Tetrahydroquinoline-based tricyclic amines as potent and selective agonists of the 5-HT2Creceptor

Article abstract of DOI:10.1016/j.bmcl.2016.11.016

The syntheses, structure-activity relationships (SARs), and biological activities of tetrahydroquinoline-based tricyclic amines as 5-HT_2Creceptor agonists are reported. An early lead containing a highly unique 6,6,7-ring system was optimized for both in vitro potency and selectivity at the related 5-HT_2Breceptor. Orally bioactive, potent, and selective 6,6,6-tricyclic 5-HT_2Cagonists were identified.

Full text of DOI:10.1016/j.bmcl.2016.11.016

Accepted Manuscript
Tetrahydroquinoline-based tricyclic amines as potent and selective agonists of
the 5-HT2C receptor
Thomas O. Schrader, Michelle Kasem, Albert Ren, Konrad Feichtinger, Bilal
Al Doori, Jing Wei, Chunrui Wu, Huong Dang, Minh Le, Joel Gatlin, Kelli
Chase, Jenny Dong, Kevin T. Whelan, Carleton Sage, Andrew J. Grottick,
Graeme Semple
PII:
S0960-894X(16)31156-8
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.11.016
BMCL 24411
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
29 September 2016
4 November 2016
7 November 2016
Please cite this article as: Schrader, T.O., Kasem, M., Ren, A., Feichtinger, K., Al Doori, B., Wei, J., Wu, C., Dang,
H., Le, M., Gatlin, J., Chase, K., Dong, J., Whelan, K.T., Sage, C., Grottick, A.J., Semple, G., Tetrahydroquinoline-
based tricyclic amines as potent and selective agonists of the 5-HT2C receptor, Bioorganic & Medicinal Chemistry
Letters (2016), doi: http://dx.doi.org/10.1016/j.bmcl.2016.11.016
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Tetrahydroquinoline-based tricyclic amines as
Leave this area blank for abstract info.
potent and selective agonists of the 5-HT2C receptor
Thomas O. Schrader, Michelle Kasem, Albert Ren, Konrad Feichtinger, Bilal Al Doori, Jing Wei, Chunrui Wu,
Huong Dang, Minh Le, Joel Gatlin, Kelli Chase, Jenny Dong, Kevin T. Whelan, Carleton Sage, Andrew J.
Grottick, and Graeme Semple

Bioorganic & Medicinal Chemistry Letters
Tetrahydroquinoline-based tricyclic amines as potent and selective agonists of
the 5-HT2C receptor
Thomas O. Schrader a,, Michelle Kasem a, Albert Ren a, Konrad Feichtinger a, Bilal Al Doori a, Jing Wei b,
Chunrui Wu b, Huong Dang a, Minh Le a, Joel Gatlin a, Kelli Chase a, Jenny Dong a, Kevin T. Whelan a,
Carleton Sage a, Andrew J. Grottick a, and Graeme Semple a
aDepartment of Medicinal Chemistry, Arena Pharmaceuticals, 6154 Nancy Ridge Drive, San Diego, CA, 92121, USA
bWuXi AppTec (Wuhan) Co Ltd., 666 Gaoxin Road, East Lake High-tech Development Zone, Wuhan 430075, China
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
The syntheses, structure-activity relationships (SARs), and biological activities of
tetrahydroquinoline-based tricyclic amines as 5-HT2C receptor agonists are reported. An early
lead containing a highly unique 6,6,7-ring system was optimized for both in vitro potency and
selectivity at the related 5-HT2B receptor. Orally bioactive, potent, and selective 6,6,6-tricyclic
5-HT2C agonists were identified.
2009 Elsevier Ltd. All rights reserved.
Keywords:
5-HT2C agonists
Lorcaserin
Tetrahydroquinoline
CNS drugs
Tricyclic amines
By the late 1990s, significant evidence had emerged
concerning the role of the 5-HT2C receptor (5-HT2CR) in
mediating satiety and food intake.1 The advent of 5-HT receptor
subtype-selective ligands was crucial to this understanding.2 In
particular, rodent studies using selective 5-HT2CR antagonists
revealed the anorectic effects of the previously FDA-approved
weight loss agents fenfluramine (Pondimin, figure 1) and its S-
isomer dexfenfluramine (Redux) were mediated, in part, by
potent agonism of the 5-HT2CR by the drugs’ primary metabolites
norfenfluramine and nordexfenfluramine respectively.3 These
data were consistent with a previous human clinical study which
demonstrated that the anorectic effect of dexfenfluramine was
attenuated with the nonselective 5-HT2CR antagonist ritanserin.4
Other antagonist experiments performed in rodents with the
nonselective 5-HT2C agonist meta-chlorophenylpiperazine
(mCPP), another clinically validated anorectic, confirmed these
results.5 Additional studies involving variably selective 5-HT2C
ligands,5 as well as data obtained from 5-HT2C receptor knockout
mice,6 suggested that agonism of the 5-HT2C receptor was a
viable mechanism for the treatment of obesity.
tissue.9 Other adverse events including hallucinations10 and
cardiovascular effects,11 are associated with agonism of another
closely related target, the 5-HT2A receptor (5-HT2AR). These
findings led to an industry-wide effort to discover and develop
selective 5-HT2CR agonists which did not affect 5-HT2BR and 5-
HT2AR function.12 The result was the 2012 FDA approval of the
5-HT2CR selective agonist lorcaserin (1, Belviq®)13 for weight
management in obese (BMI >30) or overweight (BMI 27-30)
patients with a weight-related medical condition.14 In addition to
obesity, selective 5-HT2CR agonists have therapeutic potential for
a wide variety of neuropsychiatric disorders.15 To this end, we
report the identification of a series of tetrahydroquinoline-based
tricyclic amines as potent and selective agonists of the 5-HT2C
receptor. Details of the syntheses and biological activities of
these compounds are provided herein.
Complicating matters was the 1997 withdrawal7 of both
fenfluramine and dexfenfluramine due to drug related cardiac
fibroses and related valvulopathies,8 side-effects later associated
with agonism of the related 5-HT2B receptor (5-HT2BR) in cardiac
———
Figure 1. Known anorectic agents.
 Corresponding author. Tel.: +1-858-453-7200; fax: +1-858-453-7210; e-mail: tschrader619@gmail.com

Early in our investigations the racemic 6,6,7-tricyclic amine,
1,2,3,4,5,5a,6,7-octahydro-[1,4]diazepino[1,2-a]quinoline (2,
(10g, pEC50 = 8.4). Further elaboration at the para position with
a larger benzyl substituent decreased activity significantly (10h,
pEC50 = <6) at both receptors. Of particular note is the increase
in 5-HT2CR potency observed with the 6,6,6-tricyclic analog 11
(pEC50 = 8.6). While none of the compounds displayed
meaningful improvements in 5-HT2CR versus 5-HT2BR
selectivity as compared with 2R, analogs which contained a
chloro substituent at the ortho position relative to the aniline
nitrogen (7, 10d,f) led to significant decreases in receptor
intrinsic activity (Emax≤ 20%) at the 5-HT2BR. The high binding
potency of partial agonist 10d (Emax< 5%) was confirmed in
scheme 1) was identified as a moderately potent agonist of the 5-
HT2CR (EC50 = 130 nM in IP3 accumulation assay). The
synthesis of this novel compound starts from commercially
available 2-quinolinemethanamine (3). Benzamide formation via
amide coupling was followed by hydrogenation to give the
racemic tetrahydroquinoline intermediate (4) in 53% yield (two
steps).
Selective alkylation of the aniline nitrogen was
accomplished by heating a mixture of 4 and 3-bromopropan-1-ol
in the absence of solvent at 120 °C for 1 h to yield compound 5.
Conversion of the alcohol (of 5) to the bromide is accomplished
by treatment of 5 with aqueous HBr which gives concomitant
deprotection of the benzamide group. Finally, base promoted
cyclization forms the C-ring to produce 2. To provide some
initial analogs to probe aromatic ring substitution SARs, the
racemic mono-chlorinated compounds 6 and 7 were prepared by
protecting the free nitrogen of 2 by trifluoracetylation, reaction
with NCS, and deprotection with methanolic ammonia.
[
125I]-DOI competition binding assays (5-HT2BR pKi = 8.3).
Table 1. 5-HT2CR and 5-HT2BR functional activities in intracellular IP3
accumulation assay for mCPP and tetrahydroquinoline-based tricyclic
compounds 2, 6,7, 10, and 11.
5-HT2C
R
5-HT2BR
a
b
a
b
Cmpd
R1
R2
R3
R4
pEC50
Emax
pEC50
Emax
mCPP
2
7.9 [0.1]
6.9 [0.1]
7.2 [0.2]
<5
90
99
86
7.4 [0.6]
6.0 [0.1]
6.3 [0.1]
<5
22
60
53
H
H
H
H
H
H
H
H
H
H
Cl
H
H
H
Cl
H
Cl
H
H
H
2R
H
H
H
2S
Scheme 1. Reagents and conditions: (a) PhCOOH, EDC.HCl, DMAP,
DCM, 40 °C; (b) H2, PtO2, MeOH, rt; (c) 3-bromopropan-1-ol, 120 °C; (d) 48
wt.% aq. HBr, rt; (e) Cs2CO3, ACN, rt 15 h, 81%, (f) (CF3CO)2O, Et3N,
DCM, rt; (g) NCS, ACN, 65 °C; (h) NH3, MeOH, rt.
H
Cl
H
6.7 [0.7]
7.6 [0.3]
7.6 [0.2]
6.2 [0.4]
7.4 [0.2]
8.0 [0.3]
7.1 [0.3]
6.8 [0.9]
8.4 [0.3]
<6
83
85
94
80
91
80
95
72
85
6.2 [0.2]
7.0 [0.1]
6.2 [0.2]
5.9 [0.5]
6.8 [0.2]
N.D.c
71
20
59
76
23
<5
37
7
6
H
H
H
7
H
F
H
10a
10b
10c
10d
10e
10f
10g
10h
10i
10j
11
H
OMe
OMe
OMe
H
H
We previously reported a general asymmetric route to this
class of fused tricyclic (R)-2,3,4,4a,5,6-hexahydro-1H-
pyrazino[1,2-a]quinolines and (R)-1,2,3,4,5,5a,6,7-octahydro-
[1,4]diazepino[1,2-a]quinolines.16 Starting from various N-Boc-
o-toluidines (8, Table 1) and (S)-tert-butyldimethyl(oxiran-2-
ylmethoxy)silane (9), the enantiopure analogs 2R, 10a-c,e,i-j, as
well as 6,6,6-tricylic compound 11 were prepared. The synthesis
of 2R by this method served as a stereochemical structure proof
for the individual enantiomers of 2, which had been originally
separated via chiral HPLC. Additional SAR analogs 10d,f-g
were prepared by chlorination of 10c and 10e. The benzylated
analog 10h was prepared from 10e by a four step sequence
involving bromination with NBS, Boc-protection, Negishi
coupling, and Boc-deprotection (see Supplementary Material).
H
Cl
H
Cl
H
OMe
OMe
6.2 [0.1]
8.1 [1.3]
7.2 [0.3]
<6
H
H
H
Cl OMe
Bn OMe
84
H
OMe
H
H
H
H
H
H
H
6.0 [1.5]
92
17
97
6.4 [1.5]
7.0 [0.8]
7.2 [0.2]
74
55
28
OMe 7.5 [0.2]
H
H
8.6 [0.6]
apEC50 values are geometric means of at least two experiments. Numbers in
brackets are 95% confidence intervals. bEmax % based on maximum
asymptote at 10 M relative to serotonin (5-HT). cDue to low intrinsic
activity (Emax < 5%), pEC50 was not determined (N.D).
To assess 5-HT2CR agonism, as well as potential
valvulopathogenic activities related to 5-HT2BR activation,
functional activities (pEC50 and Emax values) for prepared
compounds were determined in IP3 accumulation assays.17 As
shown in Table 1, the agonist activity of racemic compound 2 at
both 5-HT2CR and 5-HT2BR was entirely attributed to the
R-enantiomer enantiomer (2R). Other SAR analyses revealed
most single substitutions of fluoro (10a), chloro (6, 7) and
methoxy (10e,j) are well tolerated and did not result in
significant changes in 5-HT2CR activity as compared with the
parent compound 2R. Exceptions are the methoxy analogs 10b
(pEC50 = 6.2) and 10i (pEC50 = 6.0) which exhibited decreased
potency. However, in the case of 10b, 5-HT2CR activity is
The finding that 6,6,6-tricyclic compound 11 displayed the
most potent agonist activity at the 5-HT2CR without negatively
impacting selectivity versus the 5-HT2BR prompted further
investigations of this scaffold. In order to efficiently prepare
SAR analogs (12a-c,h-x, Scheme 2) with varied aromatic ring
substituents, we opted to prepare the compounds as racemates
using chemistry developed by Bernotas.18 In this method, the
lithium anion of ethyl propiolate is reacted with ortho-
fluorobenzyaldehydes (13a-g), and the resultant alcohols are
rearranged to enones 14a-g by treatment with Et3N in dioxane at
60 °C. A key one-step double cyclization is accomplished by
reaction of enones 14a-g with 1,2-diaminoethane in DMF at 60
°C to provide tricyclic ketones 15a-g. Compounds 12a-c were
prepared from 15a-c by ketone removal with TFA and Et3SiH
followed by amide reduction with either LiAlH4 or BH3.THF.
Chlorination of 12c with NCS gave the bis-halogenated
compound 12h. Methyl analog 12j was prepared from 15d (R1 =
returned by adding one (10c, pEC50 = 7.4) or two (10d, pEC50
8.0) additional chloro substituents. A similar potency increase at
the 5-HT2CR is observed when the methoxy analog 10e (pEC50
7.1) is chlorinated at the position para to the aniline nitrogen
=
=

Br) by palladium-catalyzed coupling with trimethlboroxine
followed by reductions of the carbonyl groups. The brominated
Boc-protected intermediates 16d-g, were also prepared from 15d-
g respectively by the ketone removal/amide reduction sequence,
with an added Boc-protection. The bromine atoms served as
handles to perform palladium-catalyzed coupling reactions,
which delivered analogs 12i,k-t after acid-mediated Boc-
deprotections. Aldehyde 17 was prepared from 16d via lithium-
bromine exchange and formylation with DMF. Reduction of the
aldehyde (17) with NaBH4 and conversion to the benzyl chloride
(18) allowed the synthesis of benzyl analogs 12u-v by Suzuki
couplings. Lastly, conversion of the bromine atom of 16d to
phenol 19 was accomplished by palladium-catalyzed
pinacolboration followed by oxidative deboronation. Alkylation
of 19 with cyclobutyl bromide or copper-mediated coupling with
phenylboronic acid gave 12w and 12x after Boc-group removal
with 4N HCl in dioxane.
tricycles 20a-b. An additional molecule (20c) containing a
chloro-substituent adjacent the aniline nitrogen was prepared
from 20b by chlorination with NCS.
The identification of 5-HT2CR agonists in the 6,6,6-tricyclic
series which displayed exceptional functional selectivity over the
5-HT2BR encouraged us to reexamine a number of similar
analogs in the 6,6,7-series. Therefore compounds 24 and 25
(Scheme 4) were prepared. Expanding the scope of the
Berntoas18 methodology, the synthesis of the racemic 6,6,7-ring
compound 24 employed 1,2-diaminopropane in the double-
cyclization reaction of 14d to give ketone 26 in 42% yield. The
remaining steps were performed in similar fashion as described
previously. The enantiopure compound 25, was prepared from
previously reported tricycle 2716 by a Suzuki coupling with
benzyltrifluoroborate and benzyl deprotection under transfer
hydrogenation conditions.
Scheme 3. Reagents and conditions: (a) 10% Pd/C, H2, MeOH, rt; (b) ethyl
bromoacetate, K2CO3, ACN, 80-100 °C: (c) LiAlH4, THF, rt; (d) MsCl,
DIEA, DCM; (e) NH3, H2O, ACN, 80 °C; (f) NCS, DCM, rt.
Scheme 4. Reagents and conditions: (a) 1,3-diaminopropane, DMF, 60 °C;
.
(b) TFA, Et3SiH, rt; (c) BH3 THF, reflux; (d) Boc2O, DCM, rt; (e)
.
(cyclobutylmethyl)zinc bromide, cat. Pd(dppf)Cl2 DCM, THF, 90 °C; (f) 4N
HCl, dioxane, rt; (g) Potassium benzyltrifluoroborate, cat. Pd(OAc)2, RuPhos,
K2CO3, PhMe, H2O, 115 °C; (h) 10% Pd/C, NH4+COO-, MeOH, 40 °C.
Scheme 2. Reagents and conditions: (a) ethyl propiolate, LDA, THF, 0 °C,
0.5 h then 13a-g, THF, -78 °C to rt; (b) Et3N, dioxane, 60 °C; (c) 1,2-
diaminoethane, DMF, 60 °C; (d) TFA, Et3SiH, rt; (e) LiAlH4, THF, rt; (f)
5-HT2CR and 5-HT2BR functional activities for second
generation 6,6,6- and 6,6,7-tricyclic compounds are presented in
Table 2. 5-HT2AR functional activities and binding affiinities
(pKi)19 for all 5-HT2 receptors are included for potent 5-HT2CR
agonists (pEC50 > 8) which exhibited significantly greater
potencies versus the 5-HT2BR (∆pEC50[2C-2B] ≥ 2.7). Similar to
trends observed for the compounds in Table 1, the methoxy
subsituent para (R2) to the aniline nitrogen (12a, pEC50 = 6.6)
resulted in decreased 5-HT2CR activity versus the enantiopure
parent compound (11, pEC50 = 8.6) while the para (R2) fluoro
substituent (12i, pEC50 = 8.5) had little effect. However, a small
increase in activity was observed with the meta (R3) fluoro
analog 12c (pEC50 = 9.1). Substitutions of methyl (12b) and
chloro (12h) at the position ortho (R4) to the aniline nitrogen led
to a significant decrease in 5-HT2BR signaling (Emax) as was also
observed in the 6,6,7-series. While smaller substitutions at all
aromatic positions did not result in any significant improvement
in 5-HT2CR versus 5-HT2BR selectivities, an increase in alkyl
chain length from methyl (12j, ∆pEC50[2C-2B] = 1.2) to propyl
.
BH3 THF, reflux; (g) Trimethylboroxine, cat. Pd(PPh3)4, K2CO3, dioxane, 100
°C; (h) NCS, DCM, rt; (i) Boc2O, DCM, rt; (j) Pd-catalyzed coupling
reactions, see Supplementary Material; (k) TFA, DCM, rt; (l) 4N HCl,
dioxane, rt; (m) n-BuLi, THF, -78 °C, 0.75 h then DMF; (n) NaBH4, EtOH, 0
.
°C to rt; (o) MsCl, Et3N, DCM, rt; (p) ArB(OH)2, cat. Pd(dppf)Cl2 DCM,
Na2CO3, dioxane, H2O, 90 °C; (q) bis(pinacolato)diboron, cat.
.
Pd(dppf)Cl2 DCM, KOAc, THF, 100 °C; (r) 30% aq. H2O2, THF, rt; (s)
cyclobutyl bromide, K2CO3, DMF, 65 °C; (t) PhB(OH)2, Et3N, Cu(OAc)2,
DCM, rt.
Analogs 20a-c, which contained gem-dimethyl substituents on
the saturated B-ring of the tricycle, were prepared as shown in
scheme 3. In this sequence, intramolecular reductive aminations
of nitrophenyl ketones 21a-b produced the tetrahydroquinoline
esters 22a-b. Alkylation of the nitrogen atom with ethyl
bromoacetate followed by LiAlH4 reduction gave diols 23a-b.
The C-rings are formed via reaction of 23a-b with MsCl and
DIEA, followed by amination with aqueous ammonia to deliver

(12k, ∆pEC50[2C-2B] = 3.0) at the R1 position had a significant
impact. In addition, the ∆pKi[2C-2B] and ∆pKi[2C-2A] binding
selectivities observed for 12k (1.8 and 1.2 respectively) provide
an adequate safety margin for avoiding 5-HT2BR and 5-HT2AR
agonism related side effects.21 Examination of the n-propyl
substituent at the alternate R2 (12s) and R3 positions (12t) did not
yield the same results. This discovery led us to further probe the
SARs at the R1 position. Though R1 substitutions of cyclobutyl
(12l) and phenyl (12n) did not show increases in 5-HT2CR
receptor activities or selectivities over the 5-HT2BR, extension of
these substitutions with a methylene linker (12m,o-p) did
improve the compound profiles. However, the benzyl substituent
at R1 (12o-p) led to a decrease in 5-HT2CR versus 5-HT2AR
binding (pKi) selectivities (∆pKi[2C-2A] 0.4). The
<
cyclobutylmethyl and benzyl substituents at R1 were also studied
in the 6,6,7-series (24 and 25) but the results were not as
impactful. Similar to the benzyl substituent in 6,6,6-analogs 12o-
p, the methylene linked 2-thiophene (12u) and 1,3-benzodioxole
(12v) at R1 showed good 5-HT2CR versus 5-HT2BR receptor
selectivities but suffered from decreased selectivities over the
5-HT2AR. Other R1 variants which incorporated sulfur (12r) or
oxygen (12q,w-x) linkers, were not as advantageous as the
analogous carbon linked R1 substitutions.
Table 2. 5-HT2R functional activities in intracellular IP3 accumulation assays and [125I]-DOI competition binding (pKi) data for mCPP and tetrahydroquinoline-
based tricyclic compounds 11, 12, 20, 24-25.
5-HT2C
R
5-HT2B
R
5-HT2AR
a
b
a
b
a
b
Cmpd
R1
R2
R3
R4
pEC50
Emax
pKia
pEC50
Emax
pKia
pEC50
Emax
pKia
mCPP
11
7.9 [0.1]
8.6 [0.6]
6.6 [0.7]
7.2 [0.7]
9.1 [0.2]
8.2 [0.2]
8.5 [0.4]
8.7 [0.5]
9.7 [0.2]
8.5 [0.3]
8.5 [0.2]
6.9 [0.1]
8.8 [0.2]
9.0 [0.5]
7.7 [0.6]
9.5 [0.2]
7.7 [0.3]
7.2 [0.9]
9.1 [0.7]
8.2 [0.4]
7.7 [0.1]
7.0 [3.3]
5.9 [0.4]
<6
90
97
8.1 [<0.1]
7.4 [0.6]
7.2 [0.2]
6.1 [<0.1]
7.3 [1.8]
7.5 [0.2]
8.7 [2.3]
6.9 [0.3]
7.5 [0.2]
6.7 [0.1]
7.4 [0.3]
5.7 [0.4]
5.6 [0.4]
5.6 [0.5]
6.2 [0.4]
5.9 [0.3]
7.1 [0.2]
6.2 [0.9]
6.7 [0.2]
6.4 [0.3]
5.4 [0.5]
6.1 [0.1]
5.2 [<0.1]
<5
22
28
57
10
38
12
44
65
80
112
31
72
20
32
77
46
9
8.0 [0.1]
6.6 [0.2]
12
7.6 [<0.1]
H
H
H
OMe
H
H
H
H
F
H
H
Me
H
Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
101
86
12a
12b
12c
12h
12i
H
H
H
100
94
H
H
F
H
F
H
H
H
H
H
H
H
H
H
H
H
n-Pr
H
H
H
H
96
Me
H
108
101
110
106
97
12j
n-Pr
cBu
CH2cBu
Ph
H
9.4 [0.1]
9.2 [0.1]
7.6 [0.2]
7.2 [0.2]
6.3 [0.3]
5.7 [0.2]
104
135
8.2 [0.1]
8.1 [0.1]
12k
12l
H
H
12m
12n
12o
12p
12q
12r
12s
12t
H
CH2Ph
CH2Ph
CH2OMe
SEt
H
104
102
113
101
100
87
9.2 [0.1]
9.4 [0.3]
7.4 [0.2]
7.9 [0.2]
6.9 [0.2]
7.0 [0.8]
85
8.8 [0.1]
9.2 [0.3]
F
114
H
H
H
n-Pr
H
H
Ar1
Ar2
19
24
26
9
H
106
100
98
9.5 [0.4]
9.1 [0.1]
7.6 [0.6]
6.9 [0.6]
7.1 [0.6]
6.7 [0.2]
94
64
7.6 [0.6]
8.8 [0.2]
12u
12v
12w
12x
20a
20b
20c
24
H
O-cBu
OPh
H
H
112
84
94
H
Cl
H
<6
<5
6.1 [0.2]
7.0 [1.0]
<6
84
71
8.2 [0.6]
<5
5.7 [0.3]
CH2cBu
H
F
H
H
6.9 [0.2]
116
108
CH2Ph
H
7.7 [0.1]
5.1 [0.6]
25
apEC50 and pKi values are geometric means of at least two experiments. Numbers in brackets are 95% confidence intervals. Emax % based on maximum
b
asymptote at 10 M relative to serotonin (5-HT).
Exploration of gem-dimethyl substitutions on the aliphatic B
ring revealed that compounds 20a-c displayed little or weak
agonism on the 5-HT2CR. Interestingly, the chlorinated analog
To our knowledge this represents the first example of a selective
5-HT2BR agonist and further characterization of this molecule
(20c) both in vitro and in vivo may be warranted.
20c did display appreciable agonism at the 5-HT2BR (pEC50
=
To assess both pharmacodynamic and pharmacokinetic
properties of identified potent and selective 5-HT2CR agonists
12k and 12m, their effects on acute food intake in male Sprague-
7.0, Emax = 84%) with no agonism observed at either the 5-HT2CR
or the 5-HT2AR. Good binding (pKi) selectivities were also
observed for 20c (∆pKi[2B-2C] = 2.1 and ∆pKi[2B-2A] = 2.5).

12. (a) Lee, J.; Jung, M. E.; Lee, J. Expert Opin. Ther. Pat. 2010, 20,
1429−1455. (b) Wacker, D. A.; Miller, K. J. Curr. Opin. Drug
Discovery Dev. 2008, 11, 438−45.
Dawley rats were measured (Figure 2). Oral doses of 12k (3
mg/kg) and 12m (10 mg/kg) both produced full suppression
(99%) of food intake measured at 1 h. At doses of 1 mg/kg
compound 12k produced a 66% decrease in food intake as
compared to a 27% decrease observed for 12m. In a separate
study (see Supplementary Material), the effects of compound
12m (5 mg/kg) were abrogated by preadministration (IP) with the
selective 5-HT2CR antagonist SB242084 (1 mg/kg).22 These
studies demonstrated compounds 12k and 12m were orally
bioavailable, and doses of as low as 1 mg/kg provided adequate
brain exposures to elicit 5-HT2CR mediated hypophagia.
13. www.belviq.com.
14. Smith, B. M.; Smith, J. M.; Tsai, J. H.; Schultz, J. A.; Gilson, C.
A.; Estrada, S. A.; Chen, R. R.; Park, D. M.; Prieto, E. B.;
Gallardo, C. S.; Sengupta, D.; Dosa, P. I.; Covel, J. A.; Ren, A.;
Webb, R. R.; Beeley, N. R. A.; Martin, M; Morgan, M.; Espitia,
S.; Saldana, H. R.; Bjenning, C.; Whelan, K. T.; Grottick, A. J.;
Menzaghi, F.; Thomsen, W. J. J Med. Chem. 2008, 51, 305-313.
15. (a) Di Giovanni, G.; De Deurwaerdere, P. Pharmacol.Ther. 2016,
157, 125-162. (b) Higgins, G.A., Sellers, E.M., Fletcher, P.J.
Trends Pharmacol. Sci. 2013, 34, 560-570. (c) Higgins, G. A.;
Fletcher, P. J. ACS Chem. Neurosci. 2015, 6, 1071-1088.
16. Schrader, T. O.; Kasem, M.; Sun, Q.; Wu, C.; Ren, A.; Semple, G.
Tetrahedron Lett. 2016, 57, 4730-4733.
17. The IP3 accumulation assays were performed in stably-transfected
HEK293 cell lines with low expression (low density) of 5-HT2
receptors and no detectable receptor reserve effects.
The
significance of receptor reserve effects in 5-HT2 expressing cell
lines are addressed here: (a) Cavero, I.; Guillon J. M. J.
Pharmacol. Toxicol. Methods 2014, 69, 150–61. (b) Unett, D. J.;
Gatlin, J.; Anthony, T. L.; Buzard, D. J.; Chang, S.; Chen, C.;
Chen, X.; Dang, H. T.; Frazer, J.; Le, M. K.; Sadeque, A. J.; Xing,
C.; Gaidarov, I. J. Pharmacol. Exp. Ther. 2013, 347, 645–659.
18. Bernotas, R. C. Synlett 2004, 2165-2166.
19. Competition binding (pKi) studies were performed with
[
125I]-
DOI as radioligand using HEK293 cells stably expressing
recombinant human 5-HT2 receptors. Experiments performed
with [3H]-serotonin produced variable results, where often
compounds did not displace or only partially displaced the
radioligand. This phenomena was observed across all 5-HT2R
subtypes and the results were largely inconsistent.
Figure 2. Acute food intake in rat. Data represent mean ± S.E.M. Numbers
above bar (treatment groups) represent
% inhibition versus vehicle.
**Significantly different as compared to vehicle (p <0.001).
20. The differential 5-HT2 selectivities of evaluated compounds are
lower when evaluated in [125I]-DOI (pKi) binding studies rather
than functional studies. For a discussion see ref. 14a.
In conclusion, the syntheses, SARs, and biological activities
of a series of tetrahydroquinoline-based tricyclic amines as
21. In vitro pharmacological characterization data for lorcarserin is
provided here: (a) Thomsen, W. J.; Grottick, A. J.; Menzaghi, F.;
Reyes-Saldana, H.; Espitia, S.; Yuskin, D.; Whelan, K.; Martin,
M.; Morgan, M.; Chen, W.; Al-Shamma, H.; Smith, B.; Chalmers,
D.; Behan, D. J. Pharmacol. Exp. Ther. 2008, 325, 577−587. For
clinical data related to the safety and efficacy of lorcaserin, see:
(b) Smith, S. R.; Weissman, N. J.; Anderson, C. M.; Sanchez, M.;
Chuang, E.; Stubbe,S.; Bays, H.; Shanahan, W. R. N. Eng. J. Med.
2010, 363, 245-256.
5-HT2CR receptor agonists was reported.
An early lead
containing a novel 6,6,7-ring system was optimized for in vitro
potency as well as selectivity versus the related 5-HT2BR and
5-HT2AR. Ultimately, two potent, selective, orally bioactive
6,6,6-tricyclic 5-HT2CR agonists were identified.
evaluation of these and other structurally related molecules will
be disseminated in future publication(s).
Further
22. Kennett, G. A.; Wood, M. D.; Bright, F.; Trail, B.; Riley, G.;
Holland, V.; Avenell, K. Y.; Stean, T.; Upton, N.; Bromidge, S.;
Forbes, I. T.; Brown, A. M.; Middlemiss, D. N.; Blackburn, T. P.
Neuropharmacology 1997, 36, 609-620.
References and notes
1.
(a) Burke, L. K.; Heisler, L. K. J. Neuroendocrinol. 2015, 27, 389-
398. (b) Smith, B. M.; Thomsen, W. J.; Grottick, A. J. Expert
Opin. Invest.Drugs 2006, 15, 257−266. (c) Bickerdike, M. J.;
Vickers, S. P.; Dourish, C. T. Diabetes, Obes. Metab. 1999,1, 207-
214.
Supplementary Material
Supplementary data (experimental procedures and compound
characterization data) associated with this article can be found, in
the online version, at: .
2.
3.
4.
Vickers,S. P.; Dourish, C. T. Curr. Opin. Invest. Drugs 2004, 5,
377–88.
Vickers, S. P.; Dourish, C.T.; Kennett, G.A. Neuropharmacology,
2001, 41, 200-209.
Goodall, E. M.; Cowen, P. J.; Franklin, M.; Silverstone, T.
Psychopharmacology 1993, 112, 461-466.
5.
6.
Bickerdike, M. J. Curr. Top. Med. Chem. 2003, 3, 885-897.
(a) Tecott, L. H.; Sun, L. M.; Akana, S. F.; Strack, A. M.;
Lowenstein, D. H.; Dallman, M. F.; Julius, D. Nature 1995, 374,
542-546. (b) Vickers, S. P.; Clifton, P. G.; Dourish, C. T.; Tecott,
L. H. Psychopharmacology 1999, 143, 309-314.
7.
8.
http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInfo
rmationforPatientsandProviders/ucm179871.htm.
Connolly, H. M.; Crary, J. L.; Mcgoon, M. D.; Hensrud, D. D.;
Edwards, B. S., Edwards, W. D. Schaff, H. V. N. Engl. J. Med.
1997, 337, 581-588.
9.
(a) Rothman, R. B.; Baumann, M. H.; Savage, J. E.; Rauser, L.;
McBride, A.; Hufeisen, S. J.; Roth, B. L. Circulation 2000, 102,
2836-284. (b) Fitzgerald, L. W.; Burn, T. C.; Brown, B. S.;
Patterson, J. P.; Corjay, M. H.; Valentine, P. A.; Sun, J. H.; Link,
J. R.; Abbaszade, I.; Hollis, J. M.; Largent, B. L.; Hartig, P. R.;
Hollis, G. F.; Meunier, P. C.; Robichaud, A. J.; Robertson, D. W.
Mol. Pharmacol. 2000, 57, 75−81.
10. Nichols, D. Pharmacol. Ther. 2004, 101, 131-181.
11. Dawson, P.; Moffatt, J. D. Prog. Neuropsychopharmacol. Biol.
Psychiatry 2012, 39, 244-252.