5-HT2C receptor agonists for the treatment of obesity. Biological and chemical adventures

Article abstract of DOI:10.2533/000942904777677506

Obesity is a major risk factor in the development of conditions such as hypertension, hyperglycemia, dyslipidemia, coronary artery disease and cancer. There is increasing evidence suggesting an important role for the 5-HT _2C receptor in appetite control. Collaboration between F. Hoffmann-La Roche Ltd and Vernalis Research Ltd has allowed rapid construction of a solid structure-activity relationship around a pyrroloindole core. A one-pot Sonogashira reaction followed by nucleophilic double cyclisation allows an elegant and expedient route to this central motif. Introduction of a (2S)-aminopropyl group in place of the aminoethyl endogenous ligand side-chain enhanced the affinity at the 5-HT_2C receptor and reduced affinity towards monoamine oxidase enzymes (MAO). Sulfamidate reagents were found to be very effective for the introduction of the 2-aminopropyl moiety in a stereoselective manner. The substitution at position 5 (indole numbering) was found to be crucial for both affinity and selectivity. Pyrroloindoles bearing an alkoxyether in this position exhibit promising pharmacokinetic parameters in rodent and significant reduction of food intake, after per os application.

Full text of DOI:10.2533/000942904777677506

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
613
CHIMIA 2004, 58, No. 9
Chimia 58 (2004) 613–620
© Schweizerische Chemische Gesellschaft
ISSN 0009–4293
5-HT2C Receptor Agonists for the
Treatment of Obesity. Biological and
Chemical Adventures
David Adamsb, Agnès Bénardeaua, Mike J. Bickerdikeb, Jon M. Bentleyb, Caterina Bissantza,
Anne Boursona, Ian A. Cliffeb, Paul Hebeisena, Guy A. Kennettb, Antony R. Knightb, Craig S. Mal-
colmb, Jacques Mizrahia, Jean-Marc Planchera*, Hans Richtera, Stephan Rövera, Sven Taylora,
and Steven P. Vickersb
Abstract: Obesity is a major risk factor in the development of conditions such as hypertension, hyperglycemia, dys-
lipidemia, coronary artery disease and cancer. There is increasing evidence suggesting an important role for the
5-HT2C receptor in appetite control. Collaboration between F. Hoffmann-La Roche Ltd and Vernalis Research Ltd
has allowed rapid construction of a solid structure–activity relationship around a pyrroloindole core. A one-pot
Sonogashira reaction followed by nucleophilic double cyclisation allows an elegant and expedient route to this
central motif. Introduction of a (2S)-aminopropyl group in place of the aminoethyl endogenous ligand side-chain
enhanced the affinity at the 5-HT2C receptor and reduced affinity towards monoamine oxidase enzymes (MAO).
Sulfamidate reagents were found to be very effective for the introduction of the 2-aminopropyl moiety in a stereo-
selective manner. The substitution at position 5 (indole numbering) was found to be crucial for both affinity and
selectivity. Pyrroloindoles bearing an alkoxyether in this position exhibit promising pharmacokinetic parameters in
rodent and significant reduction of food intake, after per os application.
Keywords: Obesity · Pyrroloindole · 5-HT2C Receptor agonist · Sulfamidate
1. Introduction
example, the prevalence of obesity trebled mates for the obesity drug market in the
during the past 25 years [2].
USA are ca. $5 billion by the year 2005
The healthcare burden that the obesity epi-
Obesity is a major risk factor in the de- [5].
demic now poses is highly significant [1]. velopment of conditions such as hyperten-
In 1997, the World Health Organisation sion, hyperglycemia, dyslipidemia, corona- 1.2. Existing Treatments
(WHO) officially declared human obesity ry artery disease, and cancer (Fig. 1) [3][4].
Broadly speaking, drugs are classified
to be one of the most significant health
The two main causes for the global rise by their principal mechanism of action on
problems facing mankind. This declaration in prevalence of obesity are increasingly the energy balance. When energy intake
followed the massive expansion in the pre- sedentary lifestyles and high-fat, energy (food intake) exceeds expenditure, a state
valence of obesity in almost all societies. In rich diets.
the United States of America (USA), for
of positive balance exists and vice versa.
In a state of negative energy balance,
weight loss ensues. Anti-obesity drugs are
designed with the goal to induce a state of
1.1. Scale of the Problem
The WHO has standardised its classifi- negative energy balance until the required
cation using the Body Mass Index (BMI; weight loss is reached. Anti-obesity drugs
weight [kg]/(height [m]2) as an indicator of effectively fall into four categories:
obesity. A person is now considered over-
weight with a BMI >25 and considered
obese with a BMI >30. For example, using
this measure, 55% of adults in the USA are
overweight or obese.
–
Appetite suppressants: inhibit energy
intake which act on the central nervous
system (CNS) to reduce the feeling of
hunger.
*Correspondence: Dr. J.-M. Planchera
Tel. : + 41 61 688 67 25
Fax: + 41 61 688 83 67
E-Mail: jean-marc.plancher@roche.com
aF. Hoffmann-La Roche Ltd
Pharmaceuticals Division
B092/4.18C
–
–
Inhibitors of fat absorption: reduce
energy intake and act peripherally.
Enhancers of energy expenditure: in-
crease thermogenesis/energy expendi-
ture independently of physical activity
and act peripherally.
The total cost in the USA for all obesi-
ty-related health problems exceeds $200
billion. More alarmingly, each year
300,000 people die prematurely in the USA
from obesity-related complications.
CH–4070 Basel
bVernalis Research Ltd
Oakdene Court
613 Reading Road
Wokingham RG41 5UA
United Kingdom
Obviously, there is a tremendous need
for efficient therapies. Conservative esti-
–
Stimulation of fat mobilisation: reduce
fat mass and triglyceride synthesis, inde-

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
614
CHIMIA 2004, 58, No. 9
A specific aspect in the medication of
obesity is the discrepancy between patient
expectation of up to 30% body weight loss
in the first year, possibly without dieting,
and current therapy performance of typical-
ly 5% body weight loss over the same peri-
od of time, with specific diet (compared to
a placebo group under the same diet).
This huge difference in expectation is
hampering patient compliance. Therefore
even minor side effects can become critical.
The limited efficacy might be due to re-
dundancies and compensatory effects in the
mechanism regulating energy intake. Com-
bination therapies, such as co-administra-
tion of a lipase inhibitor and an appetite
suppressant might be the key to achieve
high efficacy by synergistic or simply addi-
tive effect.
1.3. Obesity and 5-HT2C Receptor
Agonists
5-Hydroxytryptamine (Serotonin, 5-HT,
7) is a major neurotransmitter in the mam-
malian CNS, mediating a diverse range of
actions. Of the 14 mammalian serotonin re-
ceptor subtypes, all but one (5-HT3) belong
to the super-family of G-protein-coupled
receptors (GPCRs) [6]. The 5-HT2 receptor
family consists of 3 subtypes (2A, 2B, and
2C), which have been grouped together on
the basis of primary structure, secondary
messenger system and pharmacological
profile [3]. Increasing evidence suggest an
important role for Serotonin (5-HT) and the
5-HT2C receptor in appetite control, besides
other clinical indications including anxiety,
depression, OCD, migraine, sleep, epilepsy
and schizophrenia [7–9].
Fig. 1. Prevalence of obesity and diabetes among US adults. Copyright: A.H. Mokdad, E.S. Ford,
B.A. Bowman, W.H. Dietz, F. Vinicor, V.S. Bales, J.S. Marks, J. Am. Med. Asso. 2003, 289, 76.
The non-selective 5-HT2C receptor ago-
nist mCPP (8, m-chlorophenylpiperazine)
reduces appetite and body weight in obese
patients. The anorectic action of mCPP is
absent in 5-HT2C receptor knockout mutant
mice and is antagonized by the selective 5-
HT2C receptor antagonist SB-242084 9
(Fig. 3) [10–12]. Similarly, the 5-HT2 ago-
nist RO0600175 10 [13] suppresses pala-
table food intake in rats. Mice with a targe-
ted disruption of the 5-HT2C receptor have
increased food consumption and develop
late-onset obesity. In view of these findings,
a number of companies are currently in the
process of developing 5-HT2C receptor
agonists.
Fig. 2.
next two years. Phase III clinical trials
are ongoing to evaluate the potential of
Sertraline (5, Selective Serotonin Re-up-
take inhibitor -SSRI-, Pfizer) as anorec-
tic. This drug is already widely marketed
for depression, obsessive compulsive di-
sorder (OCD) and anxiety (Zoloft®).
Axokine® (Recombinant human Ciliary
Neurotrophic Factor -CNTF-, Regener-
on) and Rimonabant (6, CB-1 receptor
inverse agonist, Sanofi-Synthelabo), spe-
cifically designed to treat obesity, are
currently also in Phase III clinical trials
[2].
pendently of changes in food intake and
exercise regimes and act peripherally.
After the withdrawal of dex-fenfluramine
(Serotonin Re-uptake Inhibitor/5-HT
Recently, phase II clinical trials have
demonstrated significant weight loss with
the 5-HT2C receptor agonist BVT-933 (Bio-
vitrum, structure not published).
1
releaser, Servier) in 1999, only three drugs are
currently approved for the treatment of
obesity: Sibutramine (2, Serotonin-Nor-
adrenaline-Dopamine Re-uptake Inhibitor
-SNRI-, Abbott), Phentermine (3, Noradren-
ergic, generic, USA-market only), and Orlistat
(4, Lipase inhibitor, Roche) exemplifying how
difficult it is to tackle this pandemic (Fig. 2).
Additional drugs have entered clinical
trials and may reach the market in the
One of the most challenging goals in the
serotoninergic field is to achieve a high de-
gree of selectivity towards the different re-
ceptors, in terms of affinity and function.
In 1997, the widely prescribed appetite
suppressant dex-fenfluramine (1) was with-
drawn from the market due to treatment-re-
lated cardiac valve alterations and pulmo-

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
615
CHIMIA 2004, 58, No. 9
2.2. First Synthetic Route
The initial synthesis started from 5-me-
thoxy indole (16) and involved construction
of the third ring by radical cyclisation [16]:
Vilsmeier-Hack condensation with phos-
phorus oxychloride in N,N-dimethylforma-
mide at 40 °C introduced the aldehyde in the
3-position of the indole. N-alkylation using
3-chloro-1-bromopropane and potassium
hydroxide in dimethyl sulfoxide at room
temperature followed by Finkelstein activa-
tion with sodium iodide in refluxing acetoni-
trile provided the desired iodo intermediate
18. Tributyltin hydride-mediated 5-exo-trig
radical cyclisation in toluene at reflux led to
the desired tricyclic pyrroloindole 19 core in
poor yield (13%), along with reduced mate-
rial [16]. The side-chain was introduced by
reaction of the aldehyde with nitroethane in
the presence of ammonium acetate at 100
°C, which provided the nitroalkene 20. Re-
duction with lithium aluminium hydride in
refluxing tetrahydrofuran afforded the de-
sired racemic product (rac-12) in moderate
yield [17]. Optical resolution by chiral
HPLC was feasible only on small scale due
to poor separation (Scheme 1).
RO0600175 10
Fig. 3.
Joint efforts at F. Hoffmann-La Roche
Ltd and Vernalis Research Ltd have allowed
the construction of a solid structure–activi-
ty relationship around the pyrroloindole co-
re.
The double bond of the indole ring was
shown to be mandatory for the affinity to
the 5-HT2C receptor. Substitution of the
amino group was not tolerated; however
mono or di-N-methylation might be envi-
sioned in a pro-drug approach. The optimal
absolute configuration was found to be (S).
The (R)-antipode was one log unit less po-
tent. Further modifications of the 2-amino-
propyl side chain, with the aim to tune the
selectivity or the basicity were not success-
ful. Equally, the SAR around the benzene
ring was found to be quite narrow, with the
exception of the 5-position, where large va-
riations were allowed (Fig. 5, see also Sec-
tion 5.)
Fig. 4.
nary hypertension. A recent publication
indicates a possible role of 5-HT2B recep-
tor [14] in this alteration. 5-HT2A agonism
may mediate the effects of hallucinogens
such as LSD-25. Therefore, high 5-HT2C
receptor subtype selectivity is of prime im-
portance.
endogenous ligand of the 5-HT receptors,
shows some functional selectivity for the 5-
HT2B receptor (Fig. 4) and the O-methyl-
serotonin (11) exhibits no significant selec-
tivity, a biased screening has highlighted
the interesting properties of the pyrroloin-
dole core (i.e. 12), particularly with regard
to the 5-HT2A receptor selectivity.
3. Chemistry
3.1. Pyrroloindole Construction
Overall, the selective 5-HT2C receptor
agonist approach should lead to a better
safety profile than for the known 5-HT
releasing and 5-HT reuptake inhibitors,
such as dex-fenfluramine (1).
Further exploration focused on the 5-
position. For that purpose a new, efficient,
flexible, and preferably chiral synthetic ap-
proach was deemed mandatory. The issues
of the initial approach (i.e. low yield and te-
dious chiral HPLC enantiomer separation)
were overcome by a novel and expedient
approach that gave rapid access to key
intermediates and target compounds in a
stereochemically defined form.
Pyrrole-ring opening (13) reduces po-
tency around one log-unit while the piperi-
dine-homologue 14 lost potency almost
completely. Interestingly, the larger 7-
membered homologue 15 regained some
potency, however only modestly (Fig. 4).
The α-methyl group was incorporated
into the amino side chain in order to sup-
press metabolic side chain deamination by
monoamine oxidases (MAO). The slight in-
crease of lipophilicity also allows for better
CNS permeation [13]. However, this modi-
fication did not increase potency and selec-
tivity significantly.
2. Early SAR and Chemistry
2.1. Biased Screening and Discovery
of Pyrroloindole
Agents with high affinity for 5-HT2C re-
ceptors typically possess high affinity for 5-
HT2A receptors and rarely display >10-fold
selectivity [15]. While Serotonin (7), the
Starting from 4-bromoaniline (21) the
amine was protected as the methylcarbama-
te and selectively ortho-iodinated using N-
iodosuccinimide (NIS), leading to 22. So-

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
616
CHIMIA 2004, 58, No. 9
nogashira reaction of this material with 5-
chloro-1-pentyne allowed smooth and effi-
cient incorporation of the 5-carbon unit
(Scheme 2). Treatment of acetylene 23
with lithium hydroxide in wet DMSO at 80
ûC led to deprotection of the carbamate,
formation of the indole ring and a second
annelation to install the pyrroloindole cen-
tral motif 25 [18]. As a further simplifica-
tion the Sonogashira and annelation reac-
tions were combined in a one-pot proce-
dure with equally good yields. The
bromotricyclic compound 25 was conver-
ted to the phenol 26 by metallation with n-
BuLi, reaction with triisopropylborate and
oxidation with hydrogen peroxide. For
preparative purposes a defined alkyl side
chain could be introduced at this stage by
simple alkylation to furnish e.g. 27a. For
late stage derivatisation the phenol 26 was
protected as the thexyldimethylsilyl ether
27b.
Scheme 1.
3.2. (S)-Aminopropyl Side Chain
Incorporation
The stereocontrolled incorporation of the
aminopropyl side chain was envisioned by
reaction of an indole or the 3-lithio analogue
29 with an electrophilic chiral aminopropyl
synthon. For this purpose, the pyrroloindole
27a was iodinated in good yield using NIS
and subsequently metallated at low tempera-
ture (–78 °C). Quenching of the lithio-indole
29 with water confirmed the completion of
the metal-halogen exchange. However, at-
tempts to trap the organolithium with the me-
thanesulfonic acid 2-tert-butoxycarbonyla-
mino-propyl ester 30 met with no success
(Scheme 3). Under strongly basic conditions,
cyclisation to the N-Boc protected methyla-
ziridine 31 took place [19]. No ring opening
of aziridine 31 with lithio pyrroloindole 29 to
32 occurred even after addition of copper or
zinc salts. As a consequence more reactive
aminopropyl synthons which were devoid of
an acidic proton were designed.
Fig. 5.
Based on the synthetic equivalence of
epoxides and cyclic sulfates suggested by
Sharpless [20][21], we envisioned the cy-
clic sulfamidate 35 as an aziridine equiva-
lent with enhanced reactivity (Scheme 4).
The synthesis of 35 started from the readi-
ly available chiral (S)-alaninol which was
N-Boc-protected (33), cyclized to the 2-
oxo-[1,2,3]oxathiazolidine 34 with thio-
nylchloride and oxidised with NaIO4 in
the presence of catalytic amounts of RuO2
to give the stable crystalline alkylating
agent 35 [22]. Although described for the
sulfate analogs and for the prolinol derived
sulfamidate, the direct synthesis of the cy-
clic sulfamidate from (S)-prolinol using
sulfuryl chloride did not work, presumab-
ly due to high reactivity of the N-Boc sul-
famidate, which does not tolerate the
presence of a nucleophilic base such as
imidazole or pyridine [23][24].
Scheme 2.

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
617
CHIMIA 2004, 58, No. 9
The maximum fluorescent signal (annota-
ted as rel. eff. in Fig. 4 and the Table) ob-
tained at 1 µM was measured and compared
with the response produced by 10 µM 5-HT
(defined as 100%).
For compounds showing potency above
50% (compared with 5-HT) from this func-
tional assay, a full concentration-response
curve using the same techniques was gen-
erated (annotated as EC50 in Fig. 4). In pa-
rallel, these compounds were compared in
radioligand binding displacement assay at
recombinant human 5-HT2A, 5-HT2B, and
5-HT2C receptors expressed in CHO cells
(annotated as Ki h5-HT2A, Ki h5-HT2B, and
Ki h5-HT2C, respectively; Table). For the 5-
HT2A receptor [125I]-DOI (rac-2,5-dime-
thoxy-4-iodoamphetamine) was used as the
radioligand whereas [3H]-5-HT (cf. com-
pound 4) was used at the 5-HT2B and 5-
HT2C receptor subtypes [25]. Some results
are shown in the Table.
Scheme 3.
The presence of a heteroatom attached
to the phenyl ring is mandatory. Increasing
chain length (entry 3 versus 5) enhances se-
lectivity towards the 5-HT2B receptor sub-
type. Steric hindrance on the α-carbon is
detrimental for affinity (entry 8); only cy-
clopropyl is tolerated (entry 7). Decoration
on the β-carbon is allowed (entry 9).
However, despite various substituents on
the β- and γ-position, no significant impro-
vements were found (e.g. entry 12). Linear
hydroxyethoxy (entry 10), and methoxy-
propoxy were found to be the most intere-
sting variations in terms of potency, affini-
ty and selectivity towards 2A and 2B recep-
tor subtypes (>70-fold).
Scheme 4.
4.2. Metabolic Stability
In vitro intrinsic clearance in human he-
patocyte preparations has been determined
for the most promising compounds. From
these values the hepatic in vivo clearance
was extrapolated (Table, annotated as Cl h-
hepat) through direct scaling, this last value
being more representative of the in vivo
situation.
Phenol derivatives were generally ob-
served as the main metabolites, presumably
through oxidation of the α-carbon followed
by dealkylation. Increasing sp2 character of
this carbon (entry 7) or local polarity (ent-
ries 10–11) significantly increased metabo-
lic stability (Table).
Scheme 5.
amine as the hydrochloride salt. Differenti-
This powerful alkylating agent 35 was
reacted with the lithio-indole derived from
36, in good yields and with complete reten-
tion of configuration (Scheme 5). The
scope and limitations of this useful class of
reagents are now under investigation. This
method has been applied successfully to the
synthesis of several 5-substituted pyrrolo-
indoles (i.e. 5-bromo 37a, 5-ethoxy 37b, 5-
silyloxy 37c), including applications on a
large scale (over 50 g).
al deprotection of the silyl-protected phenol
function in 37c with NH4F in methanol
allowed for late variation of the alkoxy
substituent.
5. Modelling
5.1. Ligand Pharmacophore
One very important concept of medic-
inal chemistry is the idea of pharmaco-
phores. A pharmacophore is defined as the
minimum functionality a molecule has to
contain in order to show activity at a certain
receptor. A pharmacophore is normally
represented as a set of features (such as
positive charge, negative charge, hydropho-
4. In vitro Results
4.1. Affinity, Potency and Selectivity
The compounds were screened for func-
tional activity at recombinant 5-HT2C re-
ceptors expressed in CHO cells using a flu-
orimetric imaging plate reader (FLIPR).
Simple reaction of the N-Boc protected
amines with hydrogen chloride in ethyl ace-
tate led to precipitation of unprotected

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
618
CHIMIA 2004, 58, No. 9
Table.
bic region, aromatic ring center, H-bond
donor, H-bond acceptor) in a defined spati-
al arrangement. Molecules interacting with
a receptor in the same way should all inclu-
de these features as minimum requirement
even if the molecules have completely dif-
ferent chemical structures. By defining
such a pharmacophore, we can learn a lot
about what makes a certain compound acti-
ve at a receptor.
Using CATALYST [26] and ten known
5-HT2C agonists, a 3-point ligand pharma-
cophore was defined which consisted of a
protonated amine, an aromatic ring and an
additional hydrophobic substituent in a
defined spatial orientation (Fig. 6). This
means that all 5-HT2C agonists activating
the receptor in the same way as the known
agonists should possess at least these three
features.
The alignment of the pyrroloindole co-
re onto the pharmacophore showed that the
requirement of a hydrophobic substituent is
fulfilled by the substituent in position 5,
which can explain the relative importance
of substituents in this position for activity.
5.2. 5-HT2C Receptor Model
While the pharmacophore approach
uses only ligand information to understand
what functionalities are required for a com-
pound to be active, such information can
also be gained by investigating how
compounds of a series interact with the
receptor. In this approach, a three-dimen-
sional structure of the binding site of the
receptor is required. This is preferentially
an experimentally determined X-ray
structure obtained by cristallography, but
alternatively homology models may be
used where such information is unavailable.
Until now the only high-resolution X-
ray structure of a GPCR that has been sol-
ved is the structure of bovine rhodopsin
[27]. Structure information of other GPCRs
may, therefore only be obtained from ho-
mology models using the X-ray structure of
bovine rhodopsin as a template.
To construct this homology model, the
amino acid sequence of human 5-HT2C re-
ceptor was aligned with the sequence of bo-
vine rhodopsin. The X-ray strucuture of bo-
vine rhodopsin was then modified by chang-
ing the amino acid side chains to the
respective side-chains of the 5-HT2C receptor.
The pyrroloindole core was then docked
into the initial 5-HT2C receptor model such
that the basic amine formed a salt bridge
with an aspartate and the benzene ring
formed π-π interactions with a phenylala-
nine, interactions that are conserved in most
biogenic amine receptors. The energy of the
receptor/ligand complex was then mini-
mized by Amber 6.0 [28]. In the final mod-
el, substituents in position 5 of the
pyrroloindole are located in a large pocket
between the transmembrane domains 3, 4,
Fig. 6.

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
619
CHIMIA 2004, 58, No. 9
all tests being performed during the light
phase. Access to tap water was ad libitum.
6.2. Method
Rats were kept individually and were
deprived of food for 23 h. They were hou-
sed in plexiglass boxes with a grid on the
floor and paper was placed below the ca-
ge floor to collect the spillage. A food di-
spenser filled with a preweighed amount
of food was presented to them for 1h. At
the end of the food intake session, rats we-
re returned to their home cage.
6.3. Results
Compounds were tested for their ability
to reduce food intake in food-deprived rats.
Compound 38 (cf. Table) dose-dependently
decreased food intake during the 1 h period
of food access with a minimum efficacious
dose (MED) of 10 mg/kg p.o., as shown in
Fig. 8. The duration of action exceeded 8 h
at 30 mg/kg. After a pre-treatment time of
20 h, decrease of food intake was no longer
observed at any dose.
Fig. 7.
The anorexic effect was dose-depend-
ently blocked by pre-treatment with the
selective 5-HT2C receptor antagonist SB-
2420849, indicating a 5-HT2C receptor
mediated effect.
7. Conclusions
Obesity is becoming a major disease
around the world. Collaboration between
F. Hoffmann-La Roche Ltd and Vernalis
Research Ltd has allowed the discovery
and the rapid development of chiral pyr-
roloindoles as potent 5-HT2C receptor
agonists.
Several compounds were found to be
very selective over other 5-HT2 receptor
subtypes (>25 fold). Effective access to the
pyrroloindole motif, using a tandem Sono-
gashira/double cyclisation reaction was de-
veloped. For the first time, chiral sulfami-
dates were used for the stereospecific intro-
duction of a (2S)-aminopropyl side chain.
This synthetic route could be scaled up to
50 g of final product. Optimisation of phar-
macokinetic parameters has allowed good
in vivo efficacy after per os application in a
rodent food-deprived model.
Fig. 8. Effect of 38 on food intake in Lister hooded rats under 23 h food deprivation. 7–8 rats were
used per group. Either compound 38 (5, 10, 20, 30 mg/kg) or saline were administered orally at dif-
ferent time points (2, 4, 8 and 20 h) before the 1 h food intake session. *P <0.05 compared to vehi-
cle-treated rats (Anova followed by Dunnett’s t-test).
Received: June 18, 2004
[1] P.J. Carek, L.M. Dickerson, Drugs 1999,
57, 883.
5, and 6. This may explain why large varia-
tions are tolerated in this position (Fig. 7). In
contrast, the environment of the pyrole ring,
close to the extracellular loop between TM
4 and 5 could not be modelled robustly.
Although most of the optimization was
driven by SAR, ligand- and structure-based
modelling approaches provided valuable
information to target this GPCR.
6. In vivo Activity
[2] D. Spanswick, K. Lee, Expert. Opin.
Emerging Drugs 2003, 8, 217.
6.1. Animals
[3] Centers for Disease Control and preven-
tion; U.S. Department of Health and Hu-
man Services (http://www.cdc.gov/).
[4] A.H. Mokdad, E.S. Ford, B.A. Bowman,
W.H. Dietz, F. Vinicor, V.S. Bales, J.S.
Marks, J. Am. Med. Asso. 2003, 289, 76.
Male hooded Lister rats, weighing
160–230 g at the beginning of the experi-
ment, were used, each rat being used once.
Ambient temperature was approximately
21 oC and relative humidity 55–65%. A 12
h light–dark cycle was maintained, with

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
620
CHIMIA 2004, 58, No. 9
[5] C.P. Kordik, A.B. Reitz, J. Med. Chem. [14] R.B. Rothman, M.H. Bauman, J.E. Sa- [21] H.C. Kolb, M.S. VanNieuwenhze, K.B.
1999, 42, 181.
vage, L. Rauser, A. McBride, S.J. Hufei-
Sharpless, Chem. Rev. 1994, 2483.
[6] P. Manivet, B. Schneider, J.C. Smith, D.-
S. Choi, L. Maroteaux, O. Kellermann,
J.-M. Launay, J. Bio. Chem. 2002, 277,
17170.
sen, B.L. Roth, Circulation 2000, 102, [22] J.M. Bentley, M.J. Bickerdike, P. Hebei-
2836; L.W. Fitzgerald, T.C. Burn, B.S.
Brown, J.P. Patterson, M.H. Corjay, P.A.
Valentine, J.-H. Sun, J.R. Link, I. Abbas-
zade, J.M. Hollis, B.L. Largent, P.R.
Hartig, G.F. Hollis, P.C. Meunier, A.J.
Robichaud, D.W. Robertson, Mol. Phar- [23] J.E. Baldwin, A.C. Spicey, C. Schofield,
macol. 2000, 57,75. Tetrahedron Asym. 1990, 1, 881.
sen, G.A. Kennett, S. Lightowler, P. Mat-
tei, J. Mizrahi, T.J. Morley, J.-M. Planch-
er, H. Richter, S. Roever, S. Taylor, S.P.
Vickers, PCT Int. Appl. 2002, WO
2002051844 A1.
[7] J.C. Jerman, S.J. Brough, T. Gager, M.
Wood, M.C. Coldwell, D. Smart, D.N.
Middlemiss, Eur. J. Pharm. 2001, 414,
23.
[8] F. Jenck, M. Bös, J. Wichmann, H. Stad- [15] J. Chang-Fong, J. Addo, M. Dukat, C. [24] D. Alker, K.J. Doyle, L.M. Harwood, A.
ler, J.-M. Martin, J.-L. Moreau, Expert.
Opin. Invest. Drugs 1998, 7, 1587.
Smith, N.A. Mitchell, K. Herrick-Davis,
M. Teitler, R.A. Glennon, Bioorg. Med.
Chem. Lett. 2002, 12, 155.
McGregor, Tetrahedron Asym. 1990, 1,
877.
[25] R.H. Porter, K.R. Benwell, H. Lamb,
C.S. Malcolm, N.H. Allen, D.F. Revell,
D.R. Adams, M.J. Sheardown, Br. J.
Pharmacol. 1999, 12, 8 13.
[9] M. Issac, Drugs Fut. 2001, 26, 383.
[10] M.J. Bickerdike, S.P. Vickers, C.T. Dou- [16] C.J. Moody, C.L. Norton, J. Chem. Soc.
rish, Diabetes, Obesity & Metabolism
Perkin Trans. 1 1997, 2639.
1999, 1, 207; M.J. Bishop, B.M. Nilsson, [17] D.R. Adams, J.M. Bentley, J.R.A. Rof-
Expert. Opin. Ther. Patents 2003, 13,
1691.
fey, R.J. Hamlyn, S. Gaur, M.A.J. Dunc- [26] Molecular Simulations Inc., San Diego,
ton, J.E.P. Davidson, M.J. Bickerdike, CA, USA.
[11] P.A. Sargent, A.L. Sharpley, C. Williams,
E.M. Goodall, P.J. Cowen, Psychophar-
macology 1997, 133, 309.
I.A. Cliffe, H.L. Mansell, PCT Int. Appl. [27] K. Palczewski, T. Kumasaka, T. Hori,
2000, WO 2000012510 A1.
C.A. Behnke, H. Motoshima, B.A. Fox,
I. Le Trong, D.C. Teller, T. Okada, R.E.
Stenkamk, M. Yamamoto, M. Miyano,
Science 2000, 5480, 739.
[18] H.G. Chen, C. Hoechstetter, P. Knochel,
Tetrahedron Lett. 1989, 30, 4795.
[12] S.P. Vickers, C.T. Dourish, G.A. Kennett,
Neuropharm. 2001, 41, 200.
[19] P.R. Giles, M. Rogers-Evans, M. Soukup, J.
[13] M. Bös, F. Jenck, J.R. Martin, J.-L. Mo-
Knight, Org. Process Res. Dev. 2003, 7, 22. [28] AMBER 6, University of California, San
reau, A.J. Sleight, J. Wichmann, U. Wid- [20] Y. Gao, K.B. Sharpless, J. Am. Chem.
Francisco, 1999.
mer, J. Med. Chem. 1997, 40, 2762.
Soc. 1988, 110; 7538.