Fine-tuning the selectivity of aldosterone synthase inhibitors: Structure-activity and structure-selectivity insights from studies of heteroaryl substituted 1,2,5,6-tetrahydropyrrolo[3,2,1- ij ]quinolin-4-one derivatives

Article abstract of DOI:10.1021/jm101470k

Pyridine substituted 3,4-dihydro-1H-quinolin-2-ones (e.g., 1-3) constitute a class of highly potent and selective inhibitors of aldosterone synthase (CYP11B2), a promising target for the treatment of hyperaldosteronism, congestive heart failure, and myocardial fibrosis. Among these, ethyl-substituted 3 possesses high selectivity against CYP1A2. Rigidification of 3 by incorporation of the ethyl group into a 5- or 6-membered ring affords compounds with a pyrroloquinolinone or pyridoquinolinone molecular scaffold (e.g., 4 and 5). It was found that these molecules are even more potent and selective CYP11B2 inhibitors than their corresponding open-chain analogues. Moreover, pyrroloquinolinone 4 exhibits no inhibition of the six most important hepatic CYP enzymes as well as a bioavailability in the range of the marketed drug fadrozole. The SAR studies disclose that subtle changes in the heterocyclic moiety are responsible for either a strong or a weak inhibition of the highly homologous 11-hydroxylase (CYP11B1). These results are not only important for fine-tuning the selectivity of CYP11B2 inhibitors but also for the development of selective CYP11B1 inhibitors that are of interest for the treatment of Cushing's syndrome and metabolic syndrome.

Full text of DOI:10.1021/jm101470k

ARTICLE
pubs.acs.org/jmc
Fine-Tuning the Selectivity of Aldosterone Synthase Inhibitors:
Structure-Activity and Structure-Selectivity Insights from Studies of
Heteroaryl Substituted 1,2,5,6-Tetrahydropyrrolo[3,2,1-ij]quinolin-4-
one Derivatives
Simon Lucas,†,§ Matthias Negri,† Ralf Heim,† Christina Zimmer,† and Rolf W. Hartmann*,†,‡
†Pharmaceutical and Medicinal Chemistry, Saarland University, Campus C2.3, D-66123 Saarbr€ucken, Germany
‡Helmholtz Institute for Pharmaceutical Research Saarland (HIPS), Campus C2.3, D-66123 Saarbr€ucken, Germany
S Supporting Information
b
ABSTRACT:
Pyridine substituted 3,4-dihydro-1H-quinolin-2-ones (e.g., 1-3) constitute a class of highly potent and selective inhibitors of
aldosterone synthase (CYP11B2), a promising target for the treatment of hyperaldosteronism, congestive heart failure, and
myocardial fibrosis. Among these, ethyl-substituted 3 possesses high selectivity against CYP1A2. Rigidification of 3 by incorporation
of the ethyl group into a 5- or 6-membered ring affords compounds with a pyrroloquinolinone or pyridoquinolinone molecular
scaffold (e.g., 4 and 5). It was found that these molecules are even more potent and selective CYP11B2 inhibitors than their
corresponding open-chain analogues. Moreover, pyrroloquinolinone 4 exhibits no inhibition of the six most important hepatic CYP
enzymes as well as a bioavailability in the range of the marketed drug fadrozole. The SAR studies disclose that subtle changes in the
heterocyclic moiety are responsible for either a strong or a weak inhibition of the highly homologous 11β-hydroxylase (CYP11B1).
These results are not only important for fine-tuning the selectivity of CYP11B2 inhibitors but also for the development of selective
CYP11B1 inhibitors that are of interest for the treatment of Cushing’s syndrome and metabolic syndrome.
’ INTRODUCTION
enzymes or by blocking the actions of the effector hormones by
functional antagonism, affording a successful treatment of heart
failure and hypertension. Inhibitors of the angiotensin converting
enzyme (ACE) proved to trigger a down-regulation of circulating
aldosterone, but increased levels of aldosterone may be seen after
several months of therapy, presumably due to potassium stimu-
lated secretion.3 The persistence of aldosterone secretion despite
treatment with ACE inhibitors and the evidence of the deleter-
ious effects of aldosterone on cardiovascular function led to the
assumption that blocking the mineralocorticoid receptor might
provide additional benefit. This hypothesis has been corrobo-
rated in two recent clinical trials by using the MR antagonists
spironolactone and eplerenone in addition to the standard
Congestive heart failure (CHF) is a condition of insufficient
cardiac output and reduced systemic blood flow, which provokes
a chronic activation of the renin-angiotensin-aldosterone
system (RAAS). As a consequence, the excessive release of
angiotensin II (Ang II) and aldosterone leads to an increased
blood pressure and to a further deterioration of heart function,
mainly mediated via epithelial sodium retention by mineralocor-
ticoid receptor (MR) activation as well as Ang II mediated
vasoconstriction.1 Moreover, aldosterone is known to exert
direct effects on the heart. Activation of nonepithelial MRs
stimulates the progressive synthesis and deposition of fibrillar
collagens in fibroblasts and results in myocardial fibrosis.2 Until
today, various drug classes targeting the RAAS have been
developed in order to interrupt the circle of chronic neurohor-
monal activation, acting either by inhibition of the key regulator
Received: November 16, 2010
Published: March 08, 2011
r
2011 American Chemical Society
2307
dx.doi.org/10.1021/jm101470k J. Med. Chem. 2011, 54, 2307–2319
|

Journal of Medicinal Chemistry
ARTICLE
Chart 1. CYP11B2 Catalyzed Biosynthesis of Aldosterone
Chart 2. Title Compounds
therapy of patients with chronic congestive heart failure and in
patients after myocardial infarction, respectively.4 Aldosterone
antagonistic therapy, however, raises several issues. Spironolac-
tone can induce severe side effects due to its low selectivity
against other steroid hormone receptors.4,5 Although eplerenone
is more selective, clinically relevant hyperkalemia remains a
principal therapeutic risk.6 Moreover, the elevated plasma aldos-
terone concentrations are left unaffected on a pathological level,
promoting the up-regulation of MR expression7 and nongenomic
aldosterone effects8 on the insufficient heart.
(CYP11B1), the key enzyme of glucocorticoid biosynthe-
sis, highly homologous to CYP11B2.16 Consequent struc-
tural optimization of a hit discovered by compound library
screening performed in our laboratory led to a series of
nonsteroidal aldosterone synthase inhibitors with high selec-
tivity against other cytochrome P450s.17,18 However, several
potent and selective compounds, mostly pyridine-substituted
naphthalenes19,20 and dihydronaphthalenes,21 still showed a
strong inhibition of the hepatic drug metabolizing enzyme
CYP1A2 and no inhibitory effect on the aldosterone biosynthe-
sis in vivo.
Hence, we hypothesized a novel approach for the treatment of
hyperaldosteronism, congestive heart failure, and myocardial
fibrosis by lowering the elevated plasma aldosterone levels via
blockade of aldosterone synthase (CYP11B2), the key enzyme of
mineralocorticoid biosynthesis.9,10 Inhibition of steroidogenic
CYP enzymes had earlier been demonstrated to be very efficient
for the treatment of steroid hormone dependent diseases.
Aromatase (CYP19) and 17R-hydroxylase/17,20-lyase (CYP17)
inhibitors are first-line or upcoming therapeutics for the treat-
ment of breast or prostate cancer, respectively. Our drug design
efforts against CYP11B2 are based on our long-standing experi-
ence in the development of CYP1911 and CYP1712 inhibitors.
CYP11B2 is a mitochondrial cytochrome P450 enzyme localized
mainly in the zona glomerulosa of the adrenal gland. It catalyzes
the terminal three oxidation steps in the biogenesis of aldo-
sterone in humans via initial hydroxylation of 11-deoxycortico-
sterone at 11β-position to yield corticosterone, followed by two
subsequent hydroxylations at C18 and water release to yield
aldosterone (Chart 1).13 In addition to the potential therapeutic
utility in cardiovascular diseases, radiolabeled inhibitors of this
enzyme might be a useful tool for molecular imaging of CYP11B
expression in adrenocortical tissue and thus for the diagnosis of
adrenal tumors.14 Because of selectively binding to CYP11B2,
these compounds are also interesting for the imaging of Conn
adenomas, which are characterized by high expression of
CYP11B2.15
Currently, no crystal structure of CYP11B enzymes exist,
which could assist in solving ligand-binding and mechanistic
issues. However, various groups have developed homology
models of CYP11B1 and CYP11B2 in the past decade,18,22
focusing on different aspects in the generation of the models
and integrating most of the mutational studies from the litera-
ture. All the published models were built upon bacterial tem-
plates and hepatic CYPs. Furthermore, two pharmacophore
models have been developed, both successfully applied in a
ligand-based drug design approach.20,23 In the present study,
we describe the development of 1,2,5,6-tetrahydropyrrolo[3,2,1-
ij]quinolin-4-ones and structurally related aldosterone synthase
inhibitors (Chart 2) starting from the previously described
inhibitors 1-3 (Chart 3). A multitemplate homology model
was generated using the crystal structures of five human CYPs as
templates to assist the design of the molecules. The novelty of
this approach consists in the use of the steroidogenic CYP19
enzyme, combined with selected segments taken from other
CYPs. Moreover, we also extended our CYP11B2-pharmaco-
phore model enlarging the structural diversity of the data set.
Finally, potent aldosterone synthase inhibitors were found with
improved selectivity against the cognate CYP11B1, the steroido-
genic enzymes CYP17 (17R-hydroxylase/C17,20-lyase) and
CYP19 (aromatase), as well as the six most important drug-
metabolizing CYPs (CYP1A2, CYP2B6, CYP2C9, CYP2C19,
CYP2D6, and CYP3A4). The in vivo pharmacokinetic profiles
of some promising compounds were determined in male Wistar
rats.
A critical issue in the development of CYP11B2 inhibitors is to
accomplish selectivity against other cytochrome P450 (CYP)
enzymes because the characteristic complexation of the heme
iron is likely to occur in other CYP enzymes, too. The selectivity
issue becomes especially critical with respect to 11β-hydroxylase
2308
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
Chart 3. Considerations Leading to Compounds 4 and 5
dimethyl analogue 32a.24 Regioselective bromination was ac-
complished by treating the fused heterocycles with N-bromo-
succinimide in DMF at 0 ꢀC. The pyrrolo- or pyridoquinolinones
4-7, 13-16, and 32 were obtained by Suzuki coupling of the
arylbromides 4a, 5a, or 32a with an N-heterocyclic boronic
acid.25 Copper catalyzed N-arylation of 4a with imidazole gave
rise to the 1-imidazolyl derivative 17.26 Alternatively, the bromo-
substituted pyrroloquinolinone 4a was transformed into the
corresponding pinacol boronate 6a by treating with bis-
(pinacolato)diboron under palladium catalysis27 and was sub-
sequently used for cross-coupling with a derivative of 3-bromo-
pyridine to afford compounds 8-12 and 18-30. If not com-
mercially available, the 3-bromopyridines used in this step
were prepared as outlined in Scheme 2 either from 3-bromo-
5-methoxypyridine by demethylation and subsequent alkylation
(8a, 9a) or from 3,5-dibromopyridine by Suzuki coupling with a
substituted arylboronic acid (19a-30a). The hydroxypyridine
10 was synthesized by treating the corresponding methoxy
derivative 6 with concentrated hydrobromic acid under reflux.
Thionation of pyrroloquinolinone 4 with Lawessons reagent in
dry toluene afforded the thio-analogue 31.
’ CHEMISTRY
’ MOLECULAR MODELING
The key synthetic transformation leading to the target com-
pounds was a Suzuki coupling to connect the pyrrolo- or
pyridoquinolinone scaffold to various N-heterocyclic systems,
in most cases a derivative of 3-pyridine (Scheme 1). The
advanced intermediates 4a and 5a as well as 32a were prepared
in three consecutive steps starting from commercially available
indoline or 1,2,3,4-tetrahydroquinoline as the initial building
block. The sequence of amide formation and subsequent
Friedel-Crafts cyclization to afford 4b and 5b has been described
previously and was also used for the synthesis of the gem-
Despite a very low sequence identity (<25%), all structures of
cytochrome P450s known so far show a very similar tertiary fold
with the same number of R-helixes (A-L) and β-sheets (10).28
Moreover, they present several common and flexible structural
features like B/C-loop and F/G segment (F helix, F/G loop,
G helix), characterized by a pronounced flexibility that control the
access to the CYP active site and its volume. The F/G-segment
forms both roof and one side of the active site, and it is placed
perpendicular to the I-helix and opposite to the heme. The I-helix
Scheme 1a
a Reagents and conditions: (i) 3-chloropropanoyl chloride, acetone, reflux; (ii) AlCl3, NaCl, 150 ꢀC; (iii) NBS, DMF, 0 ꢀC; (iv) bis(pinacolato)diboron,
Pd(dppf)Cl2, KOAc, DMSO, 80 ꢀC; (v) heteroarylboronic acid, Pd(PPh3)4, aq NaHCO3, DMF, μw, 150 ꢀC or heteroarylboronic acid, Pd(PPh3)4,
aq Na2CO3, toluene/ethanol, reflux; (vi) heteroarylbromide, Pd(PPh3)4, aq Na2CO3, toluene/ethanol, reflux; (vii) imidazole, Cu2SO4, K2CO3, 180 ꢀC;
(viii) conc HBr, reflux; (ix) Lawessons reagent, toluene, reflux; (x) 3,3-dimethylacryloylchloride, acetone reflux (Het = heteroaryl, BPin = pinacol
boronate).
2309
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
Scheme 2a
a Reagents and conditions: (i) conc HBr, reflux; (ii) alkylhalide, K2CO3,
DMF, rt; (iii) arylboronic acid, Pd(PPh3)4, aq Na2CO3, toluene/ethanol,
reflux.
extends through the whole enzyme and presents a kink close to a
highly conserved acidic residue (Asp317), the catalytic Thr318,
and the heme group. Human CYP11B1 and CYP11B2 differ
only in 32 amino acids out of 503 (93% sequence identity).29
However, a variety of amino acid mutations spread through-
out CYP11B1 and CYP11B2 are known to modulate enzyme
functionalities. Moreover, residue swapping experiments,30 in
which CYP11B1 residues were exchanged with their CYP11B2
counterparts and vice versa, suggest that amino acids that are part
of the I-helix lining the active site and the N-terminal region of
the CYP11B enzymes contribute most significantly to changes in
catalytic activities. Very recently, the crystal structure of the
human CYP19 enzyme has been resolved (PDB entry 3eqm).31
It represents a very promising template for CYP11B1 and
CYP11B2 models because active site similarities (hot spots)
are feasible considering the scaffold similarity of the substrates
and the potent “dual-inhibitor“ profiles (CYP11B2 and CYP19)
seen for many potent CYP11B inhibitors (e.g., fadrozole, com-
pound 33). Thus, a multitemplate strategy was followed to build
a set of homology models of CYP11B2 using the crystal
structures of human CYP19, CYP1A2, CYP2E1, CYP2C9, and
CYP2D6 as templates. CYP19, CYP2C9, and CYP3A4 have
been trialed out as core template, necessary for a correct
alignment of all the different CYPs, with CYP2C9 resulting the
most appropriate one. Successively, regions of CYP2C9 with a
low sequence identity to CYP11B2 were exchanged with frag-
ments from other CYPs showing a much higher degree of
similarity. When no high similarity could be achieved, more
templates were selected. Intersection regions between different
crystal structures were also based on two or more templates.
Finally, the sequence identity between CYP11B2 and the “patch-
work” alignment was improved from ca. 23% (with respect to the
single CYPs) to ca. 45%. Important criteria in the choice of the
fragments were mutagenesis information (special attention was
given to residues that are important for aldosterone synthase
function) as well as ligand-steered considerations (i.e., ligands of
CYP2E1 with isoquinoline substructures were included in model
generation due to structural similarity to a series of highly potent
isoquinoline substituted derivatives20,32 developed in our
laboratory). The final models were validated by docking a series
Figure 1. Representation of most important features of the novel
pharmacophore model (A). Solid spheres represent essential features
for the inhibitory potency, whereas dashed spheres stand for optional
features. The pharmacophoric features are colorcoded: cyan, acceptor
(AA1-AA3); orange, aromatic (HY0, HY1, HY5); yellow, aromatic/
hydrophobic (HY2a/b, HY3b, HY4). In (B) and (C), compounds 4
(white) and 18 (green) are mapped to the pharmacophore model.
of nonsteroidal aldosterone synthase inhibitors with a broad
range of activity.
Two pharmacophore models based on inhibitors developed in
our group or identified by compound screening have already
been described and their predictive capacity has been
demonstrated.20,23 Within the present study, structural informa-
tion of the CYP11B2 substrates as well as of fadrozole and
chemically diverse molecules found in recent patents from
Novartis33 were additionally taken into consideration. The novel
pharmacophore model (Figure 1A and Supporting Information
Figures S1 and S2) presents several features consistent with our
previously described models but also includes two novel
2310
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
hydrophobic/aromatic features HY4 and HY5. In particular, the
essential aromatic (HY0, HY1) and acceptor atom features AA1
(responsible for binding to the heme-iron) and AA2 (forming a
hydrogen bond to Glu383), which together define the binding
geometry, as well as HY2a and HY2b, were maintained. Features
like HY3 and AA3 (Figure 1; rendered as dashed spheres) were
included in the design of the new pharmacophore model and for
the alignment of the extended inhibitor data set but were not
taken into consideration in the rational design efforts herein
because they have previously been identified to be optional,
leading to no increase of potency. For example, benzylnaphtha-
lene-based inhibitor 3320 (Chart 4) that exploits both HY3 and
AA3 has he same CYP11B2 potency (IC50 = 11 nM) as the
unsubstituted counterpart 3419 (Chart 4, IC50 = 6.2 nM). At the
same time, however, 33 shows adverse off-target effects such as
CYP2C9 and aromatase inhibition. Nevertheless, these features
are representative for the size of the active site of CYP11B2 and
could be targeted choosing core structures different from our
bicyclic scaffold.
Figure 2. Structure of the CYP11B2-inhibitor complexes of 3 (green)
and 4 (magenta). The I-helix (red) and β-sheet β4-1, both lining the
active site are rendered as cartoons, whereas the heme is shown in orange
sticks with the central iron depicted as ball. Crucial residues for the
interaction (Glu383) and for the catalytic activity (Thr318) are also
rendered as gray sticks.
’ RESULTS AND DISCUSSION
Preliminary studies aiming at the design of nonsteroidal
aldosterone synthase inhibitors performed in our laboratory have
demonstrated that 3-pyridine substituted naphthalenes provide
an ideal molecular scaffold for a strong inhibition of the target
enzyme CYP11B2 as well as high selectivity against several other
CYP enzymes (e.g., CYP11B1, CYP17, CYP19).19,21 Most of
these molecules, however, revealed two major pharmacological
drawbacks: a strong inhibition of the drug metabolizing enzyme
CYP1A2 and no inhibitory effect on the aldosterone production
in vivo in a rat model. In a recent study, we were able to show that
changing the naphthalene by a 3,4-dihydro-1H-quinolin-2-one
scaffold affords highly potent CYP11B2 inhibitors such as 1-3
(Chart 3) with pronounced selectivity against other CYP enzymes
including CYP1A2 as well as aldosterone-lowering effects in vivo.33,34
Among the latter compounds, ethyl-substituted derivative3displayed
a rather low inhibition of CYP1A2 (IC50 = 3.48 μM), thus being a
suitable starting point for further optimization.
Herein, we followed a combined ligand- and structure-based
approach for the optimization of 3 to identify new, structurally
unique molecules, to potentially improve the in vitro and
in vivo potency, and to validate our molecular modeling tools.
A homology model of CYP11B2 was used in a first design step to
rationalize the activity differences of existing compounds, to
guide the design of improved inhibitors, and in a later stage to add
steric information (excluded volumes) to our pharmacophore
hypothesis. One important observation was made by comparing
activity and binding mode of a series of N-alkyl substituted
inhibitors such as 2 and 3: while the insertion of a methyl group
increased the potency (2, IC50 = 2.6 nM) compared to its
unsubstituted derivative 1 (IC50 = 28 nM), increasing the size
of R reduced the inhibition again (3, IC50 = 22 nM) or led to a
complete loss of activity, respectively (e.g., R = isopropyl, phenyl
IC50 > 500 nM).33,35 Docking studies performed for compounds
1-3 using FlexX suggested that the decrease of inhibitory action
with increasing size of the N-alkyl substituents can be explained
by a steric clash with amino acids within β-strand 4-1, e.g. Leu382
(Figure 2). In addition, the quinolinone-based inhibitors were
found to bind in a consistent geometry defined by two key
interactions with the enzyme: (1) a strong coordinative bond to
the heme iron and (2) hydrogen bond formation to the
Chart 4. Benzyl-Substituted Inhibitor 33 and Unsubstituted
Analogue 34
backbone-amide of Glu383. Sterically demanding residues at
the quinolinone-nitrogen hinder this hydrogen bond formation,
whereas small or rigidified substituents may be tolerated. We
envisaged that minimizing the steric clash between the ethyl
group of 3 and the β4-1 strand of CYP11B2 might increase the
inhibitory potency. Eventually, these considerations led to the
design of molecules 4 and 5 (Chart 3), in which the ethyl group is
incorporated into a 5- or 6-membered ring and thereby the
assumed active conformation of the parent molecule is locked.
The inhibitory activities of the compounds were determined
as described previously using V79 MZh cells expressing either
human CYP11B2 or CYP11B1 (Table 1).10,36 The V79 MZh
cells were incubated with [14C]-deoxycorticosterone as substrate
and the inhibitor in different concentrations. The product
formation was monitored by HPTLC using a phosphoimager.
Fadrozole, an aromatase (CYP19) inhibitor with proven ability
to reduce corticoid formation in vitro37 and in vivo,38 was used as
a reference (CYP11B2, IC50 = 1 nM; CYP11B1, IC50 = 10 nM).
Both pyrroloquinolinone 4 and pyrido-condensed analogue 5
were found to be highly potent CYP11B2 inhibitors showing
IC50 values 10- to 20-fold lower (1.1 or 2.4 nM) compared to the
parent molecule 3 (22 nM), indicating that the steric clash with
the hydrogen-bond β-strand is minimized. In addition, the
selectivity ratios against CYP11B1 are significantly higher (650
and 957) than in the case of 3 (235). Because other CYPs
constitute critical off-targets for CYP11B2 inhibitors too, a CYP
2311
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
Table 1. Inhibition of Human Adrenal CYP11B2 and CYP11B1 in Vitro
IC50 valueb (nM)
%
inhibitiona
V79 11B2c
V79 11B2c
V79 11B1d
hCYP11B1
selectivity
factore
compd
n
R
Het
hCYP11B2
hCYP11B2
4
1
2
1
2
1
1
1
1
1
1
2
1
1
1
H
H
90
95
98
95
92
96
94
97
96
98
95
7
1.1
2.4
0.6
0.9
1.0
2.2
4.3
4.4
5.9
0.2
0.2
nd
715
2296
247
545
158
103
2045
1288
141
13
650
957
412
606
158
47
5
6
OMe
OMe
OEt
OiPr
OH
F
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
fadrozole
476
293
24
CF3
4-isoquinoline
4-isoquinoline
4-pyridine
65
34
170
nd
nd
5-pyrimidine
1-imidazole
81
87
97
92
97
96
80
89
81
86
86
95
92
92
80
96
48
56
28546
2077
58
510
23
89
H
1.3
0.7
1.4
0.9
3.6
2.3
18
45
2-F
43
61
3-F
490
40
350
44
4-F
2,5-F
3,4-F
3,5-F
2-OMe
3-OMe
4-OMe
3-OH
3-OCF3
3-CF3
183
496
1748
128
1374
21
51
215
97
2.4
4.6
1.4
1.2
16
53
299
15
44
37
2058
4646
333
nd
129
141
278
nd
33
1.2
nd
1.0
10
10
a Mean value of at least three experiments, standard deviation usually less than 10%; inhibitor concentration, 500 nM. b Mean value of at least three
experiments, standard deviation usually less than 25%, nd = not determined. c Hamster fibroblasts expressing human CYP11B2; substrate
deoxycorticosterone, 100 nM. d Hamster fibroblasts expressing human CYP11B1; substrate deoxycorticosterone, 100 nM. e IC50 CYP11B1/IC50
CYP11B2, nd = not determined.
counterscreen against the steroidogenic enzymes CYP17 and
CYP19 (Table 2) as well as the most important hepatic CYP
enzymes CYP1A2, CYP2B6, CYP2C9, CYP2C19, CYP2D6, and
CYP3A4 (Table 3) was performed to determine the selectivity
profiles of 4 and 5. While the pyrroloquinolinone 4 exhibits a
pronounced selectivity against all tested CYPs, compound 5
shows a moderate aromatase (69% inhibition at 0.5 μM) and
CYP1A2 inhibition (41% inhibition at 1 μM). Thus, the tricyclic
scaffolds of 4 and 5 are promising lead structures for structural
optimization, with 4 showing superior overall selectivity.
Most of the molecules derived from compounds 4 and 5
(Table 1) are highly potent aldosterone synthase inhibitors with
IC50 values in the low nanomolar range (<5 nM). In particular,
the isoquinoline derivatives 13 and 14 are the most potent
2312
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
representatives of this series showing subnanomolar IC50 values
(0.2 nM). However, replacing 3-pyridine by other nitrogen
containing heterocycles induces a decrease in inhibitory potency.
The 5-pyrimidine (16) and 1-imidazole (17) derivatives are less
The pharmacokinetic profile of selected compounds was
determined after peroral application to male Wistar rats. Plasma
samples were collected over 24 h, and the concentrations were
determined by HPLC-MS/MS. Compounds 4, 6, and 7 were
investigated in a cassette dosing approach (peroral dose = 5 mg/kg)
and compared to fadrozole (Figure 3, Table 4). All three com-
pounds show similar terminal half-lifes (t1/2 z = 2.0-2.9 h), which
active (IC50 = 56-89 nM) than the 3-pyridine analogue 4 (IC50
=
1.1 nM), and the 4-pyridine compound 15 shows no inhibition
of CYP11B2 at a concentration of 500 nM, indicating that the
3-pyridine motif is crucial for the geometry of the two key
pharmacophoric features mentioned above and thus for high
potency. The same trend was observed previously for the binding
properties of a series of substituted pyridylnaphthalenes and 3,4-
dihydro-1H-quinolin-2-ones.19,33 The gem-dimethyl group in the
quinolinone moiety in compound 32 is not tolerated in terms of
CYP11B2 potency, while the thioanalogue of 4, compound 31, is
less selective against CYP11B1 than the parent molecule. In the
present study, the chemical modification is mainly directed to the
heterocyclic moiety because both potency and selectivity have
been identified in previous investigations to be highly dependent
on the substitution pattern of the heterocycle.39 The introduc-
tion of small substituents in molecules 6-12 does not influence
the inhibitory potency (IC50 values in the range of 1.0-5.9 nM)
but decreases the selectivity against CYP11B1 compared to 4
or 5. In addition, compounds 5-7 show moderate inhibition of
some hepatic CYP enzymes (Table 3), while CYP17 and CYP19
are not affected (Table 2). Isoquinoline derivative 14 is a strong
inhibitor of hepatic CYP2C9 and CYP2C19.
are in the range of the reference compound fadrozole (t1/2 z
3.2 h). However, the absorbance of compounds 4 and 6 (tmax
=
=
4 h) occurs slower as the absorbance of fadrozole (tmax = 1 h).
Within this series, fadrozole shows the highest maximal concen-
tration (Cmax) in plasma (471 ng/mL) followed by compound 4
(317 ng/mL). The maximal amount of the other test items
Table 2. Inhibition of Human CYP17 and CYP19 in Vitro
Figure 3. Mean profile (() SEM of plasma levels (ng/mL) in rat vs
time after oral application (5 mg/kg) of compounds 4, 6, and 7
determined in a cassette dosing experiment.
% inhibitiona
% inhibitiona
compd
CYP17b
CYP19c
compd
CYP17b
CYP19c
Table 4. Pharmacokinetic Profiles of Compounds 4, 6, and 7a
4
6
6
18
69
5
14
16
20
23
26
31
5
5
<5
5
5
compd t1/2 z (h)b tmax (h)c Cmax (ng/mL)d AUC0-¥ (ng h/mL)e
3
6
<5
<5
<5
6
7
6
7
5
<5
<5
11
<5
<5
5
4
2.4
2.9
2.0
3.2
4
4
2
1
317
60
3464
557
51
11
13
<5
<5
6
7
4.9
471
a Mean value of three experiments, standard deviation less than 10%.
b E. coli expressing human CYP17; substrate progesterone, 25 μM;
inhibitor concentration, 2.0 μM; ketoconazole, IC50 = 2.78 μM.
c Human placental CYP19; substrate androstenedione, 500 nM; inhibi-
tor concentration, 500 nM; fadrozole, IC50 = 30 nM.
fadrozolef
3207
a All compounds were applied perorally (5 mg/kg) to male Wistar rats in
a cassette dosing approach. b Terminal half-life. c Time of maximal
concentration. d Maximal concentration. e Area under the curve. f Mean
value of two experiments.
Table 3. Inhibition of Selected Hepatic CYP Enzymes in Vitro
% inhibitiona
CYP1A2b,c
CYP2B6b,d
CYP2C9b,e
CYP2C19b,f
CYP2D6b,g
CYP3A4b,h
compd
10 μM
1 μM
10 μM
1 μM
10 μM
1 μM
10 μM
1 μM
10 μM
1 μM
10 μM
1 μM
4
43
83
72
87
63
38
8
41
37
47
24
9
<5
36
8
<5
27
<5
nd
nd
<5
41
40
42
63
97
79
6
19
23
34
73
39
<5
44
30
61
96
21
<5
8
<5
nd
13
nd
nd
9
<5
nd
9
<5
21
19
24
<5
43
<5
6
5
6
7
14
12
<5
37
7
nd
nd
13
11
67
nd
nd
nd
<5
14
20
a Mean value of two experiments, standard deviation less than 5%. b Recombinantly expressed enzymes from baculovirus-infected insect microsomes
(Supersomes). c Furafylline, IC50 = 2.42 μM. d Tranylcypromine, IC50 = 6.24 μM. e Sulfaphenazole, IC50 = 318 nM. f Tranylcypromine, IC50 = 5.95 μM.
g Quinidine, IC50 = 14 nM. h Ketoconazole, IC50 = 57 nM.
2313
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
(6 and 7) found in the plasma after peroral application is
significantly lower (<50 ng/mL). Using the area under the curve
(AUC0-¥) as a ranking criterion, pyrroloquinolinone 4 exhibits
8 (R = OEt, IC50 = 158 nM) < 9 (R = OiPr, IC50 = 103 nM).
At the same time, the inhibition of CYP11B2 is in a compar-
able range for all compounds (IC50 = 0.6-2.2 nM) with the
consequence that the selectivity factor decreases in the same
order from 650 (4) to 47 (9). A pronounced increase in
CYP11B1 inhibition is observed in the case of the introduction
of additional aryl moieties. Replacing 3-pyridine by 4-isoquino-
line results in a dramatic increase in CYP11B1 potency for both
the pyrrolo-condensed (13, increase by a factor of 55) and
pyrido-condensed (14, factor 68) derivative, whereas the
CYP11B2 inhibition increases to a minor degree (factor 6-12).
The same trend can be observed for several compounds with
additional aryl substituent in 5-position of the pyridine moiety
(e.g., 18, 19, 31, 25, 27, and 28). The latter compounds display a
CYP11B2 inhibition in a range of 0.7-2.4 nM, which is
the highest bioavailability (3464 ng h/mL), which is in the range
3
of the AUC0-¥ of fadrozole (3207 ng h/mL). Methoxylation of
3
the heterocycle as accomplished in 6 results in a significant
decrease of the AUC0-¥ (557 ng h/mL). A further decrease is
3
observed for the corresponding pyridoquinolinone 7 (51 ng h/mL).
3
This may be due to, for example, either excessive metabolism
(due to the electron donating ability of the methoxy substituent
and, therefore, higher sensitivity to oxidative metabolism) or to
reduced absorption.
The further optimization process, eventually leading to com-
pounds 18-30, was assisted by our pharmacophore model.
Compound 4 already shows a reasonable fit to the pharmaco-
phore hypothesis by exploiting the key pharmacophoric features
AA1, AA2, HY0, and HY1, together with overlapping HY2a and
HY2b (Figure 1B). In the new pharmacophore model, two
additional hydrophobic/aromatic areas were identified adjacent
to the pyridine ring, HY4 and HY5, which is located in the
putative meta-(5)-position. Rationalizing the given information,
we envisaged that compound 18, whose additional benzene
moiety perfectly overlaps with the aromatic moiety HY5
(Figure 1C), is a promising starting point for developing new
inhibitors. Indeed, benzene substituted analogue 18 was found to
be a potentinhibitor of CYP11B2 although no increase of potency
compared to the parent molecule 4 was observed. The same trend
was observed for the other molecules of the series, thereby
rendering the ligands less efficient. Thus, HY5 represents an
optional pharmacophoric feature (like HY3 and AA3) and is
associated with a decreased selectivity against CYP11B1. The
CYP17, CYP19, and hepatic CYP activity of selected molecules
was found to be low or moderate. Compound 20 displays a
distinct inhibition of CYP2C9 (79% at a concentration of 10 μM)
but is rather selective against other CYP enzymes (IC50 > 10 μM).
However, as presented in Table 3, none of the substituted
derivatives is as selective as the unsubstituted parent compound 4.
Given the rather constant IC50 values of CYP11B2 inhibition
of the molecules presented herein (the majority of the com-
pounds shows inhibition in the range of 1-5 nM), structure-
activity relationships with respect to the inhibition of 11β-
hydroxylase (CYP11B1) and thus selectivity become particularly
interesting. On the one hand, selectivity is influenced by the ring-
size of the carbocycle condensed to the quinolinone moiety
(HY4). Within the series of pyrido-condensed compounds (5, 7,
and 14), the CYP11B1 inhibition is significantly decreased
compared to the corresponding pyrroloquinolinone analogues
(4, 6, and 13), whereas the CYP11B2 inhibition is in a compar-
able range. This leads to an enhanced selectivity for the pyrido-
quinolinone derivatives. Pyridoquinolinone 5 is the most
selective compound of the present series (selectivity factor =
957) and is 100-fold more selective than fadrozole (selectivity
factor = 10). This experimental result is particularly noteworthy
with respect to the high homology of the two CYP11B isoforms.
On the other hand, the substitution pattern of the pyridine
moiety was found to significantly influence the CYP11B1
selectivity. The size of substituents in 5-position (HY5) of the
heterocycle plays a crucial role for the CYP11B1 inhibition. This
becomes particularly evident in the series of pyrroloquinolinone
compounds with alkoxy-substituted heterocycle. The inhibitory
potency at CYP11B1 increases with the substituent size in the
order 4 (R = H, IC50 = 715 nM) < 6 (R = OMe, IC50 = 247 nM) <
comparable to the unsubstituted parent compound 4 (IC50
=
1.1 nM). However, the CYP11B1 inhibition increases up to 36-
fold compared to 4 (IC50 = 715 nM) to IC50 values of 21-128
nM, corresponding to a rather low selectivity (factor 15-61).
Apparently, the introduction of additional aromatic rings and to a
minor degree also sterically demanding aliphatic residues (e.g.,
isopropoxy) in the pyridine moiety leads to additional interac-
tions of the inhibitors with CYP11B1. Moreover, it is striking that
meta substituents in the aryl moiety can trigger CYP11B1
selectivity again. With exception of the 3-hydroxy derivative
28, all compounds bearing a substituent in 3-position of the
benzene moiety (i.e., 20, 23, 24, 26, 29, and 30) exhibit a
decreased CYP11B1 potency with IC50 values in a range of 490-
4646 nM and thus selectivity factors of up to 350 (20). This off/
on-switch of CYP11B1 potency is of particular interest. From
unsubstituted parent compound 4 (IC50 = 715 nM), derivatiza-
tion with phenyl in the 5-position of the pyridine moiety leads to
a significant increase of inhibitory potency in compound 18
(IC50 = 58 nM), whereas the meta-fluorophenyl analogue 20
exhibits a low inhibitory potency again (IC50 = 490 nM). At the
same time, these three compounds display the same aldosterone
synthase activity (IC50 = 1.1-1.4 nM). This means that a variety
of sterically demanding substituents in the pyridine moiety is
readily tolerated in the CYP11B2 binding pocket, however, lead
to no further stabilization of the complexes formed by coordina-
tion of the heme iron by the heterocyclic nitrogen. On the other
hand, 3-pyridine substituted pyrroloquinolinone derivatives such
as 4 are per se rather poor CYP11B1 inhibitors and require an
additional benzene moiety and thus a further stabilization of the
formed complexes for basal inhibitory potency. Among these
compounds, there are highly potent CYP11B1 inhibitors, for
example, isoquinoline derivative 13 (IC50 = 13 nM) and para-
methoxyphenyl derivative 27 (IC50 = 21 nM). Obviously the
meta-substituted analogues do not adequately fit into the
CYP11B1 binding pocket or loose contact to the heme iron
while minimizing unfavorable contacts with the enzyme.
’ CONCLUSION
Herein, we have presented the rational design of a series of
pyrroloquinolinone-based inhibitors of CYP11B2, a promising
target for the treatment of aldosterone-mediated disorders such
as hypertension and congestive heart failure. The variation of the
substitution pattern was guided by a pharmacophore hypothesis
based on data from structurally diverse aldosterone synthase
inhibitors and validated by docking studies using a newly
developed homology model.
2314
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
The present study provides extensive SAR results relating to
CYP11B2 and CYP11B1 inhibition. The influence of certain
structural modifications on the 11β-hydroxylase (CYP11B1)
activity is particularly noteworthy. On the one hand, CYP11B1
inhibition is an important selectivity criterion for aldosterone
synthase inhibitors. On the other hand, selective CYP11B1
inhibitors could be used for the treatment of Cushing’s syndrome
and metabolic syndrome. Although several potent CYP11B1
inhibitors have been described previously,40 in-depth SAR stud-
ies usually focused on the CYP11B2 activity. Herein, we clearly
identified structural features important for high inhibitory
CYP11B1 potency such as sterically demanding lipophilic or
aromatic residues in the heterocyclic moiety or condensed to
the heterocycle, leading to a series of highly potent yet unselec-
tive 11β-hydroxylase inhibitors (e.g., para-methoxyphenyl deri-
vative 27, IC50 = 21 nM). Subtle structural modifications of these
compounds, such as introduction of meta-substituents into the
phenyl moiety, led to a significant loss of CYP11B1 inhibition
again, providing selective CYP11B2 inhibitors.
Finally, pyrroloquinolinone 4 emerged from the present study
as a keenly optimized lead molecule. Beside strong aldosterone
synthase inhibiting properties, this inhibitor is orally available in
the range of the marketed drug fadrozole and shows no off-target
effects against similar enzymes (e.g., steroidogenic and hepatic
CYPs). Currently, further studies with inhibitor 4 are underway
to evaluate aldosterone-lowering effects in vivo as well as
beneficial effects in appropriate disease models.
4-one (4b),24 2,3,6,7-tetrahydro-1H,5H-pyrido[3,2,1-ij]quinolin-5-one
(5b),24 5-bromopyridin-3-ol (8b),39 3-bromo-5-ethoxypyridine (8a).39
Synthesis of the Target Compounds. Procedure A. Boronic
acid (1 equiv), aryl bromide or (1.3-1.5 equiv), and tetrakis-
(triphenylphosphane)palladium(0) (5 mol %) were suspended in
toluene/ethanol 4/1 to give a 0.07-0.1 M solution of boronic acid
under an atosphere of nitrogen. To this was added a 1 N aqueous
solution of Na2CO3 (6 equiv). The mixture was then refluxed for 12-
18 h, cooled to room temperature, diluted with water, and extracted
several times with ethyl acetate. The combined extracts were dried over
MgSO4, concentrated, and purified by flash chromatography on silica gel
(petroleum ether/ethyl acetate mixtures) and/or crystallization.
Procedure B.25. Boronic acid (0.75 mmol, 1 equiv), aryl bromide or -
triflate (0.9-1.3 equiv), and tetrakis(triphenylphosphane)palladium(0)
(43 mg, 37.5 μmol, 5 mol %) were suspended in 1.5 mL of DMF in a 10
mL septum-capped tube containing a stirring magnet. To this was added
a solution of NaHCO3 (189 mg, 2.25 mmol, 3 equiv) in 1.5 mL of water,
and the vial was sealed with an Teflon cap. The mixture was irradiated
with microwaves for 15 min at a temperature of 150 ꢀC with an initial
irradiation power of 100 W. After the reaction, the vial was cooled to
40 ꢀC, the crude mixture was partitioned between ethyl acetate and
water, and the aqueous layer was extracted three times with ethyl acetate.
The combined organic layers were dried over MgSO4, and the solvents
were removed in vacuo. The coupling products were obtained after flash
chromatography on silica gel (petroleum ether/ethyl acetate mixtures)
and/or crystallization.
8-Pyridin-3-yl-1,2,5,6-tetrahydro-4H-pyrrolo[3,2,1-ij]quinolin-4-
one (4). Compound 4 was obtained according to procedure A from 4a
(5.19 g, 20.6 mmol) and 3-pyridineboronic acid (2.30 g, 18.7 mmol)
after flash chromatography on silica gel (ethyl acetate, Rf = 0.08) and
crystallization from acetone/diethylether as colorless plates (2.29 g, 9.16
mmol, 49%), mp 153-154 ꢀC. 1H NMR (500 MHz, CDCl3): δ 2.72 (t,
3J = 7.8 Hz, 2H), 3.03 (t, 3J = 7.8 Hz, 2H), 3.25 (t, 3J = 8.5 Hz, 2H), 4.13
(t, 3J = 8.5 Hz, 2H), 7.20 (s, 1H), 7.28 (s, 1H), 7.32 (ddd, 3J = 7.8 Hz, 3J =
4.8 Hz, 5J = 0.5 Hz, 1H), 7.79 (ddd, 3J = 7.8 Hz, 4J = 2.2 Hz, 4J = 1.6 Hz,
’ EXPERIMENTAL SECTION
Chemical and Analytical Methods. Melting points were mea-
sured on a Mettler FP1 melting point apparatus and are uncorrected. 1H
NMR and 13C spectra were recorded on a Bruker DRX-500 instrument.
Chemical shifts are given in parts per million (ppm), and tetramethylsilane
(TMS) was used as internal standard for spectra obtained in DMSO-d6 and
CDCl3. All coupling constants (J) are given in hertz. Reagents were used as
obtained from commercial suppliers without further purification. Solvents
were distilled before use. Dry solvents were obtained by distillation from
appropriate drying reagents and stored over molecular sieves. Flash
chromatography was performed on silica gel 40 (35/40-63/70 μM) with
petroleum ether/ethyl acetate mixtures as eluents, and the reaction progress
was determined by thin-layer chromatography analyses on Alugram SIL G/
UV254 (Macherey Nagel). Visualization was accomplished with UV light
and KMnO4 solution. All microwave irradiation experiments were carried
out in a CEM-Discover monomode microwave apparatus.
The purity of the compounds was determined by LC/MS and/or
elemental analysis and was g95%. The Surveyor-LC system consisted of
a pump, an autosampler, and a PDA detector. Mass spectrometry was
performed on a TSQ Quantum (Thermo Electron Corporation,
Dreieich, Germany). The triple quadrupole mass spectrometer was
equipped with an electrospray interface (ESI). In some cases, ionization
was accomplished by APCI. The system was operated by the standard
software Xcalibur. A RP18 100-3 column (Macherey-Nagel GmbH,
Duehren, Germany) was used as stationary phase. All solvents were
HPLC grade. The HPLC method used for the purity determination
started with a 20 min gradient from 100% water containing 1 % TFA to
100% methanol/acetonitrile (4/1) containing 1 % TFA. The injection
volume was 1-10 μL, and flow rate was set to 350 μL/min. MS analysis
was carried out at a spray voltage of 3800 V, a capillary temperature of
350 ꢀC, and a source CID of 10 V. Spectra were acquired in positive
ionization mode from 100 to 1000 m/z and full scan UV-mode.
The following compounds were prepared according to previously
described procedures: 1,2,5,6-tetrahydro-4H-pyrrolo[3,2,1-ij]quinolin-
1H), 8.54 (dd, 3J = 4.7 Hz, 4J = 1.4 Hz, 1H), 8.59 (d, 4J = 2.0 Hz, 1H). 13
C
NMR (125 MHz, CDCl3): δ 24.5, 27.7, 31.6, 45.5, 120.7, 122.4, 123.5,
124.7, 129.9, 133.5, 134.0, 136.7, 141.6, 148.1, 167.6. MS m/z 251.09
(MHþ). Purity by LC/MS: 100% (Rt = 7.00 min). Anal. calcd for
(C16H14N2O) C 76.78, H 5.64, N 11.19; found C 77.13, H 5.52, N
11.25.
9-Pyridin-3-yl-1,2,6,7-tetrahydro-5H-pyrido[3,2,1-ij]quinolin-3-one (5).
Compound 5 was obtained according to procedure B from 5a (266 mg,
1.0 mmol) and 3-pyridineboronic acid (92 mg, 0.75 mmol) after flash
chromatography on silica gel (petroleum ether/ethyl acetate, 2/3, Rf =
0.12) and crystallization from acetone/diethylether as colorless needles
1
(116 mg, 0.44 mmol, 59%), mp 122-123 ꢀC. H NMR (500 MHz,
CDCl3): δ 1.97 (m, 2H), 2.69 (t, 3J = 7.2 Hz, 2H), 2.86 (t, 3J = 6.3 Hz,
3
3
2H), 2.95 (t, J = 7.2 Hz, 2H), 3.90 (t, J = 6.1 Hz, 2H), 7.20-7.24
(m, 2H), 7.33 (ddd, 3J = 7.9 Hz, 3J = 4.7 Hz, 5J = 0.6 Hz, 1H), 7.81 (ddd,
3J = 7.9 Hz, 4J = 2.3 Hz, 4J = 1.5 Hz, 1H), 8.55 (dd, J = 4.8 Hz, 4J =
3
1.5 Hz, 1H), 8.80 (m, 1H). 13C NMR (125 MHz, CDCl3): δ 21.4, 25.4,
27.4, 31.4, 41.0, 123.5, 124.4, 125.9, 126.2, 126.4, 131.9, 133.8, 135.9,
136.2, 148.0, 148.2, 169.4. MS m/z 264.96 (MHþ). Purity by LC/MS:
99.61% (Rt = 7.93 min).
8-Isoquinolin-4-yl-1,2,5,6-tetrahydro-4H-pyrrolo[3,2,1-ij]quinolin-
4-one (13). Compound 13 was obtained according to procedure B
from 4a (252 mg, 1.0 mmol) and 4-isoquinolineboronic acid (227 mg,
0.9 mmol) after flash chromatography on silica gel (petroleum ether/
ethyl acetate, 2/3, Rf = 0.13) as colorless needles (93 mg, 0.31 mmol,
34%), mp 184-185 ꢀC. 1H NMR (500 MHz, CDCl3): δ 2.75 (t, 3J = 7.8
Hz, 2H), 3.05 (t, 3J = 7.8 Hz, 2H), 3.28 (t, 3J = 8.4 Hz, 2H), 4.16 (t, 3J =
8.4 Hz, 2H), 7.13 (s, 1H), 7.20 (s, 1H), 7.62 (m, 1H), 7.67 (m, 1H), 7.91
(d, 3J = 8.5 Hz, 1H), 8.02 (d, 3J = 7.9 Hz, 1H), 8.44 (s, 1H), 9.22 (s, 1H).
2315
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
13C NMR (125 MHz, CDCl3): δ 24.4, 27.8, 31.6, 45.4, 120.2, 124.7,
125.1, 127.1, 127.4, 127.9, 128.4, 129.3, 130.5, 132.2, 133.2, 134.2, 141.3,
142.8, 151.8, 167.6. MS m/z 300.95 (MHþ). Purity by LC/MS: 100%
(Rt = 8.72 min).
counterbalanced system and tubing to perform blood sampling in the
freely moving rat. Separate stock solutions (5 mg/mL) were prepared
for the tested compounds in labrasol/water (1:1; v/v), leading to a clear
solution. Immediately before application, the cassette dosing mixture
was prepared by adding equal volumes of the stock solutions to end up
with a final concentration of 1 mg/mL for each compound. The mixture
was applied perorally to four rats with an injection volume of 5 mL/kg
(Time 0). Blood samples (250 μL) were collected 1 h before application
and 1, 2, 4, 6, 8, and 24 h thereafter. They were centrifuged at 650g for
10 min at 4 ꢀC, and then the plasma was harvested and kept at -20 ꢀC
until LC/MS analysis. To 50 μL of rat plasma sample and calibration
standard, 100 μL of acetonitrile containing the internal standard was
added. Samples and standards were vigorously shaken and centrifuged
for 10 min at 6000g and 20 ꢀC. For the test items, an additional dilution
was performed by mixing 50 μL of the particle free supernatant with
50 μL of water. An aliquot was transferred to 200 μL sampler vials and
subsequently subjected to LC-MS/MS. HPLC-MS/MS analysis and
quantification of the samples was carried out on a Surveyor-HPLC
system coupled with a TSQ Quantum (ThermoFinnigan) triple quad-
rupole mass spectrometer equipped with an electrospray interface (ESI).
The mean of absolute plasma concentrations ((SEM) was calculated
for the four rats, and the regression was performed on group mean
values. The pharmacokinetic analysis was performed using a noncom-
partment model (PK Solutions 2.0, Summit Research Services).
Computational Methods. (1) Pharmacophore modeling. The
ligands of both the training set and the test database were built in
MOE,43 minimized, and conformers then generated using MOE’s
Conformation Import tool. Structural data of the enzyme, such as heme
complexation and conformational hindrances of active site residues,
were used to direct the preparation of both ligands and pharmacophore
queries. First, we superimposed all high-active compounds of the
training set according to the pharmacophore hypothesis of Lucas
et al.,17 and then we set up a preliminary pharmacophore query using
the Pharmacophore Consensus application of MOE. This utility calcu-
lates a query composed of all features of all molecules, with an indication
of the proportion of shared features. The whole training database was
now searched by using the preliminary query, and the results of the
search were then examined to suggest modifications to the query in
order to increase the recognition rate of highly active compounds.
Finally, the best results were obtained when the pharmacophore query
was not overconstrained, hence when AA1/HY0 and HY1 were marked
as essential and a partial match of a minimum of 7 (out of 11) features
was permitted. A visual inspection of the output of the search revealed
molecules that should be ruled out for steric reasons. An additional
volume constraint was therefore added to the query to help further
constrain the search. Using the Union feature of the Pharmacophore
Query Editor, excluded five volume spheres were positioned coincident
with atoms in the active site of the homology model of CYP11B2. (2)
Protein modeling and docking. A multitemplate strategy was followed to
build a set of homology models of CYP11B2 for which the crystal
structures of the human CYP19 (3eqm), 3A4 (1tqn), 1A2 (2hi4), 2E1
(3e6i), 2C9 (1og2), and 2D6 (2f9q) enzymes were used. CYP19,
CYP2C9, and CYP3A4 have been trialed out as core template, necessary
for a correct alignment of all the different CYPs, with CYP2C9 resulting
the most appropriate one. Sequence alignment of CYP19, 3A4, 1A2,
2E1, 2C9, and 2D6 was accomplished by means of T-COFFEE/
ESPRESSO44 and final manual adjustments. The latter was strongly
supported by secondary structure predictions by means of Jpred45
and PSIPRED.46 The homology models were generated with
MODELLER9v747 and evaluated by means of Procheck,48
Prosa2003,49,50 and DOPE score. Best ranked models were then further
validated by docking a series of aldosterone synthase within a large IC50
range using FlexX docking software.51 In this study, selected compounds
were docked into the refined homology model using FlexX-pharm. A
8-(5-Phenylpyridin-3-yl)-1,2,5,6-tetrahydro-4H-pyrrolo[3,2,1-ij]-
quinolin-4-one (18). Compound 18 was obtained according to proce-
dure A from 6a (325 mg, 1.07 mmol) and 3-bromo-5-phenylpyridine
(301 mg, 1.28 mmol) after flash chromatography on silica gel
(petroleum ether/ethyl acetate, 2/3, Rf = 0.09) as colorless plates
1
(150 mg, 0.46 mmol, 43%), mp 188-189 ꢀC. H NMR (500 MHz,
CDCl3): δ 2.73 (t, 3J = 7.7 Hz, 2H), 3.05 (t, 3J = 7.9 Hz, 2H), 3.26 (t, 3J =
8.5 Hz, 2H), 4.14 (t, 3J = 8.5 Hz, 2H), 7.26 (s, 1H), 7.34 (s, 1H), 7.42 (m,
1H), 7.49 (m, 2H), 7.63 (dd, 3J = 6.9 Hz, 4J = 1.6 Hz, 2H), 7.98 (m, 1H),
8.75 (d, 4J = 1.9 Hz, 1H), 8.78 (d, 4J = 1.9 Hz, 1H). 13C NMR (125 MHz,
CDCl3): δ 24.5, 27.8, 31.6, 45.5, 120.7, 122.4, 124.8, 127.2, 128.3, 129.1,
129.9, 132.7, 133.3, 136.7, 136.8, 137.7, 141.7, 146.5, 146.6, 167.5. MS
m/z 327.07 (MHþ). Purity by LC/MS: 99.60% (Rt = 13.74 min).
8-[5-(3-Fluorophenyl)-pyridin-3-yl]-1,2,5,6-tetrahydro-4H-pyrrolo-
[3,2,1-ij]quinolin-4-one (20). Compound 20 was obtained according
to procedure A from 6a (360 mg, 1.20 mmol) and 20a (378 mg, 1.50
mmol) after flash chromatography on silica gel (petroleum ether/ethyl
acetate, 2/3, Rf = 0.08) as colorless needles (97 mg, 0.28 mmol, 23%),
mp 181-182 ꢀC. 1H NMR (500 MHz, CDCl3): δ 2.73 (t, 3J = 7.9 Hz,
2H), 3.05 (t, 3J = 7.9 Hz, 2H), 3.27 (t, 3J = 8.5 Hz, 2H), 4.14 (t, 3J = 8.5
Hz, 2H), 7.12 (m, 1H), 7.26 (s, 1H), 7.31-7.35 (m, 2H), 7.40-7.48 (m,
2H), 7.96 (m, 1H), 8.76 (d, 4J = 2.0 Hz, 1H), 8.77 (d, 4J = 2.2 Hz, 1H).
13C NMR (125 MHz, CDCl3): δ 24.5, 27.7, 31.6, 45.5, 120.8, 114.2 (d,
2JC,F = 22.0 Hz), 115.2 (d, 2JC,F = 21.9 Hz), 122.5, 122.9, 124.8, 130.0,
130.7 (d, 3JC,F = 8.3 Hz), 132.7, 133.0, 135.5, 136.9, 139.9 (d, 3JC,F = 7.3
1
Hz), 141.9, 146.2, 147.0, 163.3 (d, JC,F = 247 Hz), 167.5. MS m/z
345.05 (MHþ). Purity by LC/MS: 98.75% (Rt = 11.62 min). Anal. calcd
for (C22H17FN2O) C 76.73, H 4.98, N 8.13; found C 76.54, H 5.19,
N 8.28.
Biological Methods. (1) Enzyme preparations. The inhibition of
CYP17 was investigated using the 50000g sediment of the E. coli
homogenate recombinantly expressing human CYP17 and progesterone
(25 μM) as substrate.41 The inhibition values were measured at an
inhibitor concentration of 2 μM. The inhibition of CYP19 at an inhibitor
concentration of 500 nM was determined in vitro with human placental
microsomes and [1β-3H]androstenedione as substrate as described by
Thompson and Siiteri42 using our modification.11,36 Inhibition of
hepatic CYP enzymes was measured using recombinantly expressed
enzymes from baculovirus-infected insect microsomes (Supersomes).
(2) Enzyme assays. The following enzyme assays were performed as
previously described: CYP17 and CYP19. (3) Activity and selectivity
assay using V79 cells. V79 MZh 11B1 and V79 MZh 11B2 cells10,36 were
incubated with [4-14C]-11-deoxycorticosterone as substrate and inhi-
bitor in at least three different concentrations. The enzyme reactions
were stopped by addition of ethyl acetate. After vigorous shaking and a
centrifugation step (10000g, 2 min), the steroids were extracted into the
organic phase, which was then separated. The conversion of the
substrate was analyzed by HPTLC and a phosphoimaging system as
described.10,36 (4) Inhibition of human hepatic CYP enzymes. The
recombinantly expressed enzymes from baculovirus-infected insect
microsomes (Supersomes) were used and the manufacturer’s instruc-
tions (www.gentest.com) were followed. (5) In vivo pharmacokinetics.
Animal trials were conducted in accordance with institutional and
international ethical guidelines for the use of laboratory animals. Male
Wistar rats weighing 260-280 g (Janvier, France) were housed in a
temperature-controlled room (20-24 ꢀC) and maintained in a 12 h
light/12 h dark cycle. Food and water were available ad libitum. The
animals were anaesthetised with a ketamine (90 mg/kg)/xylazine (10
mg/kg) mixture and cannulated with silicone tubing via the right jugular
vein. Prior to the first blood sampling, animals were connected to a
2316
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
directed heme-Fe-N interaction perpendicular to the heme plane was
set as constraint to ensure the right binding mode of the inhibitors with
the heme cofactor. All the resting ligand-protein interaction parameters
and geometries were used as the default. The protein-ligand interac-
tions were analyzed using the LigX module of MOE software.
during angiotensin-converting enzyme inhibitor therapy in essential
hypertensive patients with left ventricular hypertrophy. J. Int. Med. Res.
2001, 29, 13–21.
(4) (a) Pitt, B.; Zannad, F.; Remme, W. J.; Cody, R.; Castaigne, A.;
Perez, A.; Palensky, J.; Wittes, J. The effect of spironolactone on
morbidity and mortality in patients with severe heart failure. N. Engl. J.
Med. 1999, 341, 709–717. (b) Pitt, B.; Remme, W.; Zannad, F.; Neaton,
J.; Martinez, F.; Roniker, B.; Bittman, R.; Hurley, S.; Kleiman, J.; Gatlin,
M. Eplerenone, a selective aldosterone blocker, in patients with left
ventricular dysfunction after myocardial infarction. N. Engl. J. Med. 2003,
348, 1309–1321.
(5) (a) Hu, X.; Bolten, C. W. Adrenal corticosteroids, their receptors
and hypertension. Drug Dev. Res. 2006, 67, 871–883. (b) Hu, X.; Li, S.;
McMahon, E. G.; Lala, D. S.; Rudolph, A. E. Molecular mechanisms of
mineralocorticoid receptor antagonism by eplerenone. Mini. Rev. Med.
Chem. 2005, 5, 709–718.
’ ASSOCIATED CONTENT
S
Supporting Information. Syntheses and spectroscopic
b
data of the target compounds 6-12, 14-17, 19, 22-32, full
experimental details and spectroscopic characterization of the
reaction intermediates 4a-6a, 9a, 19a-30a, 32a, 32b, and
pharmacophore geometric properties. This material is available
free of charge via the Internet at http://pubs.acs.org.
(6) Juurlink, D. N.; Mamdani, M. M.; Lee, D. S.; Kopp, A.; Austin,
P. C.; Laupacis, A.; Redelmeier, D. A. Rates of hyperkalemia after
publication of the Randomized Aldactone Evaluation Study. N. Engl. J.
Med. 2004, 351, 543–551.
(7) Delcayre, C.; Swynghedauw, B. Molecular mechanisms of myo-
cardial remodeling. The role of aldosterone. J. Mol. Cell. Cardiol. 2002,
34, 1577–1584.
(8) (a) Wehling, M. Specific, nongenomic actions of steroid hor-
mones. Annu. Rev. Physiol. 1997, 59, 365–393. (b) L€osel, R.; Wehling, M.
Nongenomic actions of steroid hormones. Nature Rev. Mol. Cell. Biol.
2003, 4, 46–55.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: þ49-681-30270300. Fax: þ49-681-30270308. E-mail: rwh@
mx.uni-saarland.de. Homepage: http://www.PharmMedChem.de.
Present Addresses
§Gr€unenthal Pharma GmbH, Global Drug Discovery,
Zieglerstrasse 6, 52078 Aachen, Germany.
(9) Hartmann, R. W. Selective inhibition of steroidogenic P450
enzymes: current status and future perspectives. Eur. J. Pharm. Sci. 1994,
2, 15–16.
’ ACKNOWLEDGMENT
We thank Gertrud Schmitt and Jeannine Jung for their
help in performing the in vitro tests. We would also like to
acknowledge the undergraduate research participants Judith
Horzel, Helena R€ubel, and Michael Zender whose work con-
tributed to the presented results. The investigation of the hepatic
CYP profile by Dr. Ursula M€uller-Vieira and the pharmacokinetic
studies by Dr. Barbara Birk, Pharmacelsus CRO, Saarbr€ucken,
are highly appreciated. S.L. is grateful to Saarland University
for a scholarship (Landesgraduierten-F€orderung). Thanks are
due to Prof. J. J. Rob Hermans, University of Maastricht, The
Netherlands, for supplying the V79 CYP11B1 cells, and Prof.
Rita Bernhardt, Saarland University, for supplying the V79
CYP11B2 cells.
(10) Ehmer, P. B.; Bureik, M.; Bernhardt, R.; M€uller, U.; Hartmann,
R. W. Development of a test system for inhibitors of human aldosterone
synthase (CYP11B2): Screening in fission yeast and evaluation of
selectivity in V79 cells. J. Steroid Biochem. Mol. Biol. 2002, 81, 173–179.
(11) (a) Hartmann, R. W.; Batzl, C. Aromatase inhibitors. Synthesis
and evaluation of mammary tumor inhibiting activity of 3-alkylated 3-(4-
aminophenyl)piperidine-2,6-diones. J. Med. Chem. 1986, 29, 1362–
1369. (b) Le Borgne, M.; Marchand, P.; Duflos, M.; Delevoye-Seiller,
B.; Piessard-Robert, S.; Le Baut, G.; Hartmann, R. W.; Palzer, M.
Synthesis and in vitro evaluation of 3-(1-azolylmethyl)-1H-indoles
and 3-(1-azolyl-1-phenylmethyl)-1H-indoles as inhibitors of P450 arom.
Arch. Pharm. (Weinheim, Germany) 1997, 330, 141–145.
(12) (a) Hille, U. E.; Hu, Q.; Vock, C.; Negri, M.; Bartels, M.;
M€uller-Vieira, U.; Lauterbach, T.; Hartmann, R. W. Novel CYP17
inhibitors: synthesis, biological evaluation, structure-activity relation-
ships and modelling of methoxy- and hydroxy-substituted methylenei-
midazolyl biphenyls. Eur. J. Med. Chem. 2009, 44, 2765–2775. (b) Hu,
Q.; Yin, L.; Jagusch, C.; Hille, U. E.; Hartmann, R. W. Isopropylidene
substitution increases activity and selectivity of biphenylmethylene
4-pyridine type CYP17 inhibitors. J. Med. Chem. 2010, 53, 5049–5053.
(13) Kawamoto, T.; Mitsuuchi, Y.; Toda, K.; Yokoyama, Y.; Miyahara,
K.; Miura, S.; Ohnishi, T.; Ichikawa, Y.; Nakao, K.; Imura, H.; Ulick, S.;
Shizuta, Y. Role of steroid 11β-hydroxylase and steroid 18-hydroxylase in
the biosynthesis of glucocorticoids and mineralocorticoids in humans. Proc.
Natl. Acad. Sci. U.S.A. 1992, 89, 1458–1462.
(14) (a) Hahner, S.; Stuermer, A.; Kreissl, M.; Reiners, C.;
Fassnacht, M.; Haenscheid, H.; Beuschlein, F.; Zink, M.; Lang, K.;
Allolio, B.; Schirbel, A. [123I]Iodometomidate for Molecular Imaging
of Adrenocortical Cytochrome P450 Family 11B Enzymes. J. Clin.
Endocrinol. Metab. 2008, 93, 2358–65. (b) Zolle, I. M.; Berger, M. L.;
Hammerschmidt, F.; Hahner, S.; Schirbel, A.; Peric-Simov, B. New
selective inhibitors of steroid 11β-hydroxylation in the adrenal cortex.
Synthesis and structure-activity relationship of potent etomidate
analogues. J. Med. Chem. 2008, 51, 2244–2253.
(15) Bassett, M. H.; Mayhew, B.; Rehman, K.; White, P. C.;
Mantero, F.; Arnaldi, G.; Stewart, P. M.; Bujalska, I.; Rainey, W. E.
Expression profiles for steroidogenic enzymes in adrenocortical disease.
J. Clin. Endocrinol. Metab. 2005, 90, 5446–5455.
’ ABBREVIATIONS USED
Ang II, angiotensin II; AUC, area under the curve; BPin, pinacol
boronate; CHF, congestive heart failure; CYP, cytochrome
P450; CYP11B1, 11β-hydroxylase; CYP11B2, aldosterone
synthase; CYP17, 17R-hydroxylase-17,20-lyase; CYP19, aroma-
tase; Het, heteroaryl; MR, mineralocorticoid receptor; NBS, N-
bromosuccinimide; RAAS, renin-angiotensin-aldosterone sys-
tem; SAR, structure-activity relationship; HY, hydrophobic or
aromatic pharmacophore feature; AA, acceptor atom feature
’ REFERENCES
(1) Weber, K. T. Aldosterone in congestive heart failure. N. Engl. J.
Med. 2001, 345, 1689–1697.
(2) (a) Brilla, C. G. Renin-angiotensin-aldosterone system and
myocardial fibrosis. Cardiovasc. Res. 2000, 47, 1–3. (b) Lijnen, P.; Petrov,
V. Induction of cardiac fibrosis by aldosterone. J. Mol. Cell. Cardiol. 2000,
32, 865–879.
(3) (a) Struthers, A. D. Aldosterone escape during angiotensin-
converting enzyme inhibitor therapy in chronic heart failure. J. Card.
Failure 1996, 2, 47–54. (b) Sato, A.; Saruta, T. Aldosterone escape
2317
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
(16) Taymans, S. E.; Pack, S.; Pak, E.; Torpy, D. J.; Zhuang, Z.;
Stratakis, C. A. Human CYP11B2 (aldosterone synthase) maps to
chromosome 8q24.3. J. Clin. Endocrinol. Metab. 1998, 83, 1033–1036.
(17) Ulmschneider, S.; M€uller-Vieira, U.; Mitrenga, M.; Hartmann,
R. W.; Oberwinkler-Marchais, S.; Klein, C. D.; Bureik, M.; Bernhardt, R.;
Antes, I.; Lengauer, T. Synthesis and evaluation of imidazolylmethyle-
netetrahydronaphthalenes and imidazolylmethyleneindanes: Potent in-
hibitors of aldosterone synthase. J. Med. Chem. 2005, 48, 1796–1805.
(18) Ulmschneider, S.; M€uller-Vieira, U.; Klein, C. D.; Antes,
I.; Lengauer, T.; Hartmann, R. W. Synthesis and evaluation of
(pyridylmethylene)tetrahydronaphthalenes/-indanes and structurally
modified derivatives: potent and selective inhibitors of aldosterone
synthase. J. Med. Chem. 2005, 48, 1563–1575.
(19) Voets, M.; Antes, I.; Scherer, C.; M€uller-Vieira, U.; Biemel, K.;
Barassin, C.; Oberwinkler-Marchais, S.; Hartmann, R. W. Heteroaryl
substituted naphthalenes and structurally modified derivatives: selective
inhibitors of CYP11B2 for the treatment of congestive heart failure and
myocardial fibrosis. J. Med. Chem. 2005, 48, 6632–6642.
(20) Lucas, S.; Heim, R.; Negri, M.; Antes, I.; Ries, C.; Schewe, K. E.;
Bisi, A.; Gobbi, S.; Hartmann, R. W. Novel aldosterone synthase inhibitors
with extended carbocyclic skeleton by a combined ligand-based and
structure-based drug design approach. J. Med. Chem. 2008, 51, 6138–6149.
(21) Voets, M.; Antes, I.; Scherer, C.; M€uller-Vieira, U.; Biemel, K.;
Oberwinkler-Marchais, S.; Hartmann, R. W. Synthesis and evaluation of
heteroaryl-substituted dihydronaphthalenes and indenes: Potent and
selective inhibitors of aldosterone synthase (CYP11B2) for the treat-
ment of congestive heart failure and myocardial fibrosis. J. Med. Chem.
2006, 49, 2222–2231.
(22) (a) Belkina, N. V.; Lisurek, M.; Ivanov, A. S.; Bernhardt, R.
Modelling of three-dimensional structures of cytochromes P450 11B1
and 11B2. J. Inorg. Biochem. 2001, 87, 197–207. (b) Lewis, D. F. V.; Lee-
Robichaud, P. Molecular modelling of steroidogenic cytochromes P450
from families CYP11, CYP17, CYP19 and CYP21 based on the CYP102
crystal structure. J. Steroid Biochem. Mol. Biol. 1998, 66, 217–233. (c)
Krone, N.; Riepe, F. G.; G€otze, D.; Korsch, E.; Rister, M.; Commentz, J.;
Partsch, C.-J.; Gr€otzinger, J.; Peter, M.; Sippell, W. G. Congenital
adrenal hyperplasia due to 11-hydroxylase deficiency: functional char-
acterization of two novel point mutations and a three-base pair deletion
in the CYP11B1 gene. J. Clin. Endocrinol. Metab. 2005, 90, 3724–3730.
(d) Roumen, L. L.; Sanders, M. P. A.; Koen, P.; Hilbers, P. A. J.; Plate, R.;
Custers, E.; de Gooyer, M.; Smits, J. F. M.; Beugels, I.; Emmen, J.;
Ottenheijm, H. C. J.; Leysen, D.; Hermans, J. J. R. Construction of 3D
models of the CYP11B family as a tool to predict ligand binding
characteristics. J. Comput.-Aided Mol. Des. 2007, 21, 455–471.
(23) Ulmschneider, S.; Negri, M.; Voets, M.; Hartmann, R. W. Devel-
opment and evaluation of a pharmacophore model for inhibitors of
aldosterone synthase (CYP11B2). Bioorg. Med. Chem. Lett. 2006, 16, 25–30.
(24) Crabb, T. A.; Soilleux, S. L. Microbiological transformations.
Part 9: Microbiological transformations of 1,2,5,6-tetrahydropyrrolo-
[3,2,1-i,j]-quinolin-4-one and of derivatives of 1,2,3,5,6,7-hexahydro-
pyrido[3,2,1-i,j]quinoline with the fungus Cunninghamella elegans.
Tetrahedron 1986, 42, 5407–5413.
(25) Appukkuttan, P.; Orts, A. B.; Chandran, R., P.; Goeman, J. L.;
van der Eycken, J.; Dehaen, W.; van der Eycken, E. Generation of a small
library of highly electron-rich 2-(hetero)aryl-substituted phenethyla-
mines by the Suzuki-Miyaura reaction: a short synthesis of an
apogalanthamine analogue. Eur. J. Org. Chem. 2004, 3277–3285.
(26) Zhu, H.-F.; Zhao, W.; Okamura, T.; Fan, J.; Sun, W.-Y.;
Ueyama, N. Syntheses and crystal structures of 1D tubular chains and
2D polycatenanes built from the asymmetric 1-(1-imidazolyl)-
4-(imidazol-1-ylmethyl)benzene ligand with metal salts. New J. Chem.
2004, 28, 1010–1018.
(29) (a) Mornet, E.; Dupont, J.; Vitek, A.; White, P. C. Character-
ization of two genes encoding human steroid 11β-hydroxylase [P-
450(11)β]. J. Biol. Chem. 1989, 264, 20961–20967. (b) Kawamoto,
T.; Mitsuuchi, Y.; Toda, K.; Miyahara, K.; Yokoyama, Y.; Nakao, K.;
Hosoda, K.; Yamamoto, Y.; Imura, H.; Shizuta, Y. Cloning of cDNA and
genomic DNA for human cytochrome P-45011β. FEBS Lett. 1990, 296,
345–349. (c) Kawamoto, T.; Mitsuuchi, Y.; Ohnishi, T.; Ichikawa, Y.;
Yokoyama, Y.; Sumimoto, H.; Toda, K.; Miyahara, K.; Kuribayashi, I.;
Nakao, K. Cloning and expression of a cDNA for human cytochrome
P-450aldo as related to primary aldosteronism. Biochem. Biophys. Res.
Commun. 1990, 173, 309–316.
(30) (a) B€ottner, B.; Schrauber, H.; Bernhardt, R. Engineering a
mineralocorticoid- to a glucocorticoid-synthesizing cytochrome P450. J.
Biol. Chem. 1996, 271, 8028–8033. (b) B€ottner, B.; Bernhardt, R.
Changed ratios of glucocorticoid/mineralocorticoids caused by point
mutations in the putative I-helix regions of CYP11B1 and CYP11B2.
Endocr. Res. 1996, 22, 455–561. (c) Curnow, K. M.; Mulatero, P.;
Emeric-Blanchouin, N.; Aupetit-Faisant, B.; Corvol, P.; Pascoe, L. The
amino acid substitutions Ser288Gly and Val320Ala convert the cortisol
producing enzyme, CYP11B1, into an aldosterone producing enzyme.
Nature Struct. Biol. 1997, 4, 32–35. (d) Mulatero, P.; Curnow, K. M.;
Aupetit-Taisant, B.; Foekling, M. Recombinant CYP11B genes encode
enzymes that can catalyze conversion of 11-deoxycortisol to cortisol, 18-
hydroxycortisol, and 18-oxocortisol. J. Clin. Endocinol. Metab. 1998, 83,
3996–4001. (e) B€ottner, B.; Denner, K.; Bernhardt, R. Conferring
aldosterone synthesis to human CYP11B1 by replacing key amino acid
residues with CYP11B2-specific ones. Eur. J. Biochem. 1998, 252, 458–
466. (f) Bechtel, S.; Belkina, N.; Bernhardt, R. The effect of amino acid
substitutions I112P, D147E and K152N in CYP11B2 on the catalytic
activities of the enzyme. Eur. J. Biochem. 2002, 269, 1118–1127.
(31) Ghosh, D.; Griswold, J.; Erman, M.; Pangborn, W. Structural
basis for androgen specificity and estrogen synthesis in human aroma-
tase. Nature 2009, 457, 219–223.
(32) (a) Ksander, G. M.; Meredith, E.; Monovich, L. G.; Papillon, J.;
Firooznia, F.; Hu, Q.-Y. Preparation of condensed imidazole derivatives
for the inhibition of aldosterone synthase and aromatase. PCT Int. Appl.
WO2007024945, 2007. (b) Papillon J.; Ksander, G. M; Hu, Q.-Y.
Preparation of tetrahydroimidazo[1,5-a]pyrazine derivatives as aldos-
terone synthase and/or 11β-hydroxylase inhibitors. PCT Int. Appl.
WO2007139992, 2007. (c) Hu, Q.-Y.; Ksander, G. M. 4-Imidazolyl-
1,2,3,4-tetrahydroquinoline derivatives and their use as aldosterone/11-
beta-hydroxylase inhibitors. PCT Int. Appl. WO2008076860, 2008. (d)
Hu, Q.-Y.; Ksander, G. M. Organic compounds. PCT Int. Appl.
WO2008076862, 2008.
(33) Lucas, S.; Heim, R.; Ries, C.; Schewe, K. E.; Birk, B.; Hartmann,
R. W. In vivo active aldosterone synthase inhibitors with improved selectivity:
lead optimization providing a series of pyridine substituted 3,4-dihydro-1H-
quinolin-2-one derivatives. J. Med. Chem. 2008, 51, 8077–8087.
(34) Ries, C.; Lucas, S.; Heim, R.; Birk, B.; Hartmann, R. W.
Selective aldosterone synthase inhibitors reduce aldosterone formation
in vitro and in vivo. J. Steroid Biochem. Mol. Biol. 2009, 116, 121–126.
(35) (a) Hartmann, R. W.; Heim, R., Lucas, S. 6-Pyridin-3-yl-3,4-
dihydo-1H-quinoline-2-one derivatives and related compounds as inhibi-
tors of the human aldosterone synthase CYP11B2. PCT Int. Appl.
WO2009135651, 2009. (b) Hu, Q.; Hartmann, R. W. Unpublished results.
(36) (a) Denner, K.; Bernhardt, R. Inhibition studies of steroid
conversions mediated by human CYP11B1 and CYP11B2 expressed in
cell cultures. In Oxygen Homeostasis and Its Dynamics, 1st ed.; Ishimura,
Y., Shimada, H., Suematsu, M., Eds.; Springer-Verlag: Tokyo, Berlin,
Heidelberg, NY, 1998; pp 231-236. (b) Denner, K.; Doehmer, J.;
Bernhardt, R. Cloning of CYP11B1 and CYP11B2 from normal human
adrenal and their functional expression in COS-7 and V79 chinese
hamster cells. Endocr. Res. 1995, 21, 443–448.
(27) Ishiyama, T.; Murata, M.; Miyaura, N. Palladium(0)-catalyzed
cross-coupling reaction of alkoxydiboron with haloarenes: a direct
procedure for arylboronic esters. J. Org. Chem. 1995, 60, 7508–7510.
(28) Gotoh, O. Substrate recognition sites in cytochrome P450
family 2 (CYP2) proteins inferred from comparative analyses of amino
acid and coding nucleotide sequences. J. Biol. Chem. 1992, 267, 83–90.
(37) Lamberts, S. W.; Bruining, H. A.; Marzouk, H.; Zuiderwijk, J.;
Uitterlinden, P.; Blijd, J. J.; Hackeng, W. H.; de Jong, F. H. The new
aromatase inhibitor CGS-16949A suppresses aldosterone and cortisol
production by human adrenal cells in vitro. J. Clin. Endocrinol. Metab.
1989, 69, 896–901.
2318
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319

Journal of Medicinal Chemistry
ARTICLE
(38) Demers, L. M.; Melby, J. C.; Wilson, T. E.; Lipton, A.; Harvey,
H. A.; Santen, R. J. The effects of CGS 16949A, an aromatase inhibitor
on adrenal mineralocorticoid biosynthesis. J. Clin. Endocrinol. Metab.
1990, 70, 1162–1166.
(39) Heim, R.; Lucas, S.; Grombein, C. M.; Ries, C.; Schewe, K. E.;
Negri, M.; M€uller-Vieira, U.; Birk, B.; Hartmann, R. W. Overcoming
undesirable CYP1A2 inhibition of pyridylnaphthalene type aldosterone
synthase inhibitors: influence of heteroaryl derivatization on potency
and selectivity. J. Med. Chem. 2008, 51, 5064–5074.
(40) Hille, U. E.; Zimmer, C.; Vock, C. A.; Hartmann, R. W. First
selective CYP11B1 inhibitors for the treatment of cortisol-dependent
diseases. Med. Chem. Lett. 2011, 2, 2–6.
(41) (a) Ehmer, P. B.; Jose, J.; Hartmann, R. W. Development of a
simple and rapid assay for the evaluation of inhibitors of human 17R-
hydroxylase-C17,20-lyase (P450c17) by coexpression of P450c17 with
NADPH-cytochrome-P450-reductase in Escherichia coli. J. Steroid Bio-
chem. Mol. Biol. 2000, 75, 57–63. (b) Hutschenreuter, T. U.; Ehmer,
P. B.; Hartmann, R. W. Synthesis of hydroxy derivatives of highly potent
nonsteroidal CYP17 inhibitors as potential metabolites and evaluation
of their activity by a noncellular assay using recombinant enzyme. J.
Enzyme Inhib. Med. Chem. 2004, 19, 17–32.
(42) Thompson, E. A.; Siiteri, P. K. Utilization of oxygen and
reduced nicotinamide adenine dinucleotide phosphate by human pla-
cental microsomes during aromatization of androstenedione. J. Biol.
Chem. 1974, 249, 5364–5372.
(43) Molecular Operating Environment MOE 2009.10; CCGIM:
Quebec, Canada, 2008; http://www.chemcomp.com
(44) Notredame, C.; Higgins, D.; Heringa, J. T-Coffee: A novel
method for fast and accurate multiple sequence alignment. J. Mol. Biol.
2000, 302, 205–217.
(45) Cole, C.; Barber, J. D.; Barton, G. J. The JPRED 3 Secondary
structure prediction server. Nucleic Acids Res. 2008, 36, W197–W201.
(46) McGuffin, L. J.; Bryson, K.; Jones, D. T. The PSIPRED protein
structure prediction server. Bioinformatics 2000, 16, 404–405.
(47) Eswar, N.; Marti-Renom, M. A.; Webb, B.; Madhusudhan,
M. S.; Eramian, D.; Shen, M.; Pieper, U.; Sali, A. Current Protocols in
Bioinformatics, John Wiley & Sons, Inc.: New York, 2006.
(48) Laskowski, R. A.; MacArthur, M. W.; Moss, D. S.; Thornton,
J. M. PROCHECK: a program to check the stereochemical quality of
protein structures. J. Appl. Crystallogr. 1993, 26, 283–291.
(49) Wiederstein, M.; Sippl, M. J. ProSA-web: interactive web
service for the recognition of errors in three-dimensional structures of
proteins. Nucleic Acids Res. 2007, 35, W407–W410.
(50) Sippl, M. J. Recognition of Errors in Three-Dimensional
Structures of Proteins. Proteins 1993, 17, 355–362.
(51) Kramer, B.; Rarey, M.; Lengauer, T. Evaluation of the FlexX
incremental construction algorithm for protein-ligand docking. Pro-
teins: Struct., Funct., Genet. 1999, 37, 228–241.
2319
dx.doi.org/10.1021/jm101470k |J. Med. Chem. 2011, 54, 2307–2319