Aminoalkoxybenzyl pyrrolidines as novel human urotensin-II receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2005.04.074

High throughput screening of the corporate compound collection led to the discovery of a novel series of substituted aminoalkoxybenzyl pyrrolidines as human urotensin-II receptor antagonists. The synthesis, initial structure-activity relationships, and optimization of the initial hit that led to the identification of a truncated sub-series, represented by SB-436811 (1a), are described.

Full text of DOI:10.1016/j.bmcl.2005.04.074

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3229–3232
Aminoalkoxybenzyl pyrrolidines as novel human
urotensin-II receptor antagonists
Jian Jin,a,* Dashyant Dhanak,b Steven D. Knight,b Katherine Widdowson,c Nambi Aiyar,d
Diane Naselsky,d Henry M. Sarau,e James J. Foley,e Dulcie B. Schmidt,e Carl D. Bennett,f
Bing Wang,f Gregory L. Warren,f Michael L. Moore,a Richard M. Keenan,b
Ralph A. Riveroa and Stephen A. Douglasd
aHigh Throughput Chemistry, GlaxoSmithKline, 1250 South Collegeville Road, Collegeville, PA 19426, USA
bMedicinal Chemistry, Microbial, Musculoskeletal and Proliferative Diseases CEDD, GlaxoSmithKline,
1250 South Collegeville Road, Collegeville, PA 19426, USA
cMedicinal Chemistry, GlaxoSmithKline, 709 Swedeland Road, King of Prussia, PA 19406, USA
dBiology, Cardiovascular and Urogenital CEDD, GlaxoSmithKline, 709 Swedeland Road, King of Prussia, PA 19406, USA
eBiology, Respiratory and Inflammation CEDD, GlaxoSmithKline, 709 Swedeland Road, King of Prussia, PA 19406, USA
fComputational, Analytical and Structural Sciences, Discovery Research, GlaxoSmithKline,
1250 South Collegeville Road, Collegeville, PA 19426, USA
Received 16 March 2005; revised 25 April 2005; accepted 29 April 2005
Abstract—High throughput screening of the corporate compound collection led to the discovery of a novel series of substituted
aminoalkoxybenzyl pyrrolidines as human urotensin-II receptor antagonists. The synthesis, initial structure–activity relationships,
and optimization of the initial hit that led to the identification of a truncated sub-series, represented by SB-436811 (1a), are
described.
Ó 2005 Elsevier Ltd. All rights reserved.
Urotensin-II (U-II), a cyclic undecapeptide, was first
isolated in the 1960s from goby urophysis1 where it
has been proposed to be involved primarily in osmo-
regulation.2 Human urotensin-II (hU-II) and other
orthologs including mouse, rat, and monkey were subse-
quently identified and cloned.3 In 1999, hU-II was iden-
tified as a cognate ligand of human GPR-14 (hUT), an
‘‘orphan’’ 7TM receptor predominantly expressed in
the cardiovascular tissue.4 hU-II was found within vas-
cular and cardiac tissue and effectively constricted iso-
lated arteries from non-human primates.4 The potency
of hU-II as a vasoconstrictor was 10-times greater than
that of endothelin-1, making hU-II the most potent
mammalian vasoconstrictor identified to date.4 More
recently, hU-II was found to induce profound cardio-
hemodynamic effects upon systemic administration in
cat5 and in human.6 It also influenced the cardiorenal
function by acting as a potent regulator of cardiac con-
tractility7 and as a natriuretic factor.8 hU-II and hUT
are therefore proposed to be involved in the (dys)regula-
tion of cardiorenal function9, and have been implicated
in the etiology of numerous cardiorenal and metabolic
diseases including hypertension,10 heart failure,7,11
atherosclerosis,12 renal failure,13 and diabetes.14
Several non-peptidic UT ligands have recently been re-
ported.15 Herein, we describe the identification, synthe-
sis, and initial structure–activity relationships (SAR) of
a novel series of substituted 3-amino-N-(alkoxybenz-
yl)pyrrolidines as hUT antagonists. Optimization of
the initial screening hit ultimately led to the identifica-
tion of
SB-436811 (1a).
a
truncated sub-series, represented by
Keywords: Urotensin-II receptor antagonist; Aminoalkoxybenzyl
pyrrolidines.
High throughput screening (HTS) of our in-house com-
pound collection using a fluorometric imaging plate
reader (FLIPR) assay (measuring the inhibition of
*
Corresponding author. Tel.: +1 610 917 6645; fax: +1 610 917
7391; e-mail: jian.jin@gsk.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.04.074

3230
J. Jin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3229–3232
hU-II-mediated [Ca2+]i mobilization in HEK293 cells
expressing human recombinant UT receptor)16 led to
the identification of SB-328872 (2a), a single diastereo-
isomer (vide infra), as an antagonist with a pIC50 of
6.2 (Fig. 1). The compound was subsequently found to
have good hUT receptor binding affinity with a pKi of
6.9 in a [125I]hU-II radioligand binding assay using
HEK293 cell membranes stably expressing human
recombinant UT receptors.16,17 For comparison, hU–
II had a pEC50 of 9.2 in the FLIPR assay and a pKi
of 8.7 in the radioligand binding assay.4
removal, to produce resin-bound amines 6. Intermedi-
ates 6 were then coupled with various acids, followed
by nosyl-group removal, to afford resin-bound interme-
diates 7. Reductive amination of intermediates 7 and
subsequent resin cleavage produced the targeted com-
pounds 2 in good yield and purity. This solid-phase syn-
thesis was readily amenable to rapid array-based
optimization. Using the synthetic route, multiple regions
of this chemical series were explored to improve potency
and demonstrate tractable SAR.
The preferred stereochemistry at the 3-position of the
pyrrolidine and the amino acid chiral center was first
investigated by preparing the corresponding diastereo-
isomerically pure22 compounds, 2a (3S, 30S), 2b (3R,
30S), 2c (3S, 30S), 2d (3R, 30S), and 2e (3S, 30R), from com-
mercially available 3(S)- and 3(R)-(tert-butoxycarbonyl-
amino)pyrrolidine (3), and L- and D-Fmoc-Leu-OH. As
shown in Table 1, the preferred stereochemistry was
(3S, 30S). In subsequent optimization, only 3(S)-(tert-
butoxycarbonyl-amino)pyrrolidine (3) and L-amino acids
were used.
SB-328872 (2a) and its analogs were prepared using a
general solid-phase synthetic route outlined in
Scheme 1.18 Optically pure 3(S)- or 3(R)-(tert-butoxy-
carbonyl-amino)pyrrolidine (3)19 was converted to the
corresponding nosyl-protected diamine (4) HCl salts in
two steps. The diamines 4 were reacted with commer-
cially available 2,6-dimethoxy-4-polystyrenebenzyloxy-
benzaldehyde resin (DMHB resin)20 via reductive
amination to afford resin-bound amines 5. Amines 5
were coupled with various optically pure L- or D-
Fmoc-protected amino acids,21 followed by Fmoc
We next turned our attention to finding the optimum
left-hand side (LHS) moiety. A number of amide, urea,
and carbamate analogs were prepared and evaluated in
the radioligand binding assay (Table 2). In general,
modifications in this region were well tolerated with a
variety of fused bicyclic and monocyclic aromatic
groups having moderate binding affinity (pKi 6.2–6.9).
The benzothiophen-2-yl (2a) and 3,4-dichlorophenyl
(2j) groups were slightly preferred. The simple methyl
analog 2r was significantly less potent. Carbamate 2q23
was slightly more potent than its amide analog 2o, while
urea 2p23 was slightly less potent.
Figure 1. Structure of HTS hit SB-3288872 (2a).
We then investigated the central amino acid moiety (Ta-
ble 3). The SAR in this region indicated a limited degree
of tolerability for structural variation. A lipophilic
group such as isobutyl (2j), benzyl (2s), and isopropyl
(2t) was preferred, with the isobutyl group being opti-
mal. Replacing the isobutyl group with hydrogen (2u),
Table 1. Optimal stereochemistry
Scheme 1. Reagents and conditions: (a) 2-nitrobenzenesulfonyl chlo-
ride, pyridine, CH2Cl2, 0 °C to rt; (b) 4 M HCl in 1,4-dioxane, MeOH,
rt; (c) 2,6-dimethoxy-4-polystyrenebenzyloxy-benzaldehyde (DMHB
resin), Na(OAc)3BH, diisopropylethylamine, 1% of HOAc in NMP,
rt; (d) Fmoc-HNCH(R1)CO2H, 1,3-diisopropylcarbodiimide (DIC),
1-hydroxy-7-azabenzotriazole (HOAt), NMP, rt; (e) 20% of piperidine
in NMP, rt; (f) R2CO2H, DIC, HOAt, NMP, rt; (g) K2CO3, PhSH,
NMP, rt; (h) Me2N(CH2)3OPhCHO, Na(OAc)3BH, 10% of HOAc in
NMP, rt; (i) 50% of TFA in DCE, rt.
Compound
X
Diastereoisomer
hUT binding (pKi)a
2a
2b
2c
2d
2e
S
(3S, 30S)
(3R, 30S)
(3S, 30S)
(3R, 30S)
(3S, 30R)
6.9
<5.0
6.8
S
O
O
O
<5.0
5.5
a Mean of at least two determinations with standard deviation of
< 0.3.

J. Jin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3229–3232
Table 2. hUT binding affinity of LHS amide, urea, and carbamate
analogs 2f–2r
3231
Scheme 2. Reagents and conditions: (a) R3COOH, DIC, HOAt,
NMP, rt; (b) K2CO3, PhSH, NMP, rt; (c) 4-(Me2N(CH2)3O)PhCHO,
Na(OAc)3 BH, 10% of HOAc in NMP, rt; (d) 50% of TFA in DCE, rt.
Compound
R2
hUT binding (pKi)a
2a
2c
2f
Benzothiophen-2-yl
Benzofuran-2-yl
Indol-2-yl
6.9
6.8
6.3
6.2
6.3
6.4
6.9
6.7
6.4
6.2
5.7
6.3
6.0
6.6
5.6
Table 4. hUT binding affinity of truncated analogs 1a–1r
2g
2h
2i
Quinolin-2-yl
3,4-Methylenedioxyphenyl
3,4-Methylenedioxybenzyl
3,4-Dichlorophenyl
4-Bromophenyl
Thiophen-2-yl
Furan-2-yl
2j
Compound
R3
hUT binding (pKi)a
2k
2l
1a
1b
1c
1d
1e
1f
3,4-Dichlorophenyl
3,5-Dichlorophenyl
4-Chloro-3-methylphenyl
3,4-Dimethoxyphenyl
3,4-Dimethylphenyl
4-Bromophenyl
4-Bromo-3-methylphenyl
4-Cyanophenyl
6.7
5.7
2m
2n
2o
2p
2q
2r
Pyridin-4-yl
Phenyl
6.2
<5.0
<5.0
6.5
Anilino
Phenoxy
Methyl
1g
1h
1i
6.2
<5.0
<5.0
<5.0
5.6
a Mean of at least two determinations with standard deviation
of < 0.3.
4-Trifluoromethylphenyl
4-Biphenyl
Phenyl
1j
1k
1l
Thiophen-2-yl
Furan-2-yl
<5.0
<5.0
<5.0
<5.0
5.6
1m
1n
1o
1p
1q
1r
Table 3. hUT binding affinity of the middle amino acid analogs 2j,
2s–2w
Pyridin-4-yl
Methyl
Benzofuran-2-yl
3,4-Methylenedioxyphenyl
2-Naphthyl
5.6
<5.0
a Mean of at least two determinations with standard deviation of
< 0.3.
Compound
R1
hUT binding (pKi)a
modification in the truncated sub-series. For example,
while the 3,4-dichlorophenyl analog 1a and 4-chloro-
3-methylphenyl analog 1c had moderate affinity, 1d
(3,4-dimethoxyphenyl) and 1e (3,4-dimethylphenyl)
had no hUT receptor affinity. Similarly, 4-bromophenyl
analogs 1f and 1g were moderately potent, but other 4-
substituted analogs 1h, 1i, and 1j had no binding affinity.
Analogs containing an unsubstituted monocyclic aro-
matic (1k, 1l, 1m, and 1n), bicyclic aromatic (1p, 1q,
and 1r), or simple methyl group (1o) had little or no
binding affinity.
2j
Isobutyl
Benzyl
6.9
6.4
2s
2t
Isopropyl
H
6.1
2u
2v
2w
<5.0
<5.0
<5.0
4-Aminobutyl
2-Carboxyethyl
a Mean of at least two determinations with standard deviation of
< 0.3.
an aminobutyl group (2v), or a carboxyethyl group (2w)
completely abolished hUT binding affinity.
In the functional FLIPR assay, SB-436811 (1a) (Fig. 2)
was an antagonist with a pIC50 of 6.8, similar in potency
to SB-328872 (2a). The FLIPR potency of both com-
pounds was also commensurate with their receptor bind-
ing affinities further supporting the proposition that the
observed inhibition of intracellular calcium mobilization
in these cells was a result of hUT blockade. SB-436811
(1a) was also tested in monkey and rat UT binding as-
says25, and found to have moderate monkey UT binding
affinity (pKi = 6.6) but poor rat UT binding affinity
(pKi = 5.5). Although it is not clear what contributes
While exploring the central amino acid moiety, we also
prepared a number of truncated analogs (Table 4), in
which the amino acid moiety was eliminated, using the
synthetic route outlined in Scheme 2 involving the direct
coupling of resin-bound amine 5 with various acids.24
We were pleased to find that several truncated analogs,
1a, 1c, 1f, and 1g, showed moderate binding affinity with
a pKi range from 6.2 to 6.7 (Table 4). The SAR at the
LHS moiety suggested a relatively restricted scope for

3232
J. Jin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3229–3232
9. Douglas, S. A.; Dhanak, D.; Johns, D. G. Trends
Pharmacol. Sci. 2004, 25, 76.
10. (a) Matsushita, M.; Shichiri, M.; Imai, T.; Iwashina, M.;
Tanaka, H.; Takasu, N.; Hirata, Y. J. Hypertens. 2001,
19, 2185; (b) Cheung, B. M.; Leung, R.; Man, Y. B.;
Wong, L. Y. J. Hypertens. 2004, 22, 1341.
11. (a) Douglas, S. A.; Tayara, L.; Ohlstein, E. H.; Halawa,
N.; Giaid, A. Lancet 2002, 359, 1990; (b) Ng, L. L.; Loke,
I.; OÕBrien, R. J.; Squire, I. B.; Davies, J. E. Circulation
2002, 106, 2877; (c) Richards, A. M.; Nicholls, M. G.;
Lainchbury, J. G.; Fisher, S.; Yandle, T. G. Lancet 2002,
360, 545; (d) Lapp, H.; Boerrigter, G.; Costello-Boerrigter,
L. C.; Jaekel, K.; Scheffold, T.; Krakau, I.; Schramm, M.;
Guelker, H.; Stasch, J. P. Int. J. Cardiol. 2004, 94, 93.
12. (a) Bousette, N.; Patel, L.; Douglas, S. A.; Ohlstein, E. H.;
Giaid, A. Atherosclerosis 2004, 176, 117; (b) Maguire, J. J.;
Kuc, R. E.; Wiley, K. E.; Kleinz, M. J.; Davenport, A. P.
Peptides 2004, 25, 1767.
13. (a) Totsune, K.; Takahashi, K.; Arihara, Z.; Sone, M.;
Satoh, F.; Ito, S.; Kimura, Y.; Sasano, H.; Murakami, O.
Lancet 2001, 358, 810–811; (b) Shenouda, A.; Douglas, S.
A.; Ohlstein, E. H.; Giaid, A. J. Histochem. Cytochem.
2002, 50, 885; (c) Langham, R. G.; Gow, R.; Thomson, N.
M.; Dowling, J. K.; Gilbert, R. E. Am. J. Kidney Dis.
2004, 44, 826.
Figure 2. Structure of truncated analog SB-436811 (1a).
to the binding affinity difference between primate and
rodent receptors, the significant sequence differences be-
tween primate and rodent receptors3 might be a reason
for the observed receptor binding affinity difference.
In summary, a novel hU-II receptor antagonist series
represented by SB-328872 (2a) was identified via HTS.
The SAR was explored for several regions of this series
and that led to the identification of a novel truncated
sub-series represented by SB-436811 (1a). Further opti-
mization of other regions, especially the right-hand
side aminoalkoxybenzyl and central aminopyrrolidine
moieties, of both series will be the subject of future
publications.
14. (a) Totsune, K.; Takahashi, K.; Arihara, Z.; Sone, M.;
Satoh, F.; Ito, S.; Murakami, O. Clin. Sci. 2003, 104, 1; (b)
Wenyi, Z.; Suzuki, S.; Hirai, M.; Hinokio, Y.; Tanizawa,
Y.; Matsutani, A.; Satoh, J.; Oka, Y. Diabetologia 2003,
46, 972.
Acknowledgments
15. (a) Flohr, S.; Kurz, M.; Kostenis, E.; Brkovich, A.;
Fournier, A.; Klabunde, T. J. Med. Chem. 2002, 45, 1799;
(b) Croston, G. E.; Olsson, R.; Currier, E. A.; Burstein, E.
S.; Weiner, D.; Nash, N.; Severance, D.; Allenmark, S. G.;
Thunberg, L.; Ma, J.; Mohell, N.; OÕDowd, B.; Brann, M.
R.; Hacksell, U. J. Med. Chem. 2002, 45, 4950(c) For
patents, see Ref. 3.
We thank Chad Quinn and Qian Jin for providing
analytical LC/MS support.
References and notes
16. (a) For [Ca2+]i mobilization and radioligand binding assay
details, see: Dhanak, D.; Knight, S. D.; Jin, J.; Keenan, R.
M. WO Patent 2001045711-A1, 2001; Chem. Abstr. 2001,
135, 76785; (b) Dhanak, D.; Knight, S. D.; Warren, G. L.;
Jin, J.; Widdowson, K. L.; Keenan, R. M. WO Patent
2001045710-A1, 2001; Chem. Abstr. 2001, 135, 71313.
17. The radioligand binding assay was subsequently used as
the primary assay to generate SAR.
1. (a) Bern, H. A.; Lederis, K. J. Endocrinol. 1969, 45, xi; (b)
Bern, H. A.; Pearson, D.; Larson, B. A.; Nishioka, R. S.
Recent Prog. Horm. Res. 1985, 41, 533.
2. (a) Ichikawa, T.; McMaster, D.; Lederis, K.; Koyabashi,
H. Peptides 1982, 3, 859; (b) Loretz, C. A.; Howard, M.
E.; Siegel, A. J. J. Physiol. 1985, 249, G284.
3. For a recent review, see: Dhanak, D.; Neeb, M. J.;
Douglas, S. A. Annu. Rep. Med. Chem. 2003, 38, 99.
4. Ames, R. S.; Sarau, H. M.; Chambers, J. K.; Willette, R.
N.; Aiyar, N. V.; Romanic, A. M.; Louden, C. S.; Foley, J.
J.; Sauermelch, C. F.; Coatney, R. W.; Ao, Z.; Disa, J.;
Holmes, S. D.; Stadel, J. M.; Martin, J. D.; Liu, W. S.;
Glover, G. I.; Wilson, S.; McNulty, D. E.; Ellis, C. E.;
Elshourbagy, N. A.; Shabon, U.; Trill, J. J.; Hay, D. W.;
Ohlstein, E. H.; Bergsma, D. J.; Douglas, S. A. Nature
1999, 401, 282.
18. For experimental details, see Ref. 16b.
19. Purchased from TCI America, catalog number: A1172 or
A1171.
20. Purchased from Polymer Laboratories, part number:
1466-6689, 150–300 lM, 1.5 mmol/g loading.
21. Purchased from Novabiochem.
22. Confirmed via NMR studies. All new compounds in this
paper were characterized via LC/MS and/or H NMR.
1
5. Behm, D. J.; Doe, C. P.; Johns, D. G.; Maniscalco, K.;
Stankus, G. P.; Wibberley, A.; Willette, R. N.; Douglas, S.
A. Naunyn SchmiedebergÕs Arch. Pharmacol. 2004, 369,
274.
23. 2p and 2q were prepared via urea formation and carba-
mate formation of intermediate 6 with phenyl isocyanate
and phenyl chloroformate, respectively, and subsequent
reactions in Scheme 1.
6. Bo¨hm, F.; Pernow, J. Br. J. Pharmacol. 2002, 135, 25.
7. Russell, F. D.; Meyers, D.; Galbraith, A. J.; Bett, N.;
Toth, I.; Kearns, P.; Molenaar, P. Am. J. Physiol. 2003,
285, H1576.
8. Song, W.; Ashton, N.; Balment, R. J. J. Physiol. 2003,
552, P107.
24. For experimental details, see Ref. 16a.
25. For monkey and rat UT receptor radioligand binding
assay details, see: Elshourbagy, N. A.; Douglas, S. A.;
Shabon, U.; Harrison, S.; Duddy, G.; Sechler, J. L.; Ao,
Z.; Maleeff, B. E.; Naselsky, D.; Disa, J.; Aiyar, N. V. Br.
J. Pharmacol. 2002, 136, 9.