Synthesis and SAR study of 4-arylpiperidines and 4-aryl-1,2,3,6- tetrahydropyridines as 5-HT2C agonists

Article abstract of DOI:10.1016/j.bmcl.2012.01.122

A series of substituted 4-arylpiperidines and a smaller family of 4-aryl-1,2,3,6-tetrahydropyridines were synthesized and their biological activity at the 5-HT_2C receptor studied to determine whether either series showed noteworthy agonist activity. Structure-activity relationships were developed from the performed receptor binding assays and functional studies, and the results of the analysis are presented herein.

Full text of DOI:10.1016/j.bmcl.2012.01.122

Bioorganic & Medicinal Chemistry Letters 22 (2012) 2560–2564
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and SAR study of 4-arylpiperidines and
4-aryl-1,2,3,6-tetrahydropyridines as 5-HT2C agonists
Richard J. Conway a, Celine Valant b, Arthur Christopoulos b, Alan D. Robertson c, , Ben Capuano a,
,
⇑
Ian T. Crosby a,à,
⇑
a Medicinal Chemistry and Drug Action, Monash Institute of Pharmaceutical Science, 381 Royal Parade, Parkville 3052, Australia
b Drug Discovery Biology, Monash Institute of Pharmaceutical Science, 381 Royal Parade, Parkville 3052, Australia
c AMRAD Operations Pty Ltd, 576 Swan Street, Richmond, Victoria 3121, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of substituted 4-arylpiperidines and a smaller family of 4-aryl-1,2,3,6-tetrahydropyridines were
synthesized and their biological activity at the 5-HT2C receptor studied to determine whether either ser-
ies showed noteworthy agonist activity. Structure–activity relationships were developed from the per-
formed receptor binding assays and functional studies, and the results of the analysis are presented
herein.
Received 20 December 2011
Revised 25 January 2012
Accepted 30 January 2012
Available online 9 February 2012
Crown Copyright Ó 2012 Published by Elsevier Ltd. All rights reserved.
Keywords:
5-HT2C agonists
Synthesis
SAR
Anti-obesity
The compound meta-chlorophenylpiperazine (mCPP, 1, Fig. 1) is
a well known 5-HT2C receptor agonist with a proven ability to sup-
press appetite in a variety of animals including man.1–6 While it is
used regularly as a control in animal feeding trials, mCPP (1) is not
sufficiently selective for the 5-HT2C receptor to be used safely as a
therapeutic drug.7
In the 1990s, a number of simple analogues (2, 3 and 4, Fig. 1) of
mCPP (1) were reported to show modest selectivity for the 5-HT2C
receptor over 5-HT2A and 5-HT2B receptors.7,8 More recently, struc-
turally elaborated molecules (5, 6 and 7, Fig. 1) containing a central
arylpiperazine core with significantly enhanced 5-HT2C selectivity
compared to mCPP (1) have been reported.9
These observations prompted our interest in preparing selective
5-HT2C agonists as an approach towards potential therapeutics for
the treatment of obesity. We chose to explore the hypothesis that
mCPP (1) analogues lacking a piperazine core, substituted 4-arylpi-
peridines or 4-aryl-1,2,3,6-tetrahydropyridines, could exhibit suit-
able agonist activity at the 5-HT2C receptor to be used as scaffolds
for the subsequent preparation of selective appetite suppressants.
The 4-arylpiperidine series, mainly substituted 4-phenylpiperi-
dines, were prepared using the synthetic chemistry shown in
Scheme 1. The use of two equivalents of freshly prepared Grignard
reagent in step (i) generally provided the required piperidinols in
78–95% yield. Two Grignard reactions using compounds 18 and
21 resulted in significantly lower yields (29% and 17%, respectively).
The piperidinols prepared could be dehydrated in step (ii) in good
yield (57–89%) using either aqueous hydrochloric acid (6 M) or
methanesulfonic acid and toluene. Step (iv), a combined hydroge-
nation and hydrogenolysis reaction, was accomplished in good
yield (80–95%) on the hydrochloride salts of 8b–17b and 21b–
24b under standard hydrogenation conditions. Interestingly,
hydrogenation and hydrogenolysis of the 1-naphthyl analogue
(26b) gave a complex mixture of products from which small
amounts of two compounds (26d and 27d) were isolated as trifluo-
roacetate salts using preparative HPLC.
The use of a
-chloroethyl chloroformate (ACECl),10 in step (iii) of
the scheme, facilitated the preparation of four 4-aryl-1,2,3,6-tetra-
hydropyridinium chloride analogues (9c, 17c, 20c and 21c) in good
yields (71–81%). Compound 28c was prepared by dehydrating
commercially available 4-(4-bromophenyl)piperidin-4-ol hydro-
chloride (28a). Compound 29c was prepared from indole and 4-
piperidone monohydrate hydrochloride using the method of
Fonquerna et al.11 Hydrogenation of 29c provided 29d in accept-
able yield (69%). Careful hydrogenation (using smaller quantities
of catalyst and time) of tetrahydropyridines (19c and 20c) in step
⇑
Corresponding authors.
E-mail addresses: ben.capuano@monash.edu (B. Capuano), i.crosby@latrobe.
edu.au (I.T. Crosby).
Present address: Pharmaxis Ltd, 2/10 Rodborough Road, Frenchs Forest, New
South Wales 2086, Australia.
à
Present address: Dept of Chemistry, La Trobe Institute for Molecular Science, La
Trobe University, Bundoora, Victoria 3086, Australia.
0960-894X/$ - see front matter Crown Copyright Ó 2012 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2012.01.122

R. J. Conway et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2560–2564
2561
Cl
F3C
Cl
Cl
N
N
N
NH
N
NH
N
NH
F3C
Cl
N
NH
N
1
2
3
4
N
O
N
N
NH
N
NH
N
NH
F3C
N
O
F
O
N
O
5
6
7
Figure 1. Appetite suppressant meta-chlorophenylpiperazine (mCPP, 1), early analogues (2, 3 and 4) with modest selectivity for the 5-HT2C receptor and more recent
analogues (5, 6 and 7) with enhanced selectivity for the 5-HT2C receptor reported.
R
9
10
(R = H)
(R = 4-Me)
11 (R = 3-Me)
R
9c, 17c,
12 (R = 2-Me)
19c∗, 20c, 21c#
R
R
13
14
(R = 4-OMe)
(R = 3-OMe)
(iii)
(iv)
Br
15 (R = 2-OMe)
16 (R = 4-CF3)
N
+
Cl
H2
(i)
(ii)
OH
17
18
(R = 3-CF3)
(R = 2-CF3)
O
(v)
N
20 (R = 3-Cl)
21 (R = 2-Cl)
N
R
N
22
23
(R = 4-F)
(R = 3-OH)†
24 (R = 2,3-diMe)
25 (R = 4-NMe2)
9a - 18a,
20a - 26a
9b - 18b
8
20b - 26b
26
(R = 2,3-(CH)4-)
N
Cl
H2
Br
N
Br
9d - 19d
22d - 25d
(ii)
20d
#, 26d#^
27d#^ (R = 2,3-(CH2)4-)
OH
N
H2
28a
H2
28c
Cl
Cl
H
N
H
N
H
N
29
30
(iv)
(vi)
+
HO OH
N
N
H
H2
Cl
N
H2
29c
29d
Cl
Scheme 1. Reagents and conditions: (i) Mg, THF (ii) 6 M HCl (aq) or methanesulfonic acid/toluene(iii) (a) ACECl, DCM; (b) MeOH (iv) H2, 10% Pd/C (0.1–0.2 equiv), HCl
(1.0 equiv), MeOH/EtOH (v) H2, 10% Pd/C (0.05 equiv), MeOH/EtOH, (vi) (a) KOH, MeOH; (b) HCl (1.0 equiv). Notes; 23 (O-Bn protected), ⁄19c (R = 4-Cl) purchased from
Aldrich Chemicals, #isolated as trifluoroacetate salts after preparative HPLC, ^26d and 27d isolated from hydrogenation/hydrogenolysis of 26b.
(v) furnished three target 4-(chlorophenyl)piperidines (19d and
20d) in acceptable yields.
ligand affinity (expressed as a dissociation constant, Ki) values
(Table 1).
The series of compounds prepared were initially assayed using a
radioligand binding study, with screening carried out at concentra-
The synthesized 4-aryl-1,2,3,6-tetrahydropyridines were tested
in the same binding affinity study and the data obtained was used
to calculate ligand affinity (Ki) estimates (Table 2).
A review of the ligand affinity (Ki) estimates for the piperidine
series revealed that all three of the bicyclic compounds prepared
tions of 1 and 10 lM (in duplicate) to determine the percentage
inhibition (% I) of binding of [3H]-mesulergine at each concentra-
tion of test compound. The data generated were used to estimate

2562
R. J. Conway et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2560–2564
Table 1
Percent radioligand inhibition (10
l
M and 1
l
M) and estimated ligand affinity (Ki) values of 4-arylpiperidines prepared
X
R
N
H
H
N
H
H
Cl (or CF3CO2 *)
% Ia 1b
M)
(or CF3CO2 *)
Cl
Compds
X/R
% Ia 10b
(
lM)
(l
Estimated Kic (nM)
5-HT
9d
97 ( 2)
16 ( 2)
33 ( 2)
63 ( 2)
65 ( 10)
9 ( 5)
47 ( 1)
74 ( 1)
37 ( 3)
77 ( 1)
85 ( 5)
26 ( 3)
83 ( 3)
18 ( 1)
11 ( 4)
82 ( 1)
16 ( 7)
98 ( 1)
98 ( 1)
92 ( 3)
71 ( 5)
6 ( 5)
9 ( 6)
15 ( 3)
10 ( 5)
3 ( 1)
214 ( 20)
na
na
X = H
10d
11d
12d
13d
14d
15d
16d
17d
18d
19d
20d
22d
23d
24d
25d
26d
27d
29d
X = 4-Me
X = 3-Me
X = 2-Me
X = 4-OMe
X = 3-OMe
X = 2-OMe
X = 4-CF3
X = 3-CF3
X = 2-CF3
X = 4-Cl
3,130 ( 130)
3,170 ( 500)
na
6,190 ( 410)
1,800 ( 50)
8,320 ( 1400)
1,550 ( 35)
826 ( 65)
na
931 ( 70)
na
na
1,160 ( 30)
na
60 ( 3)
55 ( 5)
366 ( 24)
5 ( 4)
23 ( 2)
16 ( 6)
26 ( 1)
40 ( 2)
3 ( 2)
37 ( 4)
4 ( 7)
0 ( 5)
31 ( 2)
7 ( 3)
90 ( 1)
91 ( 1)
60 ( 2)
X = 3-Cl*
X = 4-F
X = 3-OH
X = 2,3-diMe
X = 4-NMe2
R = 4-(1-Naphthyl)*
R = 4-(5,6,7,8-Tetrahydro)-naphthalen-1-yl*
R = 4-(1H-Indol-3-yl)
a
b
c
Percentage inhibition.
Values are the mean of two experiments, standard deviation in parentheses.
Estimated Ki values are shown as na (no activity) when value is >10,000 nM.
Indicates trifluoroacetate salt.
*
Table 2
Percent radioligand inhibition (10 and 1
prepared
l
M) and estimated ligand affinity (Ki) values of 4-aryl-1,2,3,6-tetrahydropyridines
X
R
N
H
N
H
H
H
Cl
Cl
Compds
X/R
% Ia 10b
(
lM)
% Ia 1b
(
lM)
Estimated Ki (nM)
5-HT
9c
97 ( 2)
42 ( 1)
93 ( 3)
62 ( 1)
95 ( 3)
65 ( 2)
95 ( 2)
71 ( 5)
9 ( 5)
214 ( 20)
7,180 ( 550)
298 ( 22)
3100 ( 150)
280 ( 14)
2540 ( 280)
296 ( 27)
X = H
X = 3-CF3
X = 4-Cl
X = 3-Cl
X = 4-Br
17c
19c
20c
28c
29c
65 ( 3)
17 ( 3)
65 ( 2)
22 ( 5)
64 ( 5)
*
R = 4-(1H-Indol-3-yl)
a
b
*
Percentage inhibition.
Values are means of two experiments, standard deviation in parentheses.
Purchased from Sigma–Aldrich.
(26d, 27d and 29d) were more potent than any of the sixteen
substituted 4-phenylpiperidines by between a half and one order
of magnitude. Also noteworthy was the trend shown in the binding
affinity of the substituted 4-phenylpiperidines when comparing
para versus meta and ortho-substitution. In each case ortho substi-
tuted ligands had higher affinity than meta-substituted ligands,
which, in turn, had higher affinity than para-substituted phenylpi-
peridines. This is shown in Figure 2 with a comparison of the Ki
estimates derived from competition binding studies with 4-, 3-
and 2-trifluoromethyl 4-phenylpiperidines (16d, 17d and 18d,
respectively).
A review of the Ki estimates for the smaller number of 1,2,3,6-
tetrahydropyridine compounds tested (Table 2) was also impor-
tant. For the phenyl compounds, the tetrahydropyridine analogue

R. J. Conway et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2560–2564
2563
Table 3
CF3
Ki values of selected lead 4-arylpiperidines and 4-aryl-1,2,3,6-tetrahydropyridines
derived from complete competition binding curves
CF3
CF3
Compds
Structures
Kia (nM)
CF3
Cl
H
N
N
H
N
H
N
H
N
H
Cl
5-HT
445 ( 20)
Cl
Cl
H
H
H
H
NH2
HO
Cl
16d, 8,320 nM
17d, 1550 nM
18d, 826 nM
17c, 298 nM
Figure 2. Estimated Ki values for compounds 16d, 17d and 18d derived from
competition binding studies with [3H]mesulergine.
20d
26d
1430 ( 140)
N
H
H
CF3CO2
72 ( 3)
N
H
H
CF3CO2
27d
29d
43 ( 4)
N
H
CF3CO2
H
HN
Figure 3. Comparison of calcium mobilization (as a percentage of 5-HT Emax) for
tetrahydropyridine compounds 19c, 20c and 21c.
612 ( 37)
H
H
N
Cl
Cl
displayed measurably higher (>3-fold) affinity for the 5-HT2C
receptor than the corresponding piperidine analogue in each case.
The Ki estimates from the competition binding studies of 4-(3-tri-
fluoromethylphenyl)piperidine (17d) compared to 4-(3-trifluoro-
methylphenyl)-1,2,3,6-tetrahydropyri-dine (17c) are also shown
in Figure 2.
More accurate binding Ki values were determined for selected
compounds using 11 concentrations of each ligand (two repli-
cates). Assay results are displayed in Table 3.
20c
318 ( 32)
N
N
H
H
Cl
CF3
17c
29c
365 ( 22)
Competition binding studies provide a useful indication of likely
biological activity; however, they cannot readily discriminate ago-
nist from antagonist compounds. To ascertain the functional activ-
ity of the full series of compounds prepared, calcium mobilization
studies were performed, with the results (compound potency) dis-
played in Tables 4 and 5.
It was noteworthy that the calcium mobilization assays for the
4-aryl-piperidines revealed that the two naphthyl analogues (26d
and 27d) exhibited no functional activity at the 5-HT2C receptor.
This suggested the compounds were either antagonists or inverse
agonists at the receptor.
H
H
Cl
HN
632 ( 53)
H
N
H
Cl
⁄
Indicates trifluoroacetate salt.
a
Values are means of two experiments using 11 concentrations of ligand, stan-
dard deviation in parentheses.
Also of note was the fact that only meta-substituted 4-phenylpi-
peridines exhibited any functional activity in the calcium mobiliza-
tion assays. The calcium mobilization results for the
4-aryltetrahydropyridine series (Table 5) also revealed interesting
findings, which are discussed below.
4-(3-Chlorophenyl)-1,2,3,6-tetrahydropyridine (20c) was noted
to exhibit the highest 5-HT2C receptor potency (EC50 of 72 nM) of
all compounds synthesized. 4-(2-Chlorophenyl)-1,2,3,6-tetrahy-
dropyridine (21c) failed to exhibit any functional activity in the
calcium mobilization assay. Assay results for each of the three
chlorophenyl 1,2,3,6-tetrahydropyridine analogues are shown in
Figure 3. The results indicate that 4-(3-chlorophenyl)-1,2,3,6-tetra-
hydropyridine (20c) is a more potent agonist at the 5-HT2C recep-
tor than the 4-chloro analogue (19c). Interestingly, the 2-chloro
analogue (21c) showed no agonist effect at the 5-HT2C receptor
suggesting that it was either an antagonist or inverse agonist as
noted for the naphthyl piperidine analogues (26d and 27d).
Considering our results, 4-(3-chlorophenyl)-1,2,3,6-tetrahydro-
pyridine (20c) is a very potent agonist at the 5-HT2C receptor and
worthy of further investigation as a potential scaffold in future
work. The marked difference in activity when compared to its

2564
R. J. Conway et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2560–2564
Table 4
piperidine analogue (20d) is noteworthy and could be due to dif-
Functional potency determinations of 4-arylpiperidinesa
ferences in the conformations of the two molecules at C4 (sp2
hybridized versus sp3 hybridized). It has been previously reported
that piperazines with an aromatic ring attached to N4 have differ-
ent nitrogen–torsion angles (ranging from 140° to 180°) rather
than the expected sp3 hybridized torsion angle (N-torsion angle
120°).12
X
R
N
H
H
N H
H
Likewise, both of the indole compounds, the piperidine (29d)
and tetrahydropyridine (29c), exhibit noteworthy activity (128
and 163 nM, respectively) to be investigated as possible scaffolds
for the preparation of selective agonists at the 5-HT2C receptor.
In addition to the main finding of this study with regards to
5-HT2C receptor agonists, the utilization of both a binding and
functional assay as part of the discovery biology process proved
valuable in identifying and validating the two naphthalene
analogues 26d and 27d, as well as 4-(2-trifluoromethyl-
phenyl)piperidine (18d), as novel antagonists.
Cl (or CF3CO2 *)
(or CF3CO2 *)
Cl
Compds
X/R
EC50 (nM)
5-HT
9d
1.63 0.17
na
na
1995 800
na
na
na
X = H
X = 4-Me
X = 3-Me
X = 2-Me
10d
11d
12d
13d
14d
15d
16d
17d
18d
19d
20d
21d
22d
23d
24d
25d
26d
27d
29d
X = 4-OMe
X = 3-OMe
X = 2-OMe
X = 4-CF3
X = 3-CF3
X = 2-CF3
X = 4-Cl
na
na
1260 540
na
na
790 127
na
na
na
na
na
na
na
Supplementary data
X = 3-Cl*
X = 2-Cl
X = 4-F
X = 3-OH
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2012.01.122.
References and notes
X = 2,3-diMe
X = 4-NMe2
R = 4-(1-Naphthyl)*
1. Kennett, G. A.; Dourish, C. T.; Curzon, G. Eur. J. Pharmacol. 1987, 141, 429.
2. Kennett, G. G. Br. J. Pharmacol. 1991, 103, 2016.
3. Hayashi, A.; Suzuki, M.; Sasamata, M.; Miyata, K. Psychopharmacology 2005,
178.
R = 4-(5,6,7,8-Tetrahydro)-naphthalen-1-yl*
R = 4-(1H-Indol-3-yl)
128 36
4. White, C. L.; Ishihara, Y.; York, D. A.; Bray, G. A. Obesity 2007, 15, 624.
5. Walsh, A.; Smith, K.; Oldman, A.; Williams, C.; Goodall, E.; Cowen, P.
Psychopharmacology 1994, 116, 120.
6. Sargent, P. A.; Sharpley, A. L.; Williams, C.; Goodall, E. M.; Cowen, P. J.
Psychopharmacology 1997, 133, 309.
a
Values are means of two calcium mobilization experiments, standard deviation
is given in parentheses, na = no activity detected.
*
Indicates trifluoroacetate salt.
7. Bickerdike, M. J. Curr. Top. Med. Chem. 2003, 3, 885.
8. Knight, A. R.; Misra, A.; Quirk, K.; Benwell, K.; Revell, D.; Kennett, G.; Bickerdike,
M. N-S Arch. Pharmacol. 2004, 370, 114.
9. Nilsson, B. M. J. Med. Chem. 2006, 49, 4023.
10. Olofson, R. A.; Martz, J. T.; Senet, J. P.; Piteau, M.; Malfroot, T. J. Org. Chem. 1984,
49, 2081.
Table 5
Functional potency determinations of 4-aryl-1,2,3,6-tetrahydropyridinesa
11. Fonquerna, S.; Miralpeix, M.; Pagès, L.; Puig, C.; Cardús, A.; Antón, F.; Cárdenas,
Á.; Vilella, D.; Aparici, M.; Calaf, E.; Prieto, J.; Gras, J.; Huerta, J. M.; Warrellow,
G.; Beleta, J.; Ryder, H. J. Med. Chem. 2004, 47, 6326.
X
12. Andrews, P. R.; Munro, S. L. A.; Sadek, M.; Wong, M. G. J. Chem. Soc., Perkin
Trans. 2 1988, 711.
R
N
H
H
N H
H
Cl (or CF3CO2 *)
Cl
Compds
X/R
EC50 (nM)
5-HT
9c
1.63 0.17
2450 470
158 28
1760 370
72.3 8.8
na
X = H
17c
19c
20c
21c
28c
29c
X = 3-CF3
X = 4-Clb
X = 3-Cl
X = 2-Cl*
X = 4-Br
R = 4-(1H-Indol-3-yl)
3520 1700
163 46
a
Values are means of two calcium mobilization experiments, standard deviation
in parentheses.
b
Purchased from Aldrich, na = no activity detected.
Indicates trifluoroacetate salt.
*