Discovery of 1-(4-phenoxypiperidin-1-yl)-2-arylaminoethanone stearoyl-CoA desaturase 1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.03.099

A series of novel stearoyl-CoA desaturase 1 (SCD1) inhibitors were identified by scaffold design based on known SCD1 inhibitors. Large structural changes were made leading to multiple analogs with comparable or improved potency. This approach is valuable

Full text of DOI:10.1016/j.bmcl.2007.03.099

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3388–3391
Discovery of 1-(4-phenoxypiperidin-1-yl)-2-arylaminoethanone
stearoyl-CoA desaturase 1 inhibitors
Hongyu Zhao,*,ꢀ Michael D. Serby, Harriet T. Smith, Ning Cao, Tom S. Suhar,
Teresa K. Surowy, Heidi S. Camp, Christine A. Collins, Hing L. Sham and Gang Liu
Metabolic Disease Research, Global Pharmaceutical Research and Development, R4MC, AP-10, Abbott Laboratories,
100 Abbott Park Road, Abbott Park, IL 60064-6098, USA
Received 27 February 2007; revised 22 March 2007; accepted 29 March 2007
Available online 5 April 2007
Abstract—A series of novel stearoyl-CoA desaturase 1 (SCD1) inhibitors were identified by scaffold design based on known SCD1
inhibitors. Large structural changes were made leading to multiple analogs with comparable or improved potency. This approach is
valuable for generation of proprietary compounds without conducting a costly high-throughput screening.
Ó 2007 Elsevier Ltd. All rights reserved.
The quality of lead compounds and druggability of bio-
logical targets dictate the lead optimization strategies.1
When the lead is a literature compound, one needs to
drastically change its structure to secure an unquestion-
able intellectual property (IP) position.2 Scaffold hop-
ping is a computational technique that identifies
compounds with different skeletons that match the pre-
defined pharmacophore in a given database.3 Medicinal
chemists can also design and synthesize novel molecules
with a different scaffold from the lead compound and
nonetheless show similar or improved properties. Here
we report a successful application of this strategy in
the discovery of a series of potent stearoyl-CoA desatur-
ase 1 (SCD1) inhibitors based on competitor’s com-
pounds. This approach is cost and time effective
because no internal high-throughput screening is needed
and is particularly desirable for those drug discovery
programs without screening capacity.
served in SCD1 deficient Asebia mice9 and SCD1
knockout mice.10,11 Inhibition of SCD1 activity via anti-
sense oligonucleotides in diet-induced obese (DIO) mice
resulted in lower adiposity and higher energy expendi-
ture.12 Higher SCD1 activity has been linked to elevated
plasma triglyceride level in humans.13 Small molecule
SCD1 inhibitors are expected to be beneficial in treating
obesity and the related metabolic syndrome.
When we started the program, the only known class of
SCD1 inhibitors was piperidyl arylcarboxamides re-
ported by scientists from Xenon Pharmaceuticals (e.g.,
compound 1, Figure 1).14–19 To change the molecular
skeleton of 1 and yet retain its biological activity, we
first needed to hypothesize a pharmacophore model.
Since the structure of SCD1 was not known, only li-
gand-based design was possible. Compound 1 was a
fairly efficient ligand for SCD1 considering its size,
molecular complexity, and level of functionalization.
The pyridazine ring in 1 could be replaced by other het-
erocycles such as isomeric pyridines, pyrimidines, and
pyrazines without losing much of the activity, suggesting
this region might tolerate larger structural changes. The
process of deriving the pharmacophore 4 is outlined in
Figure 1. Structure 2 has similar functional groups to
those in lead 1 although their most stable conformations
would be quite different. Structure 3 is a close analog of
1 and 2 that retains all of their heteroatoms and continu-
ity of sp2 hybridized carbon atoms if linker (‘L’) is an
aromatic ring. To turn 3 into a stable molecule, pharma-
cophore 4 was envisioned. In structure 4, a more flexible
SCD1 is a microsomal enzyme that catalyzes the rate-
limiting step in the biosynthesis of monounsaturated
fatty acids from saturated fatty acids.4–6 It plays an
important role in lipid metabolism7 and body weight
control.8 Reduced adiposity, increased insulin sensitiv-
ity, and resistance to diet-induced obesity have been ob-
Keywords: Scaffold manipulation; Molecule design; SCD1 inhibitors.
*
Corresponding author. Tel.: +1 847 287 3894; fax: +1 847 788
9641; e-mail: zhao_hongyu@hotmail.com
ꢀ
Present address: Build-A-Scaffold Medchem Consulting LLC, 968
Crofton Lane, Buffalo Grove, IL 60089, USA.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.099

H. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3388–3391
3389
O
O
structure of 5 has more carbon atoms than 1, which
might lead to unfavorable interactions and desolvation
energy for binding. Indeed, a pyridine derivative 6a
showed a comparable potency to 1 and a pyrazine ana-
log 7a demonstrated an even stronger inhibitory activity
than the lead compound 1. The isomeric pyridine ana-
logs 8 and 9 were much less active.
N
N
Cl
Cl
MeHN
N
MeHN
X
N
X
IC50 0.32 μM
O
O
O
MeHN
O
O
N
L
N
X
Cl
X
Cl
Having identified two new scaffolds (those of 6a and 7a,
respectively), more thorough SAR investigations were
initiated to search for more potent inhibitors. As sum-
marized in Tables 2 and 3, a series of 2,3-substituted
pyridine and pyrazine carboxylamides were synthesized
and assayed. Although, most analogs showed good po-
tency against human SCD1, a relatively flat SAR was
observed with primary or short secondary amides being
slightly favored. Tertiary amides and branched second-
ary amides (e.g., 6j and 6k in comparison to 6c and
6e, Table 2) were much less favorable. The alkyl portion
MeHN
L
X
O
4
sp2
3
pharmacophore
O
X: O or N
L: linker
Figure 1. The logic of generating pharmacophore 4.
and hydrophobic sp3 hybridized carbon atom is intro-
duced resulting in a large conformational change and
an increase of hydrophobicity. To accommodate this
new feature, screening several polar aromatic linkers
(‘L’) was deemed necessary to find an optimal scaffold.
Table 2. SAR of pyridine-based SCD1 inhibitors
Thus several aromatic linkers (‘L’ in 4) were evaluated
first and the results are summarized in Table 1. Based
on the SAR information on 1, the 2-trifluoromethyl
and 2-chlorophenoxy groups on the right hand side
and the methyl and benzyl amides at the left hand side
were very similar and they were interchangeably used
depending on the commercial availability of the starting
materials and chemistry used. A benzene analog (inhib-
itor 5) showed a weaker potency than 1.20 The core
N
R2
O
N
N
H
R1
O
No.
R1
R2
IC50 (lM)
6a
6b
6c
6d
6e
6f
Cl
CONHBn
CONH2
0.56 0.27
0.090 0.12
0.097 0.017
0.11 0.085
0.39 0.14
0.12 0.021
0.43 0.38
CF3
CF3
CF3
CF3
CF3
CF3
CONHMe
CONHEt
CONHPr
CONH(CH2)2OH
CONHCH2(m-pyridyl)
Table 1. Evaluation of different aromatic linkers (‘L’ in 4, Fig. 1)
6g
O
6h
CF3
CONHCH2(p-pyridyl)
0.35 0.35
R2
N
N
R1
6i
CF3
CF3
CF3
CONH(CH2)2Ph
CONMe2
CONH(c-pentyl)
0.78
>10
>10
H
O
6j
6k
No.
R1
Cl
R2
IC50 (lM)
O
O
Bn
Bn
5
N
0.58
Table 3. SAR of pyrazine-based SCD1 inhibitors
H
N
N
R
O
N
N
H
Cl
N
O
6a
7a
Cl
Cl
N
H
0.56 0.27
No.
R
IC50 (lM)
7a
7b
7c
7d
7e
7f
CONHBn
CONH2
0.14 0.00
0.094 0.033
0.051 0.040
0.039 0.060
0.26 0.22
O
N
N
Bn
N
H
0.14 0.0
CONHMe
CONHEt
CONH(iBu)
CONH(CH2)2OH
CONHCH2(o-pyridyl)
0.10 0.008
0.11 0.11
H
N
O
7g
Bn
8
9
CF3
7.3 0.0
7h
7i
CONHCH2(m-pyridyl)
CONHCH2(p-pyridyl)
0.099 0.11
0.11 0.017
N
7j
CONHPh
CO2Me
CHO
CN
0.55 0.45
0.039 0.020
0.21 0.049
0.34 0.33
O
7k
7l
CF3
N
>1
H
7m
7n
N
Cl
0.030 0.0007

3390
H. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3388–3391
Table 4. SAR of the aryl at the 4-position of the piperidine ring
OEt
OEt
NHMe
O
O
b
a
NH
O
O
N
N
R
OH
N
O
N
Cl
N
N
N
N
H
F3C
H
N
O
O
15
16
9
N
H
O
Scheme 2. Reagents and conditions: (a) 1—NH2CH2CO2 tBuÆHCl,
Et3N, NMP, 150 °C, 2—HCl (4 N in dioxane); (b) 1—4-(2-trifluorom-
ethylphenoxy)piperidine hydrochloride, TBTU, Et3N, 2—LiOH, 3—
MeNH2, TBTU, Et3N.
No.
R
IC50 (lM)
7c
2-Chlorophenoxyl
Phenoxyl
2-Bromophenoxyl
0.051 0.040
6.3
10a
10b
10c
10d
10e
10f
10g
10h
0.058 0.0007
0.038 0.017
0.051 0.030
0.47 0.18
0.035 0.016
0.10 0.067
>1
the 5-position (10c and 10f) of the phenoxy ring also
seemed to increase the potency, but to a lesser extent.
A fluorine atom at the 3-position of the phenoxy ring ex-
erted negative effect on the potency (10e). Replacing the
phenoxy group with an analino group resulted in a sim-
ilarly potent inhibitor (10g), but with a benzoyl group
led to an inactive inhibitor (10h).
2,5-Dichlorophenoxyl
2-Chloro-5-fluorophenoxyl
2,3-Difluorophenoxyl
2-Chloro-3,5-difluorophenoxyl
2-Trifluoromethyl-phenylamino
Bz
of the secondary amides seemed to provide little interac-
tions, which suggested it might point to the solvent.
Furthermore, the amide functional group did not ap-
pear critical for SCD1 inhibition. For example the ester
(7k, Table 3), aldehyde, cyano, and chloro groups (7l,
7m, and 7n, respectively, Table 3) at the same position
all showed respectable potency. In fact, chloro analog
7n was one of the most active inhibitors.
The syntheses of the inhibitors are outlined in Schemes
1–3. The glycine moiety was either introduced via an
aromatic substitution reaction (Schemes 1 and 2) or a
reductive amination reaction (Scheme 3). The carboxyl
group on the aromatic ring was introduced via a palla-
dium catalyzed carbonylation reaction. The amide
bonds were formed via a standard TBTU mediated
amide-coupling reaction. The other transformations
were straightforward and well-documented in the
literature.
The SAR on the aryl groups at the 4-position of the
piperidine ring was also studied and the results are sum-
marized in Table 4. The 2-halogen clearly helped the
binding as plain phenoxy analog 10a showed only weak
inhibitory activity against SCD1. Additional halogens at
In conclusion, a series of new SCD1 inhibitors with
good potency were identified without any screening
effort. The scaffold design was based on medicinal
OMe
OMe
N
N
N
N
O
N
Cl
Cl
O
O
a
b
OH
N
N
N
Cl
N
H
H
O
O
11
12
7k
R
HN
N
O
c
O
N
N
N
H
Cl
7a-j
O
N
N
X
O
O
OH
d
e
N
N
HN
N
H
Cl
or 7n X=Cl
H2N
Cl
O
O
13
14
7m X=CN
7l X=CHO
f
OMe
O
NHMe
O
N
N
N
N
O
g
OH
N
N
H
12
N
H
R
O
O
10a-h
Scheme 1. Reagents and conditions: (a) 1—NH2CH2CO2 tBuÆHCl, Et3N, NMP, 150 °C, 2—PdCl2(dppf)-CH2Cl2, CO (60 psi), MeOH, 100 °C,
3—HCl (4 N in dioxane); (b) 4-(2-chlorophenoxy)piperidine hydrochloride, TBTU, Et3N; (c) 1—LiOH, 2—corresponding amines, TBTU, Et3N; (d)
1—N-Boc-glycine, TBTU, Et3N, 2—2-chlorophenol, Ph3P, DEAD, 3—HCl (4 N in dioxane); (e) 2-chloro-3-cyano-pyrazine (for 7l and 7m) or 11 (for
7n), Et3N, NMP, 140 °C; (f) DIABL-H; (g) 1—corresponding piperidine derivatives, TBTU, Et3N, 2—LiOH, 3—MeNH2, TBTU, Et3N.

H. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3388–3391
5. Ntambi, J. M. Prog. Lipid Res. 1995, 34, 139.
3391
OMe
O
R
6. Enoch, H. G.; Catala, A.; Strittmatter, P. J. Biol. Chem.
1976, 251, 5095.
7. Cohen, P.; Ntambi, J. M.; Friedman, J. M. Curr. Drug
Targets Immune Endocr. Metabol. Disord. 2003, 3, 71.
8. Dobrzyn, A.; Ntambi, J. M. Trends Cardiovasc. Med.
2004, 14, 77.
N
O
N
Cl
a
b
N
N
H
NH2
O
17
18 R=Cl
19 R=CF3
R
9. Zheng, Y.; Eilersten, K. J.; Ge, L.; Zhang, L.; Sundberg, J.
P.; Prouty, S. M.; Stenn, K. S.; Parimoo, S. Nat. Genet.
1999, 23, 268.
HN
R
N
O
O
N
10. Ntambi, J. M.; Miyazaki, M.; Stoehr, J. P.; Lan, H.;
Kendziorski, C. M.; Yandell, B. S.; Song, Y.; Cohen, P.;
Friedman, J. M.; Attie, A. D. Proc. Natl. Acad. Sci.
U.S.A. 2002, 99, 11482.
N
H
6a from 18
6b from 19
O
-
k
EtO
O
11. Ntambi, J. M.; Miyazaki, M. Prog. Lipid Res. 2004, 43,
91.
12. Jiang, G.; Li, Z.; Liu, F.; Ellsworth, K.; Dallas-Yang, Q.;
Wu, M.; Ronan, J.; Esau, C.; Murphy, C.; Szalkowski, D.;
Bergeron, R.; Doebber, T.; Zhang, B. B. J. Clin. Invest.
2005, 115, 1030.
OEt
5
(from 20)
or
O
N
a, b, c
or
8
(from 21)
NH2
20
NH2
21
13. Attie, A. D.; Krauss, R. M.; Gray-Keller, M. P.; Brownlie,
A.; Miyazaki, M.; Kastelein, J. J.; Lusis, A. J.; Stalenhoef,
A. F.; Stoehr, J. P.; Hayden, M. R.; Ntambi, J. M. J. Lipid
Res. 2002, 43, 1899.
14. Abreo, M.; Chafeev, M.; Chakka, N.; Chowdhury, S.; Fu,
J.; Gschwend, H. W.; Holladay, M. W.; Hou, D.; Kamboj,
R.; Kodumuru, V.; Li, W.; Liu, S.; Raina, V.; Sun, S.;
Sun, S.; Sviridov, S.; Tu, C.; Winther, M. D.; Zhang, Z.
WO 011655, 2005.
15. Fu, J.; Kodumuru, V.; Sun, S.; Winther, M.; Fine, R. M.;
Harvey, D. F.; Klebansky, B.; Gray-Keller, M. P.;
Gschwend, H. W.; Li, W. US 0119251, 2005.
16. Kamboj, R.; Zhang, Z.; Fu, J.; Seid Bagherzadeh, M.;
Sviridov, S.; Chowdhury, S.; Liu, S.; Kodumuru, V. WO
034279, 2006.
17. Kamboj, R.; Zhang, Z.; Fu, J.; Sviridov, S.; Sun, S.; Seid
Bagherzadeh, M.; Raina, V.; Hou, D.; Chowdhury, S.;
Liu, S.; Kodumuru, V.; Chakka, N. WO 034338, 2006.
18. Kamboj, R.; Fu, J.; Kodumuru, V.; Liu, S.; Sadalapure,
K.; Chakka, N.; Hou, D.; Sun, S. WO 034312, 2006.
19. Kamboj, R.; Zhang, Z.; Sviridov, S.; Raina, V.; Hou, D.;
Kodumuru, V.; Seid Bagherzadeh, M. WO 034441, 2006.
20. The activity was determined by measuring the production
of tritiated water from the desaturation of 3H (9,10)
stearoyl-CoA to oleoyl-CoA using human recombinant
SCD1. The 50 lL of reaction contains: 5 lL of a diluted
test compound, 10 lM cold stearoyl-CoA, 0.24 lM 3H
(9,10) stearoyl-CoA and 12.5 lg recombinant human
SCD1 expressed in tandem with human cyt b5/cyt b5R.
After 30 min,the reaction was stopped by 4 N HCl and
applied to 96-well plates containing pre-wetted charcoal in
a filter well. Centrifugation is used to pull the aqueous
phase through the charcoal into receiving plate. Scintilla-
tion fluid is added and samples are counted in the Perkin–
Elmer Microbeta Trilux.
Scheme 3. Reagents and conditions: (a) 1—glyoxylic acid, NaBH3CN,
2—PdCl2(dppf)-CH2Cl2, CO (60 psi), MeOH, 115 °C, 3—4-(2-chloro-
phenoxy)piperidine hydrochloride (for 18 and 5) or 4-(2-trifluoro-
phenoxy)piperidine hydrochloride (for 6j and 8), TBTU, Et3N; (b) 1—
LiOH, 2—corresponding amines, TBTU, Et3N.
chemistry principles without the knowledge of the three-
dimensional structure of the enzyme. Although, SCD1 is
likely a promiscuous protein and the success of the ap-
proach used in this study may not reproduce on less
druggable proteins, scaffold design clearly offers advan-
tages if is appropriately implemented. This method is
especially useful for programs without access to large
compound collections for screening or the 3D structure
of the target for structure-based design.
Acknowledgment
We thank protein expression/purification group at
Abbott for providing the human SCD1.
References and notes
1. Bleicher, K. H.; Bo¨hm, H. J.; Muller, K.; Alanine, A. I.
¨
Nat. Rev. Drug Disc. 2003, 2, 369.
2. Zhao, H. Drug Discovery Today 2007, 12, 149.
3. Schneider, G.; Neidhart, W.; Giller, T.; Schmid, G. Angew
Chem. Int. Ed. Engl. 1999, 38, 2894.
4. Miyazaki, M.; Ntambi, J. M. Prostaglandins Leukot.
Essent. Fatty Acids 2003, 68, 113.