Structure-based design of 7-azaindole-pyrrolidine amides as inhibitors of 11β-hydroxysteroid dehydrogenase type i

Article abstract of DOI:10.1016/j.bmcl.2012.07.070

Indole-pyrrolidines were identified as inhibitors of 11β- hydroxysteroid dehydrogenase type 1 (11β-HSD1) by high-throughput screening. Optimisation of the initial hit through structure-based design led to 7-azaindole-derivatives, with the best analogues displaying single digit nanomolar IC₅₀ potency. The modeling hypotheses were confirmed by solving the X-ray co-crystal structure of one of the lead compounds. These compounds were selective against 11β-hydroxysteroid dehydrogenase type 2 (selectivity ratio >200) and exhibited good inhibition of 11β-HSD1 (IC₅₀ < 1 μM) in a cellular model (3T3L1 adipocytes).

Full text of DOI:10.1016/j.bmcl.2012.07.070

Bioorganic & Medicinal Chemistry Letters 22 (2012) 5909–5914
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure-based design of 7-azaindole-pyrrolidine amides as inhibitors
of 11b-hydroxysteroid dehydrogenase type I
Eric Valeur a, , Serge Christmann-Franck a, , Franck Lepifre a,à, Denis Carniato a, Daniel Cravo a,§
,
⇑
Christine Charon a,à, Sophie Bozec a,§, Djordje Musil b, Per Hillertz b,–, Liliane Doare a, Fabien Schmidlin a,k
,
Marc Lecomte a, Melanie Schultz b, Didier Roche a,
a Merck-Serono S.A., 9, Chemin des Mines, 1202 Geneva, Switzerland
b Merck-Serono, Frankfurterstrasse 250, 64293 Darmstadt, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Indole-pyrrolidines were identified as inhibitors of 11b-hydroxysteroid dehydrogenase type 1 (11b-
HSD1) by high-throughput screening. Optimisation of the initial hit through structure-based design led
to 7-azaindole-derivatives, with the best analogues displaying single digit nanomolar IC50 potency. The
modeling hypotheses were confirmed by solving the X-ray co-crystal structure of one of the lead com-
pounds. These compounds were selective against 11b-hydroxysteroid dehydrogenase type 2 (selectivity
Received 17 July 2012
Accepted 19 July 2012
Available online 31 July 2012
ratio >200) and exhibited good inhibition of 11b-HSD1 (IC50 < 1 lM) in a cellular model (3T3L1
adipocytes)
Ó 2012 Elsevier Ltd. All rights reserved.
Hydroxysteroid dehydrogenases regulate steroid hormone
receptors.1 Among these, the isoform 11b-hydroxysteroid dehydro-
genase type 2 (11b-HSD2) inactivates active glucocorticoids (such
as cortisol in human and corticosterone in rodents) to cortisone
or 11-dehydrocorticosterone.2 On the other hand, the 11b-HSD1
isoform activates inert precursors (cortisone in human, 11-dehy-
drocorticosterone in rodents) to active glucocorticoids by 11-oxore-
ductase activity in liver, adipose tissue, brain, skeletal muscle,
vascular smooth muscle cells and other organs.3 The isoform 2
(11b-HSD2) expression is limited to tissues that express the miner-
alocorticoid receptor, such as kidney, gut and placenta.4
Several in vivo studies have exemplified the importance of
these converting enzymes. In particular, 11b-HSD1 knockout mice
were shown to resist hyperglycemia and to have increased hepatic
insulin sensitivity.5 Furthermore, 11b-HSD1 overexpression in
mice resulted in development of visceral adiposity, hyperglycemia
and insulin resistance.6 In parallel, 11b-HSD2 overexpression im-
proved glucose tolerance and insulin sensitivity.7
tissues, 11b-HSD1 is a potential target for treating numerous disor-
ders that may be improved by reduction of glucocorticoid action
including the metabolic syndrome. Selectivity against the 11b-
HSD2 isoform is expected to be crucial as inhibition of 11b-HSD2
is associated with serious side effects, such as hypertension.8 Many
efforts towards the discovery and development of 11b-HSD1 inhib-
itors have been carried out, and led to the discovery of several
chemical classes, that could be mainly grouped (based on the frag-
ment interacting with the catalytic residues) into amides, ureas &
carbamates, azoles & simple heteroaryls, sulfones & sulfonamides,
thiazolones, and lactams9,10 and references within. We report
herein a new class of inhibitors based on a 7-azaindole-pyrrolidine
amide scaffold.
Compound 1 was identified as an initial hit by high-throughput
screening (Fig. 1), with moderate potency on human isolated 11b-
HSD1 (IC50 of 780 nM), and a Binding Efficiency Index (BEI) of
18.3.11 The BEI index was consistently considered to identify and
focus our optimisation program on chemical functionalities carry-
ing the strongest binding contributions. For comparison purpose, a
As administration of 11b-HSD1 inhibitors is expected to de-
crease the level of cortisol and other 11b-hydroxysteroids in target
decent HTS hit (molecular weight of 350 Da, potency of 10 lM)
would exhibit a BEI of 17.1 while the BEI of a valuable lead (molec-
ular weight of 350 Da, potency of 10 nM) would reach 22.9.
Interestingly, compound 1 was based on an original indole-pyr-
rolidine scaffold, so far not reported as an 11b-HSD1 inhibitor.9
Extensive usage of public10 and in-house crystallographic informa-
tion was made in a structure-based drug design effort in order to
improve the potency of the initial hit, and analogues were synthes-
ised according to Scheme 1.
⇑
Corresponding author. Tel.: +41 22 414 95 26; fax: +41 22 414 95 58.
E-mail addresses: scfranck@hotmail.fr, serge.christmann-franck@merckgroup.
com (S. Christmann-Franck).
Currently at Novartis Pharma AG, WSJ 152-4.06.6 CH-4002 Basel, Switzerland.
Currently at Metabrain Research, 4 Avenue du Président François Mitterrand,
à
91380 Chilly-Mazarin, France.
§
Currently at Poxel Pharma, 200 avenue Jean Jaurès, 69007 Lyon, France.
Currently at AstraZeneca, Pepparedsleden 1, Mölndal SE-431 83, Sweden.
Currently at Ipsen, 5 avenue du Canada, 91966 Les Ulis Cedex, France.
Currently at Edelris, 115 avenue Lacassagne, 69003 Lyon, France.
–
Docking of the hit 1 was performed in the human 11b-HSD1–
k
adamantane sulfone co-crystal structure (PDB code 2ILT),12 using
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.07.070

5910
E. Valeur et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5909–5914
O
N
NH
MeO
1
IC50 = 780 nM
BEI = 18.3
Figure 1. Initial hit discovered by HTS.
CN
CO2Et
CHO
R1
R1
R1
(i)
(ii)
N
H
X
N
H
X
N
H
X
(iii)
O
R
N
CN
NC
R1
R1
(iv)-(vi)
N
H
X
N
H
X
X = CH, N
Scheme 1. Reagents and conditions: (i) 1,3-hexamethylenetetramine (HMTA),
AcOH, reflux, 3 h; (ii) NC-CH2-CO2CH3, piperidine, EtOH, 0 °C to rt, 5 h; (iii) KCN,
EtOH, 80 °C, 2.5 h; (iv) AcOH, H2SO4, 120 °C, 1.5 h; (v) LiAlH4, THF, 0 °C–70 °C, 18 h;
(vi) RCO2H, EDC, HOBt, Et3N, DMF, rt, 18 h.
the program GOLD 4.0.13 The docking suggested that the amide
carbonyl moiety could establish two hydrogen bonds with the hy-
droxyl groups of catalytic residues Ser170 and Tyr183 (Fig. 2) (all
11b-HSD1 co-crystallized inhibitors but the sulfonamides display
at least one of these two hydrogen bonds10).
Another hydrogen bond was observed between the nitrogen of
the indole and the backbone of Thr124. The influence of this inter-
action on the potency was assessed. First of all, according to the
postulated binding mode, the N-alkylated derivative 2 could not
establish this interaction, and was indeed confirmed experimen-
tally to be less active (Table 1). Furthermore, modulation of the
indolic hydrogen acidic character was expected to tune the
strength of the hydrogen bond and had indeed an impact on the
potency: electron donating substitutions such as 5-methoxy (com-
pound 3) were detrimental while electron withdrawing groups
such as fluorine (compound 4) increased dramatically the potency,
underlining the strong contribution of the Thr124 hydrogen bond.
A similar influence of the fluorine insertion was observed with a
tolyl group in the R2 position (compounds 7 and 8). Based on public
and proprietary structural information,11,12 we explored the intro-
duction of a conformationally restricted phenyl acetamide group
with the intention of establishing an additional interaction with
the aromatic ring of Tyr177. The resulting compound 9 and its
5-fluoroindole analog 10 showed encouraging IC50 values of
530 nM and 430 nM respectively, in agreement with our expecta-
tions and in-line with the docking-generated binding mode (com-
pound 9, Fig. 3). Compounds 5 to 10 were synthesised to explore
the influence of R2 on the activity. The introduction of a cyclohexyl
was detrimental (compound 5). Inserting a fluorine atom on the in-
dole ring (compound 6) did improve the potency as already ob-
served with the fluoro analogue 4 of compound 1, but not to the
same extent.
Figure 2. Steroid substrate binding site of human 11b-HSD1 (PDB code 2ILT) with
(a) the adamantane sulfone compound from the experimental structure and (b)
docked compound 1; for the sake of clarity, only Tyr177, Thr124 and catalytic
residues Ser170 and Tyr183, and only polar and chiral hydrogen atoms and only (S)-
enantiomer15 have been displayed; colour coding by atom is as follows: oxygen in
red, nitrogen in blue, hydrogen in white, sulfur in yellow, phosphorus in orange;
carbon atoms are coloured in orange for the experimental ligand (a), in dark green
for compound 1 (b), in light yellow for the protein and in light blue for the NADP(H)
cofactor (protonation state not determined in the X-ray structure). The population
size and number of operations per docking were set to auto with a maximum search
efficiency (200%). Water molecules 4285 and 4288 were set to toggle and spin, and
a maximum of 20 poses per ligand was set, using the Chemscore scoring function.14
could improve potency by enabling the azaindole moiety to create
a second hydrogen bond with a conserved water molecule. With
that prospect in mind, new analogues were docked and synthes-
ised (Scheme 2).
Compound 11 (Table 2), the azaindole analogue of indole 7,
reached an improved potency of 360 nM which was consistent
with the additional azaindole–water molecule hydrogen bond ob-
served in docking studies (Fig. 4).
IC50 values supported separately both structure-based modifi-
cations hypotheses (improvement of the interaction with Tyr177
In parallel, further analysis of the binding mode led to suggest
the introduction of an additional nitrogen atom on the indole. This

E. Valeur et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5909–5914
5911
Table 1
First round of optimisation around indole-pyrrolidines
O
R2
N
R1
Compounds
R1
R2
h 11b-HSD1 IC50a, nM (% inhibition @ 1
lM)
BEIb
2
nd (28)
—
N
OCH3
OCH3
MeO
F
3
4
nd (0)
100
—
N
H
19.9
—
N
H
OCH3
5
nd (0)
1300
nd (70)
800
N
H
F
F
F
6
18.7
—
N
H
7
N
H
8
18.9
19.0
18.3
N
H
9
530
N
H
10
430
N
H
a
11b-HSD1 enzyme activity was assessed in 30 mM Hepes buffer, pH 7.4, containing 1 mM EDTA, 1 mM G-6P, and a substrate mixture cortisone/NADPH (200 nM/200
lM).
Reaction was initiated by addition of 3 g of human recombinant 11b-HSD1 expressed in Escherichia Coli and terminated with addition of 18b-glycyrrhetinic acid stop
l
solution after 150 min incubation at 37 °C. Determinations of cortisol levels were monitored by HTRF assay (Cisbio International).
b
Binding Efficiency Index = pKi or pIC50/Molecular Weight (kDa), pIC50 was used in this study.
through insertion of conformationally restricted phenyl acetamide,
and addition of a nitrogen atom to the indole ring): compounds 9–
10 and 11, respectively bearing the corresponding modifications,
exhibited improved potencies over the initial hit. The combination
of these two modifications resulted in a dramatic improvement in
potency with compounds 12 and 13 displaying IC50 values of 25
and 50 nM (Table 2). Docking of 13 nicely supported all previous
hypotheses regarding interactions within the binding site (Figs. 5,
S1, S2).
Efforts were then focused on the aromatic interactions with
Tyr177. Several groups were introduced to tune the orientation
of the aryl group (Table 2) while halogens as electron withdrawing
groups were added on the phenyl acetamide (compounds 14–17).
Substitutions on the phenyl acetamide group led to single digit
nanomolar IC50 values (Table 2), presumably due to increased lipo-
philic interactions in the pocket and favorable electrostatic interac-
idea, introduction of an electron rich oxygen atom between the ori-
entating group and the phenyl (compound 18) led to a 15 fold po-
tency decrease compared to the phenyl acetamide equivalent 12.
However activity was greatly improved with the para-fluoro ana-
log compound 19 (25 nM IC50). The nature of the orientating group
seemed not to be crucial since gem-dimethyl, cyclopropane and
cyclobutane derivatives led to similar potencies (compounds 15–
17).
Furthermore, addition of a cyclopropane ring to the pyrrolidine
ring to potentially improve its metabolic stability did not alter the
activity (compounds 20 and 21).
Several further modifications were attempted to investigate
substitutions on the 7-azaindole group (Table 3). The 2-methyl
analogues 22–24 were in the same range of potency as the unsub-
stituted 7-azaindoles 14 and 16. However, a fluorine insertion on
position 5 of the azaindole ring system (compound 25) was clearly
detrimental for inhibition, with a 30 fold decrease in potency,
unlike the 5-fluoro substitution of the indole ring (Table 1,
tion between the electron-rich Tyr177
p system and electron-
depleted
p
system of the phenyl moiety.16 Following the same

5912
E. Valeur et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5909–5914
Table 2
Azaindole-pyrrolidines derivatives
O
R2
N
N
H
N
Compounds
Spacer
R2
h 11b-HSD1 IC50a, nM
BEIb
21.1
N
11
360
25
N
N
12
13
22.8
22.0
50
N
N
14
15
16
17
4
22.2
22.9
21.8
F
Figure 3. Steroid substrate binding site of human 11b-HSD1 (PDB code 2ILT) with
docked compound 9 ((S)-enantiomer represented); displayed details and colour
coding by atoms are identical to Figure 2b.
Cl
Cl
N
N
10
Boc
Boc
N
N
17
7
21.4
(i)
N
H
Cl
N
H
N
H
N
N
N
(ii), (iii), (v)
N
N
(iv), (iii), (v)
(iv), (iii), (v)
18
19
370
25
18.4
20.8
O
O
F
O
O
O
N
R
R
N
N
R
N
3
20
21
46
25
20.3
21
F
F
N
H
N
H
N
N
N
H
N
N
Scheme 2. Reagents and conditions: (i) N-Boc-pyrrolidin-3-one, KOH, MeOH, H2O,
75 °C, 18 h (ii) HCO2NH4, 10% Pd(OH)2/C, EtOH, reflux, 1 h; (iii) HCl/MeOH, rt, 12 h;
(iv) ZnEt2, CH2I2, DCM, 0 °C ->rt, 18 h; (v) RCO2H, EDC, HOBt, TEA, DMF, rt, 18 h.
a
h 11b-HSD1 experimental protocol as described in Table 1.
BEI as described in Table 1.
b
compounds 6, 8 and 10). This difference could be attributed to a
negative contribution of the 5-fluoro substituent on the pyridine-
water interaction in the case of the 7-azaindole. The importance
of this interaction was further illustrated with the 6-azaindole
derivative 26 which displayed almost one log loss of potency
(43 nM IC50).
The ternary complex structure of a racemic mixture of com-
pound 22, NADP(H) and a humanised murine 11b-HSD1 was ob-
tained by X-ray crystallography17 (the protonation state of NADP
was not determinable in the structure). The final model contains
eight molecules in the asymmetric unit that form four homodi-
mers. The overall structure of the active site mutant does not differ
significantly when compared to the murine wild type one.18 Com-
pound 22 is bound to the steroid binding site of 11b-HSD1 and is
also in contact with NADP(H). The active site amino acids and both
enantiomers of 22 are well defined in the electron density map,
sharing a similar binding mode and orientation in the active site
(Fig. S3). Figure 6 shows the binding site of the humanised murine
11b-HSD1 (6 point mutations) with the initial Fo-Fc electron den-
sity map of the (S)-enantiomer of 22 at 2.5
r superimposed. The
compound forms two hydrogen bonds to the side chain hydroxyl
groups of residues Ser170 and Tyr183. Another hydrogen bond is
established between the indolic nitrogen of the ligand and the
main chain carbonyl oxygen atom of Thr124. A further hydrophilic
contact was observed in four of the eight monomers to the main
chain amide nitrogen atom of Thr124 via a water molecule con-
served in other co-crystal structures. The azaindole methyl forms
hydrophobic contacts with Leu126 hydrophobic side chain. The
pyridine ring is involved in a
p–p aromatic stacking interaction
with the phenyl moiety of Tyr177, close to the solvent exposed
area of the protein dimer interface. Overall, the crystal structure
confirmed the binding modes observed during the dockings and
used in the compound optimisation program.
Compounds 13 and 16 were selected for further characteriza-
tion and preliminary results are presented herein (Table 4). These
compounds were selective against 11b-HSD2 (selectivity ratio
>200). We hypothesised that the increased potency observed with

E. Valeur et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5909–5914
5913
Table 3
Variation of the azaindole moiety
O
R2
N
R1
BEIb
a
Compounds R1
R2
h 11b-HSD1 IC50
,
nM
N
22
19
12
5
22.3
20.7
22.7
17.3
21.1
N
H
N
N
N
23
24
25
26
N
H
Cl
F
N
H
F
439
43
N
N
F
F
Figure 4. Steroid substrate binding site of human 11b-HSD1 (PDB code 2ILT) with
docked compound 11 ((S)-enantiomer represented); displayed details and colour
coding by atoms are identical to Figure 2b.
H
N
N
H
a
h 11b-HSD1 experimental protocol as described in Table 1.
BEI as described in Table 1.
b
Figure 5. Steroid substrate binding site of human 11b-HSD1 (PDB code 2ILT) with
docked compound 13 ((S)-enantiomer represented); displayed details and colour
coding by atoms are identical to Figure 2b.
Figure 6. Steroid substrate binding site of humanised murine 11b-HSD1 in complex
with compound 22 (PDB code 3GMD) ((S)-enantiomer); compound 22 is depicted in
heavy sticks, the protein in light sticks and water molecules in large balls; colour
coding by atoms is as follows: oxygen in red, nitrogen in blue, sulfur in yellow,
phosphorus in orange; carbon atoms are coloured in light orange for compound 22,
in light grey for the chain A of the protein, in green for chain B of the protein and in
13 and 16 on the human enzyme compared to the mouse one could
be attributed to the increased lipophilic and optimised aromatic
interactions with Tyr177 in human19, that is mutated in a non aro-
matic Gln residue in the mouse enzyme. Furthermore, compound
13 displayed good inhibition on mouse 11b-HSD1 (IC50 of
362 nM) in a cellular model (3T3L1 adipocytes), but suffered from
a poor PK profile.
light blue for the NADP(H) cofactor. Initial ligand Fo-Fc density contoured at 2.5
r is
shown in purple and initial water Fo-Fc density contoured at 3 is shown in brown.
r
The clearance (CL) was higher than the hepatic blood flow in the
mouse, suggesting very quick metabolism, supported also by the
very low recovery of unchanged drug in feces (RF results). Clearance
was improved with compound 16 but remained disappointing.
In conclusion, we have reported a new, original and promis-
ing class of 11b-HSD1 inhibitors based on a 7-azaindole-pyrroli-
dine scaffold. The most active compounds were obtained with
aromatic cylopropane carboxamides inducing a strong lipophilic
and
p–p interaction with Tyr177, and single digit nanomolar
IC50 was achieved. These compounds were selective against
11b-HSD2 and showed moderate inhibition of mouse 11b-HSD1

5914
E. Valeur et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5909–5914
Table 4
References and notes
Profile of compounds 13 and 16
1. Nobel, S.; Abrahmsen, L.; Oppermann, U. Eur. J. Biochem. 2001, 268, 4113.
Compounds
13
16
2. Tomlinson, J. W.; Walker, E. A.; Bujalska, I. J.; Draper, N.; Lavery, G. G.; Cooper,
M. S.; Hewison, M.; Stewart, P. M. Endocrinol. Rev. 2004, 25, 831.
3. (a) Brereton, P. S.; van Driel, R. R.; Suhaimi, F.; Koyama, K.; Dilley, R.;
Krozowski, Z. Endocrinology 2001, 142, 1644; (b) Bujalska, I. J.; Kumar, S.;
Stewart, P. M. Lancet 1997, 349, 1210; (c) Jellinck, P. H.; Pavlides, C.; Sakai, R. R.;
McEwen, B. S. J. Steroid. Biochem. Mol. Biol. 1999, 71, 139; (d) Moisan, M. M.;
Seckl, J. R.; Edwards, C. R. W. Endocrinology 1990, 127, 1450; (e) Ricketts, M. L.;
Verhaeg, J. M.; Bujalska, I. J.; Howie, A. J.; Rainey, W. E.; Stewart, P. M. J. Clin.
Endocrinol. Metab. 1998, 1325; (f) Whorwood, C. B.; Mason, C. B.; Ricketts, M. L.;
Howie, A. J.; Stewart, P. M. Mol. Cell. Endocrinol. 1995, 110, R7.
h 11b-HSD1 IC50a
50
10
720
1221
>10 lM
>1000
m 11 b -HSD1 IC50a
230
362
>10
>200
0.2 h
<5%
m 11 b -HSD1 3T3L1 cellular IC50b (nM)
h 11 b -HSD2 IC50c
lM
h 11 b -HSD2/ HSD1 IC50 Selectivity ratio
d
Plasma t1/2
0.3 h
<5%
Fd
CLd
20.0 L/h/kg
5.0 L/kg
61%
4.0 L/h/kg
1.5 L/kg
65%
Vssd,e
RFd,f
m SPBg
4. (a) Seckl, J. R.; Walker, B. R. Endocrinology 2001, 142, 1371; (b) Kotelevtsev, Y.
V.; Brown, R. W.; Fleming, S.; Kenyon, C.; Edwards, C. R. W.; Seckl, J. R.; Mullins,
J. J. J. Clin. Invest. 1999, 103, 683.
91%
94%
5. Morton, N. M.; Holmes, M. C.; Fiévet, C.; Staels, B.; Tailleux, A.; Mullins, J. J.;
Seckl, J. R. J. Biol. Chem. 2001, 276, 41293.
6. Masuzaki, H.; Paterson, J.; Shinyama, H.; Morton, N. M.; Mullins, J. J.; Seckl, J. R.;
Flier, J. S. Science 2001, 294, 2166.
7. Kershaw, E. E.; Morton, N. M.; Dhillon, H.; Ramage, L.; Seckl, J. R.; Flier, J. S.
Diabetes 2005, 1023, 54.
a
h 11b-HSD1 experimental protocol as described in Table 1; for murine (m) 11b-
g of microsome fraction from liver (Tebu) was used in the assay.
HSD1, 2.5
l
b
3T3L1 cells were differentiated (5 days) in adipocytes and further cultured
(4 days) in a 24 well plate format before incubation with test compounds (10-5-10-
7 M) for 150 min. Cortisol in the culture supernatant was then measured for 11b-
HSD1 activity in HTRF assay (see Table 1).
8. White, P. C.; Mune, T.; Agarwal, A. K. Endocrinol. Rev. 1997, 18, 135.
9. For reviews, see for example: (a) Fotsch, C.; Askew, B. C.; Chen, J. J. Expert Opin.
Therap. Pat. 2005, 15, 289; (b) Webster, S. P.; Pallin, T. D. Expert Opin. Therap.
Pat. 2007, 17, 1407; (c) Boyle, C. D. Curr. Opin. Drug Discov. Devel. 2008, 11, 495;
(d) Hughes, K. A.; Webster, S. P.; Walker, B. R. Expert Opin. Investig Drugs 2008,
17, 481; (e) Boyle, C. D.; Kowalski, T. J. Expert Opin. Therap. Pat. 2009, 19, 801; (f)
Ge, R.; Huang, Y.; Liang, G.; Li, X. Curr. Med. Chem. 2010, 17, 412.
10. (a) Thomas, M. P.; Potter, B. V. Future Med. Chem. 2011, 3, 367; (b) Singh, S.;
Tice, C. Curr. Pharm. Biotechnol. 2010, 11, 779; (c) Sun, D.; Wang, M.; Wang, Z.
Curr. Top. Med. Chem. 2011, 11(11), 1464.
11. Abad-Zapatero, C.; Metz, J. T. Drug Discov. Today 2005, 10, 464.
12. Sorensen, B.; Winn, M.; Rohde, J.; Shuai, Q.; Wang, J.; Fung, S.; Monzon, K.;
Chiou, W.; Stolarik, D.; Imade, H.; Pan, L.; Deng, X.; Chovan, L.; Longenecker, K.;
Judge, R.; Qin, W.; Brune, M.; Camp, H.; Frevert, E. U.; Jacobson, P.; Link, J. T.
Bioorg. Med. Chem. Lett. 2007, 17, 527.
c
11b-HSD2 enzyme activity was assessed in 50 mM Tris buffer, pH 7.8, con-
taining 1 mM MgCl2, 0.5 mM NAD, and 1
expressed in E. Coli. Reaction was initiated by addition of cortisol 20 nM and ter-
minated with addition of carbenoxolone 10 M stop solution after 60 min incu-
lg of human recombinant 11b-HSD2
l
bation at room temperature. Determinations of cortisol levels were monitored by
an HTRF assay (Cisbio International).
d
Dosed i.v. (0.2 mg/kg) and p.o. (0.5 mg/kg) in male rats in a low dose cocktail
study; mean values over 3 rats.
e
Volume of distribution at steady state.
Recovery of unchanged drug in feces (0–24 h collection).
mouse serum protein binding.
f
g
13. (a) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R. J. Mol. Biol. 1997, 267,
727; (b) Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Taylor, R. D.
Proteins 2003, 52, 609.
14. (a) Murray, C. W.; Auton, T. R.; Eldridge, M. D. J. Comput.-Aided Mol. Des. 1998,
12, 503; (b) Verdonk, M. L.; Berdini, V.; Hartshorn, M. J.; Mooij, W. T. M.;
Murray, C. W.; Taylor, R. D.; Watson, P. J. Chem. Inf. Comput. Sci. 2004, 44, 793.
15. The IC50 values were reported on racemic compounds. The docking
experiments assessed the behavior of each enantiomer. Similar binding
modes were consistently observed for both steroisomers (Figs. S1, S2).
16. Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525.
(IC50 of 1221 nM) in a cellular model (3T3L1 adipocytes). Further
optimisation of this series to improve PK profile will be reported
in due course.
Acknowledgments
17. The structure of
a humanised murine 11b-HSD1 with 6 point mutations
The authors wish to thank Proteros Biostructures GmbH for
their excellent support in solving the X-ray structure of humanised
murine 11b-HSD1 in complex with compound 22.
(M175V, Q177Y, I180V, E226A, I227V, I231V) in complex with compound 22
and NADP(H) was generated. This mutant was designed with the aim to make
the binding pocket more similar to the human isoform. The structure was
solved at a resolution of 2.28 Å, revealing the detailed binding mode of the
ligand in the active site. The atomic coordinates have been deposited in the
RCSB Protein Data Bank (www.rscb.org) under accession code 3GMD.
18. Zhang, J.; Osslund, T. D.; Plant, M. H.; Clogston, C. L.; Nybo, R. E.; Xiong, F.;
Delaney, J. M.; Jordan, S. R. Biochemistry 2005, 44, 6948.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.07.
070.
19. (a) Kim, K. W.; Wang, Z.; Busby, J.; Tsuruda, T.; Chen, M.; Hale, C.; Castro, V. M.;
Svensson, S.; Nybo, R.; Xiong, F.; Wang, M. Biochim. Biophys. Acta 2006, 1764, 824;
(b) Favia, A. D.; Masetti, M.; Recanatini, M.; Cavalli, A. PLoS One 2011, 6, e25375.