2,3-Diaryl-5-anilino[1,2,4]thiadiazoles as melanocortin MC4 receptor agonists and their effects on feeding behavior in rats

Article abstract of DOI:10.1016/S0968-0896(02)00428-5

The melanocortin-4 receptor (MC4) modulates physiological functions such as feeding behavior, nerve regeneration, and drug addiction. Using a high throughput screen based on ¹²⁵I-NDP-MSH binding to the human MC4 receptor, we discovered 2,3-diar

Full text of DOI:10.1016/S0968-0896(02)00428-5

Bioorganic & Medicinal Chemistry 11 (2003) 185–192
2,3-Diaryl-5-anilino[1,2,4]thiadiazoles as Melanocortin MC4
Receptor Agonists and Their Effects on Feeding Behavior in Rats
Kevin Pan,y Malcolm K. Scott, Daniel H. S. Lee,{ Louis J. Fitzpatrick,
x
Jeffery J. Crooke, RalphA. Rivero, Daniel I. Rosenthal,
Anil H. Vaidya, Boyu Zhao and Allen B. Reitz*
Drug Discovery Division, Johnson & Johnson Pharmaceutical Research and Development, Welsh and McKean Rds.,
PO Box 776, Spring House, PA 19477, USA
Received 29 May 2002; accepted 16 August 2002
Abstract—The melanocortin-4 receptor (MC4) modulates physiological functions such as feeding behavior, nerve regeneration, and
drug addiction. Using a high throughput screen based on 125I-NDP-MSH binding to the human MC4 receptor, we discovered
2,3-diaryl-5-anilino[1,2,4]thiadiazoles 3 as potent and selective MC4 receptor agonists. Through SAR development on the three
attached aryl rings, we improved the binding affinity from 174 nM to 4.4 nM IC50. When delivered intraperitoneally, compounds
3a, 3b, and 3c induced significant inhibition of food intake in a fasting-induced feeding model in rats. When delivered orally, these
compounds lost activity, mainly due to rapid metabolism to inactive imidoylthiourea reduction products.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
All five MC receptors exhibit diverse physiological
functions mainly through cAMP stimulation and
downstream signal transduction. Expressed in melano-
cytes of the dermis, the MC1 receptor plays a major role
in controlling the formation of the pigment melanin and
animal coat coloration. The MC2 receptor is highly
expressed in the adrenal cortex and adipose tissues, and
is believed to be responsible for the ACTH-mediated
control of glucocorticoid and mineralcorticoid produc-
tion. The MC3 receptor is expressed in the CNS, pla-
centa, gut and heart. Although the functions of the
MC3 receptor are not well understood, MSH and
ACTH peptides are believed to induce sexual behavior
in animals via the stimulation of the MC3 receptor at a
hypothalamic site. The MC4 receptor is primarily
expressed in the brain and plays an important role in
regulating feeding behavior.5 The MC4 receptor is also
involved in drug addiction,6 nerve regeneration,7,8 and
the regulation of the cardiovascular system together
withnon-melanocortin receptors.
Five melanocortin G-protein receptor subtypes (MC1–
MC5) have been cloned and characterized.1 Melanocyte-
stimulating hormone (MSH) and adrenocorticotropic
hormone (ACTH), which are produced from the pro-
teolytic cleavage of a 31–36 KD glycosylated protein
called pro-opiomelanocortin (POMC), are two natural
agonists of MC receptors. a-MSH contains the first
13N-terminal amino acids of ACTH. Homologous to
a-MSH are two additional peptides, b-MSH and g-MSH,
contained in the POMC sequence. All of these natural
ligands bind to MC receptors withdiffering affinities, with
the exception of MC2 which is activated by only ACTH.2
MC receptors are among the very few receptors that have
bothnatural agonists and antagonists. For example, the
agouti signaling protein (ASIP) is an antagonist of MC1
and MC4 receptors.3 Through a homology search of the
genes for proteins related to ASIP, another protein,
agouti-gene related protein (AGRP), was cloned and
found to be an antagonist of MC3 and MC4 receptors.4
The role of the MC4 receptor in controlling feeding
behavior and body weight is well supported by several
studies. For example, direct injection of a-MSH or
ACTH-(1–24) into the hypothalamus caused a sig-
nificant reduction of food intake in rats.9 Similar effects
were also observed when the unnatural agonistic peptide
*Corresponding author. Tel.:+1-215-628-5615; fax:+1-215-628-4985;
e-mail: areitz@prdus.jnj.com
yPresent affiliation: Shanghai Hutchinson Pharmaceuticals.
{Present affiliation: Biogen Inc., 14, Cambridge Center, Cambridge, MA.
xPresent affiliation: GlaxoSmithKline, King of Prussia, Pennsylvania.
0968-0896/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00428-5

186
K. Pan et al. / Bioorg. Med. Chem. 11 (2003) 185–192
Figure 1.
MTII was administered to C57BL/6J mice, ob/ob mice,
10
penetrating the brain–blood barrier. The intense interest
in finding small, non-peptide molecules is represented
by isoquinoline analogues 1 and 2 (Fig. 1). Withlow
micromolar affinities to the MC1 and MC4 receptors, 1
was reported to reduce dermal inflammation induced by
arachidonic acid, food consumption, and body
weight.20 Isoquinoline analogues witha 4,4-disub-
stituted piperidine moiety, suchas 2, are hMC4 receptor
agonists withsubnanomolar potency and in vivo
erectogenic effects in rodents.21
AY mice, and mice injected withneuropeptide Y. The
natural antagonist agouti is overexpressed in the CNS
of AY agouti mice that developed extreme obesity
because of overeating.11 Intracerebroventricular (icv)
administration of the MC4 receptor antagonist
SHU9119, a mimic of the MC-4 antagonist agouti,
resulted in stimulated nocturnal feeding in C57BL/6J
mice.12 Additionally, icv administration of the MC4
receptor antagonist HS014 also increased food intake in
free-feeding rats.10 Importantly, MC-4 receptor knock-
out mice developed obesity syndrome resembling some
of the characteristic features of the agouti obesity syn-
drome,13 and several MC4 receptor mutations were dis-
covered in the human population associated with
obesity.14 Thus, the MC4 receptor is an important target
for discovery of drugs to treating obesity, which repre-
sents a large and presently poorly treated medical need.15
Our interest in the area of selective MC4 agonists
prompted us to carry out MC4r high-throughput
screening (HTS) of our corporate compound collec-
tion, led to the discovery of 2,3-diaryl-5-anilino-
[1,2,4]thiadiazolium bromide 3a as an MC4 receptor
agonist. After structure–activity relationship (SAR)
studies around the central thiadiazolium scaffold and
improving the binding affinity from 174 to 4.4 nM IC50,
representative compounds were tested in a fasting-induced
rat-feeding model and inhibited food intake.
Peptide and peptidomimetic agonists for MC1–MC5
have been reported containing the critical four amino
acid pharmacophore His-Phe-Arg-Trp found in
MSH.16ꢀ18 NDP-MSH ([Nle4, D-Phe7]-a-MSH) and its
radiolabeled analogues have been used extensively to
measure the MC receptor binding affinities of non-
labeled compounds and to probe the pharmacology of
the MC receptors. Clinical studies of a potent non-
selective cyclic heptapeptide agonist at MC receptors,
melanotan-II, showed it to be effective in causing erec-
Synthetic Chemistry
The synthesis of [1,2,4]-thiadiazoles has been reported
extensively.22 In Scheme 1, substituted anilines 4 were
reacted withbenzonitriles 5 and AlCl3 at an elevated
temperature to form N-arylbenzamidines 6.23 In the
presence of a methoxy group, a strong base such as
sodium amide was used in the reaction to make 6.24
N-Arylbenzamidines 6 were reacted withsubstituted
19
tions in men withpsychogenic erectile dysfunction.
However, MSH-like peptide analogues have limited
utility due to their poor bioavailability and difficulties in
Scheme 1. (a) AlCl3/neat, 170–220 ꢁC; (b) NaNH2, toluene, reflux; (c) R3NCS, 55 ꢁC, 1,2-dichloroethane; (d) Br2, chloroform; (e) N-chloro-
succinimide, chloroform; (f) methyl triflate, dichloromethane.

K. Pan et al. / Bioorg. Med. Chem. 11 (2003) 185–192
187
Scheme 2. (a) Dithioerythritol (DTE), dichloromethane; (b) DMSO, 65 ꢁC.
isothiocyanates to form the imidoylthioureas 7,25 which
were oxidized by bromine to yield thiadiazolium salts 3.
The free bases of 3, [1,2,4]-thiadiazolines 8, were pre-
pared by treating 7 with N-chlorosuccinimide (NCS)
followed by extraction from aqueous sodium bicarbo-
nate. 2,3-Diaryl-5-(N-methyl-N-arylamino)-[1,2,4]-thia-
from human Bowes melanoma cells that overexpress
the MC4 receptor were immobilized on wheat germ
agglutinin-conjugated scintillation proximity assay
(SPA) beads, and binding of 125I-NDP-MSH was
monitored in the presence and absence of test inhibi-
tors.28 The data are reported as IC50 values in Figures
2 and 3. The in vivo effects of compounds 3a–3c were
examined for their ability to inhibit food-intake in
fasted rats (Tables 1 and 2).29 The agonist character of
the interaction of compound 3a (IC50=174 nM) at the
MC4 receptor was confirmed by increased 35S-GTPgS
binding.30 The binding affinity of 3a to the MC3
diazolium triflate
9
was prepared by selective
N-methylation of compound 8 by methyl triflate.
2-(4-Methylphenyl)-3-phenyl-5-(4-methoxyphenylamino)-
[1,2,4]-thiadiazolium bromide 3a was rapidly reduced by
dithioerythritol (DTE) back to the corresponding
thiourea (7a, Scheme 2).26 Compound 3a also under-
went a thermal rearrangement to 2-amidinobenzothia-
zole 10 (biologically inactive at the MC4 receptor), a
general reaction that we have studied in some detail.27
receptor was also determined, withan IC
2.15 mM.
value of
50
Results and Discussions
Biological Evaluation
It was reported that [1,2,4]-thiadiazoles were reduced to
imidoylthioureas by H2S and NH3.31 In our hands, 3a
was rapidly reduced to its corresponding precursor imi-
The affinities of compounds 3, 8, and 9 for the MC4
receptor were evaluated in binding assays. Membranes
Figure 2. Binding affinities of 2-aryl-3-phenyl-5-anilino[1,2,4]thiadizolium bromides 3 to the MC4 receptor—Part I.

188
K. Pan et al. / Bioorg. Med. Chem. 11 (2003) 185–192
doylthiourea 7a by DTE (Scheme 2). We had suspected
that 3a might also react with a thiol group on cysteine
residues or oxidize two thiol groups to create a disulfide
bond, to promote irreversible binding to the receptor.
This possibility was discounted after we found that the
receptor remained active in binding to 125I-NDP-MSH
after 3a was washed off.
833 nM IC50. Th e pKa of 3a was determined to be
5.67.33
We then modified aromatic substitution on the 2-phenyl
ring (R1) and the 5-anilino ring (R3) simultaneously by
preparing an array of compounds 3-I (Fig. 1). In addition
to the unsubstituted phenyl ring, methoxy, methyl, and
chloro groups were placed at 2- and 4-positions sepa-
rately. For R1 substitution, 2-methoxy or no substituent
was generally better than the others, and 2- and
4-methyl as well as 2- and 4-chloro substitution was not
favored. R3 substitution was not as discriminating as
that at R1. Electron-richsubstitution suchas methyl or
methoxy groups generally showed better activities. A
combination of 2-methoxy on R1 and 4-methyl on R3
As the first step of SAR exploration, the proton on
the 5-anilino nitrogen was replaced by a methyl
group (9). Compared to 3a, the activity was essen-
tially eliminated from a 174 nM to 30 mM IC50.
This observation strongly suggests that the proton
on the anilino nitrogen is required for binding. It is
possible that the positive charge on the thiadiazolium
ring may also be important in binding. The core tetra-
peptide (His-Phe-Arg-Trp) in the natural ligand MSH
contains a positive charge on the guanidine moiety.
The charged arg residue is thought to play an impor-
tant role in binding to Glu-94, Asp-121 and Asp-117 in
the MC1 receptor.32 When 3a was converted to its free
base 8, the apparent activity decreased from a 174 to
resulted in 3b withan IC value of 22 nM at MC4, an
50
eight-fold improvement of the binding affinity com-
pared to 3a.
After selecting 2-(2-methoxyphenyl) substitution, we
systematically altered the 3-phenyl (R2) and 5-anilino
rings (R3) to make compounds 3-II (Fig. 2). When R2
Figure 3. Binding affinities of 2-aryl-3-phenyl-5-anilino[1,2,4]thiadizolium bromides 3 to the MC4 receptor—Part II.

K. Pan et al. / Bioorg. Med. Chem. 11 (2003) 185–192
189
Table 1. Percentage change of food consumption versus baseline day
in rats (one-sample t-test)
Table 2. Percentage change of food consumption versus drug day
(vehicle) in rats (Mann–Whitney t-test)
Compd no.
Dose No. of
(mg/kg)/route animals
2 h 4 h 6 h2–6 h
Compd no.
Dose No. of
(mg/kg)/route animals
2 h 4 h 6 h2–6 h
PEG–200
0/IPa
0/POb
10/IP
30/PO
3/IP
8
8
15
8
8
8
8
8
8
5
0.0
11.2
0.0
0.0
—
3.8
1.1
6.9
ꢀ9.0
ꢀ13.8
0.0
ꢀ50.0c
ꢀ68.8c
29.7c
ꢀ1.9
29.1c
0.9
3a
3a
3b
3b
3b
3b
3b
3c
3c
10/IPa
30/POb
3/IP
15
8
8
8
8
8
8
5
7
ꢀ22.1c ꢀ12.3 ꢀ22.9c
ꢀ21.2c ꢀ11.1 ꢀ14.4c
ꢀ37.8c ꢀ37.4c ꢀ37.1c
ꢀ20.6c ꢀ45.2c ꢀ50.0c
ꢀ23.3c
ꢀ6.2
MCT
3a
3a
3b
3b
3b
3b
3b
3c
ꢀ39.0c
ꢀ42.7c
ꢀ36.2c
ꢀ71.8c
31.5c
39.7c
52.1c
0.9
1.3
2.8
0.7
1/IP
ꢀ37.8c ꢀ42.1c ꢀ44.4c
ꢀ33.3c ꢀ42.4c ꢀ51.3c
30/PO
10/PO
3/PO
10/IP
3/IP
ꢀ23.6c ꢀ12.7c
1.2
12.4c
18.5c
ꢀ4.2
ꢀ6.1
1/IP
ꢀ10.1
ꢀ9.0
ꢀ11.1
0.0
ꢀ0.9
30/PO
10/PO
3/PO
10/IP
3/IP
ꢀ10.5
ꢀ9.6
9.3
0.0
13.6
ꢀ4.2
ꢀ6.1
6.8
2.8
2.7
5.4
7.7
ꢀ2.4
ꢀ11.1
0.0
ꢀ4.0
ꢀ9.9
2.8
ꢀ4.0
aIntraperitoneal administration.
bOral administration.
cP value <0.05.
3c
7
ꢀ9.9
aIntraperitoneal administration.
bOral administration.
cP value<0.05.
When 3a–3c were administered orally, they did not
demonstrate significant inhibition on food intake. The
in vitro metabolism of 3a was evaluated using human
and rat hepatic S9 fractions (60 min incubation). In this
study, only 5% of 3a remained in the human S9 frac-
tion, and 9% in the rat S9 fraction. The major meta-
bolite was imidoylthiourea 7a (59%, human S9; 50%,
rat S9). Compound 7a was found to be inactive in the
receptor binding assay (>10000 nM IC50) and was also
inactive in the rat-feeding model (ip dosing).
was substituted by a 2-methoxy or 4-methyl group, the
binding affinities were generally reduced except witha 2-
methoxy group on R2 and a 4-methyl group on R3
resulting in 3c withan IC
of 4.4 nM, a 5-fold
50
improvement from 3b. Indeed, a 4-methyl group was the
most optimal substitution on R3, and electron-deficient
3-pyridyl or chloro substituted phenyl rings were not
generally favored.
We also replaced the 5-anilino group of 3 by 5-benzyl-
amino moieties. Compared to their direct anilino analo-
gues, 2,3-diphenyl-5-benzylaminothiadiazolium bromides
3-III where the benzyl group was optionally substituted
with methoxy, methyl, or chloro groups at the 2- or
4-positions were inactive at MC4 (800 to 3000 nM
IC50s, data not shown).
Conclusions
The MC4 receptor agonistic activity of the title com-
pounds provides an additional chemical series of small,
non-peptidic molecules withactivity in a feeding model
in rodents. We demonstrated that the 2,3-diaryl-5-ani-
lino-[1,2,4]-thiadiazoles can bind to the MC-4 receptor
withlow nanomolar affinity. Importantly, these com-
pounds inhibit food intake in rats when administered
intraperitoneally. Although the rapid metabolism of
these compounds precludes their immediate use as oral
drugs for the treatment of obesity, they have provided
useful information for further SAR development and
research.
Compounds 3b and 3c were also determined to be ago-
nists in the functional assays, and were tested along with
3a in a fasting-induced rat-feeding model. Table 1
includes data based on the change in food consumed
between the baseline relative to the drug day in indivi-
dual rats where each rat was used as its own control. A
one-sample t-test analysis was used to determine whe-
ther the change in food consumed was statistically sig-
nificant. Table 2 contains data based on the change in
food consumed from vehicle in separate rats on the drug
day. When compounds were delivered by an intraper-
itoneal route of administration, except for 3c, they
showed significant effects in reducing food intake in
animals. The inhibitory effects on the cumulative food
intake after 2, 4, and 6 hperiods increased withthe time
after compound administration, and effects on the
average food intake for the 2–6 h time block were gen-
erally greater than that at the 6 h period. This obser-
vation indicates that maximum effectiveness appeared at
least 2 hafter administration. When compound 3a and
3b were administered ip at 10 mg/kg, they were equally
effective on the inhibition of the cumulative food
intake, and 3b appeared to be more active than 3a
when comparing average food intake for the 2–6 h
block. It is interesting to see that compound 3c, which is
more active in binding, did not show significant effects
on food intake, suggesting that 3c may have some other
general properties which prevent in vivo activity.
Experimental
General. 1H NMR spectra were obtained on a 300-MHz
Bruker Avance 300 NMR spectrometer withMe Si as
4
an internal standard. The mass spectrum of each com-
pound was generated from a MicroMass Platform LC
electrospray mass spectrometer or chemical ionization on
a Hewlett-Packard 5989A mass spectrometer. Chroma-
tographic separations were performed using a Biotage
Flash 40 flash chromatography system. Most reagents
and solvents were purchased and used without further
purification.
N-(4-Methylphenyl)benzamidine (6a). Benzonitrile (10.0
g, 97.0 mmol) was slowly added to dry AlCl3 (14.23 g,
106.7 mmol) in a round-bottom-flask. After the hot
flask was cooled to rt, 4-toluidine (10.4 g, 97.0 mmol)

190
K. Pan et al. / Bioorg. Med. Chem. 11 (2003) 185–192
was added to the flask, and the reaction was heated at
160 ꢁC for 45 min. The hot mixture was poured into a
solution of 3.2 mL of concentrated HCl solution in 240
mL of mixed water and ice. Activated carbon was added
and the reaction mixture was filtered through a Celite
pad after stirring for 20 min. The filtrate was mixed with
a solution of 32 g NaOH solids dissolved in 200 mL
water. The aqueous layer was extracted with 3ꢂ150 mL
of chloroform. The combined organic layer was dried
over anhydrous MgSO4, and evaporated. The resulting
solid was washed with hexane and dried over the
vacuum to yield the product 6a as a pale white solid
Compounds 3b and 3c were prepared in a similar fash-
ion to give yellow solids.
3b (60% yield). 1H NMR (300 MHz, CDCl3) d 12.44 (s,
1H), 7.60 (d, 2H), 7.36 (t, 1H), 7.12–7.32 (m, 6H), 7.04
(m, 2H), 6.88 (t, 1H), 6.80 (d, 1H), 3.60 (s, 3H), 2.34 (s,
3H); MS (ES, MH+) m/e=374.25.
1
3c (79% of yield). H NMR (300 MHz, CDCl3) d 12.45
(s, 1H), 7.77 (d, 2H), 7.57 (d, 1H), 7.45 (t, 1H), 7.37 (t,
1H), 7.18–7.24 (m, 3H), 7.12 (t, 1H), 6.92 (t, 1H), 6.88
(t, 1H), 6.76 (d, 1H), 3.68 (s, 3H), 3.53 (s, 3H), 2.35 (s,
3H); MS (ES, MH+) m/e=404.12.
1
(11.4 g, 56%). H NMR (300 MHz, CDCl3) d 7.88 (d,
2H), 7.45 (m, 3H), 7.19 (d, 2H), 6.92(d, 2H), 4.85 (s,
2H), 2.35 (s, 3H); MS (CI, MH+) m/e=211.
2-(2-Methylphenyl)-3-phenyl-5-(4-methoxyphenylamino)-
[1,2,4]thiadiazoline (8). To a solution of imidoylthiourea
7a (1.00 g, 2.67 mmol) in 10 mL of anhydrous chloro-
form was added N-chlorosuccinimide (356 mg, 2.67
mmol). The reaction mixture was then stirred for 16 hrs,
and then diluted to 50 mL of chloroform solution and
washed twice with saturated aqueous NaHCO3 solu-
tion. The organic layer was dried over MgSO4 and the
remaining solvent was removed under vacuum. The
crude product was washed by hexane to yield the pro-
duct 8 as a white solid (457 mg, 46%). 1H NMR
(300 MHz, CDCl3) d 7.59 (d, 2H), 7.41 (t, 1H), 7.30 (t,
2H), 6.98–7.10 (m, 6H), 6.90 (d, 2H), 3.79 (s, 3H), 2.32
(s, 3H); MS (ES, MH+) m/e=374.25.
Compound 6b was prepared in a similar fashion to give
a pale white solid (34%). 1H NMR (300MHz, CDCl3) d
8.27 (s, 1H), 7.39 (t, 1H), 6.94–7.11(m, 7H), 5.55 (s, 1H),
3.90 (s, 3H), 3.82 (s, 3H); MS (CI, MH+) m/e=257.
1-(4-Methoxyphenyl)-3-[N-(4-methylphenyl)benzimidoyl]-
2-thiourea (7a). A mixture of N-(4-methylphenyl)benz-
amidine 6a (2.10 g, 10.0 mmol) and 4-methoxyphenyl
isothiocyanate (1.39 mL, 10.0 mmol) in 30 mL of anhy-
drous 1,2-dichloroethane was heated at 55 ꢁC for 16 h.
The reaction mixture was cooled and the solvent was
evaporated. The resulting residue was purified by flash
column chromatography with a mobile phase of 25%
hexane in dichloromethane. The combined fractions were
evaporated; and the resulting solid was dried over vacuum
2-(2-Methylphenyl)-3-phenyl-5-N-methyl-N-(4-methoxy-
phenyl)amino][1,2,4]thiadiazolium triflate (9). To a solu-
tion of 8 (100 mg, 0.268 mmol) in 10 mL of anhydrous
DCM was added slowly methyl triflate (36.0 uL, 0.322
mmol). After stirring for 1.5 h, the reaction was stop-
ped. After removing the solvent under vacuum, the
crude product was washed with hexane and then dried
in high vacuum to yield 9 as a yellow solid (133 mg,
1
to yield the product 7a as yellow solid (1.54 g, 41%). H
NMR (300MHz, CDCl3) d 14.19 (s, 1H), 8.10 (s, 1H),
7.50 (d, 2H), 7.27–7.39 (m, 5H), 6.96 (t, 4H), 6.64 (d, 2H),
3.83 (s, 3H), 2.24 (s, 3H); MS (ES, MH+) m/e=376.25.
Compounds 7b and 7c were prepared in a similar fash-
ion to give yellow solids.
1
92%). H NMR (300MHz, CDCl3) d 7.62 (4H), 7.45
(1H), 7.32 (4H), 7.20 (2H), 7.06 (2H), 3.87 (3H), 3.82
(3H), 2.28 (3H); 13C NMR (300MHz, CDCl3) d 180.61
(1C), 169.49 (1C), 161.26 (1C), 141.86 (1C), 135.74 (1C),
133.04 (1C), 131.65 (1C), 130.93 (4C), 128.74 (2C), 127.19
(2C), 126.70 (2C), 126.25 (1C), 116.62 (2C), 55.75 (1C),
41.56, (1C), 21.33 (1C); MS (ES, MH+) m/e=388.26.
Compound 9 was also characterized by COSY (corre-
lated spectroscopy) and HMBC (hetero multibond
coupling) NMR techniques.
1
7b (79% yield). H NMR (300 MHz, CDCl3) d 13.90 (s,
1H), 8.18 (s, 1H), 7.94 (d, 1H), 7.26–7.38 (m, 7H), 7.19
(d, 1H), 6.93 (t, 1H), 6.75 (t, 2H), 6.67 (d, 1H), 3.66 (s,
3H), 2.36 (s, 3H); MS (ES, MH+) m/e=376.29.
1
7c (61% yield). H NMR (300 MHz, CDCl3) d 14.37 (s,
1H), 8.15 (s, 1H), 7.65 (d, 2H), 7.33 (t, 1H), 7.12–7.19
(m, 3H), 6.96 (t, 1H), 6.83 (t, 2H), 6.77 (d, 1H), 6.68 (t,
1H), 6.59 (d, 1H), 3.54 (s, 3H), 3.51 (s, 3H), 2.37 (s, 3H).
MS (ES, MH+) m/e=406.09.
Melanocortin MC4 receptor binding assay. The melano-
cortin receptor MC-4-membrane was purchased from
Receptor Biology, Inc. and coupled to wheat germ
agglutinin coated polyvinyl toluene-Scintillation Proxi-
mity Assay beads (Amersham Pharmacia Inc.) for 30
min at 25 ꢁC. Into eachwell of a 96-well Opti plate
(Packard), 2.5 mg of membrane and 0.25 mg of beads
were mixed in a volume of 100 mL media. The media
was 50 mM HEPES, pH 7.4 containing 0.1% bovine
serum albumin, 2 mM CaCl2, 2 mM MgCl2 and pro-
tease inhibitors (Bohreinger Mannheim). Test com-
pounds (1.5 mL) at 1 mM in 30% DMSO—50 mM
HEPES, pH 7.4 buffer was added to separate wells on
the plate and evaluated at four doses (n=2 or 3).
Radioactive ligand 125I-NDP-melanocyte stimulating
2-(4-Methylphenyl)-3-phenyl-5-(4-methoxyphenylamino)-
[1,2,4]thiadiazol-2-ium bromide (3a). To a solution of
imidoylthiourea 7a (2.00 g, 5.33 mmol) in 15 mL of
anhydrous chloroform was slowly added bromine (330
mL, 6.40 mmol). After stirring for 16 h, the solvent was
evaporated, and the resulting solid was washed with
anhydrous ethyl ether. The crude product was recrys-
tallized from a solution of 20% water in ethanol to yield
the product 3a as a yellow solid (1.81 g, 75%). 1H NMR
(300 MHz, CDCl3) d 12.78 (s, 1H), 7.80 (d, 2H), 7.58 (d,
2H), 7.51 (d, 1H), 7.39 (t, 2H), 7.27 (m, 2H), 7.19 (d,
2H), 6.96 (d, 2H), 3.84 (s, 3H), 2.45 (s, 3H); MS (ES,
MH+) m/e=374.25.

K. Pan et al. / Bioorg. Med. Chem. 11 (2003) 185–192
191
hormone (NEN, 2000 Ci/mmol) was added to each well
(48.5 mL per well, 40 pM final concentration). The plate
was then sealed and incubated for 16 h at 25 ꢁC. NDP-
Melanocyte stimulating hormone peptide and a-melano-
cyte stimulating hormone peptide (Palomar Research
Inc, 1 mM) were used as reference inhibitor compounds
to define non-specific binding (N). Total binding (T)
was defined using 30% DMSO–50 mM HEPES, pH 7.4
buffer. Bound radioactivity for eachwell (Y) measured
at counts per min (cpm) in a TopCount (Packard, CA)
was [(TꢀY)/(TꢀN)]ꢂ100%
animals to acclimate to feeding on powdered chow
(#5002 PMI Certified Rodent Meal) during the allotted
time. The chow was made available in an open jar,
anchored in the cage by a wire, with a metal follower
covering the food to minimize spillage, and water was
available ad-libitum. Animals were fasted for 18 hprior
to testing. At the end of the fasting period, animals were
administered either a test compound or vehicle. Vehicle
and test compound were administered either orally
(5 mL/kg) 60 min prior to the experiment, sub-
cutaneously (1 mL/kg) 30 min prior to the experiment,
or intraperitoneally (1 mL/kg) 30 min prior to the
experiment. Test compounds were administered orally
as a suspension in aqueous 0.5% methylcellulose-0.4%
Tween 80, or intraperitoneally as a solution or suspen-
sion in PEG 200; compound concentrations typically
ranged from 1 to 100 mg/kg, preferably from 10 to 30
mg/kg. Food intake was measured at 2, 4, and 6 hafter
administration by weighing the special jar containing
the food before the experiment and at the specified
times. Upon completion of the experiment, all animals
were given a 1-week washout period before re-testing.
Cyclic adenosine monophosphate (cAMP) stimulation as-
say. Human Bowes melanoma cells expressing human
melanocortin MC-4 receptor were grown to confluence
in a 24-well culture plate. The cells were washed once
withHanks’ solution and eachwell received 0.5 mL of
Hanks’ solution containing 1 uM isobutylmethyl xan-
thine. Test compounds were added to the appropriate
test wells. NDP-melanocyte stimulating hormone pep-
tide (1 mM) was added to the positive control wells while
negative control wells received vehicle of 30% DMSO–
50 mM HEPES, pH 7.4 buffer. The plate was incubated
at 37 ꢁC and 5% CO2 for 30 min. The supernatant was
discarded and the cells were washed twice with Hank’s
solution. Ice cold ethanol (80%, 0.5 mL) was added to
eachwell and the plate was incubated at 4 ꢁC for 30 min
to extract cyclic AMP. Cyclic AMP content was mea-
sured using the NEN Flashplate kit. A melanocortin
receptor agonist is defined as a test compound which
resulted in a significant increase in cAMP production
compared to vehicle control in this assay.
References and Notes
1. (a) Haskell-Luevano, C.; Hadley, M. E. Drug News Per-
spectives 1999, 12, 197. (b) Wikberg, J. E. S.; Muceniece, R.;
Mandrika, I.; Prusis, P.; Lindblom, J.; Post, C.; Skottner, A.
Pharmacol. Res. 2000, 42, 393. (c) Wikberg, J. E. Exp. Opin.
Ther. Patents 2001, 11, 61.
2. Chen, W. In The Melanocortin Receptors, Cone, E. D., Ed.;
Human Press: Totowa, NJ, 2000; p 449.
3. Abdel-Malek, Z. A.; Scott, M. C.; Furumura, M.;
Lamoureux, M. L.; Ollmann, M.; Barsh, G. S.; Hearing, V. J.
J. Cell. Sci. 2001, 114, 1019.
4. (a) Yang, Y.-K.; Thompson, D. A.; Dickinson, C. J.;
Wilken, J.; Barsh, G. S.; Kent, S. B. H.; Gantz, I. Endocrinol-
ogy 1999, 13, 148. (b) Ollmann, M. M.; Wilson, B. D.; Ying-
Kui, Y.; Kerns, J. A.; Chen, Y.; Gantz, I.; Barsh, G. S. Science
1997, 278, 135.
5. (a) MacNeil, D. J.; Howard, A. D.; Guan, X.; Fong, T. M.;
Nargund, R. P.; Bednarek, M. A.; Boulet, M. T.; Weinberg,
D. H.; Strack, A. M.; Marsh, D. J.; Chen, H. Y.; Shen, C. P.;
Chen, A. S.; Rosenblum, C. I.; MacNeil, T.; Tota, M.;
MacIntyre, E. D.; van der Ploeg, L. H. T. Eur. J. Pharmacol.
2002, 440, 141. (b) Adan, R. A.; Gispen, W. H. Eur. J. Phar-
macol. 2000, 405, 13.
6. (a) Alvaro, J. D.; Tatro, J. B.; Quillan, J. M.; Fogliano, M.;
Eisenhard, M.; Lerner, M. R.; Nestler, E. J.; Duman, R. S.
Mol. Pharm 1996, 50, 583. (b) Alvaro, J. D.; Tatro, J. B.;
Duman, R. S. Life Sci. 1997, 61, 1.
7. For a review of the function of the MC4 receptor on neu-
rotrophic and neuroprotective actions, see Hol, E. M.; Gispen,
W. H.; Baer, P. R. Peptides 1995, 16, 979.
8. Adan, R. A. H.; van der Kraan, M.; Doornbos, R. P.; Bar,
P. R.; Burbach, J. P. H.; Gispen, W. H. Mol. Brain Res. 1996,
36, 37.
9. (a) Poggioli, R.; Vergoni, A. V.; Bertolini, A. Peptides 1986,
7, 843. (b) Vergoni, A. V.; Poggioli, R.; Bertolini, A. Neuro-
peptides 1986, 7, 153.
10. Fan, W.; Boston, B. A.; Kesterson, R. A.; Hruby, V. J.;
Cone, R. D. Nature 1997, 385, 165.
11. Woychik, R. P. US Patent 6080550, 2000.
G-Protein activation assay. For eachassay, membranes
expressing the melanocortin MC-4 receptor (5 mg) were
incubated for 5 min at 25 ꢁC with0.5 nM 35S-GTPgS in
100 mL of 25 mM HEPES buffer, pH7.5 containing 100
mM NaCl, protease inhibitors, 0.5 mM GDP, 5 mM
2-mercaptoethanol, 1 mM MgCl2 together with an ago-
nist suchas the test compound, 1 mM of NDP-melano-
cyte stimulating hormone or
a combination of
NDP-melanocyte stimulating hormone and test com-
pound. Basal 35S-GTPgS binding was defined by 10 mM
HEPES, pH 7.4 buffer containing 30% DMSO. The
reaction was terminated by addition of 50 mL of termi-
nation buffer containing 25 mM HEPES, pH7.5, 20 mM
MgCl2, protease inhibitors, 100 mM GDP, 100 mM
GTP, 5 mM 2-mercaptoethanol with detergents (0.5%
digitonin, 0.2% sodium deoxycholate, and 0.5% NP-
40). The membranes were solubilized for 30 min at
25 ꢁC. The 35S-GTPgS bound Gas protein was immu-
noprecipitated using anti-Gas (Santa Cruz BiotechInc.,
0.5 mg) that would be linked to anti-rabbit IgG protein
A conjugated SPA (Amersham Pharmacia Inc.). Bound
radioactivity was measured in a Topcount. Non-specific
35S-GTPgS binding was defined by 35S-GTPgS immu-
noprecipitated by normal rabbit IgG (Sigma Inc., 0.5
mg).
Rodent feeding: food intake in food-deprived rats (MC-
4). Male Long–Evans rats (180–200 g) were housed
individually and maintained on a once-a-day feeding
schedule for five days following quarantine to allow the
12. Kask, A.; Rago, L.; Mutulis, F.; Pahkla, R.; Wirkberg,
J. E. S.; Schioth, H. B. Biomed. Biophys. Res. Comm. 1998,
245, 90.

192
K. Pan et al. / Bioorg. Med. Chem. 11 (2003) 185–192
13. Huszar, D.; Lynch, C.; Fairchild-Huntress, V.; Dunmore,
J. H.; Fang, Q.; Berkemeier, L. R.; Gu, W.; Kesterson, R.;
Boston, B.; Cone, R.; Smith, F.; Campfield, L. A.; Burn, P.;
Lee, F. Cell 1997, 88, 131.
14. Yeo, G. S.; Farooqi, I. S.; Aminian, S.; Halsall, D. J.;
Stanhope, R. G.; O’Rahilly, S. Nat. Genet. 1998, 20, 111.
15. Kordik, C. P.; Reitz, A. B. J. Med. Chem. 1999, 42,
181.
16. Haskell-Leuvano, C.; Hendrata, S.; North, C.; Sawyer,
T. K.; Hadley, M. E.; Hruby, V. J.; Dickinson, C.; Gantz, I.
J. Med. Chem. 1997, 40, 2133.
17. Bednarek, M. A.; MacNeil, T.; Kalyani, R. N.; Tang, R.;
van der Ploeg, L. H. T.; Weinberg, D. H. Biomed. Biophys.
Res. Comm. 1999, 261, 209.
18. Schioth, H. B.; Muceniece, R.; Mutulis, F.; Bouifrouri,
A. A.; Mutule, I.; Wikberg, J. E. S. Neuropeptides 1999, 33,
191.
19. (a) Dorr, R. T.; Lines, R.; Levine, N.; Brooks, C.; Xiang,
L.; Hruby, V. J.; Hadley, M. E. Life Sci. 1996, 58, 1777. (b)
Wessells, H.; Fuciarelli, K.; Hansen, J.; Hruby, V. J.; Dorr,
R.; Levine, N. J. Urol. 1998, 160, 389.
20. Basu, A.; Gahman, T. C.; Girten, B. E.; Griffith, M. C.;
Hecht, C. C.; Kiely, J. S.; Silvka, S. R.; Dines, K. S. WO 1999,
9955679.
21. Nargund, R. P.; Sebhat, I. K.; Ye, Z.; Barakat, K.; Wein-
berg, D.; MacNeil, T. Book of Abstracts, Part 1, 222nd
National Meeting of the American Chemical Society, Chicago,
IL, 26–30 August 2001; American Chemical Society:
Washington, DC, 2001; MEDI 151.
Mahahan, M. P. Synthesis 1985, 83. (d) Goerdeler, J.; Lobach,
W. Chem. Ber. 1979, 112, 517.
23. Ogonor, J. I. Tetrahedron 1981, 37, 2909.
24. Singh, B.; Collins, J. C. J. Chem. Soc., Chem. Commum.
1971, 498.
25. Schwenker, G.; Kolb, R. Tetrahedron 1969, 25, 5437.
26. Compound 3a was completely reduced to 7a after mixing
with 10 mL-equiv DTE in dichloromethane for 20 min.
27. Pan, K.; Reitz, A. Manuscript submitted for publication.
28. (a) It does not appear that the Bowes melanoma cell line
expresses the MC-1 receptor: Lombardi, T.; Castellucci, M.
Virchows Archiv B. Cell Path. Mol. Path. 1990, 58, 181. (b)
Schioeth, H. B.; Kuusinen, A.; Muceniece, R.; Szardenings,
M.; Keinaenen, K.; Wikberg, J. E. S. Biochem. Biophys. Res.
Commun. 1996, 221, 807.
29. (a) Bowden, C. R.; White, K. D.; Tutwiler, G. F. J. Obe-
sity Weight Regul. 1985, 4, 5. (b) Kordik, C. P.; Luo, C.;
Zanoni, B. C.; Lovenberg, T. W.; Wilson, S. J.; Vaidya, A. H.;
Crooke, J. J.; Rosenthal, D. I.; Reitz, A. B. Bioorg. Med.
Chem. Lett. 2001, 11, 2287.
30. Compound 3a was tested in two independent GTP-gS
binding assays (n=4), and a-MSH was used as a positive con-
trol. Compound 3a showed increased binding (37.5ꢃ6.9% at 10
nM, 76.5ꢃ11.2% at 100 nM, and 106.5ꢃ11.8% at 1000 nM),
comparable to the effect seen with a-MSH (33.5ꢃ3.1% at 10
nM, 84.0ꢃ11.1% at 100 nM, and 211.3ꢃ12.5% at 1000 nM).
31. Pandey, A. K.; Singh, R.; Verma, V. K. J. India Chem Soc.
1987, 64, 675.
32. Haskell-Luevano, C.; Sawyer, T. K.; Trumpp-Kallmeyer,
S.; Bikker, J. A.; Humblet, C.; Gantz, I.; Hruby, V. J. Drug
Des. Disc. 1996, 14, 197.
33. The pKa value for 3a was measured by pION, Inc., 5
Constitution Way, Woburn, MA 01801.
22. (a) Kihara, Y.; Kabashima, S.; Uno, K.; Okawara, T.;
Yamasaki, T.; Furukawa, M. Synthesis 1990, 1020. (b) Boh-
risch, J.; Patzel, M.; Grubert, L.; Liebscher, J. Phos. Sulfur
Silicon 1993, 84, 253. (c) Chetia, J. P.; Mazumder, S. N.;