Discovery of pyrazole carboxylic acids as potent inhibitors of rat long chain l-2-hydroxy acid oxidase

Article abstract of DOI:10.1016/j.bmcl.2012.05.020

Long chain l-2-hydroxy acid oxidase 2 (Hao2) is a peroxisomal enzyme expressed in the kidney and the liver. Hao2 was identified as a candidate gene for blood pressure (BP) quantitative trait locus (QTL) but the identity of its physiological substrate and its role in vivo remains largely unknown. To define a pharmacological role of this gene product, we report the development of selective inhibitors of Hao2. We identified pyrazole carboxylic acid hits 1 and 2 from screening of a compound library. Lead optimization of these hits led to the discovery of 15-XV and 15-XXXII as potent and selective inhibitors of rat Hao2. This report details the structure activity relationship of the pyrazole carboxylic acids as specific inhibitors of Hao2.

Full text of DOI:10.1016/j.bmcl.2012.05.020

Bioorganic & Medicinal Chemistry Letters 22 (2012) 4341–4347
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of pyrazole carboxylic acids as potent inhibitors of rat long chain
L-2-hydroxy acid oxidase
Dinesh A. Barawkar a, , Anish Bandyopadhyay a, Anil Deshpande a, Summon Koul a, Sachin Kandalkar a,
⇑
Pradeep Patil a, Goraksha Khose a, Samir Vyas a, Mahesh Mone a, Shubhangi Bhosale a, Umesh Singh a,
Siddhartha De a, Ashwin Meru a, Jayasagar Gundu a, Anita Chugh a, Venkata P. Palle a,
Kasim A. Mookhtiar a, Joseph P. Vacca b, Prasun K. Chakravarty b, Ravi P. Nargund b, Samuel D. Wright b,
Sophie Roy b, Michael P. Graziano b, Doris Cully b, Tian-Quan Cai b, Sheo B. Singh 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, 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:
Long chain L-2-hydroxy acid oxidase 2 (Hao2) is a peroxisomal enzyme expressed in the kidney and the
Received 8 April 2012
Revised 28 April 2012
Accepted 2 May 2012
Available online 11 May 2012
liver. Hao2 was identified as a candidate gene for blood pressure (BP) quantitative trait locus (QTL) but
the identity of its physiological substrate and its role in vivo remains largely unknown. To define a phar-
macological role of this gene product, we report the development of selective inhibitors of Hao2. We
identified pyrazole carboxylic acid hits 1 and 2 from screening of a compound library. Lead optimization
of these hits led to the discovery of 15-XV and 15-XXXII as potent and selective inhibitors of rat Hao2.
This report details the structure activity relationship of the pyrazole carboxylic acids as specific inhibitors
of Hao2.
Keywords:
Hao2
Hypertension
Pyrazole carboxylic acid
Inhibitor
Ó 2012 Elsevier Ltd. All rights reserved.
Hypertension is a major risk factor for cardiovascular diseases,
affecting more than 25% of adult population worldwide. Its preva-
lence is expected to increase by 60% to >1.5 billion people worldwide
by 2025.1,2 While there are a number of antihypertensive agents on
the market including diuretics, beta blockers, calcium channel
blockers (CCBs), renin inhibitors, angiotensin converting enzyme
(ACE) inhibitors and angiotensin II receptor blockers (ARBs), there
is a continuousneed for innovativeantihypertensive therapyto treat
resistant hypertension, to improve blood-pressure control, and to
achieve further risk reduction beyond blood-pressure control.3,4
Hao2 is predominantly expressed in the kidney and the liver. A
physiological substrate of Hao2 remains unknown. Hao2 was first
described as L-amino acid oxidase. Subsequently L-2-hydroxy acids
was shown to be more active and the highest activity was reported
for 2-hydroxypalmitic acid. A homologous enzyme responsible for
the oxidation of short chain L-a-hydroxy acid is known as L-a-hy-
droxy acid oxidase (Hao1). Hao1 is expressed primarily in the liver
and the pancreas, and displays the highest activity for glycolic acid.
Human Hao2 contains 351 amino acids and Hao1 contains 370
amino acids. Human Hao2 shares ꢀ50% sequence similarity with
human Hao1 and ꢀ70% sequence similarity with rodent Hao2.
Long chain L
-2-hydroxy acid oxidase (Hao2)5 has been identi-
fied as a candidate gene for BP quantitative trait loci (QTL) in rats.6
Genome-wide linkage analysis in humans locates BP QTL in a de-
fined region containing Hao2 (located in Ch. 1 at 119.6 cM).7 These
findings support a potential role of Hao2 in BP regulation and as a
target for treatment of hypertension.
L-2-Hydroxy acid oxidases are flavin mononucleotide
(FMN)-dependent peroxisomal enzymes, which are members of
the flavoenzyme family that are responsible for the oxidation of a
HO
HO
N
N
N
H
N
H
number of L-2-hydroxy acids to ketoacids at the expense of molec-
O
O
ular oxygen, resulting in the formation of hydrogen peroxide.5
1
2
rHao2 IC50: 0.91 uM
rHao2 IC50: 4.27 uM
⇑
Corresponding author. Tel.: +91 20 6653 9630; fax: +91 20 6653 9620.
E-mail address: dinesh.barawkar@advinus.com (D.A. Barawkar).
Figure 1. Structures and IC50’s of screening hits 1 and 2.
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.05.020

4342
D. A. Barawkar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4341–4347
Ar
Ar
I
O
a
b, c
O
O
O
O
d
e, f
N
N
N
O
N
H
N
N
OEt
N
EtO
Ph
Ph
EtO
EtO
N
H
HO
O
3
4
5
6
7
Scheme 1. Reagents and conditions: (a) NH2NH2ꢁHCl, EtOH–H2O, 60 °C, 1 h; (b) I2, CAN, CH3CN, rt, 4 h; (c) NaH, PhCH2Br, THF 0 °C to rt, 6 h; (d) Ar-CH@CH2, Pd(OAc)2,
P(o-tolyl)3, TEA, 90 °C, 15 h; (e) Pd(OH)2, HCO2NH4, HCO2H, 60 °C, 6 h; (f) NaOH, THF–EtOH–H2O (6:3:1), 50 °C, 12 h.
Ar
Br
Ar
I
b
c, d
O
a
N
O
O
N
N
N
O
Ph
N
EtO
N
N
N
H
Ph
Ph
EtO
EtO
HO
5
8
9
10
Scheme 2. Reagents and conditions: (a) 3 or 4-Bromophenylboronic acid, Pd(PPh3)4, K2CO3, DMF, 90 °C, 12 h; (b) ArB(OH)2, Pd(PPh3)4, K2CO3, DMF, 90 °C, 12 h; (c) Pd(OH)2,
HCO2NH4, HCO2H, 60 °C, 6 h; (d) NaOH, THF–EtOH–H2O (6:3:1), 50 °C, 12 h.
Pyrazole carboxylic acids 1 and 2 (Fig. 1) were identified as
initial hits from a focused enzyme-based screen. The molecular
nature of these compounds made them an attractive starting point
to further exploration of structure-activity relationship (SAR).
Systematic lead optimization of these hits led to identification of
potent and selective novel inhibitors of Hao2 which is described
in this report.
methyl ketone derivatives 16. Oxime formation of a-ketone of
the pyruvate ester was followed by selective C-alkylation with
3-(4-fluorophenyl)benzyl bromide afforded intermediates 18,
which on heating with hydrazine hydrochloride afforded pyrazole
carboxylic acid ethyl ester derivatives 19. Saponification of the
esters 19 gave the final compounds 20 (Table 2) in overall good
yields.
Initial SAR efforts were focused on the flexible linker between
the pyrazole and hydrophobic biphenyl/phenyl moieties of com-
pounds 1 and 2 (Fig. 1). A common synthetic approach was adopted
to construct inhibitors with a two-carbon atom linker (Scheme 1)
and without any linker (Scheme 2). 4-Iodopyrazole was used as a
common intermediate. As highlighted in Scheme 1, the iodo-inter-
mediate 5 was prepared in three steps from commercially available
ethyl 2,4-dioxovalerate 3. Cyclization of 3 with hydrazine hydro-
chloride afforded 3-methylpyrazole-5-carboxylic acid ethyl ester
4. Iodination8 of 4 at C4 followed by N-benzylation provided inter-
mediate 5. Heck coupling between 5 and a styrene derivative gave
intermediate 6. Debenzylation and reduction of the double bond
followed by hydrolysis provided two-carbon linker target mole-
cules 7 (Scheme 1, Table 1).
In order to install an aromatic ring directly to the pyrazole ring,
intermediate 5 was subjected to two consecutive Suzuki reactions
(Scheme 2). This afforded bromophenyl derivatives represented by
intermediate 8 followed by biphenyl pyrazole carboxylic acid ethyl
ester derivatives 9, which on debenzylation and hydrolysis fur-
nished target molecules without any linker 10 (Table 1).
To understand the importance of the carboxylic acid on the
pyrazole ring for Hao2 inhibitory activity, compound 15-XV was
converted into amide derivatives (Table 3). Standard EDCI-HOBt-
mediated amide coupling between 15-XV and methyl amine affor-
ded amide derivative 21. 1,10-Carbonyldiimidazole (CDI) activation
followed by reaction with methanesulfonamide produced 22.
Standard EDCI-HOBt coupling of 5-aminotetrazole and 15-XV gave
23 (Scheme 5).
The unsubstituted amide derivative 24 of compound 15-XV was
synthesized in good yield from the corresponding ethyl ester
14-XV by heating with ethanolic NH3 in a sealed tube at 100 °C
(Scheme 6). To understand a requirement of the pyrazole NH for
rat Hao2 activity, both N-1 and N-2 methyl regioisomers were syn-
thesized. Methylation of 14-XV with sodium hydride and methyl
iodide followed by ester hydrolysis afforded the N2-methyl deriv-
ative 25 (Scheme 6). The structure of compound 25 was confirmed
by nuclear Overhauser effect (NOE) experiment. Methyl protons at
the C3 of the pyrazole ring showed a NOE with the N-methyl pro-
tons of the pyrazole ring, suggesting that methylation occurred on
N2 of the pyrazole ring.
A series of compounds containing one carbon atom linkers
( 15-I–15-XXXVI, Table 1) were synthesized from commercially
available ethyl 2,4-dioxopentanoate 11 (Scheme 3). Direct alkyl-
ation of 11 with selected benzyl bromides gave exclusively O-alkyl-
ated products. However, selective C-alkylation was achieved when
Cyclization of 13-XV with methyl-hydrazine followed by ester
hydrolysis afforded the N1-methyl derivative 26 (Scheme 7). The
structure of 26 was also confirmed by NOE experiment. Methyl
protons at C3 of pyrazole ring showed no nuclear Overhauser effect
with the methyl protons on the nitrogen of the pyrazole ring, sug-
gesting that the methyl group was on the N1. Although negative
data is not a proof, this result combining with the observation of
a NOE between the two methyl groups in 25 confirms the alterna-
tive methyl structure of 26.
the a-ketone of the ethyl 2,4-dioxopentanoate was converted to an
oxime 12, which upon treatment with a series of benzyl bromides
afforded C-alkylated products 13. Compounds 13 on heating with
hydrazine hydrochloride afforded pyrazole carboxylic acid ethyl
ester derivatives 14, which upon saponification provided inhibitors
15 (Table 1) with one carbon linkers in overall good yields.
Next, we examined substitutions at the C3 position of the
pyrazole moiety. Synthesis of inhibitors with substitution at C3
containing one-carbon linkers started from the corresponding
pyruvate ester derivatives 17 (Scheme 4), which in turn were syn-
thesized by condensation of diethyl oxalate with a corresponding
Hao2 inhibitors were evaluated for their in vitro enzyme inhibi-
tion activity against rat and human Hao2 and Hao1.9 A number of
compounds inhibited rat Hao2 activity with IC50 values below
0.5
showed much weaker activity against human Hao2 with
IC50 > 10 M, indicating a strong species difference of these inhib-
itors. It is unclear why such difference exists. Without having
lM (Tables 1 and 2). All compounds except 10-III (ꢀ0.9
lM)
l

D. A. Barawkar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4341–4347
4343
Table 1
SAR of two carbon and no-carbon linker analogs of the pyrazole carboxylic acid for rat Hao2 enzyme
R1
X
O
N
N
H
HO
a
a
a
a
Entry
X
R1
rHao2 IC50
M)
hHao2 IC50
M)
Entry
X
R1
rHao2 IC50
M)
hHao2 IC50
M)
(
l
(
l
(
l
(l
F
F
7-I
–CH2CH2–
–CH2CH2–
31,81
0.9
5, 16
5, 39
15-XVI
–CH2–
–CH2–
21, 73
0.5
11, 56
20, 50
F
F
7-II
15-XVII
F
F
F
7-III
–CH2CH2–
0.9
9, 45
15-XVIII
–CH2–
0.9
5, 22
F
CF3
F
F
10-I
No linker
No linker
No linker
–CH2–
1, 35
13, 65
35, 79
0.7
3, 13
19, 55
0.95
15-XIX
15-XX
–CH2–
–CH2–
–CH2–
–CH2–
0.4
1.1
0.3
2.0
8, 30
5, 14
6, 34
11, 29
CF3
OCF3
10-II
10-III
15-I
15-XXI
15-XXII
OCF
7, 14
3
CN
15-II
15-III
15-IV
–CH2–
–CH2–
–CH2–
0.2
1, 13
9, 12
5, 23
15-XXIII
15-XXIV
15-XXV
–CH2–
–CH2–
–CH2–
0.3
0.2
1.1
13, 46
18, 60
3, 15
F
CONH2
CO2H
CF3
35, 71
0.3
CF
3
CO H
2
15-V
–CH2–
–CH2–
1.3
1.3
13, 38
7, 34
15-XXVI
15-XXVII
–CH2–
–CH2–
0.4
0.4
2, 21
CF
3
OCF
3
15-VI
13, 45
N
N
15-VII
15-VIII
15-IX
–CH2–
–CH2–
–CH2–
0.7
2, 26
0, 6
15-XXVIII
15-XXIX
15-XXX
–CH2–
–CH2–
–CH2–
0.3
0.5
0.6
13, 49
10, 43
12, 41
OCF
3
CN
O
33, 85
0.5
O
O
6, 19
CN
CONH
2
15-X
–CH2–
1.1
0, 8
15-XXXI
–CH2–
10, 69
9, 49
S
15-XI
–CH2–
–CH2–
–CH2–
0.9
8, 8
15-XXXII
15-XXXIII
15-XXXIV
–CH2–
–CH2–
–CH2–
0.3
8, 29
CONH
2
CO H
2
15-XII
15-XIII
40, 69
7, 60
3, 22
3, 0
47, 87
0.2
26, 66
22, 59
O
CO H
2
F
15-XIV
15-XV
–CH2–
–CH2–
0.4
0.3
11, 25
16, 55
15-XXXV
15-XXXVI
–CH2–
–CH2–
1.1
1.7
8, 12
2, 18
F
N
a
% Inhibition at 1 and 10 lM when more than one number in a column.
precise structural information of the human enzyme and co-crystal
structure information of these compounds to both enzymes, one is
left only to speculate that they bind differently to these two en-
zymes. Further, these inhibitors were more than 150 fold selective
for rat Hao2 over rat Hao1. In order to assess the requirement of
structural flexibility for the potency, rigid linkers along with vari-
ety of orientations of the biphenyl moieties (ortho-, meta-, para-)
were investigated. A systematic evaluation of analogs with varying
chain lengths indicated that one-carbon atom linker between pyr-
azole ring and aromatic moieties was the most favorable linker

4344
D. A. Barawkar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4341–4347
R
R
O
O
O
O
O
O
a
b
O
R
N
c
d
O
N
OEt
O
O
OEt
OEt
N
N
N
H
N
H
O
HO
OEt
11
12
13
14
15
Scheme 3. Reagents and conditions: (a) MeONH2ꢁHCl, 4 Å MS, DMF, rt, 12 h; (b) RCH2Br, K2CO3, DMF, rt, 16 h; (c) NH2NH2ꢁHCl, EtOH–H2O, 60 °C, 3–12 h; (d) NaOH, THF–
EtOH–H2O (6:3:1), 50 °C, 12 h.
F
F
O
O
O
N
O
O
a
b, c
d
e
O
F
OEt
OEt
R
R
R
R
R
O
O
O
N
N
N
H
19
N
H
HO
OEt
16
17
18
20
Scheme 4. Reagents and conditions: (a) Diethyl oxalate, NaHMDS, THF, ꢂ78–0 °C; (b) MeONH2ꢁHCl, 4 Å MS, DMF, rt, 12 h; (c) 3-(4-fluorophenyl)benzyl bromide, K2CO3, DMF,
rt, 16 h; (d) NH2NH2ꢁHCl, EtOH–H2O, 60 °C, 3–12 h; (e) NaOH, THF–EtOH–H2O (6:3:1), 50 °C, 12 h.
Table 2
Pyrazole analogs with modification at C3
F
R1
O
HO
N
N
H
a
a
Entry
R1
rHao2 IC50
0.3
(
lM)
hHao2 IC50
(lM)
Entry
R1
rHao2 IC50
1.8
(l
M)
hHao2 IC50
1, 14
(lM)
CH3
CH3
CH3
CH3
20-I
5, 36
20-IV
20-V
–Ph
20-II
0.5
1.2
2, 11
1, 17
–CH2Ph
0.7
0.8
10, 8
4, 16
O
20-III
20-VI
–CH2CH2Ph
a
% Inhibition at 1 and 10 lM when more than one number in a column.
Table 3
Analogs addressing the role of pyrazole carboxylic acid and NH
a
a
Entry
rHao2 IC50
(lM)
hHao2 IC50
(l
M)
Entry
rHao2 IC50
(l
M)
hHao2 IC50 (lM)
21
22
23
9, 53a
0, 3a
0, 9a
7, 19
8, 8
0, 0
24
25
26
1.4
1, 14
2, 4
5, 3
12, 57a
4, 10a
a
% Inhibition at 1 and 10 lM when more than one number in a column.
(Table 1). A one carbon linker compound 15-I was the most active
against rat Hao2 (IC50: 0.7 M) compared to a two-carbon linker
compound (7-I, 31% inhibition at 1 M), whereas the correspond-
ing compound with no carbon linker 10-I was essentially inactive
in the rat Hao2 assay (35% inhibition at 10 M). Once we identified
that one-carbon linker compound 15-I was the most active, many
analogs of this compound were prepared. Substitutions around the
ring (ortho-, meta-, para-) of the phenyl ring led the synthesis of
compounds that represented electronically diverse structures.
We found that compounds with meta substitutions on the phenyl
ring with respect to linker were favored compared to ortho and
(15-I, IC50: 0.7 lM). Similar SAR was observed for compounds
l
having trifluoromethyl (15-III, 15-IV and 15-V), trifluoromethoxy
(15-VI and 15-VII), cyano (15-VIII and 15-IX) and unsubstituted
amide (15-X and 15-XI) substitutions on the phenyl ring. Com-
pounds having a charged carboxylic acid at both para and meta
positions (15-XII and 15-XIII) of the phenyl ring were significantly
l
l
less active (40 and 7% inhibition at 1 lM, respectively).
We further investigated effect of steric factors around biphenyl
group with varying biphenyl orientations at R1 in the linker (Table
1). Both meta (7-III, IC50: 0.9
biphenyl substitutions in two carbon linker series showed similar
activities. The meta biphenyl substitution (15-XV, IC50: 0.3 M)
in the one carbon linker series showed essentially similar activity
lM) and para (7-II, IC50: 0.9 lM)
para substitutions. Thus, 3-fluorophenyl compound 15-II (IC50
:
l
0.2 M) was more active than the corresponding 4-fluoro analog
l

D. A. Barawkar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4341–4347
4345
F
F
F
a
b
O
O
O
N
N
c
N
N
H
N
H
N
H
O
N
N
HO
S
O
H
H
21
15-XV
22
F
O
N
N
N
H
N
N
N
H
NH
23
Scheme 5. Reagents and conditions: (a) MeNH2, EDCI, HOBt, NMM, DCM, 0 °C to rt, 12 h; (b) CDI, THF, rt to 50 °C, 30 min then methanesulfonamide, DBU, rt to 50 °C, 2 h; (c)
5-aminotetrazole, EDCI, HOBt, NMM, DCM, 0 °C to rt, 12 h.
F
F
F
a
b, c
O
O
O
N
N
N
CH3
N
H
N
H
N
EtO
H2N
HO
24
25
14-XV
Scheme 6. Reagents and conditions: (a) Ethanolic NH3, sealed tube, 100 °C, 12 h; (b) dimethyl sulfate, NaH, THF, 0 °C to rt, 2 h; (c) NaOH, THF–EtOH–H2O (6:3:1), 50 °C, 12 h.
phenyl ring was synthesized. It was noted that meta and para sub-
stitutions on the distal phenyl ring were preferred over ortho sub-
F
O
O
N
stitutions for rat Hao2 activity. For example, compounds 15-XV
(IC50: 0.3 M) and 15-XVII (IC50: 0.5 M) with para and meta fluoro
substitutions on the distal phenyl ring were more active than the
corresponding ortho fluoro substituted compound 15-XVIII (IC50
0.9 M). 4-Trifluoromethoxy (15-XXI), 4-cyano (15-XXIII) and
4-carbamoyl (15-XXIV) on the distal phenyl ring retained activity
in rat Hao2 assay with IC50 of 0.3, 0.3 and 0.2 M, respectively,
a, b
F
OEt
l
l
H3C
O
O
N
:
N
HO
CH3
l
26
13-XV
l
however a 5- to 6-fold loss of activity was observed when a nega-
tively charged 4-carboxylic acid was placed on the distal phenyl
Scheme 7. Reagents and conditions: (a) CH3NHNH2, EtOH, 55 °C, 2 h; (b) NaOH,
THF–EtOH–H2O (6:3:1), 50 °C, 12 h.
(15-XXV, IC50: 1.1
carboxylic acid (–CH2CO2H) on the distal phenyl ring showed an
IC50 value of 0.4 M, suggesting that negatively charged function-
ality was somewhat unfavorable when directly attached to the
phenyl ring. Replacement of the distal phenyl with a heteroaryl
ring like 3- and 4-pyridine (15-XXVII and 15-XXVIII) also retained
lM). Compound 15-XXVI with homologated
compared to its para analog (15-XIV, IC50: 0.4
biphenyl orientation in one-carbon linker series showed significant
weaker activity (15-XVI, 21% inhibition at 1 M). Both meta and
l
M). The ortho
l
l
para biphenyl orientations without any carbon linkers (Table 1)
showed a significant loss of activity in rat Hao2 (10-II and 10-III;
rat Hao2 activity, with IC50 of 0.4 and 0.3
SAR was observed for the aryl ether at R1. Both para and meta
substituted biphenyl ethers at R1 (15-XXIX, IC50: 0.5
M and
15-XXX, IC50: 0.6 M) showed more potent activity than ortho
biphenyl ether (15-XXXI, 69% inhibition at 10 M) (Table 1).
lM, respectively. Similar
13 and 35% inhibition at 1 lM, respectively). Thus a meta biphenyl
moiety at R1 in one-carbon linker series was favored for inhibition
of rat Hao2.
l
l
In the next level of optimization, the SAR efforts focused on sub-
stitutions of the distal aryl ring (Table 1). The meta biaryl orienta-
tion was preferred over para biaryl for rat Hao2 activity. For
l
Next we explored fused aromatic systems as a replacement for
the meta-biarylmoietyat R1 (Table 1). 2-Benzothiophene (15-XXXII,
example, compounds 15-XIX (IC50: 0.4
0.3 M) were more active than the corresponding para biaryl ana-
logs 15-XX (IC50: 1.1 M) and 15-XXII (IC50: 2.0 M). A series of
meta biaryl compounds with various substitutions on the distal
l
M) and 15-XXI (IC50
:
IC50: 0.3
rior to 2-benzofuran (15-XXXIII, 47% inhibition at 1
2-naphthyl (15-XXXV, IC50: 1.1 M), respectively. 2-Quinoline at
R1 (15-XXXVI) was less active against rat Hao2 (IC50: 1.7
M).
l
M) and 1-naphthyl (15-XXXIV, IC50: 0.2
l
M) were supe-
l
l
M) and
l
l
l
l

4346
D. A. Barawkar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4341–4347
A
B
Figure 2. (A) A close-up view of docked compound 15-XV (yellow) in the active site of the enzyme. The flavin mononucleotide (FMN) ring is shown in magenta, (B) a close-up
view of docked compound 15-XXXII (bright orange) in the active site of the enzyme. The FMN ring is shown in light pink. The putative hydrogen bonds are shown as dotted
lines. Some of the important protein residues are labeled.
We further focused our SAR at the C3 position of the pyrazole
ring. Replacement of the C3 methyl group on the pyrazole ring
with various groups failed to improve activity against rat Hao2
(Table 2). Small hydrophobic groups like ethyl (20-I) and isopropyl
(20-II) at C3 of the pyrazole ring retained rat Hao2 activity similar
to the corresponding C3 methyl (15-XV). Ethoxy group (20-III) at
C3 lost approximately 4- to 5-fold activity and showed an IC50 of
1.2 lM. A bulky group at the C3 position of the pyrazole ring lost
Hao2 activity. Thus, replacement of the C3 methyl of the pyrazole
ring by phenyl group lost approximately 6- to 7-fold activity
Figure 2. A similar interaction of the carboxylic acid moiety with
these residues has also been observed in crystal structures of flavo-
cytocrome B2 (PDB ID: 1FCB)10b and glycolate oxidase (PDB ID:
1AL7)10c which are homologous to Hao2. The NH of the pyrazole
moiety forms putative hydrogen bonds with catalytic H247 and
nearby Y129 residues. The biphenyl moiety is located in the hydro-
phobic pocket and is surrounded by residues such as F79, L185,
L172, L174, E188, F23 and L161. It is interesting to note that the
biphenyl moiety docks in the pocket where the mobile loop (L4) is
situated. Since this loop is disordered and some of the residues are
missing in the crystal structure model, a preferred mode of the bind-
ing for the aromatic moiety in the active site cannot be precisely
captured by the modeling alone. However, the present model is able
to explain some of the SAR observed for this series of compounds.
In order to validate the role of carboxylic acid salt bridge inter-
actions at the active site, we synthesized compounds in which the
carboxylic acid was replaced by an amide and a sulfonamide
(21–24, Table 3). All these modifications will eliminate salt bridge
interactions with R250 and/or R164 leading to either complete loss
of activity (e.g., 21–23) or significant loss of activity (e.g., unsubsti-
(20-IV, IC50: 1.8
azole and the C3 phenyl substituent, such as compounds 20-V
(IC50: 0.7 M) and 20-VI (IC50: 0.8 M), faired better for their
inhibitory activities compared to compound 20-IV (IC50: 1.8 M).
lM). Introduction of a carbon spacer between pyr-
l
l
l
A crystal structure of rat Hao2 was reported (PDB ID: 1TB3).10a
Compounds 15-XV and 15-XXXII were docked into the rat Hao2
enzyme pocket to understand a mode of binding. The binding mod-
el suggested that a one-carbon linker between the pyrazole and
biphenyl moieties is sufficient to fill the available space and places
the biphenyl moiety in the appropriate position. A linker with
more than one carbon increases the conformational entropy of
these compounds, thereby decreasing their activity. Deletion of
the carbon linker increases the rigidity of the molecules resulting
in an unfavorable fitting of the biphenyl moiety in the active site
pocket. Hence, compounds without linker showed substantially
lower activity.
tuted amide 24, IC50 value 1.4 lM). Similarly, the requirement of
NH of pyrazole is evident from the complete loss of activity of both
N1-methyl and N2-methyl pyrazole derivatives (25 and 26, Table
3), which eliminates the formation of putative hydrogen bonds
with catalytic H247 and nearby Y129 residues.
In summary, we have systematically carried out SAR studies on
pyrazole-3-carboxylic acid derivative to identify two potent and
selective rat Hao2 inhibitors (15-XV and 15-XXXII). Both com-
pounds displayed a good rat pharmacokinetic profile with %F (38
and 54) and half life (1.7, 3.7 h), respectively and successfully de-
ployed to establish animal proof of concept towards a novel ther-
apy for hypertension as reported earlier.9 This is the first report
that outlines the rationale for design, synthesis and SAR of specific
rat Hao2 inhibitors. These inhibitors showed species specific activ-
ities and were largely inactive against human Hao2, indicating a
need for further optimization of this series or identification of
alternate lead series with dual activity.
A SAR from our study suggests that the 1,3 linkage on the phe-
nyl proximal to pyrazole is more favorable than that of the 1,4 link-
age. Computer modeling suggested that a fused aromatic system
(15-XXXII) in place of meta biphenyl (15-XV) at R1 (Table 1) is eas-
ily accommodated in the hydrophobic pocket without disrupting
the interactions of C-5 carboxylate and NH of pyrazole moieties.
This perhaps is the best explanation for a similar activity observed
for these two structurally different compounds (biphenyl and
fused aromatics).
The carboxylic acid moiety binds in the active site by forming salt
bridge interaction with basic residues R250 and R164 as shown in

D. A. Barawkar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4341–4347
5. Jones, J. M.; Morrell, J. C.; Gould, S. J. J. Bio. Chem. 2000, 275, 12590.
4347
Acknowledgments
6. (a) Lee, S. J.; Liu, J.; Qi, N.; Guarnera, R. A.; Lee, S. Y.; Cicila, G. T. Hypertens Res.
2003, 26, 75; (b) Rice, T.; Rankinen, T.; Province, M. A.; Chagnon, Y. C.; Pérusse,
L.; Borecki, I. B.; Bouchard, C.; Rao, D. C. Circulation 2000, 102, 1956; (c)
Unpublished result from Merck Research Laboratory.
7. Rice, T.; Rankinen, T.; Michael, A.; Province, M. A.; Chagnon, Y. C.; Pérusse, L.
Physiol. Genomics 2004, 16, 256.
The research work was conducted as collaborative program be-
tween Advinus Therapeutics and Merck Research Laboratories.
Advinus Publication: ADV-A-018.
8. Franco, M. I. R.; Dorronsoro, I.; Higueras, A. I. H.; Antequera, G. Tetrahedron Lett.
2001, 42, 863.
References and notes
9. Barawkar, D. A.; Meru, A.; Bandyopadhyay, A.; Banerjee, A.; Deshpande, A. M.;
Athare, C.; Koduru, C.; Khose, G.; Gundu, J.; Mahajan, K.; Patil, P.; Kandalkar, S.
R.; Niranjan, S.; Bhosale, S.; De, S.; Mukhopadhyay, S.; Chaudhary, S.; Koul, S.;
Singh, U.; Chugh, A.; Palle, V. P.; Mookhtiar, K. A.; Vacca, J.; Chakravarty, P. K.;
Nargund, R. P.; Wright, S. D.; Roy, S.; Graziano, M. P.; Singh, S. B.; Cully, D.; Cai,
T.-Q. ACS Med. Chem. Lett. 2011, 2, 919.
10. (a) Cumane, L. M.; Barton, J. D.; Chen, Z-w.; Diep Le, K. H.; David, A.; Lederer, F.;
Mathews, F. S. Biochemistry 2005, 44, 1521; (b) Xia, Z-x.; Mathews, F. S. J. Mol.
Biol. 1990, 212, 837; (c) Stenberg, K.; Lindqvist, Y. Protein Sci. 1997, 6, 1009.
1. Ong, K. L.; Cheung, B. M. Y.; Man, Y. B.; Lau, C. P.; Lam, K. S. L. Hypertension 2007,
49, 69.
2. Kearney, P. M.; Whelton, M.; Reynolds, K.; Muntner, P.; Whelton, P. K.; He, H.
Lancet 2005, 365, 217.
3. Webb, R. L.; Schiering, N.; Sedrani, R.; Maibaum, J. J. Med. Chem. 2010, 53, 7490.
4. (a) Williams, B. Br. Med. J. 2003, 326, 61; (b) Chobanian, A. V.; Bakris, G. L.;
Black, H. R.; Cushman, W. C.; Green, L. A.; Izzo, J. L., Jr.; Jones, D. W.; Materson,
B. J.; Oparil, S.; Wright, J. T., Jr.; Roccella, E. J. J. Am. Med. Assoc. 2003, 289, 2560.