Design and Synthesis of (R)-1-Arylsulfonylpiperidine-2-carboxamides as 11β-Hydroxysteroid Dehydrogenase Type1 Inhibitors

Article abstract of DOI:10.1002/cmdc.201300022

R adamantly beats S: 11β-HSD1 is a target for treating metabolic syndrome. The Risomer 5 was selected as a starting point for optimization and SAR studies. Inhibitor 8w emerged after several rounds of optimization, showing cross-species inhibition of human and mouse 11β-HSD1. It also displays a good DMPK profile invitro, and was advanced to PK/PD evaluations invivo. The results confirmed its dose-dependent activity in mice.

Full text of DOI:10.1002/cmdc.201300022

CHEMMEDCHEM
COMMUNICATIONS
DOI: 10.1002/cmdc.201300022
Design and Synthesis of (R)-1-Arylsulfonylpiperidine-2-
carboxamides as 11b-Hydroxysteroid Dehydrogenase
Type 1 Inhibitors
Guangxin Xia,[a, b] Lin Liu,[a, b] Haiyan Liu,[a] Jianxin Yu,[a] Zhenmin Xu,[a] Qian Chen,[a]
Chen Ma,[a] Ping Li,[a] Bing Xiong,*[b] Xuejun Liu,[a] and Jingkang Shen*[a, b]
Metabolic syndrome includes a series of associated metabolic
abnormalities such as insulin resistance, type 2 diabetes (T2D),
dyslipidemia, hypertension, and visceral obesity.[1] Studies have
shown that cortisol levels in liver and adipose tissues of pa-
tients with metabolic syndrome tend to be higher than in
healthy individuals.[2,3] There is also strong evidence that the
glucocorticoid action relates to metabolic diseases. Glucocorti-
coid receptor (GR) signaling depends not only on circulating
cortisol levels but also on the intracellular production of corti-
sol by reduction of cortisone, the inactive glucocorticoid. 11b-
Hydroxysteroid dehydrogenase 1 (11b-HSD1), highly expressed
in liver and adipose tissues, is the predominant enzyme re-
sponsible for converting cortisone to cortisol. Inhibition of 11b-
HSD1 has the potential to control cortisol concentrations in
the liver and adipose without affecting its systemic circulating
concentration.[4] In clinical trials, 11b-HSD1 inhibitors such as
Figure 1. Sulfonamide 11b-HSD1 inhibitors 1 and 2, and discovery of 1-aryl-
sulfonylpiperidine-2-carboxamides 3–5 as novel 11b-HSD1 inhibitors.
AZD8329 (1), PF-915275 (2), and INCB-13739 (structure un-
known) significantly improve insulin sensitivity in T2D patients,
and lower triglyceride and cholesterol levels in patients with
hyperlipidemia and hypertriglyceridemia.[5–7] Accumulation of
preclinical and clinical results has made 11b-HSD1 a promising
target for an inhibitor to combat metabolic disease.[8]
sub-pocket and cannot make good van der Waals interactions
with 11b-HSD1 (Figure 2). These contrasting interaction pat-
terns may be the cause of the significant difference in their
binding affinities. Using compound 5 as a starting point, we
discovered and elaborated a series of (R)-1-arylsulfonylpiperi-
Several classes of 11b-HSD1 inhibitors such as amides, tria-
zoles, thiazolones, and sulfonamides have been reported over
the past decade.[9] We recently performed scaffold hopping
studies and discovered a series of 1-arylsulfonylpiperidine-3-
carboxamides 3 (Figure 1).[10] Interestingly, as part of our ongo-
ing efforts in this program, piperidine-2-carboxamides also
showed inhibitory activity against 11b-HSD1, and the R enan-
tiomer 5 was found to be more active than the S enantiomer
4. Through docking simulations, we found that the piperidine
ring of compound 5 is situated in a sub-pocket formed by resi-
dues Tyr361, Leu364, Ile418, and Tyr421, whereas the piperi-
dine group of compound 4 is not located in this hydrophobic
[a] Dr. G. Xia,+ L. Liu,+ Dr. H. Liu, Dr. J. Yu, Z. Xu, Q. Chen, C. Ma, Dr. P. Li,
Dr. X. Liu, Prof. J. Shen
Central Research Institute, Shanghai Pharmaceutical Holding Co. Ltd.
898 Ha Lei Rd., Zhangjiang Hi-Tech Park, Shanghai 201203 (China)
E-mail: jkshen@mail.shcnc.ac.cn
[b] Dr. G. Xia,+ L. Liu,+ Prof. B. Xiong, Prof. J. Shen
State Key Laboratory of Drug Research
Shanghai Institute of Materia Medica, Chinese Academy of Sciences
555 Zuchongzhi Rd., Zhangjiang Hi-Tech Park, Shanghai 201203 (China)
E-mail: bxiong@mail.shcnc.ac.cn
Figure 2. Docking study of compounds 4 and 5 carried out with the Auto-
Dock Vina program[13] using default parameters. The structure of guinea pig
11b-HSD1 co-crystallized with PF-877423 (PDB ID: 3LZ6)[14] was used as the
template for preparing the binding site. The NADPH cofactor and com-
pounds 4 and 5 are shown in stick models, colored grey, yellow, and green,
respectively.
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201300022.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 577 – 581 577

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
dine-2-carboxamides as potent and selective 11b-HSD1 inhibi-
tors. Herein we report the synthesis and optimization of this
series.
Table 1. In vitro activity of derivatives with varied cycloalkylamino
groups.
(R)-1-Arylsulfonylpiperidine-2-carboxamide analogues were
prepared by two general methods, A and B. As shown in
Scheme 1 (method A), (R)-1-(tert-butoxycarbonyl)piperidine-2-
Compd
A
h11b-HSD1
IC50 [nm][a]
m11b-HSD1
Inhib. [%][b]
5
16
21
8a
8b
8c
3.7
3.1
37[b]
32
Scheme 1. Synthesis of 1-arylsulfonyl-piperidine-2-carboxamides (method A).
Reagents and conditions: a) amine, EDCI, HOBt, CH2Cl2, RT, 6 h; b) HCl(g)/1,4-
dioxane, CH2Cl2, RT, overnight; c) ArSO2Cl, Et3N, CH2Cl2, RT, 2–24 h. A=cyclo-
heptylamino, Ar: see Table 2.
42[c]
12
carboxylic acid (6) was condensed with various amines to pro-
vide corresponding amides 7. After removing the protecting
group under acidic conditions, the resulting product was then
allowed to react with arylsulfonyl chlorides to give the target
compounds 8g–p. The other synthetic route, method B, is de-
picted in Scheme 2: sulfonylation of R-configured ester 9 with
8d
8e
8 f
23[b]
15[b]
33
14
11
19
[a] Values are the average of three replicates, with a variance of <15%.
[b] Percent inhibition measured at 100 nm; values are the mean of two
measurements. [c] IC50 =173 nm.
Scheme 2. Synthesis of 1-arylsulfonyl-piperidine-2-carboxamides (method B).
Reagents and conditions: a) ArSO2Cl, Et3N, CH2Cl2, RT, 8 h; b) MeOH, 50%
NaOH(aq), reflux, 5 h; c) amine, EDCI, HOBt, CH2Cl2, RT, 2–24 h. Ar=3-chloro-2-
methylphenyl, A: see Tables 1 and 3.
Next, the effect of the aryl moiety on 11b-HSD1 inhibitory
activity was investigated (Table 2). For compounds containing
a cycloheptylamine moiety, only the naphthalen-2-yl and 5-di-
methylaminonaphthalen-1-yl derivatives (8n and 8o) showed
nanomolar inhibitory activities similar to that of 8a. Besides
good activity toward human 11b-HSD1, all compounds exhibit-
ed low activity against m11b-HSD1 (inhibition ratio <35% at
a concentration of 100 nm).
Human 11b-HSD1 shares ~79% sequence identity with the
mouse enzyme, so it might be understandable that many
known 11b-HSD1 inhibitors, including INCB-13739 and PF-
915275, display species-dependent activity, which impedes the
fast track of efficiencies of inhibitors within rodent models.
From previously reported SAR information,[5,8] it is known that
many 11b-HSD1 inhibitors with cross-species potencies have
large lipophilic groups, such as 2-norbornyl in AMG221 and
adamantyl in Merck-544. We conducted further modification
on this series by introducing bulky fused ring systems onto the
amino moiety in order to improve activity toward m11b-HSD1,
in order to be able to perform in vivo studies in a rodent
model. As listed in Table 3, (R)-(+)-bornylamine, 2-norbornyla-
mine, and adamantyl-2-amine derivatives 8s, 8t, and 8v dis-
played nanomolar inhibitory activities against h11b-HSD1 and
m11b-HSD1. As the adamantane group is prone to oxidation
arylsulfonyl chloride provided (R)-benzenesulfonamide (10).
Target compounds with varied amino moieties (5, 8a–f, and
8q–x) were obtained in two successive steps: the hydrolysis of
ester 10 and the condensation of acid and primary amines.
The inhibitory activities of this compound series against
human 11b-HSD1 (h11b-HSD1) and mouse 11b-HSD1 (m11b-
HSD1) were determined by scintillation proximity assay
(SPA),[11] and the results are listed in Tables 1–3. As summarized
in Table 1, the cycloheptylamine and cyclooctylamine deriva-
tives (8a and 8b, respectively) showed improved potencies
toward h11b-HSD1 over that of the cyclohexylamine derivative
5. The tetrahydropyran-4-ylamine and 1-benzylpiperidin-4-yla-
mine derivatives 8c and 8d exhibited dramatically lower activi-
ties against the human enzyme, indicating the well-matched
binding conformations of 8a and 8b. Whereas the azepan-1-yl-
amine analogue 8 f showed moderate activity, the azocane de-
rivative 8e did not show significant inhibition against h11b-
HSD1. However, with the exception of 8b, with an IC50 value of
173 nm against m11b-HSD1, all the cycloalkylamine com-
pounds appear to show weak inhibition toward m11b-HSD1.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 577 – 581 578

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
Table 2. In vitro activity of derivatives with varied aryl groups.
Table 3. In vitro activity of bridged hydrocarbon amine derivatives.
Compd
A
h11b-HSD1
IC50 [nm][a]
m11b-HSD1
IC50 [nm][a]
Compd
A
h11b-HSD1
IC50 [nm][a]
m11b-HSD1
Inhib. [%][b]
8a
8g
8h
3.7
83
32
ND[c]
18
8q
11
13
28[b]
8r
8s
8t
22[b]
29
3.1
6.0
24
47
8i
8j
57
28
12[b]
ND[c]
8k
8l
11
30
35
25
5
8u
8v
34
ND[c]
13
1.5
8m
24
8w
8x
1.2
15
8n
8o
8.2
9.7
25
16
32[b]
ND[c]
[a] Values are the average of three replicates, with a variance of <15%.
[b] Percent inhibition measured at 100 nm; values are the mean of two
measurements. [c] Not determined.
8p
38
29
[a] Values are the average of three replicates, with a variance of <15%.
[b] Percent inhibition measured at 100 nm; values are the mean of two
measurements. [c] Not determined.
Table 4. In vitro DMPK properties of representative compounds.[a]
Compd
HLM CLint
[mLminÀ1 mgÀ1
TDI
DI [%]
2D6
]
3A4
2C9
in vivo, which could lead to rapid elimination of these com-
pounds, we prepared 8w with an additional hydroxy group at
the vulnerable position of the adamantane ring in 8v, and this
compound maintained high potency against the human and
rodent enzymes, similar to 8v.
Based on the results of these enzyme assays, we selected
the three most potent compounds 8s, 8v, and 8w, together
with 8a, for studies of in vitro DMPK properties. As listed in
Table 4, 8w showed greater metabolic stability in human liver
microsomes (HLM) and acceptable levels of cytochrome P450
(CYP) inhibition.
8a
8s
8v
8w
393
183
175
74
No inhib.
No inhib.
No inhib.
No inhib.
93
82
72
42
49
15
34
3
62
29
0
3
[a] The procedures for evaluating metabolic stability and CYP inhibition,
including direct inhibition (DI) and time-dependent inhibition (TDI), are
described in the Supporting Information.
macodynamic (PK/PD) studies in mice. As listed in Table 5, 8w
showed a high clearance rate and low oral bioavailability. How-
ever, the bioavailability via intraperitoneal (i.p.) injection was
dramatically improved (41%), ~14-fold greater than oral bio-
To evaluate the potential of compound 8w as a lead candi-
date targeting 11b-HSD1, we performed pharmacokinetic/phar-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 577 – 581 579

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
Table 5. PK data for compound 8w in mice.[a]
Parameter
Value
CL [mLminÀ1 kgÀ1
]
130
0.76
2.9
41
760.0
Vdss [LkgÀ1
p.o. F [%]
i.p. F [%]
]
i.p. AUC0–t [nghmLÀ1
]
[a] Inhibitor 8w was dosed into male BALB/c mice by intravenous (i.v.) in-
jection (1 mgkgÀ1), oral route (p.o.) (5 mgkgÀ1), and intraperitoneal (i.p.)
injection (5 mgkgÀ1); mean values averaged over three mice. Detailed
procedures are described in the Supporting Information.
Figure 4. Tissue distribution of 8w in BALB/c mice (i.v., 5 mgkgÀ1).
availability. This allowed us to select 8w to observe the correla-
tion between in vitro activity and in vivo inhibition. For the
in vivo PD evaluation, normal mice were i.p. injected with 8w
(or vehicle). After 1 h, these mice were treated with an intrave-
nous (i.v.) injection of the exogenous substrate prednisone
(3 mgkgÀ1), which would be transformed into prednisolone via
11b-HSD1 catalysis, to evaluate the effects of inhibitors on pre-
dnisolone turnover. As illustrated in Figure 3, i.p. treatment of
BALB/c mice with 8w at 3.75, 15, and 60 mgkgÀ1 dose-de-
pendently decreased the formation of prednisolone, with re-
spective inhibition ratios of 40, 72, and 88%.
1 h after i.v. treatment, whereas the concentration in plasma
dramatically decreased 15 min after administration. Compound
8w was also eliminated moderately in adipose and liver tissue
over the subsequent 5 h, which may be the reason that 8w
showed its effects on decreasing prednisolone levels in the
in vivo assays.
In summary, we report herein a series of 1-arylsulfonylpiperi-
dine-2-carboxamides as 11b-HSD1 inhibitors. Initial enzyme
assays indicated that the R enantiomer is much more potent
than the S enantiomer, and compound 5 was selected as
a starting point for optimization and SAR studies. Inhibitor 8w,
obtained from several rounds of optimizations, showed cross-
species inhibition against human and mouse 11b-HSD1. It also
displayed an acceptable DMPK profile in vitro, and was ad-
vanced to PK/PD evaluations in vivo. The results confirmed its
dose-dependent activity in mice. Therefore, we have identified
1-arylsulfonylpiperidine-2-carboxamides as suitable compounds
for further studies as promising inhibitors of 11b-HSD1 for the
treatment of metabolic syndrome.
Acknowledgements
We are indebted to Prof. Ying Leng (Shanghai Institute of Materia
Medica) for the 11b-HSD1 assay. We are grateful for financial
support from the New Drug Creation Project of the National Sci-
ence and Technology Major Foundation of China (project
2010ZX09401-404), to Shanghai Pharmaceutical Holding, Shang-
hai Postdoctoral Sustentation Fund China (grant no. 07R214213
to G.X.), and the Program of Excellent Young Scientist of the Chi-
nese Academy of Sciences (grant no. KSCX2-EW-Q-3-01 to B.X.).
Figure 3. In vivo inhibitory activity of 8w against 11b-HSD1 in BALB/c mice
by measuring the turnover ratio and the concentration of prednisolone. The
insert shows the mean prednisolone concentration values after treatment
with three doses of inhibitor 8w, as indicated. Student’s t-test was used to
compare the differences between the dosing group and vehicle; *p<0.05
vs. vehicle.
Keywords: 11b-HSD1 · diabetes · docking · inhibitors
[1] A. J. Lusis, A. Attie, K. Reue, Nat. Rev. Genet. 2008, 9, 819–830.
[2] W. Gade, J. Schmit, M. Collins, J. Gade, Clin. Lab Sci. 2010, 23, 51–61.
[3] G. Hollis, R. Huber, Diabetes Obes. Metab. 2011, 13, 1–6.
[4] N. M. Morton, Mol. Cell. Endocrinol. 2010, 316, 154–164.
[5] J. S. Scott, J. Deschoolmeester, E. Kilgour, R. M. Mayers, M. J. Packer, D.
Hargreaves, S. Gerhardt, D. J. Ogg, A. Rees, N. Selmi, A. Stocker, J. G.
Swales, P. R. Whittamore, J. Med. Chem. 2012, 55, 10136–10147.
[6] R. Courtney, P. M. Stewart, M. Toh, M. N. Ndongo, R. A. Calle, B. Hirsh-
berg, J. Clin. Endocrinol. Metab. 2008, 93, 550–556.
High intra-adipose tissue cortisol levels have been proposed
to be a major determinant of metabolic diseases such as obesi-
ty, T2D, and metabolic syndrome, and thus it is desirable to
discover adipose tissue-specific 11b-HSD1 inhibitors.[8,12] There-
fore, we next measured the tissue distribution of 8w in BALB/
c mice (Figure 4). The concentration of 8w in subcutaneous
and celiac adipose as well as liver tissue remains high within
[7] J. Rosenstock, S. Banarer, V. A. Fonseca, S. E. Inzucchi, W. Sun, W. Yao, G.
Hollis, R. Flores, R. Levy, W. V. Williams, J. R. Seckl, R. Huber, Diabetes
Care 2010, 33, 1516–1522.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 577 – 581 580

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
[8] M. Wamil, J. R. Seckl, Drug Discovery Today 2007, 12, 504–520.
[9] C. Fotsch, M. Wang, J. Med. Chem. 2008, 51, 4851–4857.
[10] G. Xia, L. Liu, M. Xue, H. Liu, J. Yu, P. Li, Q. Chen, B. Xiong, X. Liu, J.
Shen, Mol. Cell. Endocrinol. 2012, 358, 46–52.
[14] H. Cheng, J. Hoffman, P. Le, S. K. Nair, S. Cripps, J. Matthews, C. Smith,
M. Yang, S. Kupchinsky, K. Dress, M. Edwards, B. Cole, E. Walters, C. Loh,
J. Ermolieff, A. Fanjul, G. B. Bhat, J. Herrera, T. Pauly, N. Hosea, G. Pa-
deres, P. Rejto, Bioorg. Med. Chem. Lett. 2010, 20, 2897–2902.
[11] a) S. Mundt, K. Solly, R. Thieringer, A. Hermanowski-Vosatka, Assay Drug
Dev. Technol. 2005, 3, 367–375; b) G. Xia, M. Xue, L. Liu, J. Yu, H. Liu, P.
Li, J. Wang, Y. Li, B. Xiong, J. Shen, Bioorg. Med. Chem. Lett. 2011, 21,
5739–5744.
Received: January 14, 2013
Revised: February 18, 2013
Published online on March 7, 2013
[12] E. Saiah, Curr. Med. Chem. 2008, 15, 642–649.
[13] O. Trott, A. J. Olson, J. Comput. Chem. 2010, 31, 455–461, http://vina.
scripps.edu.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 577 – 581 581