Structure-based design and optimization of potent renin inhibitors on 5- or 7-azaindole-scaffolds

Article abstract of DOI:10.1016/j.bmcl.2011.06.112

The selective inhibition of the aspartyl protease renin is of high interest to control hypertension and associated cardiovascular risk factors. Following on preceding contributions, we report herein on the optimization of two series of azaindoles to arrive at potent and non-chiral renin inhibitors. The previously discovered azaindole scaffold was further explored by structure-based drug design in combination with parallel synthesis. This results in the identification of novel 5- or 7-azaindole derivatives with remarkable potency for renin inhibition. The best compounds on both series show IC₅₀ values between 3 and 8 nM.

Full text of DOI:10.1016/j.bmcl.2011.06.112

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5487–5492
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure-based design and optimization of potent renin inhibitors
on 5- or 7-azaindole-scaffolds
Hans Matter a, , Bodo Scheiper a, Henning Steinhagen a, , Zsolt Böcskei b, Valérie Fleury c, Gary McCort c
⇑
a Sanofi-Aventis Deutschland GmbH, Chemical and Analytical Sciences, Building G878, D-65926 Frankfurt, Germany
b Sanofi-Aventis, Chemical and Analytical Sciences, 16, Rue d0Ankara, F-67080 Strasbourg, France
c Sanofi-Aventis, Cardiovascular Department,1, Avenue Pierre Brossolette, F-91385 Chilly-Mazarin, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The selective inhibition of the aspartyl protease renin is of high interest to control hypertension and asso-
ciated cardiovascular risk factors. Following on preceding contributions, we report herein on the optimi-
zation of two series of azaindoles to arrive at potent and non-chiral renin inhibitors. The previously
discovered azaindole scaffold was further explored by structure-based drug design in combination with
parallel synthesis. This results in the identification of novel 5- or 7-azaindole derivatives with remarkable
potency for renin inhibition. The best compounds on both series show IC50 values between 3 and 8 nM.
Ó 2011 Published by Elsevier Ltd.
Received 9 May 2011
Revised 24 June 2011
Accepted 26 June 2011
Available online 2 July 2011
Keywords:
Renin
Azaindole
Structure-based-design
Aspartyl-protease
The control of hypertension can considerably reduce the risk of
cardiovascular diseases like myocardial infarction, stroke, heart
failure and end-stage kidney disease.1,2 The endocrine renin-angio-
tensin system (RAS), regulating blood pressure and fluid electro-
lytes,3 is one of the major and most intensively studied systems
affecting the arterial blood pressure in humans. The aspartyl prote-
ase renin is released after activation of this system and then pro-
duces angiotensin I in a rate-limiting step. Consequently the
inhibition of this step is perceived as effective antihypertensive
strategy4, causing multiple research groups to optimize novel ser-
ies of renin inhibitors with oral bioavailability for medical therapy.
Aliskiren (SPP100) is today the first molecule of this class, which
has reached the market.5
In two recent contributions we reported on novel classes of non-
chiral indole-3-carboxamides6 and derived 4- and 6-azaindoles7 as
renin inhibitors. The combination of various phenylether or benzyl-
substituents directed towards the renin lipophilic S1 pocket com-
bined with adequate decoration at the indole or azaindole scaffold
resulted in significant improvement of binding affinity. The best
compounds from this study 1a–c display IC50 values of 0.002 or
derived by structure-based design in combination with efficient
parallel synthesis.
The superposition of X-ray crystal structures of our lead struc-
tures 1a and 1c in complex with renin (PDB 3OOT and 3SFC,
Fig. 2)6,7 reveals similar binding modes, thus forming a basis for
structure-based design. Hydrogen-bonds are observed from the
piperazine nitrogen to side chains of Asp32 and Asp215, for the car-
boxamide oxygen to Thr77-OcH and for either the OH group directly
or the 6-azaindole nitrogen indirectly contacting the flexible His287
side chain. Both phenyl rings at the indole N1- and C2-position are in-
volved in lipophilic interactions to their corresponding subpockets.
OH
H
N
NH2
O
O
O
O
NH2
O
Aliskiren
NH
NH
NH
O
O
O
N
N
N
O
0.003 lM, respectively (Fig. 1). In this publication we describe the
O
O
discovery, structure–activity-relationship and optimization of
5- or 7-azaindole-3-carboxamides as renin inhibitors, which are
N
N
N
N
N
F
F
F
⇑
Corresponding author. Tel.: +49 69 305 84329.
1a: 2 nM
1b: 2 nM
1c: 3 nM
E-mail address: Hans.Matter@sanofi-aventis.com (H. Matter).
Present address: Grünenthal Pharma GmbH, Global Drug Discovery, Zieglerstr. 6,
52099 Aachen, Germany.
Figure 1. Chemical structures of indole-3-carboxamide 1a and 6-azaindoles 1b and
1c as potent renin inhibitors.
0960-894X/$ - see front matter Ó 2011 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2011.06.112

5488
H. Matter et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5487–5492
chains towards appropriate subpockets. In particular we investi-
gated the possibility to form favorable polar interactions to the
binding site, thus for example to His287 imidazole, either directly
or via conserved water. This substitution will also modulate phys-
icochemical properties like lipophilicity of the lead series, which
might affect pharmacokinetic properties.
The synthetic pathway for the preparation of the 5- and 7-aza-
indole derivatives 3–7 is outlined in Scheme 1 for examples 5b and
5g. The 5-azaindole 2a was arylated under Copper catalysis13, then
oxidized with bromine and subsequently reduced according to the
procedure of Robinson.14 Vilsmeier–Haack formylation with con-
comitant chlorination yielded aldehyde 2b. Oxidation to the acid
and coupling with N-Boc-piperazine provided key intermediate
2c. The chloride 2c was then either coupled in a Palladium-medi-
ated reaction followed by deprotection to the corresponding ben-
zyl derivative 5g or submitted to nucleophilic substitution to
provide the arylether derivative 5b after cleavage of the protecting
group under acidic conditions. The synthesis of all other deriva-
tives 3 to 7 was carried out accordingly to this scheme. All substi-
Figure 2. Superimposed X-ray structures of indole 1a (PDB 3OOT, white carbon
atoms, 2.5 Å resolution) and 6-azaindole 1c (PDB 3SFC, orange carbon atoms, 2.1 Å
resolution) in complex with human renin. The binding site from the 1a complex is
shown as MOLCAD9 hydrogen bonding surface. Crystallographic water is indicated
by cyan spheres.
tuted piperazine derivatives for compound series
7 were
commercially available; N-Boc was also employed as protecting
group.
In particular a relevant CH–p
interaction to Tyr75 is observed in the
Table 1
S1 pocket as important contribution to affinity. The 3-fluorine sub-
stituent is situated in a lipophilic subpocket close to the Asp32 car-
boxylate, while the orientation of the 5-methyl substituent
differs.7 In contrast to 1a, the 5-methyl group in 1c points towards
the opposite site of the S1 pocket, thus contacting side chains of
Pro111 and Phe112, although dihedral angles of phenoxy- or benzyl
substituents differ by only 8°. The 7-benzyl-substituent at the 6-aza-
indole is oriented towards the more open binding site region, con-
tacting the side chains of Pro111 and Leu114. Furthermore this ring
In vitro activity for compounds 3a–3ia
NH
O
N
R
N
N
is engaged in an intramolecular p–p
interaction8 to the azaindole-
N1-phenyl ring, thus stabilizing the bioactive conformation.
Our previous discovery of potent 4- or 6-azaindoles with lower
lipophilicity prompted us to explore other scaffolds with respect to
renin activity. Docking10–12 suggested either 5- or 7-positions at
the indole core to be suitable for introducing nitrogen atoms,
which leads to frameworks orienting the privileged indole side
Compds
R
Renin IC50
1.830
(lM)
O
O
O
3a
CH3
F
3b
3c
3d
0.098
5.06
NBoc
O
O
N
N
N
N
a-d
e, f
i, h
Cl
Cl
O
O
N
H
N
N
F
F
CH3
0.231
2a
2b
2c
CH3
g, h
3e
3f
0.396
0.17
NH
O
NH
O
O
N
N
CH3
N
N
O
F
N
N
F
O
5b
5g
CH3
F
F
3g
3h
0.504
6.410
F
Scheme 1. Reagents and condition: Synthesis of compounds 5b and 5g. Reagents:
(a) Phenyl iodide, CuI,(1S,2S)-(+)-Diaminocyclohexane, K3PO4, dioxane, 110 °C, 99%;
(b) Br2, t-BuOH, H2O, rt; (c) Pd/C, H2, EtOH, rt; (d) (i) POCl3, DMF, DCM, pyridine,
0 °C?rt, (ii) POCl3, 100 °C, 28% over three steps; (e) NaClO2, NaH2PO4, t-BuOH, rt,
quant.; (f) N-Boc-piperazine, TOTU, NMM, DMF, rt, 44%; (g) (i) 3-Fluoro-2-methyl-
benzylbromide, Zn, THF, 0 °C, (ii) 9-MeO-9-BBN, ꢀ78 °C?rt, (iii) Pd(OAc)2, S-Phos,
DMF, 100 °C, 26%; (h) TFA, DCM, rt, quant.; (i) 5–Fluoro-2-methylphenol, NaH, NMP,
rt?140 °C, microwave, 77%.
Benzyl
CH3
F
3i
0.086
a
IC50 data were obtained as described in the text and in Ref. 16.

H. Matter et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5487–5492
5489
Table 2
0.006
l
M each. However, the addition of a 5-methoxy substituent
In vitro activity for compounds 4a–4ia
resulted in 4c with an IC50 value of 4.670
lM, which could be ex-
plained by the lacking hydrogen-bond donor at this position plus
additional increased steric bulk.
However, shifting the methoxy substituent to position 6 at the
7-azaindole resulted in very potent compounds, exemplified by 4d
NH
F
O
N
O
O
CH3
F
CH3 H3C
CH3
R1
R2
N
N
R2 = C
R2 = A
R2 = B
Compds
R1
R2
Renin IC50 (lM)
4a
4b
4c
4d
4e
4f
4g
4h
4i
5-OH
5-OH
A
C
C
A
B
A
C
A
B
C
0.006
0.006
4.670
0.015
0.003
1.230
0.089
1.850
0.186
0.218
5-OCH3
6-OCH3
6-OCH3
4-Ethyl
4-Ethyl
4-Propyl
4-Propyl
4-Propyl
4j
a
IC50 data were obtained as described in the text and in Ref. 16.
First we decorated the 7-azaindole scaffold15 using privileged6,7
phenylether- and benzyl-derivatives. A representative set of com-
pounds is shown in Table 1 with experimental IC50 values ( M).
Figure 3. Binding pose of compounds 4e (upper: orange carbons) from docking into
the renin binding site from the X-ray crystal structure of 1a.
l
To calculate IC50 values, the activity of an inhibitor was evaluated
in a 10 point concentration range, in duplicate (intraplate).6,7,15,16
Two to four independent experiments were performed. The IC50
is expressed as geometric mean standard deviation (SD). The
standard deviation was low for this assay.
Table 3
In vitro activity for compounds 5a–5ha
The substituents in Table 1 are oriented towards the lipophilic
NH
O
N
renin S1 pocket to exploit the CH–p interaction to Tyr75 and the
N
affinity gain by addressing appropriate subpockets. Variations of
the aromatic ring in S1 then resulted in improved affinity com-
pared to the lead 3a with a phenoxy-group and an IC50 value of
R
N
1.830
6-azaindole structures.6,7 Comparing 3a with its related indole
shows a fourfold better activity for the indole (0.420
M).6 Adding
an ortho-methyl group to the phenoxy-ring resulted in compound
3b as most active derivative from the arylether-series with an IC50
lM, as also expected from optimizing the indole and 4- or
Compds
R
F
F
Renin IC50
0.847
(lM)
l
O
CH3
5a
value of 0.098
doles (0.021
3a or 3b does not improve affinity (3c, 5.06
l
M and thus a fivefold lower affinity as for the in-
M).6 The addition of fluorine in ortho position to
M; 3d, 0.231 M).
l
O
l
l
CH3
5b
5c
0.123
0.181
As for the phenylether subseries, the corresponding S1-directed
benzyl-substituents also exhibit consistently weaker affinity, when
comparing the 7-azaindole derivatives to the indole series.6 The
unsubstituted benzyl-derivative is about 70-fold less active in
O
CH3
F
the 7-azaindole series (3h, 6.410
ortho-methyl-meta-fluoro-substituents to the benzyl ring resulted
in 3i with an IC50 value of 0.086 M, thus being twofold more ac-
lM), while the addition of
O
l
CH3
tive as the 2-phenoxy-7-azaindole-derivative 3f. However, its sig-
nificantly lower activity compared to the indole, 4-or 6-azaindole
scaffolds was unexpected (six to nine fold lower affinity).
5d
5e
5f
0.65
3.4
F
Benzyl
The inspection of the X-ray crystal structure of 1a with renin
plus docking of selected 7-azaindoles led us to decorate the 7-aza-
indole using adequate substituents. From Table 1 and our previous
SAR investigations, we identified three adequate S1-directed phen-
oxy- and benzyl-moieties for further systematic variations. Those
moieties are shown in Table 2 (R2 = A, B or C) together with a selec-
tion of renin inhibitors from this optimization. First we introduced
a small-polar 5-hydroxy group, directed towards the His287 side
chain to allow for hydrogen-bond interactions, in particular as
CH3
0.520
CH3
F
5g
5h
0.008
0.10
Cl
F
hydrogen-bond donor to His287
.
The corresponding derivatives 4a and 4b with 5-OH groups at
the 7-azaindole scaffold were highly active with IC50 values of
a
IC50 data were obtained as described in the text and in Ref. 16.

5490
H. Matter et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5487–5492
Table 4
(0.015 lM) and 4e (0.003 lM). The compatibility of this binding
In vitro activity for compounds 6a–6ia
site region with larger and more lipophilic substituents like meth-
oxy is in agreement to our previous SAR investigations at the in-
dole scaffold.6 Interestingly the analogous indoles to 4d and 4e
NH
O
N
are two to four fold less active with IC50 values of 0.024
0.012 M, respectively.
lM and
l
N
R2
R 3
O
The binding mode of 4e in the renin binding site from docking
in Figure 3 suggests a highly similar binding mode compared to
X-ray structures of related inhibitors. Characteristic interactions
essential for high-affinity binding are preserved, so for example
the piperazine-aspartate interactions as well as the bifurcated
hydrogen bond from the indole-3-carboxamide carbonyl to
N
R1
Compds
R1
R2
R3
Renin IC50
0.150
(lM)
CH3
F
6a
Cy
CH3
Thr77-OcH (distance 1.8 Å) and NH (2.2 Å). The 6-methoxy-substi-
tuent is situated in a more open binding site region surrounded by
the aromatic rings of His287 and Tyr220. Interestingly the ether oxy-
gen is only around 2.8–2.9 Å apart from the backbone amide pro-
tons of Ser219 and Tyr220. Hence, those might potentially serve as
H-bond donors for this relatively weak acceptor, while the distance
is outside the usual range for this interaction. In related X-ray
structures, structurally conserved water molecules are present in
this area, although often not close to the donors. In addition the
CH3
F
6b
6c
6d
6e
Ph
Ph
Ph
Cy
CH3
CH3
H
0.006
0.073
0.024
0.116
O
O
CH3
F
F
methoxy-oxygen is 3.6 Å away from the Ser219AO
cH in this flexi-
CH3
ble side chain, which also favors hydrogen bond interactions. The
lipophilic methoxy carbon in directed towards the plane of the aro-
matic Tyr220 side chain with a distance of 4.2 Å to the nearest aro-
matic carbon.
Although the further analysis of our X-ray structures suggested
that lipophilic substituents in position 4 of the 7-azaindole scaffold
might be favorable for affinity, the introduction of both an ethyl or
propyl-side chain did not result in an increase of affinity. While 4f
with a 4-ethyl substituent and a 2-methyl-5-fluoro-phenylether
CH3
H
F
CH3
6f
Ph
H
0.088
F
substituent has an IC50 value of 1.230
methyl-3-fluoro-benzyl substituted 4g shows an IC50 value of
0.089 M, respectively. Hence for the benzyl series, the renin affin-
lM, the corresponding 2-
O
H3C
H3C
H3C
CH3
CH3
l
6g
6h
Cy
Cy
CH3
H
0.099
0.080
ity is related to the unsubstituted 3i, while in the phenylether sub-
series, there was a significant loss of affinity, when comparing to
3e. The larger 4-propyl substitution now resulted in derivatives
O
O
with lower affinity (4j, 0.218 lM).
Next we explored a related set of S1-directed phenylether- and
benzyl variations at a 5-azaindole structure15 in order to evaluate,
whether the more polar aromatic nitrogen in this position could
have a positive influence due to hydrogen-bonding in this area of
the binding site as well as to general physicochemical properties.
The representative set of derivatives is shown in Table 3. Consis-
tently with other series, the 2-methyl-5-fluoro-phenylether substi-
tuent leads to the highly active compound 5b for the phenylether
CH3
F
6i
Cy
H
0.098
a
IC50 data were obtained as described in the text and in Ref. 16.
subseries with an IC50 value of 0.123
ity is higher for this S1 substituent in the 5-azaindole series com-
pared to the 7-azaindoles (3e, 0.396 M), while for the indole
scaffold, an affinity of 0.005 M was observed (entry 2j in Ref.6).
lM. Hence, the binding affin-
l
l
This finding is similar for the benzyl-subseries. While for the 5-
azaindoles the best affinity in Table 3 was observed for the 2-
methyl-3-fluoro-benzyl substituent directed to S1 (5g, 0.008
the affinity is lower for the 7-azaindole (3i, 0.086 M). Interestingly
for this case, the corresponding indole derivative is equipotent to 5g
(entry 2n, 0.009 M in Ref. 6).Summarizes our investigation to add
lM),
l
l
6-hydroxy- or 6-methoxy substitutions to the 5-azaindole. In addi-
tion we replaced the aromatic azaindole N1-phenyl substituent
against a larger cyclohexyl-substituent. The computational investi-
gation of tautomer ratios for 6-hydroxy-5-azaindoles suggested the
pyridone-form to be exclusively present in solution.17 This pyridone
would thus allow the 5-amide-hydrogen to act as H-bond donor for
His287. The corresponding derivative 6d is more active than the
Figure 4. Binding poses of compounds 6b (orange carbons) and 6a (white carbons)
from docking into the renin binding site from the X-ray crystal structure of 1a.
unsubstituted 5-azaindole 5b (0.024 versus 0.123 lM). Replacing
the 5-OH group versus a 5-methoxy-group, which does not form a
0.073
lM). Interestingly, this SAR trend is reversed, when compar-
ing these changes for the 2-methyl-3-fluoro-benzyl substituents.
pyridone tautomer, is found to be slightly less active (6c,

H. Matter et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5487–5492
5491
Table 5
Here the derivative 6b with a 6-methoxy-5-azaindole core and an
IC50 value 0.006 M is significantly more active as the pyridone 6f
(0.088 M). This preference for small lipophilic methoxy-substitu-
ents in the 6-position has also been observed for the corresponding
In vitro activity for compounds 7a–7ga
l
l
O
R3
O
O
4-azaindole derivatives (c.f. 4e, 0.003 lM).
N
The replacement of the N1-phenyl-ring against the larger cyclo-
R1
CH3
F
CH3 H3C
F
CH3
R 2
hexyl (R1 = Cy in Table 4) results in reduced affinity, when compar-
ing 6b to 6a with IC50 values of 0.006 and 0.150 lM, respectively.
The binding mode for both derivatives in Figure 4 from docking
suggests a better fit of the aromatic N1-substituent in 6b in the
S3 pocket, which is lined also by the aromatic ring of Phe112. The
distance of both centroids is 5.1 Å, which also suggest some
N
O
F
R1 = C
R1 = A
R1 = B
Compds
R1
R2
R3
Renin IC50
0.011
(lM)
H
N
involvement from
p p interactions in this S3 pocket. In contrast,
ꢀ
7a
A
B
C
C
C
C
C
H
the cyclohexyl-group, although situated in a lipophilic environ-
ment, is oriented more towards the open space of the pocket due
to steric constraints.
N
H
N
For the pyridone-derivative, this decrease in affinity is less dra-
matic, when comparing 6f (0.088 lM) to 6e (0.116 lM). This lower
7b
7b
7d
7e
7f
H
0.037
0.32
0.154
4.09
3.26
1.1
N
renin affinity, however, is still consistent with observations for the
other scaffolds underscoring the hydrophobic nature of this inter-
action in the renin S3-subpocket. Obviously this increased size of
the N1-substituent is compatible with only some substitution mo-
tifs, while other motifs might alter the overall binding mode, thus
repositioning this group in S3. We then evaluated N1-cyclohexyl
derivatives at the 6-hydroxy-5-azaindole scaffold further by ade-
quate S1-directed phenylether substituents.
H
N
CH3
N
H
N
H
N
H
N
Active compounds were obtained for example using a 2,6-di-
O
methyl-phenylether (6h, 0.080
phenylether motif (6i, 0.098 M). Replacing the 6-OH, which again
is present in its pyridone tautomeric form, by a 6-methoxy-substi-
tuent does not significantly affect binding affinity (6g, 0.099 M),
as observed for 6d (0.024 M, see above).
lM) or a 3-fluoro-2,6-dimethyl-
CH3
CH3
H
l
N
H
N
l
OH
OH
l
N
This N1-cyclohexyl-6-hydroxy-5-azaindole scaffold was also
chosen to explore variations at the piperazine substructure. Repre-
sentative examples are summarized in Table 5 with R1 = A, B or C
indicating the S1-directed substituents. These chiral substitutions
at the piperazine were introduced after comparing the X-ray struc-
ture of Aliskiren (PDB 2V0Z) to the indole binding pose.18 The bond
vector connecting the Aliskiren aminoethanol to its isobutyl-car-
boxyamide directed towards the renin S10-pocket is aligned with
the vector C2-Hpro-S next to the secondary amine function. Chiral
lipophilic piperazine substitutions directed towards the renin S10
pocket could also impact affinity.
H
N
7g
N
a
IC50 data were obtained as described in the text and in Ref. 16.
increased by adding a methoxy- or hydroxyl-substitution, as exem-
plified in derivatives 7e–g with IC50 values ranging from 1.1 to
4.09 lM, respectively.
Introducing the lipophilic S-methyl substituent at this scaffold
thus resulted in the potent derivatives 7a (0.011
Selected compounds were further profiled with IC50 values in
human or mouse plasma, Caco-2 permeability and physicochemi-
cal data19 (Table 6). The best compounds display IC50 values be-
tween 3 and 8 nM with only slight activity reduction by adding
human or mouse plasma. The Caco-2 Papp value for 6b on the 5-
azaindole scaffold furthermore suggests its potential for significant
intestinal absorption.
In summary we have reported the design and structure–activity
relationship of novel series of potent renin inhibitors based on 5-
or 7-azaindole scaffolds. Both scaffolds were developed further
after analysis of previous indole-3-carboxamides, 4- and 6-azain-
doles. In general structure-activity trends are correlated, while
lM) and 7b
(0.037 M). Comparing 7b to its parent structure 6i reveals a three-
l
fold increase in affinity by this chiral methyl group. However, such
an improvement was not observed for the benzyl subseries. Here
the derivative 7d (0.154
lM) is slightly less potent as the parent
compound 6e (0.116 M), again indicating a nonlinear SAR between
l
the phenylether- and benzyl-subseries. Replacing the 6-OH group
against a 6-methoxy-group in 7c resulted in lower affinity (7c,
0.32
(0.116
l
M), in agreement to observations for the parent molecule 6e
M) versus 6a (0.150 M), respectively. The renin affinity
l
l
was lower in general, if the chain length of the substituent was
Table 6
Further profiling of selected Renin inhibitorsa
Compds Renin
IC50
Renin IC50 with
M) human plasma (
Renin IC50 with
mouse plasma (
Caco-2 Permeability log P
pKa (basic, log D (pH
Ligand
efficiency
Lipophilic
ligand efficiency
(
l
lM)
lM)
(⁄10–7 cm/s)
(MoKa) MoKa)
7.4, MoKa)
4a
4b
5g
6b
0.006
0.006
0.008
0.006
0.189
0.095
0.200
0.328
0.057
0.159
0.410
0.364
4.5
nd
6.4
22
1.9
2.5
2.5
3.2
8.1
8.4
8.1
8.1
ꢀ0.5
ꢀ0.1
1.7
0.34
0.34
0.34
0.33
6.32
5.72
5.60
5.02
2.4
a
Data were obtained as described Ref. 19; nd = not determined.

5492
H. Matter et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5487–5492
9. Heiden, W.; Goetze, T.; Brickmann, J. J. Comp. Chem. 1993, 14, 246.
10. McMartin, C.; Bohacek, R. S. J. Comput.-Aided Mol. Des. 1997, 11, 333.
11. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.;
Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.;
Shenkin, P. S. J. Med. Chem. 2004, 47, 1739.
12. Modeling was done using Sybyl 8.0 (Tripos) employing the MMFF94s force
field (Halgren, T. J. Comp. Chem. 1999, 20, 720.). Protein–ligand complexes were
minimized using the quasi-Newton–Raphson procedure. For manual docking,
ligands were placed in the binding site and optimized treating all side chains
within 4 Å flexible. Automated docking was done using QXP10 or Glide.11
13. Filipski, K. J.; Kohrt, J. T.; Casimiro-Garcia, A.; Van Huis, C. A.; Dudley, D. A.;
Cody, W. L.; Bigge, C. F.; Desiraju, S.; Sun, S.; Maiti, S. N.; Jaber, M. R.; Edmunds,
J. J. Tetrahedron Lett. 2006, 47, 7677.
there is no strict linearity when transferring preferred substituents.
Hence, we explored both novel scaffolds with a privileged set of
building blocks directed to S1 in addition with other modifications
at the azaindole N1-position, the piperazine and the azaindole-6-
position.
After decoration of these scaffolds, compounds with high affin-
ity and favorable physicochemical properties were identified. In
particular, the incorporation of substituted phenoxy- or benzyl
derivatives in S1 combined with substitution at the azaindole core
led to a significant improvement of affinity. The availability of
X-ray structures of renin-inhibitor complexes during our optimiza-
tion allowed us to unveil the key determinants to affinity.
14. Robinson, R. P.; Donahue, K. M. J. Org. Chem. 1991, 56, 4805.
15. Steinhagen, H.; Scheiper, B.; Matter, H.; McCort, G. WO 2009095162; PCT Int.
Appl. 2009.
16. Inhibitory activity expressed as inhibition constant IC50 toward recombinant
human renin was determined using an assay in which the non-endogenous
Acknowledgments
fluorogenic substrate Dabcyl-c-Abu-Ile-His-Pro-Phe-His-Leu-Val-Ile-His-Thr-
EDANS is cleaved at the Leu-Val bond in analogy to angiotensinogen. Human
renin at a concentration of 5 nM is incubated with the test compound at
various concentrations and the substrate at a concentration of 10 lM (2 h at rt)
in 0.05 M Tris buffer (pH 8). The increase in fluorescence due to fluorescence
resonance energy transfer at an emission wavelength of 485 nM is recorded in
a microplate spectrofluorometer.15
The authors would like to thank Sabrina Berger, Holger Gaul,
Silvia Hein, Laetitia Martin, Andrea Müller, Mathilde Pierre and
Katja Wittmann for their technical assistance.
17. Tautomer ratios were computed using the Tauthor approach, as implemented
in MoKa (Molecular Discovery Ltd., version 1.1.0): Milletti, F.; Storchi, L.;
Sforna, G.; Cross, S.; Cruciani, G. J. Chem. Inf. Model. 2009, 49, 68.
18. Rahuel, J.; Rasetti, V.; Maibaum, J.; Rüeger, H.; Göschke, R.; Cohen, N.-C.; Stutz,
S.; Cumin, F.; Fuhrer, W.; Wood, J. M.; Grütter, M. G. Chem. Biol. 2000, 7, 493.
References and notes
1. Kearney, P. M.; Whelton, M.; Reynolds, K.; Muntner, P.; Whelton, P. K.; He, J.
Lancet 2005, 365, 217.
2. (a) Amin Zaman, M.; Oparil, S.; Calhoun, D. A. Nat. Rev. Drug Disc. 2002, 1, 621;
(b) Jensen, C.; Herold, P.; Brunner, H. R. Nat. Rev. Drug Disc. 2008, 7, 399.
3. MacGregor, G. A.; Markandu, N. D.; Roulston, J. E.; Jones, J. E.; Morton, J. J.
Nature 1981, 291, 329.
19.
a
Recombinant human renin is incubated with the inhibitor at various
concentrations plus the substrate Dabcyl- -Abu-Ile-His-Pro-Phe-His-Leu-Val-
Ile-His-Thr-EDANS at 25 M for 30 min at 37 C and 30 min at rt in human or
c
l
mouse (C57BL6) plasma. Fluorescence is recorded at an excitation wavelength
of 330 nM and an emission wavelength of 485 nM. IC50 values are calculated
from inhibition as function of concentration from multiple measurements.; b
4. Wood, J. M.; Stanton, J. L.; Hofbauer, K. G. J. Enzyme Inhib. Med. Chem. 1987, 1,
169.
5. (a) Göschke, R.; Stutz, S.; Rasetti, V.; Cohen, N.-C.; Rahuel, J.; Rigollier, P.; Baum,
H.-P.; Forgiarini, P.; Schnell, C. R.; Wagner, T.; Grütter, M. G.; Fuhrer, W.;
Schilling, W.; Cumin, F.; Wood, J. M.; Maibaum, J. J. Med. Chem. 2007, 50, 4818;
(b) Maibaum, J.; Stutz, S.; Göschke, R.; Rigollier, P.; Yamaguchi, Y.; Cumin, F.;
Rahuel, J.; Baum, H.-P.; Cohen, N.-C.; Schnell, C. R.; Fuhrer, W.; Grütter, M. G.;
Schilling, W.; Wood, J. M. J. Med. Chem. 2007, 50, 4832.
6. Scheiper, B.; Matter, H.; Steinhagen, H.; Stilz, U.; Böcskei, Z.; Fleury, V.; McCort,
G. Bioorg. Med. Chem. Lett. 2010, 20, 6268.
7. Scheiper, B.; Matter, H.; Steinhagen, H.; Böcskei, Z.; Fleury, V.; McCort, G.
Bioorg. Med. Chem. Lett. 2011, 21, in press. doi:10.1016/j.bmcl.2011.06.114.
8. (a) Burley, S. K.; Petsko, G. A. Science 1985, 229, 23; (b) McGaughey, G. B.;
Gagné, M.; Rappé, A. K. J. Biol. Chem. 1998, 273, 15458.
Permeability studies using Caco-2/TC7 cells are performed using 20 lM
compound concentration in the apical to basolateral direction with a pH
gradient (6.5–7.4) and a BSA gradient (0.5–5%). After 2 h at 37 °C, the apparent
permeability value (Papp) is obtained from quantification [⁄10ꢀ7 cm/s].;
Physico-chemical properties (pKa, log P, log D) are computed using MoKa
(Molecular Discovery Ltd., version 1.1.0.): (c) Milletti, F.; Storchi, L.; Sforna, G.;
Cruciani, G. J. Chem. Inf. Model. 2010, 47, 2172; Ligand and lipophilic ligand
efficiencies are computed according to: (d) Hopkins, A. L.; Groom, C. R.; Alex, A.
Drug Disc. Today 2004, 9, 430; Leeson, P. D.; Springthorpe, B. Nat. Rev. Drug Disc.
2007, 6, 881.