Electrochemical synthesis and: In vitro cytotoxicity study of copper(ii) carboxylates with different fatty acid alkyl chain lengths

Article abstract of DOI:10.1039/c8nj02783h

In the present study, an electrochemical technique based on the release of Cu²⁺ ions from a Cu anode in the presence of unsaturated fatty acids with different alkyl chain lengths has been used to synthesize Cu(ii) carboxylates. The fatty acids used in this study are lauric acid (C12:0) and stearic acid (C18:0). Optimum electrolysis conditions for the synthesis of Cu(ii) laurate (CuLa₂) and Cu(ii) stearate (CuSt₂) have been determined to maximize percentage yield and minimize energy consumption and loss of the Cu anode. We observe that both compounds (99.21%) are produced with lower energy consumption (～21.01 W h L^-1) and anode loss (～0.57 mg L^-1) by using the same optimum conditions of 10 V of applied voltage for 4 hours of electrolysis time in 0.1 M CH₃COONH₄ electrolyte solution. The cytotoxicity study on selected tumor cells (A549 and HeLa) shows that the synthesized compounds have moderate cytotoxic effects with IC₅₀ in the range from 19.50 to 44.67 μM. CuLa₂ with C12 alkyl chains provides better cytotoxicity effect on the selected tumor cells due to lower IC₅₀ than CuSt₂ with C18 alkyl chains. This shows that the length of alkyl chain also affects the compound toxicity towards selected tumor cells.

Full text of DOI:10.1039/c8nj02783h

NJC
View Article Online
View Journal
PAPER
Electrochemical synthesis and in vitro cytotoxicity
study of copper(II) carboxylates with different
fatty acid alkyl chain lengths†
Cite this:DOI: 10.1039/c8nj02783h
a
b
c
Norazzizi Nordin, * Badrul Hisham Yahaya and Muhammad Rahimi Yusop
2+
In the present study, an electrochemical technique based on the release of Cu ions from a Cu anode
in the presence of unsaturated fatty acids with different alkyl chain lengths has been used to synthesize
Cu(II) carboxylates. The fatty acids used in this study are lauric acid (C12 : 0) and stearic acid (C18 : 0).
2 2
Optimum electrolysis conditions for the synthesis of Cu(II) laurate (CuLa ) and Cu(II) stearate (CuSt ) have
been determined to maximize percentage yield and minimize energy consumption and loss of the Cu
anode. We observe that both compounds (99.21%) are produced with lower energy consumption
ꢀ
1
ꢀ1
(
B21.01 W h L ) and anode loss (B0.57 mg L ) by using the same optimum conditions of 10 V of
applied voltage for 4 hours of electrolysis time in 0.1 M CH COONH electrolyte solution. The
cytotoxicity study on selected tumor cells (A549 and HeLa) shows that the synthesized compounds have
moderate cytotoxic effects with IC50 in the range from 19.50 to 44.67 mM. CuLa with C12 alkyl chains
provides better cytotoxicity effect on the selected tumor cells due to lower IC50 than CuSt with C18
3
4
Received 5th June 2018,
Accepted 2nd August 2018
2
DOI: 10.1039/c8nj02783h
2
alkyl chains. This shows that the length of alkyl chain also affects the compound toxicity towards
selected tumor cells.
rsc.li/njc
method permits the preparation of many substances that cannot
be prepared using conventional methods.
Introduction
8
Over the past decade, copper(II) carboxylates have received
Since the discovery of the efficient antitumor activity of Cu in
9
considerable attention due to their unique geometrical structures, 1960, the synthesis and antitumor activity of various biologically
physicochemical properties, and their many applications in active Cu(II) complexes with different ligands have attracted
1
–4
biochemistry and in industry. Previous studies have reported significant attention from researchers. Various types of Cu(II)
that most of the complexes are synthesized using chemical complexes with different ligands exhibiting antitumor activity
1
,3–7
10–15
methods.
Herein, CuLa
2
and CuSt
2
were prepared using have been reported.
Table S1 (ESI†) shows the IC50 values
an electrochemical technique, which offers several advantages (the concentration of the compound causes 50% growth inhibition)
in inorganic synthesis: (1) it can achieve oxidizing or reducing obtained from previous studies on A549 and HeLa cells (the same
power that cannot be equaled by any ordinary chemical reagent; cells used in this study) treated with Cu(II) complexes with different
(
2) there is no contamination to the solution with a reducing active ligands. According to Marzano et al., antitumor compounds
agent (or oxidizing agent) and its corresponding oxidation that exhibit lower IC50 values in the micromolar to submicromolar
or reduction) product; (3) the selectivity of the reaction can (o10 mM) range are categorized as effective antitumor compounds,
be influenced by the applied electrode potential; and (4) this whereas compounds with IC 4 10 mM are classified as moderate
(
50
9
antitumor compounds. Cu(II) compounds with Schiff-base
ligand (2-((quinolin-8-ylimino)methyl)pyridine) discovered by
Lu et al. act as effective antitumor compounds on selected
cells due to low IC50 concentrations (o10 mM) (Table S1, ESI†).
Previous studies have reported that the toxicity of metal-based
antitumor agents not only depends on the metal used, but also
a
b
c
School of Chemical Sciences, Universiti Sains Malaysia, 11800 Gelugor,
Pulau Pinang, Malaysia. E-mail: azzizi@usm.my
10
Regenerative Medicine Cluster, Advance Medical and Dental Institute,
Universiti Sains Malaysia, Bertam, 13200 Kepala Batas, Pulau Pinang, Malaysia
School of Chemical Sciences and Food Technology, Faculty of Science and
Technology, Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor,
Malaysia
depends on the active ligand bound to the metal in the
12–17
complex.
This is because the nature of the ligand used
†
Electronic supplementary information (ESI) available: IC50 values for Cu(II)
and the type of donor atoms in these ligands can significantly
impact the biological and chemical properties of the synthesized
complexes from previous study, percentage yield and energy consumption for the
synthesis of CuLa and CuSt , FTIR, UV-Vis and XPS spectra of CuLa and CuSt
2
2
2
2
18,19
and XRD diffractogram of CuO and Cu by-products. See DOI: 10.1039/c8nj02783h compound.
A lower value of IC50 shows that the compound is
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
more reactive and can provide a better cytotoxic effect on the fatty acid solution; the fatty acid solution was mixed with
6,16,19
selected cells.
3 4
CH COONH solution at a ratio of 1 : 1 (v/v) in a simple
The use of fatty acids as ligands in the synthesis of Cu(II) undivided electrolysis cell (100 mL capacity). The reaction was
compounds can increase the toxicity of the synthesized compounds. performed at room temperature (27 1C) for 4 hours. The
Previous studies have shown that unsaturated fatty acids are resulting Cu(II) carboxylate precipitate was obtained as a blue
involved in the cell death pathway. Sparagna et al. discovered that solid in the organic phase, and it was then separated from the
20
palmitic acid showed proapoptotic effects on cardiac cells.
aqueous phase and washed with distilled water and ethanol to
A similar finding has been reported by Hardy et al. in which palmitic remove impurities. The precipitate was then dried in an oven
21
acid causes apoptosis in MDA-MB-231 breast cancer cells. Penzo (50 1C) for 2 hours and allowed to dry at room temperature.
et al. found that unsaturated fatty acids show negligible cytotoxicity Analysis of electrochemically synthesized CuLa : m.p. = 103–
2
2
2
effect on Morris Hepatoma (MH 1C1) cells, which can be 113 1C, yield = 99.69%; CuSt : m.p. = 110–120 1C, yield = 99.21%.
2
explained by the physical interaction of fatty acids with cell
membrane and its potential to alter their permeability. However,
Percentage yield and energy consumption
this permeability may vary according to the length of the alkyl The percentage yield for electrochemically synthesized CuLa2
chain in fatty acids. In the present study, we focus on optimizing and CuSt was determined using the following equation:
2
electrolysis conditions in the electrochemical synthesis of CuLa
and CuSt to minimize energy consumption, maximize the
percentage of yield and reduce the loss of Cu anode. We also
2
Actual yield ðgÞ
Yield ð%Þ ¼
ꢁ 100%
(1)
2
Theoretical yield ðgÞ
aim to identify the effect of different alkyl chain lengths in fatty Actual mass was determined by weighing the powder obtained
acid ligands of the synthesized compounds on in vitro cytotoxicity after synthesis. Energy consumption was calculated using the
against A549 and HeLa cell lines.
following equation:
À
Á
Power ðWÞ ꢁ Time ðhourÞ
Volume ðLÞ
ꢀ1
Energy consumption W h L
¼
Experimental
Chemicals and cell culture
(2)
All chemicals were of analytical grade and used without further Instrumentation
purification. Lauric acid (HLa) and stearic acid (HSt) with 99%
purity were purchased from Systerm. Cu foil (99% purity) and
ethanol (95% purity) were purchased from Sigma-Aldrich.
The synthesized compounds were characterized using several
chemical instruments. Fourier transform infrared spectroscopy
(
FTIR) (Perkin Elmer 1310) was performed in the range of 4000–
Ammonium acetate (CH
3
COONH
4
) was purchased from Hamburg
ꢀ1
400 cm using KBr pellets. Ultraviolet-Visible spectra (UV-Vis)
Chemical GmbH. Human non-small lung carcinoma (A549) and
normal human lung fibroblast (IMR90) cell lines were obtained
from American Type Culture Collection (ATCC) (No. catalogue
CCL-185t and CCL-186t), whereas Human cervical carcinoma
(
Shimadzu UV-2450) were obtained in the wavelength range
from 300 to 800 nm using 10 mm quartz cuvettes and DMSO as
the reference solvent at room temperature. X-ray photoelectron
spectroscopy (XPS) (AXIS Ultra ‘DLD’ by using the second
edition of the Kratos software) and a CHNS analyzer (Fison
EA 1108) were used to determine the elemental composition of
the synthesized complexes. Melting points were obtained in a
capillary tube using an Electrothermal IA9100 melting point
apparatus. Absorption measurements for the determination of
IC50 values were done using an Elisa Microplate Reader (BMG
Labtech) at wavelength of 570 nm.
(HeLa) cell lines were donated by Assoc. Prof. Dr Md Azman Seeni
Mohamed’s laboratory. A549, HeLa and IMR90 cells were
maintained in the Roswell Park Memorial Institute (RPMI)
medium, Dulbecco’s Modified Eagle’s Medium (DMEM) and
Modified Eagle’s Medium Alpha (a MEM), respectively. Each of
the media solutions was supplemented with 10% Foetal Bovine
Serum (FBS) and 1% antibiotic (100 units per mL penicillin and
ꢀ1
1
00 mg mL streptomycin). All cell culture reagents were purchased
from Gibco-Life Technologies unless otherwise stated.
Cytotoxicity and cell proliferation assay
The cytotoxicity and proliferative studies of A549 and HeLa cells
upon treatment with CuLa and CuSt were conducted using a
2 2
Electrochemical synthesis
CuLa
technique. The electrolysis system consisted of a Cu foil (3 ꢁ 96-well flat-bottomed microplate at a density of 1 ꢁ 10 cells per
cm, 0.1 cm thickness, Sigma-Aldrich) as the anode and mL in each well; then, they were incubated at 37 1C in 5% CO
graphite rod (6.5 mm diameter, Johnson Matthey Chemicals environment. When the cells reached B80% confluency, the
2 2
and CuSt were synthesized using an electrochemical PrestoBluet cell viability assay. First, the cells were seeded in a
5
1
2
ꢀ
1
Ltd) as the cathode in the presence of 0.1 mol L CH COONH
medium in each well was removed and replaced with a medium
3
4
as the supporting electrolyte solution. Prior to synthesis, both containing CuLa2 or CuSt2 at various concentrations in the
anode and cathode were rinsed with distilled water and a small range from 0 to 100 mM. After 72 hour of treatment, the medium
amount of acetone. A direct current (DC) power supply (TTi PSU containing tested complexes was removed; it was replaced with a
Bench CPX400A) was used throughout the electrochemical fresh medium containing 10 mL of PrestoBluet reagent (Invitrogen)
ꢀ1
synthesis. Ethanol was used as the solvent to prepare 0.1 mol L
and incubated for another 30 min. The optical density (OD) of
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
the cells was measured at days 1, 3, 5 and 7 at 570 nm wavelength anode corrosion. Therefore, Cu corrosion is expected to be the
using a FLUOstar Omega Microplate Reader (BMG Labtech). IC50 same for the synthesis of both compounds (CuLa and CuSt ).
values (the concentration of the complex caused 50% growth The supporting electrolyte plays an important role in main-
2
2
inhibition) were determined and used as a parameter to compare taining constant ionic strength, reducing cell resistance of the
8
relative cytotoxicity of the complexes. The cells incubated solution and eliminating electromigration effects. In this study,
without the tested complex were used as a negative control. some low-cost common inorganic salts with high conductivity such
Morphological changes of the cells due to the penetration of as sodium chloride (NaCl), potassium nitrate (KNO
the complex were observed under a light inverted microscope chloride (NH Cl) and ammonium acetate (CH COONH
Olympus IX51) at 20ꢁ magnification. The IC50 values of CuLa used as supporting electrolytes. The effect of the type of supporting
and CuSt on A549 and HeLa cells were also tested on IMR90 electrolyte on Cu corrosion has been presented in Fig. 1. The results
3
), ammonium
4
3
4
) have been
(
2
2
2+
cells to determine the cytotoxicity effect of complexes on showed that more Cu ions were released from the corroded
normal cells. In this assay, the IMR90 cells were treated in a anode into the aqueous phase of the CH COONH supporting
3
4
similar manner to that for A549 and HeLa cells.
electrolyte solution compared to those for other types of supporting
electrolytes (Fig. 1a). According to Bieszczad and Sanak-Rydlewska,
Statistical analysis
CH
possesses high conductivity in electrochemical synthesis. Our
results confirmed that CH COONH provides better conductivity
3 4
COONH is an acceptable supporting electrolyte material that
23
The data were expressed as mean ꢂ standard error (SE) with at
least three replicates. The statistical analysis was performed
using Statistical Package for Social Sciences (SPSS) version 17.0
with one-way ANOVA. The data obtained were considered
statistically significant when the P value was o 0.05.
3
4
and highest efficiency during electrochemical synthesis.
Electrolysis time is one of the major concerns in chemical
synthesis since longer time may lead to higher energy
2
4
consumption. In the current study, an increase in the reaction
time generally resulted in increased corrosion of the anode
2
+
Results and discussion
Electrochemical synthesis of CuLa and CuSt
(Fig. 1b circle symbol), which produced more Cu ions in the
ꢀ
1
aqueous phase. For 0.1 mol L of HLa, the Cu concentration
was generally very low for the first 3 hours due to the formation
2
2
The mechanism for the formation of CuLa
electrochemical technique is as follows:
2
or CuSt
2
using an
2
of the CuLa complex (Fig. 1b square symbol). After 3 hours, no
free fatty acid was available in the organic phase, and excess
2
+
0
2+
ꢀ
Cu ions were created in the solution. Since very long reaction
time results in increased costs due to excess energy consumption
and wasted metal species in the aqueous phase, a 4 hour
Anode: Cu - Cu + 2e
(3)
(4)
(5)
(6)
ꢀ
ꢀ
Cathode: 2e + 2H O - H + 2OH
2
2
electrolysis time was considered ideal to completely react the
ꢀ
ꢀ
2
OH + 2RCOOH - 2H
2
O + 2RCOO
2+
3 4
fatty acid with Cu ions. Electrolysis of CH COONH without
fatty acids (Fig. 1b triangle symbol) was performed as a control
experiment to prove that increasing the reaction time produced
2 2
Overall: Cu1 + 2RCOOH - Cu[RCOO] + H
2
+
2+
Here, R represents the alkyl chain of the fatty acid ligand. Cu
ions (eqn (3)) were generated in the aqueous phase by electro-
more Cu ions in the aqueous phase.
High voltages should not be applied during electrochemical
chemical oxidation of the positively charged Cu anode. At the synthesis to avoid loss of energy, high temperatures, and electrode
ꢀ
25
negatively charged graphite cathode, OH ions were generated damage. In our study, an increase in the applied voltage was
from the reduction of water (eqn (4)) and directly reacted with observed to increase corrosion percentage of the Cu anode (Fig. 1c
ꢀ
fatty acid (RCOOH) to form carboxylate ions (RCOO ions) circle symbol). During the synthesis of CuLa
(
2
(Fig. 1c square
2
+
eqn (5)) in aqueous phase. The reaction between Cu ions symbol), complete utilization of fatty acids was achieved with the
and decanoate ions resulted in the formation of CuLa
in the organic phase as a blue precipitate (eqn (6)).
2
or CuSt
2
use of 10 V of applied voltage after 4 hours with less formation of
excess Cu ions. At 5 V, no excess Cu ions formed in solution,
2+
2+
As described by the previous mechanism, the formation of but higher concentrations of free HLa were still present. Thus, the
2
+
CuLa2 and CuSt2 depends on the release of Cu ions from total reaction time at 5 V was possibly greater than 4 hours. The
electrochemical oxidation of the Cu anode. Thus, optimization use of even higher voltages (410 V) may reduce the total reaction
of electrolysis conditions should be performed to obtain optimum time but also increase the potential for the formation of certain
2
6
2 2
conditions for the synthesis of CuLa and CuSt . The electrolysis by-products such as Cu(II) oxides. Therefore, we concluded
conditions that have been tested were type of supporting that 10 V was the ideal applied voltage for the synthesis of CuLa
electrolytes, electrolysis time, applied voltage and concentration to prevent the formation of undesired compounds.
2
of supporting electrolyte. Therefore, the results of this optimization
study are expected to maximize the production of CuLa and CuSt
The main purpose of adding CH COONH during synthesis
3 4
is to increase electrolytic conductivity of the electrochemical
2
2
and minimize the waste of the Cu anode and energy consumption. system, which in turn affects the current efficiency, cell voltage,
2
7
In this study, we discuss the optimization of electrolysis conditions and energy consumption in the electrolytic cell. Increasing
for the synthesis of CuLa only. This is because the presence of fatty the concentration of CH COONH was observed to increase the
acid (HLa or HSt) in the electrolysis solution does not affect Cu corrosion percentage of the Cu anode (Fig. 1d circle symbol),
2
3
4
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
Fig. 1 Effects of (a) type of supporting electrolytes, (b) electrolysis times, (c) applied voltages, (d) supporting electrolyte concentrations on percentage of
2
+
ꢀ1
corroded anode (a–d) and Cu ion concentration (b–d) in the aqueous phase during electrochemical synthesis of CuLa . [HLa] = 0.1 mol L (a–d);
2 o
ꢀ1
[
electrolyte] = 0.1 mol L (a–c); E = 10 V (a, b and d); t = 4 h (a, c and d).
2
+
which resulted in an increase in the Cu concentration. Based (ꢂ0.31) and 99.21% (ꢂ0.33) and energy consumption values
ꢀ
1
ꢀ1
on our results, the ideal electrolyte concentration was 0.1 mol L
CH COONH for 4 hours of reaction time (Fig. 1d square symbol). Therefore, we can infer that the selected electrolysis parameters
Further increase in CH COONH concentration produced excess have maximized the formation of CuLa and CuSt (B100%
Cu ions. A major drawback of using high concentrations of percentage of yield), minimized Cu loss (0.57 mg L ) (Fig. 1)
of 20.07 (ꢂ0.07) and 21.01 W h L
(ꢂ0.04), respectively.
3
4
3
4
2
2
ꢀ1
2
+
ꢀ
1
ꢀ1
CH COONH (e.g., 40.5 mol L ) in electrochemical synthesis and optimized energy consumption (B21.0 W h L ).
3
4
is heavier cathode plating. Therefore, we determined that
ꢀ
1
Characterization
concentration of supporting electrolyte for use in the electro- The elemental compositions of the synthesized CuLa
chemical synthesis of CuLa CuSt were analyzed using CHNS. The elemental analysis for
Optimization of electrolysis conditions during the synthesis CuLa
of CuLa and CuSt was performed to minimize energy consumption C (62.37%), H (10.03%), O (13.85%), Cu (13.74%); (found): C
as well as to maximize the percentage yield. The percentage of (61.70%), (11.30%), (14.26%), Cu (12.74%), CuSt
compounds produced and energy consumption during CuLa2 (CuC H O ) (629.55 g mol ) (calculated): C (67.96%), H
0
.1 mol L
3 4
of CH COONH solution was the appropriate
2
and
2
.
2
ꢀ1
2
46 4
(CuC24H O ) (461.55 g mol ) (calculated) is as follows:
2
2
H
O
2
ꢀ1
36
70 4
and CuSt syntheses using the parameters described in Fig. 1 (11.09%), O (10.06%), Cu (10.00%); (found): C (67.34%), H
2
are shown in Table S2 (ESI†). Generally, the percentage yields (12.20%), O (10.36%), Cu (10.10%). The synthesized CuLa2
and energy consumptions (Table S2, ESI†) for CuLa
2
and CuSt
2
2
and CuSt were also characterized using FTIR, UV-Vis and
syntheses did not show significant differences. This proved that XPS spectroscopies to prove that the synthesized compounds
any difference in the length of the alkyl chain in fatty acids did are the desired compounds.
not affect percentage yield and energy consumption.
2
Fig. S1a (ESI†) shows the FTIR spectra of CuLa and free HLa
Energy consumption data (Table S2, ESI†) for all optimized as an experimental control. The FTIR spectrum of free HLa
ꢀ
1
parameters can be assigned to the anode corrosion percentages shows an intense peak at wavenumber of 1697 cm , which
in Fig. 1(a–d), where energy consumption is directly proportional represents the stretching vibration of carboxyl group in fatty
2
8
ꢀ1
to the percentage of Cu anode corrosion. The synthesis of CuLa
and CuSt using optimized electrolysis conditions (applied which represent the asymmetric (vas) and symmetric (v
voltage = 10 V; CH
electrolysis time = 4 hour) gives percentage yields of 99.69 carbon chain, respectively. After complex formation between
2
acid. Several peaks appear in the range of 2800–3000 cm ,
2
s
)
ꢀ
1
3
COONH
4
concentration = 0.1 mol L
;
stretching vibrations of methyl and methylene groups in HLa
1
–4
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
2
+
HLa and Cu ions, the FTIR spectrum for the product formed, scan XPS spectrum of C 1s (Fig. S2b, ESI†) is used as a reference
as shown in Fig. S1a (ESI†), still shows peaks in the range of for the calculation of binding energy. All three signals that
ꢀ1
2
800–3000 cm , which represent the asymmetrical (v ) and appear in the XPS spectrum of C 1s represent –COO (288.3 eV)
as
symmetrical (v ) stretching vibrations for methyl and methylene and –C–O (285.4 eV), whereas the signal at 284.5 eV represents
s
3
1,32
groups in the carbon chain of the ligand, respectively. In the carbon chain of fatty acid ligands (–C–C–).
addition, the peak representing the stretching vibrations of For oxygen signal, the O 1s spectrum in Fig. S2c (ESI†) shows
the carboxylic groups of free HLa is no longer visible. This the presence of three signals in the compound; these signals
proves that all of the HLa available in the solution has reacted represent –O–C– (532.9 eV), OQC–O (530.6 eV) and O–Cu
3
0
completely to form the desired product.
(531.5 eV). The signal representing the O–Cu bond is very
New peaks appear after the formation of the complex at important for proving the occurrence of complex formation
ꢀ
1
2+
ꢀ
wavenumbers of 1591 and 1440 cm (a and b in Fig. S1a, ESI,† between Cu ions and –COO groups on the alkyl chain of
respectively), and these peaks are very important in determining ligands. The narrow scan of Cu 2p spectrum for the synthesized
2
+
the coordination between Cu and the carboxylate group in the complex in Fig. S2d (ESI†) shows the presence of four signals.
6
ligands. Peak a in Fig. S1a (ESI†) represents the asymmetrical The signal at 953.9 eV is ascribed to 2p1/2, whereas the signal at
stretching vibration (vCOO,as), and the peak b represents the 934.1 eV is assigned to 2p3/2. The Cu 2p3/2 signal represents Cu(II)
32
symmetrical stretching vibration (vCOO,s) for the coordinated that has been coordinated to carboxylate ligands. Another two
6
carboxylate group. These peaks are important features for a signals (Fig. 3d) obtained at binding energies of 943.5 and
6
dicopper tetracarboxylate complex. Previous studies have 939.1 eV, respectively, are known as satellite signals, and they
32
suggested that the peak positions of vCOO,as and vCOO,s for the indicate the presence of Cu(II).
coordinated carboxylate group can be used to distinguish
the type of carboxylate–metal complexation structure by using
Characterization of by-products from the electrochemical
synthesis of CuLa and CuSt
As discussed in Section 3.1, the use of high applied voltage and
2
2
the equation DvCOO = vCOO,as ꢀ vCOO,s proposed by Deacon and
2
9
ꢀ1
Phillips. The value of DvCOO in the range from 150 to 170 cm
represents the bridging bidentate mode of Cu(II) coordination; supporting electrolyte concentration, which exceed the optimum
ꢀ
1
when DvCOO 4 200 cm , it shows that the synthesized compound conditions (applied voltage exceeding 10 V and the supporting
30
contains monodentate carboxylic groups. The value of DvCOO electrolyte concentration exceeding 0.1 M), for 4 hour synthesis of
ꢀ1
(
151 cm ) obtained in this study indicates that the synthesized CuLa results in the formation of undesirable by-products (CuO
2
complex has bridging bidentate coordination mode with polynuclear and Cu(OH) ). This black solid formed in the electrolysis solution
2
structure. FTIR spectrum of CuSt has a similar shape with slightly is analyzed using a powder X-ray diffractometer (PXRD). The
2
8
different wavenumbers, as shown in our previous article.
diffractogram in Fig. S3a (ESI†) shows that the resulting
Fig. S1b (ESI†) shows the UV-Vis spectra of CuLa
2
, CuSt
2
and by-products are a mixture of CuO and Cu metal when compared
free HLa. Before complex formation, HLa spectrum does not to known CuO (JCPDS 73-6371) and Cu (JCPDS 01-1241) files.
show any peaks in the wavelength range from 300 to 800 nm. The same by-products obtained in this study have also been
2
+
33
However, after complex formation between Cu ion and HLa, a reported by Yuan et al. The mechanism for the electrochemical
broad absorption band appears in the visible region (400 to formation of CuO can be proposed as follows:
9
00 nm) at 670 nm, which represents the d - d transition of
2
+
ꢀ
2
+
1,2
Step 1: Cu - Cu + 2e (at anode)
(7)
(8)
Cu ion. The d - d transition is due to the transition from
2
2
2
ꢀ
ꢀ
the filled d level (z , xy, xz/yz) to the half-filled d level d(x ꢀ y )
2 2
Step 2: 2e + 2H O - H + 2OH (at cathode)
5
,6,28
containing unpaired electrons.
In addition, the UV-Vis
2
+
ꢀ
Step 3: Cu + 2OH - Cu(OH) (in bulk solution)
(9)
2
spectra for CuLa and CuSt (Fig. S1b, ESI†) also show the presence
2
2
of a weak absorption band in the range of 350–400 nm in the UV
region (200 to 400 nm). This band refers to the dinuclear structure
of Cu(II) carboxylate in which two Cu atoms bind to four bridges
Cu(OH)
2
- CuO + H
2
O
(10)
2
+
ꢀ
Cu and OH ions are simultaneously generated on the anode
and cathode surfaces, respectively, (eqn (7) and eqn (8)) at the
same time when direct currents flow through an electrochemical
4,7
three atoms of O–C–O (Cu–O–C–O–Cu) (Fig. S1c, ESI†). There-
fore, the synthesized CuLa2 and CuSt are expected to form a
2
2
+
ꢀ
cell. Then, Cu ions react with OH ions to produce Cu(OH)
eqn (9)). Next, Cu(OH) dehydration results in the formation of
2
paddlewheel dimer structure, as shown in Fig. S1c (ESI†).
XPS analysis is carried out to determine the surface electronic
state and to identify the elemental composition of synthesized
(
2
CuO (eqn (10). The resulting Cu metal (the diffractogram in
Fig. S3a, ESI†) is also formed on the cathode surface. This
may be due to an excess concentration of Cu ions in bulk
solutions reduced back to Cu on the cathode surface, as shown
in eqn (11).
CuLa
2
and CuSt
2
. Fig. S2 (ESI†) shows the XPS spectra for wide
, whereas the signals obtained in
2
+
and narrow scans of CuLa
2
C 1s, O 1s and Cu 2p narrow scan XPS spectra for CuLa and
2
CuSt2 are summarized in Table S3 (ESI†). The synthesized
complex contains carbon, oxygen and copper (Fig. S2a, ESI†).
The percentage content of carbon is the highest compared to
those of other elements due to the alkyl chains of fatty acid
2+
ꢀ
Cu + 2e - Cu (at cathode)
(11)
The diffractograms of CuLa and CuSt in Fig. S3b (ESI†)
2
2
ligands. The carbon signal at 284.5 eV binding energy in narrow show sharp and intense diffraction peaks at low angles (2y o 201),
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
whereas the CuO diffraction peaks (Fig. S3a, ESI†) appear at high active ligands show better cytotoxicity compared to a similar
1
7–19
angle (2y 4 301). This proves that the synthesized CuLa
2
and CuSt
2
free fatty acid.
This is due to the property of the ligand used
are free from CuO contamination. Another by-product that may be and type of donor atoms in the ligand, which have significant
formed during the synthesis of CuLa and CuSt , as shown in effects on biological and chemical properties of the synthesized
2
2
34
35
eqn (9), is Cu(OH) . According to Zhang et al., the FTIR spectrum compound. This study proves that the synthesized CuLa and
2
2
ꢀ
1
of Cu(OH)
2
exhibits intense peaks at 3572 and 3315 cm , which CuSt
2
show better cytotoxicity than the selected control compounds.
correspond to the hydroxyl ions. The broad peaks at 3426 and
Fig. 2 shows that lower IC50 concentrations of CuLa
2
on
ꢀ
1
1
632 cm are assigned to O-H stretching and bending modes selected tumor cell lines were obtained compared to that for
2
3
of water, respectively. The specific peaks mentioned above do CuSt
2
. For example, the IC50 concentration for CuLa
cells is lower at 19.50 mM compared to the IC50 concentration
which confirms that the synthesized compound is free from for CuSt of 40.74 mM on the same cell line. This may be due to
2
on A549
8
2 2
not appear in the FTIR spectra of CuLa (Fig. 2a) and CuSt ,
2
Cu(OH) contamination.
the difference in the alkyl chain lengths of the ligands used,
where HLa and HSt ligands in CuLa and CuSt have C12 and
2
2
2
Cytotoxicity and anti-proliferation assays of electrochemically
synthesized CuLa and CuSt
C18 alkyl chain lengths, respectively. The increase in the alkyl chain
length in a compound can reduce the permeability of cell
2
2
36
The synthesized CuLa
2
2
and CuSt from the previous experiment membranes. Longer ligand alkyl chains are less permeable to cell
were used in a cytotoxicity study against A549 and HeLa cells membranes than shorter ligand alkyl chains. This is due to the
using a Presto Blue assay. The aim of this assay was to determine strong hydrophobic properties of fatty acid ligands with longer alkyl
36
the ability of synthesized compounds in inhibiting tumor cell chains because of the strong van der Waals forces along the chain.
growth and inducing tumor cell death. Fig. 2 shows the dose This further restricts the movement of the compound to pass
response curve for electrochemically synthesized CuLa and through the plasma membrane, thereby causing the compound to
CuSt as well as free HLa and HSt as control experiments on be less toxic to the cells studied. This shows that the length of the
A549 and HeLa cells after 72 hour of treatment. By referring to alkyl chain also affects the nature of compound cytotoxicity.
the dose response curve in Fig. 2, we can infer that the inhibition Antitumor compounds that provide lower IC50 values in the
of tumor cells treated by CuLa , CuSt
2
2
2
2
and other compounds micromolar to submicromolar (o10 mM) range are categorized
is dependent on the concentration of the compounds. IC50 as effective antitumor compounds, whereas compounds with
9
concentrations (Fig. 2) for CuLa and CuSt on both selected IC50 4 10 mM are classified as moderate antitumor compounds.
2
2
tumor cell lines are lower than that of the similar free fatty acid Since the IC50 concentrations obtained in this study for both
used (HLa and HSt). Inorganic complexes of metal ions and CuLa and CuSt on A549 and HeLa cells are in the range from
2
2
2 2
Fig. 2 Dose response curves and IC50 concentrations for (a) CuLa , (b) CuSt , (c) HLa, and (d) HSt on A549 and HeLa cells after 72 hour treatment.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
Fig. 3 Anti-proliferative activities for (a and b) A549 and IMR90 cells, (c and d) HeLa and IMR90 cells at different incubation times upon treatment with
IC50 concentrations of (a and c) CuLa and (b and d) CuSt
2
2
.
1
9.50 to 44.67 mM, we can conclude that the tested compounds
IC50 concentration obtained in Fig. 2 was then used in the
anti-proliferative activity to determine the effect of CuLa and
have moderate cytotoxic effects on selected tumor cells.
2
2 2
Fig. 4 Morphology changes of (a) A549 and IMR90 cells, (b) HeLa and IMR90 cells upon treatment with IC50 concentrations of CuLa and CuSt at
different incubation times. (1) Round-shaped cells are mostly floating; (2) protrusion of the plasma membrane; (3) shrinking cells; (4) cell lysis. Control
micrograph represents the untreated cells. Magnification = 20ꢁ.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
CuSt
NJC
ꢀ
2
on proliferation of A549 and HeLa cells within 168 hour the presence of bonding between the coordinated RCOO groups
2+
of the incubation time. The similar IC50 concentration for of fatty acids and Cu ions. The electrochemically synthesized
CuLa and CuSt was also tested on normal human fibroblasts CuLa and CuSt were used in the cytotoxicity study on selected
2
2
2
2
cells (IMR90). The main purpose was to study the proliferation A549 and HeLa cells. After 72 hour treatment of both compounds
of normal cells upon treatment with IC50 concentrations of on selected cells, the IC50 concentrations obtained were in the
CuLa
2 2
and CuSt . The results obtained are presented in Fig. 3. range from 19.50 to 44.67 mM. In the anti-proliferation study, less
Good antitumor compounds only inhibit the growth of tumor than 25% of the selected cells were still alive after 168 hours of
9
–11
cells without killing normal cells.
treatment with IC50 concentrations of both compounds. Mean-
The anti-proliferation graph of A549 and HeLa cells (Fig. 3) while, more than 60% of normal cells (IMR90 cells) were still alive
shows that the percentage of viable cells was reduced to less after being treated with similar IC50 concentrations. This shows
than 25% after being treated with IC50 concentrations of CuLa2 that the synthesized compounds induce cell death in selected
and CuSt after 168 hour of incubation time. This proved that cells without harming the non-cancerous IMR90 cells.
2
the complexes are toxic and can inhibit the growth of selected
tumor cells. With respect to anti-proliferative effects on IMR90
cells (Fig. 3), we found that more than 60% of IMR90 cells were
Conflicts of interest
still alive after being treated with IC50 concentrations of CuLa
and CuSt . These findings demonstrated that the tested com-
pounds are less toxic to the selected normal cells.
2
There are no conflicts to declare.
2
To evaluate morphological cell changes upon treatment with Acknowledgements
IC50 concentrations of CuLa and CuSt , the cells were observed
at different incubation times. The micrographs obtained are
presented in Fig. 4. The untreated tumor and normal cells
2
2
The funding from Universiti Sains Malaysia through grants
04/PKIMIA/6315104 is gratefully acknowledged. We wish to
express special thanks to Centre for Research & Instrumentation
Management (CRIM) UKM and Advanced Medical and Dental
Institute (AMDI) USM for providing research facilities.
3
(control) (Fig. 4a and b) appeared robust and adherent and
showed good cell density, suggesting normal proliferation. On
the other hand, after 168 hour treatment with IC50 concentrations
2 2
of CuLa and CuSt on A549 and HeLa cells, the morphology of
cells was altered compared to that of the control cells. The cells
progressively lost the characteristics of cell morphology and
changed to round-shaped and exhibited smaller size; moreover,
they also shrunk. For IMR90 cells, micrographs (Fig. 4) indicated
the appearance of normal and healthy IMR90 cells (control).
This further showed that the treatment of CuLa and CuSt with
2 2
IC50 concentrations did not result in toxicity to the selected
IMR90 cells.
References
1 L. Chen, H. Meng, L. Jiang and S. Wang, Chem. – Asian J.,
2011, 6, 1757.
2 S. Wang, L. Feng and L. Jiang, Adv. Mater., 2006, 18, 767.
3 R. N. Patel, D. K. Patel, K. K. Shukla and Y. Singh, J. Coord.
Chem., 2013, 66, 4131.
4 M. Iqbal, S. Ali and M. N. Tahir, J. Coord. Chem., 2018, 71, 991.
5
M. Ramos Riesco, F. J. Martınez Casado, S. Lopez-Andres,
M. V. Garcıa Perez, M. I. Redondo Yelamos, M. R. Torres,
L. Garrido and J. A. Rodrıguez Cheda, Cryst. Growth Des.,
Conclusions
2008, 8, 2547.
CuLa
2
and CuSt
2
have been successfully synthesized using an
6 K. Wojciechowski, A. Bitner, G. Bernardinelli and M. Brynda,
Dalton Trans., 2009, 1114.
7 M. Pajtasova, D. Ondrusova, E. Jona, S. C. Mojumdar,
S. L’alikova, T. Bazylakova and M. Gregor, J. Therm. Anal.
Calorim., 2010, 100, 769.
2
+
electrochemical technique based on the release of Cu ions
from electrochemical oxidation of a Cu anode with a graphite
rod cathode in the presence of 0.1 M fatty acid solution and an
aqueous solution of 0.1 M CH COONH as the supporting
3
4
electrolyte. The optimum electrolysis conditions for the synthesis
of CuLa2 and CuSt2 have been determined to maximize the
percentage yield and minimize energy consumption and loss of
Cu anode. The electrolysis conditions studied are type of
supporting electrolytes, applied voltages, electrolysis time and
8 N. Nordin, S. Z. Hasan, Z. Zakaria, N. A. Baharuddin,
W. Z. Samad, M. R. Yusop and M. R. Othman, J. Braz. Chem.
Soc., 2015, 26, 1115.
9 C. Marzano, M. Pellei, F. Tisato and C. Satini, Adv. Anticancer
Agents Med. Chem., 2009, 9, 185.
supporting electrolyte concentrations. This study showed that 10 J. Lu, Q. Sun, J. L. Li, L. Jiang, W. Gu, X. Liu, J. L. Tian and
both compounds (99.21%) were produced with lower energy
S. P. Yan, J. Inorg. Biochem., 2014, 137, 46.
11 X. Y. Qin, L. C. Yang, F. L. Le, Q. Q. Yu, D. D. Sun, Y. N. Liu
and J. Liu, Dalton Trans., 2013, 42, 14681.
ꢀ1
ꢀ1
consumption (B21.01 W h L ) and anode loss (B0.57 mg L
)
by using the same optimum conditions of 10 V applied voltage
for 4 hours of electrolysis time in 0.1 M CH COONH electrolyte 12 Z. Boulsourani, G. D. Geromichalos, K. Repana, E. Yiannaki,
3
4
solution. Characterization using FTIR, UV-Vis and XPS spectro-
scopy proved that the synthesized compounds were the desired
CuLa and CuSt . The results from spectroscopic studies confirmed
2 2
V. Psycharis, C. P. Raptopoulou, D. Hadjipavlou-Litina,
E. Pontiki and C. Dendrinou-Samara, J. Inorg. Biochem.,
2011, 105, 839.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
1
3 F. Bacher, O. Domotor, A. Chugunova, N. V. Nagy, L. Filipovic, 24 A. L. Yuvaraj and D. Santhanaraj, Mater. Res., 2014, 17, 83.
S. Radulovic, E. A. Enyedy and V. B. Arion, Dalton Trans., 2015, 25 A. Sakalis, K. Fytianos, U. Nickel and A. Voulgaropoulos,
4
4, 9071.
4 C. H. Ng, S. M. Kong, Y. L. Tiong, M. J. Maah, N. Sukram, 26 I. D. Dos Santos, J. C. Afonso and A. J. B. Dutraa, J. Braz.
M. Ahmad and A. S. B. Khoo, Metallomics, 2014, 6, 892. Chem. Soc., 2011, 22, 875.
5 A. N. Kate, A. A. Kumbhar, A. A. Khan, P. V. Joshi and 27 I. A. Sengil and M. Ozacar, J. Hazard. Mater., 2009, 161, 1369.
Chem. Eng. J., 2006, 119, 127.
1
1
1
1
1
1
V. G. Puranik, Bioconjugate Chem., 2014, 25, 102.
6 I. Iakovidis, I. Delimaris and S. M. Piperakis, Mol. Biol. Int.,
28 N. Nordin, W. Z. Samad, M. R. Yusop and M. R. Othman,
Mal. J. Anal. Sci., 2015, 19, 236.
29 G. B. Deacon and R. J. Phillips, Coord. Chem. Rev., 1980, 33, 227.
2011, 1.
7 W. J. Guo, S. S. Ye, N. Cao, J. Huang, J. Gao and Q. Y. Chen, 30 N. Nordin, W. Z. Samad, E. Kardia, B. H. Yahaya,
Exp. Toxicol. Pathol., 2010, 62, 577. M. R. Yusop and M. R. Othman, NANO, 2018, 13, 1.
8 I. Ali, W. A. Wani, K. Saleem and M. F. Hsieh, Chem. Pap., 31 Y. Avila-Torres, L. Huerta and N. Barba-Behrens, J. Chem.,
014, 68, 540. 2013, 1.
9 J. Nagaj, P. Kolkowska, A. Bykowska, U. K. Komarnicka, 32 L. Q. Xiao, X. X. Jia, L. Q. Liao and L. J. Liu, Chem. Rapid
2
A. Kyziol and M. Jezowska-Bojczuk, Med. Chem. Res., 2015,
4, 115.
0 G. C. Sparagna, D. L. Hickson-Bick, L. M. Buja and
Commun., 2014, 2, 6.
33 G. Q. Yuan, H. F. Jiang, C. Lin and S. J. Liao, J. Cryst. Growth,
2007, 303, 400.
2
2
2
2
2
J. B. McMillin, Antioxid. Redox Signaling, 2001, 3, 71.
1 S. Hardy, Y. Langelier and M. Prentki, Cancer Res., 2000,
34 Y. X. Zhang, M. Huang, F. Li and Z. Q. Wen, Int. J. Electrochem.
Sci., 2013, 8, 8645.
35 M. E. Davis, Z. Chen and D. M. Shin, Nat. Rev. Drug
Discovery, 2008, 7, 771.
36 S. Duangjit, B. Pamornpathomkul, P. Opanasopit, T. Rojanarata,
Y. Obata, K. Takayama and T. Ngawhirunpat, Int. J. Nanomed.,
2014, 2005.
60, 6353.
2 D. Penzo, C. Tagliapietra, R. Colonna, V. Petronilli and
P. Bernardi, Biochim. Biophys. Acta, 2002, 1555, 160.
3 T. Bieszczad and S. Sanak-Rydlewska, Physicochem. Probl.
Miner. Process., 2001, 35, 181.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.