Structural basis for binding of cyclic 2-oxoglutarate analogues to factor-inhibiting hypoxia-inducible factor

Article abstract of DOI:10.1016/j.bmcl.2010.08.032

Aromatic analogues of the 2-oxoglutarate co-substrate of the hypoxia-inducible factor hydroxylases are shown to bind at the active site iron: Pyridine-2,4-dicarboxylate binds as anticipated with a single molecule chelating the iron in a bidentate manner. The binding mode of a hydroxamic acid analogue, at least in the crystalline state, is unusual because two molecules of the inhibitor are observed at the active site and partial displacement of the iron binding aspartyl residue was observed.

Full text of DOI:10.1016/j.bmcl.2010.08.032

Bioorganic & Medicinal Chemistry Letters 20 (2010) 6125–6128
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structural basis for binding of cyclic 2-oxoglutarate analogues
to factor-inhibiting hypoxia-inducible factor
⇑
Ana Conejo-Garcia, Michael A. McDonough , Christoph Loenarz, Luke A. McNeill, Kirsty S. Hewitson,
⇑
Wei Ge, Benoît M. Liénard, Christopher J. Schofield, Ian J. Clifton
The Department of Chemistry, Chemistry Research Laboratory, University of Oxford, 12 Mansfield Road, Oxford OX1 3TA, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Article history:
Aromatic analogues of the 2-oxoglutarate co-substrate of the hypoxia-inducible factor hydroxylases are
shown to bind at the active site iron: Pyridine-2,4-dicarboxylate binds as anticipated with a single mol-
ecule chelating the iron in a bidentate manner. The binding mode of a hydroxamic acid analogue, at least
in the crystalline state, is unusual because two molecules of the inhibitor are observed at the active site
and partial displacement of the iron binding aspartyl residue was observed.
Received 1 June 2010
Revised 5 August 2010
Accepted 5 August 2010
Available online 10 August 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Factor inhibiting hypoxia-inducible factor
Oxygenase
Hypoxic response
Hydroxamic acid
2-Oxoglutarate analogue
2-Oxoglutarate (2OG) and ferrous iron dependent oxygenases
(2OG oxygenases) play important roles in regulating the hypoxic
response in humans via the post-translational hydroxylation of hy-
poxia-inducible factor. Prolyl- and asparaginyl- hydroxylation of
2G19, 2G1M).7,8 Because of their different roles in the hypoxic re-
sponse, the identification of inhibitors that are selective for FIH or
the PHDs is of interest. The development of selective FIH inhibitors
may be challenging because FIH is closely related to the JmjC do-
main histone demethylase family, more so than are the PHDs. Here
we report the synthesis of monocyclic 2OG analogue inhibitor tem-
plates and three crystal structures of these compounds in complex
with FIH. The results should be useful for the design of more potent
and selective FIH inhibitors.
the HIF-
interaction of HIF-
a
subunit signals for their degradation and blocks the
with transcriptional co-activators, respec-
a
tively. In humans there are three HIF prolyl hydroxylase isoforms
(PHD1,2,3) and one HIF asparaginyl hydroxylase (FIH, factor-
inhibiting HIF) (for reviews see1–3). 2OG oxygenase homologues
e
of FIH have been shown to catalyse the demethylation of N -meth-
Pyridine-2,4-dicarboxylate 4 and 3-hydroxy pyridine-2-car-
bonyl glycine 6 are inhibitors of the pro-collagen and HIF prolyl
hydroxylases.9–11 The other cyclic 2OG analogues tested were pre-
pared by synthesis (Fig. 1). Compound 3 was prepared from ethyl 3-
hydroxybenzoate as reported.12 (3-Hydroxyphenyl)(oxo)acetic acid
2 was prepared via a route involving base mediated rearrangement
of a cyanohydrin carbonate ester (Scheme 1). Aromatic cyanohy-
ylated lysine-residues in histones (for review see4). Because of
their roles in the hypoxic response and chromatin regulation,
human 2OG oxygenases are current pharmaceutical targets.
There is interest in small molecules that target specific human
2OG oxygenases both for therapeutic and functional assignment
purposes. Most reported 2OG oxygenase inhibitors are either Fe(II)
chelators and/or 2OG analogues. Relatively few structures of hu-
man 2OG oxygenases with these inhibitors are available. Struc-
tures for FIH and the histone demethylase JMJD2A in complex
with N-oxalyl amino acid 2OG analogues, including N-oxalyl gly-
drin ester (9) was prepared by the method of Thasana et al.13
,
employing ethyl chloroformate, aqueous NaCN and a phase transfer
catalyst. We found that employing THF rather than CH2Cl2 as the
organic phase resulted in higher yields and avoided the need for
purification by distillation. Thiol 5 was prepared from 3-hydroxy-
phenyl acetic acid in seven steps (Scheme 2), in a route including
benzylic bromination. The basic deprotection conditions led to iso-
lation of the disulfide of the target thiol 5. The free thiol was gener-
ated by treatment with dithiothreitol immediately prior to assays.
The cyclic 2OG analogues were then evaluated for FIH inhibition
using both 1-[14C]-2OG turnover assay (monitoring 14CO2 produc-
tion) and MALDI MS and also as inhibitors of the catalytic domain
cine and N-oxalyl-D-phenylalanine have been described (PDB-ID
1H2K, 1H2M, and 1YCI).5,6 However, few structures of human
2OG oxygenases in complex with conformationally constrained
cyclic analogues of 2OG have been reported (PDB-ID 2G1M,
⇑
Corresponding author. Tel.: +44 1865 275677; fax: +44 1865 285002.
E-mail addresses: michael.mcdonough@chem.ox.ac.uk (M.A. McDonough), ian.
clifton@chem.ox.ac.uk (I.J. Clifton).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.08.032

6126
A. Conejo-Garcia et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6125–6128
R
O
O
O
R
N
HO
OH
HO
OH HO
OH
X
OH
HO
R
N
H
X
O
O
O
O
O
O
4
1a X = CH
X = NH
6
R = OH X = N
2 R = COOH
3 R = NHOH
5
R = SH
1
6a
R = H X = N
5a R = OH
6b R = OH X = CH
Figure 1. (a) The structures of the targeted 2OG analogues 1–9.
(i)
CHO TBDMSO
(ii)
(iii)
(iv)
O
O
O
CN O
OEt
TBDMSO
OEt
HO
OH
HO
CHO TBDMSO
O
O
2
7
8
9
10
Scheme 1. Reagents and conditions: (i) t-BuMe2SiCl, dimethylaminopyridine, Et3N, CH2Cl2; (ii) ClCOOEt, benzyltrimethylammonium chloride, THF, NaCN, H2O; (iii) LiN(iPr)2,
THF, À78 °C rt; (iv) NaOH, MeOH.
O
O
O
O
(iii)
(v)
(vii)
R2O
OR1
R2 = H
BzO
HO
OH
OMe
HO
OR4
R3
14 R3 = Br
15 R3 = -SCSOEt
S
SH
)
2
2
1
11
12
R = H
16 R4 = Me, Me
17 R4 = Me, Et
18 R4 = H, H
(i)
(ii)
5
(iv)
R = Me R2 = H
1
(vi)
13 R1 = Me R2 = Bz
Scheme 2. Reagents and conditions: (i) p-toluenesulfonic acid, HC(OCH)3, MeOH; (ii) benzoyl chloride, dimethylaminopyridine, Et3N, CH2Cl2; (iii) N-bromosuccinimide,
dibenzoyl peroxide, CCl4; (iv) potassium o-ethyldithiocarbonate, MeOH, dioxane; (v) NH4OH, MeOH; (vi) NaOH, MeOH, (vii) dithiothreitol.
of PHD2 using a homogeneous time-resolved fluorescence (HTRF)
assay for comparison.14 None of the compounds tested were potent
inhibitors.9–11 The IC50 values were in the mM range under stan-
2CGN, 1H2M, 1H2L, 1H2N, and 1H2K).15,16 However, in the case
of the FIH structure in complex with 4, compared to the other
structures there is an ꢀ80° rotation of the Cb–C
c bond in His199,
dard assay conditions (2: 1 mM; 3: 18
lM; 4: 625
lM; 5 0.5 mM,
the side chain of which coordinates the Fe(II), likely due to a steric
interaction with the pyridine ring of 4. The apparent inability of 4
to bind to the Fe(II) of FIH in an un-encumbered manner may ratio-
nalise its relative lack of potency.
6: 91 M; 6a: >1 mM). Similar results were observed with PHD2
l
(EGLN1), with none of the compounds tested causing more than
a 50% drop in activity when assayed at 1 mM concentration using
the HTRF assay, with the exception that the reported PHD inhibi-
tors9–11 pyridine-2,4-dicarboxylate 4 and 3-hydroxy pyridine-2-
carbonyl glycine 6 were observed to inhibit more potently than
the other cyclic inhibitors. The relative lack of potency of 4 with
FIH compared to PHD2 is interesting given the conserved metal-
coordination chemistry in 2OG oxygenases and prompted us to
carry out structural investigations on FIH. Crystal structures were
obtained for compounds 2, 3 and 4 in complex with FIH.
Comparison of the binding modes of 4 with FIH and JMJD2A
(PDB-ID 2VD7) reveals a striking difference, in that the orientations
of the plane of the pyridine ring of 4 are different. Although, 4
binds in the same plane as 2OG/NOG in FIH, it binds in a different
plane in JMJD2A. This is likely due to the different steric constraints
within the JMJD2 subfamily active site.
Because hydroxamic acids are clinically used as metallo-en-
zyme inhibitors, we then determined the structure of FIH in com-
plex with 3. Interestingly, the electron density map implied that
two molecules of 3 were bound to the Fe(II) at the active site. In
one binding mode, 3 was observed to bind in a manner closely re-
lated to that of 2OG, 2, and 4, with the hydroxamic acid observed to
bind in the anticipated bidentate manner. The second binding
mode was unexpected because it apparently involves the partial
displacement of the Asp201 carboxylate from coordinating to the
Fe(II).
The anticipated close relationship of 2 to the conformation of
2OG observed at the FIH active site (PDB-ID 1H2L, 1H2N)15 is sup-
ported by the crystallographic analysis. Compound 2 binds to the
Fe(II) in a similar manner as 2OG with its carbonyl oxygen trans
to His199 and its ketone oxygen trans to Asp201 (Figs. 2b and
S1b). The side-chain carboxylate of 2 forms the same interactions
as that of 2OG. The benzyl ring of 2 is located in a hydrophobic
e
pocket sandwiched between Leu188, Ile281, Phe207, and C of
Lys214. This structure reveals that monocyclic 2OG analogues
can bind at the active site in a manner closely related to 2OG
and encouraged further work on other 2OG analogues.
The hydroxamate nitrogen of 3 is positioned to form a hydrogen
bond with the Asp201 side-chain oxygen stabilising its non-metal
ligated position. The unexpected second binding mode of 3 with
displacement of the Asp201 side chain from its coordination to
Fe(II) may in part reflect the apparent unusual behaviour of FIH
with respect to its metal cofactor. Previous structural and bio-
chemical analyses demonstrated that substitution of Asp201 by
glycine or alanine residues does not ablate Fe(II) binding and the
D201G variant retains hydroxylation activity (PDB-ID 3D8C and
2ILM).17
Overall the results demonstrate that cyclic 2OG analogues can
bind at the FIH active site in a closely related manner to which
its 2OG co-substrate binds. Appropriate structure guided function-
alisation of these templates should enable the generation of more
potent and selective inhibitors with precedence from work on the
As anticipated from structural work on the histone demethylase
JMJD2A,8 pyridine-2,4-dicarboxylate 4 is coordinated to the FIH ac-
tive site Fe(II) in a bidentate manner via its pyridinyl nitrogen and
2-carboxylate, with its C4 carboxylate interacting with the same
residues as the 2OG C4 carboxylate. Unanticipated electron density
at the active site was modelled and refined as a glycerol molecule
(the cryoprotectant used), which is positioned to coordinate to the
Fe(II) and the side chain of one of the Fe(II) coordinating residues,
Asp201 (Figs. 2a and S1a). Comparison of the FIHÁ4 structure with
other FIH structures reveals several differences. The pyridine ring
of 4 is observed bound to FIH in the same plane as for 2OG and
the analogues NOG, succinate, and fumarate (PDB-ID 2CGO,

A. Conejo-Garcia et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6125–6128
6127
Figure 2. Crystal structures of FIH (green sticks) in complex with (a) 4 and, (b) 2 (both as yellow sticks) with 2Fo–Fc electron density (grey mesh) contoured at 1
r
. The active
site iron atom (orange sphere) is apparently coordinated by a glycerol molecule (purple sticks) in the case of 4, (c) stereo-view of the crystal structure of FIH (green sticks) in
complex with 3 (yellow sticks) showing the two different binding modes observed for this inhibitor and the conformational change of the Asp201 side chain which no longer
coordinates to the catalytic iron atom (orange sphere) in order to accommodate one of the binding modes of 3. Comparison of the active site pocket of FIH (grey surface) with
2, (d) N-oxalylglycine (NOG) (grey sticks) and (e) N-oxalyl-D-phenylalanine (NOFD) (grey sticks). The iron atom is an orange sphere.
JMJD2 family of histone demethylases.8,18 The results also demon-
strate that hydroxamic acids can bind to the active site transition
metal centre in 2OG oxygenases in a similar manner to which they
inhibit zinc dependent hydrolases.19 This observation is of interest
because the clinically used hydroxamic acid containing histone
deacetylase inhibitor SAHA was reported to also inhibit the JMJD2
enzymes.18 Finally, the observation that two molecules of the
hydroxamic acid 3 bind at the active site metal of FIH, apparently
causing the aspartyl residue that normally coordinates to move
away from the metal, suggests that it may be possible to identify
inhibitors of 2OG oxygenases that work via novel mechanisms,
such as by ejection of the active site iron.
Acknowledgements
We thank the BBSRC, the Wellcome Trust, the European Com-
mission: Marie Curie Programme MEIF-CT-2003-500525 (A.C.-G)
and the European Union for funding.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.08.032.
References and notes
Coordinates for FIH in complex with 2, 3, and 4 have been
deposited in the PDB as 2WA3, 2WA4, and 2W0X, respectively.
Further description of the structures and details of synthesis,
assays, and crystallography are provided as Supplementary data.
1. Hirota, K.; Semenza, G. L. Biochem. Biophys. Res. Commun. 2005, 338, 610.
2. Kaelin, J. W. G. Biochem. Biophys. Res. Commun. 2005, 338, 627.
3. Schofield, C. J.; Ratcliffe, P. J. Biochem. Biophys. Res. Commun. 2005, 338, 617.

6128
A. Conejo-Garcia et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6125–6128
4. Klose, R. J.; Kallin, E. M.; Zhang, Y. Nat. Rev. Genet. 2006, 7, 715.
5. Banerji, B.; Conejo-Garcia, A.; McNeill, L. A.; McDonough, M. A.; Buck, M. R. G.;
Hewitson, K. S.; Oldham, N. J.; Schofield, C. J. J. Chem. Soc., Chem. Commun. 2005,
5438.
12. Liénard, B. M. R.; Horsfall, L. E.; Galleni, M.; Frère, J.-M.; Schofield, C. J. Bioorg.
Med. Chem. Lett. 2007, 17, 964.
13. Thasana, N.; Prachyawarakorn, V.; Tontoolarug, S.; Ruchirawat, S. Tetrahedron
Lett. 2003, 44, 1019.
6. McDonough, M. A.; McNeill, L. A.; Tilliet, M.; Papamicael, C. A.; Chen, Q.-Y.;
Banerji, B.; Hewitson, K. S.; Schofield, C. J. J. Am. Chem. Soc. 2005, 127, 7680.
7. McDonough, M. A.; Li, V.; Flashman, E.; Chowdhury, R.; Mohr, C.; Liénard, B. M.
R.; Zondlo, J.; Oldham, N. J.; Clifton, I. J.; Lewis, J.; McNeill, L. A.; Kurzeja, R. J. M.;
Hewitson, K. S.; Yang, E.; Jordan, S.; Syed, R. S.; Schofield, C. J. Proc. Natl. Acad.
Sci. U.S.A. 2006, 103, 9814.
8. Rose, N. R.; Ng, S. S.; Mecinovic, J.; Lienard, B. M. R.; Bello, S. H.; Sun, Z.;
McDonough, M. A.; Oppermann, U.; Schofield, C. J. J. Med. Chem. 2008, 51, 7053.
9. Hirsilä, M.; Koivunen, P.; Günzler, V.; Kivirikko, K. I.; Myllyharju, J. J. Biol. Chem.
2003, 278, 30772.
14. Poppe, L.; Tegley, C. M.; Li, V.; Lewis, J.; Zondlo, J.; Yang, E.; Kurzeja, R. J. M.;
Syed, R. J. Am. Chem. Soc. 2009, 131, 16654.
15. 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.
16. Hewitson, K. S.; Liénard, B. M. R.; McDonough, M. A.; Clifton, I. J.; Butler, D.;
Soares, A. S.; Oldham, N. J.; McNeill, L. A.; Schofield, C. J. J. Biol. Chem. 2007, 282,
3293.
17. Hewitson, K. S.; Holmes, S. L.; Ehrismann, D.; Hardy, A. P.; Chowdhury, R.;
Schofield, C. J.; McDonough, M. A. J. Biol. Chem. 2008, 283, 25971.
´
18. Rose, N. R.; Woon, E. C. Y.; Kingham, G. L.; King, O. N. F.; Mecinovic, J.; Clifton, I.
10. Ivan, M.; Haberberger, T.; Gervasi, D. C.; Michelson, K. S.; Günzler, V.; Kondo,
K.; Yang, H.; Sorokina, I.; Conaway, R. C.; Conaway, J. W.; Kaelin, W. G. Proc.
Natl. Acad. Sci. U.S.A. 2002, 99, 13459.
11. Majamaa, K.; Hanauske-Abel, H. M.; Günzler, V.; Kivirikko, K. I. Eur. J. Biochem.
1984, 138, 239.
J.; Ng, S. S.; Talib-Hardy, J.; Oppermann, U.; McDonough, M. A.; Schofield, C. J. J.
Med. Chem. 2009, 53, 1810.
19. Finnin, M. S.; Donigian, J. R.; Cohen, A.; Richon, V. M.; Rifkind, R. A.; Marks, P.
A.; Breslow, R.; Pavletich, N. P. Nature (London) 1999, 401, 188.