Design, synthesis, anticancer activity, molecular docking and ADME studies of novel methylsulfonyl indole-benzimidazoles in comparison with ethylsulfonyl counterparts

Article abstract of DOI:10.1039/d1nj01019k

Cancer poses a world-wide healthcare problem, demanding selective and effective therapy protocols. To address that, a vast amount of therapeutic candidates are being investigated in the field of medicinal chemistry. Accordingly, indole-benzimidazole structures have recently gained considerable interest because of their anticancer properties and estrogen receptor (ER) modulatory actions. In this study, novel methylsulfonyl indole-benzimidazole derivatives have been synthesized upon substitution of respectively the first (R1) and fifth (R2) positions of benzimidazole and indole groups. Structure and activity relationships were then studied via1H NMR, 13C NMR, mass spectral and in silico docking analyses, as well as cell viability measurements. We found that the compounds exhibited substantial affinity levels towards ER alpha (ERα). In addition, the correlation analysis of cytotoxicity profiles between ethyl- and methyl-sulfonyl indole-benzimidazoles revealed a collection of effective and consistent R1 and R2 substitutions. However, for some candidate derivatives, distinctive cytotoxicity levels and varying viability versus ERα affinity correlations were observable across the studies, suggesting that the sulfonyl side chain modifications themselves can also influence the ERα binding levels. These results demonstrated that our novel methylsulfonyl indole-benzimidazole derivatives, similar to their ethylsulfonyl counterparts, exhibit anticancer effects with potential estrogen receptor modulatory actions. This journal is

Full text of DOI:10.1039/d1nj01019k

NJC
View Article Online
View Journal
PAPER
Design, synthesis, anticancer activity,
molecular docking and ADME studies of novel
methylsulfonyl indole-benzimidazoles in
comparison with ethylsulfonyl counterparts†
a
Cite this: DOI: 10.1039/d1nj01019k
Fikriye Zengin Karadayi,‡a Murat Yaman,‡b Mehmet Murat Kisla,
‡
Ozlen Konu*bcd and Zeynep Ates-Alagoz
*
a
Cancer poses a world-wide healthcare problem, demanding selective and effective therapy protocols.
To address that, a vast amount of therapeutic candidates are being investigated in the field of medicinal
chemistry. Accordingly, indole-benzimidazole structures have recently gained considerable interest
because of their anticancer properties and estrogen receptor (ER) modulatory actions. In this study,
novel methylsulfonyl indole-benzimidazole derivatives have been synthesized upon substitution of
respectively the first (R1) and fifth (R2) positions of benzimidazole and indole groups. Structure and
activity relationships were then studied via 1H NMR, 13C NMR, mass spectral and in silico docking
analyses, as well as cell viability measurements. We found that the compounds exhibited substantial
affinity levels towards ER alpha (ERa). In addition, the correlation analysis of cytotoxicity profiles between
ethyl- and methyl-sulfonyl indole-benzimidazoles revealed a collection of effective and consistent R1
and R2 substitutions. However, for some candidate derivatives, distinctive cytotoxicity levels and varying
viability versus ERa affinity correlations were observable across the studies, suggesting that the sulfonyl side
chain modifications themselves can also influence the ERa binding levels. These results demonstrated that
our novel methylsulfonyl indole-benzimidazole derivatives, similar to their ethylsulfonyl counterparts, exhibit
anticancer effects with potential estrogen receptor modulatory actions.
Received 1st March 2021,
Accepted 24th April 2021
DOI: 10.1039/d1nj01019k
rsc.li/njc
investigations bringing up novel therapeutic options for
BC.1,2 In this regard, development of SERMs gained consider-
1. Introduction
ER positivity in breast cancer (BC) patients is higher than able attention by utilizing ERa model cell lines in preclinic
70%.1,2 ERa’s role on BC therapy arises from its potential for studies.7,8 For example, an ERa-positive luminal cell line, MCF-
inducing growth of the tumour cells in the presence of estrogen 7, has become a prominent in vitro model in conducting
which can be targeted against, using antiestrogens or selective preliminary evaluations for novel SERMs against BC.9
ER modulators (SERMs).3 ERa expression influences disease
SERMs are chemical substances that disrupt the action of
progression by regulating multiple signaling pathways while estrogen by interacting with ER in cells.10 More than 70 SERMs
maintaining the luminal phenotype, and results in increased have been discovered to this day, and they can be classified as
susceptibility of breast cancer to hormone therapy.4–6 Hence, triphenylethylenes, benzothiophenes, tetrahydronaphtalenes,
modulation of ER signaling has been at the core of many benzopyrans and indoles. In addition to the widely used first
generation SERM known as tamoxifen,11,12 there are other
second and third generation SERMs developed to reduce side
a Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Ankara University,
effects associated with tamoxifen.11,13 For instance, the affinity
Ankara, Turkey. E-mail: zeynep.ates@pharmacy.ankara.edu.tr
b Interdisciplinary Program in Neuroscience, Bilkent University, 06800, Ankara,
of third generation bazedoxifene is more selective towards the
ERa rather than ERb, hence its inhibitory effect progresses
through ERa downregulation and cell cycle arrest.11 In addi-
tion, bazedoxifene with palbociclib combination is a promising
therapeutic option in clinics, and tested against metastatic BC
at stage IV.14–16 Therefore, bazedoxifene has become a valuable
therapeutic in clinic, and served as an appropriate reference
for the studies assessing ER modulation, such as in molecular
Turkey. E-mail: konu@fen.bilkent.edu.tr
c Department of Molecular Biology and Genetics, Bilkent University, 06800, Ankara,
Turkey
d UNAM-Institute of Materials Science and Nanotechnology, Bilkent University,
06800, Ankara, Turkey
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
d1nj01019k
‡ Equal contributors.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
docking studies.17,18 Additionally, indole and benzimidazole indole-benzimidazoles possessing methylsulfonyl group, on
cores are also main structural components of some SERMs. MCF-7. Obtained cytotoxicity estimates were then compared
Besides bazedoxifene, melatonin and KB9520 carry indole with the viability scores derived from those possessing ethylsul-
moieties and pose SERM effects.3,19 Methyl or naphthyl fonyl structures which were previously studied.22 The meta-
attached benzimidazole structures also demonstrate SERM- analysis enabled us to investigate the consistencies in the
like actions on breast cancer cell lines.3,19–21
anticancer effects caused by specific side group modifications
In line with the above-mentioned findings, we have pre- across methyl-/ethylsulfonyl indole-benzimidazoles. Further-
viously identified several ethylsulfonyl indole-benzimidazole more, in silico docking analyses were proceeded with evaluating
derivatives with antiestrogenic properties.22 Interestingly, anti- ERa binding affinities to see if the actions of the methylsulfonyls
proliferative effects were distinct based on ER status where can also relate with ERa. Our findings suggest that both ethyl-
ERa-positive cell lines MCF-7 and HEPG2 resulted in overlap- and methyl-sulfonyl groups are active and may share common
ping cell viability profiles across various compound concentra- modulatory roles depending on the R1 and R2 chain diversities
tions in comparison with ERa-negative cell line MDA-MB-231. as well as the status of the sulfonyl modification. This is the first
Structure–activity relationship (SAR) studies have suggested meta-analysis approach that allows for one-to-one anticancer
that compounds possessing R1: p-fluorobenzyl group have activity comparisons between closely related counterparts of
higher anticancer activity and moderate structural affinity indole-benzimidazoles. Moreover, the study provides novel leads
towards ERa. Our results demonstrated that novel ethylsulfonyl to identification of which R1 and R2 moieties can be more
indole-benzimidazole derivatives have ER modulatory activities effective on viability of luminal breast cancer cells.
similar to SERMs, and also had in common with dioxin-
mediated aryl hydrocarbon receptor (AhR) signaling and amino
acid deprivation-mediated integrated stress response
2. Results and discussion
2.1. Biological activity
pathways.22 In a previous study, sulfonylmethyl moiety has
been also investigated for its anti-BC activity identifying a
cytotoxic agent against the MCF-7 cell line.23 In addition, Anticancer activities of the derivatives were initially evaluated
anticancer activities of methylsulfonyl benzothiazole deriva- via MTT assays on MCF-7 treated at the 20 mM dose (Fig. 1); and
tives have also been shown on a cervical cancer cell line, relative cell viabilities as well as their significance were reported
HeLA.24 These studies suggest that sulfonylmethyl groups in Table 1 and Table S1 (ESI†). First, the assessments by one-
could also modify effectivity of indole-benzimidazoles, way ANOVA/Dunnett showed that treatments with 28 out of 34
Thus, the methylsulfonyl bearing moieties can serve as candi- compounds (82%) have significantly reduced the MCF-7 viabi-
date scaffolds in providing novel alternatives. Accordingly, lity (Table S1, ESI†). Besides, 19 of them resulted in lower cell
our group designed, synthesized and tested a series of growth than the overall mean viability score (66.47%), further
Fig. 1 (A) R1 and (B) R2 based cell viability comparisons for methylsulfonyl indole-benzimidazole derivatives. Log 2 transformed percent viability scores
from each triplicate were given on the y-axis. R1, R2 and R1 Â R2 interaction effects were calculated via two-way ANOVA, as within group comparisons for
either R1 or R2 moieties were performed via multiple t-tests with Holm corrections overall. R environment, ggplot2 and ggpubr packages were utilized for
calculation and representation purposes.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
Table 1 Percentages of relative cell viabilities and adjusted p-values (one-way ANOVA/Dunnett) obtained from 20 mM concentration of indole-
benzimidazole derivatives containing methylsulfonyl. DMSO was used as the control group
Compound
R1
R2
%Viability
Adj. p-val.
Compound
R1
R2
%Viability
Adj. p-val.
23
24
25
26
27
28
29
–CH3
–CH3
–CH3
–CH3
–C2H5
–C2H5
–C2H5
–H
78.16
52.42
77.29
31.00
48.54
99.23
52.49
0.0013
41
42
43
44
45
46
47
–Br
–H
59.72
89.39
43.64
64.56
99.91
87.84
84.36
o0.0001
0.3467
–OCH3
–Cl
o0.0001
0.0008
–OCH3
–Cl
o0.0001
0.0003
–Br
o0.0001
o0.0001
40.9999
o0.0001
–H
–H
40.9999
0.0148
–OCH3
–Cl
–Cl
–Br
0.0014
30
31
32
–C2H5
–C3H7
–C3H7
–Br
52.95
43.93
91.37
o0.0001
o0.0001
0.9621
48
49
50
–H
42.46
74.21
41.82
o0.0001
o0.0001
o0.0001
–H
–OCH3
–Cl
–OCH3
33
34
35
36
37
38
–C3H7
–C4H9
–C4H9
–C4H9
–C4H9
–Cl
77.87
51.32
102.94
45.34
88.93
88.99
0.0702
51
52
53
54
55
56
–Br
53.38
60.26
61.84
45.18
45.07
76.72
o0.0001
o0.0001
o0.0001
o0.0001
o0.0001
o0.0001
–H
o0.0001
0.9991
–H
–OCH3
–Cl
–OCH3
–Cl
o0.0001
0.8268
–Br
–Br
–H
0.8307
–H
39
40
–OCH3
–Cl
43.43
73.22
o0.0001
57
58
–Cl
–Br
77.28
85.90
o0.0001
0.0139
0.0038
supporting that the substitutions caused significant cytotoxic R1:-p-chlorobenzyl has demonstrated a decrease in the anti-
effects. In addition, the results from multiple t-test/Holm and cancer activity. Although R2-based multiple pairwise tests did
two-way ANOVA revealed the significant effects of different not reveal noticeable differences, the results of ANOVA showed
substitutions at the R1 or R2 side chains. Accordingly, R1:-3,4- a significant influence of the R2 groups on the cytotoxicity levels
difluorobenzyl and -p-chlorobenzyl additions demonstrated a (Fig. 1B). Nonetheless, we also observed a significant inter-
trend for exhibiting more sensitive profiles overall (Fig. 1A). action between the R1 and R2 groups implicating that each R2
Moreover, the number of –Cl and that of –F on the benzyl group might influence cell viability dependent on the R1 and
groups by the R1 chain seemed to have an inverse relationship vice versa. Therefore, this significant interaction term might
with respect to each other but acting oppositely for –Cl and –F. actually stem from the differences generated by differential
For instance R1:-p-fluorobenzyl yielded the least anti-cancer R1:R2 stereochemistry. We also noted that the median cytotoxi-
activity across the screened derivatives (adj. p-value: 2.5 Â 10À9
)
city scores of R2:-Br bearing compounds were relatively lower
while R1:-p-difluorobenzyl was one of the most effective than the others, suggesting improvements on cytotoxicity
R1 substitutions, approaching significance (adj. p-value: 0.12). by the R2:-Br side chains. In line with this, the compound 26
Similarly, R1:-3,4-dichlorobenzyl and R1:-3,4-chlorobenzyl (R1:-methyl & R2:-Br) has shown the highest activity over all the
groups were located at the opposite ends of the viability methylsulfonyl derivatives screened in this study (Table 1).
spectrum (Fig. 1A). Moreover, the inclusion of –F by the third Judging by the cell viability values from this preliminary screen-
position of the R1:-p-fluorobenzyl chain has led to an increase ing, we focused on this lead molecule, compound 26 for which
in cytotoxicity remarkably. In contrast, an addition of –Cl to the we calculated the IC50 value (28.73 mM) (Fig. 3).
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
Fig. 2 Significant correlations observed for cellular viability between methylsulfonyl (20 mM) (x-axes) and ethylsulfonyl (16 mM) (y-axes) of indole-
benzimidazoles based on 24 h treatment of MCF-7 cells. (A) R1 and (B) R2. Pearson’s correlations were performed on the viability scores from the
biological triplicates where the lines and shadow regions represent the linear correlations and confidence intervals, respectively. Correlation plots of non-
significant groups (R1:-3,4-dichlorobenzyl, -cyclohexyl and -propyl; R2:-Br and -methoxy) are available in Fig. S1 (ESI†).
with overall comparisons between the two different sulfonyl
scaffolds (Fig. S1, ESI†). Similarly ethyl- and methylsulfonyl
derivates were negatively or positively correlated respectively for
–Cl and –H while other R2 groups did not show any significant
association (Fig. 2B and Fig. S1, ESI†). Therefore, based on
which R1 and/or R2 group present, ethyl or methyl status of the
sulfonyl groups were found to influence the viability levels
differentially.
2.2. Molecular docking
According to the MTT assay results, compound 26 was elected
as the lead candidate, and compared to the standard CPT.
According to the complex of CPT with ERa ligand binding
Fig. 3 Detailed MTT screening of eight concentration points for the lead
domain and reference study,25 this area includes various hydro-
molecule 26 on MCF-7 cell line for 24 h, across doses of 0.25 mM, 2 mM, 4 mM,
phobic residues (Fig. 4A). Therefore, incorporating hydropho-
8 mM, 10 mM, 12 mM, 16 mM and 40 mM. CPT has served as the positive control.
bic groups such as indole and benzimidazole becomes essential
for modulating the signalling mechanisms of this protein.
Additionally, in our study we compared for the first time the According to our previous docking analyses with ERa, a hydro-
viability results obtained from exposures to ethylsulfonyl deri- gen bond with some polar residues in this cavity could become
vatives (16 mM) with those of methysulfonyl derivatives (20 mM). important.22 In hindsight, interaction diagram of CPT offers
This allowed us to test whether similar modifications exhibited certain hydrophobic interactions with Leu387, Leu346, Ala350,
similar effects when they were associated with ethyl- or methyl- Met343 and Leu343 residues. Amongst them, Met343 was
sulphonyl carrying compounds. The cell viability scores previously addressed in publications. Through the examination
between different sulfonyl moieties revealed both similar as of 26’s interaction diagram (Fig. 4B), in addition to several
well as inverse profiles depending on the R1 and/or R2 (Fig. 2 steric interactions such as the ones with Leu354, Trp383 and
and Fig. S1, ESI†). For example R1:-benzyl, -butyl, -ethyl, Ala350, a hydrogen bond interaction by sulfonyl group is
-p-fluorobenzyl in addition to R2:-H demonstrated significantly present. This binding profile increased the chance of 26 in
positive correlations across the two sulfonyl moieties while R1:- being a prominent inhibitor of ERa.
methyl and -3,4-difluorobenzyl additions and R2:–Cl modifica-
The relevance of ERa binding and anticancer activity of the
tions were inversely correlated between the groups (Fig. 2A). indole-benzimidazoles has been a crucial element in guiding
Furthermore, some of the substitutions (R1:-cyclohexyl and our meta-analysis of the two types of sulfonyl groups. Accord-
-propyl; R2:-Br and -methoxy) showed no correlations, in line ingly for the methylsulfonyl derivatives, ERa docking scores
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
also observable across the ethylsulfonyl counterparts (Fig. S2A,
ESI†). Interestingly, R1:-benzyl followed an inverse correlation
regarding the viability versus docking score comparisons, as the
remaining R1 derivatives did not show a clear association in
this sense. Thus, ERa modulatory activities of the methylsulfo-
nyl compounds can differ depending on the side chain mod-
ifications. Similarly, we have seen that R2: methoxy and -H
substitutions can closely associate with ERa docking scores and
MCF-7 viabilities while R2:-Br and –Cl additions did not
(Fig. 5B).
In parallel, we have employed a similar approach with
ethylsulfonyl derivatives (Fig. S3, ESI†). By the R1 side chain,
-3,4-dichlorobenzyl groups again presented a positive relation-
ship between docking scores and cytotoxicity levels. In addi-
tion, R1: -butyl and –methyl members and R2: -H groups
indicated overlapping patterns with their methylsulfonyl coun-
terparts when comparing their viability – ERa docking associa-
tions. Moreover, R2:-Br and –Cl substitutions yielded significant
effects on ERa binding and cell viability score comparisons,
which were other distinctive features between ethyl/methylsul-
fonyl derivatives. Therefore, the findings have further imple-
mented that ERa binding and cellular viability related events
were likely to be restricted to some compound groups only.
2.3. Pharmacokinetic evaluation
As provided in the Table S2 (ESI†), corresponding inhibitor CPT
has passed the pharmacokinetic filters such as Lipinski,
Muegge and Leadlikeness, whereas BZ failed to do so. This
outcome suggests that the former compound may be formu-
lated into a drug dosage form and one shall encounter less
bioavailability problems comparing to that of the latter. Com-
pounds such as 46, 47, 51, and 54–58 have given harshly
unfavorable results since they have got stuck on these three
filters. Consequently, when these molecules get formulated,
certain pharmacokinetic problems such as low bioavailability,
biotoxicity and higher incidence of side effects may occur.
Regarding binding affinity values, all the compounds have
had similar values comparing to that of CPT. All of the
compounds including CPT have had the same amount of
H-bond donor count, whereas H-bond acceptor values were
drastically different. Interestingly, despite having less amount
of donor count than CPT, compound 26 offered an additional
H-bond. This finding may have possibly increased the activity
of this potent compound.
Fig. 4 The interaction patterns of reference CPT (A) and inhibitor 26 (B).
In this figure, light pink corresponds to Pi-alkyl and alkyl–alkyl interactions
whereas dark pink stands for Pi-Pi stacked interaction. Green is for the
conventional H-bond interaction and light blue is for van der Waals
interaction. Finally, the orange is for Pi-sulfur interaction. In panel (C),
the aligned binding pose of these ligands in the cavity and interacting
residues were given. Purple areas are for H-bond donor surfaces whilst
green areas are for acceptor surfaces.
BOILED-Egg representation is a diagram that demonstrates
the bioavailability of the input that is used in SwissADME
were further utilized in evaluating possible correlations they program. Using this diagram, one could potentially get an
might exhibit with the percent viability scores (Fig. 5 and Fig. insight about central nervous system penetration and P-
S2, ESI†). We have seen that docking scores of R1:-3,4- glycoprotein binding. In summary, all of our compounds tend
difluorobenzyl and -3,4-dichlorobenzyl groups were positively to get absorbed through the gastrointestinal tract, meaning
correlated with the viability, as R1:-ethyl moieties were also on a they get blood concentration buff. Unfortunately, some of them
similar trend (Fig. 5A). Considering prominent effects by the got stuck in BBB reabsorption hurdle. In this passage, the main
R1:-3,4-difluorobenzyl and -ethyl groups, the findings have concern will be the comparison of these compounds with CPT
implied a close relationship between ERa and MCF-7 toxicity. and BZ. Latter is a known ERa inhibitor, thus it should also be
Findings on R1:-3,4-dichlorobenzyl members supported the discussed in ADME evaluation. Herein we witness that CPT
functional interaction for ERa and toxicity events, which was and BZ would get passively reabsorbed but would not pass
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
Fig. 5 Correlation maps between in silico docking scores (kcal molÀ1) (y-axes) and in vitro log 2 transformed viability percentiles (x-axes) obtained from
methylsulfonyl derivative exposures on MCF-7 cell line in triplicates: (A) R1 and (B) R2. Each R1 and R2 groups are color coded along with the confidence
intervals, as Pearson’s correlation and significancy scores are depicted inside the plots with the respective color codes for each group. Only the plots
approaching significance are represented. Remaining plots are accessible in Fig. S2 (ESI†).
through BBB. This characteristic is ideal in terms of bioavail- context, both CPT and BZ become ideal drugs. However,
ability and safety profile since reabsorption increases blood compounds 23, 25, 26, 27, 29, 31, and 42 failed to ensure this
concentration of the drugs, and permeability through BBB characteristic since they can passively permeate, causing a risk
causes certain CNS side effects which must be avoided. In this for CNS side effects (Fig. 6).
Fig. 6 BOILED-Egg representation of indole-benzimidazole derivatives and standards CPT (Camptothecin) and BZ (Bazedoxifene). In this diagram, X-
axis is for Topological Polar Surface Area (TPSA) which indicates the polarity of compounds while the Y-axis is for the computational log P values. The
white area is for molecules to be absorbed via the gastrointestinal tract whereas Egg yolk articulates compounds that are passively permeable through
the blood–brain barrier.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
3. Experimental part
3.1. Synthetic procedures
(v.1.418)), we obtained OD570 measurements and converted
them to cell viability, by dividing blank subtracted average
OD570 values for each concentration with those of DMSO
control group OD570 averages. After multiplication with 100,
the percentage viability estimations were analyzed via statistical
approaches.
Synthesis of new compounds 23–58 were outlined in Scheme 1.
Synthesis of 1-chloro-4-(methylsulfonyl)-2-nitrobenzene (2) was
carried out with the nitration of 1-chloro-4-(methylsulfonyl)-
benzene (1) on second position using potassium nitrate
and sulfuric acid.22,26 Aromatic nucleophilic substitution of
the chlorine atom of 2 with appropriated amine to provide
the corresponding 1-substituted-amino-4-(methylsulfonyl)-2-
nitrobenzene 3–12 in good yields.27 Reduction of the nitro
group yielded 13–22 (Table 2).28
3.1.1. General procedure for synthesis of 23–58. A mixture
of the appropriate o-phenylendiamine (1 mmol), indole-3-
carboxaldehyde (1 mmol) and Na2S2O5 (40%) (2 mL) in EtOH
(4 mL) was refluxed until consumption of the starting material
(determined by TLC, 4–12 h). The reaction mixture was poured
into water, and the precipitate was filtered and washed with
water. The residue was purified by column chromatography
and crystallized with MeOH to give the final product (Scheme 1
and Table S3, ESI†).27
3.3. Material and methods
Melting points were determined and uncorrected using Buchi
SMP-20 (BuchiLabortechnik, Flawil, Switzerland) and Electro-
thermal 9100 capillary melting point apparatus (Electrother-
mal, Essex, U.K.). The 1H NMR spectra in DMSO-d6 using
Varian Mercury-400 FT-NMR spectrometer (Varian Inc., Palo
Alto, CA, USA), and the Mass spectra based on ESI(+) method
using Waters ZQ micromass LC-MS spectrometer (Waters Cor-
poration, Milford, MA, USA) were measured. For elemental
analysis and column chromatography (cc), respectively, LECO
932 CHNS (Leco-932, St. Joseph, MI, USA) instrument and Silica
gel 60 (40–63 mm particle size) were used. 1-Chloro-4-(methyl-
sulfonyl)benzene and indole-3-carboxaldehyde were obtained
commercially. The 5-substituted-indole-3-carboxaldehyde deri-
vatives were obtained by direct formylation of 5-substituted-
indole with dimethylformamide, using phosphorous oxychloride
as a catalyst.29
3.2. Biological activity assays
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
(MTT) (Molecular Probes) assay was utilized to obtain cell
viability as percentages of the control DMSO (0.1%) samples.
MCF-7 cells were seeded (96-well plate & 10 000 cells per well)
and grown for 24 hours in replicates per group. Compounds
listed in Table 1 (20 mM) were applied on cells for another 24 h.
Campthotecin (CPT) was used as the positive control (at the
concentrations of 0.25 mM and 2 mM). In addition to the large
scale screening a candidate compound (26) was screened on a
finer scale (0.25 mM, 2 mM, 4 mM, 8 mM, 10 mM, 12 mM, 16 mM
and 40 mM). The cell growth media contained DMEM-low-
glucose (phenol-free, GIBCO) supplemented with 10% FBS
(Biowest), 1% penicillin–streptomycin (LONZA), 1% non-
essential amino acid (LONZA), 2% L-glutamine (LONZA) and
1% Na-pyruvate (LONZA). Using BIO-TEK/mQuant Universal
Microplate Spectrophotometer (BIO-TEK/KC junior tool
4. Computational methods
4.1. Docking method and ADME property calculation
ERa (PDB ID:3ert, resolution: 2.5 Å) file was obtained from the
RCSB Protein database website.30 AutoDockTools 1.5.6. was
used for deleting water molecules and defining the grid
box.31 After this process, polar hydrogens and Gasteiger charges
were added and the grid was also prepared using the same
software. Assigned grid’s center was found at X = 31.583, Y =
À1.8, Z = 24.369 and dimensions were X = 40, Y = 40, Z = 48.
Spacing was defined as 0.375 Å. The 2D structures of the
compounds were drawn on ChemDraw Ultra 12.0, minimized
with MMFF94 and UFF forcefields (number of steps: 5000 with
steepest descent algorithm and convergence value of 10 Â 10À7
)
to achieve the lowest energy conformer of the compounds. This
conformer would also facilitate the occurrence of a ligand-
protein complex with the lowest energy. After the minimization
process, these files were converted to pdb files using Avogadro
software.32 Subsequently, Gasteiger charges and torsion were
added to ligand files with AutoDockTools. Prepared ligands
were docked with Autodock Vina,33 interaction diagrams were
created and interpreted using Discovery Studio Visualizer
Ligand interaction module.34
The evaluation of physicochemical properties and the pre-
diction of ADME parameters were determined with SwissADME
online tools.35 Besides, using BOILED-Egg representation,36
brain penetration and gastrointestinal absorption of synthesized
compounds were assessed. With the aid of the above-mentioned
calculations, we aimed to gather medicinal chemistry information
Scheme 1 Synthesis procedure of novel indole-benzimidazole deriva-
tives 23–58.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
Table 2 Physicochemical data for compounds 3–22
Comp. R1
1H NMR
M.p. (1C) Yield % Comp. Mass
3
4
–CH3
3.00 (d, 3H, NH–CH3), 3.17 (s, 3H, CH3), 7.15 (d, J = 9.2 Hz, 1H, CH-6), 7.90 (dd, J = 9.2 157
Hz, J = 2.4 Hz, 1H, CH-5), 8.47 (d, J = 2.0 Hz, 1H, CH-3), 8.64 (brd d, J = 4.8 Hz, 1H, NH)
86
83
13
14
200
215
À ÀÀ
–C2H5
–C3H7
–C4H9
1.23 (t, 3H, CH2–CH3), 3.21 (s, 3H, CH3), 3.45–3.52 (m, 2H, CH2), 7.25 (d, J = 9.2 Hz, 164
À ÀÀ
1H, CH-6), 7.92 (dd, J = 9.2 Hz, J = 2.0 Hz, 1H, CH-5), 8.50 (d, J = 2.4 Hz, 1H, CH-3), 8.60
(brd t, 1H, NH)
5
6
0,92 (t, 3H, CH2–CH3), 1.59–1.64 (m,2H, CH2–CH3), 3.17 (s. 3H, CH3), 3.36–3.41 (m, 116
81
87
15
16
228
243
À ÀÀ
2H, NH–CH2), 7.2À4À(dÀ, J = 9.2 Hz, 1H, CH-6), 7.88 (dd, J = 9.0 Hz, J = 2,4 Hz, 1H, CH-5),
8.47 (d, JÀ=À2À.4 Hz, 1H, CH-3), 8.6 (brd t, 1H, NH)
0.92 (t, 3H, CH2–CH3), 1.35- 1.41 (m, 2H, CH2–CH3), 1.57–1.64 (m, 2H, CH2–CH2), 3.20 122
À ÀÀ
À ÀÀ
(s, 3H, CH3), 3.44 (q, 2H, NH–CH2), 7.26 (d, J = 9.2 Hz, 1H, CH-6), 7.91À(dÀdÀ, J = 9.4 Hz,
J = 2.4 Hz, 1H,
À ÀÀ
CH-5), 8.50 (d, J = 2.4 Hz, 1H, CH-3), 8.61 (brd t, 1H, NH)
1.24 (m, 1H, cyclohexyl CH), 1.41 (m, 4H, cyclohexyl), 1.56–70 (m, 3H, cyclohexyl), 1.93 124
(m, 2H, cyclohexyl CH2), 3.18 (s, 3H, CH3), 3.73 (m, 1H, cyclohexyl CH), 7.31 (d, J = 9.2
Hz, 1H, CH-6), 7.88 (dd, J = 8.6 Hz, J = 2.4 Hz, 1H, CH-5), 8.30 (brd d, 1H, NH), 8.45
(d, J = 2.4 Hz, 1H, CH-3)
3.17 (s, 3H, CH3), 4.72 (d, 2H, CH2), 7.09 (d, J = 9.6 Hz, 1H, CH-6), 7.24–7.39 (m, 5H, 132
CH-2’,3’,4’,5’,6’), 7.84 (dd, J = 9.2 Hz, J = 2.0 Hz, 1H, CH-5), 8.52 (d, J = 2.0 Hz, 1H,
CH-3), 9.18 (t, 1H, NH)
7
-Cyclohexyl
85
17
269
8
-Benzyl
85
82
80
62
77
18
19
20
21
22
277
295
313
311
345
9
-p-Fluorobenzyl
3.18 (s, 3H, CH3), 4.71 (d, 2H, CH2), 7.09 (d, J = 9.6 Hz, 1H, CH-6), 7.15–7.20 (m, 2H), 144
7.44 (m, 2H), 7.85 (dd, J = 9.6 Hz, J = 2.4 Hz, 1H, CH-5), 8.52 (d, J = 2.0 Hz, 1H, CH-3),
9.18 (t,1H, NH)
10
11
12
-3,4-Difluorobenzyl 3.16 (s, 3H, CH3), 4.68 (d, 2H, CH2), 7.05 (d, J = 8.8 Hz, 1H, CH-6), 7.21–7.24 (m, 1H), 128
7.35–7.49 (m, 2H), 7.83 (dd, J = 9.2 Hz, J = 2.4 Hz, 1H, CH-5), 8.49 (d, J = 2.4 Hz, 1H,
CH-3), 9.16 (t,1H, NH)
-p-Chlorobenzyl
3.18 (s, 3H, CH3), 4.71 (d, 2H, CH2), 7.05 (d, J = 9.2 Hz, 1H, CH-6), 7.41 (s, 4H,
CH-2’,3’,5’,6’), 7.84 (dd, J = 8.8 Hz, J = 2.0 Hz, 1H, CH-5), 8.52 (d, J = 1.6 Hz, 1H, CH-3),
9.19 (t,1H, NH)
181
-3,4-Dichlorobenzyl 3.19 (s, 3H, CH3), 4.73 (d, 2H, CH2), 7.07 (d, J = 9.2 Hz, 1H, CH-6), 7.39 (dd, J = 8.0, Hz, 129
J = 2.0 Hz, 1H), 7.60 (d, J = 8.0 Hz, 1H), 7.70 (d, J = 2.0 Hz, 1H), 7.86 (dd, J = 9.0 Hz,
J = 2.4 Hz, 1H, CH-5), 8.53 (d, J = 2.4 Hz, 1H, CH-3), 9.21 (t,1H, NH).
about our compound set and to suggest further solutions for the exposed to 37 ethylsulfonyl indole-benzimidazole derivatives
improvement of these properties.
for 24 h (16 mM) as well as (ii) the ERa docking scores of the
respective compounds from the Glide docking software were
obtained.39 Retrieved non-versatile docking scores from ethyl-
sulfonyl data were capped to zero for making comparisons and
representations possible.
4.2. Statistical analyses
Percent viability value from each replicate was log 2 trans-
formed. Initially, viability scores from 20 mM treatments in
triplicates were compared to DMSO control group via one-way
ANOVA followed by Dunnett tests (GraphPad Prism (v. 6.01)).
Then, effects of the treatments with respect to their respective
5. Conclusions
R1 and R2 groups were evaluated by using ggplot2 and ggpubr Most of the novel derivatives yielded significant cytotoxic
packages in R statistical environment (3.6.1), while within effects, which were likely to be above 20 mM IC50 values.
group comparisons (R1 or R2) were employed via multiple However, the findings also revealed some candidate com-
t-test/Holm statistics.37,38 The analyses were further supported pounds and a subset of R1 and R2 modifications that can serve
with two-way ANOVA approaches in R. In addition, IC50 value of for lead molecule development. For instance, the compounds
the compound 26 was estimated via GraphPad Prism tool 26, 27, 48 and 50, and R1:-3,4-difluorobenzyl, -p-chlorobenzyl
with ‘‘log(inhibitor) vs. response (three parameters)’’ option. and ethyl groups as well as R2:-Br additions can become useful
Moreover, ggpubr package was utilized for drawing the scatter for further improvements.
plots for Pearson’s correlations across viability values for each
Results of the docking studies suggested that methylsulfonyl
side-chain and ERa docking scores.38 Meta-analysis was pur- indole-benzimidazole derivatives have rendered similar affinity
sued with the data all of which were retrieved from our recent values and additional interactions in docking analysis, compar-
study.22 Accordingly, (i) cell viability values of MCF-7 cells ing to that of the biological activity standard CPT. Amongst
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
these derivatives, compound 26 which ensures certain interac- methysulfonyl counterpart. Therefore, status of sulfonyl side
tions with the enzyme has shown some promise as a potential on the indole-benzimidazole scaffold has been shown to play
ERa inhibitor although with a potential risk of passing through an important part in directing molecular and cellular effects in
BBB. Pharmacokinetic parameters were also calculated to addition to the R1 and R2 positions. Yet, in line with the
evaluate whether our compounds lack certain requirements literature findings on indole-benzimidazoles, such comparative
for their pharmaceutical development, or if they need analyses on ERa binding and viability estimations can be a
any certain pharmacokinetic improvements. Major obstacles powerful approach in pinpointing candidate SERM derivatives
were observed to may involve high log P and molecular weight for further studies.43 Moreover, AutoDock and Maestro tools
related issues. In practice, adjusting the molecular weight may were utilized across the methylsulfonyl and ethylsulfonyl dock-
not be an option, yet modifying the partition coefficient (log P) ing analyses. In this regard, in silico validations on the docking
can be pursued by several strategies.40,41 As a result, both scores can be followed up by employing both platforms inter-
in vitro and in silico docking studies have provided crucial changeably and by deriving the docking correlations between
information in demonstrating potentials of novel anti- each other.44
proliferative agents and their possible interactions with ERa
in support of the previous findings with indole-benzimidazoles chain modifications that are informative for ERa binding
bearing ethylsulfonyl moieties.22
abilities of the derivatives and likely viability intervals upon
In conclusion, our findings demonstrate the prominent side
In addition to unlikely pharmacokinetic properties, general- exposure to these novel derivatives. This knowledge along with
izing SERM characteristics for the compounds has become a the literature on indole-benzimidazoles libraries can further
challenging task, due to varying interrelationships between ERa pave the way for developing stricter SERMs with improved
binding and limited cellular toxicity levels. Since some of the anti-breast cancer profiles. In this course, lead structures
side chain modifications also revealed significant effects on demand thorough in vitro and in vivo examinations for their
the cell viability percentages, the findings were in line with the modulatory activities on ERa and possible interactions with
knowledge that modulation status of ERa may influence the multiple targets.
cancer cell viability. Yet, functional studies are required to
address this issue with the lead compounds, across multiple
cell lines with varying ER genotypes.
Conflicts of interest
Initial SAR studies suggested that substituents with steric
hindrance played a pivotal role for both sulfonylethyl and
The authors declare no conflicts of interest.
sulfonylmethyl derivatives, since our previous findings sug-
gested that 3,4-difluorobenzyl group has led to increased Acknowledgements
inhibition. In a similar fashion, in this study we have con-
The work was supported by TUBITAK (grant number: 1001 –
cluded that lipophilic and bulky side chains such as 3,4-
213S037).
difluorobenzyl, p-chlorobenzyl and –Br greatly increase the
inhibition rate, thus antiestrogenic activity. Likewise, a pioneer-
ing study corporating indole and benzimidazole rings by Singla
et al. suggested that a bulky and/or lipophilic side chain is
References
crucial in interactions with ERa ligand binding domain since
this cavity primarily includes hydrophobic residues. They have
found that other lipophilic substituents such as bromine and
fluorine serve as bioisostere for –CF3, which is another lipo-
philic group that leads to a decreased pi-electron cloud in the
structure.16,42
Besides, not all the derivatives presented either positive or
inverse relationship. First of all, in silico ERa binding levels
can not completely reveal the activity state of the receptor,
demanding functional analyses to be followed with. Nonethe-
less, the action mechanisms of all the derivatives may not
be strictly ERa dependent per se. Therefore, indirect relation-
ships between ERa and cellular viability should also be
considered. In our previous study, we have observed this
paradigm with the ethylsulfonyl-based indole-benzimidazoles.
The candidates were able to modulate AhR and amino acid
depletion pathways in addition to estrogen signaling where
cell-cycle and stress pathways were strongly affected. Interest-
ingly, R1:-p-fluorobenzyl group was the most effective com-
pound family there, unlike what we observed with the
1 M. Xue, K. Zhang, K. Mu, J. Xu, H. Yang, Y. Liu, B. Wang,
Z. Wang, Z. Li, Q. Kong, X. Li, H. Wang, J. Zhu and
T. Zhuang, Oncogenesis, 2019, 8, 30.
2 R. Siersbæk, S. Kumar and J. S. Carroll, Genes Dev., 2018,
32(17–18), 1141–1154.
3 I. Paterni, C. Granchi, J. A. Katzenellenbogen and
F. Minutolo, Steroids, 2014, 90, 13–29.
4 W. Tian, F. Teng, J. Gao, C. Gao, G. Liu, Y. Zhang, S. Yu,
W. Zhang, Y. Wang and F. Xue, Cancer Biol. Med., 2019,
16(1), 55–70.
5 A. Arun, M. I. Ansari, P. Popli, S. Jaiswal, A. Dwivedi, K. Hajela
and R. Konwar, Cell Proliferation, 2018, 51(6), e12501.
6 X. W. Dou, Y. K. Liang, H. Y. Lin, X. L. Wei, Y. Q. Zhang,
J. W. Bai, C. F. Chen, M. Chen, C. W. Du, Y. C. Li, J. Tian,
K. Man and G. J. Zhang, Theranostics, 2017, 7(16),
4041–4056.
7 G. Sflomos, V. Dormoy, T. Metsalu, R. Jeitziner, L. Battista,
V. Scabia, W. Raffoul, J. F. Delaloye, A. Treboux, M. Fiche,
J. Vilo, A. Ayyanan and C. Brisken, Cancer Cell, 2016, 29(3),
407–422.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
8 A. Bouzekri, A. Esch and O. Ornatsky, FEBS Open Bio, 2019, 25 A. K. Shiau, D. Barstad, P. M. Loria, L. Cheng, P. J. Kushner,
9(9), 1652–1669. D. A. Agard and G. L. Greene, Cell, 1998, 95(7), 927–937.
9 R. Callis, A. Rabow, M. Tonge, R. Bradbury, M. Challinor, 26 Z. Ates-Alagoz, N. Coleman, M. Martin, A. Wan and
K. Roberts, K. Jones and G. Walker, J. Biomol. Screening,
2015, 20(6), 748–759.
10 J. Dandriyal, R. Singla, M. Kumar and V. Jaitak, Eur. J. Med.
Chem., 2016, 119, 141e168.
11 J. S. Lewis-Wambi, H. Kim, R. Curpan, R. Grigg, M. A. Sarker
and V. C. Jordan, Mol. Pharmacol., 2011, 80, 610e620.
12 S. J. Han, K. Begum, C. E. Foulds, R. A. Hamilton, S. Bailey,
A. Adejare, Chem. Biol. Drug Des., 2012, 80(6), 853–861.
27 H. Goker, M. Alp, Z. Ates--Alagoz and S. Yildiz, J. Heterocycl.
Chem., 2009, 46(5), 936–948.
28 H. Goker, C. Kus-, D. W. Boykin, S. Yildiz and N. Altanlar,
Bioorg. Med. Chem., 2002, 10(8), 2589–2596.
29 Z. Ates-Alagoz, Z. Buyukbingol and E. Buyukbingol, Pharma-
zie, 2005, 60, 643–647.
A. Malovannaya, D. Chan, J. Qin and B. W. O’Malley, Mol. 30 H. M. Berman, J. Westbrook, Z. Feng, G. Gilliland,
Pharmacol., 2016, 89, 14e26.
T. N. Bhat, H. Weissig, I. N. Shindyalov and P. E. Bourne,
13 F. Payton-Stewart, S. L. Tilghman, L. G. Williams and
Nucleic Acids Res., 2000, 28, 235–242.
L. L. Winfield, Biochem. Biophys. Res. Commun., 2014, 31 G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner,
450, 1358e1362.
14 R. Jeselsohn, in A Study of Palbociclib in Combination with
R. K. Belew, D. S. Goodsell and A. J. Olson, J. Comput. Chem.,
2009, 16, 2785–2791.
Bazedoxifene in Hormone Receptor Positive Breast Cancer, 32 M. D. Hanwell, D. E. Curtis, D. C. Lonie, T. Vandermeersch,
ClinicalTrials.gov, Dana-Farber Cancer Institute, 2017 as E. Zurek and G. R. Hutchison, J. Cheminf., 2012, 4, 17.
downloaded on January 22, 2017 from, https://clinical 33 O. Trott and A. J. Olson, J. Comput. Chem., 2010, 31,
trials.gov/ct2/show/record/NCT02448771. 455–461.
15 R. Lindsay, J. C. Gallagher, R. Kagan, J. H. Pickar and 34 Dassault Systemes BIOVIA, Discovery Studio Visualizer
`
`
v17.2.0.16349, Dassault Systemes, San Diego, 2017.
G. Constantine, Fertil. Steril., 2009, 92, 1045e1052.
16 R. Singla, K. B. Gupta, S. Upadhyay, M. Dhiman and 35 A. Daina, O. Michielin and V. Zoete, Sci. Rep., 2017, 7, 42717.
V. Jaitak, Eur. J. Med. Chem., 2018, 146, 206–219. 36 A. Daina and V. Zoete, ChemMedChem, 2016, 11(11), 1117–1121.
17 D. Bafna, F. Ban, P. S. Rennie, K. Singh and A. Cherkasov, 37 H. Wickham, ggplot2: Elegant Graphics for Data Analysis.
Int. J. Mol. Sci., 2020, 21, 4193.
18 D. Sharma, S. Kumar and B. Narasimhan, Chem. Cent. J.,
2018, 12, 107.
19 Y. M. Yoo and E. B. Jeung, J. Pineal Res., 2009, 47(4),
301–307.
Springer-Verlag, New York, 2016, ISBN 978-3-319-24277-4,
https://ggplot2.tidyverse.org.
38 Reference for ggpubr: https://cran.r-project.org/web/
packages/ggpubr/index.html.
¨
¨
39 L. Schrodinger, New York, NY, 2019, Schrodinger Release
20 S. Martinkovich, D. Shah, S. L. Planey and J. A. Arnott, Clin.
Interventions Aging, 2014, 9, 1437.
21 F. Payton-Stewart, S. L. Tilghman, L. G. Williams and L. L.
2019-3: Glide.
40 Y. Kawabata, K. Wada, M. Nakatani, S. Yamada and
S. Onoue, Int. J. Pharm., 2011, 420(1), 1–10.
Winfield, Biochem. Biophys. Res. Commun., 2014, 450(4), 41 A. Singh, Z. A. Worku and G. V. Mooter, Expert Opin. Drug
1358–1362. Delivery, 2011, 8(10), 1361–1378.
22 F. Zengin-Karadayi, M. Yaman, M. M. Kisla, A. G. Keskus, 42 W. O. Foye, Foye’s Principles of Medicinal Chemistry, Lippin-
O. Konu and Z. Ates-Alagoz, Bioorg. Chem., 2020, 100, 103929. cott Williams & Wilkins, 2008.
23 R. Ghodsi, E. Azizi and A. Zarghi, Iran. J. Pharm. Res., 2016, 43 M. A. Iannone, C. A. Simmons, S. H. Kadwell, D. L. Svoboda,
15(1), 169–177.
D. E. Vanderwall, S.-J. Deng, T. G. Consler, J. Shearin, J. G. Gray
24 N. P. Lad, Y. Manohar, M. Mascarenhas, Y. B. Pandit,
and K. H. Pearce, Mol. Endocrinol., 2004, 18(5), 1064–1081.
M. R. Kulkarni, R. Sharma, K. Salkar, A. Suthar and 44 R. Gaddipati, G. K. Raikundalia and M. L. Mathai, Int.
S. S. Pandit, Bioorg. Med. Chem. Lett., 2017, 27(5), 1319–1324. J. Biosci., Biochem. Bioinf., 2014, 4(2), 100–105.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021