Design, synthesis, crystal structure, biological evaluation and molecular docking studies of carbazole-arylpiperazine derivatives

Article abstract of DOI:10.1016/j.bmc.2016.09.010

Subtype-selective α₁-adrenoceptor (AR) antagonists display optimum therapeutic efficacies for the treatment of benign prostatic hyperplasia (BPH). In this study, we designed and synthesized novel carbazole-arylpiperazines derivatives (1 and 2) on the basis of the proposed pharmacophore model for α₁-AR antagonists. Structural properties were investigated using single-crystal X-ray diffraction analysis. Comparison of crystal structures with ligand-based pharmacophore models revealed that the two agents may possess antagonistic effects on α_1Dsubtype. Tissue functional assay in vitro showed that compound 2 exerted strong antagonistic activity on α_1B-AR (pA₂7.13) with a poor selectivity for α_1Aand α_1Dsubtypes. Compound 1 exhibited enhanced antagonistic effect on α_1Dsubtype (pA₂7.06) and excellent selectivity for α_1Dover α_1B(α_1D/α_1Bratio?=?79.4). To illustrate the relationship between antagonistic activity and chemical structure, molecular docking studies were performed using the homology models of α₁receptors. Binding mechanism indicated that small hydrophobic substituents attached to the arylpiperazine moiety were essential for rational design of α_1D-selective antagonists.

Full text of DOI:10.1016/j.bmc.2016.09.010

Bioorganic & Medicinal Chemistry 24 (2016) 5565–5572
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Design, synthesis, crystal structure, biological evaluation and
molecular docking studies of carbazole-arylpiperazine derivatives
b,
Wei Xu a,y, Junjun Huang b,y, Binhao Shao b, Xingjie Xu b, Renwang Jiang a, , Mu Yuan
⇑
⇑
a School of Pharmaceutical Sciences, Jinan University, Guangzhou 510632, China
b Pharmaceutical Research Center, Guangzhou Medical University, 195# Dongfengxi Road, Guangzhou 510182, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 July 2016
Revised 6 September 2016
Accepted 7 September 2016
Available online 13 September 2016
Subtype-selective
a
1-adrenoceptor (AR) antagonists display optimum therapeutic efficacies for the treat-
ment of benign prostatic hyperplasia (BPH). In this study, we designed and synthesized novel carbazole-
arylpiperazines derivatives (1 and 2) on the basis of the proposed pharmacophore model for 1-AR antag-
onists. Structural properties were investigated using single-crystal X-ray diffraction analysis. Comparison
of crystal structures with ligand-based pharmacophore models revealed that the two agents may possess
antagonistic effects on a1D subtype. Tissue functional assay in vitro showed that compound 2 exerted
a
Keywords:
strong antagonistic activity on
Compound 1 exhibited enhanced antagonistic effect on a1D subtype (pA2 7.06) and excellent selectivity
for a1D over a1B a1B ratio = 79.4). To illustrate the relationship between antagonistic activity and
a1B-AR (pA2 7.13) with a poor selectivity for a1A and a1D subtypes.
a1-Adrenoceptor antagonists
Arylpiperazine
(
a1D/
X-ray crystallography
Molecular docking
Pharmacophore model
chemical structure, molecular docking studies were performed using the homology models of a1 recep-
tors. Binding mechanism indicated that small hydrophobic substituents attached to the arylpiperazine
moiety were essential for rational design of
a1D-selective antagonists.
Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction
model for a1-AR antagonists which consisted of positive ionizable
(PI), hydrophobic features (HY) and hydrogen bond acceptor
a1-Adrenoceptors (ARs) belong to class A of the super family of
(HBA).8 Pyrrolidin-2-one derivative of arylpiperazine (III) also
G protein-coupled receptors (GPCRs) and play a key role in con-
exhibited high affinity for the
tested as
1A-AR antagonist in vivo.9 Structure–activity relation-
ship (SAR) studies of imidazo- and indol-arylpiperazine derivatives
(IV) validated the pharmacophore model for
1-AR antagonists.10
5-arylidenehydantoin-arylpiperazine and pyridine-
(V),11
arylpiperazine (VI)12 were also proved to possess good affinity
for 1-AR in vitro. Particularly, arylpiperazine-derived naftopidil
is subtype-selective antagonist with a 15-fold selectivity for a1D
a1-AR (pKi = 7.30) and was further
tracting vascular smooth and human prostate smooth muscles.1
a
Based on their distinct pharmacological properties,
classified into three subtypes, namely a1A a1B and a1D
last years, subtype-selective 1-AR antagonists were deemed to
a1-ARs are
2,3
,
.
In the
a
a
be attractive drug candidates for the treatment of benign prostatic
hyperplasia (BPH) which is a common condition that severely
impairs patient’s health in aging males.4,5
a
Arylpiperazine derivatives exhibited extensive bioactivities
including the management of BPH progression.6 Among this kind
of compounds, arylpiperazine derivatives bearing a flavone nucleus
(I) presented in Figure 1 showed similar antagonistic properties for
versus a
1B receptor.13 On the other hand, carbazole derivatives dis-
played various biological activities, such as enzyme inhibition,14,15
the anti-proliferation against different cancer cells lines,16 and the
5-HT7R blocking activity (VII).17
a
1-AR in comparison to the reference agent prazosin.7 Although
In this work, we designed novel a1-AR antagonists that had a
quinazolinone-arylpiperazine derivative (II) displayed the a1
blocking activity less than non-selective antagonist prazosin, the
compound fitted well with the ligand-based pharmacophore
-
three-carbon linker between arylpiperazine moiety and carbazole
fragments (see Fig. 2). The hypothesized pharmacophore model
suggested that the hydrophobic regions are composed of carbazole
group, the phenyl ring of the arylpiperazine moiety and the meth-
oxyl substituent attached on the arylpiperazine, and the PI feature
is the basic N atom on piperazine ring,18 and the hydroxyl group is
defined as hydrogen bond donor (HBD). Two carbazole-arylpiper-
azine derivatives were then synthesized, and fully characterized
⇑
Corresponding authors. Tel./fax: +86 020 8134 0727.
E-mail addresses: rwjiang2008@126.com (R. Jiang), mryuanmu838@sina.com
(M. Yuan).
y
They contributed equally to this work.
http://dx.doi.org/10.1016/j.bmc.2016.09.010
0968-0896/Ó 2016 Elsevier Ltd. All rights reserved.

5566
W. Xu et al. / Bioorg. Med. Chem. 24 (2016) 5565–5572
O
NH
O
O
N
N
N
N
N
O
O
O
7-(2-(4-(2-methoxyphenyl)
piperazin-1-yl)ethoxy)-2-phenyl-4H-chromen-4-one (I)
2-(2-(4-(2-methoxyphenyl)
piperazin-1-yl)ethyl)quinazolin-4(3H)-one (II)
O
N
O
N
N
N
N
N
Cl
N
N
N
O
O
O
N
N
N
N
N
6-(1H-imidazol-1-yl)-2-(4-(4-(2-methoxyphenyl)
piperazin-1-yl)butyl)pyridazin-3(2H)-one
6-(1H-indol-1-yl)-2-(4-(4-(2-methoxyphenyl)
piperazin-1-yl)butyl)pyridazin-3(2H)-one
1-(2-(4-(2-chlorophenyl)
piperazin-1-yl)ethyl)pyrrolidin-2-one (III)
IV
O
O
O
Br
N
O
N
N
O
OH
HN
OH
N
N
N
O
O
1-((6-bromopyridin-2-yl)oxy)-3-(4-(2-methoxyphenyl)
piperazin-1-yl)propan-2-ol (VI)
5-(2,4-dimethoxybenzyl)-3-(2-hydroxy-3-(4-(2-methoxyphenyl)
piperazin-1-yl)propyl)imidazolidine-2,4-dione (V)
N
N
N
O
O
N
N
O
OH
O
1-(9H-carbazol-9-yl)-3-(4-(2-methoxyphenyl)
Naftopidil
piperazin-1-yl)propan-1-one (VII)
Figure 1. Chemical structures of arylpiperazine derivatives.
the presence of TEBA. 1-phenylpiperazine derivatives were pre-
pared according to a literature method.19 The final products
(1 and 2) were afforded under reflux conditions, and further puri-
fied by silica gel column chromatography eluted by a mixture of
ethyl acetate and petroleum ether (1/5, v/v). The structures of com-
pounds 1 and 2 were characterized by their melting points, 1H
NMR, 13C NMR, element analysis and single-crystal diffraction.
2.2. X-ray crystallography
Figure 2. Visualization of pharmacophoric features for novel carbazole-arylpiper-
azine derivatives. Colour legend: green, hydrophobic features (HY); red, positive
ionizable (PI); blue, hydrogen bond donor (HBD); black, hydrogen bond acceptor
(HBA).
Compound 1 crystallizes in the monoclinic space group Cc. The
asymmetric unit contains three crystallographically independent
molecules. A representative crystal structure is presented in Fig-
ure 3. The crystal data and structural refinement of 1 are presented
in Table 1. The low R (5.32% for 1, 5.17% for 2) value in X-ray crys-
tallography validated the synthesized structures.
As depicted in Table 2, most bond lengths are within the normal
ranges, e.g., the C–C single bond in the range of 1.498(8)–1.516
(8) Å, and the O–C bond lengths from 1.382(7) to 1.419(7) Å (see
Table 2). The torsion angle for O(1)–C(13)–C(14)–C(15) is 62.1
(7)°, –59.3(7)° and –61.9(7)° for the three independent molecules
(A, B and C), respectively, which indicates that conformation A dis-
tinctly differs from another two conformers. The carbazole ring of
molecular B is almost perpendicular to [C(20), C(21), C(22), C(23), C
(24), C(25)] aromatic plane with a dihedral angle of 81.44(14)°, and
the corresponding dihedral angles of molecular A and C are 74.23
(16)° and 77.93(16)°, respectively.
by NMR (1H and 13C), elemental analysis and single crystal X-ray
diffraction analysis. The antagonistic activities towards a1-AR were
evaluated using functional assays in vitro. Molecular docking stud-
ies shed light on the relationship between antagonist structures
and bioactivities against a1-AR subtypes. The work provides valu-
able clues for the design of subtype-selective a1-AR antagonists.
2. Results and discussion
2.1. Chemistry
The title compounds 1 and 2 were synthesized in two steps
starting from the commercially available carbazole-4-ol and
2-(chloromethyl)oxirane, as depicted in Scheme 1. The condensa-
tion of carbazole-4-ol and 2-(chloromethyl)oxirane gave the
intermediate compound 4-(oxiran-2-ylmethoxy)-9H-carbazole in
The intra- and intermolecular interactions are of considerable
interest in investigating the packing structures. Intramolecular O
(2)–Hꢀ ꢀ ꢀN(2) H-bond results in the formation of the pseudo-

W. Xu et al. / Bioorg. Med. Chem. 24 (2016) 5565–5572
5567
OH
O
O
O
N
N
R
OH
a
b
HN
N
R
N
H
N
H
N
H
1
2
R= 3-OCH3
R= 2,4-OCH3
Scheme 1. Synthesis of the title compound 1 and 2. Reagents and conditions: (a) 2-(chloromethyl)oxirane, Benzyltriethylammonium chloride (TEBA), NaOH, 70 °C; (b)
CH3OH, reflux.
Figure 3. Crystallographic structure of 1 including three molecules in an asymmetric unit. Displacement ellipsoids are drawn at the 30% probability level, and intramolecular
hydrogen bonds are presented in dashed lines.
Table 1
The crystal of 2 also belongs to monoclinic system with space
Crystal data and structural refinement of compounds 1 and 2
group P21/c, and crystal data are presented in Table 1. Ortho-
Compound
1
2
and meta-methoxy stay almost coplanar with the benzene, since
torsion angles of C(21)–C(22)–O(4)–C(27) and C(20)–C(25)–O(3)–
C(26) are –174.5(2)° and 161.4(2)°, respectively. The dihedral
angle between carbazole and benzene ring is 44.0(6)°, which is
much smaller as compared to 1. The conformational discrepancy
of two X-ray structures is likely ascribed to the formation of
additional intramolecular C(18)–Hꢀ ꢀ ꢀO(3) H-bond in 2, see
Figure 4.
Intramolecular hydrogen bonds [O(2)–Hꢀ ꢀ ꢀN(2) and C(18)–
Hꢀ ꢀ ꢀO(3)] generate non-planar five-membered and six-membered
pseudo rings, respectively. As for the molecular packing (Fig. S2),
two molecules form a R21(5) motif via intermolecular hydrogen
bonds N(1)–Hꢀ ꢀ ꢀO(3) and N(1)–Hꢀ ꢀ ꢀN(3). Thus, the solid-state con-
formation and packing structure of 2 are directly related to hydro-
gen bonds.
Chemical formula
Mr
Temperature (K)
Crystal system, space
group
3(C26H29N3O3)
1294.56
293(2)
C27H31N3O4
461.55
293(2)
Monoclinic, Cc
Monoclinic, P21/c
a, b, c (Å)
17.519(4), 33.882(7),
12.440(3)
90, 104.47(3), 90
7150(3)
8.4876(17), 11.168(2),
26.600(5)
90, 97.26(3), 90
2501.2(9)
4
1.226
0.669
984
a
, b, c (°)
V (Å3)
Z
4
qcalc (g/cm3)
1.203
0.636
2760
l
(mmꢁ1
)
F(000)
Crystal size (mm3)
Radiation
0.1 ꢂ 0.1 ꢂ 0.1
0.3 ꢂ 0.2 ꢂ 0.2
Cu K
a
(k = 1.54178)
Cu Ka (k = 1.54178)
h range (°)
Tmin/Tmax
4.15–68.12
0.938/0.938
41602/12258/7537
3.35–68.21
0.852/0.875
27567/4510,/3395
Reflections collected/
unique/observed
Rint
2.3. 3D Pharmacophore model investigations
0.0329
1.071
0.0364
1.093
Goodness-of-fit on F2
Pharmacophore model is a benefit to designing the effective and
safety antagonists at a1-AR subtypes. Figure 5 listed the key phar-
macophoric features of subtype-selective antagonists on the basis
of the reported structures.8,20 It can be seen that X-ray crystallo-
graphic structures of 1 and 2 were obviously distinguished from
R1/wR2 [I> = 2
R1/wR2 [all data]
max Dqmin (e Åꢁ3
)
r
(I)]
0.0532/0.1504
0.0893/0.2056
0.30/ꢁ0.14
0.0517/0.1499
0.0644/0.1640
0.31/ꢁ0.18
D
q
/
the reported pharmacophoric model for
despite the distance between PI and HY1 was nearly equal. In com-
parison to 1B-AR antagonists model, crystal structures of 1 and 2
did not match the three pharmacophoric features. This case was
also confirmed by biological activities against 1-AR that the two
agents had low potency at the 1B receptor (see Table 4). Interest-
ingly, their crystal data were satisfied with the model for selective
a1A-AR antagonists,
five-membered rings. Intermolecular hydrogen bonds play an
important role in stabilizing the packing structure (Fig. S1 and
a
Table 3). In addition, C–Hꢀ ꢀ ꢀ
p interactions reinforce the three-
dimensional structure with a centroid distance of 3.472(0) Å
[Hꢀ ꢀ ꢀCg = 2.58 Å; \C(10B)–H(10B)ꢀ ꢀ ꢀCg = 160°; Cg is the centroid
of benzene ring; Symmetry code: ꢁ1/2 + x,1/2 ꢁ y, ꢁ1/2 + z].
a
a

5568
W. Xu et al. / Bioorg. Med. Chem. 24 (2016) 5565–5572
Table 2
Selected geometric parameters of compound 1 (Å, °)
Bonds
Dist.
Bonds
Dist.
Bonds
Dist.
O(1A)–C(6A)
O(2A)–C(14A)
O(3A)–C(23A)
N(1A)–C(4A)
N(2A)–C(19A)
C(1A)–C(2A)
C(7A)–C(8A)
C(13A)–C14(A)
1.386(7)
1.419(7)
1.381(7)
1.374(9)
1.444(8)
1.410(9)
1.381(9)
1.498(8)
(°)
O(1B)–C(6B)
O(1B)–C(6B)
O(3B)–C(23B)
N(1B)–C(4B)
N(2B)–C(19B)
C(1B)–C(2B)
C(7B)–C(8B)
C(13B)–C14(B)
1.377(7)
1.377(7)
1.381(6)
1.384(8)
1.458(8)
1.403(8)
1.390(8)
1.509(8)
(°)
O(1C)–C(6C)
O(1C)–C(6C)
O(3C)–C(23C)
N(1C)–C(4C)
N(2C)–C(19C)
C(1C)–C(2C)
C(7C)–C(8C)
C(13C)–C(14C)
1.382(7)
1.382(7)
1.391(6)
1.383(8)
1.448(7)
1.402(9)
1.390(8)
1.516(8)
(°)
Angle
Angle
Angle
O(1A)–C(13A)–C(14A)
N(1A)–C(4A)–C(5A)
C(13A)–C(14A)–C(15A)
O(1A)–C(13A)–C(14A)–C(15A)
N(2A)–C(16A)–C(17A)–N(3A)
106.3(5)
108.5(7)
113.5(6)
62.1(7)
58.5(7)
O(1B)–C(13B)–C(14B)
N(1B)–C(4B)–C(5B)
C(13B)–C(14B)–C(15B)
O(1B)–C(13B)–C(14B)–C(15B)
N(2B)–C(16B)–C(17B)–N(3B)
106.1(4)
107.5(6)
115.9(5)
–59.3(7)
57.2(8)
O(1C)–C(13C)–C(14C)
N(1C)–C(4C)–C(5C)
C(13C)–C(14C)–C(15C)
O(1C)–C(13C)–C(14C)–C(15C)
N(2C)–C(16C)–C(17C)–N(3C)
105.3(5)
107.2(6)
113.1(5)
–62.9(7)
56.9(8)
Table 3
Intermolecular hydrogen bonds for compound 1 (Å, °)
D—Hꢀ ꢀ ꢀA
D—H
Hꢀ ꢀ ꢀA
Dꢀ ꢀ ꢀA
D—Hꢀ ꢀ ꢀA(°)
N(1B)—H(1B)ꢀ ꢀ ꢀO(2A)i
O(2A)—H(2A)ꢀ ꢀ ꢀO(3B)
O(2B)—H(2B)ꢀ ꢀ ꢀO(3A)ii
O(2C)—H(2C)ꢀ ꢀ ꢀO(3C)iii
N(1A)—H(1AA)ꢀ ꢀ ꢀO(2B)IV
0.86
0.82
0.82
0.82
0.86
2.05
2.03
2.15
2.09
2.01
2.887 (7)
2.736 (4)
2.825 (6)
2.775 (6)
2.840 (8)
165
144
140
141
161
Symmetry code: (i) ꢁ1/2 + x, 1/2 + y, z (ii) x, y, 1 + z (iii) x, ꢁy, ꢁ1/2 + z (IV) ꢁ1/2 + x, ꢁ1/2 + y, ꢁ1 + z.
Figure 4. Crystallographic structure of 2 in an asymmetric unit. Displacement ellipsoids are drawn at the 30% probability level, and intramolecular hydrogen bonds
O(2)–Hꢀ ꢀ ꢀN(2) and C(18)–Hꢀ ꢀ ꢀO(3) are presented in dashed lines.
a
1D-AR antagonists. Antagonistic effects on
displayed the highest affinity to a1D in comparison with a1A and
1B, whereas 2 exhibited discrepancy results that the potency
a
1-AR indicated that 1
on a1D subtype (pA2 7.06) when compared to NAF. The results of
isolated tissue analysis in vitro revealed that the para-substituent
present on the arylpiperazine moiety (1) led to the excellent
a
against a1B receptor is over than that of a1D
.
selectivity against the a1D receptor, and two methoxyl substituents
located at the position ortho and meta (2) improved potency on a1B
.
2.4. Biological evaluation in vitro
2.5. Binding modes of ligand-a1 complexes
To evaluate the subtype selectivity, the antagonistic effects of
compounds 1 and 2 were characterized (Table 4). Compound 2,
which possesses two methoxyl groups at position 2 and 5, showed
a1-AR is a member of the G protein-coupled receptors (GPCRs)
family that are constructed by seven transmembrane (TM) helices,
strong antagonistic activity on
a
1B subtype (pA2 7.13), but had poor
N- and C-terminal fragments, and intra- and extracellular loop
a1A
,
a1D subtype selectivity over a1B
(
a1A
/
a1B ratio = 0.2, a1D
/a1B
(ECL) regions.22 The accurate 3D structures of
a1-AR with high res-
ratio = 0.5). As compared to 2, the 4-methoxyl-substituted deriva-
tive 1 exhibited enhanced antagonistic effect on a1D subtype (pA2
7.06) and excellent selectivity concerning both a1D subtype and
olutions had not been obtained, homology-modeling strategies
thus provided a viable tool for structure-based
design.
a1-AR antagonist
a1A subtype over a1B
(
a1D
/
a1B ratio = 79.4, a1A
/
a1B ratio = 3.7).
Molecular docking studies shed light on the binding mechanism
of ligand- 1-subtype complex which can explain the observed
antagonistic activities of 1 and 2. As shown in Figure 6A, compound
Moreover, the selectivity activity of 1 towards a1D subtype over
a1B was ꢃ5-fold higher than that of naftopidil (NAF) which was
served as the positive control,21 despite 1 had decreased potency
a
1 was localized in the hydrophobic pocket involving TM I, II and

W. Xu et al. / Bioorg. Med. Chem. 24 (2016) 5565–5572
5569
Distance (Å) Distance (Å)
Distance (Å)
(PI-HY3)
--
Distance (Å)
(PI-HBD)
--
(PI-HY1)
5.5
(PI-HBA)
7.1
α1A-AR antagonists
α1B-AR antagonists
α1D-AR antagonists
6.2
--
7.8
4.9
5.4
4.5
--
--
X-ray structure of 1
5.64 (A)
5.65 (B)
5.64 (C)
5.66
4.17 (A)
4.22 (B)
4.20 (C)
4.21
6.52 (A)
6.55 (B)
6.55 (C)
7.23
2.92 (A)
2.88 (B)
2.93 (C)
2.78
X-ray structure of 2
Figure 5. Visualization of pharmacophoric features of 1 and 2 based on Barbaro’s model (upper) and comparison of important distances between pharmacophoric features in
reported subtype-selective a1-AR antagonists and crystallographic structures of 1 and 2 (below).
Table 4
Functional antagonistic potency of compounds 1 and 2 (expressed as pA2) on a1-adrenoceptor subtypes of SD rat isolated tissues: vas deferens (a1A), spleen (a1B) and thoracic
aorta (a1D
)
a
Comp.
pA2 (slope)
Selectivity ratiob
a1A
a1B
a1D
a1D
/
a1A
a1D
/
a1B
a1A/a1B
1
2
5.73 0.07 (1.01)
6.36 0.10 (1.45)
7.48 0.07
5.16 0.11 (1.20)
7.13 0.21 (0.91)
6.75 0.11
7.06 0.14 (1.15)
6.85 0.10 (1.34)
7.93 0.11
21.4
3.1
2.82
79.4
0.5
15.14
3.7
0.2
5.37
Naftopidilc
a
b
c
pA2 values SEM (n = 5–8) were calculated from Schild plots, pA2 is the positive value of the line intercept derived through plotting log (Dr-1) versus log [antagonist].
Antilog of 4pA2.
Obtained in Ref. 20.
VII. The OH group formed a hydrogen bond (2.4 Å) with the car-
bonyl oxygen atom of Gln177 in the ECL2 region that has been
proved to be essential for GPCR activation.23 The carbazole and
benzene rings were mainly engaged in hydrophobic interactions
sented differential localizations. The carbazole group may play an
important role in the binding affinity for the a1A receptor as it
extended into the hydrophobic pocket. As revealed by the docking
simulations, the binding poses differences between 1 and 2 was
suggested to be associated with the decreased potency of 1 (pA2
5.16) and comparatively high potency of 2 (pA2 7.13) on the a1B
receptor. Recognition of Val197 of ECL2 improved the binding
with Phe308, Phe312 and Trp316 residues. In the 2-
a1A-adreno-
ceptor complex, the hydroxyl group formed H-bond with
a
Gln177 (2.3 Å between the oxygen atom of hydroxyl group and
the amide oxygen atom of Gln177). The major difference in the
best poses of 1 and 2 was the localization of arylpiperazine moiety
on the outer surface of TM helices.
It can be seen from Figure 6B that the carbazole rings of 1 and 2
entered into the hydrophobic region defined by Phe310, Phe311
and Phe330 residues of TM VI, and the benzene moiety contacted
via hydrophobic interactions with residues Trp121, Phe334 and
Tyr338 of TM VII. The OH group of 1 formed a hydrogen bond
(2.9 Å) with residue Ala122 of TM III; while compound 2 estab-
lished a H-bond (1.9 Å) with Val197 of ECL2, which improved the
antagonistic activity against a1B subtype.
affinity of antagonist 2 for
a1B-AR. The above results discovered
that favorable antagonistic activity against a1B receptor could be
ascribed to high hydrophobicity of the aromatic moiety and bind-
ing affinity to key residues of ECL2. Interestingly, docking modes
(Fig. 6C) demonstrated that the arylpiperazine moieties of 1 and
2 as ‘head’ directed toward helices III, V and VII, and the carbazole
group as ‘tail’ stayed the region near the entrance of the binding
pocket of
a1D, which is contrary to the binding modes of the two
antagonists to a1A and a1B-AR.
3. Conclusions
Docking results showed that 1 was located in a hydrophobic
pocket composed of TM III, V and VII in a1D subtype (Fig. 6C).
The carbazole plane was engaged in hydrophobic interactions with
residues Trp112 of TM II, Phe328 and Tyr332 of TM VII, and the
aromatic ring was also involved in hydrophobic interaction with
the region formed by Phe304 and Phe305 of TM VI. The protonated
piperazine moiety formed a weak electrostatic interaction with
Glu190 of ECL2 (4.1 Å between the nitrogen atom of piperazine
ring and carboxylic acid group of Glu190). For the title compound
2, the OH group formed an H-bond to Cys186 (2.8 Å).
BPH is a progressive condition that severely impairs the quality
of life of the aged male patients. 1-adrenoceptor antagonists were
a
seemed to be the most effective therapeutic strategy for BPH. In
this work, two novel 1-((9H-carbazol-4-yl)oxy)-3-(4-phenylpiper-
azin-1-yl)propan-2-ol derivatives were synthesized and character-
ized by NMR and X-ray crystallographic studies. Both crystals of
the title compounds 1 and 2 are monoclinic with space group Cc
and P21/c, respectively. Intra- and inter-molecular hydrogen bonds
play a significant role in sustaining the specific conformation and
three-dimensional structures. To elucidate the relationship
between structural properties and antagonistic activity, molecular
The binding modes were similar in antagonists 1 and 2 binding
to a1A receptor except that their terminal aromatic moieties pre-

5570
W. Xu et al. / Bioorg. Med. Chem. 24 (2016) 5565–5572
terminal carbazole group entered into the hydrophobic regions
surrounded by transmembrane helices, which contributed to the
improved potency against a1A and a1B subtypes; (ii) high
hydrophobicity of aromatic moiety was benefit to the binding
affinity for both a1A and
a1B-AR, but had the opposite effect on
a1D subtype, and the OH group improved the potency of a1B sub-
type through recognizing the key residues of ECL2 and (iii) intro-
ducing small hydrophobic groups to the arylpiperazine moiety
was essential for rational design of a1D-selective antagonists.
4. Experimental
4.1. Chemistry
All reagents and solvents were of analytical grade and commer-
cially available. Melting points were determined on a SGW X-4
micro melting point apparatus (Shanghai Precision & Scientific
Instrument Co., Ltd, Shanghai, China). NMR spectra were accom-
plished on a Bruker AVANCE-500 spectrometer in DMSO using
TMS as the internal standard, and coupling constants (J) were
quoted in Hz. Single-crystal X-ray diffraction data of compounds
1 and 2 were collected on a Rigaku RAPID II diffractometer with
Cu Ka radiation (k = 1.54178 Å). Flash column chromatography
was performed with silica gel (Qing Dao Ocean Chemical Factory,
300–400 mesh).
4.1.1. Synthesis of 1-((9H-carbazol-4-yl)oxy)-3-(4-(4-methoxy-
phenyl)piperazin-1-yl)propan-2-ol (1)
A mixture of 4-(oxiran-2-ylmethoxy)-9H-carbazole (100 mg,
0.44 mmol) and 1-(4-methoxyphenyl)piperazine (126.8 mg,
0.66 mmol) in methanol (10 ml) was refluxed for 2 h. When the
reaction was complete, as indicated by TLC, the solvent was evap-
orated under reduced pressure. The residue was purified by silica
gel column chromatography (petroleum ether/ethyl acetate = 5/1,
v/v) to give the desired compound 1 (119.5 mg, 63% yield) as a
white solid. mp 202–204 °C. 1H NMR (500 MHz, DMSO-d6) d
11.23 (s, 1H), 8.25 (d, J = 7.7 Hz, 1H), 7.44 (d, J = 8.0 Hz, 1H), 7.33
(t, J = 7.3 Hz, 1H), 7.28 (t, J = 8.0 Hz, 1H), 7.14 (t, J = 7.4 Hz, 1H),
7.07 (d, J = 8.0 Hz, 1H), 6.86 (d, J = 9.1 Hz, 2H), 6.79 (d, J = 9.1 Hz,
2H), 6.69 (d, J = 7.9 Hz, 1H), 5.02 (d, J = 3.4 Hz, 1H), 4.37–4.00 (m,
3H), 3.66 (s, 3H), 3.00 (t, J = 4.4 Hz, 4H), 2.77–2.53 (m, 6H).13
C
NMR (126 MHz, DMSO-d6) d 155.52, 153.32, 145.96, 141.61,
139.41, 126.99, 125.02, 123.03, 122.25, 119.05, 117.76, 114.72,
112.10, 110.85, 104.32, 100.96, 71.29, 67.31, 61.84, 55.66, 54.17,
50.16. Anal. Calcd for C26H29N3O3: C, 72.30; H, 6.72; N, 9.73. Found:
C, 72.23; H, 6.74; N, 9.59.
4.1.2. Synthesis of 1-((9H-carbazol-4-yl)oxy)-3-(4-(2,5-dimetho-
xyphenyl)piperazin-1-yl)propan-2-ol (2)
A mixture of 4-(oxiran-2-ylmethoxy)-9H-carbazole (100 mg,
0.44 mmol) and 1-(2,5-dimethoxyphenyl)piperazine (146.6 mg,
0.66 mmol) in methanol (10 ml) was refluxed for 2 h. When the
reaction was complete, as indicated by TLC, the mixture was con-
centrated under reduced pressure, and purified by silica gel col-
umn chromatography eluting with petroleum ether/ethyl acetate
(5/1, v/v) to afford compound 2 (101.5 mg, 50% yield). mp 162–
164 °C. 1H NMR (500 MHz, DMSO-d6) d 11.24 (s, 1H), 8.26 (d,
J = 7.8 Hz, 1H), 7.45 (d, J = 8.0 Hz, 1H), 7.37–7.32 (m, 1H), 7.30 (t,
J = 8.0 Hz, 1H), 7.15 (t, J = 7.4 Hz, 1H), 7.07 (d, J = 8.0 Hz, 1H), 6.81
(d, J = 8.7 Hz, 1H), 6.70 (d, J = 7.9 Hz, 1H), 6.46 (dd, J = 8.7, 2.9 Hz,
1H), 6.43 (d, J = 2.9 Hz, 1H), 5.01 (d, J = 4.5 Hz, 1H), 4.57–3.93 (m,
3H), 3.71 (s, 3H), 3.67 (s, 3H), 2.98 (s, 4H), 2.78–2.53 (m, 6H). 13C
NMR (126 MHz, DMSO-d6) d 155.00, 153.75, 146.16, 142.34,
141.08, 138.88, 126.47, 124.50, 122.51, 121.73, 118.52, 112.98,
111.57, 110.32, 105.44, 105.08, 103.78, 100.43, 70.79, 66.73,
Figure 6. Selected docking poses of the a1-AR agonists 1 (depicted in yellow) and 2
(depicted in pink) into the putative binding sites of a1A (A), a1B (B) and a1D (C)
subtypes. The two antagonists are shown in stick representation. The receptors are
shown in cartoon representation with cyan alpha helices, green beta sheets and
magenta loops. The seven TM helices are labeled by I, II, III, IV, V, VI and VII,
respectively. Dashed lines represent the hydrogen bonds or electrostatic
interactions.
docking studies using a homology model of
a1-AR was carried out.
The results obtained from docking simulations showed that: (i) the

W. Xu et al. / Bioorg. Med. Chem. 24 (2016) 5565–5572
5571
Figure 7. Amino acid sequence alignments of 3SN6, 2R4R, 2RH1, 4XEE with Alpha 1D.
61.41, 55.88, 55.13, 53.79, 49.95. Anal. Calcd for C27H31N3O4: C,
4.3. Tissue functional assays in vitro
70.20; H, 6.72; N, 9.10. Found: C, 70.33; H, 6.70; N, 9.03.
Isolated thoracic aorta rings, spleen and vas deferens of adult
mature rats (Sprague–Dawley strain, 200–350 g) were used to
test antagonistic activities of investigated compounds (1 and 2)
4.2. X-ray crystallography
Suitable crystals were obtained by slowly evaporating a mixture
of dichloromethane and n-hexane solution at ambient tempera-
ture. Colorless crystals were mounted on a thin glass fiber in a ran-
dom orientation. The data were collected by a R-AXIS RAPID II
for a1D, a1B, a1A subtype, respectively. These tissues were dis-
sected and placed into Krebs–Henseleit solution (NaCl 119 mM,
KCl 4.7 mM, CaCl2 1.9 mM, MgSO4 1.2 mM, KH2PO4 1.2 mM,
NaHCO3 25 mM, glucose 11 mM, EDTA 0.05 mM, pH 7.4) at tem-
perature 37 °C and pH 7.4 with constant oxygenation (O2/CO2,
19:1).20
diffractometer equipped with graphite-monochromatic Cu K
a
radiation (k = 1.54178 Å) by using a scan mode. The structures
x
were solved by direct methods using Olex2 software,24 and the
non-hydrogen atoms were located from the trial structure and
then refined anisotropically with SHELXL-201425 using a full-
matrix least squares procedure based on F2. The weighted R factor,
wR and goodness-of-fit S values were obtained based on F2. The
hydrogen atom positions were fixed geometrically at the calcu-
lated distances and allowed to ride on their parent atoms. Crystal-
lographic data for the structure reported in this Letter have been
deposited at the Cambridge Crystallographic Data Center and allo-
cated with the deposition numbers: CCDC 1032470 and 1028230
for compounds 1 and 2, respectively.
The thoracic aorta rings (approximately 4-mm long) were sub-
jected to an optimal tension and allowed to equilibrate for 1.5 h.
Then, noradrenaline (NA 1 mM) were added into stimulate three
times, and washed until the contractile response had become
constant. The tissues were incubated with antagonists for 0.5 h.
Concentration–response curves of NA were measured at an inter-
val of 1 h in the absence and presence of antagonists. To block
alpha2- and beta-adrenoceptors, the experiments were conducted
in the continuous presence of yohimbine (0.1 mM) and propranolol
(1 mM). The other two tissues were performed the similar experi-
mental protocols.

5572
W. Xu et al. / Bioorg. Med. Chem. 24 (2016) 5565–5572
Concentration–response curves were analyzed using GraphPad
These data include MOL files and InChiKeys of the most important
compounds described in this article.
Prism 5.0 software (GraphPad Software Inc., San Diego, CA, USA).26
As the reference of the maximal NA effect (Emax = 100%) obtained
before incubation with the tested compounds, contractile responses
to antagonists are expressed as a percentage. The data are the mean-
References and notes
1. Sarswat, A.; Kumar, R.; Kumar, L.; Lal, N.; Sharma, S.; Prabhakar, Y. S.; Pandey, S.
K.; Lal, J.; Verma, V.; Jain, A.; Maikhuri, J. P.; Dalela, D.; Gupta, G.; Sharma, V. L. J.
Med. Chem. 2011, 54, 302.
s
SEM of at least 5 separate experiments, and the final pA2 values
are obtained from the linear regression of Schild plot.27
2. Meyer, M. D.; Altenbach, R. J.; Bai, H.; Basha, F. Z.; Carroll, W. A.; Kerwin, J. F.;
Lebold, S. A., Jr.; Lee, E.; Pratt, J. K.; Sippy, K. B.; Tietje, K.; Wendt, M. D.; Brune,
M. E.; Buckner, S. A.; Hancock, A. A.; Drizin, I. J. Med. Chem. 2001, 44, 1971.
3. Nanda, K.; Naruganahalli, K. S.; Gupta, S.; Malhotra, S.; Tiwari, A.; Hegde, L. G.;
Jain, S.; Sinha, N.; Gupta, J. B.; Chugh, A.; Anand, N.; Ray, A. Eur. J. Pharmacol.
2009, 607, 213.
4. Kinsella, G. K.; Rozas, I.; Watson, G. W. J. Med. Chem. 2006, 49, 501.
5. Ventura, S.; Oliver, V.; White, C. W.; Xie, J. H.; Haynes, J. M.; Exintaris, B. Br. J.
Pharmacol. 2011, 163, 891.
6. Xu, W.; Huang, J. J.; Shao, B. H.; Xu, X. J.; Jiang, R. W.; Yuan, M. Molecules 2015,
20, 19674.
7. Jin, J.; Wang, X. B.; Kong, L. Y. Bioorg. Med. Chem. Lett. 2011, 21, 909.
8. Abou-Seri, S. M.; Abouzid, K.; Abou El Ella, D. A. Eur. J. Med. Chem. 2011, 46, 647.
9. Zare˛ba, P.; Dudek, M.; Lustyk, K.; Siwek, A.; Starowicz, G.; Bednarski, M.;
4.4. Homology modeling and molecular docking
The amino acidic sequences of the human a1A, a1B, a1D receptor
were retrieved from SwissProt database (entry code P35348,
ADA1A_HUMAN; P35368, ADA1B_HUMAN; P25100, ADA1D_HU-
MAN). The homology models of a1A and a1B receptors were gener-
ated due to our previously reported methods.6,28 To construct a1D
model, BlastP of a1D sequence against Protein Data Bank (PDB) in
combination with UCSF Chimera 1.10.2 package was performed to
search a suitable template.29 The protein structures (PDB: 3SN6,
2R4R, 2RH1, 4XEE) with high E values, scores, and wide coverages
were selected as best suitable templates (Fig. 7), and the sequence
identity with alpha 1D was 41.80% for 3SN6, 41.80% for 2R4R,
38.31% for 2RH1 and 25.63% for 4XEE, respectively. The homology
´
´
Nowinski, L.; Razny, K.; Sapa, J.; Malawska, B.; Kulig, K. Bioorg. Med. Chem. 2015,
23, 2104.
10. Strappaghetti, G.; Mastrini, L.; Lucacchini, A. Bioorg. Med. Chem. 2008, 18, 5140.
´
11. Handzlik, J.; Szymanska, E.; Wójcik, R.; Dela, A.; Jastrze˛bska-Wie˛sek, M.;
Karolak-Wojciechowska, J.; Fruzin´ ski, A.; Siwek, A.; Filipek, B.; Kiec´-
Kononowicz, K. Bioorg. Med. Chem. 2012, 20, 4245.
model of a1D subtype was generated by using the program Modeller
12. Handzlik, J.; Szyman´ ska, E.; Wójcik, R.; Dela, A.; Jastrze˛bska-Wie˛sek, M.;
Karolak-Wojciechowska, J.; Fruzin´ ski, A.; Zylewski, M. Med. Chem. 2014, 10,
144.
13. Castiglione, F.; Benigni, F.; Briganti, A.; Salonia, A.; Villa, L.; Nini, A.; Di Trapani,
E.; Capitanio, U.; Hedlund, P.; Montorsi, F. Curr. Med. Res. Opin. 2014, 30, 719.
14. Zimmermann, K.; Sang, X.; Mastalerz, H. A. Bioorg. Med. Chem. Lett. 2015, 25,
2809.
15. Akrami, H.; Mirjalili, B. F.; Khoobi, M.; Moradi, A.; Nadri, H.; Emami, S.;
Foroumadi, A.; Vosooghi, M.; Shafiee, A. Arch. Pharm. (Weinheim) 2015, 348,
366.
16. Chen, Y.; Tang, C.; Wu, Y. Org. Biomol. Chem. 2015, 13, 6773.
17. Kim, Y.; Yeom, M.; Tae, J.; Rhim, H.; Choo, H. Eur. J. Med. Chem. 2016, 110, 302.
18. MacDougall, I. J.; Griffith, R. J. Mol. Graph. Model 2006, 25, 146.
19. Kossakowski, J.; Pakosinska-Parys, M.; Struga, M.; Dybala, I.; Koziol, A. E.; Collu,
D.; Loddo, R. Molecules 2009, 14, 5189.
20. Handzlik, J.; Bojarski, A. J.; Satała, G. Eur. J. Med. Chem. 2014, 78, 324.
21. Huang, J.; He, F.; Huang, M.; Liu, X.; Xiong, Y.; Huang, Y.; Zhu, L.; Yang, Y.; Xu,
X.; Yuan, M. Eur. J. Med. Chem. 2015, 96, 83.
22. Goldfeld, D. A.; Zhu, K.; Beuming, T.; Friesner, R. A. Proc. Natl. Acad. Sci. U.S.A.
2011, 108, 8275.
23. de Graaf, C.; Foata, N.; Engkvist, O. Proteins 2008, 71, 599.
24. Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J. J. Appl. Crystallogr. 2009, 42, 339.
25. Kratzert, D.; Holstein, J. J.; Krossing, I. J. Appl. Crystallogr. 2015, 48, 933.
26. Handzlik, J.; Maciag, D.; Kubacka, M.; Mogilski, S.; Filipek, B.; Stadnicka, K.;
Kiec´-Kononowicz, K. Bioorg. Med. Chem. 2008, 16, 5982.
27. Nirogi, R. V.; Bandyala, T. R.; Reballi, V. J. Enzyme Inhib. 2015, 30, 85.
28. Xu, W.; Shao, B. H.; Xu, X. J.; Yuan, M. J. Mol. Struct. 2016, 1106, 485.
29. Nettles, J. H.; Stanton, R. A.; Broyde, J.; Amblard, F.; Zhang, H.; Zhou, L.; Shi, J.;
McBrayer, T. R.; Whitaker, T.; Coats, S. J.; Kohler, J. J.; Schinazi, R. F. J. Med.
Chem. 2014, 57, 10031.
9.10 (all parameters were left at the default values).30 The model
was then submitted to be energy optimization by using CHARM-
Ming program. Structural evaluation and stereochemical analyses
were performed using PROCHECK, PROVE, CRYST and Ramchandran
plot.31 PyMOL software was employed for checking and validating
protein structures after model refinement.
Conformation A in the crystallographic asymmetric unit of com-
pound 1 was subject to molecular docking studies based on the
comparison of the geometrical parameters between calculated
model and crystal structure. Stable conformation of compound 2
extracted from the crystal structures was saved in mol2 format.
The preparation of the pdbqt files was done by standard procedure
using AutoDock Tools 1.5.6.32 The docking procedures were per-
formed in AutoDock Vina using the default scorning function.33
The binding site was identified according to previous studies.6
Exhaustiveness was set to 100, and number of output conforma-
tions was set to 20. The searching seed was random. The calculated
geometries were ranked in terms of free energy of binding and the
best poses were selected for further analysis.
Acknowledgments
This work was supported by grants from the National Science
Foundation of Guangdong Province (No. S2013040014088), the
Postdoctoral Science Foundation of Guangzhou City (Q188) and
partially supported by the Guangdong Major Scientific and
Technological Special Project for New Drug Development
(2013A022100029).
30. Yrjölä, S.; Parkkari, T.; Navia-Paldanius, D.; Laitinen, T.; Adusei-Mensah, F.;
Savinainen, J. R.; Laitinen, J. T.; Poso, A.; Alexander, A.; Penman, J.; Stott, L.;
Anskat, M.; Irving, A. J.; Nevalainen, T. J. Eur. J. Med. Chem. 2016, 107, 119.
31. Sashidhara, K. V.; Avula, S. R.; Doharey, P. K.; Singh, L. R.; Balaramnavar, V. M.;
Gupta, J.; Misra-Bhattacharya, S.; Rathaur, S.; Saxena, A. K.; Saxena, J. K. Eur. J.
Med. Chem. 2015, 103, 418.
32. Luo, W.; Chen, Y.; Wang, T.; Hong, C.; Chang, L. P.; Chang, C. C.; Yang, Y. C.; Xie,
S. Q.; Wang, C. J. Bioorg. Med. Chem. 2016, 24, 672.
33. Trott, O.; Olson, A. J. J. Comput. Chem. 2010, 31, 455.
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2016.09.010.