Identification of 4-methyl-1,2,3,4,10,10a-hexahydropyrazino[1,2-a]indoles as 5-HT2C receptor agonists

Article abstract of DOI:10.1016/j.bmcl.2005.05.074

Synthesis and evaluation of the activity of new 4-methyl-1,2,3,4,10,10a- hexahydropyrazino[1,2-a]indoles as 5-HT_2C receptor agonists are described. Appropriately substituted, several analogs displayed selectivity against the other 5-HT₂ receptor subtypes of 1 order of magnitude or more. Selectivity was improved for several compounds versus the lead 1, increasing the therapeutic interest in this series of 5-HT_2C receptor agonists.

Full text of DOI:10.1016/j.bmcl.2005.05.074

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3604–3608
Identification of 4-methyl-1,2,3,4,10,10a-
hexahydropyrazino[1,2-a]indoles as 5-HT2C receptor agonists
a
b
d
a
S. Ro¨ver,b, D. R. Adams, A. Benardeau, J. M. Bentley, M. J. Bickerdike,
*
´
A. Bourson,b I. A. Cliffe,a P. Coassolo,b J. E. P. Davidson,a C. T. Dourish,a
P. Hebeisen,b G. A. Kennett,a A. R. Knight,a C. S. Malcolm,a P. Mattei,b A. Misra,a
J. Mizrahi,b M. Muller,c R. H. P. Porter,b H. Richter,b S. Taylorb and S. P. Vickersa
aVernalis Research Ltd, Oakdene Court, 613 Reading Road, Wokingham RG41 5UA, UK
bF. Hoffmann-La Roche Ltd, Discovery Research, 4070 Basel, Switzerland
cBasilea Pharmaceutica Ltd, 4005 Basel, Switzerland
dGlaxoSmithKline S.p.A, Centre of Excellence for Drug Discovery, Psychiatry, Via Fleming 4, 37135 Verona, Italy
Received 7 March 2005; revised 2 May 2005; accepted 11 May 2005
Available online 21 June 2005
Abstract—Synthesis and evaluation of the activity of new 4-methyl-1,2,3,4,10,10a-hexahydropyrazino[1,2-a]indoles as 5-HT2C
receptor agonists are described. Appropriately substituted, several analogs displayed selectivity against the other 5-HT2 receptor
subtypes of 1 order of magnitude or more. Selectivity was improved for several compounds versus the lead 1, increasing the ther-
apeutic interest in this series of 5-HT2C receptor agonists.
Ó 2005 Elsevier Ltd. All rights reserved.
1. Introduction
Evidence for the involvement of the 5-HT2C receptor in
appetite control comes from a number of human and
animal studies. The non-selective 5-HT2C receptor par-
tial agonist, m-chlorophenylpiperazine (mCPP), reduces
appetite and body weight in normal human volunteers3
and obese subjects,4 and induces hypophagia in rats.5
This hypophagic activity of mCPP in rats is attenuated
by the selective 5-HT2C receptor antagonist
SB-2420846 and it is absent in 5-HT2C receptor knock
out mice.7 These findings have spurred our interest in
discovering new, potent, and selective 5-HT2C receptor
agonists for controlling appetite and treating obesity.
Research in the field of serotoninergics has generated a
wealth of therapeutic agents, for example, the selective
serotonin re-uptake inhibitors (SSRIs) for depression,
5-HT1B/1D receptor agonists for treating migraine, anx-
iolytic 5-HT1A receptor partial agonists, and 5-HT3
receptor antagonists for cancer therapy induced emesis.1
More recently, there is increasing evidence for an impor-
tant role of the 5-HT2C receptor in appetite control.
Appetite suppression is one of the pharmacological
mechanisms useful in controlling obesity, which is a
major risk factor in the development of hypertension,
hyperglycemia, dyslipidemia, coronary artery disease,
and cancer. Survey data from the US show a rising prev-
alence of obesity with 31% of adults classified as obese
and 65% as either overweight or obese in 2002.2
Researchers from Roche and Vernalis, respectively,
have previously described the discovery of 5-HT2C ago-
nists based on the pyrazino[1,2-a]indole scaffold.8 While
some of these early compounds (e.g., 1) were potent
5-HT2C agonists with in vivo activity in food intake
models after po application to rats, decent selectivity
against 5-HT2A and 5-HT2B receptors was not achieved.
Indeed, selectivity is one of the most important points in
the design of 5-HT2C agonists, as cardiovascular and
psychotomimetic effects have been described for
5-HT2A agonists. Both 5-HT2A and 5-HT2B receptor
Keywords: 5-HT2C receptor agonist; Pyrazino[1,2-a]indole.
*
Corresponding author. Tel.: +41 61 688 5906; fax: +41 61 688
6459; e-mail: stephan.roever@roche.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.05.074

S. Ro¨ver et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3604–3608
3605
agonism has been implicated in the valvular pathology
seen with dexfenfluramine, a non-specific agonist,
5-HT re-uptake inhibitor, and 5-HT releaser.9
Still pyrazino[1,2-a]indoles are attractive leads combining
small size with potent in vitro and in vivo effects. So, we
reasoned that analogs of 1, especially the diastereomers
produced by appropriate substitution at the aliphatic car-
bons, might retain some of the desirable features of the
lead, but produce new compounds with improved selec-
tivity profile. This strategy proved to be the most success-
ful with methyl substitutions at position 4.
Scheme 2. Reagents and conditions: (a) NaH, DMF, then (S)-3;
(b) TFA, CH2Cl2; (c) MeOH, K2CO3; (d) LiAlH4, Et2O, reflux;
(e) NaBH4, TFA, THF.
in very much the same way from the corresponding
(R)-5-methyl-2,2-dioxo[1,2,3]oxathiazolidine-3-carboxylic
acid tert-butyl ester [(R)-3] (not shown).
The above-described synthesis method gave the
(4R,10aR)- and (4S,10aS)-enantiomers in good to excel-
lent overall yield. For the synthesis of their diastereo-
mers (4R,10aS and 4S,10aR) in more than minute
amounts, this route was not useful however.
2. Methods
Synthesis of chiral 4-methyl-substituted pyrazino[1,2-
a]indoles was possible by developing and using the pure
enantiomers of 5-methyl-2,2-dioxo[1,2,3]oxathiazoli-
dine-3-carboxylic acid tert-butyl ester [(S)-3 and (R)-3,
respectively] as a new type of chiral alkylating agent.10
The synthesis (illustrated for (S)-3 in Scheme 1) starts
from readily available chiral 1-amino-2-propanols that
are Boc-protected, cyclized to the 2-oxo[1,2,3]oxathiaz-
olidine with thionyl chloride and oxidized with NaIO4
and catalytic amounts of RuO2 to give the crystalline
and stable alkylating agents (S)-3 and (R)-3,
respectively.
A different sequence of reduction steps, as detailed in
Scheme 3, provided these target molecules in better
yield. (4R)-4-Methyl-1-oxo-dihydropyrazino[1,2-a]indoles
were reduced with magnesium in methanol (in analogy to
a method developed by Youn for reducing a,b-unsaturat-
ed esters)12 to mainly give the (4R,10aS)-4-methyl-1-oxo-
tetrahydropyrazino[1,2-a]indoles (4R,10aS/4R,10aR ra-
tio ꢀ2:1). The reaction products were not purified but
reduced directly with LiAlH4 in THF to yield the
target (4R,10aS)-4-methyl-hexahydropyrazino[1,2-a]in-
doles, which were separated by column chromatography.
Appropriately substituted indole-2-carboxylates were
alkylated by (S)-3 in excellent yield (Scheme 2). As
expected for a SN2 reaction, this step proceeded with
inversion of configuration. Boc-deprotection of the
1-alkylated indole-2-carboxylates with trifluoroacetic
acid in dichloromethane was followed by cyclization to
the corresponding 1-oxo-dihydropyrazino[1,2-a]indoles
using potassium carbonate in methanol. Reduction with
LiAlH4 in diethyl ether, in turn, gave the 4-methyl-tetra-
hydropyrazino[1,2-a]indoles, which were further reduced
with NaBH4 in a mixture of trifluoroacetic acid and THF.
The last step provided the (4R,10aR)-diastereomer in an
excess over its (4R,10aS)-diastereomer (ratio of ꢀ9:1).11
The stereoisomers were separated by conventional
column chromatography on silica gel. The synthesis of
the (4S,10aS)- and (4S,10aR)-diastereomers proceeded
The 4-methyl-hexahydropyrazino[1,2-a]indoles can, pro-
vided they are Boc-protected and substituted appropri-
ately in the aromatic ring, be processed further to
provide target molecules with more elaborate substitution
patterns, as exemplified in Scheme 4. Starting from
Boc-protected 7-bromo-4-methyl-hexahydropyrazino[1,2-
a]indole 4 Buchwald–Hartwig type chemistry gave the
7-amino compound, which could be acylated and depro-
tected to give the 7-acetylamino-4-methyl-hexahydro-
pyrazino[1,2-a]indole 12. The reaction of the same
Scheme 1. Reagents and conditions: (a) (Boc)2O, CH2Cl2; (b) SOCl2,
imidazole, CH2Cl2, À50 °C; (c) NaIO4, RuO2, EtOAc.
Scheme 3. Reagents and conditions: (a) Mg, MeOH; (b) LiAlH4,
THF, reflux.

3606
S. Ro¨ver et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3604–3608
and consequently, a number of di-substituted com-
pounds were also investigated. Of these, the
(4R,10aR)-7-chloro-6-methyl compound 17 is an excep-
tional subnanomolar agonist. Binding profiles of several
4-methyl-substituted compounds were compared with
those of their 4-desmethyl analogs, which are accessible
via the same route starting simply from ethanolamine as
starting material for the corresponding alkylating agent
and separating the racemate into enantiomers by chiral
HPLC on ChiralPak ADÒ.
Scheme 4. Reagents and conditions: (a) Pd2(dba)3, ( )-BINAP, NaOt-
Bu, toluene, benzophenone imine; Pd/C, NH4-formate, MeOH; (b)
AcCl, NEt3, CH2Cl2; (c) TFA, CH2Cl2; (d) n-BuLi, THF, CO2,
À78 °C; (e) LiAlH4, THF.
7-bromo compound with n-BuLi and carbon dioxide gave
the corresponding carboxylic acid, which was reduced
with LiAlH4 and deprotected to give the 7-hydroxymethyl
compound 13.
From these comparisons (e.g., 17–19, 18–20, Table 3), it
seems that, indeed, the 4R-methyl substituent conveys
additional potency for the 5-HT2C receptor in this
pyrazino[1,2-a]indole series of 5-HT2 receptor ligands.
Luckily, the effect of the 4R-methyl substituent on the
potency of 5-HT2A and 5-HT2B receptors is far more
variable and depends on the specific aromatic substitu-
tion pattern.
3. Results and discussion
The compounds were screened for functional activity at
human recombinant 5-HT2C receptors expressed in
CHO cells using a fluorimetric imaging plate reader
(FLIPR). The maximum fluorescent signal was mea-
sured and compared with the response produced by
10 lM 5-HT (defined as 100%).13 The active compounds
from this functional assay were then compared in radio-
Indeed by varying the aromatic substituents, appreciable
selectivity for 5-HT2C receptors against 5-HT2A and
5-HT2B receptors is achieved. Most selective compound
in this panel is the (4R,10aR)-7-cyano compound 11
(>20-fold each), which is characterized as a partial
agonist at the 5-HT2C receptor. However, also several
full agonists, that is, (4R,10aR)-7-methyl (9), (4R,
10aR)-6-methyl (14), and (4R,10aS)-7-chloro-6-methyl
(18) that show >10-fold selectivity for 5-HT2C receptors
against both 5-HT2A and 5-HT2B receptors, were
identified.
ligand binding assays at recombinant human 5-HT2A
,
5-HT2B, and 5-HT2C receptors expressed in CHO cells.
For the 5-HT2A receptor, [125I]DOI was used as the radi-
oligand, whereas [3H]5-HT was used for the 5-HT2B and
5-HT2C receptor subtypes.14 The results are shown in
Tables 1–3. Table 1 details the effect of the 4-methyl
group on the 5-HT2 receptor affinity and selectivity of
7-chloro-hexahydropyrazino[1,2-a]indoles 1, 5–8. The
(4R,10aR)-stereochemistry in compound 6 provides a
full agonist with the highest affinity for all the 5-HT2
receptors (ꢀ10-fold higher than 1) but with little selec-
tivity. The (4S,10aS)-isomer 5 is a partial agonist and
about 100-fold less potent overall than its enantiomer
6, while both the (4S,10aR)- and (4R,10aS)-isomers 7
and 8 are essentially full agonists and show appreciable
affinity for the 5-HT2C receptor with some selectivity
against 5-HT2B receptors (7- to 16-fold).
Compound 18 was tested in Wistar rats for its ability
to reduce food intake after an acute administration.15
Indeed, the compound significantly reduced food intake
in a dose-dependent manner in rats that had been food
deprived for 22 h. The minimal effective dose (MED)
after subcutaneous administration was 1 mg/kg and
after po administration MED was 10 mg/kg.
4. Conclusions
The influence of simple substituents at the 6- and
7-positions of the 4-methyl-hexahydropyrazino[1,2-a]in-
dole skeleton on receptor affinity and selectivity is given
in Table 2.
We have synthesized a novel series of chiral 4-methyl-
hexahydropyrazino[1,2-a]indoles, isolated, and charac-
terized all four stereoisomers. These compounds,
especially the (4R,10aR)-diastereomers, proved to be
high potency full agonists at 5-HT2C receptors with
the 4R-methyl substituent contributing reproducibly to
improved potency at the 5-HT2C receptor. Appropriate-
ly substituted, several analogs displayed selectivity
against the other 5-HT2 receptor subtypes of 1 order of
magnitude and more. These compounds and especially
further refined analogs with even higher selectivity and
in vivo potency have potential for use in therapy as
anti-obesity agents.
Small lipophilic substituents at the 7-position (Cl, CN,
CF3 and Me), whether electron-withdrawing or donat-
ing, all lead to high affinity for the 5-HT2C receptor,
while polar substituents lead to a comparative loss in
affinity.
The 6-methyl-substituted compound 14 shows a very
similar binding profile to the 7-methyl compound 9,

S. Ro¨ver et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3604–3608
3607
Table 1. 5-HT2 receptor efficacy and binding of 4-substituted 7-chloro-1,2,3,4,10,10a-hexahydropyrazino[1,2-a]indoles
Compound 4-Substituent Stereochemistry Relative efficacya h5-HT2C (%) Kib h5-HT2A (nM) Kic h5-HT2B (nM) Kic h5-HT2C (nM)
1
5
6
7
8
4-H
rac
99
62
97
94
87
32
430
53
2600
15
31
180
4-Me
4-Me
4-Me
4-Me
4S,10aS
4R,10aR
4S,10aR
4R,10aS
1.9
2.2
55
61
210
160
13
22
a Efficacy at 1 lM ligand concentration relative to 10 lM of 5-HT (100%).
b Displacement of [125I]DOI.
c Displacement of [3H]5-HT.
Table 2. 5-HT2 receptor efficacy and binding of substituted 4-methyl-1,2,3,4,10,10a-hexahydropyrazino[1,2-a]indoles
Compound R1 R2
Stereochemistry Relative efficacya h5-HT2C (%) Kib h5-HT2A (nM) Kic h5-HT2B (nM) Kic h5-HT2C (nM)
9
10
11
12
13
14
15
16
17
18
H
Me
4R,10aR
4R,10aR
4R,10aR
4R,10aR
100
94
77
24
97
100
96
91
98
97
22
7.5
21
11
1.3
1.4
3.8
H
CF3
CN
H
110
1100
290
43
90
H
NHAc
960
250
59
680
34
2.6
H
Me
CH2OH 4R,10aR
H
4R,10aR
4R,10aR
4R,10aS
4R,10aR
4R,10aS
Me Me
Me Me
Me Cl
Me Cl
34
102
13
67
1.7
9.5
0.3
1.9
2.6
40
3.2
19
a Efficacy at 1 lM ligand concentration relative to 10 lM 5-HT (100%).
b Displacement of [125I]DOI.
c Displacement of [3H]5-HT.
Table 3. 5-HT2 receptor efficacy and binding of 6-methyl-7-chloro-1,2,3,4,10,10a-hexahydropyrazino[1,2-a]indoles
Compound
Stereochemistry
Relative efficacya h5-HT2C (%)
Kib h5-HT2A (nM)
Kic h5-HT2B (nM)
Kic h5-HT2C (nM)
19
20
10aRd
10aSd
90
80
6.2
52
10
140
1.5
32
a Efficacy at 1 lM ligand concentration relative to 10 lM 5-HT (100%).
b Displacement of [125I]DOI.
c Displacement of [3H]5-HT.
d Absolute stereochemistry tentative, based on the comparison of ORD values with (10aS)-7-trifluoromethyl-1,2,3,4,10,10a-hexahydropyrazino[1,2-a]-
indole for which X-ray data were obtained.
Bromidge, S.; Forbes, I. T.; Brown, A. M.; Middlemiss,
D. N.; Blackburn, T. P. Neuropharmacology 1997, 36,
609.
Acknowledgments
Our warmest thanks and gratitude go to all members of
the 5-HT2C project teams of Vernalis and Hoffmann-
La Roche, respectively.
7. Tecott, L. H.; Sun, L. M.; Akana, S. F.; Strack, A. M.;
Lowenstein, D. H.; Dallman, M. F.; Julius, D. Nature
1995, 374, 542.
8. (a) Boes, M.; Jenck, F.; Martin, J. R.; Moreau, J. L.;
Mutel, V.; Sleight, A. J.; Widmer, U. Eur. J. Med. Chem.
1997, 32, 253; (b) Adams, D. R.; Bentley, J. M.; Davidson,
J.; Duncton, M. A. J.; Porter, R. H. P. WO200044753A1;
Chem. Abstr. 2000, 133, 150579.
9. (a) Gardin, J. M.; Schumacher, D.; Constantine, G.;
Davis, K. D.; Leung, C.; Reid, C. L. J. Am. Med. Assoc.
(JAMA) 2000, 283, 1703; (b) Rothman, R. B.; Baumann,
M. H.; Savage, J. E.; Rauser, L.; McBride, A.; Hufeisen,
S. J.; Roth, B. L. Circulation 2000, 102, 2836.
10. (a) Bentley, J. M.; Hebeisen, P.; Muller, M.; Richter, H.;
Roever, S.; Mattei, P.; Taylor, S. WO2002010169A1;
Chem. Abstr. 2002, 136, 167392; (b) Posakony, J. J.;
Grierson, J. R.; Tewson, T. J. J. Org. Chem. 2002, 67,
5164; (c) Bentley, J. M.; Bickerdike, M. J.; Hebeisen, P.;
Kennett, G. A.; Lightowler, S.; Mattei, P.; Mizrahi, J.;
Morley, T. J.; Plancher, J.-M.; Richter, H.; Roever, S.;
Taylor, S.; Vickers, S. P. WO2002051844A1; Chem. Abstr.
2002, 137, 78856.
References and notes
1. Jones, B. J.; Blackburn, T. P. Pharmacol. Biochem. Behav.
2002, 71, 555.
2. National Center for Health Statistics, US Centers for
Disease Control and Prevention (http://www.cdc.gov/
nchs/).
3. Walsh, A. E. S.; Smith, K. A.; Oldman, A. D.; Williams,
C.; Goodall, E. M.; Cowen, P. J. Psychopharmacology
1994, 116, 120.
4. Sargent, P. A.; Sharpley, A. L.; Williams, C.; Goodall, E.
M.; Cowen, P. J. Psychopharmacology 1997, 133, 309.
5. Vickers, S. P.; Benwell, K. R.; Porter, R. H.; Bickerdike,
M. J.; Kennett, G. A.; Dourish, C. T. Br. J. Pharmacol.
2000, 130, 1305.
6. Kennett, G. A.; Wood, M. D.; Bright, F.; Trail, B.; Riley,
G.; Holland, V.; Avenell, K. Y.; Stean, T.; Upton, N.;

3608
S. Ro¨ver et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3604–3608
11. Enantiomeric excess in the SN2 reaction was determined
by means of chiral HPLC using 2% ethanol in heptane as
mobile and Chiralpak-AD as stationary phase. Confirma-
tion of the absolute configuration of the C4-carbon was
carried out by crystallization and X-ray of (4R)-8-bromo-
7-fluoro-4-methyl-3,4-dihydro-1H-pyrazino[1,2-a]indole-
2-carboxylic acid tert-butyl ester, which was obtained in
several steps using (S)-3 in the key alkylation step. 1H
NMR and COSY allowed the assignment of equatorial
and axial protons in the pyrazine ring of the 4-methyl-
hexahydropyrazino[1,2-a]indoles 9–18 and was used to
establish the relative configuration.
12. Youn, I. K.; Yon, G. H.; Pak, C. S. Tetrahedron Lett.
1986, 27, 2409.
13. Porter, R. H. P.; Benwell, K. R.; Lamb, H.; Malcolm, C.
S.; Allen, N. H.; Revell, D. F.; Adams, D. R.; Sheardown,
M. J. Br. J. Pharmacol. 1999, 128, 13.
14. (a) 5-HT2A: McKenna, D. J.; Peroutka, S. J. J. Neurosci.
1989, 9, 3482; (b) 5-HT2B: Schmuck, K.; Ullmer, C.;
Engels, P.; Lubberts, H. FEBS Lett. 1994, 342, 85; (c) 5-
HT2C: Hoyer, D.; Engel, G.; Kalkman, H. O. Eur. J.
Pharmacol. 1985, 118, 13.
15. Vickers, S. P.; Dourish, C. T.; Kennett, G. A. Neurophar-
macology 2001, 41, 200.