Design and synthesis of 2-oxindole based multi-targeted inhibitors of PDK1/Akt signaling pathway for the treatment of glioblastoma multiforme

Article abstract of DOI:10.1016/j.ejmech.2015.10.020

Aggressive behavior and diffuse infiltrative growth are the main features of Glioblastoma multiforme (GBM), together with the high degree of resistance and recurrence. Evidence indicate that GBM-derived stem cells (GSCs), endowed with unlimited proliferative potential, play a critical role in tumor development and maintenance. Among the many signaling pathways involved in maintaining GSC stemness, tumorigenic potential, and anti-apoptotic properties, the PDK1/Akt pathway is a challenging target to develop new potential agents able to affect GBM resistance to chemotherapy. In an effort to find new PDK1/Akt inhibitors, we rationally designed and synthesized a small family of 2-oxindole derivatives. Among them, compound 3 inhibited PDK1 kinase and downstream effectors such as CHK1, GS3Kα and GS3Kβ, which contribute to GCS survival. Compound 3 appeared to be a good tool for studying the role of the PDK1/Akt pathway in GCS self-renewal and tumorigenicity, and might represent the starting point for the development of more potent and focused multi-target therapies for GBM.

Full text of DOI:10.1016/j.ejmech.2015.10.020

European Journal of Medicinal Chemistry 105 (2015) 274e288
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Design and synthesis of 2-oxindole based multi-targeted inhibitors of
PDK1/Akt signaling pathway for the treatment of glioblastoma
multiforme
Simona Sestitob 1, Giulia Nesi a , Simona Daniele , Alma Martelli , Maria Digiacomo ,
a
,
,
1
a
a
a
b
a
a
Alice Borghini , Daniele Pietra , Vincenzo Calderone , Annalina Lapucci ,
c
d
e
a
Marco Falasca , Paola Parrella , Angelantonio Notarangelo , Maria C. Breschi ,
a
a
a, *
Marco Macchia , Claudia Martini , Simona Rapposelli
a
Department of Pharmacy, University of Pisa, Via Bonanno, 6, 56126 Pisa, Italy
Alidans S.r.l., Via Vecchializia, 48, 56017 San Giuliano Terme, PI, Italy
Metabolic Signalling Group, School of Biomedical Sciences, Curtin Health Innovation Research Institute Biosciences, Curtin University, Perth, Western
Australia 6102, Australia
Laboratory of Oncology, Hospital “Casa Sollievo Della Sofferenza”, Viale Cappuccini, 1, 71013 San Giovanni Rotondo, FG, Italy
b
c
d
e
Medical Genetics Unit, IRCCS Casa Sollievo della Sofferenza Hospital, I-71013 San Giovanni Rotondo, FG, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 August 2015
Received in revised form
Aggressive behavior and diffuse infiltrative growth are the main features of Glioblastoma multiforme
(
GBM), together with the high degree of resistance and recurrence. Evidence indicate that GBM-derived
stem cells (GSCs), endowed with unlimited proliferative potential, play a critical role in tumor devel-
opment and maintenance. Among the many signaling pathways involved in maintaining GSC stemness,
tumorigenic potential, and anti-apoptotic properties, the PDK1/Akt pathway is a challenging target to
develop new potential agents able to affect GBM resistance to chemotherapy. In an effort to find new
PDK1/Akt inhibitors, we rationally designed and synthesized a small family of 2-oxindole derivatives.
5
October 2015
Accepted 8 October 2015
Available online 22 October 2015
Keywords:
Among them, compound 3 inhibited PDK1 kinase and downstream effectors such as CHK1, GS3K
a and
PDK1 inhibitors
Kinase inhibitors
Oxindole derivatives
Glioblastoma
GBM stem cells
Multi-target therapy
GS3K , which contribute to GCS survival. Compound 3 appeared to be a good tool for studying the role of
b
the PDK1/Akt pathway in GCS self-renewal and tumorigenicity, and might represent the starting point
for the development of more potent and focused multi-target therapies for GBM.
© 2015 Elsevier Masson SAS. All rights reserved.
1. Introduction
investigate new therapies, there has been little improvement in
survival rate over the last 30 years [4]. In addition, evidence in-
Glioblastoma multiforme (GBM) is the most common and lethal
primary malignancy of the central nervous system (CNS) with a
median survival rate of approximately 14 months [1]. Histologi-
cally, it is characterized by poorly differentiated astrocytes, atypical
nuclei, intense mitotic activity, vascular thrombosis, microvascular
proliferation and necrosis [2,3]. Despite of many efforts to
dicates that a population of tumor cells, i.e. cancer stem cells (CSCs),
contributes to cancer recurrence and ultimate failure of the existing
therapeutic strategies thus delineating GBM as a hierarchically
organized entity [5].
CSCs represent the core of the cancer with great capacity for
self-renewal and high level of therapy resistance, suggesting that
they should be a pivotal therapeutic target in order to achieve a
complete eradication of the tumor itself.
Abbreviation: PI3K, phosphatidylinositol-3-kinase; PDK1, phosphoinositide-
dependent kinase-1; GSK3, Glycogen synthase kinase 3; CHK1, Checkpoint kinase-
; Akt, protein kinase B; OXIDs, 2-oxindole derivatives; GBM, glioblastoma; GSCs,
Glioblastoma-derived stem cells.
As expected, GBM-derived stem cells (GSCs) display selective
resistance to several chemo- and radio-therapies [6]. In particular,
irradiated GBMs express an increased level of the stemness marker
CD133 compared to non-irradiated ones [7]. The resistance in GSCs
may dependon several mechanisms, suchas a high expression of drug
1
*
Corresponding author.
E-mail address: simona.rapposelli@farm.unipi.it (S. Rapposelli).
Equally contributed.
1
http://dx.doi.org/10.1016/j.ejmech.2015.10.020
223-5234/© 2015 Elsevier Masson SAS. All rights reserved.
0

S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
275
efflux pumps [8] and, more importantly, a high DNA repair activity.
Accumulating experiences with targeted therapies in different
tumor types clearly indicate that the understanding of the molec-
ular mechanisms of oncogenesis, as well as the correlation of ge-
netic and pathological features, are pivotal in designing more
effective personalized treatments. The loss of regulatory mecha-
nisms is the result of alterations in expression of genes involved in
cell cycle control, such as proto-oncogenes and tumor suppressor
genes [9].
study deals with (a) the design and synthesis of new inhibitors of
PDK1/Akt pathway, (b) prediction of physicochemical and ADME
(Absorption, Distribution, Metabolism, Excretion) properties, (c)
selection of lead compound(s) through a screening assay on AKT
(PKB alpha) and “PDK1 Direct kinases”(d) characterization of the
mechanism of action of selected lead-compound including kinases
profiling, and antiproliferative activity on different GBM cell lines
and finally (e) exploration of the effects induced on GSCs prolifer-
ation and differentiation.
The PI3K/Akt pathway has been identified as an important
oncogenic pathway in the malignant phenotype of GBM, because
triggers a cascade of downstream signaling events that leads to
tumor growth, survival, invasion into normal brain, and secretion of
angiogenic factors. In addition, greater activation and deregulation
of PI3K/Akt components appears to increase cell proliferation and
inhibit GSCs differentiation, thus contributing to resistance to
chemotherapy [10].
Among the compounds synthesized, the oxindole 3 effectively
reduces the viability on two different human glioblastoma cell
lines, ANGMS-CSS [18] and U118MG, which are poorly responsive to
conventional chemotherapies, such as temozolomide (TMZ).
Moreover, compound 3 inhibits GSC self-renewal and proliferation,
triggering both apoptosis and differentiation of the stem cell sub-
population. These findings suggest 3 as a novel tool to sensitize
GSCs to traditional chemotherapeutic strategies.
As a consequence, a plethora of new compounds targeting
different nodes of this pathway, such as PI3K inhibitors [11], have
been developed, and specific drugs targeting Akt/mTOR are now in
Phase I/II of clinical trials, alone or in combination with other
chemotherapeutic drugs against GBM [12,13] (Fig. 1).
2
. Results and discussion
2.1. Design and in silico prediction of pharmacokinetic properties
The increasing understanding of the key role played by PI3K/Akt
pathway in resistant cancer phenotype prompted us to design and
synthesize a small series of 2-oxindole derivatives able to inhibit
Akt phosphorylation [14] (Fig. 1). Noteworthy the oxindole is the
central nucleus of different PI3K/Akt pathway inhibitors [15e17].
Starting from these results and aiming to widen the Structure-
Activity Relationships (SAR) of this new class of compounds, this
This study started from our recent findings suggesting that 2-
oxindole derivatives (OXIDs) inhibit the PDK1/Akt pathway in
non-small cell lung cancer (NSCLC) [14]. In order to investigate
more deeply this class of compounds, we performed a systematic
medicinal chemistry optimization through a divergent synthetic
route in three steps. Recent findings demonstrated that the 2-
oxindole nucleus may be considered as a valid framework upon
Fig. 1. Structures of PI3K/Akt/mTOR inhibitors.

276
S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
which to design new ligands able to target the PI3K/Akt. Moreover,
the substituent in 5-position seems to play a detrimental role for
the affinity against different kinases involved in the PI3K/Akt
pathway [19]. In particular, the substitution of 3,4-
dimethoxytetrahydroisoquinoline alkylamido side chain of previ-
ously synthesized oxindole derivatives [14] with a tolylsulfonamido
group, led to dual Akt/mTOR inhibitors such as FC85 [19] (Fig.1). On
these basis, we planned to study more deeply the effects induced by
the substituent in 5-position on the affinity against PDK1 and Akt
kinases. Hence, herein we first explore the 5-position of OXID-
An analysis of NMR spectra allowed us to assign the configura-
tion around the double bond in 3-position of the OXID core. In
particular, the final 3-(1H-imidazole-5-yl)-adducts (3,7,10 and 13)
exist exclusively as Z-isomers, whereas the OXID-derivatives
substituted in 3-position with N-methyl-imidazol-2-yl moiety (2,
6, 9, 12) preferentially adopt the E-isomeric form [25].
On the contrary, we observed that the 2-thienyl substituted
products exist as E/Z mixture even if a sharp prevalence of Z-isomer
has been observed.
framework by the
substitution of 3,4-dimethoxytetra
2.3. Pharmacology
hydroisoquinoline-N-acetamide side chain of 1e3 with a group
showing greater flexibility and conformational freedom such as
2.3.1. Effects induced on ASPC1 cell viability
3
,4-dimethoxybenzylamino-N-acetamide moiety (4). Then, we
For a preliminary evaluation of the cytotoxicity of the synthe-
sized compounds 1e16, we decided to use the ASPC1 cells [26]. This
human pancreatic adenocarcinoma cell line is an appropriate
model to study new inhibitors of PI3K-AKT pathway, since: i) it is
characterized by an aberrant activation of PI3K/Akt pathway and ii)
shows high degree of resistance to currently available chemother-
apeutic drugs [27] and to PI3K inhibitors [26].
replaced the substituent in 5-position with an electron-rich linker
such as the arylurea- (5e10) and arylmethyleneamino-sulfonyl
moieties (11e13), or with a lesser hindered group, such as the
methanesulphonylamido ones (14e16).
Since a fast and trustworthy prediction of biological properties is
a key factor to accelerate the process of drug discovery, we estimated
in silico the main numerical and topological indices together with the
molecular transport properties (i.e. blood brain barrier penetration)
by PaDEL Descriptor [20] (MW, nHBA, nHBD, nRotB, nRotBt, and
LipinskiFailures) and by ChemAxon MarvinSketch 6.3.0 (LogP, LogD,
TPSA, and pKa) [21]. Moreover, based on aforementioned calculated
molecular features, a druglikeness central nervous system multipa-
rameter optimization (CNS MPO) algorithm recently published by
Wager et al. [22] was applied to compounds under investigation. The
CNS MPO desirability score for CNS drugs should be ꢀ4 on 0e6 scale.
According to the Lipinski's rule, molecules with good membrane
permeability have LogP ꢁ5, molecular weight ꢁ500, number of
hydrogen bond acceptors ꢁ10, and number of hydrogen bond donors
All the compounds synthesised (1e16) were tested in triplicate
at a concentration of 10 mM for 72 h (Table 2). The results show that,
within the 2-oxindoles displaying the 3,4-dimethoxytetra
hydroisoquinoline alkylamido side chain in 5-position, only com-
pound 3 reduces the cell viability to around 68%. The substitution of
this moiety with the thiomethylphenylureido- (5e7) or with the
methylsulphonylphenylureido-groups (8e10) is deleterious for the
activity against the ASPC1, whereas the replacement with a small
electron rich group such as the methanesulphonamido group
(14e16), affects only slightly the cell viability compared to com-
pound 3 (ASPC1 cell viability of 68% and 74% for 14 and 16,
respectively). On the contrary, the replacement of the 3,4-
dimethoxytetrahydroisoquinoline-N-acetamide side chain with a
substituent endowed of a greater flexibility and conformational
freedom such as the 3,4-dimethoxybenzylamino-N-acetamido
moiety (4) or with the 3,4-dimethoxyaminosulphonyl group
(11e13), induces a decrease of the antiproliferative activity (z85%
cell viability). Bearing in mind the high resistance to conventional
therapy reported for the AsPC1 cell line, the reduction of the pro-
liferative activity of around 30% induced by compounds 3, 14 and 16
prompted us to further investigate them on PDK1/Akt pathway.
ꢁ
5. On the basis of the calculated values (Table 1), we synthesized
the designed compounds since all molecules seemed to satisfy the
Lipinsky's rule of five [23].
2.2. Synthesis of new OXID-derivatives
The final compounds 1e3 were synthesised as previously re-
ported [14,19] whereas compounds 4e10 were obtained following
the synthetic procedure illustrated in Scheme 1. The catalytic
reduction of 5-nitro-2-oxindole gave the 5-amino-2-oxindole 17
which was then reacted with 4-(methylthio)phenyl-isocyanate
providing the ureido-intermediate 18. The subsequent Knoevena-
gel reaction with the appropriate carbaldehydes gave the com-
pounds 5e7. Oxidation of 5e7 with Oxone (KHSO
4 2 4
KHSO /0.5 K SO ) in MeOH-THF yielded to final products 8e10.
2.3.2. In vitro screening assay on Akt and PDK1 enzymes
Because we focused on the discovery of new molecules able to
affect the PDK1/Akt pathway, the selected OXID-derivatives 3, 14
and 16 were consequently submitted to a screening assay on AKT
(PKB alpha) and PDK1 direct kinases [28]. Compounds 3, 14 and 16
displayed an appreciable inhibitory activity on PDK1 with IC50
®
5
/0.5
The 5-amino-2-oxo-indole 17 was also submitted to a reaction with
methanesulfonyl chloride affording methansulphonamido inter-
mediate 19, which was reacted with thiophene-2-carbaldehyde,
values in the micromolar (14: IC50 ¼ 6.81
mM) and submicromolar
ranges (3 and 16: IC50 ¼ 0.98 and 0.31 M, respectively), whereas
m
(N-methyl)-imidazole-2-carbaldehyde
or
1H-imidazole-5-
no inhibition on the recombinant Akt enzyme was observed. The
data shows that the replacement of the tetrahydroisoquinoline
nucleus in 5-position of the oxindole core (3), with the meth-
anesulphonamido chain (16) seems not to affect the PDK1-affinity.
On the contrary, the substitution of the imidazole in 3-position of
16, with a thiophene (14), dramatically reduced the potency of
about 7 folds. This let us to speculate that the decrease of potency
could depend, at least in part, to the existence of 14 as a mixture of
E/Z-isomers.
carbaldehyde affording compounds 14e16, respectively. Finally,
starting from intermediate 21, easily obtained by a nucleophilic
substitution of the a-chloroacetamido-derivative 20 with the 3,4-
dimethoxybenzylamine, the reaction with thiophene-2-
carbaldehyde gave the final product 4.
The arylmethyleneamino-sulfonyl derivatives 11e13 were ob-
tained starting from 2-oxo-indole 22. (Scheme 2). Derivative 22 was
obtained by a handy and efficient one-pot reductive cyclization of
2
-nitrophenylacetic acid with iron and acetic acid [24]. The
following reaction with chlorosulphonic acid gave the 5-(chlor-
osulfonyl)indolin-2-one 23 as intermediate which reacted with 3,4-
dimethoxybenzyamine affording the adduct 24. Finally, the Knoe-
venagel reaction with the appropriate carbaldehyde gave final
product 11e13.
2.3.3. In vitro activity on GBM cell lines
In order to establish the proof of concept for the development of
Akt-PDK1 inhibitors for GBM, compound 3 was selected and
investigated on two different human GBM cell lines (ANGMS-CSS
and U118MG). These human cell lines show an aberrant activation

S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
277
Table 1
Molecular descriptors related to ADME and drug-likeness properties of the investigated compounds.
Compd
R
R
1
MWb
nHBAb
9
nHBDb
2
nRotBb
7.0
nRotBtb
9.0
LogPc
3.76
LogDc
3.74
TPSAc
pKac
6.04
LF
0
CNS MPOd
1a
475.56
108.14
2.77
2a
473.52
10
2
7.0
10.0
2.17
2.13
97.72
5.81
0
4.31
3a
4
459.50
449.52
407.51
10
7
2
3
3
7.0
9.0
6.0
9.0
11.0
7.0
1.82
3.33
4.95
1.81
2.65
4.95
105.68
121.51
123.77
2.75
7.98
0
0
0
4.18
3.06
0.74
5
6
11.14
6
7
8
9
405.47
391.45
439.51
437.47
423.44
7
7
7
9
9
3
3
3
3
3
6.0
6.0
6.0
6.0
6.0
8.0
7.0
7.0
8.0
7.0
3.06
3.02
3.16
1.28
1.23
3.03
3.02
3.16
1.24
1.23
113.33
121.31
140.99
130.57
138.53
6.5
2.75
10.93
6.5
0
0
0
0
0
3.40
3.27
1.89
3.56
3.63
1
1
1
1
0
1
2
3
2.75
456.53
454.50
440.47
7
9
9
2
2
2
7.0
7.0
7.0
9.0
10.0
9.0
3.42
1.83
2.12
3.42
1.81
2.12
130.35
119.93
127.89
10.07
6.19
2.5
0
0
0
1.80
3.73
3.74
1
1
4
5
MeSO
MeSO
2
2
320.39
318.35
5
7
2
2
3.0
3.0
4.0
5.0
1.55
1.55
111.89
101.47
9.61
6.19
0
0
3.87
5.03
ꢂ0.03
ꢂ0.06
16
MeSO
2
304.32
7
2
3.0
4.0
0.44
0.44
109.43
2.5
0
4.83
MW ¼ molecular weight; nHBA: number of Hydrogen Bond Acceptors; nHBD: number of Hydrogen Bond Donors; nRotB: Rotational Bonds; nRotBt: terminal Rotational bonds;
LogP ¼ partition coefficient; LogD ¼ distribution coefficient; TPSA ¼ topological polar surface area; pKa ¼ most basic center; LF (Lipinski failures) ¼ number of deviations from
Lipinski's rule; CNS MPO ¼ druglikeness central nervous system multiparameter optimization.
a
Ref. [10].
Calculated with PaDEL descriptors.
Calculated with MarvinSketch.
Calculated according to Ref. [13].
b
c
d
of PDK1/Akt pathway due to PTEN mutation [18,29] and are also
characterized by a high grade of resistance against TMZ, an alky-
lating agent widely used to treat GBM. The selection of the oxindole
by (a) the satisfactory and repeatable effect recorded on ASPC1 cells
(Table 2), (b) the predicted ability to penetrate the bloodebrain
barrier (BBB) (Table 1) and (c) the significant cytotoxicity previ-
ously showed in the A549 cell line, which has been reported to be
3
as prototype to further investigate in GBM cells has been driven

278
S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
Scheme 1. a. Reagents and conditions. a) (methylthio)phenyl-isocyanate, DCM/MeOH, 10 Cert, 3e4 h; b) Aryl-carbaldehyde, EtOH, piperidine, 110 C, 12 h; c) Oxone, MeOH/THF,
ꢃ
ꢃ
ꢃ
rt, 12 h; d) MsCl, H
2
O, rt; e) ClCH
2
COCl, H
2
O, rt, 3 h; f) 3,4-Dimethoxybenzylamine, K
2
CO
3
, DMF, CH
3
CN, 82 C, 12 h.
gefitinib-resistant due to the maintenance of an EGFR-independent
activity of the PI3K/Akt pathway [14]. Actually, considering the
in vitro PDK1 inhibitory activity, the antiproliferative activity on
ASPC1 cell line and the CNS MPO score of 4.2, the oxindole 3 was
selected as the best candidate.
Consequently, the antiproliferative effects induced by 3 deriva-
tive on GBM cell lines was also assessed. Noteworthy, both ANGM-
CSS and U-118MG cell lines showed to be unresponsive to the
The significant reduction of cell viability and the proapoptotic
effect of 3 in both resistance cell lines suggest a possible contri-
bution of “off-target” activities. Indeed, PDK1 inhibition has been
proposed as a highly effective mechanism for reducing metastasis
propagation, but also as an unable mechanism, at least as selective
PDK1 inhibitor, in reducing proliferation in cancer cells character-
ized by an aberrant PI3K/PDK1/AKT expression [30]. Therefore, to
elucidate the mechanism of action of OXID-derivative 3, a screening
profile on 60 kinases involved in the PI3K/Akt/mTOR pathway was
performed.
treatment with 100
pound 3 was tested at different concentrations for 72, 96, 120 and
44 h (Fig. 2). Data show that 3 inhibits cell growth in a
mM of TMZ for 72 h (data not shown). Com-
1
concentration-dependent manner, in both cell lines with GI50
values in the low micromolar range (Fig. 2). Among the two cell
lines, ANGM-CSS appears to be more responsive than U-118MG.
2
.3.4. SelectScreen kinase profiling
The study was carried out by SelectScreen Kinase Profiling
Service, Invitrogen-Life Technologies. Among almost 60 protein
kinases screened, compound 3 (1 M) showed a high inhibitory
Treating the cells for 72 h with 3 (10 mM) caused a significant
m
amount of phosphatidylserine externalisation, in the absence of 7-
AAD binding to DNA (Fig. 3A and B), demonstrating that the com-
pound induced early and late apoptosis of ANGM-CSS cells. As
depicted in Fig. 3 (panels C and D) compound 3 induced cell cycle
disturbance, strongly arresting cell-cycle in the G0-G1 phase.
activity towards GSK3 beta (70% inhibition), a significant activity
towards GSK3 alpha and CHK1 (59% and 61%, respectively)
(Table S1 Supporting Information) and a low activity towards ABL1
(55%). Compound 3 modestly inhibited (<45%) any of all the other
tested kinases, including members of both the PKC family (i.e

S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
279
Scheme 2. a. Reagents and conditions. a) Fe, AcOH, reflux, 2 h; b) ClSO
pirrolidine, 110 C.
OH, rt, 1.5 h / 68 C, 1 h; c) 3,4-dimethoxybenzylamine, EtOH, rt, 4 h; d) Aryl-carbaldehyde, EtOH,
ꢃ
2
ꢃ
Table 2
isolated from glioblastoma ANGM-CSS cells as previously reported
[19,33].
The OXID-compound induced a time-dependent inhibition of
GSC proliferation (Fig. 4). Moreover, the anti-proliferative effects
appeared to be concentration-dependent, with an IC50 value of
Cytotoxicity of compounds 1e16 on ASPC1 cell proliferation assessed by cell
counting at 10 M after 72 h (% viable cells compared to control). Data were
expressed as mean values ± SEM (n ¼ 3).
m
Compd
Cell counting ±S.D.
Compd
Cell counting ±S.D.
5
30 ± 48 nM after 7 days of incubation (Fig. 4).
1
2
3
4
5
6
7
8
82.04 ± 3.44
82.34 ± 2.87
68.14 ± 2.23
86.75 ± 0.76
77.29 ± 9.93
95.13 ± 3.46
82.94 ± 1.42
93.88 ± 5.53
9
10
11
12
13
14
15
16
90.57 ± 3.88
82.21 ± 9.43
84.18 ± 6.67
84.50 ± 4.75
n.d.
68.06 ± 11.08
96.23 ± 3.55
73.83 ± 11.19
a
2.3.6. Effects of 3 on GSC apoptosis and cell cycle block
We then investigated whether the reduction of GSC prolifera-
a
tion elicited by 3 could be associated to apoptosis. Treating GSC
with 3 for 7 days caused a significant phosphatidylserine exter-
nalization in the absence of 7-amino-actinomysin binding to DNA,
thus denoting the induction of early apoptosis (Fig. 5). In addition,
cell cycle analysis demonstrated that challenging GSCs for 7 days
with 3 induced a slight cell cycle disturbance, arresting cell-cycle
progression in the G2-M phase (Fig. 5), consistent with a CHK1
inhibition [19].
a
a
Reduction of the proliferative activity of around 30%.
AMPK) and the MAPK family (i.e. ERK, JNK). These data indicate that
inhibits PDK1 with IC50 980 nM and that, at this concentration
M) it possesses also a high inhibitory activity towards CHK1 and
3
(1 m
GSK3A and GSK3B, exhibiting IC50 values in the micromolar/sub-
micromolar (Table 3). These data revealed that OXID-derivative 3 is
a novel focused multi-target PDK1/Akt pathway inhibitor.
2
.3.7. Effects of 3 on morphology and differentiation of GSCs
The effect of 3 on GSC morphology was evaluated quantifying
the area occupied by the cells in culture plates, as well as the
outgrowth of cellular processes. Representative photographs of
control and treated cells are shown in Fig. 6. Compound 3 reduced
the area occupied by the floating spheres after 7 days of GSC
treatment, and led to a prominent outgrowth of cellular processes
(Fig. 6), suggesting that the compound could also induce GSC dif-
ferentiation. To confirm this hypothesis, the levels of stem and
differentiation markers upon stimulation with 3 were assessed by
Real Time RT-PCR analysis. In GSCs treated with 3 a significant
2.3.5. Effects of 3 on GSC proliferation
Recent evidences demonstrate that GSK-3 inhibition can inhibit
proliferation and induce differentiation of GSCs [31]; moreover, a
combined PDK1 and CHK1 inhibition has been reported to play a
pivotal role in GSC apoptosis [32]. These findings suggest that the
simultaneous inhibition of GSK-3, PDK1 and CHK1 may deplete the
GBM stem cell reservoir. Accordingly, 3 was characterized on GSCs,

280
S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
Fig. 2. Histograms represent cell viability on ANGM-CSS cell line (A) and U118MG (B) after 72, 96, 120 and 144 h of no treatment (Control), treatment with several concentrations of
compound 3 (2.5, 5, 10, 20, 40 M) or vehicle (DMSO 0.1%). The GI50 values recorded at each time, are also shown. Vertical bars indicate the standard error. The asterisk (*) indicates a
value of viability significantly different from the value of viability of the corresponding control group (*P < 0.05; **P < 0.01; ***P < 0.001).
m
decrease in the stemness markers, CD133 and Nestin, was noticed,
accompanied by a significant increase of Olig, MAP2 and GFAP
content. These data demonstrate that the compound was able to
promote GSC differentiation toward a neuronal and a glial pheno-
type. Comparing the morphological analysis and RT-PCR data
indicated that the compound triggered both apoptosis and differ-
entiation of GSCs.
the synthesis and the in vitro evaluation of a small family of OXID-
derivatives (1e16). Among the new compounds, 3 showed to be a
good candidate to further investigate in chemoresistant GBM cell
lines, because it proved to have: i) a CNS MPO score ꢀ4 (require-
ment for CNS drugs); ii) a significant cytotoxicity against a high
resistant cell line (ASPC1), in which the pathway is aberrantly
expressed, and iii) it inhibited PDK1 kinase, as well as some
downstream effectors such as CHK1, GS3Ka and GS3Kb.
Aiming at establish the proof of concept for the development of
Akt-PDK1 inhibitors for GBM, compound 3 was investigated on
different human glioblastoma cell lines: ANGMS-CSS and U118MG.
Results showed that compound 3 proved to induce a concentration-
dependent inhibition of cell growth in both cell lines, even if
ANGM-CSS appears to be more responsive than U-118MG. There-
fore, considering the significant antiproliferative effect of com-
2
.3.8. Effects of 3 on the formation of GSCs
Next, on the light of GSK3's role in GSC self-renewal [31], the
ability of 3 to affect the formation of GSCs was examined. To this
purpose, adherent ANGM-CSS cells were switched to a defined
serum-free NSC medium, and cells were allowed to growth for 9
days, in the absence or in the presence of 3. As depicted in Fig. 7, 3
reduced the number and the diameter of newly generated neuro-
spheres. These results confirmed the importance of GSK3 not only
in the maintenance of cancer stem profile, but also in their trans-
formation process.
pound 3 (GI50 ¼ 7.73
m
M in ANGCSS and GI50 ¼ 14.6
mM at 72 h
U118MG) with respect to TMZ on both cell lines, we hypothesized
that the whole cytotoxic effects resulted from a synergic inhibition
of other upstream and/or downstream effectors of PDK1 enzyme. In
order to validate such hypothesis and to determine the mechanism
responsible for the higher activity on GBM cell lines, a kinase
profiling assay has been performed.
3
. Conclusion
Since the aberrant activation of the PDK1/Akt pathway has been
Most importantly, 3 inhibited GSC self-renewal and prolifera-
tion, confirming that a combined inhibition of PDK1 and CHK1 can
represent an effective therapeutic approach to reduce the growth of
human GBM [32]. Moreover, OXID 3 promoted GSC differentiation
toward a neuronal and a glial phenotype. This effect could be due, at
reported as a pivotal player in many cancer types as well as in the
stem cell subpopulation, the possibility to develop new molecules
that target different steps involved in (and/or downstream of) this
signaling system, represents a novel strategy to improve the
treatment outcome.
The compelling need to develop new and more effective ther-
apies for the treatment of unresponsive tumors such as GBM,
prompted us to design and synthesized new inhibitors of PDK1/Akt
pathway. Recent findings demonstrated that 2-oxindole nucleus
may be considered as a good starting point to be optimized in order
to obtain new targeted ligands. Hence, in this paper we described
least in part to the inhibition GSK3b [31].
In the whole, compound 3 appears to be a good tool to inhibit
GBM proceeding, also inducing stem cell differentiation, by the
inhibition of PDK1/Akt pathway involved in promoting tumor
growth and thus contributing to anticancer therapy resistance [19].
The OXID-derivative 3 could mark the first step towards the

S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
281
Fig. 3. Compound 3 induced apoptosis and cell cycle arrest. A) ANGM-CSS cells were treated for 72 h with DMSO (control), or 3 (10
m
M). At the end of the treatment period, the cells
were collected, and the level of phosphatidylserine externalisation was evaluated using the Annexin V-staining protocol. B) The data are expressed as the percentage of apoptotic
cells (the data for the early-stage apoptotic cells shown in white, while the ones for the late-stage apoptotic/necrotic cells shown in grey) versus the total number of cells. C, D)
ANGM-CSS were treated as in A; at the end of the treatment period, cells were collected and the cell cycle was analyzed. The data were expressed as percentage of cell in the
different phases (G0/G1, G2 or S) versus total cell number. All the data shown represent the mean ± SEM of three different experiments. The significance of the differences was
determined with a one-way ANOVA with Bonferroni post-test: *p < 0.05, **p < 0.01, ***p < 0.001 vs. control.
Table 3
6
operating at 400 MHz, in a ~2% solution of DMSO-d , unless
Results from SelectScreen kinase profiling service (Invitrogen-LifeTechnologies) of 3.
otherwise stated. The >95% purity of tested compounds was
confirmed by combustion analysis. Reactions were followed by
thin-layer chromatography (TLC) on Merck aluminum silica gel (60
[ATP] tested (
mM)
Kinase
IC50(nM)
1
1
1
2
0
0
0
5
CHEK1 (CHK1)
274
884
272
998
F254) sheets that were visualized under a UV lamp. Melting points
GS3KA (GS3K alpha)
GS3KB (GS3K beta)
PDK1 direct
were determined on a Kofler hot-stage apparatus and are uncor-
rected. The elemental compositions of the compounds (C, H, N)
agreed to within (±) 0.4% of the theoretical values. When the
elemental analysis is not included, compounds were used in the
next step without further purification. Evaporation was performed
in vacuo (rotating evaporator). Sodium sulfate was always used as
development of new focused multi-target inhibitors useful in
countering the proliferation of GBM cell lines and of their stem cell
subpopulation.
®
the drying agent. Celite 545 was used as filter agent. Commercially
available chemicals were purchased from SigmaeAldrich. Com-
pounds 1e3 were synthesized as previously reported [14].
4
. Experimental procedures
4.1. Chemistry
4.2. Procedure for the synthesis of final compounds 4e7
Chemical shifts (
d) are reported in parts per million downfield
from tetramethylsilane and referenced from solvent references.
To a solution of 5-substituted-2-oxoindole (0.39 mmol) in EtOH
1
13
Coupling constants (J) are given in Hz. H NMR and C NMR spectra
of all compounds were obtained with a Varian Gemini 200 spec-
trometer operating at 200 MHz or Bruker TopSpin 3.2 spectrometer
(7 mL) was added the appropriate carbaldehyde (0.43 mmol) and a
catalytic amount of piperidine. The resulting solution was stirred
and refluxed for 12 h, then the solution was evaporated to dryness.

282
S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
Fig. 4. ANGM-CSS-derived GSCs were incubated in Neural Stem Cells (NSC) medium with the indicated concentrations of 3 for three (A) or seven days (A, B). At the end of the
treatment periods, cell proliferation was measured using the MTS assay. The data are expressed as a percentage with respect to that of untreated cells (control), and are the mean
values ± SEM of three independent experiments, each performed in duplicate. IC50 values after 7 days of treatment were calculated from sigmoid doseeresponse curve (B). The
significance of the differences was determined with a one-way ANOVA with Bonferroni post-test: ***p < 0.001 vs. control.
4.2.1. (3Z)-2-((3,4-dimethoxybenzyl)amino)-N-(2-oxo-3-
(C21
H
19
N
5
O
2
S) Calc%: C 62.21, H 4.72, N 17.27, S 7.91. Found%: C
(
thiophen-2-ylmethylene) indolin-5-yl)acetamide (4)
62.35, H 4.89, N 17.03, S 8.03.
The residual material was purified by crystallization from
AcOEt/n-hexane, affording 4 as the Z-isomer (36 mg, 0.08 mmol,
0
4.2.4. (3Z)-N-(4-(methylthio)phenyl)-N -(2-oxo-(3-(1H-imidazol-
5-yl)methylene)indolin-5-yl)urea (7)
ꢃ
1
2
1
CH
CH
0% yield): mp 167e169 C. H NMR (CDCl3):
d
3.37 (dd, 1H, J ¼ 1.8;
5.1 Hz, CH
2
); 3.55 (d, 1H, J ¼ 12.9 Hz, CH
2
); 3.75 (d, 1H, J ¼ 15.1 Hz,
The residual material was purified by crystallization from EtOH,
affording 7 as the Z-isomer (98 mg, 0.25 mmol, 63% yield): mp
2
); 3.86 (s, 3H, OMe); 3.89 (s, 3H, OMe); 4.00 (d, 1H, J ¼ 12.9 Hz,
ꢃ
1
2
); 6.66 (d, 1H, J ¼ 8.2 Hz, Ar); 6.79e6.96 (m, 3H, Ar); 7.11 (dd,1H,
215e217 C. H NMR:
d
2.44 (s, 3H, SMe); 6.82 (d, 1H, J ¼ 8.3 Hz, Ar);
J ¼ 3.6, 5.1 Hz, Ar); 7.32e7.35 (m, 1H, Ar); 7.58 (s, 1H, H-vinyl); 7.62
7.12 (dd, 1H, J ¼ 1.3, 8.3 Hz, Ar); 7.22 (d, 2H, J ¼ 8.6 Hz, Ar); 7.44 (d,
2H, J ¼ 8.6 Hz, Ar); 7.71 (s, 1H, H-vinyl); 7.79 (s, 1H, Ar); 7.86 (d, 1H,
J ¼ 1.3 Hz, Ar); 8.02 (s, 1H, Ar); 8.54 (br s, 1H); 8.70 (br s, 1H); 10.90
(
d, 1H, J ¼ 5.1 Hz); 7.76 (d, 1H, J ¼ 3.6 Hz, Ar); 8.00 (s, 1H, Ar) ppm.
1
3
C NMR (CDCl3):
37.03, 133.48, 128.95, 128.94, 128.93, 128.16, 126.94, 126.09, 124.97,
20.22, 116.49, 111.23, 110.63, 109.22, 78.63, 55.49, 54.44 ppm. Anal.
S) Calc%: C 64.13, H 5.16, N 9.35, S 7.13. Found%: C
4.25, H 5.27, N 9.58, S 7.25.
d 169.84, 167.85, 148.76, 148.14, 141.84, 137.67,
13
1
1
(br s, 1H, NH) ppm. C NMR: d 167.58,153.00,138.60,137.01,135.82,
134.00, 133.27, 128.54, 127.41, 127.03 124.93, 122.07, 120.49, 120.01,
118.58, 111.35, 109.15; 17.18 ppm. Anal. (C20H17N O S) Calc%: C
24 23 3 4
(C H N O
5
2
6
61.37, H 4.38, N 17.89, S 8.19. Found%: C 61.45, H 4.30, N 18.01, S 8.23.
4.3. General procedure for the synthesis of compounds 8-10
A solution of oxone (215 mg, 0.35 mmol) in water was added to a
0
.2.2. (Z)-N-(4-(methylthio)phenyl)-N -(2-oxo-3-(thiophen-2-
ylmethylene)indolin-5-yl)urea (5)
The residual material was purified by crystallization from EtOH,
affording 5 as the Z-isomer (94 mg, 0.23 mmol, 60% yield): mp
4
ꢃ
cooled (0 C) solution of the appropriate derivative 8e10
ꢃ
1
2
46e248 C. H NMR:
Ar); 7.11 (d, 1H, J ¼ 8.2 Hz, Ar); 7.22 (d, 2H, J ¼ 8.4 Hz, Ar); 7.44 (d,
H, J ¼ 8.4 Hz, Ar); 7.84e7.92 (m, 3H, Ar, H-vinyl); 7.96e8.01 (m, 1H,
Ar); 8.03 (s, 1H, Ar); 8.48 (br s, 1H); 8.68 (br s, 1H); 10.49 (br s, 1H,
NH) ppm. 13C NMR:
167.16, 152.65, 137.53, 137.11, 135.69, 133.93,
33.16, 129.54, 127.81, 127.16, 124.93, 121.87, 120.71, 119.91, 118.78,
11.17, 110.27, 109.27, 16.10 ppm. Anal. (C21 ) Calc%: C
1.89, H 4.20, N 10.31, S 15.74. Found%: C 61.74, H 4.27, N 10.02, S
d
2.44 (s, 3H, SMe); 6.78 (d, 1H, J ¼ 8.2 Hz,
(0.29 mmol) in 1:1 MeOH-THF (2 mL). The resulting mixture was
stirred at room temperature for 12 h. Then the solid was filtered off,
and the solution was evaporated. The residual material was diluted
with MeOH and the solid product was separated out from the
2
d
mixture. It was filtered, washed with Et
2
O, and air dried.
1
1
6
1
0
H
17
N
3
O
2
S
2
4.3.1. (3Z)-N-(4-(methylsulfonyl)phenyl)-N -(2-oxo-3-(thiophen-2-
ylmethylene)indolin-5-yl)urea (8)
ꢃ
1
5.55.
(76 mg, 0.17 mmol, 60% yield): mp 213e215 C. H NMR: d 3.16
(
s, 3H, SMe); 6.79 (d, 1H, J ¼ 8.3 Hz, Ar); 7.14 (d, 1H, J ¼ 8.3 Hz, Ar);
0
4
.2.3. (3E)-N-(4-(methylthio)phenyl)-N -(2-oxo-(3-(1-methyl-1H-
7.23 (t, 1H, J ¼ 4.3 Hz, Ar); 7.60e7.92 (m, 6H, Ar); 7.98e8.10 (m, 2H,
imidazol-2-yl)methylene)-indolin-5-yl)urea (6)
The residual material was purified by crystallization from EtOH,
affording 6 as the E-isomer (101 mg, 0.25 mmol, 64% yield): mp
Ar, H-vinyl); 8.75 (br s,1H); 9.29 (br s,1H); 10.52 (br s,1H, NH) ppm.
1
3
C NMR:
129.54, 127.75, 127.14, 125.03, 121.87, 120.98, 118.94, 118.75, 111.17,
110.27, 109.27; 52.70 ppm. Anal. (C21 ) Calc%: C 57.39, H
d 168.40, 152.56, 148.00, 137.71, 136.09, 134.12, 133.18,
ꢃ
1
2
35e237 C. H NMR:
d
2.43 (s, 3H, SMe); 3.90 (s, 3H, NMe); 6.78 (d,
17 3 4 2
H N O S
1
H, J ¼ 8.3 Hz, Ar); 7.21 (d, 2H, J ¼ 8.8 Hz, Ar); 7.36 (s, 2H, Ar); 7.42
3.90, N 9.56, S 14.59. Found%: C 57.54, H 4.01, N 10.10, S 14.78.
(
8
1
d, 2H, J ¼ 8.8 Hz, Ar); 7.53 (s, 1H, H-vinyl); 7.65 (dd, 1H, J ¼ 2.0,
.3 Hz, Ar); 8.49 (br s,1H); 8.64 (br s,1H); 9.12 (d,1H, J ¼ 2.0 Hz, Ar);
0
4.3.2. (3E)-N-(4-(methylsulfonyl)phenyl)-N -(2-oxo-3-((1-methyl-
1
3
0.47 (br s, 1H, NH) ppm. C NMR:
37.00, 138.86, 133.66, 130.76, 129.42, 128.27, 127.52, 123.70, 122.84,
21.55, 119.78, 112.57, 108.97; 34.25, 14.20 ppm. Anal.
d
168.27, 157.45, 140.26, 139.14,
1H-imidazol-2-yl)methylene)indolin-5-yl)urea (9)
ꢃ
1
1
1
(79 mg, 0.18 mmol, 62% yield): mp 229e231 C. H NMR:
d 3.15 (s,
3H, SMe); 3.90 (s, 3H, NMe); 6.81 (d, 1H, J ¼ 8.2 Hz, Ar); 7.37 (s, 2H,

S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
283
Fig. 5. GSCs were treated for 7 days with NSC medium containing DMSO (control), or 1 mM 3. At the end of treatments, the cells were collected and the degree of phosphatidylserine
externalisation was evaluated using the Annexin V protocol. B) The data were expressed as the percentage of apoptotic cells (early-apoptotic in white, late-apoptotic/necrotic in
grey) relative to the total number of cells. C, D) Cell cycle analysis at the end of treatment. The data were expressed as percentage of cell in the different phases (G0/G1, G2 or S)
versus total cell number. All the data shown represent the mean ± SEM of three different experiments. The significance of the differences was determined with a one-way ANOVA
with Bonferroni post-test: *p < 0.05, **p < 0.01 vs. control.
Ar); 7.53 (s,1H, H-vinyl); 7.61e7.71 (m, 3H, Ar); 7.81 (d, 2H, J ¼ 8.8 Hz,
Ar); 8.70 (br s, 1H); 9.10e9.18 (m, 2H, Ar, NH); 10.51 (br s, 1H, NH)
substituted-2-oxoindole (0.39 mmol) in EtOH (7 mL) following
the same procedure described above for the preparation of com-
pounds 4e7.
ppm.13C NMR:
d
168.15,153.45,143.56,139.24,138.50,137.86,134.56,
30.17, 129.71, 128.05, 127.52, 122.73, 122.19, 121.55, 119.53, 112.17,
09.04; 54.25, 35.20 ppm. Anal. (C21 S) Calc%: C 57.66, H 4.38,
1
1
19 5 4
H N O
4.4.1. (3Z)-N-(3,4-dimethoxybenzyl)-2-oxo-3-(thiophen-2-
N 16.01, S 7.33. Found%: C 57.55, H 4.42, N 16.25, S 7.02.
ylmethylene)indoline-5-sulfonamide (11)
The residual material was purified by crystallization from EtOH,
affording 11 as the Z-isomer (110 mg, 0.24 mmol, 62% yield): mp
0
4
.3.3. (3Z)-N-(4-(methylsulfonyl)phenyl)-N -(2-oxo-3-((1H-
ꢃ
1
imidazol-5-yl)methylene)indolin-5-yl)urea (10)
75 mg, 0.18 mmol, 61% yield): mp 215e217 C. H NMR:
s, 3H, SMe); 6.84 (d, 1H, J ¼ 8.2 Hz, Ar); 7.15 (d, 1H, J ¼ 8.3 Hz, Ar);
.62e7.85 (m, 6H, H-vinyl, Ar); 7.89 (s, 1H, Ar); 8.02 (s, 1H, Ar); 8.74
246e248 C. H NMR: d 3.62 (s, 6H, OMe); 3.96 (s, 2H, CH );
2
ꢃ
1
(
d
3.16
6.68e6.82 (m, 3H, Ar); 6.90 (d, 1H, J ¼ 8.1 Hz, Ar); 7.27 (t, 1H,
J ¼ 4.2 Hz, Ar); 7.62 (d, 1H, J ¼ 8.1 Hz, Ar); 7.82e7.94 (m, 1H, Ar);
(
7
(
7.82e7.84 (m, 2H, Ar, NH); 8.04 (s, 1H, H-vinyl); 8.31 (s, 1H, Ar);
13
13
br s, 1H); 9.24 (br s, 1H); 10.93 (br s, 1H, NH) ppm. C NMR:
10.99 (br s, 1H) ppm. C NMR:
d
167.23, 148.39, 147.77, 143.00,
d
1
168.85, 152.25, 144.49, 139.28, 138.46, 135.20, 133.22, 132.82,
28.01, 127.68 124.50, 122.70, 120.11, 119.45, 117.41, 110.91, 109.73;
138.66, 137.10, 135.39, 133.36, 130.38, 129.74, 127.65, 127.08, 124.41,
121.53, 119.82, 118.09, 117.69, 111.31, 109.26; 55.40; 55.20;
5
5.89 ppm. Anal. (C20
H
17
N
5
O
4
S) Calc%: C 56.73, H 4.05, N 16.54, S
20 2 5 2
46.18 ppm. Anal. (C22H N O S ) Calc%: C 57.88, H 4.42, N 6.24, S
7.57. Found%: C 57.04, H 4.28, N 16.77, S 7.78.
14.05. Found%: C 58.10, H 4.40, N 6.08, S 14.23.
4.4. Procedure for the synthesis of final compounds 11e16
4.4.2. (3E)-N-(3,4-dimethoxybenzyl)-3-((1-methyl-1H-imidazol-
2yl)methylene)-2- oxoindoline-5-sulfonamide (12)
Compounds 11e16 were synthesized from the appropriate 5-
The residual material was purified by crystallization from EtOH,

284
S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
Fig. 6. GSCs were treated for 7 days with complete NSC medium containing DMSO (control), or 3 (1
mM). (A) Representative cell micrographs after 7 days of treatment were shown.
(
B) The area of the culture plates occupied by the spheres and (C) the length of cellular processes were scored after 7 days of treatment. The counts represent the mean values ± SEM
of two independent experiments. (D) GSCs were treated as in A. At the end of treatments, Real time RT-PCR was performed using primer for the stem cell markers CD133 and Nestin,
the oligodendrocyte marker Olig, the neuronal marker MAP and for the astrocyte marker GFAP. Data were expressed as fold of change vs the levels of the control (set to 1) and are
the mean values ± SEM of three different experiments. The significance of the differences was determined with a one-way ANOVA with Bonferroni post-test: ***p < 0.001 vs control.
affording 12 as the E-isomer (109 mg, 0.24 mmol, 61% yield): mp
Ar); 9.38 (br s, 1H); 10.57 (br s, 1H, NH) ppm. 13C NMR:
137.83, 137.66, 137.01, 134.27, 131.29, 128.47, 127.25, 124.77, 123.46,
121.27, 114.54, 109.60, 38.26 ppm. Anal. (C14 ) Calc%: C
52.48, H 3.78, N 8.74, S 20.02. Found%: C 52.64, H 3.66, N 8.89, S
20.24.
d 167.09,
ꢃ
1
2
35e237 C. H NMR:
d 3.59 (s, 3H, OMe); 3.67 (s, 3H, OMe); 3.92 (s,
3
H, Me); 3.98 (d, 2H, J ¼ 6.1 Hz, CH2); 6.68e6.82 (m, 3H, Ar); 6.96
12 2 3 2
H N O S
(
d, 1H, J ¼ 8.2 Hz, Ar); 7.39 (s, 1H, Ar); 7.46 (s, 1H, Ar); 7.56 (s, 1H, H-
vinyl); 7.68 (dd, 1H, J ¼ 1.8, 8.2 Hz, Ar); 7.92 (br t, 1H, J ¼ 6.1 Hz);
13
9
.92 (d, 1H, J ¼ 1.8 Hz, Ar); 10.99 (br s, 1H) ppm. C NMR:
48.39, 147.77, 145.27, 142.14, 133.73, 130.60, 130.01, 128.59, 125.72,
23.84, 121.53, 119.69, 119.36, 111.57, 108.93; 55.49; 55.18; 46.04;
3.20 ppm. Anal. (C22 S) Calc%: C 58.14, H 4.88, N 12.33, S
.06. Found%: C 58.02, H 5.16, N 12.06, S 7.30.
d 169.27,
1
1
3
7
4.4.5. (3E)-N-(2-oxo-(3-(1-methyl-1H-imidazol-2yl)methylene)
indolin-5-yl)methanesulfonamide (15)
The residual material was purified by crystallization from EtOH,
affording 15 as the E-isomer (51 mg, 0.16 mmol, 40% yield): mp
22 4 5
H N O
ꢃ
1
1
1
7
(
1
1
4
97e199 C. H NMR:
H, J ¼ 8.2 Hz, Ar); 7.11 (dd, 1H, J ¼ 2.0, 8.2 Hz, Ar); 7.34 (s, 1H, Ar);
.38 (s, 1H, Ar); 7.52 (s, 1H, H-vinyl); 9.37 (d, 1H, J ¼ 2.0 Hz, Ar); 9.42
d 2.99 (s, 3H, Me); 3.90 (s, 3H, NMe); 6.82 (d,
4.4.3. (3Z)-3-(1-(1H-imidazol-5-yl)-ethylidene)-N-(3,4-
dimethoxybenzyl)-2-oxoindoline-5-sulfonamide (13)
The residual material was purified by crystallization from EtOH,
affording 13 as the Z-isomer (110 mg, 0.25 mmol, 64% yield): mp
1
3
br s, 1H); 10.57 (br s, 1H, NH) ppm. C NMR:
36.81, 135.45, 131.89, 131.56, 127.46, 123.59, 123.24, 121.13, 117.67,
10.02, 45.07, 38.20 ppm. Anal. (C14 S) Calc%: C 52.82, H
.43, N 17.60, S 10.07. Found%: C 52.78, H 4.73, N 17.52, S 10.25.
d 168.92, 139.65,
ꢃ
1
3
15e320 C. H NMR:
d
3.34 (s, 6H, OMe); 3.98 (d, 2H, J ¼ 8.1 Hz,
14 4 3
H N O
CH2); 6.68e6.80 (m, 3H, Ar); 7.02 (d, 1H, J ¼ 8.0 Hz, Ar); 7.40 (s, 1H,
Ar); 7.56e7.70 (m, 2H, Ar); 7.80e7.93 (m, 2H, H-vinyl, NH); 8.12 (s,
1
3
1
1
1
C
H, Ar); 10.00 (br s, 1H) ppm. C NMR:
43.18, 134.33, 133.73, 133.02, 129.96, 128.58, 127.41, 126.08, 125.63,
24.10, 122.02, 119.82, 111.28, 109.74; 55.45; 55.33; 46.15 ppm. Anal.
S) Calc%: C 57.26, H 4.58, N 12.72, S 7.28. Found%: C
7.21, H 4.73, N 12.74, S 7.37.
d 168.96, 148.33, 147.73,
4.4.6. (3Z)-N-(2-oxo-3-((1H-imidazol-5yl)methylene)indolin-5-yl)
methanesulfonamide (16)
The residual material was purified by crystallization from EtOH,
affording 16 as the Z-isomer (55 mg, 0.18 mmol, 45% yield): mp
21
20
H N
4
O
5
5
ꢃ
1
2
05e207 C. H NMR:
d
2.94 (s, 3H, Me); 6.87 (d, 1H, J ¼ 8.1 Hz, Ar);
4.4.4. (3Z)-N-(2-oxo-(3-(thiophen-2-ylmethylene)indolin-5-yl)
7.05 (dd, 1H, J ¼ 1.7, 8.1 Hz, Ar); 7.50 (d, 1H, J ¼ 1.7 Hz, Ar); 7.72 (s,
methanesulfonamide (14)
1H, Ar); 7.83 (s, 1H, Ar); 8.03 (s, 1H, H-vinyl); 9.43 (br s, 1H); 11.02
13
The residual material was purified by crystallization from EtOH,
affording 14 as the Z-isomer (48 mg, 0.15 mmol, 38% yield): mp
(br s, 1H, NH) ppm. C NMR:
125.79, 123.73, 123.11, 122.29, 114.14, 110.22, 109.17, 38.27 ppm.
Anal. (C13 S) Calc%: C 51.31, H 3.97, N 18.41, S 10.54. Found%:
C 51.44, H 4.05, N 18.29, S 10.37.
d 170.04, 139.14, 137.77, 131.18, 127.85,
ꢃ
1
1
85e187 C. H NMR:
d
2.94 (s, 3H, Me); 6.73e6.90 (m, 1H, Ar);
12 4 3
H N O
7.00e7.36 (m, 2H, Ar); 7.54 (s, 1H, Ar); 7.79e8.29 (m, 3H, H-vinyl,

S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
285
Fig. 7. Effects of 3 on the formation of GSCs. Adherent ANGM-CSS cells were switched to NSC medium, and cells were allowed to growth for 9 days, in the absence or in the
presence of 1 M 3 (A) Representative picture of ANGM-CSS neurospheres after 9 days of incubation. The number of the newly formed spheres (B) and the mean diameter (C) were
m
scored using ImageJ. Data represent the mean ± SEM of three pictures of two independent experiments. The significance of the differences was determined with a one-way ANOVA
with Bonferroni post-test: ***p < 0.001 vs control.
4.5. 5-amino-1,3-dihydro-2H-indol-2-one (17)
2
1.00 mmol) in H O (10 mL) was added MsCl (137 mg, 1.2 mmol) at
RT; stirring was continued at room temperature until TLC analysis
shows disappearance of the starting material. Then the solution
was evaporated to dryness. The residual material was diluted
with MeOH and the solid product was filtered out from the
The commercial 5-nitroindolin-2-one (1.00 g, 5.61 mmol) was
hydrogenated in EtOH (70 mL) in the presence of 10% PdeC
315 mg, 2.97 mmol) for 4 h. Then the catalyst was filtered off
(
through Celite, the Celite rinsed with additional EtOH and the so-
mixture, washed with Et
0.38 mmol, 38% yield): mp 196e198 C. H NMR:
Me); 3.47 (s, 2H, CH
); 6.77 (d, 1H, J ¼ 8.2 Hz, Ar); 7.04 (d, 1H,
J ¼ 8.2 Hz, Ar); 7.09 (s, 1H, Ar); 9.37 (br s, 1H); 10.37 (br s, 1H, NH)
2
O, and air dried. (19). (86 mg,
ꢃ
1
lution was evaporated, to give 17 as a brown solid (698 mg,
d 2.88 (s, 3H,
1
4
6
.71 mmol, 84% yield). H NMR (CD
3
OD):
d
3.43 (s, 2H, CH
2
);
O)
2
8 8 2
.63e6.65 (m, 2H, Ar); 6.73e6.74 (m, 1H, Ar) ppm. Anal. (C H N
13
Calc%: C 64.85, H 5.44, N 18.91; Found%: C 65.03, H 5.49, N 18.83.
.6. 1-(4-(methylthio)phenyl)-3-(2-oxoindolin-5-yl)urea (18)
To solution of 5-amino-1,3-dihydro-2H-indol-2-one 17
ppm. C NMR:
14.84, 112.82, 40.73, 37.41 ppm. Anal. (C
7.78, H 4.45, N 12.38, S 14.17; Found%: C 47.91, H 4.68, N 12.51, S
4.28.
d
169.74, 135.61, 133.17, 131.18, 126.85, 123.51,
1
9 10 2 3
H N O
S) Calc%: C
4
1
4
a
(
(
1
148 mg, 1.0 mmol) in 10 mL DCM/MeOH (9:1), 1-isocyanato-4-
methylsulfanyl)benzene (182 mg, 1.1 mmol) was added slowly at
0
4
.8. 2 -chloro-N-(2-oxo-2,3-dihydro-1H-indol-5-yl)acetamide
ꢃ
0
C. The mixture was stirred at the room temperature for 4 h.
(20)
Solid product 18 was separated out from the mixture. It was
filtered, washed with DCM, and air dried. (330 mg, 0.96 mmol, 96%
To a stirred solution of 5-amino-2-oxindole 17 (122 mg,
ꢃ
1
yield): mp 212e214 C. H NMR:
CH
); 6.72 (d, 1H, J ¼ 8.2 Hz, Ar); 7.12e7.23 (m, 3H, Ar); 7.37 (s, 1H,
Ar); 7.41 (d, 2H, J ¼ 8.8 Hz, Ar); 8.48 (br s, 1H); 8.62 (br s, 1H); 10.24
br s, 1H, NH) ppm. 13C NMR:
167.93, 158.66, 145.19,142.75, 133.57,
29.87, 128.45, 127.67, 126.25, 113.33, 109.09; 108.90; 36.71;
5.10 ppm. Anal. (C16 S) Calc%: C 61.32, H 4.82, N 13.41, S
d
2.43 (s, 3H, SMe); 3.52 (s, 2H,
ꢃ
.82 mmol) in acetone (5 mL) and DMF (1.2 mL), cooled at 0 C, was
0
2
added dropwise 2-chloroacetyl chloride (93 mg, 0.82 mmol). The
reaction mixture was stirred at RT for additional 3 h, then the
solvent was evaporated affording derivative 20, which was used
(
1
1
1
d
1
without further purification (180 mg, 0.80 mmol, 98% yield).
NMR: 3.47 (s, 2H, CH ); 4.21 (s, 2H, CH
.33 (d,1H, J ¼ 8.2 Hz, Ar); 7.49 (s,1H, Ar); 10.15 (br s,1H, NH); 10.33
H
15 3 2
H N O
d
2
2
Cl); 6.73e6.78 (m, 1H, Ar);
0.23; Found%: C 61.05, H 4.67, N 13.23, S 10.23.
7
(
13
br s, 1H, NH) ppm. C NMR:
125.95, 118.82, 116.61, 108.73, 43.36, 35.90 ppm. Anal.
ClN ) Calc%: C 53.47, H 4.04, N 12.47. Found%: C 53.67, H
4.19, N 12.59.
d 175.92, 163.90, 139.70, 132.22,
4
.7. N-(2-oxoindolin-5-yl)methanesulfonamide (19)
(C
10
H
9
2 2
O
To a stirred solution of 5-amino-2-oxindole 17 (148 mg,

286
S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
4
.9. 2-((3,4-dimethoxybenzyl)amino)-N-(2-oxoindolin-5-yl)
1H, J ¼ 6.2 Hz); 10.76 (br s, 1H) ppm. 13C NMR:
d 168.20, 148.35,
147.75, 143.02, 134.39, 133.50, 128.04, 127.67, 126.08, 122.41, 111.89,
acetamide (21)
18 2 5
111.31, 109.46; 55.49; 55.20; 46.10; 34.10 ppm. Anal. (C17H N O S)
A solution of K
2
CO
3
(1.19 g, 8.61 mmol) in CH
3
CN (8 mL), heated
Calc%: C 56.43, H 5.01, N 7.73, S 8.85; Found%: C 56.26, H 5.32, N
7.80, S 8.92.
ꢃ
at 80 C, was mixed with a solution of compound 20 (879 mg,
3
.91 mmol) in DMF (1.3 mL) and with a solution of 3,4-
dimethoxybenzylamine (654 mg, 3.91 mmol) in CH
The suspension was stirred at 80 C for 12 h, then the potassium
3
CN (2 mL).
4.13. Molecular descriptor calculation
ꢃ
carbonate was filtered off and the solvent removed. The residue
was diluted with CH Cl and washed with water and brine. The
2 2
organic layer was dried and concentrated.
The crude product was purified by crystallization from AcOEt/n-
hexane, affording compound 21 (462 mg, 1.30 mmol, 33% yield) as a
Molecular descriptors related to ADME and drug-likeness
properties of the investigated compounds were calculated by us-
ing publicly available software. In particular, the following de-
scriptors: MW, nHBA, nHBD, nRotB, nRotBt, and Lipinski Failures
were calculated with PaDEL Descriptor [20]; LogP, LogD, TPSA, and
pKa were calculated with ChemAxon MarvinSketch 6.3.0 [21]; CNS
MPO was calculated according to Wager et al. [22].
1
yellow solid. H NMR (CDCl
3
(
3
):
); 3.86 (s, 3H, OMe); 3.88 (s, 3H, OMe); 6.76e7.93
m, 4H, Ar); 7.22 (d,1H, J ¼ 8.6 Hz, Ar); 7.58 (s,1H, Ar); 8.02 (br s,1H,
2 2
d 3.52 (s, 2H, CH ); 3.78 (s, 2H, CH );
.81 (s, 2H, CH
2
1
3
NH); 9.18 (br s, 1H, NH) ppm. C NMR (CDCl
49.13, 148.86, 139.07, 132.92, 129.16, 125.05, 123.00, 122.55, 120.45,
11.93, 110.66, 56.94, 56.23, 53.70, 51.80, 36.56 ppm. Anal.
) Calc%: C 64.21, H 5.96, N 11.82. Found%: C 63.97, H
3
):
d
178.05, 169.00,
5. In vitro biological studies
1
1
5.1. Cell lines
19 21 3 4
(C H N O
5
.59, N 11.59.
The human AsPC1 cell line derived from human pancreatic
adenocarcinoma. Cells were grown in RPMI-1640 with 10% fetal
4.10. Indolin-2-one (22)
bovine serum, 1% glutamine and 1% penicillinestreptomycin (Sig-
ꢃ
maeAldrich). Cells were cultured in Petri dishes at 37 C in 5% CO
2
Iron chips (1.6 g, 29.40 mmol) were added in one portion to
commercial 2-nitrophenylacetic acid (2.0 g, 11.04 mmol) in of
and 95% air. U118MG and ANGM-CSS cell lines, derived from human
glioblastoma (ATCC, Manassas, VA, USA and IRCCS-CASA, Sollievo
Della Sofferenza Hospital, San Giovanni Rotondo, FG, Italy), were
both cultured in DMEM:HAMS F12 (1:1) (SigmaeAldrich) supple-
glacial acetic acid (25 mL). The resulting mixture was heated to
ꢃ
1
00 C for 4 h, then concentrated to dryness, sonicated in AcOEt and
filtered to remove the insolubles. The filtrate was washed twice
with 1 N HCl, H O, brine, dried over anhydrous sodium sulfate and
concentrated, affording compound 22. (1.40 g, 10.49 mmol, 95%
mented with 2 mM of L-glutamine (Sigma Aldrich),10% Fetal Bovine
2
Serum (FBS, SigmaeAldrich), 100 units/ml penicillin and 100
streptomycin (P/S) in tissue culture flasks at 37 C in a humidified
m
g/ml
ꢃ
ꢃ
1
yield): mp 92e94 C. H NMR (CDCl
3
):
d
3.56 (s, 2H, CH
2
); 6.90 (d,
2
atmosphere of 5% CO .
1
8
H, J ¼ 7.7 Hz, Ar); 6.98e7.06 (m, 1H, Ar); 7.19e7.26 (m, 2H, Ar);
.95 (br s, 1H) ppm. 13C NMR (CDCl
):
d
179.25, 143.38, 128.54,
NO ) Calc%: C
5.1.1. Cell proliferation assay on ASPC1
Cells were plated in 12-well sterile plastic plates, allowed to
attach for 24 h, and then treated in duplicate with indicated com-
3
125.98, 125.12, 122.93, 110.66; 37.10 ppm. Anal. (C
8
H
7
2
72.16, H 5.30, N 10.52; Found%: C 72.32, H 5.35, N 10.48.
pounds at 10 mM. After 72 h, the number of viable cells was assessed
4.11. 2-oxoindoline-5-sulfonyl chloride (23)
by manual cell counting. Changes in cell growth were expressed as
a percentage relative to cells treated with DMSO (control).
Chlorosulfonic acid (27 mL, 408 mmoli) was slowly added
dropwise to indolin-2-one 23 (13.3 g, 100 mmol). The reaction
5.1.2. Z'Lyte assay on selected kinases
ꢃ
temperature was maintained below 30 C during the addition. Later
The assay was performed by Invitrogen corporation on AKT1
(PKB alpha), PDK1 direct, CHK1, GSK3A and GS3KB as reported
below: (AKT1) The 2ꢄ AKT1 (PKB alpha)/Ser/Thr 06 mixture is
ꢃ
the reaction mixture was stirred at rt for 1.5 h, and then at 68 C for
1
h, cooled, and poured carefully into water. The precipitate formed
was collected by vacuum filtration, washed with H O and dried in a
2
2
prepared in 50 mM HEPES pH 7.5, 0.01% BRIJ-35, 10 mM MgCl ,
vacuum oven to give compound 23, which was used without
further purification (3.02 g, 13.03 mmol, 98% yield): mp
1 mM EGTA. The final 10
AKT1 (PKB alpha) and 2
m
m
L kinase reaction consists of 1.15e25 ng
M Ser/Thr 06 in 50 mM HEPES pH 7.5,
, 1 mM EGTA. After the 1 h kinase re-
L of a 1:2048 dilution of Development Re-
ꢃ
1
7
1
80e185 C. 1H NMR:
d
3.46 (s, 2H, CH
2
); 6.74 (d, 1H, J ¼ 7.7 Hz, Ar);
0.01% BRIJ-35, 10 mM MgCl
action incubation, 5
2
13
.42e7.46 (m, 2H, Ar); 10.48 (br s, 1H) ppm. C NMR:
43.68, 139.71, 124.55, 121.32, 107.39; 35.16 ppm. Anal.
ClNO S) Calc%: C 41.48, H 2.61, N 6.05, S 13.84; Found%: C
1.57, H 2.72, N 6.14, S 13.92.
d
175.89,
m
agent A is added. (PDK1 Direct) The 2ꢄ PDK1 Direct/Ser/Thr 07
(
C
8
H
6
3
mixture is prepared in 50 mM Tris pH 8.5, 0.01% BRIJ-35, 10 mM
4
MgCl2, 1 mM EGTA, 0.02% NaN3. The final 10
consists of 7.36e38.7 ng PDK1 Direct and 2 M Ser/Thr 07 in 50 mM
Tris/HEPES pH 8.0, 0.01% BRIJ-35, 10 mM MgCl , 1 mM EGTA, 0.01%
NaN . After the 1 h kinase reaction incubation, 5 L of a 1:32,768
mL kinase reaction
m
4.12. N-(3,4-dimethoxybenzyl)-2-oxoindoline-5-sulfonamide (24)
2
3
m
To a stirred solution of 2-oxoindoline-5-sulfonyl chloride 23
530 mg, 2.29 mmol) in EtOH (10 mL) was added a solution 3,4-
dilution of development reagent A is added. CHEK1 (CHK1): The 2ꢄ
(
CHEK1 (CHK1)/Ser/Thr 19 mixture is prepared in 50 mM HEPES pH
dimethoxybenzylamine (418 mg, 2.50 mmol) in EtOH (2 mL). The
suspension was stirred at RT for 4 h, then the mixture was
concentrated and the solid collected by vacuum filtration. The re-
sidual material was purified by crystallization from MeOH, afford-
7.5, 0.01% BRIJ-35, 10 mM MgCl
Reaction consists of 2.04e35 ng CHEK1 (CHK1) and 2
in 50 mM HEPES pH 7.5, 0.01% BRIJ-35, 10 mM MgCl2, 1 mM EGTA.
After the 1 h Kinase Reaction incubation, 5 L of a 1:256 dilution of
Development Reagent A is added. GSK3A (GSK3 alpha): The 2ꢄ
GSK3A (GSK3 alpha)/Ser/Thr 09 mixture is prepared in 50 mM
HEPES pH 7.5, 0.01% BRIJ-35, 10 mM MgCl2, 1 mM EGTA. The final
2
, 1 mM EGTA. The final 10
mL Kinase
mM Ser/Thr 19
m
ꢃ
ing compound 24 (664 mg, 1.83 mmol, 80% yield): mp 227e232 C.
1
H NMR:
.88 (d, 1H, J ¼ 6.2 Hz, CH
J ¼ 8.1 Hz, Ar); 7.51 (s, 1H, Ar); 7.60 (d, 1H, J ¼ 8.1 Hz, Ar); 7.89 (br t,
d
3.52 (s, 2H, CH
2
); 3.65 (s, 3H, OMe); 3.69 (s, 3H, OMe);
3
2
); 6.68e6.82 (m, 3H, Ar); 6.90 (d, 1H,
10 mL Kinase Reaction consists of 0.21e1.69 ng GSK3A (GSK3 alpha)

S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
287
and 2
mM Ser/Thr 09 in 50 mM HEPES pH 7.5, 0.01% BRIJ-35, 10 mM
or GSCs were treated with DMSO (control) or compound 3 (1 mM)
MgCl2,1 mM EGTA. After the 1 h Kinase Reaction incubation, 5
a 1:512 dilution of Development Reagent A is added. GSK3B (GSK3
m
L of
for 72 h or seven days, respectively. At the end of the treatment
periods, the percentages of living, apoptotic and dead cells were
acquired and analyzed by Muse™ Cell Analyzer as previously
described [19].
beta): The 2ꢄ GSK3B (GSK3 beta)/Ser/Thr 09 mixture is prepared in
5
0 mM HEPES pH 7.5, 0.01% BRIJ-35,10 mM MgCl2,1 mM EGTA. The
final 10
mL Kinase Reaction consists of 0.17e0.88 ng GSK3B (GSK3
beta) and 2
mM Ser/Thr 09 in 50 mM HEPES pH 7.5, 0.01% BRIJ-35,
5.1.8. Cell cycle analysis in ANGM-CSS cells and in GSCs
1
0 mM MgCl2, 1 mM EGTA. After the 1 h Kinase Reaction incuba-
The measurement of the percentage of cells in the different cell
phases was performed using the Muse™ Cell Analyzer (Merck
KGaA, Darmstadt, Germany). Briefly, U87MG or GSCs were treated
tion, 5 mL of a 1:512 dilution of Development Reagent A is added.
5
.1.3. Protein kinase profiling
Effect of the indicated compounds on the activity of various
kinases was assessed by SelectScreen Kinase Profiling Service
Invitrogen-Life Technologies, Paisley, UK). Assays were performed
with DMSO or compound 3 (1
respectively. Cells were collected and centrifuged at 300 ꢄ g for
5 min. The pellets were washed with PBS and suspended in 100
mM), for 72 h or seven days,
m
l
(
of PBS, and cells were slowly added to 1 mL of ice cold 70% ethanol
ꢃ
using 1uM of compound 3 and ATP concentration as indicated in
the corresponding table.
and maintained o/n at ꢂ20 C. Then, a cell suspension aliquot
(containing at least 2 ꢄ 105 cells) was centrifuged at 300 ꢄ g for
5
min, washed once with PBS and suspended in the fluorescent
5.1.4. Cell proliferation assay on ANGM-CSS and U118MG
reagent (Muse™ Cell Cycle reagent) [34].
Cells were cultured up to about 80% confluence in culture me-
dium and 24 h before the experiment, cells were seeded onto 96-
5.1.9. RNA extraction and Real Time PCR analysis in GSCs
GSCs were treated with DMSO (control), or different concen-
trations of compound 3 (1 nMe1mM) for seven days. At the end of
3
3
well plates at a density per well of 10 for U118MG and 3 ꢄ 10
for ANGM-CSS (only for U118MG it was necessary to pre-coat every
well with gelatine 1% from porcine skin, SigmaeAldrich). After 24 h
to allow cell attachment, the medium was replaced in each 96-well
plate and the cells were treated for 72 h, 96 h, 120 h and 144 h with
treatments, cells were collected, and total RNA was extracted using
®
Rneasy Mini Kit (Qiagen, Hilden, Germany) according to manu-
facturer's instructions. cDNA synthesis was performed with 500 ng
of RNA using i-Script cDNA synthesis kit (BioRad, Hercules, USA)
following manufacturer's instructions. RT-PCR reactions consisted
G51 compound (40
DMSO 0.1%) or control (no-treatment). At the end of each treat-
ments cell viability was assessed using the Cell Proliferation Re-
agent WST-1 (4-[3-(4-iodophenyl)-2-(4-nitrophenyl)-2H-5-
mM, 20 mM, 10 mM, 5 mM and 2.5 mM), vehicle
(
®
®
of 25
10 M forward and reverse primers, 3
reactions were performed for 40 cycles using the following tem-
m
L Fluocycle II SYBR (Euroclone, Milan, Italy), 1.5
m
L of both
m
m
L cDNA, and 19 L of H
m
2
O. All
tetrazolium]-1,3-benzene disulphonate) (Roche, Mannheim, Ger-
many) based on the cellular cleavage of the WST-1 to formazan.
WST-1 was added at 1/10 of the total volume and after 60 min of
ꢃ
ꢃ
perature profiles: 98 C for 30 s (initial denaturation); T C (see
Table S2) for 30 s (annealing); and 72 C for 3 s (extension) [33].
ꢃ
ꢃ
incubation at 37 C, the absorbance was measured at 450 nm with a
multiplate reader (Enspire; PerkinElmer, Wellesley, MA, USA).
5.1.10. Quantitation of the occupied area and the cellular processes
of neurospheres
5
.1.5. GSC isolation from ANGM-CSS cells
To obtain the GSC subpopulation, approximately 2.0 ꢄ 10
GSCs were plated in complete growth medium (day 0) and
treated for the indicated days with compound 3. Photographs of the
neurospheres were taken at days 0, 7, and 21. Three different wells
were analyzed for each condition, and 15 images of each well were
captured [19,33]. The response of the cultures to the various
treatments was quantified by measuring the area occupied by
neurospheres that had formed, using the Image J program (version
1.41; Bethesda, MD, USA). The cellular processes extending from
the six to eight differentiating neurospheres per condition in three
independent experiments were evaluated.
6
ANGM-CSS cells were suspended in 1 mL of a defined serum-free
Neural Stem Cell (NSC) medium [33]. After 5e6 days of culture,
GSCs (called “neurospheres”) were collected, suspended in NSC
medium, dissociated into single cells and plated for the assays. The
method accuracy for GSC isolation was verified in previous experi-
ments [19,33]. For the long-term treatment of cells, NSC or complete
medium containing drugs was replaced every two to three days.
5
.1.6. Cell proliferation assays on GSCs
GSCs were seeded at a density of 3 ꢄ 10 cells/well in 24 well-
4
5.1.11. Neurosphere formation assay
plate, and then treated from one to seven days with fresh growth
The ability of ANGM-CSS to initiate neurosphere formation were
assessed by harvesting, washing, and resuspending monolayer cells
in serum-free NSC medium. Cells were seeded into 96-well at
medium containing different concentrations of compound 3
(
0.1 nMe100 mM). Following the treatment period, GSCs were
4
collected and transferred in 96 well-plate, and cell proliferation was
determined using the MTS assay according to manufacturer's in-
struction. The absorbance of formazan at 490 nM was measured in
a colorimetric assay with an automated plate reader (Victor Wallac
2 ꢄ 10 cells/well and incubated with DMSO (0.5%, control) or
compound 3 (1 mM) for 9 days without disturbing the plates, and
without replenishing the medium. At the end of the treatment
period, pictures of the neurospheres were taken. Three different
wells were analyzed for each condition and 3 images of each well
were captured. The number and the diameter of the newly formed
neurosphere were estimated using the Image J program (version
1.41; Bethesda, MD, USA).
2
, Perkin Elmer). The results were calculated by subtracting the
mean background from the values obtained from each evaluation
and were expressed as the percentage of the control (untreated
cells). Sigmoid doseeresponse curve was generated, from which
the IC50 values were derived.
6. Statistical analysis
5.1.7. Annexin V and 7-AAD staining in ANGM-CSS cells and in GSCs
Dual staining with Annexin V coniugated to fluorescein-
All the experiments were performed in duplicate and repeated
isothiocyanate (FITC) and 7-amino-actinomysin (7-AAD) was per-
formed using the commercially available kit (Muse Annexin V and
Dead Cell Kit; Merck KGaA, Darmstadt, Germany). ANGM-CSS cells
three times. ASPC1 data were expressed as mean values ± S.D. and
analyzed by Student's t test; the level of significance was set at
P < 0.05.

2
88
S. Sestito et al. / European Journal of Medicinal Chemistry 105 (2015) 274e288
[14] G. Nesi, et al., Synthesis of novel 3,5-disubstituted-2-oxindole derivatives as
The nonlinear multipurpose curve-fitting program Graph-Pad
antitumor agents against human nonsmall cell lung cancer, ACS Med. Chem.
Lett. 4 (12) (2013) 1137e1141.
Prism (GraphPad Software Inc., San Diego, CA) was used for data
analysis and graphic presentations. All data are presented as the
mean ± SEM. Statistical analysis was performed by one-way anal-
ysis of variance (ANOVA) with Bonferroni's corrected t-test for
post-hoc pair-wise comparisons. P < 0.05 was considered statisti-
cally significant. The parameter of GI50 was calculated as the molar
concentration of the tested compound evoking a half reduction in
cell viability.
[
[
[
[
[
15] M.D.M. AbdulHameed, et al., Combined 3D-QSAR modeling and molecular
docking study on indolinone derivatives as inhibitors of 3-phosphoinositide-
dependent protein kinase-1, J. Chem. Inf. Model. 48 (9) (2008) 1760e1772.
16] I. Islam, et al., Indolinone based phosphoinositide-dependent kinase-1 (PDK1)
inhibitors. Part 1: design, synthesis and biological activity, Bioorg. Med. Chem.
Lett. 17 (14) (2007) 3814e3818.
17] I. Islam, et al., Indolinone based phosphoinositide-dependent kinase-1 (PDK1)
inhibitors. Part 2: optimization of BX-517, Bioorg. Med. Chem. Lett. 17 (14)
(2007) 3819e3825.
18] A. Notarangelo, et al., Establishment and genetic characterization of ANGM-
CSS, a novel, immortal cell line derived from a human glioblastoma multi-
forme, Int. J. Oncol. 44 (3) (2014) 717e724.
19] S. Daniele, et al., Combined inhibition of AKT/mTOR and MDM2 enhances
glioblastoma multiforme cell apoptosis and differentiation of cancer stem
cells, Sci. Rep. 5 (2015).
Acknowledgment
We thank the International Society for Drug Development
(
ISDD, Milano) who partially supported this research. The authors
[20] C.W. Yap, PaDEL-descriptor: an open source software to calculate molecular
descriptors and fingerprints, J. Comput. Chem. 32 (7) (2011) 1466e1474.
[21] I.V. Tetko, et al., Virtual computational chemistry laboratoryedesign and
description, J. Computer-aided Mol. Des. 19 (6) (2005) 453e463.
[22] T.T. Wager, et al., Moving beyond rules: the development of a central nervous
system multiparameter optimization (CNS MPO) approach to enable align-
ment of druglike properties, ACS Chem. Neurosci. 1 (6) (2010) 435e449.
declare no competing financial interest.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2015.10.020.
[23] D.E. Clark, In silico prediction of bloodebrain barrier permeation, Drug Discov.
Today 8 (20) (2003) 927e933.
[
24] L. Sun, et al., Design, synthesis, and evaluations of substituted 3-[(3-or 4-
carboxyethylpyrrol-2-yl) methylidenyl] indolin-2-ones as inhibitors of VEGF,
FGF, and PDGF receptor tyrosine kinases, J. Med. Chem. 42 (25) (1999)
5120e5130.
References
[
1] T.A. Wilson, M.A. Karajannis, D.H. Harter, Glioblastoma multiforme: state of
the art and future therapeutics, Surg. Neurol. Int. 5 (2014) 64.
2] P. Kleihues, W. Cavanee, World Health Organization Classification of Tumors:
Pathology and Genetic: Tumors of the Nervous System, IARC, Press, 2000 (Ref
Type: Generic).
[25] L. Sun, et al., Synthesis and biological evaluations of 3-substituted indolin-2-
ones: a novel class of tyrosine kinase inhibitors that exhibit selectivity to-
ward particular receptor tyrosine kinases, J. Med. Chem. 41 (14) (1998)
2588e2603.
[
[
[
3] A. Behin, et al., Primary brain tumours in adults, Lancet 361 (9354) (2003)
[26] T. Asano, et al., The PI 3-kinase/Akt signaling pathway is activated due to
323e331.
aberrant pten expression and targets transcription factors NF-
kB and c-Myc in
4] R. Stupp, et al., Effects of radiotherapy with concomitant and adjuvant
temozolomide versus radiotherapy alone on survival in glioblastoma in a
randomised phase III study: 5-year analysis of the EORTC-NCIC trial, Lancet
Oncol. 10 (5) (2009) 459e466.
pancreatic cancer cells, Oncogene 23 (53) (2004) 8571e8580.
[27] I. Murtaza, et al., Suppression of cFLIP by lupeol, a dietary triterpene, is suf-
ficient to overcome resistance to TRAIL-mediated apoptosis in chemoresistant
human pancreatic cancer cells, Cancer Res. 69 (3) (2009) 1156e1165.
[28] E.A. Jares-Erijman, T.M. Jovin, FRET imaging, Nat. Biotechnol. 21 (11) (2003)
1387e1395.
[29] M. Verreault, et al., Combined RNAi-mediated suppression of rictor and EGFR
resulted in complete tumor regression in an orthotopic glioblastoma tumor
model, PloS One 8 (3) (2013) e59597.
[
[
[
[
[
5] S. Colak, J.P. Medema, Cancer stem cellseimportant players in tumor therapy
resistance, FEBS J. 281 (21) (2014) 4779e4791.
6] A. Eramo, et al., Chemotherapy resistance of glioblastoma stem cells, Cell
Death Differ. 13 (7) (2006) 1238e1241.
7] S. Bao, et al., Glioma stem cells promote radioresistance by preferential acti-
vation of the DNA damage response, Nature 444 (7120) (2006) 756e760.
8] A. Shervington, C. Lu, Expression of multidrug resistance genes in normal and
cancer stem cells, Cancer Investig. 26 (5) (2008) 535e542.
9] D. Chen, et al., Discovery of potent N-(isoxazol-5-yl) amides as HSP90 in-
hibitors, Eur. J. Med. Chem. 87 (2014) 765e781.
[30] K. Nagashima, et al., Genetic and pharmacological inhibition of PDK1 in cancer
cells characterization of a selective allosteric kinase inhibitor, J. Biol. Chem.
286 (8) (2011) 6433e6448.
[31] S. Korur, et al., GSK3b regulates differentiation and growth arrest in glio-
blastoma, PloS One 4 (10) (2009) e7443.
[
[
[
10] I. Vivanco, C.L. Sawyers, The phosphatidylinositol 3-kinaseeAKT pathway in
human cancer, Nat. Rev. Cancer 2 (7) (2002) 489e501.
11] P.Y. Wen, et al., Current clinical development of PI3K pathway inhibitors in
glioblastoma, Neuro-oncology 14 (7) (2012) 819e829.
12] K. Shih, et al., Phase I study of the combination of BKM120 and bevacizumab
in patients with relapsed/refractory glioblastoma multiforme (GBM) or other
refractory solid tumors, in: ASCO Meeting Abstracts, 2013.
[32] M. Signore, et al., Combined PDK1 and CHK1 inhibition is required to kill
glioblastoma stem-like cells in vitro and in vivo, Cell Death Dis. 5 (2013)
e1223ee1223.
[33] S. Daniele, et al., Modulation of A1 and A2B adenosine receptor activity: a new
strategy to sensitise glioblastoma stem cells to chemotherapy, Cell Death Dis.
5 (11) (2014) e1539.
[34] S. Daniele, et al., Apoptosis therapy in cancer: the first single-molecule co-
activating p53 and the translocator protein in glioblastoma, Sci. Rep. 4 (2014).
[13] X. Ma, Y. Hu, Targeting PI3K/Akt/mTOR cascade: the medicinal potential,
updated research highlights and challenges ahead, Curr. Med. Chem. 20 (24)
(
2013) 2991e3010.