(Phenylpiperazinyl)cyclohexylureas: Discovery of α1a/1d-selective adrenergic receptor antagonists for the treatment of benign prostatic hyperplasia/lower urinary tract symptoms (BPH/LUTS)

Article abstract of DOI:10.1016/j.bmcl.2007.11.068

Benign prostatic hyperplasia/lower urinary tract symptoms (BPH/LUTS) can be effectively treated with α₁ adrenergic receptor antagonists. Unfortunately, currently marketed α₁ blockers produce CV-related side effects that are caused by

Full text of DOI:10.1016/j.bmcl.2007.11.068

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 640–644
(Phenylpiperazinyl)cyclohexylureas: Discovery
of a1a/1d-selective adrenergic receptor antagonists
for the treatment of benign prostatic hyperplasia/lower
urinary tract symptoms (BPH/LUTS)
George Chiu,*, Shengjian Li,à Peter J. Connolly,à Virginia Pulito,§
Jingchun Liu and Steven A. Middleton–
Johnson & Johnson Pharmaceutical Research and Development L.L.C., PO Box 300, 1000 Route 202 South, Raritan, NJ 08869, USA
Received 17 September 2007; revised 15 November 2007; accepted 19 November 2007
Available online 22 November 2007
Abstract—Benign prostatic hyperplasia/lower urinary tract symptoms (BPH/LUTS) can be effectively treated with a1 adrenergic
receptor antagonists. Unfortunately, currently marketed a1 blockers produce CV-related side effects that are caused by the subtype
non-selective nature of the drugs. To overcome this problem, it was postulated that an a1a/1d subtype-selective antagonist would
bring more benefit for the treatment of BPH/LUTS. As a continuation of our effort to develop selective a1a/1d ligands, a series
of (phenylpiperazinyl)cyclohexylureas was synthesized and evaluated for the ability to bind to three cloned human a1-adrenergic
receptor subtypes. Several trans isomers were shown to have equal affinity for both a1a, and a1d subtypes, with 14- to 47-fold selec-
tivity versus the a1b subtype and >15-fold selectivity versus dopamine D2.
Ó 2007 Elsevier Ltd. All rights reserved.
Because the proportion of older adults is steadily
increasing and is predicted to grow in the 21st century,
healthcare providers and the pharmaceutical industry
have begun to pay greater attention to age-related dis-
eases such as benign prostatic hyperplasia (BPH). BPH
is a disease affecting the prostate, a walnut-sized male
sex auxiliary gland located just below the bladder and
surrounding the urethra. Excessive growth of the pros-
tate with age will lead to BPH, which eventually causes
obstruction of the bladder outlet and results in the man-
ifestation of lower urinary tract symptoms (LUTS).
These symptoms include increased frequency of urina-
tion, decreased urine stream, increased urgency and feel-
ing of irritation, and sensation of incomplete bladder
emptying.1,2 Several methods for the treatment of
BPH/LUTS are available. Besides surgical resection of
the prostate (prostatectomy), drug intervention is a pop-
ular alternative. Pharmaceutical therapies for BPH/
LUTS have been developed to treat the two pathological
components in BPH, increased size and elevated muscle
tone of prostate gland. In the first class of medicines are
the 5-a-reductase inhibitors (e.g., finasteride and
dutasteride), which work by reducing the size of the
prostate. The second class is comprised of the a1-adren-
ergic receptor antagonists (e.g., tamsulosin and terazo-
sin), which work by relaxing prostate smooth muscle.
One potential advantage of the a1 blockers is that they
can provide effective relief of symptoms in a very short
period of time, compared with the typically slow onset
of 5-a-reductase inhibitors. Unfortunately, the benefit
of using a1 blockers for treatment of BPH/LUTS is
shadowed by the fact that all a1 drugs currently on the
market also produce potentially serious cardiovascu-
lar(CV)-associated side effects, specifically orthostatic
hypotension.3,4
Keywords: BPH/LUTS; a1 adrenergic receptors; a1 blockers; a1a/1d
subtype selective antagonist; (Phenylpiperazinyl)cyclohexylureas.
*
Corresponding author. Tel.: +1 908 231 4697; fax: +1 908 231
4766; e-mail: George.Chiu@sanofi-aventis.com
Present address: Sanofi-Aventis, BW-D-203A, 1041 Route 202-206,
PO Box 6800, Bridgewater, NJ 08807, USA.
à
§
Present address: Johnson & Johnson Pharmaceutical Research and
Development LLC, 8 Clarke Drive, Cranbury, NJ 08512, USA.
Present address: Wyeth Research, 865 Ridge Road, Monmouth
Junction, NJ 08852, USA.
–
Molecular biology has classified the a1-adrenergic
receptor into a1a, a1b, and a1d subtypes;5–7 current
Present address: Hoffmann-La Roche, Inc., 340 Kingsland Street,
Bldg. 76/5E11, Nutley, NJ 07110, USA.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.068

G. Chiu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 640–644
641
a1-blocking drugs have been shown to have approxi-
mately equal affinity for each subtype.8 It is speculated
that the orthostatic hypotension side effect of current
a1 blockers is caused by their lack of subtype selectivity.
Among the a1-adrenoceptor subtypes, a1a was deter-
mined to play a major role in the control of human pros-
tatic smooth muscle contraction, but the precise
contribution of each subtype to adverse side effects has
not yet been clearly defined.8 Many a1a-adrenoceptor
subtype-selective antagonists have since been discov-
ered, and they have demonstrated the ability to relax
prostate muscle without causing CV side effects in ani-
mals.9 Surprisingly, in subsequent human clinical trials,
these a1a-selective compounds were not proven to be
effective in relieving LUTS, in sharp contrast to their
subtype non-selective counterparts.10 This finding
strongly suggests that in addition to the a1a subtype,
other a1 receptor subtypes may be implicated in BHP/
LUTS. Recently, evidence has emerged indicating that
the a1d subtype is also involved in the mediation of
LUTS.11 Experimental data also suggest that the a1b
subtype may be associated with CV-related side effects.12
These results, combined with the fact that tamsulosin (1,
Fig. 1), a moderately a1a/1d-selective drug, is capable of
treating both BPH and LUTS, led to the formation of
new hypothesis that an antagonist with a balanced
a1a/1d selectivity profile will be efficacious yet will pro-
duce fewer side effects.13 Unfortunately, convincing evi-
dence for this hypothesis has been hindered by the fact
that no a1 antagonists with high a1a and a1d affinity
and good selectivity versus a1b are currently available.
3,14b and (phenylpiperidinyl)cyclohexylsulfonamides
414c (Fig. 1). Many of these compounds exhibited
equally high affinity for both a1a- and a1d-adrenocep-
tors, with very good selectivity against the a1b subtype.
As a part of this research, in this paper, we report the
design and synthesis of (phenylpiperazinyl)cyclohexylu-
reas 5 and evaluation of their binding affinities for
cloned human a1a, a1b, and a1d-adrenergic receptor sub-
types and the dopamine D2 receptor. As can be seen in
the following results, some of these compounds show
single-digit nanomolar affinities for a1a and a1d sub-
types, with more than 10-fold selectivity versus the a1b
subtype. The data also reveal a configuration–activity-
relationship unique to this urea series.
The (phenylpiperazinyl)cyclohexylureas were generally
prepared by the sequence shown in Scheme 1. Readily
available 1-(2-isopropoxyphenyl)piperazine was sub-
t
jected to reductive alkylation with Boc-protected 4-
aminocyclohexanone to give a cis/trans mixture of
substituted 1,4-diaminocyclohexane intermediates.
Treatment with trifluoroacetic acid produced the free
cyclohexylamine, which was acylated by various phenyl
or benzyl isocyanates. Final chromatographic separa-
tion gave the individual cis and trans stereoisomers for
all cases except the analogues prepared from 2-pheno-
xyphenylisocyanate (R = 2-PhO-C6H4), which were
tested as the cis/trans mixture.15
The resulting series of analogues was prepared from
the deprotected cyclohexylamines and phenyl or benzyl
isocyanates to examine the effects of aromatic substitu-
tion on binding affinities for the three cloned human
a1-adrenoceptor subtypes and for the dopamine D2
receptor.16 The receptor binding results are summa-
rized in Table 1, with binding data for tamsulosin (1)
included for reference. We previously reported that
the cis and trans isomers in the (phenylpiperazi-
nyl)cyclohexylphthalimide (2)14a and (phenylpiperazi-
nyl)cyclohexylsulfonamide (3)14b series exhibited
substantial differences in binding affinity and selectivity
We initiated a research program to validate this hypoth-
esis by first discovering a1a/1d-selective compounds, then
studying them in established animal models. Our pri-
mary goal was to design and synthesize potent and
a1a/1d subtype-selective antagonists with better selectiv-
ity profiles than marketed drug tamsulosin. In our previ-
ous papers, we reported the discovery of three new series
of a1 antagonists: (phenylpiperazinyl)cyclohexylphthali-
mides 2,14a (phenylpiperazinyl)cyclohexylsulfonamides
SO2NH2
O
O
O
O
R
N
N
N
N
H
O
O
tamsulosin (1)
R1
2
O
O
O
O
R2
R2
S
N
H
O
N
N
X
N
N
N
R4
R3
R1
X: N, 3
X: C, 4
5
Figure 1. Structures of tamsulosin (1), (2), (3), (4), and (5).

642
G. Chiu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 640–644
O
O
O
NHBoc
b
N
N
N
NHBoc
NH
a
O
O
O
N
R
H
N
c
N
N
NH2
N
N
H
O
O
N
O
O
N
R
H
R
H
d
N
N
N
N
N
N
+
H
H
Scheme 1. Reagents and conditions: (a) Na(AcO)3BH, HOAc, CH2Cl2, rt, 8 h, 40–65% yield; (b) CF3CO2H/CH2Cl2, rt, 2 h, 90–100% yield; (c)
RNCO/CH2Cl2, 50–85% yield; (d) SiO2 column or prep TLC.
profile, with cis isomers generally having higher affini-
ties and selectivities. However, in contrast to these pre-
vious findings, the trend in the phenylurea series is
reversed, with the trans isomers (especially trans-9,
-12, -13, -14, -15, -16, and -17) exhibiting higher a1
binding affinities than their cis counterparts. This
Table 1. Binding profile of (phenylpiperazinyl)cyclohexylureas (Kia, nM)
O
O
N
N
N
H
N
H
n
X
Compound
Config.16,17
X
n
a1a
a1b
a1d
D2
1
6
7
—
Mixture
cis
Tamsulosin
2-OPh
2-Cl
—
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0.19
38
2.0
430
373
39
0.2
8.4
46
—
22
48
11
878
33
trans
cis
2-Cl
3.0
73
8
9
2-MeO
2-MeO
3-Cl
155
38
223
107
277
71
390
921
108
84
trans
cis
8.7
29
22
trans
cis
3-Cl
5.1
110
74
1.5
92
10
11
12
13
14
15
16
17
18
2,4-diMeO
2,4-diMeO
2-MeO-5-Cl
2-MeO-5-Cl
2,4-diF
2,4-diF
2,5-diF
2,5-diF
2,6-diF
2,6-diF
3,4-diF
3,4-diF
2,6-diCl
2,6-diCl
2,4,6-triCl
2,4,6-triCl
3-F
475
214
207
81
288
892
99
trans
cis
20
64
101
34
trans
cis
2.7
26.5
2.7
52
1.2
35.5
0.69
199
0.46
22
14
475
76
trans
cis
3.1
26
259
36.4
299
36
trans
cis
1.7
52
1.2
92
trans
cis
5.3
51
4.8
72
2.4
127
54
963
60
trans
cis
3.6
33
1.6
16
289
50
55
trans
cis
6.2
51
2.6
25
4.7
103
6.5
183
120
483
112
706
69
trans
cis
6.5
82
3.7
62
trans
3-F
11
15
a Ki values for a1a, a1b, and a1d were obtained by displacement of [125I]-HEAT from recombinant human receptors. Ki values for D2 were obtained by
displacement of [125I]-spiperone from recombinant human receptors.

G. Chiu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 640–644
643
opposite configuration–activity-relationship clearly
indicates that phenylurea analogues occupy the antag-
onist binding sites in a very different manner than both
phthalimides and sulfonamides. In general, the trans
isomers bind with up to 4-fold higher affinity for a1d
than a1a, and this trend appears to be independent of
the substitution pattern on the phenyl ring. However,
the presence of an electron donating methoxy substitu-
ent leads to relatively weaker affinity for both a1d and
a1a, even in the favored trans series (6, trans-8, trans-
10, and trans-11). Several phenylureas having fluorine
or chlorine substituents show excellent, single-digit
nanomolar a1a and a1d affinities with 14- to 47-fold
selectivity versus the a1b subtype (trans-9, trans-12,
trans-13, trans-15, and trans-17). Compounds trans-
12, trans-13, and trans-17 in particular show equally
high affinity for both a1a and a1d subtypes and are 5-
to 34-fold selective versus a1b; however, they also bind
strongly to the D2 receptor. Overall, the best two com-
pounds in this series are trans-9 and trans-15. They
have roughly equal affinity for both a1a and a1d sub-
types and are selective with a1a/a1b ratios of 14- and
17-fold, and a1d/a1b ratios of 30- and 47-fold, respec-
tively. Trans-9 and trans-15 are not as potent as tam-
sulosin, but they have a1a/a1b selectivity similar to
tamsulosin and substantially better a1d/a1b selectivity.
Trans-9 and 15 also show better selectivity versus the
D2 receptor (15- to 56-fold) than trans-12, trans-13,
and trans-17. It was disappointing to discover that
selectivity versus a1b and D2 receptors in the phenylu-
rea series is not as good as our previously reported
compounds, especially the sulfonamide analogues (3
and 4), which have selectivity ratios close to and some-
times better than 100-fold. Clearly, improvement of the
selectivity profile of the phenylurea series should be a
goal of future research.
trans isomers have higher affinity than cis isomers)
opens the door to a new a1 antagonist pharmacophore
that warrants future research.
Acknowledgments
We express our gratitude to Ms. Amy Madan, Ms. Sally
Varga, and Ms. Aida Howell for their technical assis-
tance in this research.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2007.11.068.
References and notes
1. Rosini, M.; Bolognesi, M.; Giardina, D.; Minarini, A.;
Tumiatti, Vincenzo; Melchiorre, C. Curr. Top. Med.
Chem. 2007, 7, 147.
2. (a) Michelotti, G. A.; Schwinn, D. A. Curr. Urol. Rep.
2004, 5, 258; (b) Cutis, N. J. Urology 2003, 62, 34.
3. Lowe, F. C. Clin. Ther. 2004, 26, 1701.
4. Muramatsu, I.; Suzuki, F.; Tanaka, T.; Yamamoto, H.;
Morishima, S. J. Pharm. Soc. Jap. 2006, 126, 187.
5. Bruno, J. F.; Whittaker, J.; Song, J.; Berelowitz, M.
Biochem. Biophys. Res. Commun. 1991, 179, 1485.
6. Ramarao, C. S.; Kincade Denker, J. M.; Perez, D. M.;
Galvin, R. J.; Riek, R. P.; Graham, R. M. J. Biol. Chem.
1992, 267, 21936.
7. Hirasawa, A.; Horie, K.; Tanaka, T.; Takagaki, K.;
Murai, M.; Yano, J.; Tsujimoto, G. Biochem. Biophys.
Res. Commun. 1993, 195, 902.
8. Forray, C.; Bard, J. A.; Wetzel, J.; Chiu, G.; Shapiro, E.;
Tang, R.; LePor, H.; Hartig, P. R.; Weinshank, R. L.;
Branchek, T. A.; Gluchowski, C. Mol. Pharmacol. 1994,
45, 703.
9. (a) Chiu, G.; Forray, C.; Gluchowski, C. Chem. Biol. Drug
Design 2006, 676, 437; (b) Wong, W. C.; Chiu, G.; Wetzel,
J. M., et al. J. Med. Chem. 1998, 41, 2643; (c) Lagu, B.;
Tian, D.; Chiu, G.; Nagarathram, D.; Fang, J.; Shen, Q.;
Forray, C.; Ransom, R. W.; Chang, R. S. L.; Vyas, K. P.;
Zhang, K.; Gluchowski, C. Bioorg. Med. Chem. Lett.
2000, 10, 175.
In addition to the phenylureas, the related benzylurea
analogues 18 were also prepared and tested. Like all of
the phenylureas, trans-18 is more potent than its cis
counterpart. One encouraging observation is the much
weaker D2 affinity of trans-18 than most trans-phenylu-
reas. This may be a pharmacophore feature that can be
explored in the future.
In conclusion, we have designed and synthesized a series
of (phenylpiperazinyl)cyclohexylureas as an expansion
of our efforts to develop a1a/1d-selective adrenergic
receptor antagonists as new drugs for the treatment of
BPH/LUTS. The binding affinities and selectivities of
these compounds were evaluated in cloned human a1a,
a1b, and a1d adrenergic receptor subtypes and the dopa-
mine D2 receptor. The effect of aromatic substitution on
affinity and selectivity was investigated. We discovered
several compounds (trans-9, trans-12, trans-13, trans-
15, and trans-17) that showed equally potent affinity
for both a1a and a1d adrenoceptor subtypes. Two of
these compounds (trans-9 and trans-15) also had 14- to
47-fold selectivity versus the a1b subtype and >15-fold
selectivity versus D2. Although from the perspective of
a1a/1d antagonism, urea analogues are not as selective
as our previously developed sulfonamide analogues,
their unique configuration–activity-relationship (i.e.,
10. Blue, Jr. D.R.; Grino, P.B.; Jung, D.T.; et al, [abstract] In:
Proceedings of the Fifth International Consultation on
BPH; June 25, 2000.
11. (a) Ilampel, C.; Dolber, P. C.; Smith, M. P., et al. J. Urol.
2002, 167, 1513; (b) Gu, B.; Reiter, J. P.; Schwinn, D. A.,
et al. J. Urol. 2004, 172, 758; (c) Smith, M. S.; Scharnbra,
U. M.; Wilson, K. H.; Page, S. O.; Schwinn, D. A. Brain.
Res. Mol. Brain Res. 1999, 63, 254.
12. Cavalli, A.; Lattion, A. L.; Hummler, E., et al. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 11589.
13. (a) Schwinn, D. A.; Price, D. T.; Narayan, P. Mayo Clinic
Proc. 2004, 79, 1423; (b) Roehrborn, C. G.; Schwinn, D.
J. Urol. 2004, 171, 1029; (c) Sarma, P. K. S.; Tiwari, A.;
Pal, A. Exp. Opin. Ther. Patents 2005, 15, 1333; (d)
Daniels, D. V.; Gever, J. R.; Jasper, J. R.; Kava, M. S.,
et al. Eur. J. Pharmacol. 1999, 370, 337; (e) Lowe, F. C.
Clin. Ther. 2004, 26, 1701.
14. (a) Li, S.; Chiu, G.; Pulito, V. L.; Liu, J.; Connolly, P. J.;
Middleton, S. A. Bioorg. Med. Chem. Lett. 2007, 17, 1646;

644
G. Chiu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 640–644
(b) Chiu, G.; Li, S.; Connolly, P. J.; Pulito, V. L.; Liu, J.;
4.50–4.75 (m, 1 H), 4.95–5.25 (m, 1 H), 6.70–7.10 (m, 6 H),
7.15–7.45 (m, 2 H), 8.15 (d, J = 6.9 Hz, 1 H); MS: 471/473
(M+1). Trans-7: NMR: d (CDCl3) 1.05–1.34 (m, 3 H), 1.38
(d, J = 6 Hz, 6 H), 1.41–2.25 (m, 5 H), 2.25–2.45 (m, 1 H),
2.60–2.85 (m, 4 H), 3.00–3.30 (m, 4 H), 3.50–3.80 (m, 1 H),
4.50–4.75 (m, 1 H), 4.75–5.20 (m, 1 H), 6.70–7.10 (m, 6 H),
7.15–7.50 (m, 2 H), 8.14 (d, J = 6.9 Hz, 1 H).; MS: 471/473
(M+1). Cis-18: NMR: d (CDCl3) 1.32 (d, J = 6.1 Hz, 6H),
1.61 (m, 4H), 1.7–1.8 (m, 4H), 2.30 (m, 1H), 2.70 (m, 4H),
3.11 (m, 4H), 3.82 (m, 1H), 4.33 (d, J = 6.2 Hz, 2H), 4.59
(m, 1H), 4.60 (bs, NH, 1H), 5.08 (bs, NH, 1H), 6.7–7.4 (m,
8H);MS: 469 (M+1). Trans-18: NMR: d (CDCl3) 1.10 (m,
2H), 1.33 (d, J = 6.0 Hz, 6H), 1.40 (m, 2H), 1.95 (bd, 2H),
2.10 (bd, 2H), 2.26 (m, 1H), 2.72 (m, 4H), 3.12 (m, 4H),
3.51 (m, 1H), 4.32 (bs, NH, 1H), 4.35 (d, J = 6.1 Hz, 2H),
4.60 (m, 1H), 4.80 (bs, NH, 1H), 6.7–7.4 (m, 8H); MS: 469
(M+1).
Middleton, S. A. Bioorg. Med. Chem. Lett. 2007, 17, 3292;
(c) Chiu, G.; Li, S.; Connolly, P. J.; Pulito, V. L.; Liu, J.;
Middleton, S. A. Bioorg. Med. Chem. Lett. 2007, 17, 3930.
15. Most cis/trans pairs of urea analogues were separable.
Both cis and trans isomers give distinct yet consistent
NMR patterns. The isomer having a higher Rf value by
TLC was determined by 2D NMR to be the cis isomer.
This result was confirmed by X-ray crystallography. See
supplemental materials in Supporting Information
section.
16. Many analogues in previously reported series showed
affinity for the D2 receptor. See Ref. 14.
17. NMR and MS data of representative compounds cis-7,
trans-7, cis-18, and trans-18: Cis-7: NMR: d (CDCl3) 1.37
(d, J = 6 Hz, 6 H), 1.50–2.10 (m, 8 H), 2.20–2.38 (m, 1 H),
2.55–2.90 (m, 4 H), 3.00–3.30 (m, 4 H), 3.90–4.10 (m, 1 H),