Novel pyrazolopyridine derivatives as potential angiogenesis inhibitors: Synthesis, biological evaluation and transcriptome-based mechanistic analysis

Article abstract of DOI:10.1016/j.ejmech.2016.05.035

Modified purine derivatives exemplified by pyrazolopyrimidines have emerged as highly selective inhibitors of several angiogenic receptor tyrosine kinases. Herein, we designed and synthesized a new series of substituted pyrazolopyridines and explored their ability to influence crucial pro-angiogenic attributes of endothelial cells. Four of the synthesized compounds, possessing analogous substitution pattern, were found able to inhibit at low micromolar concentrations endothelial cell proliferation, migration and differentiation, constitutively or in response to Vascular Endothelial Growth Factor (VEGF) and to attenuate VEGF-induced phosphorylation of VEGF receptor-2 and downstream kinases AKT and ERK1/2. Administration of effective compounds in mice delayed the growth of syngeneic Lewis lung carcinoma transplants and reduced tumor microvessel density, without causing toxicity. Genome-wide microarray and gene ontology analyses of treated endothelial cells revealed derivative 18c as the most efficient modulator of gene expression and mitotic cell cycle/cell divisiong along with ? cholesterol biosynthesis? as the most significantly altered biological processes.

Full text of DOI:10.1016/j.ejmech.2016.05.035

European Journal of Medicinal Chemistry 121 (2016) 143e157
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Novel pyrazolopyridine derivatives as potential angiogenesis
inhibitors: Synthesis, biological evaluation and transcriptome-based
mechanistic analysis
Maria Michailidou a, Vassiliki Giannouli b, Vasilios Kotsikoris a, Olga Papadodima c,
Georgia Kontogianni c, Ioannis K. Kostakis b, Nikolaos Lougiakis b,
,
**
Aristotelis Chatziioannou c, Fragiskos N. Kolisis d, Panagiotis Marakos b, Nicole Pouli b
,
,
*
Heleni Loutrari a
a GP Livanos and M Simou Laboratories, 1st Department of Critical Care Medicine & Pulmonary Services, Evangelismos Hospital, Faculty of Medicine,
National and Kapodistrian University of Athens, 3 Ploutarchou St., 106 75, Athens, Greece
b Department of Pharmaceutical Chemistry, Faculty of Pharmacy, National and Kapodistrian University of Athens, Panepistimiopolis-Zografou, Athens,
15771, Greece
c Metabolic Engineering and Bioinformatics Group, Institute of Biology, Medicinal Chemistry and Biotechnology, National Hellenic Research Foundation, 48
Vassileos Constantinou Ave., 116 35, Athens, Greece
d School of Chemical Engineering, National Technical University of Athens, 9 Iroon Polytechneiou, Zografou, 157 80, Athens, Greece
a r t i c l e i n f o
a b s t r a c t
Article history:
Modified purine derivatives exemplified by pyrazolopyrimidines have emerged as highly selective in-
hibitors of several angiogenic receptor tyrosine kinases. Herein, we designed and synthesized a new
series of substituted pyrazolopyridines and explored their ability to influence crucial pro-angiogenic
attributes of endothelial cells. Four of the synthesized compounds, possessing analogous substitution
pattern, were found able to inhibit at low micromolar concentrations endothelial cell proliferation,
migration and differentiation, constitutively or in response to Vascular Endothelial Growth Factor (VEGF)
and to attenuate VEGF-induced phosphorylation of VEGF receptor-2 and downstream kinases AKT and
ERK1/2. Administration of effective compounds in mice delayed the growth of syngeneic Lewis lung
carcinoma transplants and reduced tumor microvessel density, without causing toxicity. Genome-wide
microarray and gene ontology analyses of treated endothelial cells revealed derivative 18c as the most
efficient modulator of gene expression and “mitotic cell cycle/cell division” along with “cholesterol
biosynthesis” as the most significantly altered biological processes.
Received 13 April 2016
Received in revised form
15 May 2016
Accepted 18 May 2016
Available online 20 May 2016
Keywords:
Pyrazolo[3,4-c]pyridines
Receptor tyrosine kinase inhibitors
Angiogenesis
Vascular endothelial growth factor
Microarray
Gene ontology
© 2016 Elsevier Masson SAS. All rights reserved.
Abbreviations: VEGF, Vascular Endothelial Growth Factor; VEGFR, Vascular Endothelial Growth Factor receptor(s); EC, endothelial cells; RTK, receptor Tyr kinase; FDA,
United States Food and Drug Administration; HUVEC, human umbilical vein endothelial cells; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; LLC, Lewis
lung carcinoma; GO, gene ontology; qRT-PCR, real-time quantitative reverse PCR; HMGCS, hydroxy-3-methylglutaryl-CoA synthase 1; MSMO1, methylsterol monooxygenase;
INSIG1, insulin induced gene 1; SQLE, squalene epoxidase; UBE2C, ubiquitin-conjugating enzyme E2C; CCNB2, cyclin B2; CCNA2, cyclin A2; CDC20, cell division cycle 20;
HMOX1, heme oxygenase 1; Pddba, bis(dibenzylideneacetone)palladium; X-Phos, 2-dicyclohexylphosphino-20,40,60-triisopropylbiphenyl; Et2O, diethyl ether; NIS, N-
iodosuccinimide.
* Corresponding author.
** Corresponding author.
E-mail addresses: mmichail@med.uoa.gr (M. Michailidou), v_giannouli@yahoo.gr (V. Giannouli), vasiliskotsikoris@yahoo.gr (V. Kotsikoris), opapadod@eie.gr
(O. Papadodima), gkontogianni@eie.gr (G. Kontogianni), ikkostakis@pharm.uoa.gr (I.K. Kostakis), nlougiak@pharm.uoa.gr (N. Lougiakis), achatzi@eie.gr (A. Chatziioannou),
kolisis@chemeng.ntua.gr (F.N. Kolisis), marakos@pharm.uoa.gr (P. Marakos), pouli@pharm.uoa.gr (N. Pouli), elloutrar@med.uoa.gr (H. Loutrari).
http://dx.doi.org/10.1016/j.ejmech.2016.05.035
0223-5234/© 2016 Elsevier Masson SAS. All rights reserved.

144
M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
1. Introduction
for the complete investigation of the bioactivity and the discovery
of new/modified derivatives with improved properties.
Angiogenesis, the formation of new blood vessels from pre-
existing vasculature, has become an important therapeutic target
because of its crucial role in several pathologies, characteristically
in tumor progression and metastasis [1]. As a result the discovery of
new compounds possessing potent anti-angiogenic activity has
been the focus of intense current research [2e6]. Numerous growth
factors are involved in tumor angiogenesis such as Vascular
Endothelial Growth Factor (VEGF) [7], epidermal growth factor [8],
platelet-derived growth factor [9], and basic fibroblast growth
factor [10]. Yet, the members of VEGF family including VEGF-A, B, C,
D, E and placenta growth factor are considered to be the most
critical endothelium-specific mediators of angiogenesis [7]. VEGF
exerts its biological effects by binding to and activating VEGF re-
ceptors (VEGFR) which are almost exclusively expressed in endo-
thelial cells (EC). There are three types of VEGFR, namely, VEGFR1
(Flt-1), VEGFR2 (KDR/Flk-1) and VEGFR3. Among them, VEGFR2
plays a major role in VEGF-A-mediated vascular cell biology [11,12].
Binding of VEGF-A to VEGFR2 results in the activation of several
signal transduction events. These involve receptor dimerization
and autophosphorylation at specific Tyr residues followed by
phosphorylation of downstream proteins including PKC, phos-
As part of our involvement in the design and synthesis of new
potentially bioactive purine analogues [29e31], we report here the
synthesis and evaluation of a number of novel pyrazolo[3,4-c]pyr-
idine derivatives as angiogenesis inhibitors. To this end we initially
examined the ability of this class of compounds to affect essential
pro-angiogenic functions of EC, constitutively or upon stimulation
with VEGF in cell-based assays and next to modulate tumor
angiogenesis and growth in a mouse cancer model. Finally, we
analyzed the global effect of best candidates on EC transcriptome in
an effort to uncover potential action mechanisms. Our study may
provide a basis for further evaluation of the most promising de-
rivative(s) in anti-angiogenic therapeutic interventions, especially
for cancer treatment.
2. Results and discussion
2.1. Chemistry
The target compounds were prepared using 2-amino-5-nitro-4-
picoline (1, Scheme 1) [32] as starting material. This compound was
diazotized and the resulting pyridinone 2 [33] was treated with
phosphorus oxychloride and converted to the chloropicoline 3 [34].
The nitroderivative 3 was then reduced using stannous chloride
as the reducing agent to give the aminopyridine 4 which was
acetylated to the acetamide 5. This acetamide was then heated at
reflux in benzene with isoamyl nitrite, in the presence of acetic
anhydride [35,36]. This furnished a mixture of the corresponding 1-
and 2-acetylpyrazolo[3,4-c]pyridines through a rearrangement of
the intermediate N-nitroso compound. These isomers were not
isolated, but the acetyl groups were easily cleaved upon treatment
with methanolic ammonia to provide the pyrazolopyridine 6 [36].
Compound 6 was subsequently treated with 4-methoxybenzyl
chloride in the presence of sodium hydride to provide derivative
7, together with an amount of the corresponding N2 regio-isomer.
From a brief study of the reaction conditions we resulted that the
use of a polar solvent (DMF) at room temperature favors the for-
mation of N1 over N2 isomer, which was thus prepared in 2:1 ratio.
Both isomers were separated and identified using NOE spectro-
scopic data. More precisely, in the case of the N1-isomer we
recorded a clear cross-coupling of both the benzylic and o-phenyl
protons with H-7. Compound 7 was then converted to the corre-
sponding N-oxide 8, using m-CPBA as oxidizing agent. The rear-
rangement of the N-oxide in the presence of phosphorous
oxychloride, produced the 5,7-dichloropyrazolopyridine 9, which
was used for the nucleophilic substitution of the 7-chloro group
using suitable secondary amines, in order to provide compounds
10a-c. The chlorides 10 were converted to the target derivatives
11a-d through a Suzuki-type coupling with aniline or 4-(4-
methylpiperazin-1-yl)aniline, in the presence of cessium carbon-
ate, using bis(dibenzylideneacetone)palladium (Pddba) [37] as
catalyst and 2-dicyclohexylphosphino-20,40,60-triisopropylbiphenyl
(X-Phos) as ligand [38]. 1H and 13C NMR spectra of target com-
pounds 11a-d are provided in Supplementary data.
pholipase Cg, PI3K, 3-phosphoinositide-dependent kinase 1, AKT,
focal adhesion kinase, ERK1/2, Src and p38MAPK [13e16]. More
specifically, activation of PI3KeAKT cascade by VEGFR2 has been
shown to play a causal role in VEGF-stimulated survival, migration
and sprouting of endothelial cells in vitro and angiogenesis in vivo,
whereas VEGF-induced ERK1/2 and p38MAPK signaling have been
mostly associated with regulation of EC proliferation and motility,
respectively [11,12]. Consequently, VEGF and VEGFR2 signaling
system is recognized as an attractive target for anti-angiogenic
intervention and a variety of approaches are currently being
assessed in cancer clinical trials. These include the use of soluble
VEGFR, monoclonal antibodies against VEGF or VEGFR and small
molecule inhibitors of VEGFR Tyr kinase activity [17e19].
In recent years targeted inhibition of several oncogenic receptor
Tyr kinases (RTK) by small molecules that compete with the ATP for
binding in the catalytic domain has been introduced as a systemic
treatment strategy for cancer. Actually, an array of RTK inhibitors
are being studied in clinical trials whereas several of them such as
sorafenib and sunitinib (Chart 1) have already been approved by
the United States Food and Drug Administration (FDA) for indica-
tion of specific types of cancer [20]. Nevertheless limitations in the
use of these targeted therapies, mostly associated with clinical
resistance and toxicity [21], prompt the need for developing new
drug candidates with improved bioactivity and a more favorable
safety profile. Pyrazolo[3,4-d]pyrimidines through their selectivity
toward different RTK have proved to be a very promising chemical
class in the discovery of new lead compounds to treat cancer and
pathological angiogenesis [22,23].
Focusing on VEGFR inhibitors, a great number of ATP site-
targeted ligands have been produced using either analogue syn-
thesis approaches or a combination of more sophisticated methods,
including structure-based drug design and fragment-based strate-
gies. Among them, some disubstituted 6-aminopurines [24] and
trisubstituted pyrazolo[4,3-d]pyrimidines [25] inhibit tumor
angiogenesis and cell migration, whereas benzothiopyrano[4,3-d]
pyrimidines [5] and 3-aminopyrazolo[3,4-b]pyridine ureas [26]
target kinases of the VEGF pathway. Closely related to the later,
pyrrolo[3,2-d]pyrimidine derivatives display strong inhibitory ac-
tivities against both VEGFR and fibroblast growth factor receptor
kinases [27] and possess anti-proliferative activity against VEGF-
stimulated human umbilical vein endothelial cells (HUVEC) [28].
In this respect the study of bioisosters provides an important tool
The corresponding 3-phenyl analogues (18a-c, Scheme 2), were
prepared from the intermediate chloride 6, which was successively
converted to the iodide 12 upon treatment with N-iodosuccinimide
and then to the 4-methoxybenzyl derivative 13 [39]. This iodide
provided the 3-phenylanalogue 14, through a Suzuki-type coupling
using phenylboronic acid in the presence of tetrakis(-
triphenylphosphine)palladium (0) and sodium bicarbonate.
Following a synthetic procedure analogous to the above mentioned
one, analogue 14 was converted to the target derivatives 18a-c. 1H
and 13C NMR spectra of target compounds 18a-c are provided in

M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
145
Chart 1. Representative structures of FDA approved anti-angiogenic drugs.
Scheme 1. Reagents and conditions: a) NaNO2, H2SO4, H2O; b) POCl3, 110 ꢀC; c) SnCl2$2H2O, HCl(c.), 55 ꢀC; d) Ac2O, CH2Cl2, r.t.; e) i) AcOK, Ac2O, isoamyl nitrite, benzene, reflux, ii)
NH3(g.), MeOH, r.t.; f) i) NaH, DMF, r.t., ii) 4-methoxybenzyl chloride, DMF, r.t.; g) m-CPBA, CHCl3, r.t.; h) POCl3, THF, r.t.; i) piperazine (for 10a) or pyrrolidine (for 10c), microwave
irradiation, 300 W, 160 ꢀC, or N-methylpiperazine (for 10b), reflux; j) aniline (for 11a, 11b) or 4-(4-methylpiperazin-1-yl)aniline (for 11c, 11d), X-Phos, Pddba, CsCO3, toluene, reflux.

146
M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
Scheme 2. Reagents and conditions: a) NIS, MeOH, r.t.; b) i) NaH, DMF, r.t., ii) 4-methoxybenzyl chloride, DMF, r.t.; c) phenylboronic acid, Pd(PPh3)4, NaHCO3, toluene/ethanol/H2O
(10/1/0.2), reflux; d) m-CPBA, CHCl3, r.t.; e) POCl3, THF, r.t.; f) piperazine (for 17a) or N-methylpiperazine (for 17b), microwave irradiation, 300 W, 160 ꢀC; g) aniline (for 18a) or 4-(4-
methylpiperazin-1-yl)aniline (for 18b, 18c), X-Phos, Pddba, CsCO3, toluene, reflux.
Supplementary data.
proliferation assay that is based on DNA content (not shown).
Furthermore, all four compounds were found to drastically
decrease VEGF-165-induced HUVEC chemotactic migration in
Transwell assays (Fig. 2B) and abolish the stimulatory action of
VEGF-165 on HUVEC characteristic ability to differentiate into
capillary-like structures when cultured on a Matrigel-coated sur-
face (Fig. 2C).
2.2. Biological evaluation
2.2.1. Cellular assays and angiogenesis models
The newly synthesized pyrazolopyridines 10a, 11a-d, 17a and
18a-c were initially screened for their ability to modulate the
proliferative and migratory competence of EC, at two representa-
Because VEGFR2 is the main mediator of VEGF-A-triggered
endothelial growth, motility and functional differentiation [11,12]
these results pointed to the hypothesis that in VEGF-A-induced
HUVEC test pyrazolopyridines exert their inhibitory actions
mainly by targeting RTK activity of VEGFR2 in HUVEC. However
interference of these derivatives with other endothelial RTK cannot
be certainly excluded. As a matter of fact in preliminary in vitro data
using a commercially available assay kit we found that the selected
compounds were able to significantly suppress the total protein Tyr
kinase activity in treated HUVEC, though they displayed a quite
different effect within the examined range of concentrations
(Supplementary data, Fig. S1). In order to directly evaluate our
hypothesis about VEGFR2 targeting, we subsequently assessed the
effect of active compounds on the extent of VEGFR2 autophos-
phorylation at key Tyr residues (Tyr951, Tyr996, Tyr1059 and
Tyr1175) after stimulation of HUVEC with VEGF-165. As shown in
Fig. 3A all test compounds significantly inhibited the relative
phosphorylation levels of Tyr1175, known to be essential for the
activation of VEGFR2 signal transduction and subsequent control of
endothelial functions [13,15], without affecting the amount of total
VEGFR2 in cell lysates. No significant changes were observed in the
relative phosphorylation levels of Tyr951, Tyr996 (with the excep-
tion of 18b) and Tyr1059. To further investigate the consequences
of the observed inhibitory effect on VEGFR2 signal transduction, we
next examined whether these compounds could modulate the
relative phosphorylation of three key downstream substrates,
namely ERK1/2, AKT and p38MAPK. As shown in Fig. 3B compound
tive concentrations (0.2 and 2 mM), under non-induced conditions
i.e. in response to culture medium supplemented with a low con-
tent of fetal bovine serum (2.5e5%). As shown in Fig. 1 from the
nine compounds tested, only the 3-phenylsubstituted derivatives
17a and 18a-c were found able to significantly inhibit both the EC
growth (A) and migration (B) in the applied in vitro assays, indi-
cating that 3-phenyl group substitution is required for anti-
angiogenic activity. Thus, these four derivatives were subse-
quently chosen for a more detailed evaluation of their anti-
angiogenic actions and underlying mechanisms. IC50 values for
cytostatic activity of the best inhibitors determined under similar
non-induced conditions were found to be in the low micromolar
range (Table 1).
Next, in order to explore whether the four selected anti-
angiogenic pyrazolopyridines could specifically attenuate the
VEGF-A-driven activation of pro-angiogenic endothelial pheno-
types, we examined the capability of compound-treated EC to
proliferate, migrate, and develop tube-like networks in response to
exogenously added VEGF-165, the most abundant variant of VEGF-
A. Cell growth studies conducted by the MTT method (Fig. 2A)
demonstrated that derivatives 17a, 18a and 18c were able to
significantly impair the stimulatory effect of VEGF-165 on HUVEC
growth, thus indicating an inhibitory action in cell proliferation
and/or survival. However compound 18c proved to be the most
effective inhibitor in the examined range of concentrations
(0.5e2.0 m
M). Similar results were obtained using CyQUANT® cell

M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
147
Fig. 1. Screening for angiogenesis modulators. (A) Effects of compounds on EC proliferation. EA.hy926 cells were treated with compound or DMSO (CTL) in medium containing 2.5%
FBS for 48 h. Data represent mean % of cell number relative to CTL SEM of replicates from two independent experiments (*p < 0.05, **p < 0.01, ***p < 0.001). (B) Effects of
compounds on EC migration. HUVEC pre-treated with compound or DMSO (CTL) were allowed to migrate towards medium containing 5% FBS or starvation medium (Basal). Graphs
represent mean % of migrating cells relative to CTL SEM from replicates of two independent experiments (*p < 0.05, **p < 0.01,***p < 0.001). In the lower panel, representative
microphotographs of migrating cells (magnification ꢁ200; scale bar: 100
mm).
Table 1
IC50 values (
potential of the four selected inhibitors in vivo, using a well-
characterized late-stage preclinical model based on Lewis lung
carcinoma (LLC) cell transplantation into immunocompetent mice.
Administration of test compounds into tumor-bearing C57BL/6
mice significantly reduced the rate of tumor growth, with the
exception of derivative 18a that was less effective (Fig. 4A). None of
the compounds created signs of toxicity as assessed by the clinical
examination and haematoxylin-eosin staining of tissue sections
from vital organs (not shown). Furthermore, IHC analysis of tumor
sections using CD31-specific antibody staining showed a concom-
itant significant decrease of microvessel density in tissues from
mice treated with active compounds as compared to tumors from
vehicle-treated animals (Fig. 4B). Collectively, derivatives 17a, 18b
and 18c proved to be capable of delaying LLC tumor growth in mice,
to some extent, through a remarkable inhibition of tumor vascu-
larization, in good correlation with our preceding in vitro data and
were selected for further investigation.
m
M) for anti-proliferating activitya of compounds 17a and 18a-c.
17a
18a
18b
18c
7.2 2.4
4.3 0.3
14.1 0.9
12.0 0.6
a
EC were treated with test compounds at
a
range of concentrations
(0.125e50 mM). Cell proliferation was assessed by the 3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyltetrazolium bromide (MTT) method. Values represent mean SEM.
treatment resulted in a marked inhibition of VEGF-165-induced
phosphorylation of AKT and ERK1/2, known to promote endothe-
lial cell survival, migration and proliferation [11], but did not affect
the phosphorylation of p38MAPK. Overall, these data provide direct
evidence that in VEGF-simulated EC test compounds efficiently
impede RTK activity of VEGFR2 consequently leading to distur-
bance of activation of ERK1/2 and AKT signaling and subsequent
inhibition of related angiogenic functions.
Finally, we examined the anti-tumor and anti-angiogenic

148
M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
Fig. 2. Effects of best inhibitors on VEGF-induced EC functions. (A) EC proliferation. HUVEC were treated with compound or DMSO vehicle in starvation medium containing or not
VEGF-165 for 48 h. Data represent mean % of cell number relative to control (vehicle in starvation medium, CTL) SEM of replicates from at least two independent experiments
(*p < 0.05, **p < 0.01; ***p < 0.001 in relation to vehicle-VEGF-165). (B) EC migration. HUVEC pre-treated with compound or DMSO were allowed to migrate towards medium
supplemented with VEGF-165 or starvation medium (CTL). Data represent the mean % of migrated cells relative to CTL SEM of replicates from at least two independent ex-
periments (***p < 0.001 in relation to vehicle-VEGF-165). (C) EC differentiation. HUVEC cultured on Matrigel-coated surfaces were treated with compound or DMSO in the presence
or not of VEGF-165 for 24 h. Graphs represent mean % of tube number relative to control (vehicle in medium, CTL) SEM of replicates from two independent experiments (*p < 0.05;
**p < 0.01; ***p < 0.001 in relation to vehicle-VEGF-165). On the right panel, representative microphotographs of tube-like networks (magnification ꢁ100; scale bar: 200
mm).
2.2.2. Genome-wide microarray and gene ontology (GO) analyses
In an effort to explore the mechanisms underlying the pheno-
typic changes imposed by the best three candidates, we next
examined by genome-wide microarrays the differential gene
expression profile of treated EC at two different time points, 6 h and
24 h.
As demonstrated in Table 2, derivative 18c had the most pro-
nounced effect on gene expression affecting a total number of 256
genes (12 were common in both time points), followed by 17a and
18b (33 and 17 altered genes, respectively). Complete lists of
significantly modified genes, are provided in Supplementary data
(Tables S1, S2 and S3) while the entire microarray dataset has been
deposited in NCBI’s Gene Expression Omnibus (GEO) and is acces-
sible through GEO accession number GSE75588.
We next analyzed the union of differentiated probes, meaning
those that were significant at least once, by hierarchical clustering.
As shown in the heat map representation of gene expression fold-
change values (Fig. 5), there is a significant cluster of under-
expressed genes common in all three compounds following 24 h
of exposure. Moreover, pathway analysis, exploiting the functional
annotations of GO, revealed that this cluster of altered genes affects
cellular functions pertinent to mitosis and cell division (“mitotic
cell cycle”, “cell division”, “mitotic nuclear division”, etc), suggest-
ing a down-regulating effect on cell proliferation, in concurrence
with our preceding experimental results (Figs. 1A and 2A).
We then focused on the dataset from cells treated with 18c for
24 h, because it enclosed the highest number of differentially
expressed genes and performed a detailed functional analysis in
order to gain insight into the affected molecular processes. Fig. 6
depicts bar plots of the top enriched processes with the amount
of differentially expressed genes per GO term. The biological pro-
cesses that were mostly deregulated by 18c fall into two basic
categories: a. those that, consistent with our experimental data,
were related to mitosis and included in vast majority under-
expressed genes, such as “mitotic cell cycle” involving 39 genes
(Supplementary data, Table S4) and b. those generally related to
cholesterol and lipid metabolism which contained almost exclu-
sively over-expressed genes, such as “cholesterol biosynthetic
process” involving 17 genes (Supplementary data, Table S5).
Although additional mechanistic studies are certainly required, this
finding corroborates the acquired phenotypic data since cholesterol
homeostasis and in general endothelial cell metabolism have been

M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
149
Fig. 3. Effects of 17a and 18a-c on VEGF-induced phospho-signaling. Western blot analysis of protein extracts from HUVEC pre-treated with 2
mM of test compound or DMSO vehicle
for 2 h and then exposed to VEGF-165 or vehicle containing medium (CTL) for 5 min. (A) Detection of phosphorylated VEGFR2 at indicated Tyr residues. (B) Detection of phos-
phorylated AKT, ERK1/2 and p38MAPK. Blots shown are representative of at least three independent experiments. Graphs are mean data from densitometry analysis and
demonstrate relatively changed phosphorylation levels of shown protein by test compound when the level of VEGF-induced protein phosphorylation in the absence of compound,
was normalized as 1.0. All phosphoprotein signals had been normalized to their respective total proteins. Data are mean values SEM (*p < 0.05; **p < 0.01; ***p < 0.001, n ¼ 3).
Fig. 4. In vivo effects of 17a and 18a-c on LLC tumor growth and vascularization. (A) Tumor volume was evaluated in mice injected with test compounds (0.2 mg/kg of body weight)
or DMSO vehicle (CTL). (B) Microvessel density of LLC tumors at end-point as assessed by IHC using CD31 specific antibody. Data represent mean SEM relative to control group
(n ¼ 10e17; *p < 0.05). In the lower panel, representative microphotographs (magnification ꢁ400; scale bar: 50
mm).

150
M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
Table 2
(HMOX1). As shown in Fig. 7 for all genes tested the direction and
magnitude of changes measured by qRT-PCR were consistent with
the results obtained from the microarray analysis.
Differentially expressed genes by pyrazolopyridines 17a, 18b-c.a
6 h
24 h
Under-expressed Over-expressed Under-expressed Over-expressed
3. Conclusion
17a
18b
4
2
10
1
17
7
2
7
18c 14
42
112
100
A series of novel pyrazolopyridines were synthesized following
a procedure involving the use of a suitable aminopicoline, which
upon chemical modifications was ring closed to result in a central
pyrazolo[3,4-c]pyridine core, that was successively substituted to
provide the target derivatives. Nine compounds were initially
evaluated for their activity as inhibitors of EC growth and migration
under basal conditions. Subsequently, the anti-angiogenic effects of
the best candidates, namely 17a and 18a-c, against EC proliferation,
migration and differentiation were further demonstrated under
VEGF-stimulated conditions. At the level of cell signal transduction
the selected inhibitors attenuated VEGF-induced phosphorylation
of Tyr-1175 in RTK domain of VEGFR2 and accordingly negatively
regulated the VEGF-mediated phosphorylation of downstream ki-
nases AKT and ERK1/2. Moreover, derivatives 17a and 18b-c proved
to be able to cause in vivo a significant regression of LLC growth and
a prominent reduction of tumor microvessel density, without
obvious side-effects. Genome-wide microarray profiling and sub-
sequent GO analysis of EC in response to 17a and 18b-c identified a
commonly targeted cluster of under-expressed genes related to cell
division and mitosis and revealed 18c as the most powerful de-
rivative in inducing gene expression alterations. Additionally, in
18c-treated EC, cholesterol biosynthesis and lipid metabolism were
recognized amongst the top deregulated cellular processes with
relevant GO categories mainly comprising over-expressed genes. As
a
Cut-off: p-value < 0.05 & |log2FC| > 0.5.
recently shown to critically impact the angiogenic process [40e43].
Involvement of VEGFR2 signaling into the observed metabolic
changes is also speculated as this pathway has been previously
connected with the activation of sterol regulatory element binding
proteins (transcription regulators of lipid biosynthesis), an event
that is proved to be indispensable for the unobstructed progression
of VEGFA-induced angiogenesis [44]. In conclusion, the imposed
changes on the gene expression profile of treated HUVEC indicate
dysregulation of cell cycle and cholesterol dynamics as two sig-
nificant mechanisms underlying the angiogenesis modulating ac-
tion of derivative 18c. Subsequently, a number of genes, involved in
these top deregulated cellular functions (see Supplementary data
Tables S4 and S5) were selected to independently confirm, by real-
time quantitative reverse PCR (qRT-PCR), the transcriptional
changes observed in the microarray experiment. The genes chosen
were those encoding for 3-hydroxy-3-methylglutaryl-CoA synthase
1
(HMGCS1), methylsterol monooxygenase (MSMO1), insulin
induced gene 1 (INSIG1) squalene epoxidase (SQLE), ubiquitin-
conjugating enzyme E2C (UBE2C), cyclin B2 (CCNB2), cyclin A2
(CCNA2), cell division cycle 20 (CDC20) and heme oxygenase 1
Fig. 5. Hierarchical clustering of the gene expression fold change values (in log2 scale) in HUVEC treated with compounds 17a, 18b and 18c versus DMSO for 6 h and 24 h. The
bracket indicates the top enriched GO terms as indicated by the StRAnGER2 analysis, affected by the specific cluster of under-expressed genes common for all compounds at 24 h.

M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
151
Fig. 7. Validation of microarray data using qRT-PCR in HUVEC treated with compound
18c or DMSO (control) for 24 h. Relative expression for each gene is shown as fold
change versus control. qRT-PCR data represent mean SEM (n ¼ 3; *p < 0.05).
4. Experimental procedures
4.1. Chemistry
4.1.1. General information
Melting points were determined on a Büchi apparatus and are
uncorrected. FT-IR spectra were recorded on a Perkin-Elmer RX1
FT-IR spectrometer. 1H NMR spectra and 2D spectra were recorded
on a Bruker Avance III 600 or a Bruker Avance DRX 400 instrument,
whereas 13C NMR spectra were recorded on a Bruker Avance III 600
or a Bruker AC 200 spectrometer in deuterated solvents and were
referenced to TMS (d
scale). The signals of 1H and 13C spectra were
unambiguously assigned by using 2D NMR techniques: 1H1H COSY,
NOESY, HMQC, and HMBC. Mass spectra were recorded with a LTQ
Orbitrap Discovery instrument, possessing an Ionmax ionization
source. Flash chromatography was performed on Merck silica gel 60
(0.040e0.063 mm). Analytical thin layer chromatography (TLC) was
carried out on precoated (0.25 mm) Merck silica gel F-254 plates.
The purity of all the synthesized compounds was >95% as ascer-
tained by elemental analysis. Elemental analyses were undertaken
using a PerkinElmer PE 240C elemental analyzer (Norwalk, CT, U.S.)
and the measured values for C, H, and N were within 0.4% of the
theoretical values. Stock solutions of compounds for biological ex-
periments were prepared in DMSO (St. Louis, MO, USA) and stored
at ꢂ20 ꢀC. Working dilutions contained up to 0.1% v/v DMSO.
Fig. 6. Bar plot diagram of differentially expressed genes per GO term in HUVEC
treated with compound 18c versus DMSO for 24 h. Functional analysis performed
using StRAnGER2 depicts the statistically significant altered biological processes.
far as structure-activity relationships are concerned, a thorough
consideration of acquired biological data allowed a preliminary
estimation of optimal structural requirements for anti-angiogenic
activity. It is noticeable that the best three inhibitors possess a 3-
phenyl group, whereas two of the most interesting analogues
(18b and 18c) are endowed with a quite similar substitution
pattern. Overall, in this study we present convincing preclinical
evidence and mechanistic insights into the action of a new type of
anti-angiogenic pyrazolopyridines. Our results encourage future
assessment of the most promising derivative, 18c, as a lead for the
development of novel agents against angiogenesis-dependent
disorders. To this end a detailed structure-activity relationship
analysis of new synthesized analogues along with estimation of
bioavailability, pharmacokinetics and toxicity profiles would be
definitely required.
4.1.2. Synthesis of 5-chloro-1-(4-methoxybenzyl)-1H-pyrazolo[3,4-
c]pyridine (7)
Sodium hydride (260 mg, 6.52 mmol, 60% in paraffin oil) was
added at 0 ꢀC, under argon to a solution of the pyrazolopyridine 6
[36] (500 mg, 3.26 mmol) in dry DMF (20 mL) and the reaction
mixture was stirred at room temperature for 30 min. The reaction
was then cooled to 0 ꢀC, a solution of 4-methoxybenzylchloride
(0.66 mL, 4.89 mmol) in DMF (2 mL) was added dropwise and
the mixture was stirred at room temperature for 2 h. The bulk of the
solvent was removed in vacuo, water was added to the residue and
it was extracted with dichloromethane (3 ꢁ 50 mL). The organic
extracts were dried (Na2SO4) and evaporated to dryness to provide
both N1 and N2 regio isomers which were separated by column
chromatography (silica gel) using a mixture of cyclohexane/ethyl
acetate (5/1, v/v) as the eluent.

152
M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
4.1.2.1. Data for N1-isomer: 5-chloro-1-(4-methoxybenzyl)-1H-pyr-
using a mixture of dichloromethane/methanol (from 100/5 up to
azolo[3,4-c]pyridine (7). Yield 57%. Mp 117e119 ꢀC (EtOAc). 1H NMR
100/12, v/v) as the eluent. Yield: 82%. Mp 77e79 ꢀC (CH2Cl2/Et2O).
(600 MHz, CDCl3)
d
3.79 (s, 3H, OCH3), 5.61 (s, 2H, CH2), 6.87 (d,
IR (Nujol)
n
3242, 1612, 1585, 1558, 1510, 1310, 1266, 1245, 1174,
2.96 (m, 4H,
J ¼ 8.5 Hz, 2H, H-30, H-50), 7.22 (d, J ¼ 8.5 Hz, 2H, H-20, H-60), 7.64 (s,
1085, 1028, 803, 732 cmꢂ1. 1H NMR (400 MHz, CDCl3)
d
1H, H-4), 8.03 (s, 1H, H-3), 8.62 (s, 1H, H-7). 13C NMR (151 MHz,
CDCl3) d 53.7, 55.3,114.4,115.2,127.4,129.0,131.1,132.0,133.2,135.4,
piperazine H), 3.14 (m, 4H, piperazine H), 3.59 (s, 3H, OCH3), 5.59 (s,
2H, CH2), 6.67 (d, 2H, J ¼ 8.6 Hz, H-30, H-50), 7.00 (d, 2H, J ¼ 8.6 Hz,
H-20, H-60), 7.11 (s, 1H, H-4), 7.85 (s, 1H, H-3). 13C NMR (50 MHz,
140.8, 159.7. HR-MS (ESI) m/z: Calcd for
C14H13N3OCl:
[M1 þ H]þ ¼ 274.0742, found 274.0746.
CDCl3) d 45.4, 49.8, 51.7, 55.0, 109.3, 113.9, 128.4, 129.0, 129.4, 133.0,
133.7, 138.4, 148.9, 159.0. HR-MS (ESI) m/z: Calcd for C18H21ClN5O:
4.1.2.2. Data for N2-isomer: 5-chloro-2-(4-methoxybenzyl)-2H-pyr-
[M1 þ H]þ
¼
358.1429, found 358.1434. Anal. Calcd for
azolo[3,4-c]pyridine. Yield 28%. Mp 92e94 ꢀC (EtOAc). 1H NMR
C18H20ClN5O: C, 60.42; H, 5.63; N, 19.57. Found: C, 60.74; H, 5.46; N,
(600 MHz, CDCl3)
d
3.81 (s, 3H, OCH3), 5.58 (s, 2H, CH2), 6.91 (d,
19.77.
J ¼ 8.5 Hz, 2H, H-30, H-50), 7.30 (d, J ¼ 8.5 Hz, 2H, H-20, H-60), 7.51 (s,
1H, H-4), 7.88 (s, 1H, H-3), 9.07 (s, 1H, H-7). 13C NMR (151 MHz,
4.1.6. Synthesis of 5-chloro-1-(4-methoxybenzyl)-7-(4-
CDCl3)
d
55.35, 57.99, 112.79, 114.56, 121.96, 126.42, 126.51, 129.96,
methylpiperazin-1-yl)-1H-pyrazolo[3,4-c]pyridine (10b)
140.41, 143.94, 144.58, 160.12.
A mixture of compound 9 (988 mg, 3.21 mmol) and 1-
methylpiperazine (3.2 g, 31.9 mmol) was refluxed for 20 min, un-
der argon. The mixture was then poured into water (200 mL) and
extracted with dichloromethane (3 ꢁ 100 mL). The crude product
was purified by column chromatography (silica gel) using a mixture
of dichloromethane/methanol (from 100/5 up to 100/10, v/v) as the
4.1.3. Synthesis of 5-chloro-1-(4-methoxybenzyl)-1H-pyrazolo[3,4-
c]pyridine-6-oxide (8)
3-Chloroperbenzoic acid (1.77 g, 10.24 mmol) was added to a
solution of pyrazolopyridine 7 (2 g, 7.31 mmol) in chloroform
(120 mL) and the reaction mixture was stirred at room tempera-
ture, away from light, for three days. Upon completion of the re-
action, the solvent was evaporated and a solution of sodium
carbonate (150 mL, 5% w/v) was added to the residue. The white
precipitate was filtered under vacuum, washed with a small
quantity of water (10 mL) and air-dried, to provide 1.9 g of pure N-
eluent to provide pure 10b as an oil. Yield: 89%. IR (film)
2921, 2860, 1602, 1558, 1514, 1459, 1249, 1167, 1085, 1028, 798 cmꢂ1
1H NMR (400 MHz, CDCl3)
2.28 (s, 3H, CH3-piperazine), 2.66 (m,
n 2955,
.
d
4H, piperazine H), 3.35 (m, 4H, piperazine H), 3.72 (s, 3H, OCH3),
5.69 (s, 2H, CH2), 6.79 (d, 2H, J ¼ 8.6 Hz, H-30, H-50), 7.15 (d, 2H,
J ¼ 8.6 Hz, H-20, H-60), 7.21 (s, 1H, H-4), 7.95 (s, 1H, H-3). 13C NMR
oxide 8. Yield 90%. Mp 156e8 ꢀC (EtOH). IR (Nujol)
1514, 1245, 1177, 1143, 1062, 1024, 797 cmꢂ1 1H NMR (400 MHz,
n
1612, 1585,
(50 MHz, CDCl3) d 46.1, 50.4, 53.0, 54.6, 55.0, 108.9, 113.8, 128.6,
.
129.0, 129.4, 132.9, 133.8, 138.4, 148.6, 158.9. HR-MS (ESI) m/z: Calcd
for C19H23ClN5O: [M1 þ H]þ ¼ 372.1586, found 372.1589. Anal.
Calcd for C19H22ClN5O: C, 61.37; H, 5.96; N, 18.83. Found: C, 61.45;
H, 5.99; N, 18.96.
methanol-d4)
d 3.80 (s, 3H, OCH3), 5.47 (s, 2H, CH2), 6.87 (d, 2H,
J ¼ 8.0 Hz, H-30, H-50), 7.20 (d, 2H, J ¼ 8.0 Hz, H-20, H-60), 7.83 (s, 1H,
H-4), 7.99 (s, 1H, H-3), 8.58 (s, 1H, H-7). 13C NMR (50 MHz, meth-
anol-d4)
d 52.3, 55.5, 114.4, 118.1, 120.3, 123.9, 128.8, 129.8, 133.7,
134.8,135.7,159.3. Anal. Calcd for C14H12ClN3O2: C, 58.04; H, 4.17; N,
14.50. Found: C, 57.96; H, 4.21; N, 14.43.
4.1.7. Synthesis of 5-chloro-1-(4-methoxybenzyl)-7-(pyrrolidin-1-
yl)-1H-pyrazolo[3,4-c]pyridine (10c)
This compound was prepared by a procedure analogous to that
of 10a. Purification was effected by column chromatography (silica
gel) using a mixture of cyclohexane/ethyl acetate (from 90/10 up to
70/30, v/v) as the eluent to provide pure 10c as an oil. Yield: 91%. IR
4.1.4. Synthesis of 5,7-dichloro-1-(4-methoxybenzyl)-1H-pyrazolo
[3,4-c]pyridine (9)
Phosphorous oxychloride (2.2 mL, 24.35 mmol) was added
dropwise, at 0 ꢀC, to a solution of the N-oxide 8 (1.5 g, 4.87 mmol) in
dry tetrahydrofuran (20 mL) and the reaction mixture was stirred at
room temperature for 12 h. Then, it was poured into crushed ice
(100 mL), made alkaline with a solution of sodium carbonate (5% w/
v) and extracted with dichloromethane (3 ꢁ 100 mL). The com-
bined organic layers were dried (Na2SO4) and evaporated to dry-
ness. The crude product was purified by column chromatography
(silica gel) using a mixture of cyclohexane/ethyl acetate (75/25, v/v)
as the eluent, to provide 1.4 g of the dichloro compound 9. Yield
(film)
n
2954, 2922, 2873, 1612, 1562, 1513, 1458, 1248, 1175, 1074,
1.95 (m, 4H, pyrroli-
1033, 797 cmꢂ1. 1H NMR (400 MHz, CDCl3)
d
dine H), 3.53 (m, 4H, pyrrolidine H), 3.74 (s, 3H, OCH3), 5.63 (s, 2H,
CH2), 6.78 (d, 2H, J ¼ 8.6 Hz, H-30, H-50), 7.07 (d, 2H, J ¼ 8.6 Hz, H-20,
H-60), 7.10 (s, 1H, H-4), 7.92 (s, 1H, H-3). 13C NMR (50 MHz, CDCl3)
d
24.8, 51.1, 53.9, 55.2, 107.0, 114.0, 128.7, 129.2, 130.2, 132.9, 134.0,
138.5, 148.1, 159.1. HR-MS (ESI) m/z: Calcd for C18H20ClN4O:
[M1 þ H]þ
¼
343.1320, found 343.1326. Anal. Calcd for
18H19ClN4O: C, 63.06; H, 5.59; N, 16.34. Found: C, 63.26; H, 5.70; N,
C
93%. Mp 101e3 ꢀC (Et2O/n-hexane). IR (Nujol)
n
1605, 1582, 1507,
3.79
16.16.
1245, 1174, 1096, 1028, 797 cmꢂ1. 1H NMR (600 MHz, CDCl3)
d
(s, 3H, OCH3), 5.92 (s, 2H, CH2), 6.85 (d, 2H, J ¼ 8.5 Hz, H-30, H-50),
4.1.8. General procedure for the synthesis of 5,7-bis substituted 1-
(4-methoxybenzyl)-1H-pyrazolo[3,4-c]pyridines 11aed
7.20 (d, 2H, J ¼ 8.5 Hz, H-20, H-60), 7.59 (s, 1H, H-4), 8.09 (s, 1H, H-3).
13C NMR (151 MHz, CDCl3)
d
53.9, 55.2, 114.0, 114.1, 128.6, 128.6,
Aniline (0.164 mL, 1.8 mmol) or 4-(4-methylpiperazin-1-yl)an-
iline (210 mg, 1.1 mmol) was added under argon to a solution of the
7-aminosubstituted-5-chloro-pyrazolopyridines 10aec (1 mmol)
in dry toluene (25 mL), followed by addition of cesium carbonate
(1.3 g, 4 mmol), 2-dicyclohexylphosphino-20,40,60-triisopropylbi-
phenyl (X-phos, 24 mg, 0.05 mmol) and bis(dibenzylideneacetone)
palladium(0) (29 mg, 0.05 mmol) and the resulting mixture was
refluxed for 18 h. Upon completion of reaction, the mixture was
filtered through a celite pad, the filtrate was vacuum-evaporated,
extracted with ethyl acetate e water and the extracts were dried
and evaporated to dryness. The crude product was then purified by
column chromatography (silica gel) to provide pure compounds
11aed.
132.3, 132.4, 132.7, 133.1, 138.6, 159.3. HR-MS (ESI) m/z: Calcd for
C
14H12Cl2N3O: [M1 þ H]þ ¼ 308.0352, found 308.0355. Anal. Calcd
for C14H11Cl2N3O: C, 54.57; H, 3.60; N, 13.64. Found: C, 54.43; H,
3.51; N, 13.79.
4.1.5. Synthesis of 5-chloro-1-(4-methoxybenzyl)-7-(piperazin-1-
yl)-1H-pyrazolo[3,4-c]pyridine (10a)
A mixture of compound 9 (200 mg, 0.65 mmol) and piperazine
(560 mg, 6.5 mmol) was irradiated at 160 ꢀC (300 W, Milestone
Start E) for 3 min. Upon cooling the mixture was poured into water
(100 mL) and extracted with dichloromethane (3 ꢁ 50 mL). The
crude product was purified by column chromatography (silica gel)

M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
153
4.1.8.1. 1-(4-Methoxybenzyl)-N-phenyl-7-(piperazin-1-yl)-1H-pyr-
azolo[3,4-c]pyridin-5-amine (11a). This compound was prepared
according to the general procedure described above. Purification
was effected using a mixture of dichloromethane/methanol (from
100/5 up to 100/14, v/v) as the eluent. Yield: 79%. Mp 131e132 ꢀC
piperazine H), 3.48 (m, 4H, pyrrolidine H), 3.71 (s, 3H, OCH3), 5.57
(s, 2H, CH2), 6.11 (brs, 1H, NH-aniline, D2O exch), 6.47 (s, 1H, H-4),
60.076 (d, 2H, J ¼ 8.7 Hz, H-30, H-50), 6.90 (d, 2H, J ¼ 8.9 Hz, H-300, H-
5 ), 7.11 (d, 2H, J ¼ 8.7 Hz, H-20, H-60), 7.18 (d, 2H, J ¼ 8.9 Hz, H-200, H-
600), 7.76 (s, 1H, H-3). 13C NMR (50 MHz, CDCl3)
d 24.6, 46.1, 49.9,
(EtOAc/n-pentane). IR (Nujol)
1232, 1170, 1033, 812, 730 cmꢂ1. 1H NMR (600 MHz, methanol-d4)
3.04 (m, 4H, piperazine H), 3.25 (m, 4H, piperazine H), 3.71 (s, 3H,
n
3264, 3169, 1608, 1571, 1518, 1250,
50.8, 53.6, 55.2, 87.3, 113.9, 117.4, 121.9, 127.3, 128.8, 129.9, 133.7,
133.8, 134.8, 146.8, 147.5, 148.8, 158.9. HR-MS (ESI) m/z: Calcd for
d
C
C
29H36N7O: [M1 þ H]þ ¼ 498.2976, found 498.2979. Anal. Calcd for
OCH3), 5.68 (s, 2H, CH2), 6.77 (s, 1H, H-4), 6.79 (d, 2H, J ¼ 8.6 Hz, H-
30, H-50), 6.84 (m, 1H, H-400), 7.07 (d, 2H, J ¼ 8.7 Hz, H-20, H-60), 7.22
(t, 2H, J ¼ 7.2 Hz, H-300, H-500), 7.41 (d, 2H, J ¼ 7.4 Hz, H-200, H-600), 7.89
29H35N7O: C, 69.99; H, 7.09; N, 19.70. Found: C, 70.23; H, 7.22; N,
19.61.
(s, 1H, H-3). 13C NMR (151 MHz, methanol-d4)
d
46.4, 52.6, 54.1,
4.1.9. Synthesis of 5-chloro-1-(4-methoxybenzyl)-3-phenyl-1H-
pyrazolo[3,4-c]pyridine (14)
55.8, 93.1, 115.0, 119.0, 121.3, 127.7, 129.5, 129.9, 131.5, 134.6, 135.6,
144.5, 149.5, 149.8, 160.7. HR-MS (ESI) m/z: Calcd for C24H27N6O:
[M1 þ H]þ ¼ 415.2241, found 415.2242. Anal. Calcd for C24H26N6O:
C, 69.54; H, 6.32; N, 20.27. Found: C, 69.77; H, 6.20; N, 20.53.
Phenylboronic acid (416 mg, 3.42 mmol) and palladium(0) tet-
rakis(triphenylphosphine) (180 mg, 0.15 mmol) were added to a
solution of the iodo compound 13 (1.24 g, 3.1 mmol) in a mixture of
toluene (100 mL) and ethanol (10 mL), followed by addition of an
aqueous solution (2 mL) of sodium hydrogen carbonate (780 mg,
9.30 mmol) and the resulting mixture was refluxed for 16 h. The
solvents were then vacuum-evaporated, the residue was extracted
with dichloromethane e water, and the combined extracts were
dried (Na2SO4) and evaporated to dryness. The crude mixture was
purified by column chromatography (silica gel), using a mixture of
dichloromethane/ethyl acetate (99/1, v/v) as the eluent, to provide
pure 14 (1.08 g, yield 100%). Mp 73e75 ꢀC (MeOH). 1H NMR
4.1.8.2. 1-(4-Methoxybenzyl)-7-(4-methylpiperazin-1-yl)-N-phenyl-
1H-pyrazolo[3,4-c]pyridin-5-amine (11b). This compound was pre-
pared according to the general procedure described above. Purifi-
cation was effected using a mixture of dichloromethane/methanol
(from 100/2 up to 100/8, v/v) as the eluent. Yield: 77%. Mp
117e118 ꢀC (Et2O/n-pentane). IR (Nujol)
1262, 1238, 1170, 1143, 1031, 1007, 848, 732 cmꢂ1
(600 MHz, CDCl3) 2.81 (s, 3H, CH3-piperazine), 3.14 (brs, 2H,
n
3167, 1599, 1565, 1510,
.
1H NMR
d
piperazine H), 3.37 (brs, 2H, piperazine H), 3.70 (brs, 2H, piperazine
H), 3.75 (s, 3H, OCH3), 3.99 (brs, 2H, piperazine H), 5.57 (s, 2H, CH2),
6.28 (brs, 1H, NH-aniline, D2O exch), 6.78 (d, 2H, J ¼ 8.6 Hz, H-30, H-
50), 6.87 (s, 1H, H-4), 7.00e7.07 (m, 3H, H-400, H-20, H-60), 7.22 (d, 2H,
J ¼ 7.6 Hz, H-200, H-600), 7.31 (t, 2H, J ¼ 7.6 Hz, H-300, H-500), 7.88 (s, 1H,
(400 MHz, CDCl3) d 3.71 (s, 3H, OCH3), 5.56 (s, 2H, CH2), 6.83 (d, 2H,
H-20, H-60, J ¼ 8.5 Hz), 7.23 (d, 2H, H-30, H-50, J ¼ 8.5 Hz), 7.40 (t, 1H,
H-400, J ¼ 7.5 Hz), 7.49 (t, 2H, H-300, H-500, J ¼ 7.5 Hz), 7.83 (s, 1H, H-4),
7.91 (d, 2H, H-200, H-600, J ¼ 7.5 Hz), 8.58 (s, 1H, H-7). 13C NMR
(50 MHz, CDCl3)
d 53.5, 55.0, 114.1, 114.5, 126.8, 127.3, 128.3, 128.4,
H-3). 13C NMR (50 MHz, CDCl3)
d
46.3, 50.5, 53.0, 55.0, 55.3, 90.9,
128.8, 129.0, 131.8, 133.3, 136.5, 141.1, 142.8, 159.4. HR-MS (ESI) m/z:
Calcd for C20H17ClN3O: [M1 þ H]þ ¼ 350.1055, found 350.1061.
Anal. Calcd for C20H16ClN3O: C, 68.67; H, 4.61; N, 12.01. Found: C,
68.44; H, 4.49; N, 11.87.
114.0, 118.5, 121.3, 126.8, 128.8, 129.2, 129.8, 133.7, 133.8, 142.2,
147.3, 148.2, 159.0. HR-MS (ESI) m/z: Calcd for
C25H29N6O:
[M1 þ H]þ ¼ 429.2397, found 429.2402. Anal. Calcd for C25H28N6O:
C, 70.07; H, 6.59; N, 19.61. Found: C, 69.98; H, 6.47; N, 19.69.
4.1.10. Synthesis of 5-chloro-1-(4-methoxybenzyl)-3-phenyl-1H-
pyrazolo[3,4-c]pyridine-6-oxide (15)
This compound was prepared by a procedure analogous to that
of 8. The crude product was purified by column chromatography
(silica gel), using a mixture of cyclohexane/ethyl acetate (from 80/
20 up to 0/100, v/v) as the eluent, to provide 15 (420 mg, yield 57%).
4.1.8.3. 1-(4-Methoxybenzyl)-7-(4-methylpiperazin-1-yl)-N-(4-(4-
methylpiperazin-1-yl)phenyl)-1H-pyrazolo[3,4-c]pyridin-5-amine
(11c). This compound was prepared according to the general pro-
cedure described above. Purification was effected using a mixture
of dichloromethane/methanol (from 100/1 up to 100/5, v/v) as the
eluent. Yield: 88%. Mp 136e138 ꢀC (Et2O/n-pentane). IR (Nujol)
3392, 3283, 1605, 1568, 1514, 1245, 1174, 1143, 1031, 1004, 817,
722 cmꢂ1 1H NMR (600 MHz, methanol-d4)
2.43 (s, 3H, CH3-
n
Mp 142e144 ꢀC (EtOH). IR (Nujol)
1136, 1068, 1028, 722 cmꢂ1. 1H NMR (400 MHz, CDCl3)
n
1609, 1585, 1514, 1249, 1174,
3.73 (s, 3H,
d
.
d
OCH3), 5.46 (s, 2H, CH2), 6.81 (d, 2H, H-20, H-60, J ¼ 8.7 Hz), 7.21 (d,
2H, H-30, H-50, J ¼ 8.7 Hz), 7.40 (t,1H, H-400, J ¼ 7.4 Hz), 7.48 (t, 2H, H-
300, H-500, J ¼ 7.4 Hz), 7.83 (d, 2H, H-200, H-600, J ¼ 7.8 Hz), 7.99 (s, 1H,
piperazine), 2.44 (s, 3H, CH3-piperazine), 2.75 (m, 8H, piperazine
H), 3.16 (m, 4H, piperazine H), 3.32 (m, 4H, piperazine H, over-
lapping with methanol-d4), 3.72 (s, 3H, OCH3), 5.65 (s, 2H, CH2),
6.66 (s, 1H, H-4), 6.79 (d, 2H, J ¼ 8.7 Hz, H-30, H-50), 6.95 (d, 2H,
J ¼ 8.9 Hz, H-300, H-500), 7.06 (d, 2H, J ¼ 8.6 Hz, H-20, H-60), 7.33 (d,
2H, J ¼ 8.6 Hz, H-200, H-600), 7.86 (s, 1H, H-3). 13C NMR (151 MHz,
H-4), 8.58 (s, 1H, H-7). 13C NMR (50 MHz, CDCl3)
d 53.9, 55.3, 114.5,
117.4, 119.1, 123.6, 126.7, 127.1, 129.0, 129.1, 129.2, 131.2, 136.1, 136.5,
144.0, 159.8. Anal. Calcd for C20H16ClN3O2: C, 65.67; H, 4.41; N,
11.49. Found: C, 65.86; H, 4.57; N, 11.12.
methanol-d4)
d 45.9, 46.2, 51.3, 51.4, 54.1, 55.8, 55.9, 56.1, 115.0,
115.1, 119.2, 121.2, 129.5, 129.7, 131.5, 134.5, 138.2, 146.8, 149.0, 160.7.
HR-MS (ESI) m/z: Calcd for C30H39N8O: [M1 þ H]þ ¼ 527.3241,
found 527.3243. Anal. Calcd for C30H38N8O: C, 68.41; H, 7.27; N,
21.28. Found: C, 68.30; H, 7.44; N, 21.49.
4.1.11. Synthesis of 5,7-dichloro-1-(4-methoxybenzyl)-3-phenyl-
1H-pyrazolo[3,4-c]pyridine (16)
This compound was prepared by a procedure analogous to that
of 9. The crude product was purified by column chromatography
(silica gel), using a mixture of cyclohexane/ethyl acetate (90/10, v/
v) as the eluent, to provide pure 16 (3.2 g, yield 84%). Mp
4.1.8.4. 1-(4-Methoxybenzyl)-N-(4-(4-methylpiperazin-1-yl)phenyl)-
7-(pyrrolidin-1-yl)-1H-pyrazolo[3,4-c]pyridin-5-amine
(11d).
126e127 ꢀC (Et2O/n-hexane). IR (Nujol)
1247, 1147, 1024, 722 cmꢂ1. 1H NMR (400 MHz, CDCl3)
n
1612, 1584, 1510, 1266,
3.76 (s, 3H,
This compound was prepared according to the general procedure
described above. Purification was effected using a mixture of
dichloromethane/methanol (from 100/2 up to 100/10, v/v) as the
d
OCH3), 5.95 (s, 2H, CH2), 6.83 (d, 2H, H-20, H-60, J ¼ 8.7 Hz), 7.24 (d,
2H, H-30, H-50, J ¼ 8.7 Hz), 7.45 (t,1H, H-400, J ¼ 7.6 Hz), 7.52 (t, 2H, H-
300, H-500, J ¼ 7.7 Hz), 7.83 (s, 1H, H-4), 7.87 (d, 2H, H-200, H-600,
eluent. Yield: 83%. Mp 122e124 ꢀC (Et2O/n-pentane). IR (Nujol)
n
3392, 1605, 1568, 1517, 1293, 1245, 1174, 1143, 1031, 1004, 926, 814,
725 cmꢂ1. 1H NMR (400 MHz, CDCl3)
1.93 (m, 4H, pyrrolidine H),
2.35 (s, 3H, CH3-piperazine), 2.59 (m, 4H, piperazine H), 3.17 (m, 4H,
J ¼ 7.6 Hz). 13C NMR (50 MHz, CDCl3)
d 54.1, 55.3, 114.2, 114.4, 127.4,
128.8,128.9,129.1,129.2,131.0,131.5,132.7,133.9,139.2,144.1,159.4.
d
HR-MS (ESI) m/z: Calcd for C20H16Cl2N3O: [M1 þ H]þ ¼ 384.0665,

154
M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
found 384.0669. Anal. Calcd for C20H15Cl2N3O: C, 62.51; H, 3.93; N,
10.94. Found: C, 62.39; H, 3.99; N, 11.21.
methanol (from 100/4 up to 100/12, v/v) as the eluent. Yield: 68%.
Mp 217e218 ꢀC (EtOAc). IR (Nujol)
3290, 3187, 1626, 1534, 1510,
1269, 1249, 1170, 1143, 1109, 1031, 902, 807, 722 cmꢂ1 1H NMR
(600 MHz, methanol-d4) 2.35 (s, 3H, CH3-piperazine), 2.63 (m, 4H,
n
.
4.1.12. Synthesis of 5-chloro-1-(4-methoxybenzyl)-3-phenyl-7-
(piperazin-1-yl)-1H-pyrazolo[3,4-c]pyridine (17a)
d
piperazine H), 3.08 (m, 4H, piperazine H), 3.13 (m, 4H, piperazine
H), 3.29 (brs, 4H, piperazine H), 3.71 (s, 3H, OCH3), 5.72 (s, 2H, CH2),
6.80 (d, 2H, J ¼ 8.7 Hz, H-30, H-50), 6.92 (s, 1H, H-4), 6.95 (d, 2H,
J ¼ 8.9 Hz, H-3000, H-5000), 7.15 (d, 2H, J ¼ 8.7 Hz, H-20, H-60), 7.38 (m,
3H, H-400, H-2000, H-6000), 7.47 (t, 2H, J ¼ 7.6 Hz, H-300, H-500), 7.83 (d,
This compound was prepared by a procedure analogous to that
of 10a. It was purified by column chromatography using a mixture
of dichloromethane/methanol (from 100/2 up to 100/10, v/v) as the
eluent. Yield: 91%. Mp 127e129 ꢀC (EtOAc/n-pentane). IR (Nujol)
n
3378, 3180,1609,1582,1544,1510,1286,1249,1174,1113,1072,1021,
2H, J ¼ 7.4 Hz, H-200, H-600). 13C NMR (151 MHz, methanol-d4)
d 46.2,
939, 810, 695 cmꢂ1 1H NMR (400 MHz, CDCl3)
.
d
3.17 (m, 4H,
46.4, 51.5, 52.4, 54.3, 55.8, 56.2, 91.9, 115.0, 119.1, 121.1, 128.2, 128.7,
129.3,129.6,130.0,131.5,133.2,134.6,138.1,145.3,147.0,149.8,151.4,
piperazine H), 3.36 (m, 4H, piperazine H), 3.74 (s, 3H, OCH3), 5.74 (s,
2H, CH2), 6.79 (d, 2H, J ¼ 8.7 Hz, H-30, H-50), 7.17 (d, 2H, J ¼ 8.7 Hz,
H-20, H-60), 7.42 (t, 1H, J ¼ 7.4 Hz, H-400), 7.50 (t, 2H, J ¼ 7.5 Hz, H-300,
H-500), 7.55 (s, 1H, H-4), 7.87 (d, 2H, J ¼ 7.4 Hz, H-200, H-600). 13C NMR
160.7.
HR-MS
(ESI)
m/z:
Calcd
for
C35H41N8O:
[M1 þ H]þ ¼ 589.3398, found 589.3405. Anal. Calcd for C35H40N8O:
C, 71.40; H, 6.85; N, 19.03. Found: C, 71.61; H, 6.99; N, 18.92.
(50 MHz, CDCl3)
d 45.3, 51.5, 53.4, 55.4, 109.9, 114.1, 127.4, 128.6,
128.7, 129.1, 129.3,131.0, 131.1, 132.4, 139.3, 145.1, 159.2. HR-MS (ESI)
m/z: Calcd for C24H25ClN5O: [M1 þ H]þ ¼ 434.1742, found 434.1746.
Anal. Calcd for C24H24ClN5O: C, 66.43; H, 5.57; N, 16.14. Found: C,
66.68; H, 5.77; N, 16.01.
4.1.16. Synthesis of 1-(4-methoxybenzyl)-7-(4-methylpiperazin-1-
yl)-N-(4-(4-methylpiperazin-1-yl)phenyl)-3-phenyl-1H-pyrazolo
[3,4-c]pyridin-5-amine (18c)
This compound was prepared according to the general proce-
dure described above for compounds 11aed. It was purified by
column chromatography using a mixture of dichloromethane/
methanol (from 100/3 up to 100/10, v/v) as the eluent. Yield: 80%.
4.1.13. Synthesis of 5-chloro-1-(4-methoxybenzyl)-3-phenyl-7-(4-
methylpiperazin-1-yl)-1H-pyrazolo[3,4-c]pyridine (17b)
This compound was prepared by a procedure analogous to that
of 10a. It was purified by column chromatography using a mixture
of dichloromethane/methanol (from 100/1 up to 100/8, v/v) as the
Mp 161e162 ꢀC (Et2O/n-pentane). IR (Nujol)
n
3324, 3174, 1602,
1577, 1554, 1514, 1303, 1252, 1174, 1147, 1011, 817, 723 cmꢂ1. 1H NMR
(400 MHz, CDCl3) 2.36 (s, 3H, CH3-piperazine), 2.40 (s, 3H, CH3-
d
eluent to provide pure 17b as foam. Yield: 89%. IR (film)
2941, 2839, 2798, 1609, 1585, 1554, 1510, 1463, 1266, 1177, 1031,
820 cmꢂ1. 1H NMR (400 MHz, CDCl3)
2.39 (s, 3H, CH3-piperazine),
n
3044,
piperazine), 2.61 (m, 4H, piperazine H), 2.67 (brs, 4H, piperazine H),
3.17 (m, 4H, piperazine H), 3.34 (brs, 4H, piperazine H), 3.73 (s, 3H,
OCH3), 5.68 (s, 2H, CH2), 6.21 (brs, 1H, D2O exch, NH), 6.77 (d, 2H,
J ¼ 8.6 Hz, H-30, H-50), 6.90e6.94 (m, 3H, H-4, H-3000, H-5000),
7.18e7.24 (m, 4H, H-20, H-60, H-2000, H-6000), 7.34 (t, 1H, J ¼ 7.3 Hz, H-
400), 7.43 (t, 2H, J ¼ 7.4 Hz, H-300, H-500), 7.83 (d, 2H, J ¼ 7.4 Hz, H-200,
d
2.66 (m, 4H, piperazine H), 3.38 (m, 4H, piperazine H), 3.70 (s, 3H,
OCH3), 5.73 (s, 2H, CH2), 6.79 (d, 2H, J ¼ 8.6 Hz, H-30, H-50), 7.21 (d,
2H, J ¼ 8.6 Hz, H-20, H-60), 7.35e7.50 (m, 3H, H-300, H-400, H-500), 7.54
(s, 1H, H-4), 7.88 (d, 2H, J ¼ 7.0 Hz, H-200, H-600). 13C NMR (50 MHz,
H-600). 13C NMR (50 MHz, CDCl3)
d 46.2, 46.3, 49.9, 50.5, 53.1, 55.1,
CDCl3)
d
46.1, 50.5, 53.1, 54.7, 55.1, 109.3, 113.8, 127.2, 128.4, 128.6,
55.3, 89.5, 113.9, 117.5, 121.2, 127.1, 127.7, 127.8, 128.8, 130.0, 131.7,
133.5, 134.8, 144.1, 146.7, 148.5, 149.3, 158.9. HR-MS (ESI) m/z: Calcd
for C36H43N8O: [M1 þ H]þ ¼ 603.3554, found 603.3558. Anal. Calcd
for C36H42N8O: C, 71.73; H, 7.02; N, 18.59. Found: C, 71.87; H, 7.00;
N, 18.39.
128.8, 129.1, 130.7, 132.2, 139.1, 144.8, 148.9, 159.0. HR-MS (ESI) m/z:
Calcd for C25H27ClN5O: [M1 þ H]þ ¼ 448.1899, found 448.1901.
Anal. Calcd for C25H26ClN5O: C, 67.03; H, 5.85; N, 15.63. Found: C,
66.96; H, 5.89; N, 15.52.
4.1.14. Synthesis of 1-(4-methoxybenzyl)-N,3-diphenyl-7-
4.2. Biological studies
(piperazin-1-yl)-1H-pyrazolo[3,4-c]pyridin-5-amine (18a)
This compound was prepared according to the general proce-
dure described above for compounds 11aed. It was purified by
column chromatography using a mixture of dichloromethane/
methanol (from 100/2 up to 100/8, v/v) as the eluent. Yield: 73%.
4.2.1. Cell culture
EA.hy926 immortalized human EC (kindly provided by Prof. A.
Papapetropoulos, National and Kapodistrian University of Athens),
were maintained in DMEM containing 10% FBS, 1% L-glutamine, 2%
Mp 142e143 ꢀC (EtOAc). IR (Nujol)
1245, 1174, 1028, 905, 810, 752, 695 cmꢂ1
methanol-d4)
n
3399, 3174, 1599, 1561, 1514,
1H NMR (600 MHz,
3.18 (m, 4H, piperazine H), 3.36 (brs, 4H, piperazine
hypoxanthine-aminopterin-thymidine mixture and antibiotics
(10 U/mL penicillin, 100 mg/mL streptomycin). HUVEC were iso-
lated from the vein of fresh umbilical cords, collected from 3 to 5
donors, following collagenase digestion as previously described
[45] and maintained in pooled cultures up to passage 4 in M199
.
d
H), 3.71 (s, 3H, OCH3), 5.73 (s, 2H, CH2), 6.82 (d, 2H, J ¼ 8.7 Hz, H-30,
H-50), 6.86 (m, 1H, H-4000), 7.06 (s, 1H, H-4), 7.14 (d, 2H, J ¼ 8.7 Hz, H-
20, H-60), 7.23 (t, 2H, J ¼ 7.2 Hz, H-3000, H-5000), 7.38 (t,1H, J ¼ 7.4 Hz, H-
400), 7.44e7.50 (m, 4H, H-300, H-500, H-2000, H-6000), 7.86 (d, 2H,
including 15% FBS,1% L-glutamine, 5 U/mL heparin, 0.2 mg/mL ECGS
and antibiotics. HUVEC identification was regularly certified from
typical cobblestone cell morphology and the presence of endo-
thelial markers (CD31 and vWF) as detected by cytospin-
immunostaining. Lewis lung carcinoma cells (LLC) were obtained
from American Type Culture Collection (Manassas, VA, USA) and
J ¼ 7.4 Hz, H-200, H-600). 13C NMR (151 MHz, methanol-d4)
d 45.8,
51.4, 54.4, 55.8, 93.8, 115.1, 119.1, 121.4, 128.3, 128.8, 129.4, 129.5,
129.9, 130.1, 131.5, 133.2, 134.4, 144.4, 145.4, 149.1, 150.5, 160.8. HR-
MS (ESI) m/z: Calcd for C30H31N6O: [M1 þ H]þ ¼ 491.2554, found
491.2557. Anal. Calcd for C30H30N6O: C, 73.45; H, 6.16; N, 17.13.
Found: C, 73.21; H, 6.01; N, 17.33.
maintained in DMEM containing 10% FBS, 1% L-glutamine and an-
tibiotics. All cell cultures were initiated from authenticated master
frozen stocks, incubated at 37 ꢀC in a humidified atmosphere
containing 5% CO2 and regularly tested for mycoplasma contami-
nation. All reagents were purchased from Biochrom (Berlin,
Germany).
4.1.15. Synthesis of 1-(4-methoxybenzyl)-N-(4-(4-methylpiperazin-
1-yl)phenyl)-3-phenyl-7-(piperazin-1-yl)-1H-pyrazolo[3,4-c]
pyridin-5-amine (18b)
This compound was prepared according to the general proce-
dure described above for compounds 11aed. It was purified by
column chromatography using a mixture of dichloromethane/
4.2.2. EC proliferation and viability test
The synthesized derivatives were tested in vitro for their anti-

M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
155
proliferative properties against EC by using the MTT assay that is
based on the reduction of soluble MTT into blue-purple formazan
4.2.5. Western blot
Serum-starved HUVEC cultured in 6-well plates were treated
with 2 M of test compound or vehicle for 2 h followed by exposure
crystals by metabolically active cells [46]. Briefly, EA.hy926 or
m
HUVEC plated in 96-well plates at 4 ꢁ 103 cells/well in 100
m
L of
to VEGF-165 (50 ng/mL) or PBS for 5 min. Cells were solubilized
with lysis buffer (50 mM Tris-HCl, pH 7.6; 150 mM NaCl; 50 mM
NaF; 0.5% sodium deoxycholate; 1 mM EDTA; 0.1 mM EGTA; 1%
Triton-X; 0.1% sodium dodecyl sulfate (SDS); 1 mM phenyl-
methanesulfonyl fluoride; 1 mM Na3VO4; 0.1% protease inhibitor
cocktail) and cell lysates containing equal amounts of protein were
resolved by SDS-PAGE and immunoblotting as described before
[49]. Primary antibodies used (Cell Signaling Technology, Danvers,
MA, USA) specifically recognize the phosphorylated forms of
VEGFR2 (Tyr 951, 996, 1056, 1175), AKT (Ser473), p38MAPK
(Thr180/Tyr182) and ERK1/2 (Thr202/Tyr204, Thr185/Tyr187) or
the total endogenous levels of the corresponding proteins. Immu-
noreactive bands were visualized by enhanced chemiluminescence
detection (SuperSignal™ West Pico Chemiluminescent Substrate
kit, Thermo Scientific, Waltham, MA, USA). Gel-Pro Analyzer soft-
ware (Media Cybernetics, Inc., Rockville, MD, USA) was used for
densitometry.
complete medium were allowed to adhere for 24 h and then sub-
jected to serum-starvation in medium containing 0.25% BSA (star-
vation medium) for 4 h. Subsequently, the cells were exposed to
synthetic compounds or 0.1% v/v DMSO vehicle in presence of 2.5%
FBS, 50 ng/mL of VEGF-165 or starvation medium alone for
24e48 h. At the end of treatment, MTT (Sigma, St. Louis, MO, USA)
was added at 0.5 mg/mL and the cells further incubated for 4 h at
37 ꢀC. The formazan crystals were solubilized by the addition of
0.1 N HCl in anhydrous isopropanol and the absorbance was
measured on a microtiter plate reader at 595 nm with correction at
630 nm. The number of viable cells was determined by interpola-
tion of OD values to respective standard cell growth curves.
Cell proliferation was also assessed under similar treatment
conditions by using the CyQUANT® cell proliferation assay kit
(Invitrogen, Waltham, MA, USA), according to manufacturer’s in-
structions. Briefly, after treatment completion medium was
removed from the wells, cells were carefully washed once with PBS,
and then 200
mL of CyQUANT GR dye/cell lysis buffer was added to
4.2.6. LLC mouse tumor growth model
each well and incubated for 5 min at room temperature, protected
from light. The sample fluorescence was measured in a fluores-
cence microplate reader using excitation and emission filters at 480
and 520 nm, respectively and a reference standard curve was used
for converting sample fluorescence to cell numbers.
Finally to check the potential cytotoxicity of test compounds
under conditions applied for each of the following biological assays,
cells were treated as specified and the number of viable cells was
measured by the MTT assay.
In vivo experiments were approved by the Veterinary Admin-
istration Bureau, Prefecture of Athens, Greece under compliance to
the national law and the EU Directive 2010/63/EU for animal ex-
periments. Male 6e8 week-old C57BL/6 mice (Hellenic Pasteur
Institute, Athens, Greece) were assigned to separate groups of 5e6
mice and inoculated sc into the right flank with 5 ꢁ 105 viable LLC
cells. Injections of test compounds at 0.05e0.2 mg/kg of body
weight or vehicle were administered ip, thrice a week for 2 weeks,
starting 24 h before implantation of tumor cells. Every other day
tumor growth was monitored by electronic caliper measurement as
described before [50] and mice were weighed and examined clin-
ically for general condition. At end-point, animals were sacrificed,
tumors and vital organs were excised and tissue samples were fixed
in formalin.
4.2.3. IC50 determination
IC50 values for compound anti-proliferative activity were
measured by the MTT assay as described above. Cells cultured in
DMEM containing 2.5% FBS were exposed for 48 h to serial 1:2
dilutions of test compounds covering a concentration range of
0.125e50
mM. IC50 values (mean SEM) were calculated from eight
4.2.7. IHC and histopathology
or more measured points using non-linear regression analysis.
Slides of paraffin-embedded tissue sections were prepared and
treated according to standard protocols. Next, sections were
deparafinized in xylene following rehydration in reducing degrees
of absolute ethanol (100%, 96%, 70%) followed by washing in
distilled water. For assessment of microvessel density, slides with
tumor tissue sections were initially exposed to normal goat serum
for 30 min at RT for blocking of non-specific binding sites and then
incubated overnight at 4 ꢀC with an anti-CD31 antibody (ab28364,
Abcam, Cambridge, UK) at 1:100 [51]. Immunoreactivity was
detected using the Vectastain ABC and DAB Peroxidase Substrate kit
(Vector Laboratories, Burlingame, CA, USA) according to the
manufacturer instructions. Immunoreactive cells were quantified
in six different non-overlapping high-power fields. For estimation
of compound toxicity, slides with tissue sections from vital organs
(kidney, lung, spleen, liver, heart) were stained with haematoxylin-
eosin and examined microscopically for histopathological
alterations.
4.2.4. Cell migration and Matrigel tube formation assays
HUVEC migration was evaluated using 8
mm-pore size mem-
brane Transwell chambers for 24-well plates (ThinCerts™, Greiner
Bio-One International GmbH, Kremsmuenster, Austria) as previ-
ously described [47]. Briefly, confluent cultures of serum-starved
HUVEC grown in 6-well plates were pre-treated o/n with test
compounds or vehicle solutions in medium containing 5% FBS. Cells
were transferred to the upper chamber (6 ꢁ 104 cells/well) and left
to migrate towards 5% FBS, VEGF-165 (50 ng/mL) or starvation
medium added in the bottom chamber, for 6 h. Cells that have
migrated to the lower side of the membrane were fixed with ice-
cold ethanol, stained with haematoxylin for 5 min followed by
eosin for 2 min, photographed and manually counted in five fields
of view/well.
Formation of HUVEC tube-like structures was assessed by
Matrigel assay as previously described [48]. Briefly, pre-starved
HUVEC were seeded at 2 ꢁ 104 cell per well in 24-well plates
4.2.8. RNA isolation and qRT-PCR
pre-coated with 50
mL of growth factor-reduced Matrigel (BD Bio-
Serum-starved confluent HUVEC grown in 6-well plates were
sciences, San Jose, CA, USA). Test compound or vehicle solutions
were then added and cells further incubated for 24 h. Cells were
fixed with ice-cold ethanol, stained with haematoxylin for 5 min
followed by eosin for 2 min, photographed and the number of tube-
like structures was scored in at least 5 fields of view/well.
Compound concentrations used in all these assays were previ-
ously shown to be non-cytotoxic.
treated with 2 mM of test compound or vehicle in presence of 2.5%
FBS for 6 h or 24 h. Total RNA was then isolated using the pureLink
RNA Mini Kit (Ambion, Carlsbad, CA, USA) according to the manu-
facturer instructions. The quantification and quality analysis of RNA
was performed on a Bioanalyzer 2100 (Agilent, Santa Clara, CA,
USA).
For qRT-PCR, RNA (250 ng) was subjected to reverse

156
M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
transcription using the Enhanced Avian First Strand synthesis kit
(Sigma, St Louis, MO, USA) and qRT-PCR was performed using the
KAPA SYBR Fast One-step qRT-PCR protocol (KapaBiosystems,
Wilmington, MA, USA). Amplifications were performed in triplicate
on a MJ RESEARCH PTC-200 qRT-PCR system, as follows: 95 ꢀC for
5 min and 40 cycles of 95 ꢀC for 5 s, 58 ꢀC for 20 s and 72 ꢀC for 5 s.
The ribosomal protein RPS18S gene was used as the endogenous
reference gene for all normalizations. Primers were provided by
Eurofins Genomics (Ebersberg, Germany). The sequence of primers
was as follows:
Conflict of interest disclosure
The authors declare no competing financial interest.
Authors declaration
All authors have participated significantly in the completion of
the manuscript. OP, AC, FNK, PM, NP, HL: conceived and designed
the experiments; MM, VG, VK, OP, GK, IKK, NL, HL: performed the
experiments; MM, VG, VK, OP, GK, IKK, NL, AC, FNK, PM, NP, HL:
analyzed the data; MM, PM, NP, HL: wrote the manuscript.
The final manuscript and authorship have been approved by all
authors.
HMGCS1: Forward primer: 50-CATTAGACCGCTGCTATTCTGTC-30
and Reverse primer: 50-TTCAGCAACATCCGAGCTAGA-30,
MSMO1: Forward primer: 50-ACATGGGAAAACCAATGGAA-30
and Reverse primer: 50-TTCCAAATGGAGCCTGAAAC-30,
INSIG1: Forward primer: 50-GCACTGCATTAAACGTGTGG-30 and
Reverse primer: 50-GCAGCACTGAAATGAATGGA-30,
Acknowledgments
SQLE: Forward primer: 50-TTAGAGGAGAAATGCCAAGGAA-30
and Reverse primer: 50-CACTGATGAAGGAGGAAGGAAG-30,
UBE2C: Forward primer: 50-TGATGTCTGGCGATAAAGGGATT-30
and Reverse primer: 50-GTGATAGCAGGGCGTGAGGAA-30,
CCNB2: Forward primer: 50-CCTCCCTTTTCAGTCCGC-30 and
Reverse primer: 50-CTCCTGTGTCAATATTCTCCAAATC-30,
CCNA2: Forward primer: 50-CTGCATTTGGCTGTGAACTAC-30 and
Reverse primer: 50-ACAAACTCTGCTACTTCTGGG-30,
We thank Z. Kollia (Research Unit for animal standards, Evan-
gelismos Hospital, Athens, Greece) for animal care assistance. We
also gratefully acknowledge financial support from NSRF
2007e2013 National action, Cooperation sub-action II: SYN-
ERGASIA 2009 PROGRAMME (09SYN-21-1078).
Appendix A. Supplementary data
CDC20: Forward primer: 50-GGCACCAGTGATCGACACATTCG-
CAT-30 and Reverse primer: 50-GCCATAGCCTCAGGGTCTCATCTGCT-
30,
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2016.05.035.
HMOX1: Forward primer: 50-CTGCGTTCCTGCTCAACATC-30 and
Reverse primer: 50-GGGGCAGAATCTTGCACTTT-30,
References
RPS18S: Forward primer:50-TCGGAACTGAGGCCATGA-30 and
[1] J. Folkman, Angiogenesis: an organizing principle for drug discovery? Nat. Rev.
Drug Discov. 6 (2007) 273e286.
[2] L. Wang, G. Chen, X. Lu, S. Wang, S. Han, Y. Li, G. Ping, X. Jiang, H. Li, J. Yang,
C. Wu, Novel chalcone derivatives as hypoxia-inducible factor (HIF)-1 inhib-
itor: synthesis, anti-invasive and anti-angiogenic properties, Eur. J. Med.
Chem. 89 (2015) 88e97.
[3] M. Romero, M. Zanuy, E. Rosell, M. Cascante, J. Piulats, M. Font-Bardia,
J. Balzarini, E. De Clerq, M.D. Pujol, Optimization of xanthatin extraction from
Xanthium spinosum L. and its cytotoxic, anti-angiogenesis and antiviral
properties, Eur. J. Med. Chem. 90 (2015) 491e496.
[4] D. Ravishankar, K.A. Watson, S.Y. Boateng, R.J. Green, F. Greco, H.M.I. Osborn,
Exploring quercetin and luteolin derivatives as antiangiogenic agents, Eur. J.
Med. Chem. 97 (2015) 259e274.
Reverse primer: 50-GAACCTCCGACTTTCGTTC-30
Ct values were automatically calculated by the machine’s soft-
DD
ware and data analysis was performed by using the 2ꢂ CT method
[52].
4.2.9. Microarray hybridizations and functional analysis
Synthesis of cDNA and biotinylated cRNA was performed with
the Illumina TotalPrep RNA Amplification Kit (Illumina, San Diego,
CA, USA) using 500 ng of total RNA. Hybridization was performed
onto Illumina Human-HT12 platform according to the manufac-
turer’s instructions. The raw data were analyzed using packages
lumi [53] and limma [54] in R computational environment. Briefly,
method bgAdjust was used for correction, which estimates the
background based on control probe information for between array
correction, followed by variance adjustment based on bead level
and mean-variance using variance stabilizing transformation (vst)
for within array correction, and finally data normalization using the
robust spline method (rsn), which combines the characteristics of
quantile and loess algorithms. Significantly differentiated probe
sets in compound-treated as compared to DMSO-treated cells were
identified using a moderated paired t-test on linear models from
limma, by applying the following thresholds: p-value < 0.05 and a
fold change >|0.5| in log2 scale. For functional analysis highlighting
significant cellular processes based on GO, we used StRAnGER2 as
described before [55].
[5] S. Salerno, A.M. Marini, G. Fornaciari, F. Simorini, C. La Motta, S. Taliani,
ꢀ
S. Sartini, F. Da Settimo, A.N. García-Argaez, O. Gia, S. Cosconati, E. Novellino,
P. D’Ocon, A. Fioravanti, P. Orlandi, G. Bocci, L. Dalla Via, Investigation of new
2-aryl substituted Benzothiopyrano[4,3-d]pyrimidines as kinase inhibitors
targeting vascular endothelial growth factor receptor 2, Eur. J. Med. Chem. 103
(2015) 29e43.
[6] Y. Shan, H. Gao, X. Shao, J. Wang, X. Pan, J. Zhang, Discovery of novel VEGFR-2
inhibitors. Part 5: exploration of diverse hinge-binding fragments via core-
refining approach, Eur. J. Med. Chem. 103 (2015) 80e90.
[7] N. Ferrara, H.-P. Gerber, J. LeCouter, The biology of VEGF and its receptors, Nat.
Med. 9 (2003) 669e676.
[8] C.K. Goldman, J. Kim, W.L. Wong, V. King, T. Brock, G.Y. Gillespie, Epidermal
growth factor stimulates vascular endothelial growth factor production by
human malignant glioma cells: a model of glioblastoma multiforme patho-
physiology, Mol. Biol. Cell 4 (1993) 121e133.
[9] H. Li, L. Fredriksson, X. Li, U. Eriksson, PDGF-D is a potent transforming and
angiogenic growth factor, Oncogene 22 (2003) 1501e1510.
[10] C. Lieu, J. Heymach, M. Overman, H. Tran, S. Kopetz, Beyond VEGF: inhibition
of the fibroblast growth factor pathway and antiangiogenesis, Clin. Cancer
Res. 17 (2011) 6130e6139.
[11] A.-K. Olsson, A. Dimberg, J. Kreuger, L. Claesson-Welsh, VEGF receptor sig-
nalling in control of vascular function, Nat. Rev. Mol. Cell Biol. 7 (2006)
359e371.
[12] M. Shibuya, L. Claesson-Welsh, Signal transduction by VEGF receptors in
regulation of angiogenesis and lymphangiogenesis, Exp. Cell Res. 312 (2006)
549e560.
[13] T. Takahashi, S. Yamaguchi, K. Chida, M. Shibuya, A single autophosphor-
ylation site on KDR/Flk-1 is essential for VEGF-A-dependent activation of PLC-
4.2.10. Statistical analysis of cellular assays
The Mann-Whitney test was applied to determine the statistical
significance of the data obtained and to compare the means be-
tween groups using GraphPad Prism 5.00 for Windows (GraphPad
Software, San Diego, CA, USA). A p value of <0.05 represented a
statistically significant difference.
g
and DNA synthesis in vascular endothelial cells, EMBO J. 20 (2001)
2768e2778.
[14] H. Zeng, S. Sanyal, D. Mukhopadhyay, Tyrosine residues 951 and 1059 of
vascular endothelial growth factor receptor-2 (KDR) are essential for vascular
permeability factor/vascular endothelial growth factor-induced endothelium
migration and proliferation, respectively, J. Biol. Chem. 276 (2001)

M. Michailidou et al. / European Journal of Medicinal Chemistry 121 (2016) 143e157
157
32714e32719.
[36] D. Chapman, J. Hurst, Pyrazolopyridines. Part 5. Preparation and reactions of
pyrazolo[3,4-c]pyridines, J. Chem. Soc. (1980) 2398e2404.
[37] T. Ukai, H. Kawazura, Y. Ishii, J.J. Bonnet, J.A. Ibers, Chemistry of
€
[15] K. Holmqvist, M.J. Cross, C. Rolny, R. Hagerkvist, N. Rahimi, T. Matsumoto,
L. Claesson-Welsh, M. Welsh, The adaptor protein Shb binds to tyrosine 1175
in vascular endothelial growth factor (VEGF) receptor-2 and regulates VEGF-
dependent cellular migration, J. Biol. Chem. 279 (2004) 22267e22275.
[16] T. Matsumoto, S. Bohman, J. Dixelius, T. Berge, A. Dimberg, P. Magnusson,
L. Wang, C. Wikner, J.H. Qi, C. Wernstedt, J. Wu, S. Bruheim, H. Mugishima,
D. Mukhopadhyay, A. Spurkland, L. Claesson-Welsh, VEGF receptor-2 Y951
signaling and a role for the adapter molecule TSAd in tumor angiogenesis,
EMBO J. 24 (2005) 2342e2353.
dibenzylideneacetone-palladium(0) complexes
: I. Novel tris(dibenzylide-
neacetone)dipalladium(solvent) complexes and their reactions with qui-
nones, J. Organomet. Chem. 65 (1974) 253e266.
[38] X. Huang, K.W. Anderson, D. Zim, L. Jiang, A. Klapars, S.L. Buchwald, Expanding
Pd-catalyzed CꢂN bond-forming processes: the first amidation of aryl sul-
fonates, aqueous amination, and complementarity with Cu-catalyzed re-
actions, J. Am. Chem. Soc. 125 (2003) 6653e6655.
[17] N. Ferrara, K.J. Hillan, W. Novotny, Bevacizumab (Avastin), a humanized anti-
VEGF monoclonal antibody for cancer therapy, Biochem. Biophys. Res. Com-
mun. 333 (2005) 328e335.
[18] Q.S.-C. Chu, Aflibercept (AVE0005): an alternative strategy for inhibiting
tumour angiogenesis by vascular endothelial growth factors, Expert Opin.
Biol. Ther. 9 (2009) 263e271.
[19] F. Musumeci, M. Radi, C. Brullo, S. Schenone, Vascular endothelial growth
factor (VEGF) receptors: drugs and new inhibitors, J. Med. Chem. 55 (2012)
10797e10822.
[20] K.J. Gotink, H.M.W. Verheul, Anti-angiogenic tyrosine kinase inhibitors: what
is their mechanism of action? Angiogenesis 13 (2010) 1e14.
[21] G. Bergers, D. Hanahan, Modes of resistance to anti-angiogenic therapy, Nat.
Rev. Cancer 8 (2008) 592e603.
[39] E.J. Hanan, C. Eigenbrot, M.C. Bryan, D.J. Burdick, B.K. Chan, Y. Chen, J. Dotson,
R.A. Heald, P.S. Jackson, H. La, M.D. Lainchbury, S. Malek, H.E. Purkey,
G. Schaefer, S. Schmidt, E.M. Seward, S. Sideris, C. Tam, S. Wang, S.K. Yeap,
I. Yen, J. Yin, C. Yu, I. Zilberleyb, T.P. Heffron, Discovery of selective and
noncovalent diaminopyrimidine-based inhibitors of epidermal growth factor
receptor containing the T790M resistance mutation, J. Med. Chem. 57 (2014)
10176e10191.
[40] J. Xu, Y. Dang, Y.R. Ren, J.O. Liu, Cholesterol trafficking is required for mTOR
activation in endothelial cells, Proc. Natl. Acad. Sci. U. S. A. 107 (2010)
4764e4769.
[41] L. Fang, S.-H. Choi, J.S. Baek, C. Liu, F. Almazan, F. Ulrich, P. Wiesner, A. Taleb,
E. Deer, J. Pattison, J. Torres-Vazquez, A.C. Li, Y.I. Miller, Control of angio-
genesis by AIBP-mediated cholesterol efflux, Nature 498 (2013) 118e122.
[42] F. Adah, H. Benghuzzi, M. Tucci, Cholesterol production inhibitor (statin)
increased angiogenesis in surgically created femoral defect in an animal
model, Biomed. Sci. Instrum. 50 (2014) 54e61.
[43] G. Eelen, P. de Zeeuw, M. Simons, P. Carmeliet, Endothelial cell metabolism in
normal and diseased vasculature, Circ. Res. 116 (2015) 1231e1244.
[44] R.-H. Zhou, M. Yao, T.-S. Lee, Y. Zhu, M. Martins-Green, J.Y.-J. Shyy, Vascular
endothelial growth factor activation of sterol regulatory element binding
protein: a potential role in angiogenesis, Circ. Res. 95 (2004) 471e478.
[45] M. Hatziapostolou, P. Katsoris, E. Papadimitriou, Different inhibitors of
plasmin differentially affect angiostatin production and angiogenesis, Eur. J.
Pharmacol. 460 (2003) 1e8.
[22] M. Chauhan, R. Kumar, Medicinal attributes of pyrazolo[3,4-d]pyrimidines: a
review, Bioorg. Med. Chem. 21 (2013) 5657e5668.
[23] S. Schenone, M. Radi, F. Musumeci, C. Brullo, M. Botta, Biologically driven
synthesis of pyrazolo[3,4-d]pyrimidines as protein kinase inhibitors: an old
scaffold as a new tool for medicinal chemistry and chemical biology studies,
Chem. Rev. 114 (2014) 7189e7238.
€
[24] C. Peifer, S. Bühler, D. Hauser, K. Kinkel, F. Totzke, C. Schachtele, S. Laufer,
Design, synthesis and characterization of N9/N7-substituted 6-aminopurines
as VEGF-R and EGF-R inhibitors, Eur. J. Med. Chem. 44 (2009) 1788e1793.
ꢁ
[25] S.B. Weitensteiner, J. Liebl, V. Krystof, L. Havlícek, T. Gucký, M. Strnad, R. Fürst,
A.M. Vollmar, S. Zahler, Trisubstituted pyrazolopyrimidines as novel angio-
genesis inhibitors, PLoS One 8 (2013) e54607.
[46] T. Mosmann, Rapid colorimetric assay for cellular growth and survival:
application to proliferation and cytotoxicity assays, J. Immunol. Methods 65
(1983) 55e63.
[47] H. Loutrari, S. Magkouta, A. Papapetropoulos, C. Roussos, Mastic oil inhibits
the metastatic phenotype of mouse lung adenocarcinoma cells, Cancers 3
(2011) 789e801.
[26] Y. Dai, K. Hartandi, N.B. Soni, L.J. Pease, D.R. Reuter, A.M. Olson, D.J. Osterling,
S.Z. Doktor, D.H. Albert, J.J. Bouska, K.B. Glaser, P.A. Marcotte, K.D. Stewart,
S.K. Davidsen, M.R. Michaelides, Identification of aminopyrazolopyridine
ureas as potent VEGFR/PDGFR multitargeted kinase inhibitors, Bioorg. Med.
Chem. Lett. 18 (2008) 386e390.
[27] Y. Oguro, N. Miyamoto, T. Takagi, K. Okada, Y. Awazu, H. Miki, A. Hori,
K. Kamiyama, S. Imamura, N-Phenyl-N0-[4-(5H-pyrrolo[3,2-d]pyrimidin-4-
yloxy)phenyl]ureas as novel inhibitors of VEGFR and FGFR kinases, Bioorg.
Med. Chem. 18 (2010) 7150e7163.
[28] Y. Oguro, N. Miyamoto, K. Okada, T. Takagi, H. Iwata, Y. Awazu, H. Miki,
A. Hori, K. Kamiyama, S. Imamura, Design, synthesis, and evaluation of 5-
methyl-4-phenoxy-5H-pyrrolo[3,2-d]pyrimidine derivatives: novel VEGFR2
kinase inhibitors binding to inactive kinase conformation, Bioorg. Med. Chem.
18 (2010) 7260e7273.
[48] M. Michailidou, H. Brown, D. Lefley, A. Evans, S. Cross, R. Coleman, N. Brown,
I. Holen, Microvascular endothelial cell responses in vitro and in vivo: mod-
ulation by zoledronic acid and paclitaxel? J. Vasc. Res. 47 (2010) 481e493.
[49] H. Loutrari, M. Hatziapostolou, V. Skouridou, E. Papadimitriou, C. Roussos,
F.N. Kolisis, A. Papapetropoulos, Perillyl alcohol is an angiogenesis inhibitor,
J. Pharmacol. Exp. Ther. 311 (2004) 568e575.
[50] S. Magkouta, G. Stathopoulos, I. Psallidas, A. Papapetropoulos, F. Kolisis,
C. Roussos, H. Loutrari, Protective effects of mastic oil from Pistacia lentiscus
variation chia against experimental growth of lewis lung carcinoma, Nutr.
Cancer 61 (2009) 640e648.
[51] N. Weidner, J.P. Semple, W.R. Welch, J. Folkman, Tumor angiogenesis and
metastasis - correlation in invasive breast carcinoma, New Engl. J. Med. 324
(1991) 1e8.
[52] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using
real-time quantitative PCR and the 2ꢂDDCT method, Methods 25 (2001)
402e408.
[29] V.N. Kourafalos, P. Marakos, N. Pouli, L.B. Townsend, The synthesis of 4-
deazaformycin A, J. Org. Chem. 68 (2003) 6466e6469.
ꢁ
ꢁ
[30] S.S. Kadam, L. Maier, I. Kostakis, N. Pouli, J. Tousek, M. Necas, P. Marakos,
R. Marek, Synthesis and tautomerism of substituted pyrazolo[4,3-c]pyrazoles,
Eur. J. Org. Chem. 2013 (2013) 6811e6822.
[31] T. Tite, N. Lougiakis, P. Marakos, N. Pouli, The synthesis of 6-deazaformycin A,
Synlett 18 (2009) 2027e2930.
[32] L.N. Pino, W.S. Zehrung, Preparation of pure 2-aminonitropyridines and 2-
aminonitropicolines. Rapid separations by sublimation, J. Am. Chem. Soc. 77
(1955) 3154e3155.
[33] A.G. Burton, P.J. Halls, A.R. Katritzky, The selective 3- and 5-nitration of 2-
pyridones, Tetrahedron Lett. 12 (1971) 2211e2212.
[53] P. Du, W.A. Kibbe, S.M. Lin, lumi: a pipeline for processing Illumina micro-
array, Bioinformatics 24 (2008) 1547e1548.
[54] M.E. Ritchie, B. Phipson, D. Wu, Y. Hu, C.W. Law, W. Shi, G.K. Smyth, Limma
powers differential expression analyses for RNA-sequencing and microarray
studies, Nucleic Acids Res. 43 (2015) e47ee47.
[34] A. Roe, R.B. Seligman, The preparation of 3-fluoroisonicotinic acid and related
compounds, J. Org. Chem. 20 (1955) 1729e1732.
[35] P. Marakos, N. Pouli, D.S. Wise, L.B. Townsend, A new and facile method for
the preparation of 3-substituted pyrazolo[3,4-c]pyridines, Synlett 1997
(1997) 561e562.
[55] E.D. Pilalis, A.A. Chatziioannou, Prioritized functional analysis of biological
experiments using resampling and noise control methodologies, in: Pro-
ceedings IEEE 13th International Conference on Bioinformatics and Bioengi-
neering (BIBE), 2013, pp. 1e3.