Tri N-Heterocyclic Carbene Trinuclear Silver(I) complexes: Synthesis and In Vitro cytotoxicity studies

Article abstract of DOI:10.1016/j.molstruc.2020.128890

Synthesis of ethylene linked tris benzimidazolium bromide salts that serve as precursors for the tri-NHC ligands in the respective tri-NHC trinuclear silver(I) complexes is described (NHC = N-heterocyclic carbene). Different wingtip substituents were selected (benzyl, n-butyl, cyclopentyl, 2-methylene benzonitrile, n-decyl) to furnish five new tris benzimidazolium salts (1-5) and the respective tri-NHC trinuclear silver(I) complexes (6-10). The synthesized and characterized compounds (1-10) were evaluated for their cytotoxic potential on human colon, human breast and human epitheloid cervix cancer cell lines by determining their IC₅₀ values and evaluating their antiproliferative activity against the selected cell lines. The tested compounds were found to be good anticancer agents against breast cancer and cervical cancer cell lines and selectively good anticancer agents for the human colon cancer cell line.

Full text of DOI:10.1016/j.molstruc.2020.128890

Journal of Molecular Structure 1222 (2020) 128890
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstr
Tri N-Heterocyclic Carbene Trinuclear Silver(I) complexes: Synthesis
and In Vitro cytotoxicity studies
Tabinda Fatimaa,d, Rosenani A. Haque , Ashfaq Ahmad , Loiy Elsir Ahmed Hassan ,
a
b
c
c
c
a,∗
Mohamed B Khadeer Ahamed , AMS Abdul Majid , Mohd.R. Razali
a
School of Chemical Sciences, Universiti Sains Malaysia, 11800 USM, Penang, Malaysia.
b
Department of Pharmacology and Toxicology, Virginia Commonwealth University, Richmond, 23298-0613, USA.
c
EMAN Research and Testing Laboratory, School of Pharmacy, Universiti Sains Malaysia, 11800 USM, Penang, Malaysia.
Department of Chemistry, COMSATS University Islamabad, Islamabad Campus, 45550, Pakistan.
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Synthesis of ethylene linked tris benzimidazolium bromide salts that serve as precursors for the tri-NHC
ligands in the respective tri-NHC trinuclear silver(I) complexes is described (NHC = N-heterocyclic car-
bene). Different wingtip substituents were selected (benzyl, n-butyl, cyclopentyl, 2-methylene benzoni-
trile, n-decyl) to furnish five new tris benzimidazolium salts (1-5) and the respective tri-NHC trinuclear
silver(I) complexes (6-10). The synthesized and characterized compounds (1-10) were evaluated for their
cytotoxic potential on human colon, human breast and human epitheloid cervix cancer cell lines by de-
termining their IC50 values and evaluating their antiproliferative activity against the selected cell lines.
The tested compounds were found to be good anticancer agents against breast cancer and cervical cancer
cell lines and selectively good anticancer agents for the human colon cancer cell line.
Received 18 February 2020
Revised 10 July 2020
Accepted 11 July 2020
Available online 12 July 20202
Keywords:
NHC ligands
Tris benzimidazolium salts
Antiproliferative activity
Trinuclear silver(I)
©
2020 Elsevier B.V. All rights reserved.
1
. Introduction
[10]. After introducing a new strategy towards the synthesis of tris
azolium salts (Fig. 1) [10] we put forth our efforts to synthesize tris
benzimidazolium salts having different wingtip substituents (ben-
zyl, n-butyl, cyclopentyl, 2-methylene benzonitrile, n-decyl). The
resulting salts serve as precursors for the tri-NHC ligands which
were generated by their complexation with silver(I) ions.
The N-heterocyclic carbenes are well recognized class of lig-
ands in organometallic chemistry due to their ability to form sta-
ble complexes with low and high oxidation state metals [1–3]. This
class of ligands offer the advantages of structural diversity and
various coordination modes over other known ligands. In addition
to the widespread research on the synthesis and applications of
mono and bis N-heterocyclic carbenes, the introduction of tris and
tetrakis NHC complexes has further opened up a new avenue and
so far there is still a very few reports addressing these poly NHC
systems.
The tri-carbene metal complexes reported so far have been
studied for catalysis [7,8,11,12], luminescence [9] and transmetalla-
tion [13] properties. However, the biological studies on such com-
plexes still remain elusive in contrast to the anticancer and an-
tibacterial properties studied for the mono and di-NHC metal com-
plexes [14-18]. The Ag(I)-NHC complexes, compared to the earlier
known ionic silver drugs like silver sulfazide and silver nitrate,
were found to have better efficacy as they involve slow and sus-
tained release of silver ions [19]. Taking advantage of this capa-
bility of Ag(I)-NHC complexes, apart from the potent antibacterial
properties the anticancer perspective of mono- and di- Ag(I)-NHC
complexes have also been investigated in several studies. The stud-
ied anticancer mechanism of action of these complexes showed
that they are not genotoxic as no modifications in the cell cycle
have so far been observed [20,21]. Further investigations by us-
ing fluorescent Ag(I)-NHC complexes confirmed that these com-
plexes target mitochondria and offers apoptotic cell death [22].
Among the poly NHC Ag(I) complexes the Ag(I)-tetra NHC com-
plexes were investigated for antiproliferative activity. These Ag(I)
The literature known tris azolium salts have been reported to
be synthesized in a single step reaction, by reacting an azole with
a tripodal precursor [4–8]. A multistep scheme for the synthesis of
tris-NHC ligands has been reported recently [9]. A new designed
multistep scheme for the synthesis of tris benzimidazolium salts
was introduced by our group. Our designed scheme was different
from the previous and the recent reported tri NHC proligands as it
involved the reaction of unsubstituted benzimidazole with the syn-
thesized 3-(2-bromoethyl)-1-substituted benzimidazolium bromide
∗
Corresponding author.
E-mail address: mohd.rizal@usm.my (Mohd.R. Razali).
https://doi.org/10.1016/j.molstruc.2020.128890
0
022-2860/© 2020 Elsevier B.V. All rights reserved.

2
T. Fatima, R.A. Haque and A. Ahmad et al. / Journal of Molecular Structure 1222 (2020) 128890
complexes were found to be anticancer agents by inhibiting the
cell migration and colony formation of cancer cells [23].
8 × CH of cyclopentyl, m), 4.99 (2H, -CH-, m), 5.23 (8H, 2 × N-
2
CH -CH -N, s), 7.42-8.13 (12H, Ar-H, m), 10.12 (2H, NCHN, s), 10.17
2
2
As we have studied and reported the cytotoxicity of mono-,
di- and tetra-NHC silver(I) complexes against various cancer cell
lines [24–26], thus in keeping the continuation of our studies, the
tri-NHC silver(I) complexes were prepared and evaluated for their
anticancer potential against various cancer cell lines. The present
study aimed to probe whether the tri-NHC metal complexes having
three Ag(I) ions per complex, furnishes greater number of silver(I)
ions with a more sustained release in the cancer cells as compared
to mono-NHC Ag(I) complexes having one, di- and tetra-NHC Ag(I)
complexes having two silver centres per complex. Further in or-
der to get insight into the effect of substituents on the delivery
of the tri-NHC Ag(I) complexes into the cancer cells different sub-
stituents (benzyl/ n-butyl/ cyclopentyl/ 2-methylene benzonitrile/
n-decyl) have been selected.
(1H, NCHN, s).13C NMR (125 MHz, DMSO-d ) δ ppm: 22.9, 29.1,
6
31.3, 45.6, 59.6 (CH ); 112.5, 114.1, 127.4, 127.8, 130.6, 131.2 (Ar-C);
2
143.1, 144.5 (NCHN).
2.2.3. 2-Methylenebenzonitrile substituted tris benzimidazolium
bromide (4)
Salt 4 was prepared following the same procedures as described
for salt 2 except that salt II was replaced by IV (0.73 g, 1.62
mmol), benzimidazole (0.12 g, 0.83 mmol). Salt 4 was collected as
white precipitate. Yield: 0.40 g (57%). M.p.: 275-279°C. Anal. Calc.
for C41H35Br3N8: C, 55.97; H, 3.98; N, 12.74. Found: C, 56.31; H,
4.04; N, 12.47. FTIR (KBr) ʋ (cm 1): 3159 (Carom-H), 2968 (Caliph
H), 2228(C≡N), 1616, 1491, 1446 (Carom=Carom), 1568 (Carom-N),
1287 (C
-N). 1H NMR (500 MHz, DMSO-d6) δ ppm: 5.02 (8H,
−
-
aliph
2
× -CH -CH -, s), 5.80 (4H, 2 × CH -Ar, m), 7.46-8.20 (20H, H-Ar,
2
2
2
m), 10.01 (2H, NCHN, s), 10.30 (1H, NCHN, s).13C NMR (125 MHz,
2. Experimental
DMSO-d ) δ ppm: 45.6, 46.2, 49.2 (CH ); 111.7, 112.9, 113.1, 113.8,
6
2
116.5, 127.6, 127.8, 129.4, 129.9, 130.8, 131.2, 133.7, 134.8 (Ar-C);
2
.1. Materials and instruments
143.1, 144.3 (NCHN).
All the chemicals and solvents were of analytical grade and
2
.2.4. n-decyl substituted tris benzimidazolium bromide (5)
were used as received. FTIR spectra were recorded with Perkin
Elmer 2000 spectrometer. 1H and 13C NMR spectra were recorded
Salt 5 was prepared following the same procedures as described
for salt 2 except that salt II was replaced by V (1.04 g, 2.50 mmol),
benzimidazole (0.15 g, 1.21 mmol). Salt 5 was collected as white
precipitate. Yield: 0.60 g (60%). Anal. Calc. for C45H65Br ClN : C,
with Bruker 500 MHz spectrometer. Elemental analysis was per-
formed on Perkin Elmer series II, 2400 microanalyzer. Melting
points were taken using Stuart Scientific SMP-1 (UK) instrument.
3
6
61.05; H, 7.34; N, 9.49. Found: C, 61.38; H, 6.98; N, 9.63. FTIR (KBr)
ʋ (cm 1): 3166, 3109 (Carom-H); 2926, 2855 (Caliph-H), 1614, 1458
−
2
.2. Synthesis
(
Carom=Carom), 1572 (Carom-N), 1271 (Caliph-N). 1H NMR (500 MHz,
DMSO-d ) δppm: 0.85 (6H, 2 × CH , t, J = 7.5 Hz), 1.22-1.25 (28H,
6
3
The precursor salts 3-(2-bromoethyl)-1-(benzyl/ n-butyl/ cy-
1
4 × -CH -, m), 1.77-1.79 (4H, 2 × -CH -, m), 4.44 (4H, 2 × -
2
2
clopentyl/ 2-methylene benzonitrile/ n-decyl) benzimidazolium
bromides (I-V), proligand 1 (Benzyl substituted tris benzimida-
zolium bromide salt) and respective silver(I) complex 6 were pre-
pared according to the reported procedure [27].
CH -N, t, J = 7.0 Hz), 5.22 (8H, 2 × -CH -CH -, s), 7.46-7.80 (8H,
2
2
2
Ar-H, m), 8.09-8.20 (4H, Ar-H, m), 10.16 (2H, NCHN, s), 10.37 (1H,
NCHN, s). 13C NMR (125 MHz, DMSO-d ) δ ppm: 13.8(CH ); 22.1,
6
3
2
5.6, 28.4, 28.6, 28.7, 28.9, 31.2, 45.5, 46.6 (CH ); 111.2, 113.4, 113.7,
2
Synthesis of tris benzimidazolium bromides (2-5)
1
26.5, 130.7, 131.4(Ar-C); 142.8, 144.2 (NCHN).
Synthesis of tri-NHC trinuclear Ag(I) complexes (7-10)
2
.2.1. n-Butyl substituted tris benzimidazolium bromide (2)
Compound II (1.30 g, 3.60 mmol) and benzimidazole (0.21 g,
2
.2.5. n-Butyl substituted tri NHC-Ag(I) complex (7)
1
.8 mmol) were dissolved in a mixture of 1,4-dioxane (20 mL) and
Salt 2 (0.25 g, 0.35 mmol) was dissolved in methanol (50 mL)
methanol (2 mL). The reaction mixture was heated under reflux
for 32 hours. The tris benzimidazolium bromide salt appeared as
white precipitate, which was filtered and washed with 1,4-dioxane
and dichloromethane sequentially. The product obtained was dried
and silver oxide (0.23 g, 1.00 mmol) was added to it. The reac-
tion mixture was stirred at room temperature for 2 days. The re-
action was carried out in a round bottom flask wrapped with alu-
minium foil. After completion of reaction time, the product was
in fume cupboard. Yield: 0.71 g (51%). Anal. Calc. for C33H41Br N :
3
6
separated from the insolubles like AgBr and unreacted Ag O by fil-
2
C, 52.03; H, 5.38; N, 11.03. Found: C, 52.43; H, 5.42; N, 10.80. FTIR
tering through a column of celite. The obtained silver complex is
in bromide form so it was subjected to metathesis using potassium
hexafluorophosphate (0.18 g, 1.00 mmol). The mixture was stirred
for 2 hours and the resulting precipitates were filtered, washed
with distilled water and left to dry at room temperature. The sil-
ver complex 7 was collected as white precipitate. Yield: 0.27 g
(
KBr) ʋ (cm 1): 3166, 3105 (Carom-H), 2967, 2877 (Caliph-H), 1614,
−
1
489, 1469 (Carom=Carom), 1568 (Carom-N), 1282 (Caliph-N). 1H NMR
(
500 MHz, DMSO-d ) δ ppm: 0.92 (6H, 2 × CH , t, J = 7.5 Hz), 1.29
6
3
(
4H, 2 × CH -CH -, m), 1.84 (4H, 2 × CH -CH -CH , m), 4.45 (4H,
3
2
2
2
2
2
8
× -CH -N, t, J = 7.0 Hz,), 5.28 (8H, 2 × N-CH -CH -N, s), 7.47-
2
2
2
.04 (12H, Ar-H, m), 9.91 (2H, NCHN, s), 10.20 (1H, NCHN, s).13
C
(
43%). M.p.: 254-257°C. Anal. Calc. for C66H76 Ag3F18 N12P .H O: C,
3 2
NMR (125 MHz, DMSO-d ) δ ppm: 13.0 (CH ); 18.9, 30.2, 45.5, 47.0
6
3
43.69; H, 4.30; N, 9.26. Found: C, 43.30; H, 3.85; N, 9.19. FTIR (KBr)
(
(
CH ); 112.6, 113.6, 127.2, 127.8, 130.8, 131.2 (Ar-C); 143.2, 144.4
−1
2
ʋ (cm ): 3063 (Carom-H), 2958, 2873 (Caliph-H), 1614, 1481, 1450
NCHN).
Carom=Carom), 1271 (Caliph-N). 13C NMR (125 MHz, acetonitrile-d3)
(
δ ppm: 12.6 (CH ); 19.1, 31.8, 43.8, 49.5, 50.1 (CH ); 108.9, 110.6,
3
2
2
.2.2. Cyclopentyl substituted tris benzimidazolium bromide (3)
116.0, 121.2, 124.8, 133.7 (Ar-C); 187.8 (C-Ag, br d).
Salt 3 was prepared following the same procedures as described
for salt 2 except that salt II was replaced by III (1.05 g, 2.60 mmol),
benzimidazole (0.16 g, 1.30 mmol). Salt 3 was collected as white
precipitate. Yield: 0.57 g (57%). M.p.: 286-290°C. Anal. Calc. for
C35H41Br N : C, 53.50; H, 5.09; N, 10.70. Found: C, 53.77; H, 5.23;
2.2.6. Cyclopentyl substituted tri NHC-Ag(I) complex (8)
Complex 8 was prepared following the same procedures as
described for complex 7 except that salt 2 was replaced by 3
(0.41 g, 0.50 mmol) and silver oxide (0.35 g, 1.50 mmol). The
silver complex 8 was collected as white precipitate. Yield: 0.42
g (46%). M.p.: 272-277°C. Anal. Calc. for C70H76Ag3F18 N12P3: C,
46.62; H, 4.22; N, 9.32. Found: C, 46.97; H, 4.61; N, 9.56. FTIR
3
6
N, 10.88. FTIR (KBr) ʋ (cm 1): 3163, 3102 (Carom-H); 2968, 2884
−
(
Caliph-H), 1618, 1488, 1431 (Carom=Carom), 1561 (Carom-N), 1275
(
Caliph-N). 1H NMR (500 MHz, DMSO-d ) δ ppm: 1.73-2.27 (16H,
6

T. Fatima, R.A. Haque and A. Ahmad et al. / Journal of Molecular Structure 1222 (2020) 128890
3
KBr) ʋ (cm 1): 3117, 3063 (Carom-H), 2960, 2880 (Caliph-H), 1614,
−
(
2.3.3. Preparation of cell culture
-N). 13C NMR (125 MHz,
The cell culture was prepared as reported previously [26].
Briefly, the first step in the preparation is the growth of HCT
1
477, 1446 (Carom=Carom), 1237 (C
aliph
acetonitrile-d ) δ ppm: 22.9, 24.2, 31.4, 43.9, 49.4, 50.7(CH ),
3
2
116 cell lines under incubation conditions. Only those cells which
6
0.5(CH); 108.8, 110.6, 121.3, 123.9, 125.0, 133.0(Ar-C), 184.8- 185.2
(d, J(C109-Ag) = 209.0 Hz and d, J(C-107Ag) = 180.6 Hz], 186.6(C-
reached about 70-80% level of confluency were selected for plating
purposes. The medium used in the plates previously was aspirated
and the cells used were washed with phosphate buffer saline so-
lution 2-3 times. After washing with phosphate buffer saline (PBS)
solution, trypsin was added evenly on the cell surfaces. The cells
were incubated for 1 minute at 37°C. Cell segregation was facil-
itated by simply tapping the flask containing the cells and then
observing them under a microscope. A volume of 5 mL of fresh
media was added to inhibit the trypsin activity. After counting, a
[
Ag, br d).
2
.2.7. 2-Methylenebenzonitrile substituted tri NHC-Ag(I) complex (9)
Complex 9 was prepared following the same procedures as de-
scribed for complex 7 except that salt 2 was replaced by 4 (0.21 g,
.24 mmol) and silver oxide (0.17 g, 0.71 mmol). The silver com-
plex 9 was collected as white precipitate. Yield: 0.24 g (51%). M.p.:
67-171°C. Anal. Calc. for C82H64Ag3F18 N16 P3: C, 48.46; H, 3.15;
0
1
N, 11.03. Found: C, 48.04; H, 2.94; N, 10.71. FTIR (KBr) ʋ (cm 1):
−
5
final density of 2.5 × 10 cell/mL was observed (cells were counted
3
121, 3064 (Carom-H), 2941 (Caliph-H), 2224 (C≡N), 1607, 1477, 1446
using hemocytometer, 10 uL of cell suspension was mixed with 10
uL of tryptophan blue and cells in 4 chambers were counted us-
ing microscope) which was later inoculated to the wells of 96 well
plate (100μL cells/well). These seeded plates were then incubated
at standard atmospheric conditions for cell culture.
Carom=Carom), 1271 (Caliph-N). 13C NMR (125 MHz, acetonitrile-d )
(
3
δ ppm: 44.6, 47.2, 49.2 (CH ); 111.7, 112.9, 113.9, 117.6, 127.5, 128.7,
2
129.9, 130.8, 131.2, 133.7, 134.5, 139.8 (Ar-C).
2
.2.8. n-Decyl substituted tri NHC-Ag(I) complex (10)
Complex 10 was prepared following the same procedures as
2
.3.4. MTT assay
described for complex 7 except that salt 2 was replaced by 5
0.29 g, 0.32 mmol) and silver oxide (0.22 g, 0.96 mmol). The
MTT assay was carried out as reported previously by our group
(
[
26]. Briefly, 100 μL of the cells were seeded in all 96 wells of
silver complex 10 was collected as white precipitate. Yield: 0.81
the microplate and incubated overnight for cell attachment under
CO2 incubator. Later on, 100 μL of each test substance was added
in each well and the plate was labelled accordingly. Different di-
lutions of the test substance were prepared to observe dose de-
pendent response. After the addition of the test substance into the
plate containing cancer cells, the plate was incubated at 37°C with
5% CO2 environment for 72 hrs (MTT cell viability assay can be per-
formed for 48-72 hrs, time of experiment depends upon the pop-
ulation doubling time of cells, normally time is allowed for two
populations). After 72 hrs of treatment, a volume of 20 μL of MTT
reagent was added into each well and it was again incubated for 4
hrs. After this period of incubation, 20 μL of DMSO (MTT lysis so-
lution) was added in each well. The Plates were further incubated
for 5 minutes under the same environment of incubation. Finally,
absorbance was taken at 570 nm and 620 nm using a microplate
reader (Epoch, BioTek, USA). The data was analyzed for the cell vi-
ability and % inhibition of proliferation of test substances. The re-
sults were presented as percent viability compared to the negative
control (mean ± SD, n=3) [28].
18
g (50%). M.p.: 142-146°C. Anal. Calc. for C90H124Ag3F P3N12: C,
5
0.69; H, 5.82; N, 7.88. Found: C, 50.50; H, 5.37; N, 7.54. FTIR
− ): 3159, 3064 (Carom-H), 2930, 2850 (Caliph
(
KBr) ʋ (cm 1
-H), 1614,
1
496, 1450 (Carom=Carom), 1279 (Caliph-N). 13C NMR (125 MHz,
acetonitrile-d ) δ ppm: 13.1 (CH ); 22.1, 25.5, 26.0, 27.5, 28.4, 31.3,
3
3
3
8.9, 40.3, 43.2, 45.7, 46.9 (CH ); 106.6, 108.7, 112.4, 120.9, 121.7,
2
126.8 (Ar-C).
2
2
.3. In vitro cytotoxicity studies
.3.1. Materials and equipment
The equipment used in this study were, Rosewell Park Memo-
rial Institute (RPMI 1640) cell culture media was obtained
from ScienCell, USA. Microplate reader (Epoch, BioTek, USA),
Methylthiazolyldiphenyl-tetrazolium bromide (MTT) reagent, heat
inactivated foetal bovine serum (HIFBS) were purchased from
GIBCO, UK. Pencillin/streptomycin solution and 5-fluorouracil were
purchased from Sigma-Aldrich. Crystal violet, PBS and Tamoxifen
were purchased from Sigma-Aldrich. DMEM medium was pur-
chased from Gibco/Life technology, UK. ATCC cell lines were pur-
chased from Bio-Focus Saintifik Sdn Bhd. Malaysia.
2
.3.5. Statistical analysis
The percentage inhibition of proliferation data was analyzed for
Dissecting microscope (Motic, Taiwan), Eppendorf tips (Axy-
GEN, USA), Eppendorf tube 1.5 mL (Eppendorf, Germany), incu-
bator (Binder Fisher Scientific, Germany), inverted fluorescent mi-
croscope (Olympus, Japan), inverted light microscope (Matrix Op-
tic (M) Sdn. Bhd, Japan), Image J software http://rsb.info.nih.gov/ij/,
laminar flow (Class II) (ESCO- BSC, Singapore), micro pipette (Ep-
pendorf, USA), refrigerator (SAMSUNG, Japan), serological pipette
significance by two-way ANOVA and then compared by Bonfer-
roni tests using GraphPad Prism Software, Version 5.01. One-way
ANOVA was performed to study the overall mean percentage of
inhibition of proliferation, cell migration and colony formation as-
says. The results presented are mean ± SD and were considered
significant at p < 0.05.
10, 5 mL (TPP®,USA), water bath (PROTECH-Electronic, Malaysia),
6
-well cell culture plate (Costar Corning, USA).
3. Results and discussion
2
.3.2. Cell lines and environmental conditions
3.1. Synthesis and characterization
Human colorectal tumor (HCT 116) cell lines, human breast can-
cer cell line (MCF-7), human cervical cancer cell line (HeLA cell
line) and human endothelial cell line (EA.hy926 cell line) were pur-
chased from Rockville, MD, USA by Bio-Focus Saintifik Sdn Bhd.
Malaysia. HCT 116 cell line was kept in RPMI 1640 cell culture me-
dia with the supply of 10% HIFBS and 10% Pencillin/streptomycin
solution. Temperature was maintained at 37°C. The cell passage
number for HCT-116 was 9-11, MCF-7 was 20-23 and Ea.hy 926 was
The tri-NHC proligands (1-5) and respective trinuclear sil-
ver(I) complexes (6-10) were synthesized by
a multistep de-
signed scheme first reported by our group (Scheme 1-3)
Figure Scheme. 2[10]. The structure of the proligands (2-5) is pro-
posed to be similar to the reported single crystal structure of salt
1 [10], supported by 1H-, 13C-NMR, IR and elemental analysis.
The 1H- and 13C-NMR provide characteristic evidence for the
successful synthesis of the desired tris benzimidazolium salts and
the respective Ag(I) complexes. In 1H NMR of tris benzimidazolium
5
. The cell densities used for seeding cells of HCT-116, MCF-7 and
Ea.hy926 were 10,000, 5000 and 20,000 respectively.

4
T. Fatima, R.A. Haque and A. Ahmad et al. / Journal of Molecular Structure 1222 (2020) 128890
Scheme 1. Synthesis of 3-(2-bromoethyl)-1-substitutedbenzimidazolium bromides (I-V).
4
Scheme 2. Synthesis of tris benzimidazolium bromides (1-5).
salts (1-5), the acidic protons of the three pre-carbenic centres ex-
hibit two singlets in the most downfield region within 10.1-10.3
ppm [18], as these three protons are under two different chemical
environments. Resonances of all the bridging dimethylene protons
appear in the form of a singlet around 5.2 ppm in all of the salts
which signify the non-rigidity of the structures [29]. In addition
to the characteristic and common proton resonances of the core
structures, the proton resonances of the terminal substituents in
each salt appear in their expected regions, thus signifying the for-
mation of the desired salts. In 13C NMR the most distinctive res-
onances are displayed by the pre-carbenic carbons (NCHN) in the
most deshielded region in the range of 142-145 ppm [30,31].
The generation of carbene centres is also supported by 1H NMR
spectra of each complex as no resonances could be observed for
the acidic protons around 10 ppm in all spectra [32,33]. The res-
onances of the rest of the protons of the ligands are broadened
and overlapped, which implies that all the tri-NHC trinuclear Ag(I)
complexes have a highly flexible structure as previously reported
[
10]. The evidence for the formation of Ccarbene-Ag(I) bond is pro-
vided by 13C-NMR analysis in which the complexes show doublets
of doublets between 183-199 ppm with the coupling constants of
1
80.6 Hz for C-107 Ag and 209 Hz for C-109Ag for the Ccarbene-Ag(I)
coupling [34,35]. The Ag-carbene coupling was only observed in
the 13C-NMR spectrum of compounds 7 and 8 to any resolvable de-
gree while the compounds 9 and 10 had carbene resonances that
were sufficiently broadened to be undetectable. The structures of
all the tri-NHC trinuclear silver(I) complexes could not be further
supported by X-ray crystallography since all efforts to grow single
crystals were unsuccessful. As reported before [11], we postulate
that the tri-NHC trinuclear silver(I) complexes in this work exist
as trinuclear structures from two tri-NHC ligands with linear ge-
ometry, the evidence of structural symmetry in the NMR spectrum
provides clear support of the proposed structure. The coordination
of three carbene centres with three silver(I) ions in a tri-metal di-
ligand arrangement is provided by elemental analysis which sup-
ports the stoichiometry [Ag L ]·3PF for the complexes (vide supra)
3
2
6
(
6-10).
Scheme 3. Synthesis of tri-NHC trinuclear silver(I) complexes (6-10).

T. Fatima, R.A. Haque and A. Ahmad et al. / Journal of Molecular Structure 1222 (2020) 128890
5
Fig. 1. The previously reported tris azolium (benzimidazole/imidazole) salts (a); and the respective tri-NHC trinuclear silver(I)- complexes (b).
3
.2. In vitro cytotoxic activities of tri-NHC trinuclear silver(I)
fect of the compounds is in accordance with the reported results of
complexes and their respective proligands
mono-NHC trinuclear Ag(I) complexes [36]. The non-active tri-NHC
trinuclear Ag(I) complexes and the proligands may be attributed to
the different substitutions within these compounds. The results in
the present study are in accordance with the earlier findings that
the substitution patterns/steric bulk on the benzimidazolium ring
has a major influence on the anticancer potential [37] although a
few other studies have reported that the flexibility, nature of the
NHC ligands, hydrogen bonding capabilities of the NHC-Ag(I), the
number of silver centres and the charge can influence the anti-
cancer potential [38]. It can be deduced from the IC50 values that
the inactivity of these proligands (1-4) and their respective com-
plexes (6-8) might be caused by the steric bulk and an enhanced
stability of the molecules that might inhibit the activity by pre-
venting an adequate release of silver ions. The proligand 5 and its
respective complex 10 containing long, 10-carbon chains, however,
showed IC50 values of 14.98 μM and 0.26 μM, respectively, as com-
pared to that of the standard drug Table 1.
3
.2.1. In vitro cytotoxic activities of tri-NHC trinuclear silver(I)
complexes and their respective proligands against the human colon
cancer (HCT-116) cell line
It has been reported in the previous studies that the mono-
NHC trinuclear Ag(I) imidazolium complexes showed weak anti-
cancer activity against the human colon cancer cell lines [36].
Based on the observations and results of the anticancer activity
of mono- and di- NHC Ag(I) complexes and their respective pro-
ligands [24,25], the anticancer activity of the trinuclear Ag(I)-NHC
complexes derived from tris benzimidazolium salts is expected to
be enhanced by increasing the number of silver centres and the
NHC units in the complexes. In order to explore the cytotoxic ac-
tivity of the proligands and complexes containing three NHCs units
and three Ag(I) ions, aforementioned proligands (1-5) and their re-
spective complexes (6-10) were synthesized successfully in a pro-
grammed and schematic way. The tri-NHC trinuclear Ag(I) com-
plexes have not been studied previously for their antitumor activ-
ity and has yet to be assessed, thus implying that the antitumor
behaviour of these compounds could not be predicted. Thus, in or-
der to evaluate the anticancer potential of the tri-NHC trinuclear
Ag(I) complexes, the proligands (1-5) and the complexes (6-10)
containing different substituents were tested against the human
colon cancer cell line, HCT116. In the initial screening, all the syn-
thesized proligands and the complexes were evaluated for the IC50
values. Herein, most of the compounds are inactive on HCT116 ex-
cept for the proligand 5 and complexes 9 and 10 (Table 1). This ef-
The results support the fact that incorporation of long carbon
side chains may facilitate drug absorption and increase the cyto-
toxic activity of the tri NHC compounds, a similar phenomenon
observed in the mono- and di-NHC proligands and their respective
complexes [24,25]. Complex 9 containing 2-methylenebenzonitrile
has an IC50 approximately three times smaller than that of 5-
FU, thus implying that a functional group does play a role in
cytotoxic activity. The activity of this complex is in accordance
with those of the previously reported nitrile functionalized an-
ticancer drugs in in vitro and in vivo experiments on different
cell lines [39,40].

6
T. Fatima, R.A. Haque and A. Ahmad et al. / Journal of Molecular Structure 1222 (2020) 128890
Table. 1
IC50 values of the proligands (1-5) and their respective complexes (6-10) on HCT 116 and EA.hy926 using the standard drug 5-FU (IC50 = 10.2 μM), MCF-7
using the standard drug Tamoxifen (IC50 = 7.5μM) and HeLa cell lines using the standard drug 5-FU (IC50 = 10.2 μM) as a positive control.
Proligands
Tri NHC-Ag(I)
complexes
IC50 (μM) HCT 116
EA.hy926
IC50 (μM) MCF-7
IC50 (μM) HeLa cell lines
5-FU (IC50 = 10.2 μM)
Tamoxifen (IC50 = 7.5μM)
5-FU (IC50 = 10.2 μM)
1
2
3
4
5
6
NA
NA
NA
NA
3.2
0.26
NA
NA
NA
NA
35
6.16
12.3
9.4
NA
NA
NA
NA
1.42
7.78
NA
NA
NA
NA
5.43
2.77
8.65
0.40
10.89
1.51
7
NA
0.02
8
NA
14.83
18.66
0.27
9
NA
9.38
11
10
14.98
±
NA = not active
Fig. 2. Dose dependent proliferative effects of complex 9, 10 and proligand 5 against the human colon cancer cell line in comparison with the standard drug 5-FU.
The IC50 values of the synthesized tri-NHC compounds (pro-
ligands 1-4 and complexes 6-8) indicate that these compounds
are inactive against the human colon cancer cell lines, hence
their effects on the cell proliferation are not considered in this
study. The complex 9 containing a nitrile functional group shows
a dose dependent inhibition in the proliferation of cancer cells
on HCT116 (Fig. 2). Even though proligand 5 shows a percentage
inhibition of proliferation on HCT116 cell lines, this effect how-
ever is less prominent than that of the standard drug 5-FU. How-
ever, an exploration of these responses using complex 10 demon-
strates a significant higher percentage inhibition in cell prolifera-
tion in low, medium and high doses as compared to the respec-
tive proligand and the standard drug 5-FU (Fig. 2). The photomi-
crographic images of HCT116 cell lines and EA.hy926 cell lines af-
ter treated with proligands 5 and complexes 9 and 10 are shown
in Fig. 3. These compounds are found to be less toxic on the nor-
mal cell lines (EA.hy926), but relatively toxic on the HCT 116 cell
line.
3
.2.2. In vitro cytotoxic activities of tri-NHC trinuclear silver(I)
complexes and their respective proligands against human breast
cancer cell line (MCF-7)
The results of cytotoxic studies of the tri-NHC trinuclear sil-
ver(I) complexes and the respective proligands on HCT116 indicate
that these compounds are inactive, hence in order to find whether
these compounds are also inactive on other cancer cell lines, they
were further evaluated against breast cancer cell line (MCF-7). As
reported in the literature compounds which are inactive on one
cancer cell line are found to be active on the other cancer cell lines
Fig. 3. Images of human colorectal cancer cell lines and human endothelial cell
lines showing the effects of proligands 5 and complexes 9 and 10. (cell images
were taken under a light microscope at × 200 magnification with a digital cam-
era at 48 hours after treatment). Arrows points out the altered morphology of cells
(1 = cell shrinkage, 2 = dead cells, 3 = detached floating cells not seen, 4 = loss of
[
41]. Unlike the IC50 values on HCT116 cell lines, in which only the
contact between cells).
proligand 5 and the complexes 9 and 10 are active, the IC50 val-

T. Fatima, R.A. Haque and A. Ahmad et al. / Journal of Molecular Structure 1222 (2020) 128890
7
Fig. 4. Dose dependent antiproliferative effects of complexes 7, 8, 9, 10 and proligand 5 against breast cancer cell line in comparison with the standard drug, Tamoxifen.
ues of all the tri-NHC compounds show large variability, though
only proligand 5 and all trinuclear complexes show activity against
MCF-7 (Table 1).
Another significant result is that the complex 10 shows greater
percentage inhibition of proliferation as compared to that of the
respective proligand 5, although both show dose dependent re-
sponses on MCF-7. All complexes show cytotoxicity against the
breast cancer cells which is in accordance with the results of a
previous study which reports the imidazolium-based mononuclear
silver(I)-NHC complexes as anticancer agents against breast cancer
cell lines [41].
It is of great interest that proligand 5 and the respective com-
plex 10 show an IC50 of 5.2 and 28 times, respectively, lesser when
compared to that of the standard drug, Tamoxifen (IC50 = 7.5μM).
Furthermore, the antiproliferation activity of the proligand 5 and
complexes (6-10) with different doses in the range of 1.56-50 μM,
along with the same dilution of the standard drug Tamoxifen were
also examined. For complex 6, although the IC50 value is compa-
rable to that of the standard drug, the antiproliferation activity is
not promising. Besides, complex 7 with n-butyl side chains shows
a dose dependent response (Fig. 4). This complex displays antipro-
liferative activity on the breast cancer cell lines at lower, medium
and higher doses and these responses are higher in percentage
as compared to that of Tamoxifen (Fig. 4). Similarly, complex 8
shows significant percentage inhibition of proliferation but lower
than that of the standard drug. This observation could be due to
the presence of steric bulk of the substituent group in complex 8.
Complex 9, containing 2-methylene benzonitrile showed dose de-
pendent responses of antiproliferation activity in a fashion similar
to that of complex 8.
3.2.3. In vitro cytotoxic activities of tri-NHC trinuclear silver(I)
complexes and their respective proligands against the human
epitheloid cervix carcinoma (HeLa) cell line
All the synthesized proligands and their respective tri-NHC trin-
uclear Ag(I) complexes were also investigated for their activity on
HeLa cell lines, in which their IC50 values were also calculated. All
complexes show an IC50 values similar to those observed on the
MCF-7 cell line. The IC50 values on HeLa cell line vary depend-
ing upon the type of substituents present in the structure moiety
(Table 1).
Interestingly, among all the studied proligands, only proligand
5 shows activity on the HeLa cell line with an IC50 of 5.43 μM,
which is an indicative of a high lipophilicity of this proligand.

8
T. Fatima, R.A. Haque and A. Ahmad et al. / Journal of Molecular Structure 1222 (2020) 128890
Fig. 5. Dose dependent antiproliferative effects of complex 6, 7, 8, 9, 10 and proligand 5 in comparison with the standard drug 5-FU.
Aforementioned, increasing the carbon chain length will enhance
the lipophilicity, thereby facilitating a greater bioavailaibility of the
proligand in the cell [42]. The active compounds, after screening
for their IC50 values, were evaluated for their proliferation inhibi-
tion activities against the human epitheloid cervix cancer cell line,
keeping in view the fact that the silver complexes are reported be-
fore to have antiproliferative potential against the HeLa cell line
than those of the standard drug at lower and medium doses, but
comparable to that of the latter at the highest dose (Fig. 5). Com-
plex 7, exhibits a dose dependent inhibition of proliferation of the
HeLa cancer cells. Noteworthy, the respective proligand is inactive
to HeLa, thus suggesting that not only the NHC moiety is respon-
sible for the overall responses but such anticancer activities can
be attributed to the additional silver ions present in the complex.
Complex 8 shows greater responses in terms of percentage inhibi-
[
43]. Complex 6, shows dose dependent responses that are better

T. Fatima, R.A. Haque and A. Ahmad et al. / Journal of Molecular Structure 1222 (2020) 128890
9
tion of proliferation at lower and medium doses but comparable
to 5-FU at higher dose while complex 9 shows inhibition of prolif-
eration comparable to the similar standard drug (Fig. 5). The only
proligand in this series that shows activity on the HeLa cell lines
is the proligand 5 that shows higher percentage inhibition of pro-
liferation at lower and medium doses as compared to the standard
drug used. Additionally, complex 10 shows higher percentage inhi-
bition of proliferation at all doses compared to the respective pro-
ligand 5. At the maximum doses (25 and 50 μM), the percentage
inhibition of proliferation of complex 10 and 5-FU are more com-
parable than proligand 5. All the complexes 6-10 show potential
cytotoxicity against the cervical cancer which is in contrast with
the previous findings of mono-NHC mononuclear Ag(I) complexes
on the same cell lines [41]. This thus suggesting that anticancer
activity can be attributed to both the additional silver ions present
in the complex and the NHC moiety. The observed phenomenon
of increased cytotoxicity of the NHC-Ag(I) complexes, in compari-
son to their respective proligands is in line with the previously re-
ported data [44]. The mechanism of cytotoxicity of these tri NHC-
Ag(I) complexes can be explored in future studies.
[2] T. Weskamp, V.P.W. Bohm, W.A. Herrmann, J. Organomet. Chem. 600 (2000)
2–22.
1
[3] F.E. Hahn, Chem.Rev. 118 (2018) 9455–9456.
[4] X. Hu, I. Castro-Rodriguez, K. Olsen, K. Meyer, Organomet. 23 (2004) 755–764.
[5] A. Rit, T. Pape, F.E. Hahn, J. Am. Chem. Soc. 132 (2010) 4572–4573.
[
6] X. Hu, I. Castro-Rodriguez, K. Meyer, J. Am. Chem. Soc. 125 (2003)
2237–12245.
1
[7] E. Mas-Marza, M. Poyatos, M. Sanau, E. Peris, Inorg. Chem. 43 (2004)
2213–2219.
[8] C.E. Ellul, G. Reed, M.F. Mahon, S.I. Pascu, M.K. Whittlesey, Organometallics 29
2010) 4097–4104.
(
[9] Y.W. Zhang, R. Das, Y. Li, Y.Y. Wang, Y.F. Han, Chem.-A Eur. J. 25 (2019)
5472–5479.
[10] T. Fatima, R.A. Haque, M.R. Razali, J. Mol. Struct. 1141 (2017) 346–350.
11] S. Gonell, M. Poyatos, E. Peris, Angewandte Chemie 52 (2013) 7009–7013.
[
[
12] A. Biffis, G.G. Lobbia, G. Papini, M. Pellei, C. Santini, E. Scattolin, C. Tubaro, J.
Organomet. Chem. 693 (2008) 3760–3766.
[13] X. Hu, Y. Tang, P. Gantzel, K. Meyer, Organometallics 22 (2003) 612–614.
[
[
14] S.Y. Hussaini, R.A. Haque, M.R. Razali, J. Organomet. Chem. 882 (2019) 96–111.
15] S. Budagumpi, R.A. Haque, S. Endud, G.U. Rehman, A.W. Salman, Eur. J. Inorg.
Chem. 2013 (2013) 4367–4388.
[16] S. Kankala, N. Thota, F. Bjorkling, M.K. Taylor, R. Vadde, R. Balusu, Drug De-
velop. Res. 80 (2019) 188–199.
[
17] S.A. Patil, S.A. Patil, R. Patil, R.S. Keri, S. Budagumpi, G.R. Balakrishna, M. Tacke,
Future Med. Chem. 7 (2015) 1305–1333.
[18] P.O. Asekunowo, R.A. Haque, M.R. Razali, Rev. Inorg. Chem. 37 (2017) 29–50.
19] S. Tan, Y. Yan, P. Lee, K. Lim, Future Med. Chem. 2 (2010) 1591–1608.
[
The cytotoxicity results of the present study on tri NHC-Ag(I)
complexes when compared to similar substituted mono, di and
tetra-NHC-Ag(I) complexes previously studied by our group indi-
cates a decrease in activity for tri and tetra NHC-Ag(I) complexes.
The factors responsible for such an unexpected decrease in anti-
cancer potential are the poor solubility of these bulkier complexes
that reduces their bioavailability and efficacy, in addition an inad-
equate release of Ag(I) ions due to an increase in the NHC units in
the ligand molecule that might tightly hold the silver ions.
[
20] L. Eloy, A.S. Jarrousse, M.L. Teyssot, A. Gautier, L. Morel, C. Jolivalt, T. Cresteil,
S. Roland, Chem. Med. Chem. 7 (2012) 805–814.
[21] J. Dinda, T. Samanta, A. Nandy, K.D. Saha, S.K. Seth, S.K. Chattopadhyay,
C.W. Bielawski, New J. Chem. 38 (2014) 1218–1224.
[22] C. Saturnino, I. Barone, D. Iacopetta, A. Mariconda, M.S. Sinicropi, C. Rosano,
A. Campana, S. Catalano, P. Longo, S. Andò, Future Med. Chem.
2213–2229.
8 (2016)
[23] T. Fatima, R.A. Haque, M.A. Iqbal, A. Ahmad, L.E.A. Hassan, M. Taleb-Agha,
M.B.K. Ahamed, A.A. Majid, M.R. Razali, J. Organomet. Chem. 853 (2017)
122–135.
[
[
[
24] T. Fatima, R.A. Haque, M.R. Razali, A. Ahmad, M. Asif, M.B. Khadeer Ahamed,
A.M.S. Abdul Majid, Appl. Organomet. Chem 31 (2017) e3735.
25] T. Fatima, R.A. Haque, M.R. Razali, A. Ahmad, M.A. Iqbal, M. Asif,
M.B.K. Ahamed, A. Abdul Majid, J. Coord. Chem. 69 (2016) 3367–3383.
26] T. Fatima, R.A. Haque, M.A. Iqbal, A. Ahmad, L.E.A. Hassan, M. Taleb-Agha,
M.B.K. Ahamed, A.M.S.A. Majid, M.R. Razali, J. Organomet. Chem. 853 (2017)
4. Conclusion
This study accounts for the successful synthesis and charac-
1
22–135.
27] T. Fatima, R.A. Haque, M.A. Iqbal, M.R. Razali, J. Organomet. Chem. 831 (2017)
0–54.
terization of tris benzimidazolium bromide salts (1-5) and their
respective trinuclear silver(I) complexes (6-10). The cytotoxic po-
tential of the tested compounds (1-10) determined by IC50 values
and antiproliferative activity against human colon, human breast
and human epitheloid cervix cancer cell lines revealed these com-
pounds to be more cytotoxic towards breast cancer and cervical
cancer cell lines and selectively toxic against human colon cancer
cell line. On all studied cell lines, the complexes are found to be
cytotoxic compared to their proligands, which signifies the ability
of silver ions to kill cancer cells. Among all the active compounds,
complex 10 containing n-decyl substituent exhibits pronounced ac-
tivity on the selected cell lines which imply the role of enhanced
lipophilicity on drug absorption in addition to the silver ions.
[
5
[28] A.F. Aisha, K.M. Abu-Salah, Z. Ismail, A.M.S.A. Majid, BMC Complement. Altern.
Med. 12 (2012) 104.
[
29] O. Sanchez, S. González, M. Fernández, A.R. Higuera-Padilla, Y. Leon, D. Coll,
A. Vidal, P. Taylor, I. Urdanibia, M.C. Goite, Inorg. Chim. Acta 437 (2015)
143–151.
[
30] B. Yigit, Y. Gok, I. Ozdemirr, S. Gunal, J. Coord. Chem. 65 (2012) 371–379.
[
31] C. O’Beirne, H.T. Althani, O. Dada, J. Cassidy, K. Kavanagh, H. Muller-Bunz,
Y. Ortin, X. Zhu, M. Tacke, Polyhedron 149 (2018) 95–103.
[32] S. Patil, K. Dietrich, A. Deally, B. Gleeson, H. Muller Bunz, F. Paradisi, M. Tacke,
Helvetica Chim. Acta 93 (2010) 2347–2364.
33] C. O’Beirne, N.F. Alhamad, Q. Ma, H. Muller-Bunz, K. Kavanagh, G. Butler,
X. Zhu, M. Tacke, Inorg. Chim. Acta 486 (2019) 294–303.
34] B. Bildstein, M. Malaun, H. Kopacka, K. Wurst, M. Mitterback, K.-H. Ongania,
G. Opromolla, P. Zanello, Organometallics 18 (1999) 4325–4336.
35] J.C. Garrison, W.J. Youngs, Chem. Rev. 105 (2005) 3978–4008.
36] A.W. Salman, R.A. Haque, Eur. J. Chem. 7 (2016) 115–120.
[
[
[
[
Declaration of Competing Interest
[
37] R.A. Haque, S.Y. Choo, S. Budagumpi, M.A. Iqbal, A.A.-A. Abdullah, Eur. J. Med.
Chem. 90 (2015) 82–92.
[
38] S. Ray, R. Mohan, J.K. Singh, M.K. Samantaray, M.M. Shaikh, D. Panda, P. Ghosh,
J. Am. Chem. Soc. 129 (2007) 15042–15053.
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared to
influence the work reported in this paper.
[39] F. Iduna, C. Jindrich, M. Martin, C.S. Lara, H. Ralf, N.K. Breandan, L.R. Alison,
H. Frauke, L. Grainne, J.Q. Susan, M. Ian, T. Matthias, Lett. Drug Design Discov.
9
(2012) 815–822.
[
[
40] H.Z. Zulikha, R.A. Haque, S. Budagumpi, A.M.A. Majid, Inorg. Chim. Acta 411
Acknowledgement
(2014) 40–47.
41] M.Doug A, H.Khadijah M, P.Matthew J, D.Andrew J, Y.Yang H, Y.Wiley J, Met-
al-Based Drugs, 2008 (2008).
The authors thank USM for financial support through funding
[42] S.N. Riduan, Y. Zhang, Chem. Soc. Rev. 42 (2013) 9055–9070.
1001/PKIMIA/8011089.
[43] R. Sukirtha, K.M. Priyanka, J.J. Antony, S. Kamalakkannan, R. Thangam, P. Gu-
nasekaran, M. Krishnan, S. Achiraman, Process Biochem. 47 (2012) 273–279.
References
[44] N.A. Johnson, M.R. Southerland, W.J. Youngs, Molecules 22 (2017) 1263.
[
1] J. Cheng, L. Wang, P. Wang, L. Deng, Chem. Rev. 118 (2018) 9930–9987.