Novel carbazole derivatives as NPY Y1 antagonists

Article abstract of DOI:10.1016/j.bmcl.2006.11.034

The synthesis of a series of carbazole derivatives and their SAR at the NPY Y1 receptor is described. Modulation of physicochemical properties by appropriate decoration led to the identification of a high-affinity NPY Y1 antagonist that shows high brain penetration and modest oral bioavailability.

Full text of DOI:10.1016/j.bmcl.2006.11.034

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1043–1046
Novel carbazole derivatives as NPY Y1 antagonists
Colin P. Leslie,* Romano Di Fabio,* Francesca Bonetti, Manuela Borriello,
Simone Braggio, Giovanna Dal Forno, Daniele Donati, Alessandro Falchi,
Damiano Ghirlanda, Riccardo Giovannini, Francesca Pavone,
Angelo Pecunioso, Giorgio Pentassuglia, Domenica A. Pizzi,
Giovanna Rumboldt and Luigi Stasi
GlaxoSmithKline Medicines Research Centre, Via A. Fleming 4, Verona 37135, Italy
Received 2 October 2006; revised 10 November 2006; accepted 12 November 2006
Available online 16 November 2006
Abstract—The synthesis of a series of carbazole derivatives and their SAR at the NPY Y1 receptor is described. Modulation of
physicochemical properties by appropriate decoration led to the identification of a high-affinity NPY Y1 antagonist that shows high
brain penetration and modest oral bioavailability.
Ó 2006 Elsevier Ltd. All rights reserved.
Neuropeptide Y (NPY) is a 36 amino acid peptide which
was first isolated in 1982 from porcine brain.1 It is wide-
ly distributed in both the central nervous system (CNS)2
and in peripheral neurones3 and has a range of biologi-
cal actions. These include stimulation of food and water
intake,4 regulation of vascular tone5 and control of
mood.6 Several NPY receptor subtypes have been char-
acterised at a molecular level (Y1, Y2, Y4, Y5 and Y6)
but to date it has not been possible to unambiguously
ascribe the role of each subtype in the pharmacological
responses to NPY.
described the preparation and structure–activity rela-
tionships (SAR) of a series of tetrahydrocarbazole
NPY Y1 antagonists.12 Herein, we disclose a comple-
mentary series of highly brain penetrant, bioavailable
carbazole derivatives that are potent and selective
NPY Y1 antagonists.
The core carbazole scaffold was synthesised as outlined
in Scheme 1. Oxidation of the functionalised toluene 1,
followed by esterification, gave 2. The aryloxy group
was introduced by nucleophilic aromatic substitution
to give 3. Reduction of the nitro group was followed
by iodination to give the key aniline intermediate 5.
Carbazole formation was accomplished via sequential
palladium-catalysed Suzuki coupling, with 2-bro-
mophenylboronic acid, and intermolecular Buchwald
amination to give 7. Deprotonation of 7 with sodium
hydride or cesium carbonate followed by treatment with
a suitable alkylating agent gave 8. Ester hydrolysis was
followed by amide coupling and any eventual deprotec-
tions to give final products 9. For convenience, the order
in which the two points of diversity were elaborated was
inverted (hydrolysis and amide formation prior to alkyl-
ation) when exploring the SAR of the N-carbazole
sidechain.
Over the past decade many researchers have strived to
discover potent selective NPY Y1 receptor antagonists
in order to elucidate the role of this receptor subtype
and evaluate the potential of such agents to treat obesity
in man. Although a number of such ligands have been
reported (BIBP3226,7 SR120819A,8 PD160170,9
LY357897,10 J-11581411), none are ideal tools as they
all share pharmacokinetic weaknesses that limit CNS
exposure following oral administration.
As part of our efforts to prepare novel NPY Y1 ligands
with improved pharmacokinetic properties, we recently
Binding affinities for all new compounds were deter-
mined by measuring their ability to displace [125I]pep-
tide YY from cloned human Y1 receptors expressed
in CHO cells. The affinity of selected compounds
Keywords: Neuropeptide Y; NPY Y1 antagonist; CNS; Obesity.
*
Corresponding authors. Tel.: +390458218546; fax: +390458218196
(C.P.L.); tel.: +3904582188879; fax: +390458218196 (R.D.F.); e-mail
addresses: colin.p.leslie@gsk.com; romano.m.di-fabio@gsk.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.034

1044
C. P. Leslie et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1043–1046
MeO2C
MeO2C
MeO2C
I
MeO2C
a,b
c
d
e
NO2
NH2
NH2
NO2
NO2
O
O
O
F
F
Cl
Cl
Cl
2
1
4
3
5
f
Br
R3R2NOC
MeO2C
MeO2C
MeO2C
i,j
h
g
N
N
N
H
NH2
R1
R1
O
O
O
O
Cl
Cl
Cl
Cl
9
8
7
6
Scheme 1. Reagents and conditions: (a) Na2Cr2O7, H2SO4, H2O, 0–100 °C, 0.5 h, 62%; (b) SOCl2, MeOH, rt, 18 h, 93%; (c) 4-chlorophenol, K2CO3,
o-xylene, reflux, 2 h, 97%; (d) Fe, NH4Cl, EtOH/H2O, reflux, 3 h, 95%; (e) IPyBF4, CF3CO2H, CH2Cl2, rt, 0.5 h, 84%; (f) 2-bromophenylboronic
acid, cat. Pd(PPh3)4, Na2CO3, toluene, 85 °C, 3 h, 99%; (g) stoich. Pd(PPh3)4, Na2CO3, toluene, reflux, 6 h, 74% ; (h) NaH or Cs2CO3, DMF, 90 °C,
R1Br or R1OMs; (i) NaOH, THF/MeOH, 60 °C, 2 h, 98%; (j) EDC, HOBt, R2R3NH, DMF, rt.
for human Y2 and Y5 receptors was measured
analogously, giving IC50 values greater that 10 lM in
all cases. In vitro functional antagonism was confirmed
for selected compounds by measuring their ability to
inhibit [35S]GTPcS binding.
B/P = 0.12). It was hypothesised that the strong basic
character of the secondary amine (calculated
pKa > 10) of the piperidinylmethyl chain was a major
contributing factor to the poor pharmacokinetic profile
of 10. Consequently, a focused exploration around the
parent piperidylmethyl sidechain was made with mod-
ifications targeting reduction of the pKa of the amine.
These studies indicate that affinity can be maintained
or even improved whilst reducing the basicity of the
nitrogen (Table 1, compounds 11 and 19). However,
it was found that affinity dropped off sharply when
the basicity was reduced below a certain limit. This
can be noted by contrasting the gain in potency upon
methylation of 10 or 12 (Table 1, compounds 11 and
13, respectively) with the drop in affinity observed
upon both methylation of 14 and cyclopropanation
of 12 (Table 1, compounds 15 and 17, respectively).
In agreement with this evidence, all attempts to replace
the amino group with alternative hydrogen bond do-
nors or acceptors led to inactive products (data not
shown). This behaviour is consistent with a cationic
charge interaction of the amine with acidic residues
in the receptor.
Clearance, half-life, volume of distribution and area un-
der the curve were measured after intravenous adminis-
tration of a 1 mg/kg DMSO/PEG400/H20 5:50:45
solution of the test compound to male Sprague–Dawley
rats. Brain penetration was determined by measuring the
blood/plasma partition ratio (B/P) 5 min after the
administration. Oral bioavailability (F) was calculated
using the normalised area under the curve measured fol-
lowing oral administration of a 5 mg/kg methocel sus-
pension of the test compound to male Sprague–
Dawley rats.
Throughout our SAR explorations special attention was
given to a number of physicochemical parameters which
we believed would be important in confering good per-
meability properties. In particular (i) the overall size of
the molecule was to be kept as low as possible, (ii) logP
should be maintained in the range of 2–6, (iii) the pKa of
any basic residues should be kept below 9, (iv) the pKa
of any acidic residues should be kept above 5, (v) the
number of rotatable bonds was to be kept as low as pos-
sible. We began by examining the SAR of the N-carba-
zole sidechain. For this investigation the amide
substitution was fixed as piperazine in light of the
SAR that had emerged around the tetrahydrocarbazole
and related scaffolds.12,13
As expected an improvement in the pharmacokinetic
parameters was seen with compounds bearing less basic
amine groups. Indeed, with other parameters effectively
constant a trend towards greater brain penetration was
observed with decreasing pKa: for example, the brain/
plasma partitions of compounds 11, 13 and 15 were
measured as 0.8, 4.2 and 8.2, respectively.
From this series of compounds the best compromise be-
tween high receptor affinity and good pharmacokinetic
properties was reached with compound 13, which offers
potent antagonist activity and good CNS exposure after
oral administration (F = 19%, B/P = 4.2). Furthermore,
The piperidinylmethyl sidechain (Table 1, 10) was
found to give high affinity whilst maintaining a rela-
tively compact molecular footprint. Unfortunately, 10
showed poor permeability properties in vivo (F = 0%,

C. P. Leslie et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1043–1046
Table 1. Structure–activity relationship of N-carbazole sidechain Table 2. Structure–activity relationship of the amide group
1045
O
O
N
R1R2N
N
N
N
R1
O
O
NH
F
Cl
Compound NR1 R2
Cl
Compound R1
NPY Y1
binding
affinity
Calculated
pKa for R1a
NPY Y1
binding
affinity
Calculated
pKa for NR1R2
IC50 (nM)
IC50 (nM)
12
51
8.3
10
16
11.0
N
N
HN
N
21
22
23
24
25
832
9772
2882
1905
7762
8.1
7.9
7.0
7.3
—
11
11
51
9.7
9.2a
7.9a
7.8a
6.5a
11.0
8.0a
7.9
N
N
HO
N
12
HN
F3C
N
N
F
F
F
F
N
13
28
N
H
N
N
N
14
53
HN
N
N
15
95
N
HO
16
107
186
43
N
H
piperazine, with most alternatives leading to significant-
ly less potent analogues. Only the open-chain derivative
28 (Table 3) maintained a similar binding affinity, sug-
gesting that the binding pocket occupied by the pipera-
zine is highly demanding in terms of the size and
orientation of the amide substitutent which it can
accommodate. Simple methylation of 28 is sufficient
to give a fourfold drop in affinity. The requirement
for a basic amine in this region was confirmed by the
inactivity of amide derivatives such as 25 and 32
(Tables 2 and 3, respectively).
17
N
F
18
HN
F
O
19
14
7.4
N
Me2N
N
20
50
9.7
a see Ref. 14.
In summary, a novel series of carbazole NPY Y1 recep-
tor antagonists were prepared and characterised. The
SARs at the N-carbazole sidechain and at the C4 amide
substituent have been explored, evidencing the require-
ment of a basic nitrogen in each position. A trend to-
wards improved bioavailability and brain penetration
was observed upon reduction of the basicity of the
amine on the N-carbazole sidechain. Reduction of
basicity below a certain limit led to a dramatic drop in
binding affinity.
the plasma clearance of 13 in rat is moderately low
(Clp = 15 mL/min/kg), leading to a long half-life (11 h)
in this species.
Our attention then turned to the exploration of the
amide substitution. During this investigation we chose
to maintain an emphasis on reducing the basicity of
the amine group in the light of the positive impact that
this strategy had made on permeability properties when
applied to the N-carbazole sidechain. The SAR was
constructed in part with the carbazole bearing the side-
chain of 12 (Table 2) and in part with the carbazole
bearing the epimeric sidechain of 14 (Table 3). Despite
an extensive exploration of the amide position, no
improvement could be made over the originally chosen
The culmination of the present study was the identifica-
tion of compound 13, a potent, orally bioavailable,
highly brain penetrant antagonist of the NPY Y1 recep-
tor. As such 13 represents an invaluable tool for study-
ing the centrally mediated effects of this receptor and
should help clarify the scope of Y1 antagonists as phar-
maceutical agents.

1046
C. P. Leslie et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1043–1046
Table 3. Structure–activity relationship of the amide group
Pickel, V. M.; Dimaggio, D. A.; Hotchkiss, A. J.; Crowly,
W. R.; Zukowska-Grojec, Z. Peptides 1985, 6, 755.
3. Tatemoto, K.; Mann, M. J.; Shimizu, M. Proc. Natl.
Acad. Sci. U.S.A 1992, 89, 1174.
O
R1R2N
4. Stanley, B. G.; Leibowitz, S. F. Proc. Natl. Acad. Sci.
U.S.A 1985, 82, 3940.
5. Boublik, J. H.; Scott, N. A.; Brown, M. R.; Rivier, J. E.
J. Med. Chem. 1989, 32, 597.
N
O
NH
F
Cl
6. Heilig, M.; Widerlov, E. Acta Psychiatr. Scand. 1990, 82,
95.
7. Rudolf, K.; Eberlein, W.; Engel, W.; Wieland, H. A.;
Willim, K. D.; Entzeroth, M.; Wienem, W.; Beck-Sickin-
ger, A. G.; Doods, H. N. Eur. J. Pharmacol. 1994, 271,
R11.
Compound
NR1 R2
NPY Y1
binding
affinity
Calculated
pKa for NR1R2
IC50 (nM)
14
53
8.3
N
N
8. Serradeil-Le Gal, C.; Valette, G.; Rouby, P.-E.; Pellet, A.;
Oury-Donat, F.; Brossard, G.; Lepsy, L.; Marty, E.;
Neliat, G.; Conite, P.; Maffrand, J.-P.; Le Fur, G. FEBS
Lett. 1995, 362, 192.
9. Wright, J.; Bolton, G.; Creswell, M.; Downing, D.;
Georgic, L.; Heffner, T.; Hodges, J.; MacKenzie, R.;
Wise, L. Bioorg. Med. Chem. Lett. 1996, 6, 1809.
10. Hipskind, P. A.; Lodd, K. L.; Nixon, J. A.; Britton, T. C.;
Bruns, R. F.; Catlow, J.; Dieckman-McGinty, D. K.;
Gackenheimer, S. L.; Gitter, B. D.; Iyengar, S.; Schober,
D. A.; Simmons, R. M. A.; Swanson, S.; Zarrinmayeh, H.;
Zimmerman, D. M.; Gehlert, D. R. J. Med. Chem. 1997,
40, 3712.
11. Kanatani, A.; Hata, M.; Mashiko, S.; Ishihara, A.;
Okamoto, O.; Haga, Y.; Ohe, T.; Kanno, T.; Murai, N.;
Ishii, Y.; Fukuroda, T.; Fukami, T.; Ihara, M. Mol.
Pharmacol. 2001, 59, 501.
12. Di Fabio, R.; Giovannini, R.; Bertani, B.; Borriello, M.;
Bozzoli, A.; Donati, D.; Falchi, A.; Ghirlanda, D.; Leslie,
C. P.; Pecunioso, A.; Rumboldt, G.; Spada, S. Bioorg.
Med. Chem. Lett. 2006, 16, 1749.
N
26
27
28
29
30
31
32
871
467
8.2
9.0
8.9
9.0
9.0
10.3
—
N
N
N
N
50
N
H
N
191
N
N
H
263
N
1862
12022
N
N
H
13. Unpublished results, GSK.
14. pKa’s were calculated using ACD/Labs software (http://
www.acdlabs.com). As this software does not discriminate
between the axial and equatorial fluoropiperidine isomers,
pKa’s for compounds 12, 13, 14, 15 and 17 were calculated
by correcting the values for the corresponding non-
fluorinated piperidines according to van Niel, M. B.;
Collins, I.; Beer, M.S.; Broughton, H. B.; Cheng, S. K. F.;
Goodacre, S. C.; Heald, A.; Locker, K. L.; MacLeod, A.
M.; Morrison, D.; Moyes, C. R.; O’Connor, D.; Pike, A.;
Rowley, M.; Russel, N.; Sohal, B.; Stanton, J. A.;
Thomas, S.; Verrier, H.; Watt, A. P.; Castro, J. L.
J. Med. Chem. 1999, 42, 2087 (À1.83 pKa units for an axial
fluorosubstituent and À3.24 pKa units for an equatorial
fluorosubstituent). The authors believe that the calculated
pKa for compound 17 may be an overestimation as
experimentally determined pKa values of cyclopropanated
amines can show a 2 pKa unit drop compared to the non-
alkylated amines (for example, see Gillaspy, M. L.; Lefker,
B. A.; Hada, W. A.; Hoover, D. J. Tetrahedron Lett. 1995,
41, 7399).
N
Acknowledgments
The authors thank the Computational, Analytical and
Structural Sciences group, the In Vitro Pharmacology
and the DMPK laboratories of GlaxoSmithkline SpA,
Verona, Italy, for their assistance in the characterisation
of the compounds described herein.
References and notes
1. Tatemoto, K.; Carlquist, M.; Mutt, V. Nature 1982, 296,
659.
2. O’Donohue, T. L.; Chronwall, B. M.; Pruss, R. M.;
Mezey, E.; Kiss, J. Z.; Eiden, L. E.; Massari, J.; Tessel, E.;