8-Piperazinyl-2,3-dihydropyrrolo[3,2-g]isoquinolines: Potent, selective, orally bioavailable 5-HT1 receptor ligands

Article abstract of DOI:10.1016/j.bmcl.2005.06.042

The novel 8-piperazinyl-2,3-dihydropyrroloisoquinoline template was synthesized in nine steps. The template was N-substituted to give a series of compounds showing binding to human cloned 5-HT_1A, 5-HT_1B and 5-HT_1D receptor

Full text of DOI:10.1016/j.bmcl.2005.06.042

Bioorganic & Medicinal Chemistry Letters 15 (2005) 4370–4374
8-Piperazinyl-2,3-dihydropyrrolo[3,2-g]isoquinolines:
Potent, selective, orally bioavailable 5-HT1 receptor ligands
Tom D. Heightman,a,* Laramie M. Gaster,a Sarah L. Pardoe,a Jean-Pierre Pilleux,b
Michael S. Hadley,b Derek N. Middlemiss,b Gary W. Price,b Claire Roberts,b
Claire M. Scott,b Jeannette M. Watson,b Laurie J. Gordon,c Vicky A. Holland,c
Jenifer Powles,c Graham J. Riley,c Tania O. Stean,d Brenda K. Trail,d Neil Upton,d
Nigel E. Austin,b Andrew D. Ayrton,e Tanya Colemand and Leanne Cutlerd
aHigh Throughput Chemistry, Discovery Research, GlaxoSmithKline, New Frontiers Science Park,
Third Avenue, Harlow Essex CM19 5AW, UK
bPsychiatry Centre of Excellence for Drug Discovery, GlaxoSmithKline, New Frontiers Science Park,
Third Avenue, Harlow Essex CM19 5AW, UK
cDiscovery Research Biology, GlaxoSmithKline, New Frontiers Science Park, Third Avenue, Harlow Essex CM19 5AW, UK
dNeurology and GI Centre of Excellence for Drug Discovery, GlaxoSmithKline, New Frontiers Science Park,
Third Avenue, Harlow Essex CM19 5AW, UK
eGlaxoSmithKline Drug Metabolism and Pharmacokinetics, Ware SG12 0DP, UK
Received 11 May 2005; revised 9 June 2005; accepted 9 June 2005
Available online 21 July 2005
Abstract—The novel 8-piperazinyl-2,3-dihydropyrroloisoquinoline template was synthesized in nine steps. The template was N-
substituted to give a series of compounds showing binding to human cloned 5-HT1A, 5-HT1B and 5-HT1D receptors with pKiꢀs great-
er than 9 and selectivities up to 1000-fold against other serotonin, dopamine and adrenergic receptors. Several compounds were
shown to possess weak partial agonist activity in cloned receptors, which translated to antagonism in in vitro studies.
Ó 2005 Elsevier Ltd. All rights reserved.
Although the precise mechanism of antidepressant ac-
tion remains poorly understood, many currently pre-
scribed treatments for depression are believed to
enhance neurotransmission by increasing extracellular
levels of serotonin (5-hydroxytryptamine, 5-HT) in the
brain.1 Neuronal 5-HT release is modulated by presyn-
aptic inhibitory 5-HT1 autoreceptors. 5-HT1A receptors
are mainly somatodendritic, and cause a reduction in the
rate of cell firing and 5-HT release when stimulated.2
5-HT1B receptors are mainly located on cell terminals,
where they mediate a reduction in the amount of 5-HT
released on each firing event.3 The role of terminal
5-HT1D receptors is less clear, although there is recent
evidence to suggest that they play a role in modulating
extracellular brain 5-HT levels in guinea pig.4 The com-
bined activation of all three receptor subtypes by extra-
cellular 5-HT thus has the potential to cause a decrease
in the amount and frequency of 5-HT released into the
synaptic cleft. Stimulation of these autoreceptors may
be responsible for the delayed onset of action of most
antidepressants which work wholly or partly by inhibit-
ing 5-HT reuptake.5 The two to four week onset time for
these antidepressants is consistent with the time taken
for these autoreceptors to desensitize in rats upon treat-
ment with fluoxetine or citalopram.6 Therefore, a drug
which antagonizes all 5-HT1 inhibitory autoreceptors
simultaneously should acutely mimic their chronic
desensitization, and thereby lead to an immediate and
sustained increase in levels of synaptic 5-HT for each
transmission event.5 Such a drug could well provide a
much needed rapidly acting antidepressant.
We have previously reported the in vitro pharmacology
of SB-272183, the first selective 5-HT1A/B/D receptor
antagonist.7 In this paper we describe the synthesis
Keywords: 5-HT1A; 5-HT1B; 5-HT1D; Receptor antagonist.
*
Corresponding author. Tel.: +44 1279 622031; fax: +44 1279
627779; e-mail: Tom.D.Heightman@gsk.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.06.042

T. D. Heightman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4370–4374
4371
and pharmacology of a new series of highly potent and
selective 5-HT1A/B/D receptor partial agonists/
antagonists.
The novel dihydropyrrolo[3,2-g]isoquinoline template
10 was proposed from a pharmacophore alignment of
the indoline SB-272183 (1) and a series of naphthylpip-
erazines,8 e.g., 2, with reported 5-HT1 pKiꢀs > 6.2.
NMe
N
Cl
Scheme 2. Reagents and conditions: (a) DMFDMA, DMF, 120 °C,
quant.; (b) 4-cyanophenyl hydrazine, EtOH, 97%; (c) NaOH aq,
dioxane.
H
O
N
O
NMe
N
N
N
N
1
2
Cl
Analogues of the N-methyl piperazine moiety were
introduced into intermediate triflate 7 and further deriv-
atized using analogous procedures to those outlined in
Scheme 1. The 4-cyanophenyl pyrazole carboxylic acid
intermediate 26 used in the synthesis of 22 was prepared
by combination of 4-cyanophenyl hydrazine with the
dimethylaminomethylene oxopentanoate ester 249 fol-
lowed by saponification, as per Scheme 2.
The synthesis is illustrated in Scheme 1. Indole-6-car-
boxylic acid (3) was converted to the N-monobenzylated
indoline 4 by dibenzylation; reduction of the five-mem-
bered ring; and saponification of the ester, all under
standard conditions.
To construct the pyridyl ring, a procedure related to the
Pomeranz–Fritsch method was used via the intermediate
2,2-dimethoxyethyl amide 5. Treatment of 5 with
Brønsted acids (H2SO4, TFA or HCl) gave rise to a mix-
ture of cyclized regioisomers, presumably by partially
favouring N-protonation of the indoline leading to
intramolecular delivery of the proton to the acetal, so
that the reactive oxonium species is formed in proximity
to the 7-position of the indoline. The use of boron triflu-
oride-diethyl etherate in refluxing THF gave a single
regioisomer 6, easily distinguished from the alternative
regioisomer by the presence of two new singlets in the
aromatic region of the 1H NMR. This is consistent with
selective boron complexation of the acetal oxygen
atoms, leading to formation of the reactive oxonium
intermediate in the less hindered region of the molecule,
and favouring reaction at the 5-position of the indoline.
Attempts to cyclize analogues of 5 with N-acyl protect-
ing groups were not successful.
Compounds were tested for their activity at 5-HT1
receptors using standard binding assays with [3H]5-
HT or [3H]8-OH-DPAT as the radioligand (Table 1).
Substituted benzamides showed clear SAR with 4-sub-
stituents showing higher affinity than two or three iso-
mers. Heterocycles were also tolerated in this position
with some loss of potency, but this was restored with
biaryl groups.
The N-methyl piperazinyl group proved relatively insen-
sitive to steric modifications: removal of the methyl
group (31) resulted in only a slight loss of potency, but
a dramatic loss of selectivity against the b2 adrenorecep-
tor (Table 2).
Increasing the size of the N-alkyl substituent caused the
affinity to diminish incrementally (32, 33). The 3,5-di-
methylpiperazine 34 showed somewhat reduced activity
particularly at the 5-HT1A receptor, but replacement of
the N-methyl piperazine with either enantiomer of diaz-
abicyclooctane was well tolerated (35, 36).
The hydroxy group of isoquinoline 6 was readily activat-
ed by triflation, and displaced by N-methyl piperazine.
Debenzylation was effected by catalytic hydrogenation
under strongly acidic conditions, giving the key interme-
diate 9, which was coupled to form amides and ureas un-
der standard conditions.
Compounds in this series showed a wide range of effica-
cies vs 5-HT for 5-HT1A and 5-HT1B receptor subtypes
Scheme 1. Reagents: (a) NaH, BnBr, NMP, 90%; (b) NaBH3CN, AcOH, 96%; (c) aq NaOH, dioxane, quant.; (d) EDC, HOBT, DMF, 98%; (e)
BF3ÆOEt2, THF, 87%; (f) Tf2O, pyridine, 63%; (g) N-methyl piperazine, 93%; (h) H2/Pd/C, MeOH, HCl, 81%; (i) RCO2H, DIC, HOBt, DMF.

4372
T. D. Heightman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4370–4374
Table 1. Receptor binding affinities of reference compound 1 and amide derivatives 11–22 in the 1-(4-methyl piperazinyl) series
Me
N
O
N
R
N
N
Compound
R
pKia
Selectivityb
5-HT1A
5-HT1B
8.1 0.1
5-HT1D
1
—
8.0 0.1
8.7 0.1
30
11
9.1 0.1
9.0 0.1
9.5 0.1
9.3 0.1
9.3 0.1
9.1 0.1
320
12
13
>1000
30
8.0 0.1
7.8 0.1
7.8 0.1
14
8.3 0.2
8.9 0.1
8.6 0.1
60
15
16
8.5 0.1
8.4 0.2
9.6 0.2
9.1 0.2
9.3 0.1
8.4 0.2
400
130
17
18
19
8.3 0.1
8.2 0.3
8.2 0.2
8.7 0.2
7.8 0.2
8.3 0.1
8.1 0.2
8.6 0.2
8.0 0.1
250
30
160
20
9.5 0.1
9.0 0.1
8.9 0.2
250
21
22
9.2 0.4
9.2 0.3
9.2 0.4
500
160
8.9 0.3
9.0 0.4
9.2 0.3
a Values are means of at least three experiments.
b Selectivity versus 5-HT2A 2B 2C
, 6 and 7; dopamine D2, D3; adrenergic a1b and b2.
,
,

T. D. Heightman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4370–4374
4373
Table 2. Receptor binding affinities of amino group analogues 31–36 from the cyanophenyl pyrazole series
Compound
R
pKia
Selectivityb
5-HT1A
5-HT1B
5-HT1D
31
32
33
34
8.9 0.3
8.6 0.3
8.8 0.1
8.4 0.1
8.1 0.1
8.6 0.2
4
320
100
320
8.5 0.1
8.2 0.1
7.5 0.1
8.6 0.1
8.1 0.1
8.1 0.1
35
36
8.8 0.1
8.9 0.1
9.0 0.2
8.7 0.1
9.0 0.1
8.5 0.2
200
100
a Values are means of at least three experiments.
b Selectivity versus 5-HT2A 2B 2C
, 6 and 7; dopamine D2, D3; adrenergic a1b
,
,
Table 4. In vivo pharmacokinetics (rat, n = 1)10
in the GTPcS functional assay7 (Table 3). For both aryl
and heteroaryl compounds, a para-substituent tended to
engender higher intrinsic activity, whilst an ortho-substi-
tuent adjacent to the linking carboxamide showed lower
intrinsic activity approaching silent antagonism, sug-
gesting a role for the torsional angle of the ring-C@O
bond. Despite this intrinsic activity in the recombinant
GTPcS assay, compounds with intrinsic activities below
ꢀ0.5 behaved as antagonists in 5-HT1A and 5-HT1B
receptor-mediated native tissue assays. Hence, the phen-
yl pyrazole 22 reversed the inhibition of rat dorsal raphe
nucleus cell firing by the 5-HT1A agonist 8-OH-DPAT
(pA2 ꢀ 7); and it potentiated the electrically stimulated
release of 5-HT from guinea pig cortex slices, which
has been previously demonstrated to be mediated by
the 5-HT1B receptor.7
Compound
CLb (ml/min/kg)
Brain:blood
Fpo (%)
11
12
16
18
19
21
22
29
19
75
39
48
32
28
1.3
0.7
0.2
0.5
0.1
1.1
0.5
NDa
ND
ND
55
ND
40
50
a Not determined.
andlimited brain penetration. In contrast, the smaller het-
erocycles showed more acceptable clearance; however
whilst the dimethyloxazole 18 showed good brain pene-
tration, the closely related methyl-trifluoromethyl pyra-
zole 19 did not. The addition of the phenyl ring restored
the brain penetration and provided compounds with
low to moderate clearance and promising oral bioavail-
ability. In particular, compound 22 showed moderate to
good bioavailability across species: rat 50%, guinea pig
20%, dog 89% and cynomolgous monkey 58%.
After intravenous infusion, compounds in this series
showed a wide range of clearance rates in rat, from less
than 25% of liver blood flow to approximately 100%,
and an equally broad range of brain penetration (Table
4).10 The 4-cyanobenzamide 12 showed a lower rate of
clearance but also lower brain penetration than the isos-
teric 4-chlorobenzamide 11, consistent with its lower lipo-
philicity. The quinolinyl amide 16 showed rapid clearance
Compound 22 was further studied in in vivo pharmaco-
dynamic models based on 5-HT1B receptor antagonism.
In rats, 22 reversed the effects of the selective 5-HT1B
receptor agonist SKF-99101 on electroshock seizure
threshold, with an oral ID50 of 3.5 1.3 mg/kg.11 In
guinea pigs, 22 reversed SKF-99101-induced hypother-
mia with an oral ID50 of 0.2 mg/kg.12 Compound 22
and related compounds are under further investigation
in preclinical models of depression.
Table 3. Partial agonist efficacy of analogues 11–14, 21 and 22 in the
5-HT1A and 5-HT1B GTPcS functional assays
Compound
GTPcS
5-HT1A
5-HT1B
11
12
13
14
21
22
0.69 0.03
0.84 0.05
Inactive
0.75 0.06
0.86 0.04
0.23 0.02
0.42 0.06
0.59 0.04
0.47 0.03
References and notes
0.42 0.04
0.54 0.05
0.33 0.03
1. Cowen, P. J. In Selective Serotonin Re-uptake Inhibitors;
Feighner, J. P., Boyer, W. F., Eds.; John Wiley & Sons:
Chichester, 1996, pp 63–86.
Values expressed as fraction of 5-HT efficacy.

4374
T. D. Heightman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4370–4374
2. (a) Craven, R.; Grahame-Smith, D.; Newberry, N. Eur. J.
Wyman, P.; Parker, C.; Price, G. W.; Middlemiss, D. N;
Br. J. Pharmacol. 2001, 133, 797.
8. Chenard, B.L.; Macor, J.E.; Segelstein, B.E.; PCT Int.
Appl. WO9421619; Chem. Abstr. 1994, 122, 314570.
9. Menozzi, G.; Mosti, L.; Schenone, P. J. Heterocycl. Chem.
1987, 24, 1669.
10. DMPK procedures. Clearance and brain penetration
determination, iv dosing in male Sprague–Dawley rats,
constant infusion over 1 h to achieve target dose of
1 mg free base/kg; oral dosing, solution or suspension
administered at target dose of 3 mg free base/kg. Blood
and brain homogenate samples analysed by HPLC/MS/
MS.
11. Procedures for electroshock seizure threshold experiments
described in: Stean, T. O.; Atkins, A. R.; Heidbreder, C.
A.; Quinn, L. P.; Trail, B. K.; Upton, N. Br. J. Pharmcol
2005, 144, 628.
Pharmacol. 1994, 271, R1; (b) Haj-Dahmane, S.; Hamon,
M.; Lanfumey, L. Neuroscience 1991, 41, 495; (c) Sprouse,
J. S.; Aghajanian, G. K. Synapse 1987, 1, 3.
3. (a) Roberts, C.; Watson, J.; Burton, M.; Price, G.; Jones,
B. Br. J. Pharmacol. 1996, 117, 384; (b) Engel, G.;
Gothert, M.; Hoyer, D.; Schlicker, E.; Hillenbrand, K.
Naunyn-Schmiederbergꢀs Arch. Pharmacol. 1986, 332, 1.
4. (a) Bonaventure, P.; Voorn, P.; Luyten, W. H. M. L.;
Jurzak, M.; Schotte, A.; Leysen, J. E. Neuroscience 1998,
9, 641; (b) Bonaventure, P.; Langlois, X.; Leysen, J. E.
Neurosci. Lett. 1998, 254, 113.
5. (a) Cryan, J. F.; Leonard, B. E. Human Psychopharma-
cology 2000, 15, 113; (b) Artigas, F. TIPS 1993, 14, 262.
6. Czachura, J. F.; Rasmussen, K. Naunyn-Schmiederbergs
Arch. Pharmacol. 2000, 362, 266; Moret, C.; Briley, M.
Eur. J. Pharmacol. 1990, 180, 351.
7. Procedures for cell firing and electrically stimulated release
described in: Watson, J.; Roberts, C.; Scott, C.; Kendall,
I.; Collin, L.; Day, N.; Harries, M. H.; Soffin, E.; Davies,
C. H.; Randall, A. D.; Heightman, T. D.; Gaster, L.;
12. Procedure for SKF-99101-induced hypothermia described
in: Hagan, J. J.; Slade, P. D.; Gaster, L.; Jeffrey, P.;
Hatcher, J. P.; Middlemiss, D. N. Eur. J. Pharmacol. 1997,
331, 169.