Discovery and SAR development of thienopyridones: A class of small molecule AMPK activators

Article abstract of DOI:10.1016/j.bmcl.2007.04.011

AMP-activated protein kinase (AMPK) is well established as a sensor and regulator of intracellular and whole-body energy metabolism. A high-throughput screen was performed in order to identify chemotypes that are bound by AMPK. A novel thienopyridone comp

Full text of DOI:10.1016/j.bmcl.2007.04.011

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3254–3257
Discovery and SAR development of thienopyridones: A class
of small molecule AMPK activators
Gang Zhao,* Rajesh R. Iyengar, Andrew S. Judd, Barbara Cool, William Chiou,
Lemma Kifle, Ernst Frevert, Hing Sham and Philip R. Kym
Metabolic Disease Research, Abbott Laboratories, 100 Abbott Park Road, Abbott Park, IL 60064, USA
Received 13 February 2007; revised 30 March 2007; accepted 4 April 2007
Available online 10 April 2007
Abstract—AMP-activated protein kinase (AMPK) is well established as a sensor and regulator of intracellular and whole-body
energy metabolism. A high-throughput screen was performed in order to identify chemotypes that are bound by AMPK. A novel
thienopyridone compound (1) was identified and subsequently optimized. The structure–activity relationships that emerged from
this effort are described.
Ó 2007 Elsevier Ltd. All rights reserved.
Obesity and associated diseases like type 2 diabetes, the
metabolic syndrome, hypertension, and atherogenic
dyslipidemia represent major health risks around the
world. An estimated 194 million people have either type
1 or 2 diabetes, according to the International Diabetes
Federation. Type 2 diabetes is the most common and
fastest growing form of the disease and is often
complicated by obesity. AMPK (adenosine monophos-
phate-activated protein kinase), a heterotrimeric serine/
threonine kinase, is well established as a key sensor
and regulator of intracellular and whole-body energy
metabolism.1 Activation of AMPK alters carbohydrate
and lipid metabolism to increase fatty acid oxidation
and glucose uptake and decrease fatty acid and choles-
terol synthesis.2 Through its central role in the regula-
tion of glucose and lipid metabolism, AMPK is
emerging as an attractive molecular target for the treat-
ment of diabetes, metabolic syndrome, and obesity.2,3
As part of our effort to identify and assess the effects of
AMPK activation on appropriate metabolic parameters,
a high-throughput screening was performed from which
several structurally diverse AMPK activators were iden-
tified. This report describes the systematic evaluation of
the thienopyridone agonist 1, in addition to the struc-
ture–activity relationships (SAR) of the resultant
analogs.
We initiated our efforts with a resynthesis of the HTS hit
1 (Fig. 1), followed by the preparation of several direct
analogs. The synthesis of the thienopyridone com-
pounds is outlined in Scheme 1.5a,6 Briefly, acetophe-
none derivatives were treated with ethyl cyanoacetate,
sulfur, and morpholine, heated at 60 °C to afford the
2-amino-4-aryl-thiophene-3-carboxylic acid ethyl esters
2. The appropriate ethyl ester 2 was then reacted with
cyanoacetic acid chloride, which was made fresh by
treating cyanoacetic acid with PCl5 in dichloromethane,
There have been a number of reports of small molecule
regulators of AMPK, such as AICAR, metformin and
rosiglitazone.4 These are all indirect activators, however,
and thus to further elucidate the physiological conse-
quences of AMPK activation, our laboratory sought
to identify small molecules that directly interact with
this enzyme.
OH
CN
S
N
O
H
1
AMPK Rat liver EC50 (μM) 38
Keywords: AMPK; Obesity; Diabetes.
*
Corresponding author. Tel.: + 1 847 272 9398; fax: + 1 847 938
1674; e-mail: zhaoxing2@yahoo.com
Figure 1. High-throughput screen hit: thienopyridone compound 1.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.04.011

G. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3254–3257
3255
R1
Table 1. SAR of 3-aryl thienopyridones
R1
R1
COOEt
OH
a
b
CN
O
O
R2
S
c
NH2
R1
2
S
N
H
R1
Compound R1
R2
AMPK Rat liver
a
EC50 (lM)
OH
COOEt
CN
O
O
1
5
H
H
38
38
S
S
3-OH
4-OH
H
CN
N
H
N
H
6
H
20
20
1, 5-13
3
7
3,4-OCH2O
4-F
4-Cl
H
8
H
84
81
d
e
9
H
10
11
12
13
14
15
16
17
18
19
4-Br
H
72
55
R1
R1
4-COOMe
4-Allyl
4-OCH2OCH3
H
H
40
8
OH
H
COOEt
O
c
CN
O
H
Cl
Br
Cl
Br
NO2
3.7
5.8
2
R2
R2
H
S
CN
S
N
H
4-OH
4-F
H
N
H
10
Inactive
4
14-19
H
2-OH-C6H4 Inactive
Scheme 1. Reagents and conditions: (a) CH2(CN)CO2Et, S, morphol-
ine, EtOH, 60 °C (20–45%); (b) CH2(CN)CO2H, PCl5, Et3N, DCM
(90%); (c) NaH, THF, reflux (80%); (d) SOCl2, DCM (90%); (e)
C5H5NHBr3, AcOH or NCS/NBS, AcOH (30–60%).
a All compounds were >95% pure by HPLC and characterized by 1H
NMR and HRMS. Values represent an average of at least two
determinations.
in the presence of triethylamine at 60 °C to provide the
2-(2-cyano-acetylamino)-4-aryl-thiophene-3-carboxylic
acid ethyl esters 3. Following this step, the ethyl esters 3
were treated with sodium hydride in tetrahydrofuran at
temperatures ranging from 25 to 80 °C to provide the fi-
nal thienopyridone compounds. The final products were
very polar and typically required reverse-phase HPLC
purification.
Encouraged by the increase in AMPK activation affor-
ded by the appropriately substituted phenyl analogs,
we turned our efforts towards the incorporation of func-
tionality at the 2-position of the thienopyridine core. 2-
Chloro and 2-bromo substitution was effected according
to Scheme 2. Briefly, 4-phenylthiophene intermediate 3
was treated with thionyl chloride in dichloromethane
to afford the chlorinated intermediate 4 in 90% yield.
Treatment with sodium hydride rapidly afforded the fi-
nal cyclized product in good yields. Alternatively, the
Compound 1, which showed modest AMPK activity (rat
liver EC50, 38 lM),7 was our starting point for optimiza-
tion (Table 1). The SAR investigation was initiated with
the functionalization of the unsubstituted 3-phenyl ring.
Derivatization of this ring with either a meta- or para-
hydroxy group either retained or had a twofold
improvement of activity of parent 1, respectively (see 5
and 6, Table 1). Tying these hydroxyls into a five-mem-
bered ring to produce the methylenedioxyphenyl analog
7 also improved the activity twofold, resulting in an
EC50 of 20 lM. Other fused ring systems resulted in a
diminution of AMPK activity, along with various
ortho-substitutions (data not shown). Since the para-
substitution position appeared to be the most produc-
tive position for increasing the potency, several more
analogs with substituents at this position were prepared.
Replacement of the hydroxyl group in 3 with fluoro,
chloro or bromo afforded analogs that had lowered po-
tency as compared to parent 1. While no gain in potency
was observed with the para ester 11, the allyl-substituted
analog 12 resulted in a compound equipotent to 1. Fi-
nally, capping the para-hydroxyl moiety of 6 with a
methoxymethyl group afforded the most potent analog
in this group of compounds, as analog 13 showed a sig-
nificantly improved EC50 of 8 lM.
OH
O
CN
O
CN
O
a
S
S
N
H
N
H
20
R2
R2
OH
OH
N
CN
O
CN
O
b
R1
R1
S
S
N
H
21: R1 = H, R2 = H
22: R1 = H, R2 = Cl
23: R1 = F, R2 = Br
Scheme 2. Reagents and condition: (a) Diazomethane, diethyl ether
(50%); (b) NaH, MeI, THF, reflux (40%).

3256
G. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3254–3257
Table 2. SAR of 5-substituted thienopyridones
2-chlorinated analogs could also be prepared by the
treatment of analogs 1 and6 with N-chlorosuccinimide,
although the yields in this transformation were low
(30%). The 2-brominated analogs were prepared in a
straightforward manner by the treatment of analogs 1
or 8 with NBS, affording the desired products in about
60% yield.
OH
R
S
N
O
H
a
Compound
R
AMPK Rat liver EC50 (lM)
Evaluation of these analogs in our binding assay re-
vealed that functionalization with electron withdrawing
groups at the 2-position significantly improved AMPK
activation. As demonstrated in Table 1, analogs 14
and 15 had EC50 values of 3.7 and 5.8 lM, respectively.
Compound 16, the 2-chloro-substituted analog of 6,
showed a tenfold improvement in activity relative to
the parent analog. This trend was further present in ana-
log 17. In contrast, incorporation of larger electron-
withdrawing groups, such the nitro moiety present in
18,8 resulted in a loss of potency. The same result was
observed with the phenyl-substituted compound 19.9
25
26
27
28
29
30
31
Ph
Inactive
180
183
3-OMe-C6H4
2-Thiophene
–COOMe
–NMe2
Cl
175
Inactive
88
Inactive
H
a All compounds were >95% pure by HPLC and characterized by 1H
NMR and HRMS. Values represent an average of at least two
determinations.
We next evaluated the requirement of the hydroxyl moi-
ety on the thienopyridone core. To this end, the chemi-
cal reactivity of the acidic-OH and -NH were
investigated. While basic alkylation conditions (sodium
hydride, methyl iodide) exclusively afforded N-alkylated
products, O-alkylation required more neutral conditions
(CH2N2) for an efficient and selective transformation
(Scheme 2). Finally, the amino analog 24 was prepared
in an analogous manner to 1 (see Scheme 3). However,
all of these modifications proved unproductive, as ana-
logs 20–24 were inactive against rat liver AMPK.
As shown in Table 2, only the chloride substituent could
effectively replace the cyano group in the pyridone ring
of 1 (analog 30). Other substitutions resulted in a dimi-
nution or complete loss of AMPK activity up to the con-
centrations tested.
The systematic investigation of the HTS lead structure 1
identified several productive areas for optimization. Spe-
cifically, 3-phenyl ortho-substitution led to improve-
ments as much as tenfold, while 2-substitution
afforded analogs with improvements over fortyfold.
Subsequent reports will disclose further optimization
of these analogs, as well as their evaluation in in vivo
models of glucose lowering.
Our final exploration was directed towards replacement
of the 5-cyano functionality on the pyridone ring. As
shown in Scheme 3, this was achieved analogously to 1
by switching from cyanoacetyl chloride to various
substituted acyl chlorides in the coupling step prior to
cyclization. Our attempts to generate the 5-carboxylic
acid analogs were unsuccessful, as the attempted hydro-
lysis of the nitrile under acidic conditions only afforded
the corresponding decarboxylated product 31.
References and notes
1. (a) Hardie, D. G. Endocrinology 2003, 144, 5179; (b)
Hardie, D. G.; Scott, J. W.; Pan, D. A.; Hudson, E. R.
FEBS Lett. 2003, 546, 113; (c) Kahn, B. B.; Alquier, T.;
Carling, D.; Hardie, D. G. Cell Metab. 2005, 1, 15; (d)
Hawley, S. A.; Boudeau, J.; Reid, J. L.; Mustard, K. J.;
Udd, L.; Makela, T. P.; Alessi, D. R.; Hardie, D. G. J. Biol.
2003, 2, 28.
Br
Br
2. (a) Carling, D.; Zammit, V. A.; Hardie, D. G. FEBS Lett.
1987, 223, 217; (b) McGarry, J. D.; Brown, N. F. Eur. J.
Biochem. 1997, 244, 1; (c) Saha, A. K.; Schwarsin, A. J.;
Roduit, R.; Masse, F.; Kaushik, V.; Tornheim, K.; Prentki,
M.; Ruderman, N. B. J. Biol. Chem. 2000, 275, 24279; (d)
Hardie, D. G.; Hawley, S. A. Bioessays 2001, 23, 1112.
3. (a) Winder, W. W.; Hardie, D. G. Am. J. Physiol. 1999,
277, E1; (b) Moller, D. E.; Kaufman, K. D. Annu. Rev.
Med. 2005, 56, 45; (c) Schimmack, G.; DeFronzo, R. A.;
Musi, N. Diab. Obes. Metab. 2006, 8, 591.
4. (a) Vincent, M. F.; Erion, M. D.; Gruber, H. E.; van den
Berghe, G. Diabetologia 1996, 39, 1148; (b) Young, M. E.;
Radda, G. K.; Leighton, B. FEBS Lett. 1996, 382, 43; (c)
Fryer, L. G.; Parbu-Patel, A.; Carling, D. J. Biol. Chem.
2002, 277, 25226; (d) Zou, M. H.; Kirkpatrick, S. S.; Davis,
B. J.; Nelson, J. S.; Wiles, I. W.; Schlettner, U.; Neumann,
D.; Brownlee, M.; Freeman, M. B.; Goldman, M. H.
J. Biol. Chem. 2004, 279, 43940.
NH2
CN
NH2
CN
O
a,b
S
N
S
H
24
OH
COOEt
NH
c
R
O
S
S
O
N
H
25-33
R
Scheme 3. Reagents and conditions: (a) CH2(CN)CO2H, PCl5, Et3N,
DCM (90%); (b) NaOEt, EtOH, reflux (43%); (c) NaH, THF, reflux or
KHMDS, THF, PhCH3, 0 °C–rt, (20–70%).

G. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3254–3257
3257
5. (a) Iyengar, R. R.; Judd, A. S.; Zhao, G.; Kym, P. R.;
8.
SIham, H. L.; Gu, Y.; Liu, G.; Liu, M.; Zhao, H.; Clark, R.
F.; Frevert, E. U.; Cool, B. L.; Zhang, T.; Keyes, R. F.;
Hansen, T. M.; Xin, Z. US 2,005,038,068, 2005; U.S. Pat.
Appl. Publ. 2005, p 86; (b) Cool, B.; Zinker, B.; Chiou, W.;
Kifle, L.; Cao, N.; Perham, M.; Dickinson, R.; Adler, A.;
Gagne, G.; Iyengar, R.; Zhao, G.; Marsh, K.; Kym, P. R.;
Jung, P.; Camp, H. S.; Frevert, E. Cell Metab. 2006, 3, 403;
(c) Andersson, S. N.; Cool, B. L.; Kifle, L.; Chiou, W.;
Egan, D. A.; Barrett, L. W.; Richardson, P. L.; Frevert, E.
U.; Warrior, U.; Kofron, J. L.; Burns, D. J. J. Biomol.
Screen. 2004, 9, 112.
1. Fuming HNO3
Ac2O/AcOH/CHCl3
20-30% yield
O
OH
O
CN
O
O2N
S
2. NaH, THF
reflux
80-90% yield
NH
S
N
H
CN
O
3
18
9.
6. Thiophene synthesis: Gewald, K.; Schinke, E.; Boettcher,
H. Chem. Ber. 1966, 99, 94.
7. The rat liver EC50 value was measured by a 96-well
AMPK assay. The AMPK activity was measured by
monitoring phosphorylation of the SAMS peptide
substrate (20 lM in standard assays and 100 lM in
O
B
OH
O
OH
OH
CN
CN
O
Br
CsCO3 / Pd(PPh3)4
S
S
DME/PhCH3/EtOH/H2O
(10 :1 : 6 : 3)
O
N
H
N
H
OH
Microwave 160 oC, 5 min
additivity assays) following
protocol.5c
a
previously described
15
19