h
%
0.5
8.9
0.009
0.178
2–4
4
6
56
1.1
1.8
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 11, 2017
1 were examined in two monkeys dosed intravenously at 3 mg/kg.
Chemical structures and pharmacokinetic parameters are shown in
Fig. 1 and Table 3, respectively. Conversion of these potential prodrugs to 1 was evaluated by comparing the dose-normalized AUC
values of 1 obtained after dosing 1 (16 ␮g 䡠 min 䡠 kg/mg 䡠 ml), 2 (0.4
␮g 䡠 min 䡠 kg/mg 䡠 ml), or 3 (0.3 ␮g 䡠 min 䡠 kg/mg 䡠 ml) i.v. at 3 mg/kg.
Conversion to 1 was low (⬍3%) with both derivatives. The Clp, Vdss
and t1/2 values were 124 ml/min/kg, 22.3 l/kg, and 3.3 h for 2 and 64
ml/min/kg, 3.5 l/kg, and 2.4 h for 3. After oral administration of 2 and
3 at 10 mg/kg, dose-normalized plasma AUC values of 1 (1.1 and 1.8
␮g 䡠 min 䡠 kg/mg 䡠 ml, respectively) were not higher than observed
with 1 (2.7 ␮g 䡠 min 䡠 kg/mg 䡠 ml) administered orally at the same
dose; therefore, neither entity was an effective prodrug. Dose-normalized plasma AUC values for intact 2 (0.5 ␮g 䡠 min 䡠 kg/mg 䡠 ml) and
3 (8.9 ␮g 䡠 min 䡠 kg/mg 䡠 ml), however, differed substantially, and oral
bioavailabilities were estimated at 6 and 56%, respectively.
The pharmacokinetics of 4 were evaluated in rats, and data are
shown in Table 4. The Clp, Vdss, and t1/2 values were 20.1 ml/min/kg,
5.8 l/kg, and 5.7 h for 4 compared with 62 ml/min/kg, 23 l/kg, and 8 h
for 1. After oral dosing at 10 mg/kg, Cmax (395 ng/ml) was 3-fold
higher than 1 (126 ng/ml), and the oral bioavailability (30%) was
approximately twice that of 1 (17%). The hepatic extraction of 4 also
was evaluated in rats by comparing systemic AUC values following
femoral and portal vein infusions. These data also are presented in
Table 4. The hepatic extraction of 4 was 33%, and the plasma
clearance of 4 was not dependent on the vehicle in which the drug was
administered (ethanol/PEG 400/saline versus plasma).
The pharmacokinetics of 4 were evaluated subsequently in rhesus
monkeys. Data are presented in Table 4. The Clp, Vdss, and t1/2 values
were 8 ml/min/kg, 3 l/kg, and 7 h for 4 compared with 19 ml/min/kg,
6 l/kg, and 9 h for 1. After oral dosing at 10 mg/kg, Cmax (374 ng/ml)
was approximately 12-fold higher than 1 (32 ng/ml), and the oral
bioavailability (23%) was 4-fold higher than that of 1 (⬃5%).
775
776
STEARNS ET AL.
TABLE 4
Pharmacokinetic data in rats and monkeys dosed with 4
Rat
Monkey
Rats
i.v. Dose
AUCnorm
Clp
t1/2
Vdss
mg/kg
␮g 䡠 min 䡠 kg/mg 䡠 ml
ml/min/kg
h
l/kg
0.2 (N ⫽ 3)a
3 (N ⫽ 3)
3 (N ⫽ 2)
47.8 ⫾ 19.3b
50.6 ⫾ 8.0
125.8 ⫾ 31.7
20.9 ⫾ 8.4
20.1 ⫾ 3.4
8.2 ⫾ 2.1
5.7 ⫾ 0.2
7.2 ⫾ 1.3
5.8 ⫾ 0.9
3.3 ⫾ 0.7
p.v. Dose
AUCnorm
Eh
Tmax
F
mg/kg
␮g 䡠 min 䡠 kg/mg 䡠 ml
%
0.2 (N ⫽ 3)a
32.2 ⫾ 4.7
33
Oral Dose
AUCnorm
Cmax
mg/kg
␮g 䡠 min 䡠 kg/mg 䡠 ml
␮g/ml
h
%
10 (N ⫽ 3)
10 (N ⫽ 2)
15.3 ⫾ 3.6
28.7 ⫾ 1.3
0.395 ⫾ 0.127
0.374 ⫾ 0.074
5⫾1
6⫾3
30 ⫾ 7
23
Rat
Monkey
AUC values was proportional to the systemic dose over the concentration
range from 0.2(0.3) to 3 mg/kg in both rats and monkeys. Hepatic
extraction (Eh) was calculated by comparison of the plasma AUC values
obtained by each route of administration. On this basis, the hepatic
extraction of 1 in the rat and monkey was 30 and 47%, respectively.
Results from the study in rats suggested that hepatic clearance was the
primary contributor to the first-pass effect because fg ⫽ (F/(fh 䡠 fabs) ⫽
87%) was high, and consequently, extraction by the gut was low. Therefore, the primary factor that limits the oral bioavailability of 1 in the rat
is poor oral absorption. By using the same reasoning, the product fg 䡠 fabs
could be estimated in the monkey, since both the oral bioavailability (4%)
and hepatic extraction (47%) have been determined experimentally; thus,
in the monkey, fg 䡠 fabs ⫽ F/fh ⫽ 8%, which is considerably lower than
the 24% (fg 䡠 fabs) estimated in the rat. Two factors can explain this result;
either 1) oral absorption in monkeys is less than in rats or 2) the
contribution from intestinal first-pass metabolism is greater in monkeys.
Studies were not performed to substantiate either hypothesis.
Attempts were made to improve the oral bioavailability of 1 by
synthesizing potential prodrugs. To determine whether N-alkylation of 1
was a viable prodrug strategy, the pharmacokinetics of the N-ethyl (2)
and morpholine (3) derivatives of 1 were examined in i.v.-dosed monkeys. An estimate of the extent of conversion of the prodrugs to 1 was
made by comparing plasma AUC values of 1 after i.v. administration of
either 1, 2, or 3. Conversion to 1 was low (⬍3%) with both derivatives.
After oral administration of 2 and 3, the plasma AUC value of 1 (1.1 and
1.8 ␮g 䡠 min 䡠 kg/mg 䡠 ml, respectively) was not higher than that observed
when 1 (2.7 ␮g 䡠 min 䡠 kg/mg 䡠 ml) was given at the same dose; therefore,
neither derivative was an effective prodrug. The oral bioavailability of
intact 3, however, was high (56%). The increased oral bioavailability of
3 in comparison to 1 was assumed to be due to improved absorption, a
result that demonstrated the potential for improving the oral bioavailability of this class of compounds.
Consideration of the prodrug pharmacokinetic data led to the hypothesis that increased absorption might be achieved by decreasing
the number of hydrogen bonding sites within the ethanolamine functional group. One strategy was to change the linkage to the pyridine
moiety from the 3- to the 2-position so that the pyridyl-nitrogen atom
was positioned to hydrogen bond with the ethanolamine hydroxyl
group as shown in Fig. 2. This analog (4) was synthesized, and its
pharmacokinetics were evaluated in rats and monkeys. Plasma clearance values were lower, and oral bioavailability was higher in both
species. To determine whether the increased oral bioavailability of 4
was due to improved absorption or decreased clearance, hepatic
extraction of 4 was evaluated in rats. The hepatic extraction of 4
(33%) was similar to that of 1 (30%). Having obtained values for F
(30%) and fh (1 - Eh ⫽ 67%) in rats, fg 䡠 fabs could be estimated as
described previously. This estimate (F/fh ⫽ fg 䡠 fabs ⫽ 45%) suggested
that, at least in rats, the absorption of 4 was approximately 2-fold
greater than that of 1 (24%).
In conclusion, the oral bioavailability of 1, a ␤3-adrenergic receptor
agonist, is limited by poor oral absorption in rats and monkeys. The
oral bioavailability of this compound was improved by making a
slight modification to its structure, resulting in the 2-pyridyl analog 4,
that was designed to increase intramolecular hydrogen bonding interactions and thereby minimize intermolecular interactions that may
limit the oral absorption of this compound class.2
References
Arch JRS, Ainsworth AT, Cawthorne MA, Piercy V, Sennit MV, Thody VE, Wilson C, and
Wilson S (1984) Atypical ␤-adrenoreceptor on brown adipocytes a target for anti-obesity
drugs. Nature (Lond) 309:163–165.
FIG. 2. Comparison of internal hydrogen bonding in 1 and 4.
2
Subsequently, the disposition of 1 and 4 were evaluated in brain and plasma
after i.v. dosing in CF1 mdr1a ⫹/⫹ and ⫺/⫺ mice (data not shown). Plasma
concentrations of both drugs were similar in each strain of mice, but the concentrations in brain were 5- to 7-fold higher in the ⫺/⫺ strain. This data suggested
that both 1 and 4 were P-glycoprotein substrates and that the increased oral
absorption of 4 was not due to a difference in P-glycoprotein substrate selectivity
but rather to an improvement in the intrinsic membrane permeability of this
analogue.
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 11, 2017
p.v., portal vein.
a
Dose administered by infusion.
b
Values represent mean ⫾ S.D.
PHARMACOKINETICS OF ␤3-ADRENERGIC RECEPTOR AGONISTS
containing a substituted indoline-5-sulfonamide pharmacophore. Bioorg Med Chem Lett
9:1869 –1874.
Mathvink RJ, Tolman SJ, Chitty D, Candelore MR, Cascieri MA, Colwell LF Jr, Deng L, Feeney
WP, Forrest MJ, Hom GJ, et al. (2000a) Potent, selective 3-pyridylethanolamine ␤3-adrenergic
receptor agonists possessing a thiazole benzenesulfonamide pharmacophore. Bioorg Med
Chem Lett 10:1971–1973.
Mathvink RJ, Tolman SJ, Chitty D, Candelore MR, Cascieri MA, Colwell LF Jr, Deng L, Feeney
WP, Forrest MJ, Hom GJ, et al. (2000b) Discovery of a potent, orally bioavailable ␤3adrenergic receptor agonist, (R)-N-[4-[2-[[2-hydroxy-2-(3-pyridinyl)-ethyl]amino]ethyl]phenyl-4-[4-[4-(trifluoromethyl)phenyl]thiazol-2-yl]benzenesulfonamide. J Med Chem 43:3832–
3836.
Naylor EM, Colandrea VJ, Candelore MR, Cascieri MA, Colwell LF, Deng L, Feeney WP,
Forrest MJ, Hom GJ, MacIntyre DE, et al. (1998) 3-pyridylethanolamine: potent and selective
human ␤3-adrenergic receptor agonists. Bioorg Med Chem Lett 8:3087–3092.
Naylor EM, Parmee ER, Colandrea VJ, Perkins L, Brockunier L, Candelore MR, Cascieri MA,
Colwell LF, Deng L, Feeney WP, et al. (1999) Human ␤3-adrenergic receptor agonists
containing imidazolidinone and imidazolone benzenesulfonamides. Bioorg Med Chem Lett
9:755–758.
Ok HO, Reigle LB, Candelore MR, Cascieri MA, Colwell LF, Deng L, Feeney WP, Forrest MJ,
Hom GJ, MacIntyre DE, et al. (2000) Substituted oxazole benzenesulfonamides as potent
human ␤3-adrenergic receptor agonists. Bioorg Med Chem Lett 10:1531–1534.
Parmee ER, Naylor EM, Perkins L, Colandrea VJ, Ok HO, Candelore MR, Cascieri MA, Deng
L, Feeney WP, Forrest MJ, et al. (1999) Human ␤3-adrenergic receptor agonists containing
cyclic ureidobenzenesulfonamides. Bioorg Med Chem Lett 9:749 –754.
Pi-Sunyer FX (1993) Medical hazards of obesity. Ann Intern Med 119:655– 660.
Shih TL, Candelore MR, Cascieri MA, Chiu SHL, Colwell LF, Deng L, Feeney WP, Forrest MJ,
Hom GJ, MacIntyre DE, et al. (1999) L-770,644: a potent and selective human ␤3-adrenergic
receptor agonist with improved oral bioavailability. Bioorg Med Chem Lett 9:1251–1254.
Tang W, Stearns RA, Miller RR, Ngui JS, Mathvink RJ, Weber AE, Kwei GY, Strauss JR,
Keohane CA, Doss GA, et al. (2002) Metabolism of a thiazole benzenesulfonamide derivative,
a potent and selective agonist of the human ␤3 adrenergic receptor, in rats: identification of a
novel isethionic acid conjugate. Drug Metab Dispos 30:778 –787.
Weber AE (1998) ␤3-adrenergic receptor agonists for the treatment of obesity. Annu Rep Med
Chem 33:193–202.
Weber AE, Ok HO, Alvaro RF, Candelore MR, Cascieri MA, Chiu SHL, Deng L, Forrest MJ,
Hom GJ, Hutchins JE, et al. (1998) 3-Pyridyloxypropanolamine agonists of the ␤3-adrenergic
receptor with improved pharmacokinetic properties. Bioorg Med Chem Lett 8:2111–2116.
Wickelgren I (1998) Obesity: how big a problem. Science (Wash DC) 280:1364 –1367.
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 11, 2017
Arch JRS, Cawthorne MA, Coney KA, Gusterson BA, Piercy V, Sennitt MV, Smith SA, Wallace
J, and Wilson S (1990) ␤-adrenoreceptor-mediated control of thermogenesis, body composition and glucose homeostasis, in Obesity and Cachexia (Rothwell NJ and Stock MJ eds) pp
241–268, John Wiley & Sons, Inc, New York.
Arch JRS and Wilson S (1996) Prospects for ␤3-adrenoreceptor agonists in the treatment of
obesity and diabetes. Int J Obes 20:191–199.
Biftu T, Feng DD, Liang G-B, Kuo H, Qian X, Naylor EM, Colandrea VJ, Candelore MR,
Cascieri MA, Colwell LF Jr, et al. (2000) Synthesis and SAR of benzyl and phenoxymethylene
oxadiazole benzenesulfonamides as selective ␤3-adrenergic receptor agonist antiobesity
agents. Bioorg Med Chem Lett 10:1431–1434.
Brockunier LL, Parmee ER, Ok HO, Candelore MR, Cascieri MA, Colwell LF Jr, Deng L,
Feeney WP, Forrest MJ, Hom GJ, et al. (2000) Human ␤3-adrenergic receptor agonists
containing 1,2,3-triazole-substituted benzenesulfonamides. Bioorg Med Chem Lett 10:2111–
2114.
Dow RL (1997) ␤3-adrenergic agonists: potential therapeutics for obesity. Exp Opin Invest Drugs
6:1811–1825.
Feng DD, Biftu T, Candelore MR, Cascieri MA, Colwell LF Jr, Deng L, Feeney WP, Forrest MJ,
Hom GJ, MacIntyre DE, et al. (2000) Discovery of an orally bioavailable alkyl oxadiazole
␤3-adrenergic receptor agonist. Bioorg Med Chem Lett 10:1427–1429.
Fisher MH, Amend AH, Bach TJ, Barker JM, Brady EJ, Candelore MR, Carroll D, Cascieri MA,
Chiu SHL, Deng L, et al. (1998) A selective human ␤3-adrenergic receptor agonist increases
metabolic rate in rhesus monkeys. J Clin Invest 101:2387–2393.
Gibaldi M and Perrier D (1982) Nonlinear pharmacokinetics, in Pharmacokinetics, pp 271–318,
Marcel Dekker Inc., New York.
Hom GJ, Forrest MJ, Bach TJ, Brady E, Candelore MR, Cascieri MA, Fletcher DJ, Fisher MH,
Iliff SA, Mathvink R, et al. (2001) ␤3-adrenoreceptor agonist-induced increases in lipolysis,
metabolic rate, facial flushing and reflex tachycardia in anesthetized rhesus monkeys. J Pharmacol Exp Ther 297:299 –307.
Kordik CP and Reitz AB (1999) Pharmacological treatment of obesity: therapeutic strategies.
J Med Chem 42:181–201.
Kuczmarski RJ, Flegal KM, Campbell SM, and Johnson CL (1994) Increasing prevalence of
overweight among US adults: the National Health and Nutrition Examination Surveys 1960 –
1991. J Am Med Assoc 270:205–211.
Lowell BB and Flier JS (1997) Brown adipose tissue, ␤3-adrenergic receptors and obesity. Annu
Rev Med 48:307–316.
Mathvink RJ, Barritta AM, Candelore MR, Cascieri MA, Deng L, Tota L, Strader CD, Wyvratt
MJ, Fisher MH, and Weber AE (1999) Potent, selective human ␤3-adrenergic receptor agonists
777