Discovery of a potent and selective small molecule hGPR91 antagonist

Article abstract of DOI:10.1016/j.bmcl.2011.04.091

GPR91, a 7TM G-Protein-Coupled Receptor, has been recently deorphanized with succinic acid as its endogenous ligand. Current literature indicates that GPR91 plays role in various pathophysiology including renal hypertension, autoimmune disease and retinal angiogenesis. Starting from a small molecule high-throughput screening hit 1 (hGPR91 IC₅₀: 0.8 μM) - originally synthesized in Merck for Bradykinin B₁ Receptor (BK₁R) program, systematic structure-activity relationship study led us to discover potent and selective hGPR91 antagonists e.g. 2c, 4c, and 5g (IC₅₀: 7-35 nM; >1000 fold selective against hGPR99, a closest related GPCR; >100 fold selective in Drug Matrix screening). This initial work also led to identification of two structurally distinct and orally bio-available lead compounds: 5g (%F: 26) and 7e (IC₅₀: 180 nM; >100 fold selective against hGPR99; %F: 87). A rat pharmacodynamic assay was developed to characterize the antagonists in vivo using succinate induced increase in blood pressure. Using two representative antagonists, 2c and 4c, the GPR91 target engagement was subsequently demonstrated using the designed pharmacodynamic assay.

Full text of DOI:10.1016/j.bmcl.2011.04.091

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3596–3602
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of a potent and selective small molecule hGPR91 antagonist
Debnath Bhuniya a, , Dhananjay Umrani a, Bhavesh Dave a, Deepak Salunke a, Gagan Kukreja a,
⇑
Jayasagar Gundu a, Minakshi Naykodi a, Nadim S. Shaikh a, Prasad Shitole a, Santosh Kurhade a,
Siddhartha De a, Sreemita Majumdar a, Srinivasa B. Reddy a, Suhas Tambe a, Yogesh Shejul a,
Anita Chugh a, Venkata P. Palle a, Kasim A. Mookhtiar a, Doris Cully b, Joseph Vacca b,
Prasun K. Chakravarty b, Ravi P. Nargund b, Samuel D. Wright b, Michael P. Graziano b,
Sheo B. Singh b, Sophie Roy b, Tian-Quan Cai b
a Drug Discovery Facility, Advinus Therapeutics, Quantum Towers, Plot-9, Phase-I, Rajiv Gandhi InfoTech Park, Hinjewadi, Pune 411 057, India
b Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
GPR91, a 7TM G-Protein-Coupled Receptor, has been recently deorphanized with succinic acid as its
endogenous ligand. Current literature indicates that GPR91 plays role in various pathophysiology includ-
ing renal hypertension, autoimmune disease and retinal angiogenesis. Starting from a small molecule
Received 17 March 2011
Revised 18 April 2011
Accepted 22 April 2011
Available online 28 April 2011
high-throughput screening hit 1 (hGPR91 IC50: 0.8 lM)—originally synthesized in Merck for Bradykinin
B1 Receptor (BK1R) program, systematic structure-activity relationship study led us to discover potent
and selective hGPR91 antagonists e.g. 2c, 4c, and 5g (IC50: 7–35 nM; >1000 fold selective against hGPR99,
a closest related GPCR; >100 fold selective in Drug Matrix screening). This initial work also led to iden-
Keywords:
GPR91
Succinate
tification of two structurally distinct and orally bio-available lead compounds: 5g (%F: 26) and 7e (IC50
:
180 nM; >100 fold selective against hGPR99; %F: 87). A rat pharmacodynamic assay was developed to
characterize the antagonists in vivo using succinate induced increase in blood pressure. Using two rep-
resentative antagonists, 2c and 4c, the GPR91 target engagement was subsequently demonstrated using
the designed pharmacodynamic assay.
hGPR91 antagonist
Structure–activity relationship
Pharmacodynamic effect
Ó 2011 Elsevier Ltd. All rights reserved.
Succinate, a Kreb’s cycle intermediate, has been recently char-
acterized by He et al. as an endogenous ligand of GPR91, a 7TM
G-Protein-Coupled Receptor (GPCR) and a novel target for drug dis-
covery.1 The closest protein related to GPR91 is another 7TM GPCR
called GPR99 sharing 33% protein sequence identity. GPR99 has
also been deorphanized with another tricarboxylic acid (TCA) cycle
activity and subsequent activation of renin-angiotensin system
(RAS).1 Succinate is accumulated in extra-cellular spaces under
hyperglycemic or ischemic condition. It has been hypothesized
that such an ischemic condition may lead to manifestation of mito-
chondrial dysfunction and the release of succinate as a signal to
activate GPR91 to regulate local blood flow in order to match met-
abolic demands.6,8 Therefore, antagonists of the receptor may
prove helpful in reversing pathophysiology under the clinical set-
ting of renal hypertension and diabetic nephropathy.
In addition, independent works from various research groups
have indicated other potential therapeutic implication of GPR91
antagonism.9–11 GPR91 is highly expressed in quiescent hepatic
stellate cells (HSC)9 and spleen dendritic cells (DCs).10 In DCs, suc-
cinate induced agonism of GPR91 lead to production of proinflam-
matory cytokines. Succinate induced signaling through GPR91 in
liver stellate cells as well as in spleen dendritic cells could lead
to antigen-driven responses of T-cells which correlates to inflam-
mation and autoimmune diseases.9,10 Function of succinate and
GPR91 (present in retinal ganglion neurons) have also been de-
scribed in establishing neovascular network during retinal angio-
genesis and in response to injury.5,11
intermediate,
Whereas, succinate activates GPR91 through both Gi and Gq medi-
ated signaling pathways (EC50 [Ca2+]i: 28–56
M in HEK293 cells
stably expressing hGPR91), -ketoglutarate seems to act exclu-
sively via Gq signaling (EC50 [Ca2+]i: 32–69 M).1,2 Interestingly,
a
-ketoglutarate, by the same research group.1
l
a
l
these two receptors also have distinctions in their expression pat-
tern in various tissues. Whereas, GPR91 has been shown to express
in tissues like kidney, liver, adipose, spleen, heart, retina and intes-
tine, GPR99 is largely expressed in kidney, smooth muscle and tes-
tis.1,3–5
Role of Succinate in hypertension, at least in animals, has been
demonstrated through several in vivo and in vitro mechanistic
studies.1,6–8 Succinate treatment leads to increase in mean arterial
pressure (MAP) via GPR91 mediated action on plasma rennin
Despite a rapid advancement of GPR91 biology, pharmacologi-
cal validation of the target could not been achieved mainly due
⇑
Corresponding author. Tel.: +91 20 66539600; fax: +91 20 66539620.
E-mail address: debnath.b@advinus.com (D. Bhuniya).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.04.091

D. Bhuniya et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3596–3602
3597
to the fact that GPR91 antagonists were not available as tool com-
pounds.12 We report herein identification of the first potent and
selective small molecule GPR91 antagonists towards unfolding
therapeutic potential of the target.
From a high-throughput screening of internal sample collection
at Merck, using hGPR91 over-expressed CHO-K1 stable cell-line, 1
tractable hit for SAR development for what the initial goal was to
improve hGPR91 potency while maintaining the metabolic stabil-
ity (for SAR summary, see Fig. 1, I-1). We found that meta position
of the terminal phenyl ring (termed as R3) in 2b was very respon-
sive for developing a SAR. Small hydrophobic substitutions like –
Me, –Cl, –CF3 (examples 2c–e), among the several sterically and
electronically tuned substitutions tested, significantly improved
the potency often by 30 fold. Though the smallest substitution
(e.g., –F) at the para position (termed as R4) maintained the po-
tency, any larger substitution at that position led to loss of activity
(compare data of 2c & 2e vs 2u). The m-CF3 (R3) and p-F (R4) substi-
tuted compound 2e was considered as an optimized compound
(IC50: 22 nM and CL: 0.1 nmol/min/mg RLM) from the early SAR
study. Further improvement in hGPR91 potency was achieved in
2j (ꢀ3 fold compared to 2d) by addition of a methyl group at ben-
zylic carbon (termed as R5) however without any improvement of
metabolic stability. For further SAR information on biphenyl tail
part (I-1), Table S1 of Supplementary data is referred.
Next set of SAR iteration was planned on propylene linker in 2b
with an intension of arresting a possible CYP mediated metabolism
at the aliphatic chain (Fig. 1, I-2; see Supplementary data Table S2
for preliminary SAR information). Upon constraining the propylene
linker with para-phenylene moiety indeed resulted in metaboli-
cally very stable compound 3 (CL: 0.02 nmol/min/mg RLM) but
completely lost hGPR91 activity. We rationalized that the pheny-
lene–cabonyl moiety might have destroyed the required flexibility
in the linker region. Therefore, a methylene linker was introduced
between the phenylene and cabonyl group which in combination
with the most prefered tail part (R3 = CF3, R4 = F, R5 = (S)-Me)
(Fig. 1) was identified as a hit (IC50 of 0.8 lM with complete inhi-
bition of succinate induced increase in intra-cellular Ca2+ pool)
which served as a starting point for hit-to-lead identification.13
The hit 1 was originally synthesized and evaluated for Bradykinin
B1 Receptor (BK1R) inhibitory activity at Merck. The modest BK1R
activity associated with 1 (IC50: 2.5 lM) was expected to be dialed
out during the drug optimization simply by removal of carboxylic
ester present in its biphenyl part.14 To drive medicinal chemistry
around the hit 1, the structure–activity relationship (SAR) studies
were divided into three categories: (a) around the biphenyl tail,
(b) the propylene-carbonyl linker, (c) and the naphthyridin-2-yl
head (Fig. 1). SAR exercise to optimize the tail part provided us a
major break through, whereas a little variation of naphthyridine
part was found to be too intolerant to develop any SAR. Key find-
ings of initial SAR work, carried out with the linker and tail parts,
are summarized in this communication.
In one of the early investigation of SAR, we noticed that the hit 1
lost its hGPR91 activity upon changing its methyl carboxylate into
the corresponding acid 2a. However, we were inspired to see that
the unsubstituted biphenyl compound 2b was almost equipotent
to the hit 1. Additionally, the BK1R activity was completely dialed
out in 2b. Hence 2b (IC50: 1
lM; metabolic clearance in rat liver
microsome CL: 0.17 nmol/min/mg RLM) was considered as a more
HTS hit
:
μ
1
Naphthyridin-2-yl
Head part
IC50: 0.8 M (hGPR91 CHO-K1; Gq signaling)
N
N
A compound having Bradykinin B1 R (BK1R) IC50: 2.5
M
μ
Propylenecarbonyl
Linker part
O
NH
F
F
O
OMe
F
O
OH
2a
(inactive)
Biphenyl
Tail part
F
N
N
CF3
F
Me
3
(inactive)
NH
2c
(30 nM)
2e
(22 nM)
O
O
NH
I-1
I-2
R5
R3
Me
Cl
Cl
R4
2u
(300 nM)
2d
(40 nM)
R7
R6
: R3, R4, R5 = H
2b
(1000 nM)
O
NH
Me
Me
: R6 = R7 = H
4c
CF3
Cl
(7 nM)
: R6 = R7 = F
4h
(19 nM)
2j
F
(11 nM)
Figure 1. Structure of HTS hit 1 and the initial SAR summary (the tail and linker parts).

3598
D. Bhuniya et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3596–3602
resulted in potent and stable compound 4c (IC50: 7 nM; CL:
the two differently connected pyrimidines H1 and H2 (e.g. 5a vs
5b). Among the others, pyrazole analog 5d also exhibited good po-
tency and stability (IC50: 210 nM; CL: 0.18 nmol/min/mg RLM).
From this optimization study, 5g was identified as next improved
compound having demonstrated satisfactory oral bioavailability
in rats (%F: 26; vehicle: supension in 0.5% CMC), very good plasma
0.2 nmol/min/mg RLM). Substitution like –Me, –F, –CF3, –OH, –
NH2 were also attempted at
a
-C of amide functionality (R6 & R7)
to follow the effect on hGPR91 potency as well as impact on met-
abolic stability. Except for the case of -NH2 (e.g., 4f), the hGPR91
potency was more or less maintained in these analogs (e.g. 4d–e,
4g–j) in addition to retention of metabolic stability. Antagonists
4c and 4h (IC50: 19 nM, CL: 0.14 nmol/min/mg RLM) were consid-
ered as optimized compounds from that particular SAR work.
At that stage of the program, we considered few potent hGPR91
antagonists which were selective against hGPR99 for pharmacoki-
netic (PK) evaluation before embarking on pharmacodynamic (PD)
or in vivo target engagement studies. Unfortunately, lead com-
pounds from series 2 and 4 did not offer satisfactory oral bioavail-
ability (tested in Wistar rats) although some of them exhibited low
plasma clearance reflecting the metabolic stability seen in RLM.
None the less, reasonable plasma exposure was achieved upon
administration of these compounds via intra-peretonial (i.p.) route.
In vitro and PK profile of representative leads 2c, 2d, 2j and 4c are
summarized in Table 1.
concentration (Cmax: 37 lM and AUC0–24 h: 69 lM h at 30 mg/kg
p.o. dose), and low plasma clearance (4.7 mL/min/kg). In addition,
5g was found to be selective against hGPR99 and a pannel of >120
off-targets (see Table 1). It is important to note here that going
from 2e to 5g, c Log P value reduced from 4.87 to 3.56 resulting
in improvement of oral bioavailability.
In another SAR exercise, effect of bio-isosteric replacement of
amide functionality was studied in structure
2 (Fig. 2, I-4;
Table S4). Out of three isosteric heterocycles H8-10 tested, 1,3,4-
oxadiazol-2,5-ylene H8 was found to be superior over the other
two in retaining hGPR91 activity. The compound 6a (R3 = Cl) was
a potent hGPR91 antagonist (IC50: 40 nM) and selective against
hGPR99 (IC50: >30 lM). However, 6a (c Log P: 5.34) did not offer
us adequate oral bioavailability; hence it was considered for fur-
ther optimization to improve its drug-like properties by reducing
its lipophylicity. It was envisaged that the biphenyl-oxadiazole
part in 6a might be made less hydrophobic by curtailing one phe-
nyl group. Accordingly, the linker length was increased by a bond
to compensate the length of the designed molecule, and another
round of SAR iteration was carried out on substitutions in the ter-
minal ring (R3 & R4 of 7 in Table S4). Compound 7d–f (R3 = CF3, Cl,
CF3; R4 = CF3, OCF3, Cl respectively) were found to have satisfactory
hGPR91 potency (160–180 nM) and metabolic stability (CL: 0.17–
0.3 nmol/min/mg RLM). Once again, a meta-substitution (R3) was
found to improve the hGPR91 potency (7b–c vs 7d–f). Interest-
ingly, unlike in series 2, larger substitutions like Cl, CF3, OCF3 at
para-position (R4) of the terminal phenyl ring was preferred in this
series (7d–f vs 7g) presumably because of shorter length of these
compounds compared to series 2–5. In vitro and PK data of repre-
sentative leads 7e (R3 = Cl; R4 = OCF3) and 7f (R3 = CF3; R4 = Cl) are
presented in Table 1 which clearly indicate that these compounds
are selective against hGPR99, and have the desirable PK character-
istics. As for example, 7e (c Log P: 4.53) demonstrated low plasma
Needless to mention here that there was a strong need to im-
prove oral bioavailability of lead compounds from the series 2
and 4 for the purpose of their robust utility as pharmacological
tool. Therefore our next round of optimization was focused on bal-
ancing hGPR91 potency, selectivity against hGPR99, metabolic sta-
bility and oral bioavailability. The SAR efforts along this direction
are summarized in Figure 2 (I-3 and I-4). Further SAR data are pre-
sented in Supplementary data Tables S3 and S4.
We envisaged that the penultimate phenylene ring in the
biphenyl tail part of 2e could be considered as next point of SAR
(I-3) where the aromatic ring would be replaced with a heteroaro-
matic ring. As presented in Table S3 nine heteroarylene moieties
(H1–H9), tethered to the most optimimally substituted terminal
phenyl ring (R3 = CF3, R4 = F), were considered for SAR investiga-
tion. We were delighted to notice that the pyrimidin-2,5-ylene
(H1) and isoxazol-3,5-ylene (H5) analogs 5a and 5g respectively
could retain the potency with satisfactory metabolic stability
(IC50: 49, 35 nM; CL: 0.2, 0.12 nmol/min/mg RLM respectively). It
was interesting to note a 4 fold difference in potency between
C4
N
C4
N
7c
CF3
O
(>3000 nM)
N
O
N
5a
N
N
F
(49 nM)
N
N
Cl
7b
(3000 nM)
OCF3
N
C4
N
O
NH
CF3
F
N
O
C3
I-4
I-3
Cl
5b
R3
R4
O
N
(200 nM)
N
N
OCF3
7e
(180 nM)
Cl
N
N
CF3
F
6a
C4
: R3 = R4 = H
2b
(40 nM)
O
5d
(210 nM)
(1000 nM)
N
CF3
Cl
N
C4
N
: R3 = CF , R4 = F
2e
(22 nM)
3
O
O
N
7f
CF3
F
CF3
F
: R3 = Cl, R4 = H
(160 nM)
N
2d
7g
(2000 nM)
(40 nM)
5g
(35 nM)
Figure 2. SAR summary to identify leads with desirable oral PK properties.

D. Bhuniya et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3596–3602
3599
Table 1
In vitro and PK profile of selected compounds
Antagonists hGPR91a
IC50 (nM)
rGPR91a
IC50 (nM)
hGPR99a
CLb (nmol/ Rat PKc
IC50 M) min/mg)
(l
2cd
30 2 (n = 14)
7
3 (n = 4) >30
1.2
0.3
i.v. (5 mg/kg): Cmax 19 1.4
i.p. (50 mg/kg): Cmax 3.7 1.6
C2 h 3.3 0.8 M; C4 h 2.4 0.3
i.v. (5 mg/kg): Cmax 82.7 10.4
i.p. (50 mg/kg): Cmax 17.8 4.3
l
M; CL 64.7 9 mL/min/kg; Vss 0.5 0.0 L/kg; t1/2 0.1 0.01 h.
M; Tmax 1.8 0.3 h; AUC0–8 h 17.4 2.3 M h;
M; C8 h 1.1 0.3 M.
M; CL 4.4 0.6 mL/min/kg; Vss 0.2 0.05 L/kg; t1/2 1.2 0.3 h.
M; Tmax 3.3 1.2 h; AUC0–8 h 119 29 M h; C8 h 15 M.
M; CL 14.8 0.6 mL/min/kg; Vss 0.7 0.1 L/kg; t1/2 0.8 0.2 h.
M; CL 36 5.0 mL/min/kg; Vss 1.0 0.2 L/kg; t1/2 0.9 0.2 h.
M; Tmax 4.9 3 h; AUC0–24 h 4.3 3.2 M h;
M; C4 h 0.33 0.1 M; C8 h 0.28 0.2 M.
M; CL 4.7 2.1 mL/min/kg; Vss 0.36 0.1 L/kg; t1/2 1.1 0.3 h.
M; Tmax 0.0 h; AUC0–8 h 267 40 M h; C2 h 22 M;
M.
p.o. (30 mg/kg): Cmax 37 4.5
AUC0–24 h 69 12 M h; t1/2 0.9 0.1 h; %F 26.
i.v. (5 mg/kg): Cmax 9.3 0.9 M; CL 51 3 mL/min/kg; Vss 1.9 0.9 L/kg; t1/2 0.9 0.6 h.
i.v. (5 mg/kg): Cmax 56 M; CL 2.0 0.2 mL/min/kg; Vss 0.2 0.0 L/kg; t1/2 1.3 0.2 h.
p.o. (30 mg/kg): Cmax 72 2.8 M; Tmax 1.5 0.8 h; C2 h 70 0.2 M; C4 h 40.5 7.2 M;
AUC0–24 h 471.5 198 M.h; t1/2 3.4 2.0 h; %F 87.
i.v. (5 mg/kg): Cmax: 92 34 M; CL: 4.8 1.6 mL/min/kg; Vss: 0.1 0.03 L/kg; t1/2: 0.3 0.1 h.
i.p. (50 mg/kg): Cmax 21 M; Tmax 2.7 1.2 h; AUC0–8 h 304 91 M h; C2 h 19 M;
C4 h 20 M; C8 h 15 M.
l
l
l
l
l
l
l
2d
40
9
115 25
100 15
7
2
2
l
5 l
2j
11
7
1
0.6
0.42
0.2
i.v. (5 mg/kg): Cmax 19
i.v. (3 mg/kg): Cmax 9.8
4 l
4 l
4cd
2 (n = 5)
—
>30
e
i.p. (100 mg/kg): Cmax 0.4 0.1
l
l
C2 h 0.24 0.08
l
l
l
5gd
35 3 (n = 5)
135 63
>30
0.12
i.v. (5 mg/kg): Cmax 40
3 l
i.p. (50 mg/kg): Cmax 22
4
l
2
l
4 l
C4 h 19
2
lM; C8 h 19
5 l
l
M; Tmax 0.5 0.0 h; C2 h 11.3 5.7 lM; C4 h 2.6 0.7 lM;
l
6a
7e
40
1
67
—
5
5
1
0.3
0.17
l
e
180 28
>20
6 l
l
l
l
l
7f
160 0.0
435 22
>30
0.3
l
6
l
l
3 l
7
l
5 l
a
A brief description of hGPR91, rGPR91, and hGPR99 assay development and screening protocol is available as supplementary material. Details will be reported in a
separate communication. IC50 values are presented as average SD from at least two independent 9 point titration experiments. Data collections were done in triplicate.
b
Measured as intrinsic metabolic clearance of tested drug (CL, nmol/min/mg of RLM). Based on our RLM metabolic stability assay protocol (see Supplementary data), a CL
range of 0.1–0.3 was indicative of good stability; 0.4–0.6 as moderate stability.
c
Vehicle information: (i) Solution prepared using DMAC (10–20%) + CrEL (10%) + PEG400 (10%) in saline or DMAC (10%) + ethanol (10%) + PG (10%) + MQ water (70%) for
i.v. administration (5 mL/kg dose volume). (ii) Suspension prepared using CMC (0.5%) + Tween 80 (0.5%) in water for i.p. and oral dosing. PK parameters are presented as mean
value of three animals SD. Supplementary data referred for detailed PK experiments.
d
Demonstrated >100 fold selectivity against >120 off-targets (MDS PanLab).
rGPR91 data not generated for these compounds. However, the data of their close analogs indicate for no species difference between rat and human.
e
clearance (2.0 mL/min/kg), excellent bioavailability (%F: 87) and
drug exposure (Cmax: 72 lM, AUC0–24 h: 471 lM h; t1/2: 3.4 h) upon
oral administration (30 mg/kg p.o.) in Wistar rats.
GPR91
antagonist/vehicle, IP
Saline
SA1
SA2
SA3
To study GPR91 target engagement in vivo, a pharmacodynamic
(PD) assay was developed based on the initial report of He et al.1
Details of the PD assay development and pharmacological charac-
terization of GPR91 antagonists will be a subject matter of another
report. In brief, bolus administration of succinate (1 mg/kg i.v.) to
overnight fasted and anesthetized male Wistar rats led to a tran-
sient increase of mean arterial pressure (MAP). Rats exhibiting an
increasing of MAP by at least 10 mm Hg above the saline-induced
5 min
120 min
30min
120 min
Figure 3. Protocol for the pharmacodynamic (PD) assay. Saline, SA1, SA2 and SA3:
Vehicle and succinate doses (1 mg/kg i.v. bolus) followed by measurement of
MAP. The antagonistic effects are measured from percent inhibition of
independently at SA2 and SA3 time points in comparison to the
SA1 time point.
D
D
MAP
DMAP measured at
change in MAP, were considered for the PD study (DMAP >10 mm
Hg as the PD window). A schematic diagram of the PD assay proto-
col is outlined in Figure 3.
To estimate the effect of GPR91 antagonists on succinate in-
duced increase in MAP, two time points were considered at 2
and 4 h (SA2 and SA3) post-drug dose. An effect from a pre-drug
dose time point SA1 was chosen as the study control. PD data of
two representative antagonists 2c and 4c are presented here in
Figure 4. Compounds were dosed i.p. which allowed us to do drug
delivery when the animals were under anesthetized condition. In
addition, i.p. dosing offered sustained drug exposure during the
PD study. 2c (50 mg/kg, i.p.) demonstrated 71 and 67% inhibition
of succinate-induced rise in
and SA3 time points) post drug dose. Plasma drug levels were
found to be 1.8 and 2.0 M respectively. Drug concentration in kid-
DMAP measured at 2 and 4 h (SA2
l
ney was found to be 7.0
l
M measured soon after SA3 time point. 4c
Vehicle
2c
4c
15.0
15.0
25.0
20.0
10.0
15.0
10.0
-59%
-67%
-71%
-76%
*
10.0
*
5.0
5.0
*
*
5.0
0.0
0.0
0.0
SA1
SA2
SA3
SA1
SA2
SA3
SA1
SA2
SA3
Figure 4. In vivo target engagement by the GPR91 antagonists 2c and 4c in the Pharmacodynamic (PD) assay. Data are shown as mean S.E.M. ⁄Significantly different as
compared to SA1. Unpaired t-test, p <0.05. Numbers above the bars indicate% inhibition of succinate (SA) induced increase in mean arterial pressure (DMAP) at that time
point.

3600
D. Bhuniya et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3596–3602
(100 mg/kg, i.p.) led to 59 and 76% inhibition of
D
MAP at 2 and 4 h
observed kidney exposure in the range of 7–8 lM can explain
(drug exposure: 0.2 & 0.3 M in plasma; 8 M in kidney). Both the
compounds had shown rat plasma protein binding 99%. Consider-
ing their intrinsic rGPR91 potency in the range of 10–30 nM, the
l
l
the PD effect. These preliminary results strongly suggest that these
small molecule GPR91 antagonists indeed have engaged the target
under the in vivo condition.
NH 2
11
R5
T
CHO
NH 2
R5
O
O
i, ii
O
O
iii
+
(L)
OEt
(L)
OH
N
N
(L)
N
N
N
H
N
T
9
10a-d
8
1-5
T
(L)
For
Het
10a
For
10a, 10c-d
R1
(CH2)3
(CH2)4
9a
9b
R3
R4
R2
R3
R4
9c
9d
5
1-4
Het
N
N
O
R6 R7
N
N
N
N
H5
N
H1
N
H2
N
H4
H3
N
N
N
S
N
N
S
N
H8
H7
H9
H6
S
O
Scheme 1. General synthetic route to amide series of compounds 1–5. Reaction condition: (i)
(iii) EDCI, HOBt, NMM, DMF, rt, 18 h, 40–85%.
L-Proline, EtOH, reflux, 20 h. 42–55%; (ii) LiOHÁH2O, THF, H2O, rt, 16 h, 70–83%;
(L) =
(CH2)3
Ar1; R3 = Cl; R4 = H
6a:
6b:
6c:
ii
N
N
(L)
N
N
(L)
Ar1; R3 = Cl; R4 = F
iii
N
N
(L)
Ar1; R3 = CF ; R4 = H
3
NH
HN
O
O
N
N
S
H8
N
N
(L) =
(CH2)4
H9
O
13
Ar
Ar2
7a-h:
Ar
Ar
(CH2)3
(L) =
i
Ar1; R3 = Cl; R4 = H
6d:
Ar
O
O
H
2N
R3
R4
N
(L)
OH
N
N
(L) =
(L) =
R3
R4
H
Ar
12
(CH2)3
(CH2)4
10a
10b
Ar1
Ar2
2,4-dimethoxy
benzylamine
i
(CH2)3
Ar1; R3 = Cl; R4 = F
O
(L) =
H2N
N
6e:
H
Ar
12
S
O
iv
iii
v
(L)
NH
N
(L)
NH
N
N
N
(L)
N
N
N
N
N
H10
14
15
MeO
OMe
OMe
MeO
Ar
Scheme 2. General synthetic routes to amide bio-isosteric compounds 6–7. Reaction condition: (i) EDCI, HOBt, NMM, DMF, rt, 18 h, 40–85%; (ii) POCl3, ethylene dichloride,
100 °C, 3 h., 45%; (iii) Lawessons reagent, THF, reflux, 16 h, 99%; (iv) Hg(OAc)2, THF, 0 °C, 2 h, 40%; (v) TFA, DCM, 0–rt, 18 h, 68%.

D. Bhuniya et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3596–3602
3601
General and typical synthetic routes employed to access the
GPR91 antagonists are outlined in Scheme 1 and Scheme 2. 1,8-
Naphthyridine heterocycle was constructed having appropriate
2- substitution via Friedländer heterocyclization15 between 2-ami-
no-3-formylpyridine 816 and appropriately chosen methylketoest-
er 9. Subsequent ester hydrolysis provided us the carboxylic acid
10 as a common building block for synthesis of both the amide
(1-5) as well as amide bio-isosteric compounds (6-7).
tion of key compounds; additional information on animal studies)
associated with this article can be found, in the online version, at
doi:10.1016/j.bmcl.2011.04.091.
References and notes
1. He, W.; Miao, F. J.-P.; Lin, D. C.-H.; Schwandner, R. T.; Wang, Z.; Gao, J.; Chen, J.-
L.; Tian, H.; Ling, L. Nature 2004, 429, 188.
2. Gonzalez, N. S.; Communi, D.; Hannedouche, S.; Boeynaems, J.-M. Purinergic
Signalling 2004, 1, 17.
3. Regard, J. B.; Sato, I. T.; Coughlin, S. R. Cell 2008, 135, 561.
4. Hakak, Y.; Lehmann-Bruinsma, K.; Phillips, S.; Le, T.; Liaw, C.; Connolly, D. T.;
Behan, D. P. J. Leukoc. Biol. 2009, 85, 837.
The scaffold 10 was reacted with appropriately chosen amine
11 using EDCI/HOBt promoted amide coupling condition to obtain
the antagonists 1–5 (Scheme 1).17–19
The acid intermediates 10a–b, synthesized in Scheme 1, were
used for the synthesis of amide bio-isosters 6 and 7 shown in
Scheme 2. Starting from 10a–b, and reacting with acylhydrazine
12, another general building block 13 was obtained by treatment
with the amide coupling reagent. Appropriate diacylhydrazine
scaffold 13 was then converted to oxadiazole compounds 6a–c
and 7a–h by intramolecular cyclization followed by dehydration
in presence of POCl3.20 The thiadiazole analog of 6a, that is 6d,
was obtained from the intermediate 13 by treatment with Lawes-
son’s reagent. The corresponding triazole analog 6e was obtained
from the intermediate 10a in four steps: via formation of the cor-
responding N-protected amide 14 followed by thioamide 15, and
subsequently by Hg(OAc)2 induced condensation of the intermedi-
ate 15 with appropriate acylhydrazine 12, followed by deprotec-
tion of 2,4-dimethoxy benzyl group.21–23
5. Sapieha, P.; Sirinyan, M.; Hamel, D.; Zaniolo, K.; Joyal, J.-S.; Cho, J.-H.; Honoré,
J.-C.; Kermorvant-Duchemin, E.; Varma, D. R.; Tremblay, S.; Leduc, M.;
Rihakova, L.; Hardy, P.; Klein, W. H.; Mu, X.; Mamer, O.; Lachapelle, P.; Polo,
A. D.; Beauséjour, C.; Andelfinger, G.; Mitchell, G.; Sennlaub, F.; Chemtob, S.
Nature Med. 2008, 14, 1067.
6. Hebert, S. C. Nature 2004, 429, 143.
7. Sadagopan, N.; Li, W.; Roberds, S. L.; Major, T.; Preston, G. M.; Yu, Y.; Tones, M.
A. Am. J. Hypertens. 2007, 20, 1209.
8. (a) Toma, I.; Kang, J. J.; Sipos, A.; Vargas, S.; Bansal, E.; Hanner, F.; Meer, E.; Peti-
Peterdi, J. J. Clin. Invest. 2008, 118, 2526; (b) Peti-Peterdi, J.; Kang, J. J.; Toma, I.
Nephrol. Dial. Transplant. 2008, 23, 3047–3049; (c) Vargas, S. L.; Toma, I.; Kang,
J. J.; Meer, E. J.; Peti-Peterdi, J. J. Am. Soc. Nephrol. 2009, 20, 1002; (d) Peti-
Peterdi, J. Kidney Int. 2010, 78, 1214.
9. Correa, P. R. A. V.; Kruglov, E. A.; Thompson, M.; Leite, M. F.; Dranoff, J. A.;
Nathanson, M. H. J. Hepatol. 2007, 47, 262.
10. Rubic, T.; Lametschwandtner, G.; Jost, S.; Hinteregger, S.; Kund, J.; Carballido-
Perrig, N.; Schwärzler, C.; Junt, T.; Voshol, H.; Meingassner, J. G.; Mao, X.;
Werner, G.; Rot, A.; Carballido, J. M. Nature Immunol. 2008, 9, 1261.
11. Kermorvant-Duchemin, E.; Sapieha, P.; Sirinyan, M.; Beauchamp, M.; Checchin,
D.; Hardy, P.; Sennlaub, F.; Lachapelle, P.; Chemtob, S. Doc. Opthalmol. 2010,
120, 51.
In conclusion, starting from
a near micromolar active
high-throughput screening hit 1, systematic structure-activity
relationship study led to the identification of several potent and
selective hGPR91 antagonists. Two independent series were devel-
oped, namely amides 2–5, and amide-bio-isosteric analogs 6–7.
Initial hurdle of poor oral bioavailability, observed in series 2–4
and 6, was subsequently overcome in series 5 and 7 respectively
(Table 1). A new pharmacodynamic (PD) assay was set up in anes-
thetized Wistar rats based on succinate induced increase in mean
arterial pressure. Employing two representative tool compounds
2c and 4c, in the PD study, GPR91 target engagement was demon-
strated in vivo (Fig. 4). To the best of our knowledge, this is the first
report of the discovery and pharmacodynamic characterization of
GPR91 antagonists. The compounds presented here (e.g., 5g, 7e)
may serve as pharmacological tools for establishing animal
proof-of-concept of GPR91 modulation in several potential thera-
peutic indications such as: renal hypertension, diabetic nephropa-
thy, autoimmune disease, retinal angiogenesis, and diabetic
retinopathy. In addition, we hope this communication will serve
as a catalyst for further studies leading to better understanding
of the target and eventually to a new line of therapy. A complete
medicinal chemistry and pharmacological validation of GPR91 will
be disclosed elsewhere in the form of full paper.
12. We are not aware of any prior art literature of GPR91 antagonist. However,
there are patent applications claiming method of screening of drug candidates
modulating GPR91 and the method of treatment to various therapeutic
indications. Please see, (a) Li, K.; Wang, S.-W.; Hu, G.; Yao, Z. PCT
international publication no. WO 2005/010152 A2. (b) Golz, S.; Brüggemeier,
U.; Geerts, A.; Summer, H. PCT international publication no. WO 2005/050220
A1. (c) Carballido Herrera, J. M.; Lametschwandtner, G.; Werner, G.; Rot, A. PCT
international publication no. WO 2006/117193 A2. (d) Hakak, Y.; Liu, L.;
Bruinsma, K. PCT international publication no. WO 2009/011885 A1.
13. Supplementary data referred for
a brief description of GPR91 antagonist
screening assay protocol used in HTS as well as for the SAR work presented
here.
14. For literature on BK1R antagonists related to 1, please see: (a) Kuduk, S. D.;
Marco, C. N. D.; Chang, R. K.; Wood, M. R.; Kim, J. J.; Schirripa, K. M.; Murphy, K.
L.; Ransom, R. W.; Tang, C.; Torrent, M.; Ha, S.; Prueksaritanont, T.; Pettibone,
D. J.; Bock, M. G. Bioorg. Med. Chem. Lett. 2006, 16, 2791; (b) Wood, M. R.;
Kuduk, S. D.; Bock, M. G.; Chang, R. K. US Patent application publication no. US
2006/0128765 A1.; (c) Wood, M. R.; Su, D.-S.; Wai, J. M.-C. US Patent
application publication no. US 2006/0173023 A1.; (d) Kuduk, S. D.; Marco, C.
N. D.; Chang, R. K.; Wood, M. R.; Schirripa, K. M.; Kim, J. J.; Wai, J. M.-C.;
DiPardo, R. M.; Murphy, K. L.; Ransom, R. W.; Harrell, C. M.; Reiss, D. R.;
Holahan, M. A.; Cook, J.; Hess, J. F.; Sain, N.; Urban, M. O.; Tang, C.;
Prueksaritanont, T.; Pettibone, D. J.; Bock, M. G. J. Med. Chem. 2007, 50, 272.
15. (a) Meissner, R. S.; Perkins, J. J.; Duong, L. T.; Hartman, G. D.; Hoffman, W. F.;
Huff, J. R.; Ihle, N. C.; Leu, C. –T.; Nagy, R. M.; Naylor-Olsen, A.; Rodan, G. A.;
Rodan, S. B.; Whitman, D. B.; Wesolowski, G. A.; Duggan, M. E. Bioorg. Med.
Chem. Lett. 2002, 12, 25; (b) Dormer, P. G.; Eng, K. K.; Farr, R. N.; Humphrey, G.
R.; McWilliams, J. C.; Reider, P. J.; Sager, J. W.; Volante, R. P. J. Org. Chem. 2003,
68, 467.
16. (a) Turner, J. A. J. Org. Chem. 1983, 48, 3401; (b) Rivera, N. R.; Hsiao, Y.; Cowen, J.
A.; McWilliams, C.; Armstrong, J.; Yasuda, N.; Hughes, D. L. Synth. Commun.
2001, 31, 1573.
Acknowledgments
17. N-[(3’-Methylbiphenyl-4-yl)methyl]-4-([1,8]naphthyridin-2-yl)butyramide,
2c.
Mp: 140–142 °C. 1H NMR (400 MHz, CDCl3): d 2.30 (quint, J = 7 Hz, 2H), 2.39
(t, J = 7 Hz, 2H), 2.41 (s, 3H), 3.14 (t, J = 7 Hz, 2H), 4.48 (d, J = 5.6 Hz, 2H), 6.42
(br t, J = 5.6 Hz, 1H, -NH-), 7.16 (d, J = 7.2 Hz, 1H), 7.32 (t, J = 7.6 Hz, 1H), 7.34–
7.38 (aromatics, 4H), 7.42 (d, J = 8.4 Hz, 1H), 7.46 (dd, J = 8.2, 4.4 Hz, 1H), 7.53
(d, J = 8.4 Hz, 2H), 8.12 (d, J = 8.3 Hz, 1H), 8.17 (dd, J = 8.1 Hz, 2 Hz, 1H), 9.07
(dd, J = 4.4, 2 Hz, 1H). 13C NMR (100 MHz, DMSO-D6): d 21.08, 24.86, 34.86,
37.76, 41.72, 120.90, 121.62, 122.55, 123.64, 126.53 (2C), 127.16, 127.74 (2C),
127.90, 128.75, 137.26, 137.61, 137.97, 138.71, 138.83, 139.90, 153.16, 155.29,
165.47, 171.80. HPLC purity: 99.6%. MS (m/z): 396.0 [M+1] base peak. HRMS
(m/z): Calcd For [M+1]: 396.2075; Found: 396.2094.
Described work was carried out as a part of collaborative pro-
gram between Advinus Therapeutics and Merck Research Labora-
tories. We thank all the members of much larger GPR91 team,
business alliance leaders, and senior management from both the
organizations. We thank Dr. Mahesh Mone for analytical support,
and Dr. Anup Ranade for managing intellectual property. Advinus
publication no. ADV-A-012.
18. N-[(S)-1-(4’-Fluoro-3’-trifluoromethylbiphenyl-4-yl)ethyl]-2-[4-
Supplementary data
([1,8]naphthyridin-2-yl)phenyl]acetamide, 4c. Mp: 198–200 °C. 1H NMR
(400 MHz, DMSO-D6): d 1.42 (d, J = 6.9 Hz, 3H), 3.59 (s, 2H), 4.98 (quintet,
J = 7.1 Hz, 1H), 7.43 (d, J = 8.2 Hz, 2H), 7.48 (d, J = 8.2 Hz, 2H), 7.56–7.65
(aromatics, 2H), 7.69 (d, J = 8.2 Hz, 2H), 7.96 (br d, J = 6 Hz, 1H), 8.00–8.05
(aromatics, 1H), 8.27 (d, J = 8.4 Hz, 3H), 8.48 (dd, J = 8.0, 1.8 Hz, 1H), 8.56 (d,
J = 8.6 Hz, 1H), 8.70 (d, J = 7.9 Hz, 1H), 9.09 (dd, J = 4.0, 1.8 Hz, 1H). 13C NMR
Supplementary data (hGPR91 antagonistic screening assay pro-
tocol; series wise hGPR91 IC50 along with metabolic stability data
of representative compounds (Table S1–S4); spectral characteriza-

3602
D. Bhuniya et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3596–3602
(100 MHz, DMSO-D6): d 22.52, 42.16, 47.81, 116.60–117.40 (m, –CF3), 117.74
21. For 1,2,4-triazole synthesis, see: Boeglin, D.; Cantel, S.; Heitz, A.; Martinez, J.;
Fehrentz, J.-A. Org. Lett. 2003, 5, 4465.
(J = 20 Hz), 119.49, 121.28, 121.60, 122.06, 123.99, 125.05 (d, J = 3.9 Hz),
126.67 (2C), 126.95 (2C), 127.43 (2C), 129.62 (2C), 133.28 (d, J = 8.5 Hz),
136.09, 136.41, 137.05 (d, J = 3 Hz), 137.28, 138.71 (J = 7.4 Hz), 144.79, 153.93,
155.49, 158.31 (d, J = 252 Hz), 158.91, 168.96. HPLC purity: 99.1%. Chiral
purity: 100% (chiral HPLC). MS (m/z): 530.0 [M+1] base peak. HRMS (m/z):
22. 2-{3-[5-(3’-Chlorobiphenyl-4-yl)-[1,3,4]oxadiazol-2-yl]propyl}[1,8]naphthyridine
6a. Mp: 162–164 °C. 1H NMR (DMSO-D6): d 2.33 (quint, J = 7.3 Hz, 2H), 3.06 (t,
J = 7.4 Hz, 2H), 3.12 (t, J = 7.3 Hz, 2H), 7.47 (dt, J = 7.7, 1.6 Hz, 1H), 7.51 (t,
J = 7.7 Hz, 1H), 7.54 (d, J = 8.5 Hz, 1H), 7.55 (d, J = 8.6 Hz, 1H), 7.71 (dt, J = 7.5,
1.6 Hz, 1H), 7.80 (t, J = 1.6 Hz, 1H), 7.89 (d, J = 8.5 Hz, 2H), 7.99 (d, J = 8.5 Hz,
2H), 8.34 (d, J = 8.3 Hz, 1H), 8.37 (dd, J = 8.2, 2.0 Hz, 1H), 8.99 (dd, J = 4.2, 2.0 Hz,
1H). 13C NMR (DMSO-D6): d 24.36, 25.05, 37.17, 120.97, 121.72, 122.70, 123.01,
125.59, 126.61, 127.02 (2C), 127.75 (2C), 128.14, 130.97, 133.94, 137.29,
137.72, 140.95, 141.47, 153.25, 155.29, 163.61, 164.84, 166.73. MS (m/z): 427.0
[M (35Cl) +1] base peak. HPLC purity: 99%.
23. 2-{4-[5-(3-Chloro-4-trifluoromethoxyphenyl)-[1,3,4]oxadiazol-2-yl]butyl}-
[1,8]naphthyridine, 7e. Mp: 80–82 °C. 1H NMR (400 MHz, CDCl3): d 1.98 (quint,
J = 7.5 Hz, 2H), 2.08 (quint, J = 7.5 Hz, 2H), 3.00 (t, J = 7.4 Hz, 2H), 3.12 (t,
J = 7.4 Hz, 2H), 7.38 (d, J = 8.2 Hz, 1H), 7.42-7.48 (aromatics, 2H), 7.96 (dd,
J = 8.2, 1.9 Hz, 1H), 8.10 (d, J = 8.4 Hz, 1H), 8.13 (d, J = 1.8 Hz, 1H), 8.15 (dd,
J = 8.0, 1.7 Hz, 1H), 9.07 (dd, J = 4.0, 1.7 Hz, 1H). 13C NMR (100 MHz, DMSO-D6):
d 33.96, 34.89, 37.30, 47.14, 129.42 (J = 258 Hz, –CF3), 130.39, 131.13, 132.05,
133.34, 133.68, 136.50, 136.69, 138.19, 146.77, 147.13, 155.48, 162.68, 164.80,
171.46, 175.04, 176.94. HPLC purity: 99.27%. MS (m/z): 448.9 [M(35Cl) +1] base
peak.
Calcd for [M+1]: 530.1855; Found: 530.1854. [
CHCl3).
a
]
D
= +127.15 (25 °C, c 0.3,
19. N-[{3-(3-trifluoromethyl-4-fluorophenyl)isoxazol-5-yl}methyl]-4-
([1,8]naphthyridin-2-yl)butyramide, 5g. Mp: 168–170 °C. 1H NMR (400 MHz,
DMSO-D6): d 2.04 (quintet, J = 7.4 Hz, 2H), 2.23 (t, J = 7.4 Hz, 2 H), 2.95 (t,
J = 7.5 Hz, 2H), 4.43 (d, J = 5.7 Hz, 2H), 7.11 (s, 1H), 7.52 (d, J = 8.4 Hz, 1H), 7.54
(dd, J = 8.1, 4.4 Hz, 1H), 7.64 (t, J = 9.5 Hz, 1H), 8.19 (d, J = 7.0 Hz, 1H), 8.22–8.28
(m, 1H), 8.34 (d, J = 8.4 Hz, 1H), 8.39 (dd, J = 8.1, 1.9 Hz, 1H), 8.56 (br t,
J = 5.7 Hz, –NH–), 8.99 (dd, J = 4.4, 1.9 Hz, 1H). 13C NMR (100 MHz, DMSO-D6):
d 24.61, 34.47, 34.66, 37.61, 100.23, 117.0–118.0 (m, CF3), 118.24 (J = 21 Hz),
120.96, 121.72, 122.57, 123.80, 125.42-125.48 (m), 125.80 (d, J = 3.1 Hz),
133.54 (d, J = 9.3 Hz), 137.34, 137.72, 153.23, 155.32, 159.7 (J = 256 Hz),
160.07, 165.47, 172.12, 172.21. HPLC purity: 99.33%. MS (m/z): 459.3 [M+1]
base peak. HRMS (m/z): Calcd for [M+1]: 459.1444; Found: 459.1459.
20. For oxadiazole synthesis, see: (a) Cao, S.; Qian, X.; Song, G.; Huang, Q. J. Fluorine
Chem. 2002, 117, 63; (b) Chang, E.-M.; Lin, C.-J.; Wong, F. F.; Yeh, M.-Y.
Heterocycles 2006, 68, 733.