GPR109a agonists. Part 1: 5-Alkyl and 5-aryl-pyrazole-tetrazoles as agonists of the human orphan G-protein coupled receptor GPR109a

Article abstract of DOI:10.1016/j.bmcl.2009.03.014

5-Alkyl and aryl-pyrazole-tetrazoles have been identified as a new class of selective, small-molecule, agonists of the human G-protein-coupled receptor GPR109a, a high affinity receptor for the HDL-raising drug nicotinic acid.

Full text of DOI:10.1016/j.bmcl.2009.03.014

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2121–2124
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
GPR109a agonists. Part 1: 5-Alkyl and 5-aryl-pyrazole–tetrazoles as agonists
of the human orphan G-protein coupled receptor GPR109a
a
a
a
a
Jason E. Imbriglio a, , Sookhee Chang , Rui Liang , Subharekha Raghavan , Darby Schmidt ,
Abby Smenton a, Scott Tria a, Thomas O. Schrader c, Jae-Kyu Jung c, Craig Esser a, Andrew K. P. Taggart b,
Kang Cheng b, Ester Carballo-Jane b, M. Gerard Waters b, James R. Tata a, Steven L. Colletti a
*
a Department of Medicinal Chemistry, Merck Research Laboratories, Merck & Co., Inc., PO Box 2000, Rahway, NJ, 07065-0900, United States
b Department of Cardiovascular Diseases, Merck Research Laboratories, Merck & Co., Inc., PO Box 2000, Rahway, NJ, 07065-0900, United States
c Department of Medicinal Chemistry, Arena Pharmaceuticals, 6166 Nancy Ridge Drive, San Diego, CA 92121, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
5-Alkyl and aryl-pyrazole–tetrazoles have been identified as a new class of selective, small-molecule,
agonists of the human G-protein-coupled receptor GPR109a, a high affinity receptor for the HDL-raising
drug nicotinic acid.
Received 24 February 2009
Revised 4 March 2009
Accepted 4 March 2009
Available online 9 March 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Niacin
GPR109a
Atherosclerosis
Nicotinic acid (niacin) 1 has been a leading treatment for dyslip-
ideamia and for the prevention of atherosclerosis for over 40
years.1 Long term clinical studies have revealed niacin’s ability to
reduce mortality from coronary heart disease.2 In spite of niacin’s
clinical significance, patients treated with niacin show low compli-
ance of use due to an intense flushing side effect.3,4 A number of
drug discovery programs have focused on the development of a
‘flush-free’ niacin-like therapy. Despite considerable effort in the
field, the absence of a niacin-related target and/or mechanism of
action have limited such investigations.
plasma FFA that is observed in wild type mice. Furthermore, these
mice also failed to show a niacin-induced cutaneous flush. As a re-
sult, it has been suggested that the niacin-associated flush and
anti-lipolytic effects are both mediated through GPR109a.5a
In light of these recent findings, we initiated a drug discovery
program focused on the identification of a ‘flush-free’ agonist of
the niacin receptor. Our lead identification effort identified pyra-
zole–tetrazole 2 as a promising candidate.8–10 While pyrazole 2
displayed only modest in vitro affinity for GPR109a (Table 1), its
in vivo pharmacology was intriguing (see Fig. 1).11
Recently, a G-protein-coupled receptor, GPR109a, was identi-
fied as a molecular target for niacin.5 Mechanistic studies have sug-
gested that the benefits of niacin therapy may result from the
activation of GPR109a located on adipocytes.6 Recent experiments
have shown that niacin activation of GPR109a results in a reduc-
tion of intracellular cAMP and it is believed that this reduction in
cAMP levels effectively inhibits lypolysis by the negative modula-
tion of hormone sensitive lipase (HSL). As a result, a concomitant
decrease in plasma free fatty acid (pFFA) levels is observed and it
is thought that this reduction in pFFA is crucial to the lipid modu-
lation and resulting therapeutic value of niacin.7
In a mouse plasma free-fatty acid (pFFA) assay, mice treated
with 10 mpk of pyrazole 2 showed a net reduction in plasma free
fatty acid levels comparable to that observed upon treatment with
100 mpk of niacin. Furthermore, in a mouse vasodilation-assay
(mVD), mice treated with 30 mpk of 2 failed to show any signs of
the characteristic cutaneous flushing response observed with nia-
cin treatment. This putative decoupling of the free fatty acid and
cutaneous flushing effect characteristic of niacin treatment capti-
vated out attention. As a result, pyrazole 2 became the focal point
of our medicinal chemistry effort.
The initial goal of the program focused on improving the in vitro
profile of pyrazole 2. With this goal in mind, we proceeded to ex-
plore the SAR around the framework of pyrazole 2. Early in our ef-
forts, we discovered that both the pyrazole and tetrazole moieties
were essential in retaining affinity for the receptor. In addition, it
was discovered that substituents in either the C4 or C6 positions
In control experiments, GPR109a knockout mice, when treated
with niacin therapy, failed to show the characteristic reduction in
* Corresponding author.
E-mail address: jason_imbriglio@merck.com (J.E. Imbriglio).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.014

2122
J. E. Imbriglio et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2121–2124
Table 1
O
Agonism of GPR109a by C5-substituted pyrazole–tetrazolesa
CO2H
i-iii
O
N
H
N
H
N
N
EtO
CO2H
N
4
N
Bn
3
R2
R2
R1
N
1
2
8 and 9
IC50 (lM)
CN
TfO
iv-vii
R1
R2
b
Compd
N
mNBA
hNBA
N
1
2
—
H
—
0.14
1.2
0.14
1.5
Bn
5
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
Tetrazole
8a
8b
8c
8d
8e
8f
8g
8h
9a
9b
9c
9d
9e
9f
9g
9h
9i
9j
9k
9l
Ethyl
n-Propyl
n-Butyl
1.3
4.6
Scheme 1. Reaction conditions: (i) LDA, di-t-butyl oxalate; (ii) BnNHNH2–HCl; (iii)
TFA; (iv) NHS, EDC; (v) NH4OH; (vi) cyanuric chloride; (vii) LDA, Comin’s reagent.
0.26
0.21
0.06
3.6
0.8
2.5
0.69
0.34
0.08
4.8
0.87
5.2
2.5
0.43
16
7.7
5.0
2.7
2.4
ent-n-Butyl
n-Pentane
n-hexane
Cyclopropyl
Cyclopentyl
Ph
3-Me–Ph
2-Me–Ph
4-Cl–Ph
concomitant removal of the benzyl protecting group to afford the
C5-pyrazole derivatives 8a–h and 9a–p (see Scheme 2).
9.0
0.31
11.0
4.2
4.0
1.7
1.6
0.85
3.1
0.52
0.09
0.34
0.14
0.08
0.07
0.04
0.07
As a means of identifying compounds suitable for in vivo anal-
ysis, we explored the in vitro activity of this class of pyrazole–tet-
razoles. As illustrated in Table 1, the C5-alkyl-pyrazoles showed a
markedly improved affinity for both the human (GPR109a) and
murine (Puma-G) variants of the receptor. Ethyl derivative 8a
4-F–Ph
2,4-F–Ph
2,5-F–Ph
2-Cl–Ph
3,4-F–Ph
2,3-F–Ph
2-F–Ph
3-Cl–Ph
3-F–Ph
2,3,5-F–Ph
ent 2,3,5-F–Ph
3,5-F–Ph
0.76
1.9
(1.3
affinity to pyrazole 2, while propyl derivative 8b (0.26
0.69 M) provided a modest boost in potency with both the mouse
and human receptors. Further extension of the carbon side-chain,
as in n-butyl derivative 8c (0.21 M and 0.34 M), showed an
incremental boost in affinity as the racemate; however, a chiral
resolution of 8c, yielding the active stereoisomer 8d (0.06
and 0.08
M)12 provided the most potent agonist within the C5-al-
kyl series. Additional elongation of the alkyl chain, as in pentane
derivative 8e (3.6 M and 4.8 M) and hexane derivative 8f
(0.8 M and 0.87 M) showed a precipitous drop in affinity, when
compared to 8d.
The saturated carbocyclic derivatives, pyrazoles 8g (2.5
M) and 8h (9.0 M and 2.5 M), provided little improvement
in potency when compared to the linear alkyl derivatives, whereas
the C5-phenyl derivative 9a (0.31 M and 0.43 M) showed prom-
l
M and 4.6
l
M) demonstrated nearly comparable in vitro
lM and
0.61
0.14
0.41
0.29
0.14
0.08
0.04
0.07
l
l
l
9m
9n
9o
9p
l
M
l
a
Values are an average of two or greater individual assays.
[3H]-Nicotinic acid binding competition assay.
l
l
b
l
l
l
M and
N
5.2
l
l
l
N
N
NH
4
l
l
5
6
ise as a scaffold for further structural modifications.
N
N
In this regard, a positional scan of a variety of functional groups
on the phenyl ring system of 9a exposed a tight binding pocket
within this region of the molecule.13 It was discovered that phenyl
rings bearing small alkyl groups or halogens were tolerated in this
locale. As illustrated in Table 1, the methyl substituted ring sys-
H
Figure 1. Lead candidate pyrazole–tetrazole 2.
tems 9b (11 lM and 16 lM) and 9c (4.2 lM and 7.7 lM) showed
of the saturated carbon backbone of 2 were poorly tolerated. In
light of these observations, we focused our attention around the
C5 position of the molecule.
modest affinity for the receptor across both species, while halogen
substitution about the phenyl ring showed improved in vitro po-
tency. In particular, the placement of either a fluoro or chloro
In order to explore the C5 position of the molecule most judi-
ciously, we utilized triflate 5 as a late stage synthetic intermediate
that allowed for the installation of both alkyl and aryl substituents
into the framework of the molecule. As outlined in Scheme 1, our
synthesis toward compound 5 began with acylation of the com-
mercially available cyclopentenone 3 followed by a condensation
reaction with benzyl hydrazine to install the pyrazole. Hydrolysis
of the enol-ether and t-butyl ester revealed ketone 4 which was
further elaborated to the nitrile. Finally, the C5 ketone was con-
verted to the desired enol-triflate intermediate 5.
Enol-triflate 5 was converted to the desired C5-alkyl or C5-aryl
substituted pyrazole 6 via Suzuki or Stille couplings with the
appropriate boronic acid or organostannane. With the desired C5
substituent in place, the tetrazole functionality was furnished
through a [3+2] cycloaddition with sodium azide to afford 7. Final-
ly, hydrogenation provided the saturated carbon backbone with
group in the 3-position of the phenyl ring as in 9l (0.14
0.29 M) and 9m (0.08 M and 0.14 M) resulted in a significant
improvement in potency. Whereas, substitution of a chloro group
in either the 2 or 4-position as in 9h (3.1 M and 1.9 M) and 9d
M) or a fluoro group as in 9k (0.34 M and
M) and 9e (1.7 M and 2.7 M) provided only a marginal
lM and
l
l
l
l
l
(4.0
0.41
lM and 5.0
l
l
l
l
l
improvement in potency. Amongst the disubstituted derivatives,
the 3,5-fluoro derivative 9p (0.7 M and 0.7 M) demonstrated im-
proved affinity over the 2,4-difluoro derivative 9f (1.6 M and
2.4 M), 2,5-derivative 9g (0.85 M and 0.76 M) 3,4-derivative
9i (0.52 M and 0.61 M) and 2,3-derivative 9j (0.09 M and
0.14 M). Further fluoro substitution, as in enantiomer 9o of the
l
l
l
l
l
l
l
l
l
l
2,3,5-trifluoro derivative 9n proved to be the most potent com-
pound in the C5-aryl-series.
In order to set the stage for in vivo analysis, mouse pharmaco-
kinetics (mPK) of lead candidates were investigated (Table 2). In

J. E. Imbriglio et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2121–2124
2123
CN
CN
TfO
i
R1
ii
N
N
N
N
Bn
Bn
5
6
N
N
N
N
N
N
N
N
R1
NH
R1
NH
iii
N
H
N
Bn
8a-h, R1-alkyl
9a-p, R1aryl
7
Scheme 2. Reaction conditions: (i) R1B(OH)2, Pd(PPh3)4; (ii) NaN3, ZnBr2; (iii) H2, Pd-C.
0.3
the alkyl series, n-propyl derivative 8b showed a lower clearance
(30 mL/kg/min), longer half life (2.9 h), and an improved bioavail-
ability (51%). when compared with niacin. In contrast, compound
8d, with better in vitro affinity than 8b, demonstrated an inferior
mPK profile when compared with 8b.
Within the aryl series of molecules, the fluoro-derivative 9m
showed improved mPK with good bioavailability (57%), lower
clearance (47 mL/kg/min) and a longer half-life (1.8 h) when com-
pared with niacin. Similarly, the 2,3,5-trifluoro-derivative 9o
showed a longer half-life (2.4 h) and lower clearance (27 mL/kg/
min) with a modest bioavailability (10%).
P = 0.06
0.2
0.1
0.0
P < 0.001
With a promising group of GPR109a agonists in hand, we were
eager to investigate this new series of C5-substituted pyrazole–tet-
razoles in an in vivo setting. In this effort, we chose to explore the
pharmacology of 9o in our mouse pFFA and mouse vasodilation as-
say. As illustrated in Figure 2, six male C57 B1/6 mice were each
dosed at 10 mpk for 15 min with 9o and compared to mice dosed
100 mpk with niacin for an effect on plasma free-fatty acids. In the
case of compound 9o, we observed a characteristic 24% reduction
in plasma free fatty acid levels, while niacin, dosed at 100 mpk,
showed a characteristic 58% reduction of plasma free fatty acids.
Having demonstrated the ability of agonist 9o to effectively
lower plasma free fatty acids in our mouse pharmacodynamics
model, we proceeded to measure the cutaneous vasodilation effect
of 9o on male C57 B1/6 mice, as a surrogate for the flushing side
effect experienced with niacin treatment. In this effort, compound
9o, administered to 8 mice at 30 mpk for 15 min monitored with a
laser doppler imager, showed no evidence of cutaneous flushing
(Fig. 3).
Figure 2. Mouse pFFA reduction studies.
80
60
40
20
0
Vehicle, n=7
NA 30 mg/kg, n=13
9o
30 mg/kg, n=8
-5
0
5
10
15
Time (min)
While nicotinic acid (NA), administered in 10 mice at 30 mpk
showed, on average, a characteristic 80% change in perfusion, typ-
ical of a flushing response in the mouse.14
Figure 3. Mouse vasodilation studies.
Table 2
Mousea pharmacokinetics of pyrazole–tetrazoles
In summary, we have identified a new class of C5-alkyl and
aryl-pyrazole–tetrazoles that act as agonists of GPR109a. We pro-
ceeded to show that selected members of this family of compounds
have improved mouse pharmacokinetic profiles when compared to
niacin. Most notably, as part of an effort to identify and develop a
flush-free niacin-like agonist of GPR109a, we have shown that ago-
nist 9o effectively reduces plasma free-fatty acids without eliciting
a niacin-like flush in our mouse model. Current efforts are focused
on further pre-clinical validation of the pyrazole–tetrazole series of
GPR109a agonists.
Fb (%)
Clp (mL/kg/min)
Vdss (L/kg)
t1/2 (h)
c
Compds
1
100
51
34
57
10
145
30
21
47
27
1.0
0.02
2.9
1.5
1.8
2.4
8b
8d
9m
9o
1.55
0.70
1.79
1.1
a
C57BL/6-mice.
Dose: 1 mg/kg iv; 2 mg/kg po.
t1/2 = plasma half-life (0–8 h).
b
c

2124
J. E. Imbriglio et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2121–2124
6. (a) Tunaru, S.; Kero, J.; Schaub, A.; Wufka, C.; Blaukat, A.; Pfeffer, K.;
Acknowledgment
Offermanns, S. Nat. Med. 2003, 9, 352–355; (b) Zhang, Y.; Schmidt, R. J.;
Foxworthy, P.; Emkey, R.; Oler, J. K.; Large, T. H.; Wang, H.; Su, E. W.; Mosior, M.
K.; Eacho, P. I.; Cao, G. Biochem. Biophys. Res. Commun. 2005, 334, 729–732.
7. Offermanns, S. Trends Pharmacol. Sci. 2006, 27, 384.
8. Identification of leads was based on the ability of compounds to decrease cAMP
levels in forskolin-stimulated cells stably transfected with the niacin receptor
(GPR109a).
The authors would like to thank Dr. Daniel Blom of Merck
Research Labs for helpful discussions.
References and notes
9. Van Herk, T.; Brussee, J.; van den Nieuwendijk, A. M. C. H.; van der Klein, P. A.
M.; Ijzerman, A. P.; Stannek, C.; Burmeister, A.; Lorenzen, A. J. Med. Chem. 2003,
46, 3945–3951.
10. Semple, G.; Schrader, T.; Skinner, P. J.; Colletti, S. L.; Gharbaoui, T.; Imbriglio, J.
E.; Jung, J. K.; Liang, R.; Raghavan, S.; Schmidt, D.; Tata, J. R. U.S. Patent, 2005.
11. Semple, G.; Skinner, P. J.; Gharbaoui, T.; Shin, Y. J.; Jung, J. K.; Cherrier, M. C.;
Webb, P. J.; Tamura, S. Y.; Boatman, P. D.; Sage, C. R.; Schrader, T. O.; Chen, R.;
Colletti, S. L.; Tata, J. R.; Waters, G.; Cheng, K.; Taggart, A. K.; Cai, T. Q.; Jane, E.
C.; Behan, D. P.; Connolly, D. T.; Richman, J. G. J. Med. Chem. 2008, 51, 5105–
5108.
1. Carlson, L. A. J. Int. Med. 2005, 258, 94–114.
2. Tavintharan, S.; Kashyap, M. L. Curr. Atheroscler. Rep. 2001, 3, 74.
3. Knopp, R. H. Am. J. Cardiol. 2000, 86, 51L–56L.
4. For a discussion on the clinical use of the D2 receptor antagonist laropripant as
a means of suppressing the nicotinic acid-induced vasodilation in humans,
refer to: Paolini, J. F.; Mitchel, Y. B.; Reyes, R.; Kher, U.; Lai, E.; Watson, D. J.;
Norquist, J. M.; Meehan, A. G.; Bays, H. E.; Davidson, M.; Ballantyne, C. M. Am. J.
Cardiol. 2008, 101, 625.
5. (a) Lorenzen, A.; Stannek, C.; Lang, H.; Andrianov, V.; Kalvinsh, I.; Schwabe,
U. Mol. Pharmacol. 2001, 59, 349; (b) Wise, A.; Foord, S. M.; Fraser, N. J.;
Barnes, A. A.; Elshourbagy, N.; Eilert, M.; Ignar, D. M.; Murdock, P. R.;
Steplewski, K.; Green, A.; Brown, A. J.; Dowell, S. J.; Szekeres, P. G.; Hassall,
D. G.; Marshall, F. H.; Wilson, S.; Pike, N. B. J. Biol. Chem. 2003, 278, 9869; (c)
Soga, T.; Kamohara, M.; Takasaki, J. M.; Shun-Ichiro, S.; Tetsu, O.; Takahide,
H.; Hideki, M. A.; Matsushime, H.; Furuichi, K. Biochem. Biophys. Res.
Commun. 2003, 303, 364.
12. The enantiomer of 8 d showed only modest affinity for the receptor, 7
(hNBA) and 5 M (mNBA) respectively, illustrating a clear stereochemical
preference for groups substituted in this C5 position.
lM
l
13. A variety of functional groups including alkyl, naphthyl, trifluoromethoxy,
trifluoromethyl, sulfonyl, and methoxy substituted ring systems were not
tolerated in this region.
14. Measured using a Periscan PIMII Laser Doppler Perfusion Imager system.