Pyrrolidine-pyrazole ureas as potent and selective inhibitors of 11β-hydroxysteroid-dehydrogenase type 1

Article abstract of DOI:10.1016/j.bmcl.2011.02.111

A High Throughput Screening campaign allowed the identification of a novel class of ureas as 11β-HSD1 inhibitors. Rational chemical optimization provided potent and selective inhibitors of both human and murine 11β-HSD1 with an appropriate ADME profile an

Full text of DOI:10.1016/j.bmcl.2011.02.111

Bioorganic & Medicinal Chemistry Letters 21 (2011) 2244–2251
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pyrrolidine-pyrazole ureas as potent and selective inhibitors
of 11b-hydroxysteroid-dehydrogenase type 1
Olivier Venier a, , Cécile Pascal a, Alain Braun a, Claudie Namane a, Patrick Mougenot a, Olivier Crespin a,
⇑
François Pacquet a, Cécile Mougenot a, Catherine Monseau a, Bénédicte Onofri a, Rommel Dadji-Faïhun a,
Céline Leger a, Majdi Ben-Hassine a, Thao Van-Pham a, Jean-Luc Ragot a, Christophe Philippo a,
Stefan Güssregen b, Christian Engel b, Géraldine Farjot a, Lionel Noah a, Karima Maniani a, Eric Nicolaï a,
⇑
a Sanofi-aventis R&D, 1 Avenue Pierre Brossolette, 91385 Chilly-Mazarin, France
b Sanofi-aventis Deutschland GmbH, Industriepark Hoechst, 65926 Frankfurt am Main, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
A High Throughput Screening campaign allowed the identification of a novel class of ureas as 11b-HSD1
inhibitors. Rational chemical optimization provided potent and selective inhibitors of both human and
murine 11b-HSD1 with an appropriate ADME profile and ex vivo activity in target tissues.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 18 January 2011
Revised 25 February 2011
Accepted 27 February 2011
Available online 23 March 2011
Keywords:
11b-Hydroxysteroid-dehydrogenase type 1
11b-HSD-1 inhibitors
Metabolic syndrome
Urea
Pyrrolidine-pyrazole
Pharmacodynamic ex vivo assay
OH
11b-Hydroxy steroid dehydrogenase type 1 (11b-HSD1) con-
verts the biologically inactive glucocorticoids, cortisone (in man)
and 11-dehydrocorticosterone (in rodents), into their biologically
active metabolites, respectively cortisol and corticosterone, which
are known to act as functional antagonists of insulin in several tar-
get organs and tissues such as the liver, muscle, and adipose tissue
(Fig. 1).1 There is evidence implicating an excess of cortisol in tis-
sues as a primary driver of insulin resistance and a critical point
for disease intervention.2 Liver- or adipose tissue-specific overex-
pression of 11b-HSD1 in transgenic mice produces a phenotype
closely resembling human type 2 diabetes mellitus.3 Reduction of
intracellular corticosterone levels in rodents as a result of pharma-
cological inhibition of 11b-HSD1 reverses manifestations of altered
metabolic parameters including ectopic fat storage, diabetes, dysl-
ipidemia and atherosclerosis.4 Both 11b-HSD1 expression and
activity are up-regulated in the adipose tissue of obese, insulin-
resistant or diabetic humans.5 Inhibition of 11b-HSD1 therefore
offers potential as a novel therapy to lower intracellular cortisol
concentrations and thereby enhance insulin sensitivity and hepatic
lipid catabolism in type 2 diabetes, obesity, and hyperlipidemia.
OH
O
O
11βHSD1
HO
O
OH
NADP
NAD
NADPH
OH
H
H
H
H
H
H
NADH
11βHSD2
O
O
Cortisone
Cortisol
Figure 1. Interconversion of cortisone and cortisol by 11b-HSD1 and 11b-HSD2
enzymes.
Two isoforms of 11b-HSD are known, the products of distinct
genes: 11b-HSD1 is primarily expressed in the liver and adipose
tissue, the other isoform, 11b-HSD2, is located mainly in the kid-
ney.6 11b-HSD2 converts active glucocorticoids (such as cortisol
and corticosterone) into inactive cortisone or 11-dehydrocorticos-
terone. Any potential drug aimed at inhibiting 11b-HSD1 should
therefore be selective for this isoform since potent inhibition of
11b-HSD2 may result in sodium retention, hypokalemia, and
hypertension.7
In the last decade, several classes of 11b-HSD1 inhibitors have
been identified by the pharmaceutical industry and academic
groups.8 These included sulfonamides such as BVT.2733 or
⇑
Corresponding authors.
E-mail addresses: Olivier.Venier@sanofi-aventis.com (O. Venier), Eric.Nicolai@
sanofi-aventis.com (E. Nicolaï).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.02.111

O. Venier et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2244–2251
2245
Table 1
Cl
NC
Variations of the tetrahydroquinoline urea: compounds 2a–m
N
H
O
N
H
N
N
N
R
N
N
S
N
NH2
S
O
O
S
O
O
O
BVT.2733
PF-915275
Compound
R
h/m 11b-HSD1 IC50 (nM)
N
O
N
2a
42/325
N
NH
H
N
N
H
N
S
N
2b
2c
>2000/ND
Merck 544
AMG 221
N
H
N
Figure 2. Representative 11b-HSD1 inhibitors.
>2000/ND
N
N
PF-915275,9,10 triazoles such as the Merck compound 544,11 and
thiazolones such as Amgen compound AMG221 (Fig. 2).12
At the time we started our research program, few drug candi-
dates had entered clinical trials: PF-915275 had been discontinued
for formulation issues13 and BVT3498 was stopped for unknown
reasons.14 Nevertheless, in 2008, proof of concept of therapeutic
efficacy of 11b-HSD1 inhibition in type 2 diabetic patients has been
obtained with INCB013739 showing improvement of blood glucose
parameters and lipid profile.15 Furthermore, several classes of 11b-
HSD1 inhibitors including INCB013739 displayed limited activity
on rodent enzymes, making it difficult to assess them in appropri-
ate animal models for optimal drug candidate selection. In meta-
bolic disease pathogenesis, 11b-HSD1 activity is up-regulated in
adipose tissue and the liver, thus inhibition in both tissues is a
likely prerequisite for producing significant benefits.16 In order to
identify new drug candidates with an optimal PK/PD profile we fo-
cused our efforts on compounds displaying inhibition of both hu-
man and rodent enzymes, thus allowing early ex vivo testing in
rodent.
2d
2e
2f
127/1000
20/113
NH
N
NH
N
436/ND
N
N
N
O
O
2g
2h
2i
>2000/ND
>2000/ND
378/ND
O
N
N
N
N
O
N
N
2j
93/ND
N
2k
>2000/ND
O
N
N
N
A HTS campaign using the Sanofi-aventis chemical library and
sorting of the resulting hits resulted in the identification of two po-
tent and selective inhibitors, compounds 1 and 2a, which displayed
mixed human/rodent 11b-HSD1 activity and good selectivity over
11b-HSD2 (Fig. 3).
2l
>2000/ND
66/73
N
2m
Unfortunately, these compounds include an imidazole moiety
which is known as a potential CYP3A4 binder17 and experimental
data confirmed that both 1 and 2a were strong CYP3A4 inhibitors
ND: not determined.
(IC50 <1 l
M). To overcome this issue,18 we prepared a series of ana-
The oxadiazole derivatives 2g–k showed variable inhibition po-
tency depending of the structure and substitution of the oxadiazole
with the 1,2,4-oxadiazole derivative 2j being the most potent. The
3-pyridyl derivative 2m was in the same range of activity as 2a
with an IC50 of 66 nM. Replacement by an oxazole (2k) and a tria-
zole (2l) dramatically decreased the activity. Finally, the 4-substi-
logs where the imidazole ring was replaced by alternative surro-
gate heterocyclic moieties (Table 1). The synthesis of compounds
2a–m is depicted in Schemes 1–3.
Compound activity was measured in vitro using recombinant
human (h) and mouse (m) 11b-HSD1 enzymes. The conversion of
radiolabeled cortisone into cortisol was detected using an anti-cor-
tisol antibody in a miniaturized scintillation proximity assay
(SPA).31 As shown in Table 1, changing the position of one nitrogen
in the imidazole ring of 2a resulted in a dramatic loss of activity
(2b), indicating a key interaction between this nitrogen and the
protein. No improvement was observed with the imidazoline 2c.
tuted pyrazole derivative 2e displayed
a twofold potency
enhancement compared to 2a. Introduction of a methyl group at
the nitrogen atom of the pyrazole (2f) led to a 20-fold loss in po-
tency. This result highlighted in part the favorable effect of a
hydrogen bond donor at that position (H-bond interaction between
the NADPH cofactor and the heterocyclic NH was confirmed by the
X-ray co-crystal structure, see Fig. 4a). In addition, this effect can
be in part associated with unfavorable steric interaction. As ex-
pected, non-imidazole compounds 2e–f displayed reduced CYP3A4
O
N
N
O
N
N
inhibitory activity with IC50’s above 10 lM. This first round of opti-
N
N
NH
NH
mization allowed the identification of 2e as a new lead with potent
activity in human and rodent enzymes and no CYP3A4 inhibition
issue and prompted us to consider the pyrazole moiety as the best
imidazole surrogate even though 2e displayed limited metabolic
stability (%lability: 57/79 in human/mouse liver microsomes).32
In order to check if piperidine was the optimal heterocyclic
spacer, we kept the pyrazole moieties of 2d–e and assessed differ-
ent piperidine replacements. We first evaluated the tropane
O
2a
1
h11βHSD1 IC50= 42nM
m11βHSD1 IC50= 325nM
h11βHSD2 IC50>10.000nM
h11βHSD1 IC50= 66nM
m11βHSD1 IC50= 72nM
h11βHSD2 IC50>10.000nM
Figure 3. Structure and inhibitory activities of lead compounds 1 and 2a.

2246
O. Venier et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2244–2251
Scheme 1. Synthesis of inhibitors 2a,d,e,k and 2f. R is as defined in Table 1. Reagents and conditions: (a) 0 °C, triphosgene,19 Et3N, 1,2,3,4-tetrahydroquinoline, CH2Cl2 then
compound 3 (2d 87%, 2k 41%), or p-nitrophenyl(1,2,3,4-tetrahydroquinoline)carbamate,20 Et3N, CH2Cl2 (2e, 28%) or methylimidazolium (1,2,3,4-tetrahydroquinoline)car-
bamate iodide,21 Et3N, CH2Cl2 (2a 16%); (b) 2e, BEMP supp., MeI, CH3CN, 55%.22
Scheme 2. Synthesis of inhibitors 2b,c,h,i. Reagents and conditions: (a) triphosgene, Et3N, 2-tetrahydroquinoline, CH2Cl2/CH3CN then compound 4, 100%; (b) TFAA,23 Et3N,
CH2Cl2, 0 °C, 82%; (c) ethylenediamine, S8, reflux, 46%;24 (d) MnO2, dioxane, 41%;25 (e) ethanolamine HCl, Et3N, MeOH, reflux, 84%; (f) acetyl chloride, DMF, 43%;26 (g) triethyl
orthoformate, reflux, 43%.27
O
O
N
O
O
N
O
N
a
N
b
HN
OH
OEt
OEt
10
8
9
c
O
N
O
N
d
N
N
2j
H
N
H
N
O
N
O
N
N
N
O
N
O
N
N
f
e
N
N
N
O
N
O
2l
2g
11
Scheme 3. Synthesis of inhibitors 2g,j,l. Reagents and conditions: (a) diphosgene, Et3N, 2-tetrahydroquinoline, CH2Cl2 then compound 8, 100%; (b) NaOH, EtOH, 50 °C, 93%;
(c) N-hydroxy-acetamidine, TBTU, HOBt, DIPEA, DMF, 95 °C, 46%;28 (d) acetic acid hydrazide, EDC, HOBt, Et3N, CH2Cl2/CH3CN, 33%; (e) supp. BEMP, TsCl, THF, reflux, 51%;29 (f)
methylamine HCl, methylamine 2 M in MeOH, 130 °C, microwave, 31%.30
analogs 12a and 12b, the synthesis of which is illustrated in
Scheme 4. This chemical scheme is an alternative strategy, where
the pyrazole group is introduced after urea formation. The prepa-
ration of the pyrrolidine analog 12c is illustrated in Scheme 5. In
this synthetic scheme, the carbon–carbon bond formation between
the pyrrolidine and the pyrazole is performed before the urea for-
mation step.
The effect of these piperidine modulations on human/mouse
enzyme inhibition and metabolic stability are summarized in Table
2. The endo tropane derivative 12a compared favorably with its
piperidine analog 2d. On the other hand, the exo conformer 12b
is fivefold less potent on human enzyme. The pyrrolidine deriva-
tive 12c displayed the most potent activity in both human
(IC50 = 4 nM) and mouse enzyme (IC50 = 25 nM). However, it exhib-
ited insufficient metabolic stability for in vivo testing.
As the benzylic position of the tetrahydroquinoline was identi-
fied as a spot for oxidative metabolism we tried to modulate this
position by introduction of steric hindrance or heteroatoms in or-
der to increase the metabolic stability.
Compounds 23a and 23c were obtained from commercially
available building blocks using the synthetic route presented in
Scheme 1. The synthesis of the spirocyclopropyl derivative 23b is
depicted in Scheme 6. The synthetic route for preparing tetrahy-
droquinoxaline derivatives 23d–e is shown in Scheme 7.
The effect of tetrahydroquinoline modifications on human/
mouse inhibition potency and metabolic stability are presented
in Table 3. The dimethyl derivative 23a displayed a very potent
enzymatic inhibition of human and mouse 11b-HSD1 but still a
very low metabolic stability, showing that hindering the metabol-
ically labile benzylic position had no effect and that metabolic

O. Venier et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2244–2251
2247
O
N
a
b
Boc
N
Boc
N
Boc
N
N
OTf
a
c
O
Boc
N
OTf
O
R
b
N
OTf
NH
NH
N
20
18
19
14 R=Boc
13
16
c
R=H
15
d
O
N
e
N
R
N
O
O
N
N
NH
N
NH
NH
N
N
e
N
N
N
21
R=Boc
12c
d
22R=H
17
12a Endo +12b Exo
Scheme 5. Synthesis of inhibitor 12c. Reagents and conditions: (a) LiHMDS, N-
phenylbis(trifluoromethanesulfonimide), THF, À78 °C, 28%; (b); Pd(PPh3)4, N-Boc-
pyrazole-4 boronic acid pinacol ester, K2CO3, DME, 90 °C, 85%; (c) H2, Pd/C, EtOH,
81%; (d) HCl 4 M dioxane, 100%; (e) triphosgene, Et3N, 1,2,3,4-tetrahydroquinoline,
CH2Cl2 then compound 22, 58%.
Scheme 4. Synthesis of inhibitors 12a and b. Reagents and conditions: (a) LDA, THF,
N-phenylbis(trifluoromethanesulfonimide), À78 °C, 77%; (b) HCl 4 M dioxane,
100%;33 (c) triphosgene, Et3N, 1,2,3,4-tetrahydroquinoline, CH2Cl2 then compound
15, 88%; (d) Pd(PPh3)4, 1H-pyrazole-3-boronic acid, NaHCO3, DME, 90 °C, 58%; (e)
H2, Pd/C, EtOH, 18%.
On the other hand, replacing the benzylic methylene by an oxy-
gen atom in 23c improved metabolic stability while keeping good
inhibitory activity of the human enzyme but not of that of the ro-
dent. Replacement of the oxygen by a nitrogen as in 23d further
enhanced the metabolic stability over 12c (13% vs 48% lability in
lability was more likely due to high lipophilicity. This was con-
firmed by the cyclopropyl derivative 23b which displayed the same
features as 23a.
Table 2
Modulation of the piperidine moiety: compounds 12a–c
O
N
het.
N
NH
N
Compound
Pyrazole
Het
11b-HSD1 IC50 (nM) h/m
% Metabolic lability h/m
ND/ND
N
12a
40/51
200/ND
4/25
NH
N
N
N
N
12b
ND/ND
48/74
NH
NH
N
12c
ND: not determined.
Boc
N
Boc
N
R
N
O
N
N
a
d
b
NH
N
O
24
25
26 R=Boc
23b
c
27
R=H
Scheme 6. Synthesis of inhibitor 23b. Reagents and conditions: (a) PPh3MeBr, t-BuOK, THF, reflux, 79%; (b) Et2Zn, ClCH2I, Et2O;34 (c) HCl 4 M dioxane, 21% two steps; (d)
triphosgene, Et3N, CH2Cl2 then compound 22, 21%.
O
NO2
O
NO2
O
NO2
O
O
O
d
H
N
N
N
c
a
b
N
N
N
N
N
H
Boc
Boc
32
31
28
29
d
O
N
O
N
O
N
N
N
N
b
NH
NH
NH
N
N
N
N
N
N
Boc
H
23e
30
23d
Scheme 7. Synthesis of inhibitors 23d–e. Reagents and conditions: (a) 4-nitrophenyl chloroformate, TEA, THF, 100%; (b) 4 M HCl in dioxane, 100%; (c) NaBH(OAC)3,
formaldehyde, CH2Cl2, 46%; (d) compound 22, THF, 100 °C, microwave, 45–80%.

2248
O. Venier et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2244–2251
Table 3
Modulation of the tetrahydroquinoline ring: compounds 23a–e
H
N
O
N
N
N
X
Compound
X
11b-HSD1 IC50 (nM) h/m
% Metabolic lability h/m
48/74
CaCO2 Permeability (10À7 cm sÀ1
277
)
12c
–CH2–
4/25
8/7
23a
94/ND
362
23b
18/21
78/79
ND
23c
23d
23e
–O–
–NH–
–NMe–
27/490
7/55
9/26
4/44
13/30
70/ND
281
139
ND
ND: not determined.
Figure 4a. Crystal structure of human 11b-HSD1 with compound 2a. The key hydrogen bonding interactions of 2a to the enzyme are shown as magenta dashed lines with
distances given in angstroms. The color scheme is red for oxygen atoms, blue for nitrogen atoms, maroon for phosphorus atoms and yellow for sulfur atoms. Carbon atoms are
dark gray for the ligand and purple for protein and NADP cofactor. Water molecules are omitted for clarity. Resolution is 2.72 Å.
Figure 4b. X-ray crystal structure of compound 2a in human 11b-HSD1. The shape of the binding site is shown as a Connolly surface with the electrostatic potential as
calculated by the MAESTRO software36 being mapped onto the surface. Areas of positive potential are shown in blue, areas of negative potential are shown in red.
human liver microsomes, 30% vs 74% lability in mouse liver micro-
somes at 20 min), maintained an excellent potency on both human
and mouse enzymes and gave a good estimated intestinal perme-
ability. The N-methyl derivative 23e showed a drop in metabolic
stability while keeping a good inhibitory activity.
Compound 23d was considered a good lead to be assessed in a
pharmacodynamic assay in mice in order to determine its ability to
inhibit 11b-HSD1 in target tissues.35 Despite a favorable in vitro
ADME profile, 23d did not show any inhibition of 11b-HSD1 in
the liver and adipose tissue at 2 h post administration to mice at

O. Venier et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2244–2251
2249
H
O
N
N
a
c
N
NC
NC
N
R
N
N
N
N
NH
Boc
34
35R=H
R=Boc
33a
b
28
d
d
O
N
f
O
NH
N
N
HOOC
N
HOOC
NH
N
N
N
H
NH
36
37
38
c
e
O
N
O
O
N
O
N
N
N
N
N
N
H
N
H2N
NH
NH
33c
33b
Scheme 8. Synthesis of inhibitors 33a,b,c. Reagents and conditions: (a) 4-fluorobenzonitrile, t-BuOK, DMSO, 59%; (b) 4 M HCl in dioxane, 100%; (c) triphosgene, Et3N, CH2Cl2
then compound 22, 22–48%; (d) 10 N aq HCl, 100 °C, 15–97%; (e) aq NH3, EDC, HOBt, dioxane, 22%; (f) cyclopropylamine, Et3N, EDC, HOBt, dioxane, 78%.
H
N
a
b
N
Boc
N
N
H
N
N
MeOOC
MeOOC
Boc
28
39
40
c
O
O
N
e
N
N
N
N
N
N
N
H
N
NH
ROOC
NH
O
41 R=Me
42 R=H
d
33d
Scheme 9. Synthesis of inhibitor 33d. Reagents and conditions: (a) NaBH(OAc)3, methyl 4-formylbenzoate, CH2Cl2, 63%; (b) 4 M HCl in dioxane, 93%; (c) triphosgene, Et3N,
CH2Cl2 then compound 22, 66%; (d) LiOH, H2O, THF, MeOH, 74%; (e) cyclopropylamine, DIPEA, EDC, HOBt, CH2Cl2, 22%.
H
N
N
N
N
N
a
b
MeO
MeO
N
Boc
N H
N
N
N
Boc
28
43
44
c
O
N
N
MeO
N
N
N
N
NH
33e
Scheme 10. Synthesis of inhibitor 33e. Reagents and conditions: (a) LiHMDS, THF, 2-chloro-5-(4-methoxy-phenyl)-pyrimidine, 56%; (b) 4 M HCl in dioxane, 49%; (c)
triphosgene, Et3N, CH2Cl2 then compound 22, 75%.
30 mg/kg per os. As reasons for this lack of activity could be high
clearance and/or inappropriate tissue distribution, modulation of
the physico-chemical properties of the tetrahydroquinoxaline ser-
ies was carried out in order to identify compounds with increased
exposure in fat and the liver. For this, we considered possibilities
for extending the scaffold.
Ser170 responsible for substrate ketone reduction (Fig. 4a). The tet-
rahydroquinoline moiety of 2a is oriented toward a solvent access
channel which makes an extension possible from its benzylic posi-
tion (Fig. 4b). This prompted us to use the nitrogen atom of 23e as
an attachment point to extend molecules into the solvent access
channel.
In order to identify appropriate intervention points from which
we could extend 23d, we used a co-crystal structure of compound
2a bound to human 11b-HSD1 (pdb code 3QQP). This structure
indicates that the inhibitor 2a binds to the steroid binding site with
its central urea interacting with the key residues Tyr183 and
The synthesis of extended compounds 33a–g is depicted in
Schemes 8–10.
The effects of tetrahydroquinoxaline extension are summarized
in Table 4. Gratifyingly, our hypothesis was confirmed experimen-
tally with the first extended compound, 33a, which retained good

2250
O. Venier et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2244–2251
Table 4
Extension of the tetrahydroquinoxaline urea: compounds 33a–e
H
N
O
N
N
N
N
R
Compound
X
11b-HSD1 IC50 (nM) h/m
% Metabolic lability h/m
CaCO2 Permeability (10À7 cm sÀ1
)
CN
O
33a
37/75
59/54
190
33b
33c
100/55
67/90
ND/ND
ND
14
NH2
O
25/36
N
H
O
33d
73/108
43/49
79/60
40/10
ND
N
H
N
N
O
33e
194
ND: not determined.
Table 5
We then took advantage of structural biology data to design and
assay new extended compounds which addressed the solvent ac-
cess channel of the protein. These extended compounds not only
retained potent inhibitory activity on both human and rodent en-
zymes but displayed improved metabolic stability and ex vivo
activity in target tissues as illustrated by the lead compound 33e.
Further optimization of 33e has been performed (results to be pub-
lished) leading to identification of new drug candidates with an
optimal PK/PD profile.38
Ex vivo inhibition of 11b-HSD1 activity in target tissues after oral administration of
compound 33e at 30 mg/kg to mice
Liver
90%
Fat
% Inhibition 2 h post dosing
70%
mixed human/mouse inhibitory activity.37 Unfortunately, the met-
abolic lability of 33a was too high for ex vivo evaluation. Increasing
the polarity of the molecule by changing the cyano group of 33a
into a primary amide led to 33b which displayed reduced potency
against the human enzyme. Substitution of the amide with a cyclo-
propyl provided 33c which restored activity on the human enzyme
and improved metabolic stability. Unfortunately introduction of
such a polar amide moiety led to a drop in intestinal permeability
(amide 33b,c vs cyano 33a). The benzylic analog 33d was equipo-
tent to 33c but displayed poor metabolic stability. Further exten-
sion by introduction of a bi-aryl moiety gave rise to 33e which
turned out to be a potent inhibitor (IC50 <50 nM on human and
mouse enzymes) with a high permeability value, and low meta-
bolic lability in mouse and human microsomes. Furthermore, addi-
tional experiments showed that 33e was selective over 11b-HSD2
References and notes
1. Tomlinson, J. W.; Walker, E. A.; Bujalska, I. J.; Draper, N.; Lavery, G. G.; Cooper,
M. S.; Hewison, M.; Stewart, P. M. Endocr. Rev. 2004, 25, 831.
2. (a) Thieringer, R.; Hermanowski-Vosatka, A. Expert Rev. Cardiovasc. Ther. 2005,
3, 911; (b) Wamil, M.; Seckl, J. R. Drug Discovery Today 2007, 12, 504.
3. (a) Kotelevtsev, Y.; Holmes, M.; Burchell, A.; Houston, P. M.; Schmoll, D.;
Jamieson, P.; Best, R.; Brown, R.; Edwards, C. R. W.; Seckl, J. R.; Mullins, J. J. Proc.
Natl. Acad. Sci. U.S.A. 1997, 94, 14924; (b) Masuzaki, H.; Paterson, J.; Shinyama,
H.; Morton, N. M.; Mullins, J. J.; Seckl, J. R.; Flier, J. S. Science 2001, 294, 2166.
4. (a) Barf, T.; Vallgårda, J.; Emond, R.; Häggström, C.; Kurz, G.; Nygren, A.;
Larwood, V.; Mosialou, E.; Axelsson, K.; Olsson, R.; Engblom, L.; Edling, N.;
Rönquist-Nii, R.; Öhman, B.; Alberts, P.; Abrahmse, L. J. Med. Chem. 2002, 45,
3813; (b) Alberts, P.; Engblom, L.; Edling, N.; Forsgren, M.; Klingstrom, G.;
Larsson, C.; Ronquist-Nii, Y.; Ohman, B.; Abrahmsen, L. Diabetologia 2002, 45,
1528; (c) Yeh, V. S. C.; Kurukulasuriya, R.; Fung, S.; Monzon, K.; Chiou, W.;
Wang, J.; Stolarik, D.; Imade, H.; Shapiro, R.; Knourek-Segel, V.; Bush, E.;
Wilcox, D.; Nguyen, P. T.; Brune, M.; Jacobson, P.; Link, J. T. Bioorg. Med. Chem.
Lett. 2006, 16, 5555; (d) Wan, Z. K.; Chenail, E.; Xiang, J.; Li, H. Q.; Ipek, M.; Bard,
J.; Svenson, K.; Mansour, T. S.; Xu, X.; Tian, X.; Suri, V.; Hahm, S.; Xing, Y.;
Johnson, C. E.; Li, X.; Qadri, A.; Panza, D.; Perreault, M. J.; Tobin, F.; Saiah, E. J.
Med. Chem. 2009, 52, 5449.
5. (a) Paulmyer-Lacroix, O.; Boullu, S.; Oliver, C.; Alessi, M. C.; Grino, M. J. Clin.
Endocrinol. Metab. 2002, 87, 2701; (b) Kannisto, K.; Pietiläinen, K. H.; Ehrenborg,
E.; Rissanen, A.; Kaprio, J.; Hamsten, A.; Yki-Järvinen, H. J. Clin. Endocrinol.
Metab. 2004, 89, 4414; (c) Rask, E.; Olsson, T.; Söderberg, S.; Andrew, R.;
Livingstone, D. E.; Johnson, O.; Walker, B. R. J. Clin. Endocrinol. Metab. 2001, 86,
1418.
6. Van Uum, S. H.; Lenders, J. W.; Hermus, A. R. Semin. Vasc. Med. 2004, 4, 121.
7. 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.
(IC50 >10 lM).
Thus the extended lead 33e displayed appropriate in vitro activ-
ity on both human and rodent enzymes, and favorable ADME and
selectivity profiles for further characterization in animals.
It was therefore used as a tool to validate the ‘extended com-
pound’ hypothesis. We determined its ex vivo activity in mouse
target tissues (Table 5). Interestingly, 33e produced a significant
inhibition of 11b-HSD1 in both the liver and fat tissue, 2 h after oral
administration at 30 mg/kg. Compound 33e thus constituted a sat-
isfactory starting point to elaborate further optimization aimed at
increasing the duration of action in both target tissues to give an
optimal drug candidate.
8. Hale, C.; Wang, M. Mini-Rev. Med. Chem. 2008, 8, 702.
9. Barf, T.; Vallgårda, J.; Emond, R.; Häggström, C.; Kurz, G.; Nygren, A.; Larwood,
V.; Mosialou, E.; Axelsson, K.; Olsson, R.; Engblom, L.; Edling, N.; Rönquist-Nii,
Y.; Öhman, B.; Alberts, P.; Abrahmsén, L. J. Med. Chem. 2002, 45, 3813.
10. Bhat, B. G.; Hosea, N.; Fanjul, A.; Herrera, J.; Chapman, J.; Thalacker, F.; Stewart,
P. M.; Rejto, P. J. Pharmacol. Exp. Ther. 2007, 324, 299.
In conclusion, starting from compact imidazolylpiperidine ureas
with strong CYP3A4 inhibition liability, we moved to pyrazolylpyr-
olidine urea analogs devoid of CYP3A4 issues but with poor ex vivo
activity in mouse.

O. Venier et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2244–2251
2251
11. Gu, X.; Dragovic, J.; Koo, G. C.; Koprak, S. L.; LeGrand, C.; Mundt, S. S.; Shah, K.;
Springer, M. S.; Tan, E. Y.; Thieringer, R.; Hermanowski-Vosatka, A.; Zokian, H.
J.; Balkovec, J. M.; Waddell, S. T. Bioorg. Med. Chem. Lett. 2005, 15, 5266.
12. Caille, S.; Cui, S.; Hwang, T. L.; Wang, X.; Faul, M. M. J. Org. Chem. 2009, 74,
3833.
30. Olson, S.; Aster, S. D.; Brown, K.; Carbin, L.; Graham, D. W.; Hermanowski-
Vosatka, A.; LeGrand, C. B.; Mundt, S. S.; Robbins, M. A.; Schaeffer, J. M.;
Slossberg, L. H.; Szymonifka, M. J.; Thieringer, R.; Wright, S. D.; Balkovec, J. M.
Bioorg. Med. Chem. Lett. 2005, 15, 4359.
31. The standard error for compd 33e tested in nine separate assays is
representative of the error in this enzymatic assay: geometric average of
IC50 = 43.4 nM with a SEM (standard error to the mean) of 6.1 nM.
32. Compound is incubated 20 min with human or mouse hepatic microsomal
fraction. Liability% result corresponds to disappearance of tested compound at
the end of the experiment.
33. Comins, D. L.; Dehghani, A. Tetrahedron Lett. 1991, 32, 5697.
34. Charette, A. B.; Beauchemin, A. In Organic Reactions; Overman, L. E., Ed.; Wiley:
New York, 2001; Vol. 58, pp 1–65.
35. Male C57Bl6/J from Charles River Laboratory, fed ad libitum under standard
regimen with free access to water, were used in the study (n = 5 per group, 25–
30 g at the time of experiment). The compound was administered orally at the
13. www.clinicaltrials.gov; August 4th, 2010.
14. Wess, L. Biovitrum replenishes. BioCentury 2005, March 28, A13.
15. Gathercole, L. L.; Stewart, P. M. J. Steroid Biochem. Mol. Biol. 2010, 122, 21.
16. Webster, S. P.; Ward, P.; Binnie, M.; Craigie, E.; McConnell, K. M. M.; Sooy, K.;
Vinter, A.; Seckl, J. R.; Walker, B. R. Bioorg. Med. Chem. Lett. 2007, 17, 2838.
17. Tang, C.; Chiba, M.; Nishime, J.; Hochman, J. H.; Chen, I.-W.; Williams, T. M.;
Lin, J. H. Drug Metab. Dispos. 2000, 28, 680.
18. Hooper, W. D. In Handbook of Drug Metabolism; Woolf, T. F., Ed.; Marcel Dekker:
New York, 1999; pp 229–238.
19. Waal, N. D.; Yang, W.; Oslob, J. D.; Arkin, M. R.; Hyde, J.; Lu, W.; McDowell, R. S.;
Yu, C. H.; Raimundo, B. C. Bioorg. Med. Chem. Lett. 2005, 15, 983.
20. Mehrotra, M. M.; Heath, J. A.; Rose, J. W.; Smyth, M. S.; Seroogy, J.; Volkots, D.
L.; Ruhter, G.; Schotten, T.; Alaimo, L.; Park, G.; Pandey, A.; Scarborough, R. M.
Bioorg. Med. Chem. Lett. 2003, 12, 1103.
21. Batey, R. A.; Santhakumar, V.; Yoshina-Ishii, C. S.; Taylor, D. Tetrahedron Lett.
1998, 39, 6267.
22. Xu, W.; Mohan, R.; Morrissey, M. M. Bioorg. Med. Chem. Lett. 1998, 8, 1089.
23. Iyobe, A.; Uchida, M.; Kamata, K.; Hotei, Y.; Kusama, H.; Harada, H. Chem.
Pharm. Bull. 2001, 49, 822.
dose of 30 mg/kg, as
a suspension of Tween 80 in methylcellulose 0.6%
(0.5:99.5; vol/vol), and under a volume of 10 mL/kg. Treated animals were
sacrificed by cervical dislocation, and 50 mg of liver and 200 mg of
subcutaneous fat tissue were removed for subsequent determination of
11beta-HSD1 activity. Tissue samples were immediately weighed, placed in
ice-cold phosphate buffered saline (PBS) containing protease inhibitors, and
then cut in small pieces before being incubated with radiolabeled 3H-
cortisone. The reaction was stopped by adding 500 lL of HCl 1 N/methanol
24. Mirkhani, V.; Moghadam, M.; Tangestaninejada, S.; Kargara, H. Tetrahedron Lett.
2006, 47, 2129.
(1:1, vol/vol) solution, and the supernatants were frozen on dry ice. After
defrosting and a clarification step (centrifugation and filtration), the samples
25. DeWald, H. A.; Beeson, N. W.; Hershenson, F. M.; Wise, L. D.; Downs, D. A.;
Heffner, T. G.; Coughenour, L. L.; Pugsley, T. A. J. Med. Chem. 1988, 31, 454.
26. Cheng-Ming, C.; Ming-Shiu, H.; Min-Tsang, H.; Chun-Wei, K.; Suja, T. D.; Jen-
Shin, S.; Hua-Hao, C.; Yu-Sheng, C.; Kak-Shan, S. Org. Biomol. Chem. 2008, 6,
3399.
were injected on an HPLC (WATERS 2695 with a SunFire R column C18 3.5
lm
4.6 Â 75 mm), coupled to
a
flow scintillation analyzer (Perkin–Elmer
radiomatic 625T) to quantify the relative amount of 3H-cortisone (11beta-
HSD1 substrate) and 3H-cortisol (11beta-HSD1 product).
36. Maestro 9.1, Schrödinger, LLC., New York, USA.
27. Palin, R.; Clark, J. K.; Evans, L.; Houghton, A. K.; Jones, P. S.; Prosser, A.; Wishart,
G.; Yoshiizumi, K. Bioorg. Med. Chem. 2008, 16, 2829.
28. Poulain, R. F.; Tartar, A. L.; Déprez, B. P. Tetrahedron Lett. 2001, 42, 1495.
29. Brain, C. T.; Brunton, S. A. Synlett 2001, 382.
37. Hillgren, K. M.; Kato, A.; Brochardt, R. T. Med. Res. Rev. 1995, 15, 83.
38. Some of the molecules described in this article are included in WO2008000950
patent application.