Inhibition of hypoxia-induced gene transcription by substituted pyrazolyl oxadiazoles: Initial lead generation and structure-activity relationships

Article abstract of DOI:10.1002/cmdc.201300357

The transcription factors hypoxia-inducible factor-1 and -2 (HIF-1 and HIF-2) orchestrate a multitude of processes that allow tumor cells to survive under conditions of low oxygen and nutrients, and that lead to resistance to some apoptotic pathways and facilitate invasion and metastasis. Therefore, inhibition of transactivation by HIF has become an attractive target in cancer research. Herein we present the results of a cell-based screening approach that led to the discovery of substituted 1H-pyrazole-3-carboxamides. Chemical optimization of the hit class with respect to potency and metabolic stability is described; it resulted in novel 5-(1H-pyrazol-3-yl)-1,2,4-oxadiazoles that inhibit the hypoxia-induced accumulation of HIF-1α and HIF-2α. The HIF inhibitory potency in the screening cell system was improved from IC ₅₀ 190 to 0.7 nM, and significant parts of the SAR are disclosed. For a key compound, the ability to suppress the hypoxia-induced expression of HIF target genes was studied in A549 human lung adenocarcinoma cells. The same compound shows a favorable pharmacokinetic profile in rats after i.v. and p.o. administration. Suppressing HIF target genes: Substituted 5-(1H-pyrazol-3-yl)-1, 2,4-oxadiazoles are presented as a novel chemotype to specifically inhibit the hypoxia-induced transcription of target genes of the transcription factor hypoxia-inducible factor (HIF). The new chemotype was derived from 1H-pyrazole-3-carboxamides that had been discovered from a cell-based screen. We present the optimization of the potency and metabolic stability of the initial screening hit. Copyright

Full text of DOI:10.1002/cmdc.201300357

CHEMMEDCHEM
COMMUNICATIONS
DOI: 10.1002/cmdc.201300357
Inhibition of Hypoxia-Induced Gene Transcription by
Substituted Pyrazolyl Oxadiazoles: Initial Lead Generation
and Structure–Activity Relationships
Michael Hꢀrter,*[a] Karl-Heinz Thierauch,[b] Stephen Boyer,[c] Ajay Bhargava,[c]
Peter Ellinghaus,[d] Hartmut Beck,[a] Susanne Greschat-Schade,[e] Holger Hess-Stumpp,[b] and
Kerstin Unterschemmann[b]
The transcription factors hypoxia-inducible factor-1 and -2
(HIF-1 and HIF-2) orchestrate a multitude of processes that
allow tumor cells to survive under conditions of low oxygen
and nutrients, and that lead to resistance to some apoptotic
pathways and facilitate invasion and metastasis. Therefore, in-
hibition of transactivation by HIF has become an attractive
target in cancer research. Herein we present the results of
a cell-based screening approach that led to the discovery of
substituted 1H-pyrazole-3-carboxamides. Chemical optimiza-
tion of the hit class with respect to potency and metabolic sta-
bility is described; it resulted in novel 5-(1H-pyrazol-3-yl)-1,2,4-
oxadiazoles that inhibit the hypoxia-induced accumulation of
HIF-1a and HIF-2a. The HIF inhibitory potency in the screening
cell system was improved from IC50 190 to 0.7 nm, and signifi-
cant parts of the SAR are disclosed. For a key compound, the
ability to suppress the hypoxia-induced expression of HIF
target genes was studied in A549 human lung adenocarcino-
ma cells. The same compound shows a favorable pharmacoki-
netic profile in rats after i.v. and p.o. administration.
erodimeric transcription factors consist of a constitutively ex-
pressed HIF-1b subunit (also called ARNT) and one of the
highly regulated HIF-1a or HIF-2a subunits. Although constant-
ly synthesized, the a subunits undergo rapid degradation in
well-oxygenated tissue (normoxic conditions), a process that
starts with oxygen-dependent hydroxylation of specific proline
residues of the HIF-1a/2a proteins, catalyzed by prolyl hydrox-
ylases (most importantly PHD-2), followed by binding to the
von Hippel–Lindau (VHL) E3 ligase, which marks HIF-1a/2a
proteins for proteasome-mediated degradation.[3] Prolyl hy-
droxylase activity is decreased under hypoxic conditions, lead-
ing to an accumulation of HIF-1a and HIF-2a, which translo-
cate into the nucleus where they heterodimerize with HIF-1b
and induce the transcription of HIF target genes by binding to
so-called hypoxia-response elements in their respective pro-
moter regions.[4] It is the action of the protein products of the
HIF target genes that leads to the profound changes observed
in hypoxic tumor cells: shift from mitochondrial respiration to
glycolysis for ATP production, altered catabolism to maintain
cell proliferation, induction of angiogenesis, facilitation of inva-
sion and metastasis, and suppression of certain apoptotic
mechanisms.[5] Therefore, the HIF pathway has become an at-
tractive target in anticancer research, and a number of com-
pounds have been reported recently that modulate HIF target
gene expression by various molecular modes of action.[6] Of
the compounds that can be regarded as rather specific HIF in-
hibitors, only EZN-2968 is currently under evaluation in clinical
studies (phase 1). EZN-2968 is a locked nucleic acid (LNA) anti-
sense oligonucleotide that targets HIF-1a mRNA.[7]
A common feature of growing tumors are areas of transient
and chronic hypoxia, in which tumor cells are distant from
a supportive blood vessel and thus suffer from a lack of
oxygen (as well as nutrients and the removal of metabolic
waste). It has been recognized that hypoxic tumor cells are
often resistant to chemo- and radiotherapy and give rise to
a more invasive tumor type. Clinically, tumor hypoxia is associ-
ated with aggressive disease and poor prognosis.[1] The adap-
tation of tumor cells to their hostile hypoxic environment is
orchestrated mainly by the action of the transcription factors
hypoxia-inducible factor-1 and -2 (HIF-1 and HIF-2).[2] The het-
We set out to identify novel HIF pathway inhibitors by
screening our company’s compound collection, using the
human colorectal carcinoma cell line HCT-116 stably transfect-
[a] Dr. M. Hꢀrter, Dr. H. Beck
[e] Dr. S. Greschat-Schade+++
Department of Medicinal Chemistry
Research Pharmacokinetics Group
Bayer HealthCare Pharmaceuticals, Global Drug Discovery
Postfach 101709, 42096 Wuppertal (Germany)
E-mail: michael.haerter@bayer.com
Bayer HealthCare Pharmaceuticals, Global Drug Discovery
[+] Current address: Bristol-Myers Squibb, 5 Research Parkway, Wallingford, CT
06492 (USA)
[b] Dr. K.-H. Thierauch, Dr. H. Hess-Stumpp,+++ Dr. K. Unterschemmann+++
[
++] Current address: Shakti BioResearch, 1 Bradley Road, Suite 401, Wood-
bridge, CT 06525 (USA)
+++] Currently at various functions within Bayer HealthCare Pharmaceuticals,
Therapeutic Research Group, Oncology
Bayer HealthCare Pharmaceuticals, Global Drug Discovery
[c] Dr. S. Boyer,+ Dr. A. Bhargava++
[
Global Drug Discovery
Former Exploratory Research Group, Small Molecules Cancer
Bayer Pharmaceuticals (USA)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201300357.
[d] Dr. P. Ellinghaus
Biomarker Research Group, Clinical Science
Bayer HealthCare Pharmaceuticals, Global Drug Discovery
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 61 – 66 61

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
ed with a plasmid containing four times the hypoxia response
element of the VEGF promoter in front of a luciferase reporter
gene construct. Cultivation of the cells under hypoxic condi-
tions (1% O2 for 16 h) led to hypoxia/HIF-induced expression
of luciferase, which was suppressed by the test compounds to
varying extents. Compound 1 (Figure 1) was found as the rep-
resentative example of a confirmed hit cluster of 1H-pyrazole-
3-carboxamides. It inhibits the luciferase signal with an IC50
value of 190 nm.
Table 2. Regiochemistry at the anilinic phenyl ring and its effect on HIF–
Luc potency.
Compd
RA
Regioisomer
IC50 [nm][a]
4
ÀOCF3
ÀOCF3
ÀOCF3
ÀiPr
ÀiPr
ÀiPr
ÀF
para
meta
ortho
para
meta
ortho
para
meta
ortho
para
meta
ortho
25
5000
>10000
39
8100
>10000
150
12
13
14
15
16
17
18
19
20
21
22
ÀF
890
7200
240
3200
ÀF
ÀCl
Figure 1. Representative example from the screening hit cluster “1H-pyra-
zole-3-carboxamides”. HIF–Luc IC50: inhibitory potency in a cell-based report-
er assay; see text and Experimental Section for explanations.
ÀCl
ÀCl
10000
[a] See footnotes in Table 1 for details.
Optimization of the screening hit started with variations of
the para substituents of the anilinic and the benzylic phenyl
rings, RA and RB, both with respect to their chemical nature
and their regiochemistry. The results for some of the RA ana-
logues that we prepared are summarized in Tables 1 and 2; re-
sults for some of the RB analogues can be found in Table 3. A
significant gain in potency was observed when the bromine
was replaced by N-bound piperidine (compound 2, 3.7 nm,
Table 1). Other substituents led to more potent compounds as
well. In general, it seems that bulky and/or lipophilic substitu-
ents RA increase potency, whereas more polar groups or less
bulky substituents have the opposite effect (Table 1). There is
a strong preference of RA for the para position, as is evident
from the examples listed in Table 2.
para position over both meta and ortho (Table 3). We invested
less effort into the optimization of RB at that stage, as we
became aware of the hydrolytic instability of some of the 1H-
pyrazole-3-carboxamides; we therefore redirected our atten-
tion from optimizing RB to addressing the stability issue. Cleav-
age of the amide bond was observed in vitro in rat plasma,
which translated into above liver blood flow clearance in vivo
(HTS hit 1, Figure 2). We were able to replace the amide bond
with an oxadiazole ring, which not only resolved the clearance
problem, but also improved potency (compound 30, Figure 2).
Therefore, we abandoned the initial HTS structure class and
turned our attention to further optimizing the new class of 5-
(1H-pyrazol-3-yl)-1,2,4-oxadiazoles.
The number of substituents RB investigated at the benzylic
phenyl ring was smaller, and less SAR information was ob-
tained. Apparently, there is only a weak preference for the
In the new series, the strong preference of the substituent
RA (now attached to the oxadiazole-bound phenyl ring) for the
para position was confirmed (Table 4), as was the weak prefer-
Table 3. Effect of the substituent RB of the benzylic phenyl ring on HIF–
Table 1. Effect of the para substituent RA of the anilinic phenyl ring on
Luc potency.
HIF–Luc potency; representative examples out of 98 analogues in total.
Compd
RA
IC50 [nm][a]
Compd
RA
IC50 [nm][a]
Compd
RA
Regioisomer
IC50 [nm][a]
23
24
3
25
26
27
28
29
MeOÀ
MeOÀ
MeÀ
MeÀ
ClÀ
para
meta
para
ortho
para
para
para
para
11
29
15
14
31
37
56
170
2
3.7
7
50
3
4
5
6
ÀtBu
15
25
36
42
8
9
10
11
À(CO)NMe2
ÀOMe
170
220
330
460
ÀOCF3
ÀcHex
ÀSF5
ÀSO2Me
ÀH
HÀ
FÀ
[a] IC50: inhibitory potency in the cell-based HIF–Luc reporter assay. The
relative error of the IC50 values is in the range of Æ40%. See main text
and Experimental Section for further explanations.
iPrÀ
[a] See footnotes in Table 1 for details.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 61 – 66 62

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
Table 6. Effect of the substituent RA of the oxadiazole-bound phenyl ring
on HIF–Luc potency; representative examples out of 220 analogues in
total.
Compd
RA
IC50 [nm][a]
Compd
RA
IC50 [nm][a]
Figure 2. Strong decrease of in vivo clearance by replacement of the central
amide bond with an oxadiazole ring in the screening hit 1. CL: plasma clear-
ance in mice after i.v. administration of 1 mgkgÀ1 test compound.
36
3
44
30
37
38
39
4
8
45
46
47
30
40
40
Table 4. Regiochemistry at the oxadiazole-bound phenyl ring and its
effect on HIF–Luc potency.
10
40
41
42
10
20
20
48
49
50
50
60
Compd
RA
Regioisomer
IC50 [nm][a]
31
32
ÀCF3
ÀCF3
para
meta
40
>10000
200
[a] See footnotes in Table 1 for details.
43
20
51
500
[a] See footnotes in Table 1 for details.
ence of RB likewise for para (Table 5). We then synthesized
a number of analogues with different substituents in either
para position: approximately 220 RA analogues and roughly 70
RB analogues were prepared and tested for their HIF–Luc po-
tency. The results for some illustrative examples (with a certain
bias for the more potent compounds) are listed in Tables 6 and
7.
preferred. The methylamino derivative 52 (2 nm, first entry in
Table 7) is tenfold more potent than the corresponding methyl
analogue 33 (19 nm, first entry in Table 5). The rather low po-
tencies of some of the analogues with ionizable groups may
be due to their low target affinities, or may also be explained
by a diminished ability to cross cell membranes (or both).
The combination of the two most potent substituents RA
and RB, i.e., 2,2,2-trifluoro-1,1-dimethylethyl of compound 36
(Table 6) and methylamino of compound 52 (Table 7), gave
compound 62, which is the most potent compound of the
entire series, with a HIF–Luc IC50 value of 0.7 nm (Figure 3).
Nevertheless, it was compound 33 (Table 5) that caught our
attention very early during the course of the lead optimization,
and we decided to investigate this compound in some detail.
The ability of compound 33 to suppress HIF target gene ex-
pression in vitro was demonstrated by using human lung ade-
nocarcinoma cell line A549, cultured under hypoxic conditions
(1% O2 for 16 h). In line with the IC50 values on the reporter
gene signal observed in the HCT-116 HIF–Luc screening cell
line, compound 33 was also able to suppress hypoxia-induced
gene expression in the non-transfected lung carcinoma cell
line A549 at concentrations above 10 nm. Figure 4 displays the
concentration-dependent decrease in mRNA levels of the HIF
target genes CA-IX, EGLN-3, and VEGF, as quantified by RT-PCR.
There is no effect of compound 33 on the expression level of
the non-HIF target gene EGLN-2 (Figure 4). Additional gene ex-
pression data are available in the Supporting Information, as
are western blots that show decreased HIF-1a and HIF-2a pro-
The previously made observation holds true that bulky and
lipophilic groups RA improve potency, whereas compounds
with smaller substituents or polar groups (such as alcohols,
amides, and carbamates) are less potent. The highest increase
in potency was achieved with the 2,2,2-trifluoro-1,1-dimethyl-
ethyl derivative 36 (3 nm, first entry in Table 6) which was in-
troduced as a potentially metabolically stable version of a tert-
butyl group. At the benzylic phenyl ring, some polar substitu-
ents RB are well tolerated, whereas bulkiness seems not to be
Table 5. Regiochemistry at the benzylic phenyl ring and its effect on
HIF–Luc potency in the oxadiazole series.
Compd
RB
Regioisomer
IC50 [nm][a]
33
34
35
MeÀ
MeÀ
MeÀ
para
meta
ortho
19Æ7
100
500
[a] See footnotes in Table 1 for details.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 61 – 66 63

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
Table 7. Effect of the substituent RB of the benzylic phenyl ring on HIF–
Luc potency; representative examples out of 70 analogues in total.
Figure 3. The most potent compound of the lead series was derived from
the initial screening hit 1 (Figure 1) by substitution of the amide bond for
an oxadiazole ring, the introduction of a methylamino group in place of
a methyl, and the introduction of a 2,2,2-trifluoro-1,1-dimethylethyl group
replacing a bromine.
Compd
RB
IC50 [nm][a] Compd
RB
IC50 [nm][a]
52
2
5
57
58
40
53
54
100
500
Compound 33 suppresses hypoxia-induced expression of HIF
target genes in vitro, whereas it does not have an effect on
the expression levels of non-HIF target genes. The molecular
target and thus the exact mode of action of compound 33
(and the entire pyrazolyl oxadiazole class) is currently under in-
vestigation. Compound 33 has a favorable pharmacokinetic
profile in rat after intravenous and oral dosing. The structure–
activity relationships reported herein show that further optimi-
zation of the lead is possible and likely promising, with com-
pound 62 (HIF–Luc IC50 =0.7 nm) being the most potent
member of the series to date.
10
59
55
56
15
40
60
61
2000
4000
[a] See footnotes in Table 1 for details.
tein concentrations in the pres-
ence of compound 33. A more
detailed description of the phar-
macology of compound 33 will
be the subject of a separate
communication.
Key pharmacokinetic proper-
ties of compound 33 are sum-
marized in Table 8. After intrave-
nous administration to rats, com-
pound 33 displays low clearance
and a very high volume of distri-
bution, resulting in a long half-
life of 30 h. When applied orally,
the
absolute
bioavailability
amounts to a favorable 74% and
the half-life is, again, close to
30 h.
In conclusion, we have pre-
sented the discovery of com-
pound 33 (HIF–Luc IC50 =19 nm)
as a novel lead structure in the
field of HIF pathway inhibitors. It
was developed from the initial
screening hit 1 (HIF–Luc IC50 =
190 nm) by substitution of an
amide bond for an oxadiazole
ring and replacement of a bro-
mine by
a
trifluoromethoxy
Figure 4. Effect of HIF pathway inhibitor 33 on the expression levels of HIF target genes (CA-IX, EGLN-3, and
VEGF) and a non-HIF target gene (EGLN-2) in hypoxic A549 human lung adenocarcinoma cells. The cells were in-
cubated for 16 h under hypoxic conditions (1% pO2) with various concentrations of 33. Gene expression was
group which improved the enzy-
matic hydrolytic stability signifi-
cantly and increased potency. quantified by RT-PCR.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 61 – 66 64

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
er’s instructions, and after five minutes the luminescence was mea-
sured. Cells that were incubated under normoxic conditions served
as background controls. Most IC50 values given in this report are
the average of two to four independent measurements and were
rounded to two significant figures. The relative error of the IC50
values is in the range of Æ40% on average.
Table 8. Pharmacokinetic parameters of compound 33 after intravenous
(i.v.) and oral (p.o.) application of solutions to rats.[a]
Intravenous[b]
[dose=0.36 mgkgÀ1
Oral[b]
[dose=3 mgkgÀ1
]
]
CLblood [LhÀ1·kgÀ1
VSS [LkgÀ1
1/2 [h]
]
0.58
6.9
30
AUCnorm [kghLÀ1
]
1.92
0.16
29
]
Cmax,norm [kgLÀ1
t1/2 [h]
]
Suppression of HIF target genes in vitro: Human bronchial carcino-
ma cells (A549 cell line) were incubated for 16 h with variable con-
centrations of the test substances (0.001–10 mm) under either nor-
moxic conditions or under 1% O2 (see HIF–Luciferase assay above).
The total RNA was isolated from the cells and transcribed into
cDNA, and the mRNA expression of HIF target genes (and non-HIF
target genes as controls) was analyzed by real-time PCR. mRNA ex-
pression levels of the HIF target genes are stronger under hypoxic
conditions. Active test substances suppress the low constitutive
mRNA expression levels of HIF target genes under normoxic condi-
tions, but display a much stronger effect under hypoxic conditions.
Expression levels of non-HIF target genes are unaffected by hypo-
xia and the presence of test compounds. Data are displayed as rel-
ative mRNA expression, for which the expression under hypoxic
conditions and in the absence of test compound is set at 100%.
Numbers of independent measurements: CA-IX (7), EGLN-3 (2),
VEGF (2), EGLN-2 (1).
t
F [%]
74
[a] CL: clearance; VSS: volume of distribution; t1/2 : half-life; AUCnorm: dose-
normalized exposure (area under the curve); Cmax,norm: dose-normalized
maximum plasma concentration; F: absolute oral bioavailability.
[b] 5 mLkgÀ1 vehicle: plasma/DMSO=98:2 (i.v.) or solutol/ethanol/
water=40:10:50 (p.o.).
Experimental Section
Chemistry: Compound 33 was prepared in three steps from com-
mercially available starting materials according to Scheme 1. Exper-
imental details and analytical data are given in the Supporting In-
formation, where the syntheses and analytical data of compounds
1, 30, and 62 can also be found.
HIF–Luciferase assay: Human colorectal carcinoma cells (HCT-116
cell line) were stably transfected with a plasmid containing a lucifer-
ase reporter under the control of an HIF-responsive sequence con-
sisting of a fourfold repeat of the hypoxia-responsive element of
the human VEGF promoter. These cells were sown in microtiter
plates: 20000 cells per well in RPMI 1640 medium with 10% fetal
calf serum (FCS) and 100 mgmLÀ1 hygromycin. Incubation was car-
ried out overnight under standard conditions: 21% O2, 5% CO2,
378C, humidified. The following morning, the cells were incubated
with various concentrations of test substances (0–10 mm) in a hypo-
xia chamber with 1% O2. After 24 h, Bright-GloTM reagent (Prome-
ga, Wisconsin, USA) was added in accordance with the manufactur-
Animal experiments: All animal experiments were performed in ac-
cordance with German law on animal protection as revised in its
current version. Permission is granted to Nonclinical Research Phar-
macokinetics of Bayer Pharma AG by the animal protection agency
of the state of North Rhine-Westphalia (Landesamt fꢂr Natur,
Umwelt und Verbraucherschutz Nordrhein-Westfalen) to which the
experiments had been notified.
Acknowledgements
The authors thank their co-workers for their dedicated and skillful
work. Especially Karin Smyk, Peter Voits, Kirstin Seifert, Jana Wꢀt-
zold, Sandra Zickelbein, Uwe Appel, Claus Jorzik, Philip Wickens,
and Wenlang Fu, who conducted the chemical syntheses and car-
ried out the HIF–Luc in vitro screen. The authors also thank
Donald Bierer for the preparation of some parallel synthesis libra-
ries, which helped to establish the early structure–activity rela-
tionships. The support by management functions (especially by
Dieter Zopf, Michael Brands, Stefan Jaroch, Klaus Bosslet, Helmut
Haning, and Karl Ziegelbauer) and the contributions by service
functions of the various institutes are also gratefully acknowl-
edged.
Keywords: cancer · hit optimization · hypoxia-inducible factor
(HIF) · inhibitors · transcription factors
[1] a) P. Vaupel, A. Mayer, Cancer Metastasis Rev. 2007, 26, 225–239; b) G. L.
Semenza, Oncogene 2010, 29, 625–634.
[2] a) G. L. Semenza, G. L. Wang, Mol. Cell. Biol. 1992, 12, 5447–5454; b) G. L.
Wang, G. L. Semenza, J. Biol. Chem. 1995, 270, 1230–1237.
[3] a) C. J. Schofield, P. J. Ratcliffe, Nat. Rev. Mol. Cell Biol. 2004, 5, 343–354;
b) M. Ivan, K. Kondo, H. Yang, W. Kim, J. Valiando, M. Ohh, A. Salic, J. M.
Asara, W. S. Lane, W. G. Kaelin, Jr., Science 2001, 292, 464–468; c) G. L. Se-
menza, Physiology 2009, 24, 97–106; d) Y. V. Liu, G. L. Semenza, Cell Cycle
2007, 6, 656–659; e) M. Y. Koh, T. R. Spivak-Kroizman, G. Powis, Trends
Biochem. Sci. 2008, 33, 526–534; f) I. K. Nordgren, A. Tavassoli, Chem. Soc.
Rev. 2011, 40, 4307–4317.
Scheme 1. Synthesis of compound 33. Reagents and conditions: a) AcOH,
908C, 4 h, chromatographic separation of regioisomers (48% yield); b) NaOH
(1m in H2O), EtOH, 708C, 15 h, acidification with HCl(aq) (97%); c) EDCI/HOBT,
DMF, RT, 1 h, then 1408C, 2 h (69%).
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 61 – 66 65

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
[4] a) A. Loboda, A. Jozkowicz, J. Dulak, Mol. Cells 2010, 29, 435–442; b) B.
Keith, R. S. Johnson, M. C. Simon, Nat. Rev. Cancer 2012, 12, 9–22;
c) S. M. Choi, H. Oh, H. Park, FEBS J. 2008, 275, 5618–5634; d) A. Ortiz-
Barahona, D. Villar, N. Pescador, J. Amigo, L. del Peso, Nucleic Acids Res.
2010, 38, 2332–2345.
[5] a) K. Hirota, G. L. Semenza, Crit. Rev. Oncol. Hematol. 2006, 59, 15–26;
b) M. J. Calzada, L. del Peso, Clin. Transl. Oncol. 2007, 9, 278–289; c) G. L.
Semenza, Semin. Cancer Biol. 2009, 19, 12–16; d) H. Harada, M. Hiraoka,
Curr. Signal Transduction Ther. 2010, 5, 188–196; e) N. M. Mazure, M. C.
Brahimi-Horn, J. Pouyssꢃgur, Bull. Cancer 2011, 98, 40–46; f) N. Rohwer,
T. Cramer, Drug Resist. Updates 2011, 14, 191–201.
c) D. T. Jones, A. L. Harris, Expert Opin. Ther. Targets 2012, 16, 463–480;
d) R. Wang, S. Zhou, S. Li, Curr. Med. Chem. 2011, 18, 3168–3189; e) Y.
Hu, J. Liu, H. Huang, J. Cell. Biochem. 2013, 114, 498–509; f) H. Minegishi,
T. Matsukawa, H. Nakamura, ChemMedChem 2013, 8, 265–271.
[7] R. B. Cohen, A. Olszanski, J. Figueroa, H. Hurwitz, F. M. Lokiec, K. Rezai,
N. B. Berkowitz, A. Buchbinder, American Association for Cancer Research
(AACR) 102nd Annual Meeting, April 2–6, 2011, (Orlando, FL, USA), Ab-
stract LB-407.
Received: September 6, 2013
Revised: November 8, 2013
[6] a) H. S. Ban, Y. Uto, H. Nakamura, Expert Opin. Ther. Pat. 2011, 21, 131–
146; b) Y. Xia, H.-K. Choi, K. Lee, Eur. J. Med. Chem. 2012, 49, 24–40;
Published online on November 27, 2013
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 61 – 66 66