Design and synthesis of a series of novel pyrazolopyridines as HIF 1-α prolyl hydroxylase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2006.08.017

Recently resolved X-ray crystal structure of HIF-1α prolyl hydroxylase was used to design and develop a novel series of pyrazolopyridines as potent HIF-1α prolyl hydroxylase inhibitors. The activity of these compounds was determined in a human EGLN-1 assa

Full text of DOI:10.1016/j.bmcl.2006.08.017

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5687–5690
Design and synthesis of a series of novel pyrazolopyridines as
HIF 1-a prolyl hydroxylase inhibitors
Namal C. Warshakoon,* Shengde Wu, Angelique Boyer, Richard Kawamoto,
Sean Renock, Kevin Xu, Matthew Pokross, Artem G. Evdokimov, Songtao Zhou,
Carol Winter, Richard Walter and Marlene Mekel
Procter & Gamble Pharmaceuticals Inc, 8700 Mason-Montgomery road, Mason, OH 45040, USA
Received 12 June 2006; revised 26 July 2006; accepted 1 August 2006
Abstract—Recently resolved X-ray crystal structure of HIF-1a prolyl hydroxylase was used to design and develop a novel series of
pyrazolopyridines as potent HIF-1a prolyl hydroxylase inhibitors. The activity of these compounds was determined in a human
EGLN-1 assay. Structure-based design aided in optimizing the potency of the initial lead (2, IC50 of 11 lM) to a potent (11l,
190 nM) EGLN-1 inhibitor. Several of these analogs were potent VEGF inducers in a cell-based assay. These pyrazolopyridines
were also effective in stabilizing HIF-1a.
Ó 2006 Elsevier Ltd. All rights reserved.
Hypoxia inducible factor (HIF) is a transcriptional com-
plex that plays a key role in mammalian oxygen homeo-
stasis and regulates a host of hypoxic response genes
that regulate angiogenic, glycolytic, and erythropoietic
processes.1–3 The subunit components, HIF-1a and
HIF-1b (ARNT) are constitutively expressed and
regulation is achieved by the selective destruction of
HIF-1a. HIF-1a is a major regulatory point of cellular
response to hypoxia (decreasing partial pressure of
oxygen).
controlled by oxygen concentration, under hypoxia the
hydroxylation of HIF-1a is inhibited, and HIF-1a binds
to ARNT to form a functional transcriptional activator
that turns on genes with hypoxic response elements (e.g.,
VEGF, EPO, glycolytic enzymes).7 Therefore, inhibition
of HIF-1a prolyl hydroxylases, resulting in HIF-1a sta-
bilization and has a potential to be a viable therapeutic
approach for ischemic diseases including myocardial
infarction, stroke, peripheral arterial disease, heart fail-
ure, diabetes, and anemia.
In the presence of oxygen, a family of non-heme iron
containing prolyl hydroxylases (EGLN-1, EGLN-2,
and EGLN-3) effects post-translational modification of
HIF-1a by hydroxylation of proline564 and proline402
in the oxygen dependent degradation domain (ODD)
via a reaction that requires 2-oxoglutarate (2OG) and
ascorbate.4–6 Proline hydroxylation then targets HIF-
1a subunits for proteasomal degradation via binding
to the VHL (Von Hippel Lindau tumor suppressor pro-
tein), elongin C/B, CuI2, Rbx1 ubiquitin ligase complex.
Since the post-translational hydroxylation of HIF-1a is
Neither crystal structure of the procollagen nor that of
HIF prolyl hydroxylases has been reported in the litera-
ture. However, atomic resolution structures have
been obtained for other 2-oxoglutarate dioxygenases
including bacterial proline 3-hydroxylase, deacetoxy-
cephalosporin C synthase, clavaminate synthase, and
asparaginyl hydroxylase.8 The similarity of cofactor
requirements for the catalytic step suggests that these
structural data should provide significant information
concerning the active site of HIF prolyl hydroxylase to
aid synthesis of small molecule inhibitors.
We sought to identify suitable small molecule HIF-1a
prolyl hydroxylase inhibitors based on structure-based
design approach utilizing the recently solved crystal
structure of EGLN-19 in complex with the isoquinoline
inhibitor 110 (Fig. 1).
Keywords: Prolyl hydroxylase inhibitors; HIF1-a; Hypoxia; Ischemia;
Peripheral arterial disease (PAD); Anemia; Pyrazolopyridines.
*
Corresponding author. Tel.: +1 513 622 4467; fax: +1 513 622
0523; e-mail: Warshakoon.nc@pg.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.08.017

5688
N. C. Warshakoon et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5687–5690
procedure to the synthesis of 2, was condensed with eth-
yl pyrazole 4. However, this reaction did not go to com-
pletion, and the separation of the product and the
starting material was extremely difficult. Conversely,
pyridine N-oxide 8 reacted with pyrazole 4 and pro-
duced desired condensed product 9 in good yields.
Hydrogenation of
9 under microwave conditions
cleaved both the benzyl ester and the N-oxide. Alkyl-
ation of 10 followed by the hydrolysis of the ethyl ester
generated the analogs 11a–h (Table 1).
The benchmark isoquinoline 1 has an IC50 of 1.4 lM in
the EGLN-1 enzyme assay. In comparison, compound 2
showed reasonable activity in the assay (12 lM) spur-
ring us to initiate a structure-based medicinal chemistry
program around the novel pyrazolopyridine scaffold.
Figures 2 and 3 show that by reaching out toward the
outer region of the active site, ligands can interact favor-
ably with the active site residues Tyr310, Arg322,
Met299, and Gln239. The modeling studies suggested
that an appropriate substitution placed at the 5-position
of the pyridyl nucleus has the ability to interact with
these residues, and therefore, brings about the optimal
binding energy of the ligands and the active site of
EGLN-1.
Figure 1. Binding of compound 1 to the catalytic domain of EGLN-1.
Among the key interactions between compound 1 and
the EGLN-1 active site are the salt bridge between the
terminal carboxylic acid and Arg383, the coordination
of iron by the isoquinoline nitrogen and the side chain
amide, and also the hydrogen bond between the pheno-
lic hydroxyl of the inhibitor and Tyr303.
One such analog (compound 11a) is expected to have a
strong coordination to the iron using both the pyridyl
and pyrazole nitrogens, while the oxygen atom on the
side-chain ether linkage builds a hydrogen bonding
interaction with OH of Tyr310. The phenyl ring inter-
acts with the side chain of Gln239 in a lipophilic fashion
(Fig. 3).
In our structure-based drug design approach, pyrazolo-
pyridine 2 emerged as a possible lead for novel HIF-1a
prolyl hydroxylase inhibitors. The predicted binding
mode of this compound is shown in Figure 2. In this
binding orientation the nitrogens in the pyridine and
pyrazole bind to iron, while the carboxylate interacts
with Arg383.
The pyrazolopyridine analogs were screened in an
EGLN-1 enzyme assay (Table 1). The activity of these
compounds was then measured in cell-based VEGF
induction assay, which measures events downstream of
HIF-1a stabilization (Table 1). In this study, we focused
our structure–activity relationship development on the
phenyl ring of the distal ether linkage. As suggested by
the modeling studies at the outset, the analog 11a
showed good activity against EGLN-1. A range of ring
substitutions (both electronic and positional) was toler-
ated by the enzyme. Thus, both electron-donating and -
withdrawing substituents displayed good activity. The
meta- and para-substitutions both showed comparable
activity; however, in the case of the chloride substituent,
the para-substituted pyrazolopyridine 11b had improved
activity as compared to the ortho- and meta-substitu-
tions (11c and d). Compound 11l, in which the distal
phenyl ring was substituted with a 2-cyano-phenyl
group, had nanomolar activity against the enzyme. In
addition to the interactions within the active site, de-
scribed earlier, we judiciously positioned the nitrile
group at the 2-position to build a hydrogen bonding
interaction with Trp258 at the top of the active site
(Fig. 4). Interestingly, the amide linkage at the 5-posi-
tion of the pyridyl nucleus had an adverse effect on the
activity against EGLN-1 (data not shown).
The reasonable binding mode of pyrazolopyridine 2
(Fig. 2) prompted us to probe this structural class more
carefully. The synthesis of 2 is outlined in the scheme be-
low (Scheme 1).
Additional analogs beginning with 6 were prepared
according to the scheme shown below (Scheme 2). The
commercially available, 2-chloro-5-hydroxymethylpyri-
dine 6 was protected as its benzyl ether and, in a similar
In the cell-based VEGF assay, these pyrazolopyridines
were able to induce significant VEGF induction. In
Figure 2. The predicted mode of binding between 2 and EGLN-1.

N. C. Warshakoon et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5687–5690
5689
a
b
N
N
N
CO2H
N
N
N
5
CO2Et
N
Cl
CO2Et
3
HN
2
N
4
Scheme 1. Reagents and conditions: (a) Cs2CO3/DMF/140 °C (65%); (b) KOH/EtOH/water (90%).
a
b
HO
BnO
BnO
N
Cl
N
Cl
N
O
8
Cl
6
7
BnO
c
d
HO
N
O
N
N
CO2Et
N
N
N
CO2Et
9
10
R
O
e
N
N
N
CO2H
11
Scheme 2. Reagents and conditions: (a) BnBr/THF/NaH (75%); (b) MeReO3/H2O2/CH2Cl2 (75%); (c) Cs2CO3/DMF/ethylpyrazole-3-carboxylate/
140 °C/30 min/MW (80%); (d) Pd(OH)2/cyclohexene/EtOH/120 °C/20 min/MW (70%); (e) 1—RX/NaH/THF (65–70%), 2—KOH/EtOH/water
(80%).
Table 1. Biological activity of pyrazolopyridines against EGLN-1 and
VEGF
O
R
N
N
N
CO2H
11
Entry
R
EGLN-1–IC50
(lM)
VEGF EC50
(lM)
11a
11b
11c
11d
11e
11f
11g
11h
11i
H
2.7
0.58
1.7
2.4
2.4
1.6
1.6
2.1
2.9
1.7
2.8
0.19
25
p-Chloro
m-Chloro
o-Chloro
p-Methoxy
m-Methoxy
p-Cyano
m-Methyl
7.4
3.8
na
1.0
>100
na
Figure 3. The predicted mode of binding between 11a and EGLN-1.
na
11
m-Trifluoromethyl
p-Isopropyl
3,5-Dimethoxy
11j
na
na
11k
11l
>100
CN
addition, these EGLN inhibitors were effective in stabi-
lizing HIF-1a. Therefore, this study serves to establish
our initial hypothesis of stabilization of HIF-1a via
inhibiting HIF-1a prolyl hydroxylase and subsequently
leading to increase in VEGF induction. Potency changes
noted in the enzyme assay were not consistently seen in
the cell-based VEGF induction assay. The apparent
discrepancies probably reflect the difficulties of reconcil-
ing the physical properties of the compounds and
Figure 4. The predicted mode of binding between 11l and EGLN-1.

5690
N. C. Warshakoon et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5687–5690
compatibility with differing conditions and requirements
of isolated enzyme and cellular assays.
the cells. The conditioned media were analyzed for
VEGF with a Quantikine human VEGF immunoassay
kit (R&D Systems, Minneapolis, MN). Optical density
measurements at 450 nm were recorded using the Spec-
tra Max 250 (Molecular Devices, Sunnyvale, CA). Data
defined as % of DFO stimulation were used to calculate
EC50 values with GraphPad Prism 4 software.
In summary, this study indicates the ability to design po-
tent, nonpeptidic, HIF-1a prolyl hydroxylase inhibitors
using a structure-based drug design approach. This nov-
el class of pyrazolopyridines is currently under investiga-
tion for its PK and in vivo properties. The results will be
disclosed elsewhere.
Acknowledgments
EGLN-1 activity assay. The EGLN-1 enzyme activity
was determined using mass spectrometry (matrix-assist-
ed laser desorption ionization, time-of-flight MS, and
MALDI-TOF MS). The HIF-1a peptide corresponding
to residues 556–574 (DLDLEALAPYIPADDDFQL)
was used as substrate. The reaction was conducted in
a total volume of 50lL containing Tris–Cl (5 mM, pH
7.5), ascorbate (120 lM), 2-oxoglutarate (3.2 lM),
HIF-1a (8.6 lM), and bovine serum albumin (0.01%).
EGLN-1, quantity predetermined to hydroxylate 20%
of substrate in 20 min, was added to start the reaction.
Where inhibitors were used, compounds were prepared
in dimethylsulfoxide at 10-fold final assay concentra-
tion. After 20 min at room temperature, the reaction
was stopped by transferring 10 lL of reaction mixture
to 50 lL of a mass spectrometry matrix solution (a-cya-
no-4-hydroxycinnamic acid, 5 mg/mL in 50% acetoni-
trile/0.1% TFA, 5 mM NH4PO4). Two microliters of
the mixture was spotted onto a MALDI-TOF MS target
plate for analysis with an Applied Biosystems (Foster
City, CA) 4700 Proteomics Analyzer MALDI-TOF
MS equipped with a Nd:YAG laser (355 nm, 3 ns pulse
width, 200 Hz repetition rate). Hydroxylated peptide
product was identified from substrate by the gain of
16 Da. Data defined as percent conversion of substrate
to product were analyzed in GraphPad Prism 4 to calcu-
late IC50 values.
NCW wishes to thank Dr. Joseph Gardner for his con-
tinuous support and guidance. Ms. Anne Russell and
Ms. Marcia Ketcha are gratefully acknowledged for
analytical instrumentation support. The authors wish
to acknowledge Chuiying Li for EGLN-1 cloning stud-
ies and Michelle Tscheiner for the synthesis of HIF-1a
peptide.
References and notes
1. Mack, C. A.; Magovern, C. J.; Budenbender, K. T.; Patel,
S. R.; Schwatz, E. A.; Zanzonico, P.; Ferris, B.; Sanborn,
T.; Isom, O. W.; Crystal, R. G.; Rosengart, T. K. J. Vasc.
Surg. 1998, 27, 699.
2. Bruick, R. K.; McKnight, S. L. Genes Dev. 2001, 15, 2497.
3. Safran, M.; Kaelin, W. G., Jr. J. Clin. Invest. 2003, 111,
779.
4. Semenza, G. L. J. Appl. Physiol. 2000, 88, 1474.
5. Semenza, G. L. Cell 2001, 107, 1.
6. Kondo, K.; Kaelin, W. G., Jr. Exp. Cell Res. 2001, 264, 117.
7. Semenza, G. Biochem. Pharmacol. 2002, 64, 993.
8. (a) Cunliffe, C. J.; Franklin, T. J.; Hales, N. J.; Hill, G. B.
J. Med. Chem. 1992, 35, 2652; (b) Valegard, K.; Van
Scheltinga, A. C.; Lloyd, M. D.; Hara, T.; Ramaswamy,
S.; Parrakis, A.; Thompson, A.; Lee, H. J.; Baldwin, J. E.;
Schofield, C. J.; Hajdu, J.; Anderson, I. Nature 1998, 394,
805; (c) Zhang, Z.; Ren, J.; Harlos, K.; McKinnon, C. H.;
Clifton, I. J.; Schofield, C. J. FEBS Lett. 2002, 517, 7; (d)
Elkins, J. M.; Hewitson, K. S.; McNeill, L. A.; Seibel, J.
F.; Schlemminger, I.; Pugh, C. W.; Ratcliffe, P. J.;
Schofield, C. J. J. Biol. Chem. 2003, 278, 1802.
9. Evdokimov, et al., Nat. Struct. Mol. Biol. 2006, under
review.
VEGF ELISA. HEK293 cells were seeded in 96-well
poly-lysine coated plates at 20,000 cells per well in
DMEM (10% FBS, 1% NEAA, and 0.1% glutamine).
Following overnight incubation, the cells were washed
with 100lL of Opti-MEM (Gibco, Carlsbad, CA) to re-
move serum. Compound in DMSO was serially diluted
(beginning with 100 lM) in Opti-MEM and added to
10. Guenzler, P.; Volkmar, K.; Stephen, J.; Liu, D. Y.; Todd,
W. PCT. In. Appl. WO 2005007192, 2005.