Identification of an indol-based multi-target kinase inhibitor through phenotype screening and target fishing using inverse virtual screening approach

Article abstract of DOI:10.1016/j.ejmech.2019.01.066

A series of 1,3,5-substituted indole derivatives was prepared to explore the anti-proliferative activity against a panel of human tumour cell lines. A 5-carboxamide derivative (27) emerged as the most potent compound of this series, inhibiting the HeLa cell growth at sub-micromolar concentrations. Target fishing of 27 using a combination of inverse virtual screening (IVS) approach and ligand-based shape similarity study identified the top-ranked targets for 27 as belonging to kinome. These results were further confirmed by in vitro binding assays, leading to the identification of 27 as multi-target kinase inhibitor. The compound 27 was further characterized for its antiproliferative activity by in cell studies, showing a mechanism of action involving modification of the cell cycle, increase in ROS release and caspase 3-expression and decrease in ERK expression.

Full text of DOI:10.1016/j.ejmech.2019.01.066

Accepted Manuscript
Identification of an indol-based multi-target kinase inhibitor through phenotype
screening and target fishing using inverse virtual screening approach
Carmine Ostacolo, Veronica Di Sarno, Gianluigi Lauro, Giacomo Pepe, Simona
Musella, Tania Ciaglia, Vincenzo Vestuto, Giuseppina Autore, Giuseppe Bifulco,
Stefania Marzocco, Pietro Campiglia, Isabel M. Gomez-Monterrey, Alessia Bertamino
PII:
S0223-5234(19)30086-8
DOI:
https://doi.org/10.1016/j.ejmech.2019.01.066
EJMECH 11073
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 22 November 2018
Revised Date: 25 January 2019
Accepted Date: 27 January 2019
Please cite this article as: C. Ostacolo, V. Di Sarno, G. Lauro, G. Pepe, S. Musella, T. Ciaglia, V.
Vestuto, G. Autore, G. Bifulco, S. Marzocco, P. Campiglia, I.M. Gomez-Monterrey, A. Bertamino,
Identification of an indol-based multi-target kinase inhibitor through phenotype screening and target
fishing using inverse virtual screening approach, European Journal of Medicinal Chemistry (2019), doi:
https://doi.org/10.1016/j.ejmech.2019.01.066.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Identification of an indol-based multi-target kinase inhibitor through phenotype screening
and target fishing using inverse virtual screening approach
Carmine Ostacolo1#, Veronica Di Sarno2#, Gianluigi Lauro2, Giacomo Pepe2, Simona Musella2,
Tania Ciaglia2, Vincenzo Vestuto2, Giuseppina Autore2, Giuseppe Bifulco2, Stefania Marzocco2,
Pietro Campiglia2, Isabel M. Gomez-Monterrey1*, Alessia Bertamino2*.
1 Department of Pharmacy, University of Naples Federico II, Via D. Montesano 49, 80131 Naples,
Italy
2 Department of Pharmacy, University of Salerno, Via G. Paolo II, 84084, Fisciano (SA), Italy
Corresponding Authors
*For Alessia Bertamino, Tel: +39-089-969242; email address abertamino@unisa.it
*For Isabel M. Gomez-Monterrey, Tel: +39-081-678633; email address: imgomez@unina.it
# These authors contributed equally to the work.

ACCEPTED MANUSCRIPT
Abstract
A series of 1,3,5-substituted indole derivatives was prepared to explore the anti-proliferative
activity against a panel of human tumor cell lines. A 5-carboxamide derivative (27) emerged as the
most potent compound of this series, inhibiting the HeLa cell growth at sub-micromolar
concentrations. Target fishing of 27 using a combination of inverse virtual screening (IVS)
approach and ligand-based shape similarity study identified the top-ranked targets for 27 as
belonging to kinome. These results were further confirmed by in vitro binding assays, leading to the
identification of 27 as multi-target kinase inhibitor. The compound 27 was further characterized for
its antiproliferative activity by in cell studies, showing a mechanism of action involving
modification of the cell cycle, increase in ROS release and caspase 3-expression and decrease in
ERK expression.
Keywords: 1,3,5-Substituted indole derivatives; Kinase multi-target inhibitor; Target fishing; In
cell studies
Introduction
Progresses in the biology and “omic” science have increased our understanding on the robustness of
complex pathological networks underlying the major chronic diseases and cancer, suggesting that
multi-target drugs may exhibit higher clinical efficacy than single-target drugs [1-3]. This new
appreciation of compounds with poly-pharmacological activities, reinforced by the observation that
known one-target therapeutic agents are in fact multi-target ligands, has supposed a paradigm shift
in the drug discovery field [3-5]. The rational development of this new multi-target-based drug
approach is still a challenge for the scientific community although it has the potential to
,
complement the well know target-based and phenotype-based drug discovery approaches [6, 7]. In
fact, multi-target drugs could address the limitations produced either by the election of a wrong

ACCEPTED MANUSCRIPT
target or by the difficulty to optimize a compound, providing also opportunities for the
application/innovation in the computational and chemoinformatic fields [3, 4, 7-11].
Design of multi-target drugs has proven to be effective in oncology, infectious and CNS diseases [7,
12-14]. In cancer field, small molecules as sorafenib, sunitinib, dasatinib, pazopanib and
regorafenib among others, are now used for the treatment of some leukaemia and solid tumours [15-
20]. These compounds act simultaneously targeting various tyrosine kinase receptors (RTKs) and
small kinases, validating the protein-kinase groups (PKs, kinome) as anticancer molecular targets
[7, 20-22]. Now, the structural similarity of the catalytic site of the different kinases could be
considered as unique opportunity in the design of multi-active kinome inhibitors. Instead, molecules
structurally similar to known kinase inhibitors could be considered as hit/tool compounds in the
detection of new targets or target combinations, and therefore of new biological activities.
During a research program aimed to identify selective and potent varicella zoster virus (VZV)
inhibitors, [23] we observed that a series of 1,3,5-trisubstituted indole derivatives showed a
remarkable cytotoxic activity in the human cervical epitheloid carcinoma cell lines (HeLa) used as
host in the viral activity assay. Starting from these data, new indole derivatives were designed,
synthesized and their cytotoxic activity was tested against a wide panel of cancer cell lines. Giving
the structural similarity of the most potent compound identified in the phenotypic assay with indol-
base kinase inhibitors [24-27], our lead compound was subjected to 1) a molecular computational
study using inverse virtual screening (IVS) approaches to predict its potential PK molecular targets
and 2) an in vitro binding assay on a selected kinase panel to confirm the putatively identified
targets and assess the compound potency. In addition, in cell studies allowed us to further
characterize the biological activity of the selected compound.
Results and Discussion
1.1 Design and Synthesis

ACCEPTED MANUSCRIPT
As indicated above, during the screening for the antiviral activity of an indol-based derivatives
series we have identified analogues of compound 9 and 10 as potential cytotoxic agent (see
structures in the Scheme 1) [23]. Preliminary results of cytotoxic activity (Table 1) showed a strong
toxicity over non-tumour cells for the propyl derivative 9. Thus, our attention has been focused on
the 1-methyl derivative 10, considered as our hit compounds for further modifications. First, we
have designed different modifications at the C-3 indole position, including variation of both the
nature and the dimension of the biphenyl group as well as alkylation of the methyl amino group.
The relevance of the aryl amino methyl group at C-5 indole position on the anti-proliferative
activity of these compounds has been addressed introducing diverse amino acyl group. Methylation
and acylation of the methyl amino group at C-3 was also considered in this series.
The 1,3,5-trisubstituted indole derivatives 9, 10 and 14-17 were obtained as depicted in scheme 1.
Indole-5-carboxaldehyde (1) was reacted with propyl or methyl iodide in DMF using NaH as base
to obtain N1-alkylated intermediates 2 and 3 with excellent yields. Both compounds were treated
with formaldehyde and commercially available amines in Mannich reaction’s conditions giving the
3-amino alkyl intermediates 4-8. Reductive amination of their 5-aldehyde groups with 4-
fluoroaniline, using Na(CH3COO)3BH as reducing agents, lead to the hits 9 and 10. Treatment of
the intermediates 10-13 with formaldehyde in the same conditions gave the final N-methylated
compounds 14-17.

ACCEPTED MANUSCRIPT
R'
NH
R'
NH
F
O
O
O
a
b
c
H
H
H
N
H
N
R
2, R = Propyl
3, R = Methyl
N
N
R
N
H
R
1
4, R = Propyl
5-8, R = Methyl
9,
R = Propyl,
10-13, R = Methyl
4, 5, 9,10,14, R' =
6,11,15, R' =
R'
NH
F
R'
N
F
O
d
N
N
H
H
7,12,16, R' =
8,13,17, R' =
N
N
10-13
14-17
Scheme 1. Synthesis of the 5-(4-fluoroanilinemethyl) indol derivatives 9, 10, 14-17. Reagents: a) alkyliodide (1.5 eq),
NaH (1.5 eq), DMF, 0°C for 30 minutes then RT for 12h; b) CH2O (2.0 eq), RNH2 (2.0 eq), TFA (2.0 eq), DCM, RT, 4-
6h c) 4-F-C6H4NH2 (2.0 eq), CH2Cl2/CH3COOH (5:1, v:v), reflux, 1.5h then Na(CH3COO)3BH (1.8 eq), reflux, 3-5h; d)
CH2O (1.5 eq), MeOH dry, RT for 1.5h then NaBH4 (3.0 eq), RT, 3h.
On the other hand, N-(4-fluorophenyl)-1-methyl-1H-indole-5-carboxamide derivatives 20-26 were
synthesized according to scheme 2A. First, N-1 methylation of indole-5-carboxylic acid (18) and
coupling with 4-fuorophenylaniline yielded the 5-carboxamide intermediate 19. Mannich reaction
of 19 with formaldehyde and different secondary or primary amines led to final piperazine
derivatives 20 and 21, and the secondary amine intermediates (22-26), which were then N-
methylated as above described to obtain the final compounds 27-31.

ACCEPTED MANUSCRIPT
A
F
R
F
N
O
O
O
N
c
a, b
HO
N
N
H
H
N
N
N
H
c
18
19
20, R = CH3 21, R =
R1
F
O
NH
N
H
R1
N
d
22-26
22,27
23,28
25,30
26,31
R1
N
F
OMe
OMe
OMe
O
O
N
H
24,29
N
27-31
B
O
O
O
NH
N
a, b
R
c
d
18
R
R
N
N
N
32-34
35-37
38-40
NH
NH
32,35,38, R =
33,36,39 R =
34,37,40 R =
NH
Scheme 2. Synthesis of the 5-carboxamide derivatives, 20, 21, 27-31, and 38-40. Reagents and conditions: a) CH3I (1.5
eq), NaH (1.5 eq), DMF, 0°C for 30 minutes then RT for 12h b) 4-F-C6H4NH2 (1.2 eq), or 4-CH3- C6H4NH2 (1.2 eq), or
piperidine (1.2 eq), or ethylamine (1.2 eq), PyBop (1.2 eq), DIPEA (2.4 eq), DCM, RT; 12h c) 1-methylpiperazine (2.0
eq) or 1-([1,1'-biphenyl]-4-ylmethyl)piperazine (2.0 eq), or 4-(Ph-Ph)CH2CH2NH2 (2.0 eq), or 4-(PhOPh)CH2NH2 (2.0
eq), or 4-(Ph-Ph)CH2NH2 (2.0 eq), or PhCH2CH2NH2 (2.0 eq), or 3,4,5-(OMe)3PhCH2NH2 (2.0 eq), CH2O (2.0 eq), ,
TFA (2.0 eq), DCM, RT, 4-6h d) CH2O (1.5 eq), MeOH dry, RT for 1.5h then NaBH4 (3.0 eq), RT, 3h.
To complete this SAR study of the 5- carboxamide series a number of 5-carboxamide analogues of
the compound 27 were also synthesized using the same chemical route approach (Scheme 2B). N-
methylation of 18 and coupling with 4-methyl aniline, piperidine, or ethylamine yielded the
intermediates 32-34. Mannich reaction of these intermediates with biphenylethylamine and
formaldehyde led to intermediates 35-37, which are subsequently N-methylated to final compounds
38-40. The effect of the biphenyl moiety position was addressed introducing it at N-1 indole

ACCEPTED MANUSCRIPT
position (Scheme 3). The product 42 was obtained after N-1 substitution of 18 using
biphenylmethyl iodide [28] and coupling of this intermediate with 4-fluoro-aniline. Final
compounds 43 and 44 were obtained as rotamer mixtures by treatment of the corresponding
intermediates 22 and 35 with acetyl chloride, used as acylation agent.
F
O
O
HO
N
H
O
N
a
N
b
HO
42
41
N
H
18
R
R
O
O
NH
N
c
N
N
O
H
H
N
N
43 R = F
44 R = CH3
22 R = F
35 R = CH3
Scheme 3. Synthetic route for the 5-carboxamide target compounds 42-44. Reagents and conditions: a) 4-biphenyl
methyl iodide (1.5 eq), NaH (1.5 eq), DMF, 0°C for 30 minutes then RT for 12h b) 4-F-C6H4NH2 (1.2 eq), PyBop (1.2
eq), DIPEA (2.4 eq), DCM, RT; 12h c) CH3COCl (1.2 mmol), TEA (1.2 mmol), DCM, RT, 1h.
1.2 Pharmacological results
All synthesized compounds were tested for assessing their antiproliferative activity against the
human breast adenocarcinoma (MCF-7), cervical epitheloid carcinoma (HeLa), human prostate
cancer (PC3), lymphocyte cells (Jurkat) as tumor cell lines and normal rat intestinal crypt cells
(IEC-6) as non tumorigenic control cell line. Cell viability was assessed 72 h after treatment with
the tested compounds using the MTT assay. The IC50 values are summarized in Table 1, where was

ACCEPTED MANUSCRIPT
also included the cytotoxic activity of the doxorubicin used as reference compounds. We
investigated the biological effects given by the use of different substituents at the positions N-1, C-3
and C-5 of the indole ring. As indicated above, the propyl derivative 9 showed high potency but
was affected by a high toxicity over IEC-6 cells. Most of the N-methyl derivatives showed
remarkable and selective antiproliferative activity against MCF-7 and HeLa cell lines. Results
obtained with both series, the 5- methyl amino and 5-carboxamide, highlighted the importance and
the specificity given by substitution at the C-3 position. In particular, a biphenyl ethyl moiety linked
to the aminomethyl group appeared pivotal to attain the antiproliferative activity. Indeed,
compounds 9, 14, 17, 22, 27, 47 and 48 featuring IC50 in the micromolar range in at least one
tumour cell line (Table 1), were the most potent compounds. Substitution of the biphenyl ethyl
moiety by phenoxyphenyl methyl (15 and 28), biphenyl methyl (16 and 29), piperazinyl (20, 21), or
phenyl ethyl (30) led to less potent derivatives. Remarkably the substitution of this group by the
most sterically restricted naphtyl methyl (17) resulted in a recovery of the cytotoxic activity but
associated with higher toxicity (IC50 IEC-6 cells = 0.59±0.05 µM). In the 5-carboxamide series,
compound 31, containing a trimethoxybenzyl instead of the biphenyl ethyl group, maintained a
micromolar activity against HeLa and PC3 cell lines. Specificity of the C-3 side chain was also
investigated by substitution on the nitrogen of the aminomethyl group. Methylation of this group
increased the antiproliferative activity. Derivatives 14 and 27, in fact, showed an increased potency
in comparison with their analogues 10 and 22. Contrariwise, the greater rigidity introduced in that
position by the presence of an acetyl group led to a loss of activity of the N-acetyl derivatives (43
and 44) compared to the corresponding N-methyl analogues (27 and 38). As observed in Table 1,
compound 27 belonging to the 5-carboxamide series showed a high potency in all the cell lines
tested, selectivity over HeLa cells (IC50 = 0.33 ±0.03 µM), and reduced toxicity over IEC-6 (IC50 =
6.43±1.2 µM). The introduction of an electron donor group in the aryl ring (compound 38) or the
substitution of aryl by alky groups (39) were well accepted, maintaining activity in the micromolar
range against MCF-7 cell lines and tolerable toxicity over IEC-6. The 5-ethylcarboxamide

ACCEPTED MANUSCRIPT
derivative 40 showed a more toxic profile
,
while switching the biphenyl groups from position C-3 to
position N-1 (compound 42) led to weak cytotoxic selectivity over MCF-7 and PC3 cell lines. For
these reasons compound 27 was selected for the following pharmacological studies.

ACCEPTED MANUSCRIPT
Table 1. Cytotoxic activity of indole derivatives on selected tumor cell lines.
IC50
HeLa
µ
M (± SEM)
Compound
R
R1
CH3CH2CH2
CH3
R2
H
R3
MCF-7
1.26±0.5
PC-3
Jurkat
IEC-6
4-(Ph-Ph)CH2CH2
4-(Ph-Ph)CH2CH2
4-(Ph-Ph)CH2CH2
4-(PhOPh)CH2
4-(Ph-Ph)CH2
7.23±0.1
14.66±3.9
2.92±0.1
9.98±1.8
8.18±0.2
5.41±0.7
2.08±0.2
72.68±3.3
56.03±3.6
84.3±1.6
6.21±0.02
72.00±0.3
26.47±2.7
35.65±0.5
34.43±0.5
1.33±0.7
76.98±3.1
4.76±2.1
48.01±3.2
37.25±1.1
9
H
76.35±0.5
2.15±0.6
10
14
15
16
CH3
CH3
CH3
CH3
CH3
CH3
36.97±0.6
20.19+0.46
CH3
CH3
1-naphtylmethyl
CH3
4.55±0.07
24.70±1.0
0.95±0.5
9.41±2.7
3.68±0.3
>100
0.59±0.1
4.24±1-0
17
20
24.81±3.9
89.23±6.9
4-(Ph-Ph)CH2CH2
4-(Ph-Ph)CH2CH2
4-(Ph-Ph)CH2CH2
4-(PhOPh)CH2
63.89±3.6
7.07±1.1
39.73±4.6
24.3±6,1
0.33±0.1
16.56±2.3
37.54±1.7
45.3±5.2
>100
7.49±1.5
8.34±3.5
6.43±1.2
15.51±5.1
21
22
27
28
4-F-PhNH
4-F-PhNH
4-F-PhNH
CH3
CH3
CH3
H
83.78±8.2
2.7±0.27
51.78±7.3
CH3
CH3
2.16±0.53
20.69±4.1
16.86±2.7
30.78±7.3

ACCEPTED MANUSCRIPT
4-F-PhNH
4-F-PhNH
4-F-PhNH
4-CH3PhNH
piperidine
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
4-(Ph-Ph)CH2
4-PhCH2CH2
54.38±5.6
14.68±2.7
13.21±0.1
3.04±1.21
3.04±1.38
12.20±1.8
10.81±0.5
82.56±8.31
12.77±2.61
13.97±0.12
23.56±2.3
19.56±2.3
4.21±0.1
69.81±3.5
73.17±7.3
>100
>100
>100
>100
5.88±0.2
12.46±2.7
7.9±2.7
29
30
47.70±3.2
6.45±0.4
50.3±3.8
52.19±7.9
87.50±5.6
8.36±0.8
>100
CH3
3,4,5-(OMe)3PhCH2,
4-(Ph-Ph)CH2CH2
4-(Ph-Ph)CH2CH2
4-(Ph-Ph)CH2CH2
42.1±0.1
68.01±1.9
60.41±3.5
79.24±5.5
>100
31
CH3
6.01±3.5
19.60±3.2
4.66±0.5
13.19±2.4
>100
38
CH3
39
CH3CH2NH
4-F-PhNH
4-F-PhNH
4-CH3-PhNH
CH3
40
4-(Ph-Ph)CH2
CH3
67.64±2.5
>100
42
COCH3
COCH3
4-(Ph-Ph)CH2CH2
4-(Ph-Ph)CH2CH2
>100
43
CH3
>100
>100
>100
42.4±0.93
7.82±0.81
44
1.44±0.14
1.07+0.72
2.35±0.11
doxo

ACCEPTED MANUSCRIPT
1.3 Inverse Virtual Screening
In order to tentatively identify the molecular target involved in the antiproliferative mechanism of
derivative 27, we performed an in silico Inverse Virtual Screening (IVS) target fishing
computational approach. [29-35]
In details, IVS allows the analysis of different binding hypotheses between a single ligand and a
high number of targets through structure-based molecular docking experiments. The case-study
compound 27 and twenty-three “blank” compounds (the latters needed for the “normalization” of
the predicted binding affinities, See Experimental Section) were then screened against on a panel of
312 proteins involved in cancer and inflammation events. In this way, the most promising
interacting targets of 27 were selected considering: a) the related normalized ligand/target binding
affinities (See Experimental Section); b) the analysis of the predicted binding mode with the
receptor counterpart; c) evaluating the superposition with co-crystallized known binders of the
identified proteins, when available. In more details, after normalizing the predicted binding
affinities of each 27/protein complex (See Experimental Section), we obtained a ranking from the
highest to the lowest normalized value. This procedure allowed a first selection of the most
promising interacting targets, and then each selected ligand/protein complex was further analysed
by visual inspection. Also, for each selected target, the obtained docking poses of 27 were
compared with the 3D structure of the co-crystallized compound, when available, through the
computation of their ligand based “shape similarity” (See Experimental Section). [36] Specifically,
this parameter indicates the similarity between two chemical species and it ranges between 0 (no
one atom matching between screened and reference compounds) to 1 (all atoms matching). Then, it
is useful for quickly assessing the aptitude of the case-study compound in advantageously binding
the receptor, following the principle that similar molecules could likely show similar binding modes
on the protein counterpart. Combining all these information, the obtained results from the IVS
highlighted the selection of a large set of kinases as promising interacting targets of 27 (See Table
S2, Supporting Information), suggesting a kinome modulation mechanism by the case-study

ACCEPTED MANUSCRIPT
compound. Specifically, among the first 28 top-ranked targets identified by IVS, we identified 14
kinases whose predicted binding with 27 was experimentally evaluated by specific biological
profiling.
1.4 In vitro kinase profiling of 27
To challenge the potential kinase inhibitory activity, compound 27 was assayed in vitro using a
radiometric protein kinase assay (33PanQinase Activity Assay, Proqinase, Germany) and 40
different protein kinases, including the 14 kinases identified by IVS.
Table 2: Ranked IC50 values for compound 27 in protein kinase assays
Target kinase
Target kinase
IC50 (µM)
IC50 (µM)
FGF-R3 wt
TRK-B
SRPK1
6.3
6.3
7.8
7.9
8.8
ERK7
MET wt
ERK1
61.0
61.0
65.0
66.0
66.0
69.0
74.0
79.0
79.0
81.0
88.0
88.0
90.0
99.0
100
ERBB4
ERK2
SRC (GST-HIS-tag)
EGF-R wt
FGF-R4
MAP3K1
ABL1 wt
MEK2
CHK2
WEE1
10.0
11.0
11.0
11.0
13.0
13.0
16.0
18.0
21.0
23.0
24.0
31.0
35.0
43.0
48.0
KIT wt
PDK1
FGF-R1 wt
CLK1
TRK-A wt
MAP3K10
AKT2 aa107-481
FGF-R2
PDGFR-beta
c-RAF
p38-alpha
ERK5
CK2-alpha2
MAP3K9
MEK1 wt
VEGF-R3
AKT2 aa1-481
CK2-alpha1
FAK aa2-1052
ERBB2 wt
VEGF-R2
ROCK1
>100
>100
>100
>100
>100
VEGF-R1
PDGFR-alpha wt
As showed in Table 2, the results obtained by the in vitro assays were in good accordance with the
prediction by IVS. The most relevant affinities data regarded FGF-R3 wt, TRK-B, SRPK1
(predicted by IVS computational screening), ERBB4, SRC (GST-HIS-tag), EGF-R wt (IC50: 6.3-10
µM
). Interestingly, compound 27 also showed a medium micromolar activity (10 µM < IC50 < 50
µM) over further three targets identified by IVS (namely PDK1, CLK1, and AKT2), thus

ACCEPTED MANUSCRIPT
confirming this computational methodology as a useful tool for quickly predicting the targets of
interaction of a bioactive case-study compound. The careful analysis of molecular docking poses
related to the ligand/macromolecule complexes also showed a good accommodation of compound
27 onto the binding sites of SRPK1, PDK1, CLK1, and AKT2 (See Figures S1-S4, Supporting
Information), with the establishment of different contacts (H-bonds, π- π and polar interactions)
with the key residues of the investigated proteins. On the other hand, the comparison of the binding
modes of 27 predicted by docking calculations with those experimentally determined for the ligands
co-crystallized with SRPK1, PDK1, CLK1, and AKT2 highlighted different degrees of similarity
and spatial accommodations in the specific binding pockets (e.g. similar patterns of interactions in
the cases of SRPK1 and PDK1, and different accommodations for CLK1 and AKT2). The
selectivity showed by compound 27 over HeLa cell line could be justified on the basis of its
efficacy over a series of kinase targets, especially growth factors and their receptors such as FGF,
TRK-B and EGFR, which interplay a pivotal role in the progression of cervix carcinoma [37-39],
particularly affecting the epithelial–mesenchymal transition [37, 40, 41]. It is worth of note that the
FGF signalling, strongly affected by compound 27, is also responsible for the proliferation, survival
and invasiveness of several malignancies [42, 43]. Finally, several tyrosine kinases inhibited by
compound 27, particularly TRKB, FGF, EGFR and EGF, act as a mediator of the neurotrophin
signalling, promoting the growth of brain tumour stem cells [44].
2. In vitro pharmacological characterization of compound 27
2.1 Antiproliferative and pro-apoptotic activity and effects on cell cycle distribution
In order to further characterize the pharmacological profile of compound 27, HeLa and IEC-6 cells
viability was assessed after treatment with 27 and the well-known kinase inhibitor TAK-632 [45]
(0.1-0.6 µM) for 24, 48 and 72h. Our data indicated that cell viabilities were affected in a
concentration and time dependent manner (Figure S5, see supporting information). Compound 27

ACCEPTED MANUSCRIPT
showed a higher antiproliferative effect in HeLa cells (Figure S5A-C) and a lower toxic effect in
IEC-6 cells (Figure S5D-F) than the reference TAK-632.
In order to investigate the mechanism(s) underlying the antiproliferative effect, apoptotic death was
performed on HeLa and IEC-6 cells treated with compound 27 and TAK-632 by cytofluorimetric
assays. Our results indicated that 27 is more active on HeLa cells compared with TAK-632 at the
two highest tested concentrations (0.3,0.6 µM; P<0.001 vs 27; Figure 1A). On the contrary, the
effect of 27 on IEC-6 was lower respect to TAK-632 (P<0.001 vs TAK-632; Figure 1C).
The influence of 27 on cell cycle distribution was also determined. As shown in Figure 1B and 1D,
HeLa and IEC-6 cells exposure for 24 h to 27 and TAK-632 (0.1-0.6 µM) resulted in an increase of
the S phase cell cycle distribution (P<0.01 vs control for IEC-6 cells), but mostly G0/G1 phase cell
cycle distribution (P<0.01 vs control for both the cellular lines), accompanied by a significant
decrease in G2 phase compared to untreated cells (P<0.01 vs control; for both the cellular lines).
The effect of 27 on cell cycle distribution was more pronounced in Hela cells and less in IEC-6
cells respect to the effects of TAK-632, thus indicating a stronger pro-apoptotic activity on Hela
cells (Figure 1B) respect to the non tumorigenic IEC-6 cells (Figure 1D).

ACCEPTED MANUSCRIPT
A
B
100
G1
S
90
80
70
60
50
40
30
20
10
0
150
G2
§§§
##
§§§
###
§§§
100
50
0
°°°
***
°°°
***
***
0.6
***
***
0.1
***
0.1
[µM]
[µM]
c
0.3
0.6
0.3
C
0.1
0.3
0.6
0.1
0.3
0.6
27
TAK 632
27
TAK 632
C
D
100
90
80
70
60
50
40
30
20
10
0
G1
S
150
100
50
§§§
òòò
###
§§§
§§§
##
§§§
òò
###
§§§
òò
###
§§§
##
G2
°°°
***
°°°
***
***
0.6
***
**
0
[µM]
[µM]
c
0.1
0.3
0.1
0.3
0.6
C
0.1
0.3
0.6
0.1
0.3
0.6
27
TAK 632
27
TAK 632
Figure 1: Effects of 27 and TAK632 on cell apoptosis on HeLa cells (panel A) and on IEC-6 cells (panel C). Effects of
27 and TAK632 on cell cycle distribution on HeLa cells (panel B) and on IEC-6 cells (panel D). Values are expressed
as mean±s.e.m. of % hypodiploid nuclei or of % cells distribution of three independent experiments. ***, ** denote
P<0.001 and P < 0.01 respectively vs control (untreated cells); °°° denotes P <0.001 vs TAK693. ### ## denote P<0.001
,
and P<0.01 respectively vs control phase G1, òòò and òò denote P<0.001 and P<0.01 respectively vs control phase S. §§§
denotes P<0.001 vs control phase G2. Cell nuclei were analysed with a Becton Dickinson FACScan flow cytometer
using the CellQuest program and the DNA content of the nuclei was registered on a logarithmic scale. Cellular debris
was excluded from the analysis by raising the forward scatter threshold, then the percentage of cells in the hypodiploid
region (sub G0/G1) was calculated. For cell cycle analysis samples were analysed with a FACScan flow cytometer
(Becton Dickinson, CA) using Mod FitLT program.
Considering the importance of ROS as mediators involved in the antitumoral drugs activity, we
investigated the effect of compound 27 and TAK-632 on oxidative stress in HeLa (Figure 2 A) and
IEC-6 cells (Figure 2 B), evaluating intracellular ROS production. Both 27 and TAK-632 at all
tested concentrations (0.1-0.6 µM) induced an increase in ROS release (P<0.05 vs control; Figure 2
A and B). In particular, 27 showed a major ROS production compared to TAK-632 at the two
highest concentrations (0.3, 0.6 µM; P<0.001 vs 27; Figure 3 A) in HeLa cells. Instead, in IEC-6
cells TAK-632 induced a major increase of ROS release compared to 27 (Figure 2 B).

ACCEPTED MANUSCRIPT
B
A
100
80
60
40
20
0
100
80
***
0.6
°°°
***
***
0.6
60
°°°
***
40
***
***
***
0.6
***
0.1
*
20
0
***
***
0.1
*
[µM]
[µM]
0.1
0.3
0.3
c
0.3
0.6
0.1
0.3
c
27
27
TAK 632
TAK 632
Figure 2: Effects of 27, TAK632 on intracellular ROS release on HeLa cells (panel A) and on IEC-6 cells (panel B).
Values are expressed as mean±s.e.m. of ROS release of three independent experiments. ***, * denote P<0.001 and P
<0.05 respectively vs control (untreated cells); °°° denotes P <0.001 vs TAK693. ROS release was analysed with a
Becton Dickinson FACScan flow cytometer using the CellQuest program.
2.2 Effect of 27 on the intracellular apoptotic mediators
Caspase-3 activation by proteolytic cleavage is a marker and an irreversible step in the apoptosis
cascade. Treatment with 27 and TAK-632 (0.1-0.6 µM) increased caspase-3 expression both in
HeLa and IEC-6 cells (P<0.01 vs control; Figure 3 A, B respectively). On the other hand, no effect
on intracellular PARP levels was evidenced for compound 27 and TAK-632 both in tumour and non
tumorigenic cell lines (Figure 3 C, D).

ACCEPTED MANUSCRIPT
A
B
100
80
60
40
20
0
100
80
60
40
***
0.6
***
0.6
***
0.6
***
***
***
0.6
20
0
***
***
0.1
**
**
[µM]
[µM]
c
0.3
0.1
0.3
c
0.1
0.3
0.1
0.3
27
TAK 632
27
TAK 632
D
C
100
80
60
40
20
0
100
80
60
40
20
0
[µM]
c
[µM]
0.1
0.3
0.6
0.1
0.3
0.6
c
0.1
0.3
0.6
0.1
0.3
0.6
TAK 632
27
27
TAK 632
Figure 3: Effects of 27 and TAK632 on caspase-3 and PARP expression on HeLa cells (panel A, C) and on IEC-6 cells
(panel B, D) respectively. Values are expressed as mean±s.e.m. of caspase-3 or PARP expression of three independent
experiments. ***, ** denote P<0.001, P<0.01 respectively vs control (untreated cells); Caspase-3 and PARP expression
was analysed with a Becton Dickinson FACScan flow cytometer using the CellQuest program.
To further verify the kinome modulating properties of compound 27, the intracellular levels of ERK
and pERK after exposure to 27 and TAK-632 were assessed. While compound 27 significantly
reduced the levels of both proteins in HeLa cells (P<0.05 vs control; Figure 4 A and 4C), TAK-632
affected ERK expression at all tested concentrations (P<0.01 vs control; Figure 4A), but pERK
levels were reduced only at the highest concentration assayed (0.6 µM, P<0.001 vs control; Figure
4C). Similarly, in the non tumor cell line, while 27 was unable to modulate the ERK and pERK
levels (Figure 4B and 4D), TAK-632 decreased the level of both proteins at the highest
concentration (P<0.05 vs control; Figure 4B, 4D). Decrease in ERK and pERK expression induced
by the cell treatment with 27 further confirmed the results obtained by binding studies, considering
that all the targeted kinases are correlated to ERK intracellular pathway.

ACCEPTED MANUSCRIPT
A
B
100
80
60
40
20
0
100
80
60
40
**
***
***
20
***
0.1
***
***
0.6
*
0
[µM]
0.3
0.1
0.3
0.6
[µM]
0.6
c
0.1
0.3
0.6
0.1
0.3
c
27
TAK 632
27
TAK 632
D
100
80
60
40
20
0
C
100
80
60
40
20
0
*
***
***
***
0.6
**
[µM]
0.6
[µM]
0.6
0.1
0.3
27
0.6
0.1
0.3
c
0.1
0.3
0.1
0.3
c
27
TAK 632
TAK 632
Figure 4: Comparative effects of 27, TAK632 on ERK and p-ERK expression on HeLa cells (panel A, C) and on IEC-6
cells (panel B, D) respectively. Values are expressed as mean±s.e.m. of ERK or p-ERK expression of three independent
experiments. ***, **, * denote P<0.001, P<0.01 and P <0.05 respectively vs control (untreated cells); ERK and p-ERK
expression was analysed with a Becton Dickinson FACScan flow cytometer using the CellQuest program.
3. Conclusions
In this work we discuss the identification of compound 27, a new indole-based compound, as multi-
kinases inhibitor. The antiproliferative mechanism arising from its multi-kinase inhibition profile
was suggested by IVS studies and confirmed by binding assays on a panel of 40 kinases. In fact,
compound 27 displayed an interesting inhibition against a number of kinases considered important
regulators of cancer progression. In tumour cells, but not in normal cell, 27 modulates the ERK and
pERK levels suggesting a selective action on up-activate kinases correlated to ERK intracellular
pathway. In addition, 27 provokes apoptotic death and it is able to increase the ROS production,
presenting a multi-mechanism anti-proliferative profile. Further studies are ongoing to both identify
new analogues of 27 endowed with anti-tumour activity and selectivity over tumour cell lines and to
examine in depth the multi-kinase inhibitor activity of this series of derivatives.

ACCEPTED MANUSCRIPT
5. Experimental protocols
5.1 General
Reagents, starting materials, and solvents were purchased from Sigma-Aldrich (Milan, Italy) and
used as received. Reactions were carried out with magnetic stirring in round-bottomed flasks unless
otherwise noted. Purifications were conducted on the Biotage Isolera One flash purification system,
using prepacked KP-sil columns, (Biotage, Uppsala, Sweden). Microwave assisted closed vessel
reactions were performed in a Biotage Initiator+ reactor, using 10-20 ml vials type and external
temperature sensor. Analytical thin layer chromatography (TLC) was performed on precoated glass
silica gel plates 60 (F254, 0.25 mm, VWR International). 1D-NMR spectra were recorded with
Bruker Avance (400 MHz) spectrometer, at room temperature. Spectra were referenced to residual
1
13
1
13
chloroform (7.24 ppm, H; 77.23 ppm, C) or methanol (3.31 ppm, H; 49.15 ppm, C).Chemical
shifts are reported in values (ppm) relative to internal Me4Si, and values are reported in hertz
δ
J
(Hz). The following abbreviations are used to describe peaks: s (singlet), d (doublet), dd (double
doublet), t (triplet), q (quartet), and m (multiplet). HR-MS experiments were performed using an
LTQ-Orbitrap-XL-ETD mass spectrometer (Thermo Scientific, Bremen, Germany), using
electrospray ionization.
5.2 General method for the synthesis of 1-alkyl-1H-indole-5-carbaldehyde (2, 3) and 1-alkyl-1H-
indole-5-carboxylic acid derivatives (18a, 41)
Indole-5-carboxaldehyde or indole-5-carboxylic acid (1 or 18, 1.0 mmol) was dissolved in
anhydrous DMF under magnetic stirring, and the temperature was set to 0 °C. To this solution, 1.5
mmol of NaH was added portionwise and the mixture was allowed to react for 30 min. Then, 1.5
mmol of alkyl iodide [methyl iodide or propyl iodide, or 4-[phenyl]iodomethylbenzene] in DMF
was added dropwise and the reaction was warmed to room temperature and maintained under
stirring for further 12 h. Then, reaction was quenched by 10% aqueous solution of citric acid and

ACCEPTED MANUSCRIPT
washed with brine. Organic layer was separated, dried over anhydrous Na2SO4, filtered, and
evaporated in vacuo. Crude products were purified by column chromatography using n-
hexane/ethyl acetate (4:1 v:v) as mobile phase, to obtain intermediate compounds 2, 3, 18a and 41
with 89, 92, 90 and 78% of yields respectively.
5.2.1 1-propyl-1H-indole-5-carbaldehyde (2)
Yield 89%. 1H NMR (400 MHz, CDCl3)
δ; 0.93 (t, 3H, CH3, J = 12.0 Hz); 1.83-1.95 (m, 2H, C
H2);
4.13 (t, 2H, CH2, J = 12.0 Hz); 6.67 (d, 1H, aryl,
J
= 4.0 Hz); 7.20 (d, 1H, aryl, = 4.0 Hz); 7.42 (d,
J
1H, aryl,
J = 8.0 Hz, aryl ); 7.79 (d, 1H, aryl, J = 8.0 Hz, aryl ); 8.16 (s, 1H, aryl); 10.07 (s, 1H,
CHO). HR-MS m/z calcd for C12H14NO [(M + H)]+: 188.1070, found 188.1076.
5.2.2 1-methyl-1H-indole-5-carbaldehyde
Yield 92%. 1H NMR (400 MHz, CDCl3)
1H, aryl, = 4.0 Hz); 7.40 (d, 1H, aryl,
(3)
δ; 3.78 (s, 3H, CH3); 6.71 (d, 1H, aryl,
J = 4.0 Hz); 7.23(d,
J
J
= 8.0 Hz, aryl ); 7.80 (d, 1H, aryl, = 8.0 Hz, aryl ); 8.16
J
(s, 1H, aryl); 10.04 (s, 1H, CHO). HR-MS m/z calcd for C10H10NO [(M + H)]+: 160.0757, found
160.0762.
5.2.3 1-methyl-1H-indole-5-carboxylic acid (18a
Yield 84%. 1H NMR (400 MHz, CD3OD)
)
δ
; 3.85 (s, 3H, C
H3); 6.58 (d, 1H, aryl,
J
= 4.0 Hz); 7.26
(d, 1H, aryl,
8.33 (s, 1H, aryl). HR-MS m/z calcd for C10H10NO2 [(M + H)]+: 176.0706, found 176.0710.
J = 4.0 Hz); 7.43 (d, 1H, aryl, J = 8.0 Hz, aryl ); 7.87 (d, 1H, aryl, J = 8.0 Hz, aryl );
5.2.4 1-([1,1'-biphenyl]-4-ylmethyl)-1H-indole-5-carboxylic acid
Yield 78%. 1H NMR (400 MHz, CD3OD)
(41)
δ
; 5.39 (s, 2H, CH2); 6.61 (d, 1H, aryl, J = 4.0 Hz); 7.22−
7.46 (m, 10H, aryl); 8.40 (s, 1H, aryl). HR-MS m/z calcd for C22H18NO2 [(M + H)]+: 328.1332,
found 328.1339.

ACCEPTED MANUSCRIPT
5.3 General method for coupling reactions. Synthesis of 1-alkyl-1H-indole-5-carboxamide
derivatives (19, 32-34, and 42).
To a solution of 1-alkyl-1H-indole-5-carboxylic acids 18a or 41 (1.0 mmol) in dichloromethane,
different aromatic or aliphatic amines (1.2 mmol), PyBop (1.2 mmol) and DIPEA (2.4 mmol) were
successively added. Stirring was continued at room temperature for 12 h. Afterward, the reaction
mixture was diluted with dichloromethane (20 mL), and the resulting solution was washed
successively with 10% citric acid (2 x 25 mL), 10% NaHCO3 (2 x 25 mL), and water (2 x 25 mL),
dried over Na2SO4 and evaporated to dryness. Flash chromatography of the residues, using a
mixture of ethyl acetate/n-hexane as eluent systems, yielded, in each case, the corresponding 5-
carboxamide indole derivatives in 56-64% of yield.
5.3.1 N-(4-fluorophenyl)-1-methyl-1H-indole-5-carboxamide
Yield 56%. 1H NMR (400 MHz, CD3OD)
7.14 (m, 2H, aryl); 7.30 (d, 1H, aryl, = 4.0 Hz); 7.50 (d, 1H, aryl,
aryl); 7.80 (d, 1H, aryl, = 4.0 Hz); 8.24 (s, 1H, aryl). HR-MS m/z calcd for C16H14FN2O [(M +
(19)
δ; 3.88 (s, 3H, C
H3); 6.60 (d, 1H, aryl,
J = 4.0 Hz); 7.09-
J
J
= 12.0 Hz), 7.69-7.74 (m, 2H,
J
H)]+: 269.1085 found 269.1090.
5.3.2 1-methyl-N-(4-tolyl)-1H-indole-5-carboxamide (32)
Yield 60%. 1H NMR (400 MHz, CDCl3)
δ
; 2.37 (s, 3H, C
H
3); 3.85 (s, 3H, C
= 8.0 Hz); 7.39 (d, 1H, aryl,
= 8.0 Hz); 8.19 (s, 1H, aryl). HR-MS m/z
H
3); 6.60 (d, 1H, aryl,
J
= 4.0 Hz); 7.15 (d, 1H, aryl,
J
= 4.0 Hz); 7.20 (d, 2H, aryl,
J
J
= 8.0
Hz); 7.58 (d, 2H, aryl,
J
= 8.0 Hz), 7.79 (d, 1H, aryl,
J
calcd for C17H17N2O [(M + H)]+: 265.1335 found 265.1341.
5.3.2 (1-methyl-1H-indol-5-yl)(piperidin-1-yl)methanone (33)

ACCEPTED MANUSCRIPT
Yield 64%. 1H NMR (400 MHz, CDCl3)
δ; 1.53-1.76 (m, 6H, CH2); 3.51-3.73 (m, 4H, CH2); 3.82
(s, 3H, C
H3); 6.51 (d, 1H, aryl, J = 4.0 Hz); 7.12 (d, 1H, aryl, J = 4.0 Hz); 7.28-7.35 (m, 2H, aryl);
7.71 (s, 1H, aryl). HR-MS m/z calcd for C15H19N2O [(M + H)]+: 299.2118 found 299.2125.
5.3.3 N-ethyl-1-methyl-1H-indole-5-carboxamide
Yield 62%. 1H NMR (400 MHz, CDCl3)
3.83 (s, 3H, C 3); 6.21 (s, 1NH); 6.56 (d, 1H, aryl,
(d, 1H, aryl, = 8.0 Hz); 7.68 (d, 1H, aryl, = 8.0 Hz); 8.08 (s, 1H, aryl). HR-MS m/z calcd for
(34)
δ; 1.27 (t, 3H, CH3, J = 12.0 Hz); 3.50-3.59 (m, 2H, C
H2);
H
J
= 4.0 Hz); 7.12 (d, 1H, aryl, = 4.0 Hz); 7.34
J
J
J
C12H15N2O [(M + H)]+: 203.1179 found 203.1184.
5.3.4 1-([1,1'-biphenyl]-4-ylmethyl)-N-(4-fluorophenyl)-1H-indole-5-carboxamide (42)
1
Isolated as yellowish oil in 60% yield. H NMR (400 MHz, CDCl3)
δ
; 5.43 (s, 2H, C
= 12.0 Hz, aryl); 7.35-7.47 (m,
= 8.0 Hz, aryl); 7.88 (s,
H2); 6.70 (d,
1H,
J = 4.0 Hz, aryl); 7.09 (t, 2H, J = 8.0 Hz, aryl); 7.20 (d, 2H, J
5H, aryl); 7.55-7.59 (m, 4H, aryl); 7.61-7.67 (m, 2H, aryl); 7.74 (d, 1H,
1NH); 8.24 (s, 1H, aryl). 13C NMR (100 MHz, CDCl3)
J
δ: 50.1, 103.1, 110.0, 115.5, 120.7, 120.9,
121.8, 126.4, 127.2, 127.6, 128.5, 128.8, 130.0, 135.8, 138.2, 160.5, 166.7. HR-MS m/z calcd for
C28H22FN2O [(M + H)]+: 421.1711 found 421.1719
5.4 General method for Mannich reactions. Synthesis of 3-(aminomethyl)-1-alkyl-1H-indole-5-
carbaldehyde derivatives (4-8) and 3-(aminomethyl)-1-alkyl-1H-indole-5-carboxamide derivatives
(20-26, 35-37).
Formaldehyde (2.0 mmol), trifluoroacetic acid (2.0 mmol) and different commercially available
amine (2.0 mmol) were dissolved in dichloromethane and the solution was stirred at room
temperature for 30 minutes. Then, a solution in dichloromethane of the opportune 1,5-disubstituted
indole derivatives
2, 3, 19, 32-34 (1.0 mmol) was added and the mixture was stirring for 4-6 hours,
following the course of reaction by TLC. When the starting compound disappeared, the reaction

ACCEPTED MANUSCRIPT
was quenched by 10% aqueous solution of sodium bicarbonate and washed with brine, dried over
anhydrous Na2SO4 and filtered. Organic phase was evaporated in vacuum and 3-aminomethyl
indole derivatives were obtained after flash chromatography using a mixture of
dichloromethane/methanol (95/5 v:v) as eluent with 65-77% of yield.
5.4.1 3-(((2-([1,1'-biphenyl]-4-yl)ethyl)amino)methyl)-1-propyl-1H-indole-5-carbaldehyde
Yield 71%. 1H NMR (400 MHz, CDCl3)
3.01 (t, 2H, CH2, J = 8.0 Hz); 3.08 (t, 2H, CH2
2); 7.11 (s, 1H, aryl); 7.33-7.40 (m, 4H, aryl); 7.50 (t, 2H, aryl,
8.0 Hz); 7.65 (d, 2H, aryl, = 8.0 Hz); 7.81 (d, 1H, aryl, = 8.0 Hz); 8.25 (s, 1H, aryl); 10.08 (s,
(4)
δ
; 0.95 (t, 3H, CH3, J = 12.0 Hz); 1.85-1.96 (m, 2H, C
= 8.0 Hz); 3.81 (s, 3H, C 3); 4.12-4.19 (m, 4H,
= 8.0 Hz); 7.58 (d, 2H, aryl,
H2);
,
J
H
C
H
J
J =
J
J
1H, CHO). HR-MS m/z calcd for C27H29N2O [(M + H)]+: 397.2274, found 397.2280.
5.4.2 3-(((2-([1,1'-biphenyl]-4-yl)ethyl)amino)methyl)-1-methyl-1H-indole-5-carbaldehyde (5)
Yield 73%. 1H NMR (400 MHz, CDCl3)
Hz); 3.76 (s, 3H, C 3); 4.08 (s, 2H, C 2); 7.09 (s, 1H, aryl); 7.32-7.38 (m, 4H, aryl); 7.46 (t, 2H,
aryl, = 8.0 Hz); 7.57 (d, 2H, aryl, = 8.0 Hz); 7.62 (d, 2H, aryl, = 8.0 Hz); 7.82 (d, 1H, aryl,
δ; 2.95 (t, 2H, CH2, J = 8.0 Hz); 3.07 (t, 2H, CH2, J = 8.0
H
H
J
J
J
J =
8.0 Hz); 8.22 (s, 1H, aryl); 10.05 (s, 1H, CHO). HR-MS m/z calcd for C25H25N2O [(M + H)]+:
369.1961, found 369.1968.
5.4.3 1-methyl-3-(((4-phenoxybenzyl)amino)methyl)-1H-indole-5-carbaldehyde
(6)
Yield 74%. 1H NMR (400 MHz, CDCl3)
δ; 3.83 (s, 3H, CH3); 3.89 (s, 2H, CH2); 4.06 (s, 2H, CH2);
6.99-7.14 (m, 5H, aryl); 7.31-7.41 (m, 6H, aryl); 7.81 (d, 1H, aryl,
J = 8.0 Hz); 8.21 (s, 1H, aryl);
10.06 (s, 1H, CHO). HR-MS m/z calcd for C24H23N2O2 [(M + H)]+: 371.1754, found 371.1760.
5.4.4 3-((([1,1'-biphenyl]-4-ylmethyl)amino)methyl)-1-methyl-1H-indole-5-carbaldehyde (7)

ACCEPTED MANUSCRIPT
Yield 77%. 1H NMR (400 MHz, CDCl3)
δ; 3.79 (s, 3H, CH3); 3.91 (s, 2H, CH2); 4.06 (s, 2H, CH2);
7.26 (s, 1H, aryl); 7.36-7.59 (m, 10H, aryl); 7.81 (d, 1H, aryl,
J = 8.0 Hz); 8.18 (s, 1H, aryl); 10.05
(s, 1H, CHO). HR-MS m/z calcd for C24H23N2O [(M + H)]+: 355.1805, found 355.1810.
5.4.5 1-methyl-3-(((naphthalen-1-ylmethyl)amino)methyl)-1H-indole-5-carbaldehyde
(8)
Yield 65%. 1H NMR (400 MHz, CDCl3)
δ; 3.82 (s, 3H, CH3); 4.14 (s, 2H, CH2); 4.35 (s, 2H, CH2);
7.14 (s, 1H, aryl); 7.37-7.53 (m, 5H, aryl); 7.80-7.91 (m, 3H, aryl); 8.12-8.15 (m, 2H, aryl); 9.97 (s,
1H, CHO). HR-MS m/z calcd for C22H21N2O [(M + H)]+: 329.1648, found 329.1654
5.4.6
20
N-(4-fluorophenyl)-1-methyl-3-((4-methylpiperazin-1-yl)methyl)-1H-indole-5-carboxamide
(
)
1
Isolated as an orange wax in 27% yield. H NMR (400 MHz, CD3OD)
δ
; 2.31 (s, 3H, C
3); 7.06-7.11 (m, 3H, aryl); 7.36 (d, 1H, aryl,
= 8.0 Hz), 8.23 (s, 1NH); 8.34 (s, 1H,
H3); 2.49-
2.61 (m, 8H, CH2); 3.78 (s, 2H, CH2); 3.83 (s, 3H, CH
J
= 12.0 Hz), 7.66.7.72 (m, 2H, aryl); 7.81 (d, 1H, aryl,
J
aryl). 13C NMR (100 MHz, CD3OD)
δ
31.6, 44.4, 51.8, 52.0, 54.1, 109.0, 110.5, 114.6, 114.9,
119.2, 120.8, 122.8, 122.9, 125.3, 127.9, 130.9, 139.0, 157.8, 161.0, 168.6. HR-MS m/z calcd for
C22H26FN4O [(M + H)]+: 381.2085 found 381.2094.
5.4.7
3-((4-([1,1'-biphenyl]-4-ylmethyl)piperazin-1-yl)methyl)-N-(4-fluorophenyl)-1-methyl-1H-
21
indole-5-carboxamide
(
)
Isolated as yellow oil in 30% yield. 1H NMR (400 MHz, CD3OD)
(s, 2H, C 2); 3.81 (s, 2H, C 2); 3.85 (s, 3H, C 3); 7.12 (t, 2H, = 8.0 Hz, aryl); 7.29-7.49 (m, 7H,
= 8.0 Hz), 8.36 (s,
32.9, 33.0, 51.9, 52.1, 52.7, 24.23, 58.6, 109.49, 115.4,
δ; 2.48-2.76 (m, 8H, CH2); 3.57
H
H
H
J
aryl); 7.58 (t, 4H,
J = 7.8 Hz, aryl); 7.69-7.73 (m, 2H, aryl); 7.83 (d, 1H, aryl, J
1H, aryl). 13C NMR (100 MHz, CD3OD)
δ
115.6, 119.4, 121.4, 121.9, 122.0, 126.0, 126.9, 127.1, 127.5, 128.7, 129.2, 130.2, 134.7, 139.0,

ACCEPTED MANUSCRIPT
139.2, 140.8, 158.0, 160.4, 166.8,. HR-MS m/z calcd for C34H34FN4O [(M + H)]+: 533.2711 found
533.2720.
5.4.8 3-(((2-([1,1'-biphenyl]-4-yl)ethyl)amino)methyl)-N-(4-fluorophenyl)-1-methyl-1H-indole-5-
carboxamide
Isolated as yellowish oil in 68% yield. 1H NMR (400 MHz, CDCl3)
2.94 (t, 2H, = 8.0 Hz, C 2); 3.70 (s, 3H, C 3); 3.89 (s, 2H, C 2); 6.87-6.95 (m, 3H, aryl); 7.19-
(22)
δ; 2.85 (t, 2H, J = 8.0 Hz, CH2);
J
H
H
H
7.23 (m, 3H, aryl); 7.28-7.36 (m, 4H, aryl); 7.47-7.50 (m, 4H, aryl); 7.54-7.57 (m, 2H, aryl); 7.71
= 8.0 Hz), 8.44 (s, 1NH). 13C NMR (100 MHz, CDCl3)
(d, 1H, aryl,
J
δ 33.5, 42.4, 52.9, 58.1,
109.6, 115.0, 115.4, 118.5, 122.2, 126.4, 127.7, 129.3, 129.5, 130.7, 133.5, 139.2, 141.1, 158.2,
160.9, 168.7. HR-MS m/z calcd for C31H29FN3O [(M + H)]+: 478.2289 found 478.2296.
5.4.9
N-(4-fluorophenyl)-1-methyl-3-(((4-phenoxyphenethyl)amino)methyl)-1H-indole-5-
carboxamide
(23
)
Yield 71%. 1H NMR (400 MHz, CD3OD)
Hz); 3.87 (s, 3H, C 3); 4.37 (s, 2H, C 2); 6.94 (t, 4H,
aryl); 7.23 (d, 2H, = 7.6 Hz, aryl); 7.33 (t, 2H, = 8.0 Hz, aryl); 7.48 (s, 1H, aryl); 7.53 (d, 1H,
δ; 2.96 (t, 2H, CH2
,
J
= 8.0 Hz); 3.22 (t, 2H, CH2
, J = 8.0
H
H
J
=8.0 Hz, aryl); 7.10 (t, 4H,
J
= 8.0 Hz,
J
J
aryl,
J = 8.0 Hz), 7.68-7.73 (m, 2H, aryl); 7.86 (d, 1H, J=8.0 Hz, aryl); 8.39 (s, 1H, aryl). HR-MS
m/z calcd for C31H29FN3O2 [(M + H)]+: 494.2238 found 494.2242.
5.4.10
3-((([1,1'-biphenyl]-4-ylmethyl)amino)methyl)-N-(4-fluorophenyl)-1-methyl-1H-indole-5-
24
carboxamide
(
)
Yield 70%. 1H NMR (400 MHz, CDCl3)
7.02 (t, 2H, = 7.8 Hz, aryl); 7.09 (s, 1H, aryl); 7.36 (d, 1H,
Hz, aryl); 7.43-7.48 (m, 4H, aryl); 7.57 (d, 4H, = 8.0 Hz, aryl); 7.66-7.70 (m, 2H, aryl); 7.82 (d,
δ; 3.78 (s, 3H, C
H3); 3.95 (s, 2H, C
H2); 4.04 (s, 3H, C
H3);
J
J
= 8.0 Hz, aryl); 7.38 (d, 1H,
J
= 8.0
J

ACCEPTED MANUSCRIPT
1H,
J
= 7.8 Hz, aryl); 8.27 (s, 1H, aryl); 8.33 (s, 1NH). HR-MS m/z calcd for C30H27FN3O [(M +
H)]+: 464.2133 found 464.2138.
5.4.11 N-(4-fluorophenyl)-1-methyl-3-((phenethylamino)methyl)-1H-indole-5-carboxamide
(25
)
Yield 74%. 1H NMR (400 MHz, CDCl3)
Hz); 3.37 (s, 3H, C 3); 4.15 (s, 2H, C
δ
; 2.46 (t, 2H, CH2
,
J
= 8.0 Hz); 3.06 (t, 2H, CH2
,
J
= 8.0
H
H
2); 6.88-7.04 (m, 5H, aryl); 7.07-7.28 (m, 4H, aryl); 7.79-
7.84 (m, 3H, aryl); 8.28 (s, 1H, aryl), 9.40 (s, 1NH). HR-MS m/z calcd for C25H25FN3O [(M + H)]+:
402.1976 found 402.1981.
5.4.12
N-(4-fluorophenyl)-1-methyl-3-(((3,4,5-trimethoxybenzyl)amino)methyl)-1H-indole-5-
(26)
carboxamide
Yield 69%. 1H NMR (400 MHz, CDCl3)
δ; 3.75 (s, 3H, CH3); 3.83 (s, 6H, CH3); 3.84 (s, 3H, CH3);
3.90-3.97 (s, 4H, C 2); 6.62 (s, 2H, aryl); 7.01-7.07 (m, 3H, aryl); 7.28-7.32 (m, 2H, aryl); 7.69-
H
7.71 (m, 1H, aryl); 7.78 (d, 1H,
J = 8.0 Hz, aryl); 8.26 (s, 1H, aryl); 8.67 (s, 1NH). HR-MS m/z
calcd for C27H29FN3O4 [(M + H)]+: 478.2137 found 478.2143.
5.4.13
3-(((2-([1,1'-biphenyl]-4-yl)ethyl)amino)methyl)-1-methyl-N-(p-tolyl)-1H-indole-5-
35
carboxamide
(
)
Yield 67%. 1H NMR (400 MHz, CDCl3)
2H, = 8.0 Hz, C 2); 3.50 (s, 3H, C 3); 4.05 (s, 2H, C
aryl); 7.36 (t, 1H, aryl, = 8.0 Hz), 7.43 (t, 2H, aryl, = 8.0 Hz), 7.47 (d, 2H, aryl, J
δ
; 2.33 (s, 3H, C
H
3); 2.78 (t, 2H,
2); 7.05 (s, 1H, aryl); 7.09-7.15 (m, 5H,
= 8.0 Hz); 7.53
J = 8.0 Hz, CH2); 3.06 (t,
J
H
H
H
J
J
(d, 2H, aryl,
J = 8.0 Hz); 7.59 (d, 2H, aryl, J = 8.0 Hz); 7.77 (d, 1H, aryl, J = 8.0 Hz); 8.29 (s, 1H,
aryl); 9.10 (s, 1NH). HR-MS m/z calcd for C32H32N3O [(M + H)]+: 474.2540 found 474.2545.
5.4.14
(3-(((2-([1,1'-biphenyl]-4-yl)ethyl)amino)methyl)-1-methyl-1H-indol-5-yl)(piperidin-1-
36
yl)methanone
( )

ACCEPTED MANUSCRIPT
Yield 70%. 1H NMR (400 MHz, CDCl3)
3.18 (t, 2H, CH2 = 8.0 Hz); 3.27-3.42 (m, 4H, C
2H, = 8.0 Hz, aryl); 7.23 (d, 2H, = 8.0 Hz, aryl); 7.31 (d, 1H,
6H, aryl); 7.54 (d, 1H, = 8.0 Hz, aryl); 7.74 (s, 1H, aryl). HR-MS m/z calcd for C30H34N3O [(M +
δ
; 1.44-1.63 (m, 6H, C
H
2); 2.81 (t, 2H, CH2
3); 4.20 (s, 2H, C
= 8.0 Hz, aryl); 7.33-7.50 (m,
,
J
= 8.0 Hz);
,
J
H2); 3.62 (s, 3H, C
H
H2); 7.11 (d,
J
J
J
J
H)]+: 452.2696 found 452.2702.
5.4.15 3-(((2-([1,1'-biphenyl]-4-yl)ethyl)amino)methyl)-N-ethyl-1-methyl-1H-indole-5-carboxamide
(
37)
Yield 72%. 1H NMR (400 MHz, CDCl3)
Hz); 3.06 (t, 3H, CH2, J = 8.0 Hz); 3.48-3.52 (m, 5H, CH2 and C
δ
; 1.30 (t, 3H, CH3, J = 8.0 Hz); 2.64 (t, 3H, CH2, J = 8.0
3); 4.07 (s, 2H, C 2); 6.96 (s,
= 8.0 Hz, aryl); 8.11 (s, 1H,
H
H
1NH); 7.09-7.12 (m, 4H, aryl); 7.33-7.58 (m, 7H, aryl); 7.73 (d, 1H,
J
aryl). HR-MS m/z calcd for C27H30N3O [(M + H)]+: 412.2383 found 412.2388
5.5 General method for reductive amination reaction. Synthesis of (1-alkyl-1H-indole-3,5-
diyl)dimethanamine indole derivatives (9-13).
1-alkyl-3-aminomethyl-5-carboxyaldehyde indole intermediates (4-8, 1.0 mmol) were dissolved in
a solution of DCM/CH3COOH (5:1 v/v) at room temperature. To each solution an amount of 2.0
mmol of 4-fluoroaniline was added and the mixture was warmed to reflux for 1.5 h. Then, an
amount of 1.8 mmol of sodium triacetoxyborohydride was added portionwise and the mixture was
allowed to reflux for further 3−5 h. After cooling to room temperature, NaOH 1 N was added. The
organic phase was separated and extracted one more time with the alkaline solution. Then it was
dried over Na2SO4, filtered, and concentrated in vacuo. The crude products were purified by column
chromatography using mixtures of DCM/MeOH as eluent leading to the corresponding 3,5-
diaminomethyl indole derivatives with 59-68% of yields.