Self-emulsifying drug delivery systems comprising chlorhexidine and alkyl-EDTA: A novel approach for augmented antimicrobial activity

Article abstract of DOI:10.1016/j.molliq.2019.111649

The aim of the study was to develop a self-emulsifying drug delivery systems (SEDDS) comprising chlorhexidine (CX) and monododecylamide-EDTA (alkyl-EDTA) exhibiting enhanced microbicidal properties. SEDDS containing 20% Captex 300, 45% Cremophor EL, 17% Tween 80 and 18% DMSO were loaded with 3% (m/v) alkyl-EDTA (F_A) and formulations of 1% CX and 1.5% (m/v) alkyl-EDTA (F_A-CX_1% and F_A-ED_1.5%) were prepared separately as well as in combination (F_A-CX_1%ED_1.5%). Resazurin assay was performed to assess the biocompatibility of SEDDS in concentrations of 0.5% and 1%. Alkyl-EDTA SEDDS were also evaluated for their binding affinity towards Ca²⁺ and Mg²⁺ as well as for augmented antimicrobial properties using E. coli as model germ. Biocompatibility assay of SEDDS revealed that >85% of cells remained viable after 4 h. SEDDS F_A efficiently bound 770 ± 19 μmol/g of calcium and 784 ± 16 μmol/g of magnesium, respectively. SEDDS F_A-CX_1%ED_1.5% showed a 34.3- and 12.9-fold enhanced antimicrobial effect compared to F_A-ED_1.5% and FA-CX_1%, respectively. The combination of CX with alkyl-EDTA in SEDDS seems therefore to be a promising strategy to enhance the antimicrobial activity of this antiseptic drug.

Full text of DOI:10.1016/j.molliq.2019.111649

Journal of Molecular Liquids 295 (2019) 111649
Contents lists available at ScienceDirect
Journal of Molecular Liquids
journal homepage: www.elsevier.com/locate/molliq
Self-emulsifying drug delivery systems comprising chlorhexidine and
alkyl-EDTA: A novel approach for augmented antimicrobial activity
Aamir Jalil a, Mulazim Hussain Asim a,c, Zeynep Burcu Akkus a, Martin Schoenthaler b,
Barbara Matuszczak b, Andreas Bernkop-Schnürch a,
⁎
a
Center for Chemistry and Biomedicine, Department of Pharmaceutical Technology, Institute of Pharmacy, University of Innsbruck, Innrain 80/82, 6020 Innsbruck, Austria
Center for Chemistry and Biomedicine, Department of Pharmaceutical Chemistry, Institute of Pharmacy, University of Innsbruck, Innrain 80/82, 6020 Innsbruck, Austria
Department of Pharmaceutics, Faculty of Pharmacy, University of Sargodha, 40100 Sargodha, Pakistan
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 July 2019
Received in revised form 23 August 2019
Accepted 26 August 2019
Available online 27 August 2019
The aim of the study was to develop a self-emulsifying drug delivery systems (SEDDS) comprising chlorhexidine
(CX) and monododecylamide-EDTA (alkyl-EDTA) exhibiting enhanced microbicidal properties. SEDDS contain-
ing 20% Captex 300, 45% Cremophor EL, 17% Tween 80 and 18% DMSO were loaded with 3% (m/v) alkyl-EDTA
(FA) and formulations of 1% CX and 1.5% (m/v) alkyl-EDTA (FA-CX1% and FA-ED1.5%) were prepared separately
as well as in combination (FA-CX1%ED1.5%). Resazurin assay was performed to assess the biocompatibility of
SEDDS in concentrations of 0.5% and 1%. Alkyl-EDTA SEDDS were also evaluated for their binding affinity towards
Ca2+ and Mg2+ as well as for augmented antimicrobial properties using E. coli as model germ. Biocompatibility
Keywords:
SEEDS
assay of SEDDS revealed that N85% of cells remained viable after 4 h. SEDDS FA efficiently bound 770
19
Chlorhexidine
Alkyl-EDTA
Chelation
Antimicrobial
E. coli
μmol/g of calcium and 784 16 μmol/g of magnesium, respectively. SEDDS FA-CX1%ED1.5% showed a 34.3- and
12.9-fold enhanced antimicrobial effect compared to FA-ED1.5% and FA-CX1%, respectively. The combination of
CX with alkyl-EDTA in SEDDS seems therefore to be a promising strategy to enhance the antimicrobial activity
of this antiseptic drug.
© 2019 Elsevier B.V. All rights reserved.
1. Introduction
in combination with potentiating agents for enhanced efficacy against
pathogenic microbes [9]. The likely most prominent example for such
Antimicrobial agents were considered ‘wonder drugs’ of the last cen-
tury revolutionizing the treatment of microbial infections [1]. Over the
last decades, however, pathogenic bacteria developed resistance caus-
ing a global crisis narrowing the choices for the treatment of microbial
infections [2]. The slow and intermittent state of innovating new anti-
microbial agents has additionally worsened the situation. Moreover,
pharmaceutical industry is not eagerly interested in the discovery and
development of new antimicrobial agents for economic reasons [3,4].
In comparison to antiseptics, antibiotics are more prone to resistance
because of their irrational and intensive use. An imperative approach
to slow down microbial resistance is therefore the use of fewer antibi-
otics and more antiseptics wherever applicable [5,6].
‘potentiators’ for antiseptics is EDTA [10]. As a chelating agent it exhibits
high affinity towards metal ions with two amino and four carboxylate
binding sites [11]. Because of this property it affects permeability of
the membrane of microbes by binding Ca2+ and Mg2+ resulting in de-
stabilization. This mechanism of action causes an augmented effect
when used together with antiseptics such as CX by enhancing their pen-
etration into microbial cells [12]. Such combinations are likely useful in
preventing the development of resistant strains by providing broad-
spectrum coverage and improving therapeutic efficacy [13]. The cat-
ionic CX and the anionic EDTA, however, are incompatible in aqueous
media forming insoluble and consequently inactive complexes [14,15].
A way out of this dilemma might be the design of lipophilic formulations
instead of aqueous formulations allowing the combined use of CX and
EDTA without these shortcomings. Among such lipophilic formulations,
self-emulsifying drug delivery systems (SEDDS) seem to be advanta-
geous as they are capable of forming spontaneously emulsions of large
interfacial surface area in aqueous media [16,17]. Moreover, inside the
oily droplets of SEDDS CX remain non-ionized exhibiting likely im-
proved penetrating properties into microbial cells.
In parallel, new strategies have been explored focusing on the addi-
tion of potentiating and sensitizing agents that synergistically work
with antiseptics to enhance their bactericidal or bacteriostatic efficacy
[7,8]. Broad spectrum antiseptics like chlorhexidine (CX) were studied
⁎
Corresponding author at: Centre for Chemistry and Biomedicine, Department of
Pharmaceutical Technology, Institute of Pharmacy, University of Innsbruck, Innrain 80/
82, 6020 Innsbruck, Austria.
It was therefore the aim of this study to develop SEDDS comprising
CX and EDTA in order to combine the advantages of both compounds
E-mail address: andreas.bernkop@uibk.ac.at (A. Bernkop-Schnürch).
https://doi.org/10.1016/j.molliq.2019.111649
0167-7322/© 2019 Elsevier B.V. All rights reserved.

2
A. Jalil et al. / Journal of Molecular Liquids 295 (2019) 111649
for an enhanced antimicrobial effect. In order to achieve this goal, the
hydrophobic complexing agent ethylenediaminetetraacetic acid dode-
cyl amide (alkyl-EDTA) was synthesized and characterized by 1H NMR
and LC-MS for confirmation of structure and mass, respectively. The
combination of CX and alkyl-EDTA was incorporated into SEDDS and
characterized for size and binding affinity towards calcium and magne-
sium. Moreover, antimicrobial activity of SEDDS containing CX and
alkyl-EDTA was evaluated using E. coli as representative bacterial strain.
2.3. 1H NMR spectroscopy
The 1H NMR of the monoanhydride-EDTA and alkyl-EDTA was re-
corded on Varian Bruker NMR spectrometer (400 MHz). DMSO-d6 was
used as solvent for collecting 1H NMR spectra of all samples. The
resulting chemical shifts were stated in parts per million (ppm) and
tetramethylsilane (TMS) was used as the internal standard.
2.4. Molecular mass determination by LC-MS
2. Materials and methods
Molecular mass of alkyl-EDTA was determined by using Chromaster
5610 MS detector (Hitachi) controlled by MSD system manager soft-
ware (version 2.1). The conditions provided for measuring the molecu-
lar mass of alkyl-EDTA were as follows: ionization potential (V): 2000;
ion injection (eV): 2.0; counter gas flow (L/min): 1.0; AIF temperature
(°C): 140 and ion source temperature (°C): 80. Sample solutions were
prepared by dissolving alkyl-EDTA in methanol as mobile phase to
reach a final concentration of 500 ng/mL. In the mass range of m/z 450
→ 470 negative electrospray ionization (ESI) mode was run to calculate
molecular mass of alkyl-EDTA. The sample was directly infused using a
syringe pump with flow rate of 2 μL min−1 in order to obtain a clear
mass spectrum without any background noise.
2.1. Materials
Ethylenediaminetetraacetic dianhydride, ethylenediaminetetraace-
tic acid (EDTA), lysogeny broth medium (LB medium), nutrient agar
medium, N,N-dimethyl formamide (DMF), diethyl ether, dodecylamine,
calcium chloride (CaCl2), magnesium chloride (MgCl2), chlorhexidine
(CX), TLC plate, Captex-300, PEG 400, dimethyl sulfoxide (DMSO),
Tween 80 and Miglyol 840, Cremophor EL and sodium hydroxide
(NaOH) were purchased form Sigma-Aldrich (Vienna, Austria), chlor-
hexidine digluconate (CHG) 20% (w/v) from Santa Cruz Biotechnology
Incorporation (Texas, USA).
2.5. SEDDS development and characterization
2.2. Synthesis of ethylenediaminetetraacetic acid mono-dodecylamide (al-
kyl-EDTA)
For SEDDS preparation, Captex 300 and Miglyol 480 were used as oil
while Cremophor EL and Tween 80 were employed as surfactants and
polyethylene glycol as well as dimethyl sulfoxide (DMSO) served as
co-solvents. The components were homogenized in ratios as listed in
Table 1 via mixing at 1200 rpm at 40 °C using a thermomixer
(Thermomixer comfort, Eppendorf, Germany). These SEDDS pre-
concentrates were diluted in 1:100 in 50 mM phosphate buffer pH 7.4
by vortexing. Furthermore, mean droplet size, polydispersity index
and zeta potential were determined by dynamic light scattering using
a Zetasizer Nano ZSP (Malvern Instruments, Worcestershire, UK).
Monoalkylamide-EDTA was synthesized in two steps. In first step,
hydrolysis of EDTA dianhydride was performed to form monoanhydride
ethylenediaminetetraacetic acid (monoanhydride-EDTA). For this pur-
pose, 2.7 mmol of EDTA dianhydride was dissolved in dry DMF (5 mL)
at 80 °C in a 15 mL three neck flask equipped with a reflux condenser
under argon environment and constant stirring. Water (50 μg,
2.5 mmol) was diluted in 1 mL of dry DMF and added dropwise in
course of 50 min into the reaction mixture. The reaction mixture was
stirred for 2 h at 80 °C and afterwards stirred at room temperature for
12 h. The precipitate was collected through filtration, washed with dry
DMF and dried under vacuum for 24 h at 70 °C.
In the second step monoanhydride-EDTA (338 mg, 1.2 mmol) and
dodecylamine (220 mg, 1.2 mmol) were dissolved in dry DMF
(15 mL) in a 50 mL three necked round-bottomed flask equipped
with reflux condenser. The mixture was stirred for 10 h at 100 °C
under inert environment in order to protect the reaction from atmo-
spheric moisture and oxygen. The reaction mixture was allowed to
cool down to room temperature and transferred to 100 mL round bot-
tom flask. Afterwards, 45 mL of water was added to precipitate the tar-
get compound. The precipitate was collected through centrifugation
and washed 2–3 times with water. Afterwards precipitate was dis-
solved in 1 M aqueous sodium hydroxide (12 mL) at pH 10 and added
to diethyl ether (10 mL) under shaking. The ethereal phase containing
unreacted alkyl amine was removed. 2 mL of 1 M HCl solution was
added to neutralize the water phase and heated at 80 °C to remove re-
maining traces of ether. In order to precipitate the target compound, pH
of the solution was adjusted to 4 by dropwise addition of 1 M HCl. The
precipitate was filtered, washed with demineralized water and dried
for 48 h at 80 °C in vacuum [18]. For identification of target compound,
solutions of the reactants EDTA, dodecyl amine and purified product
monododecylamide-EDTA (alkyl-EDTA) were prepared for spot appli-
cation on TLC plate. After spot applications chromatographic plates
were developed using mixture of methanol, acetic acid and ethyl ace-
tate as the eluent in a ratio of 3:1:2, respectively. The plate was then in-
cubated in iodine chamber at 90°for 5 min in order to stain the
separated compounds. Another plate having been developed in the
same way was sprayed with ninhydrin reagent in order to detect pri-
mary amines.
2.6. Determination of distribution coefficient (Log D)
The distribution coefficient (Log D) of CX between SEDDS pre-
concentrate and 0.1 M PBS pH 6.8 as release medium was determined
by quantifying the maximum solubility of CX in SEDDS pre-
concentrate and in the release medium [19]. Log D was calculated by
using following equation:
SSEDDS
Log D ¼ log
SRM
whereSSEDDS = maximum solubility in SEDDS pre-concentrateSRM
maximum solubility in the release medium
=
2.7. Cytotoxicity assay
Cytotoxicity of SEDDS formulations as listed in Table 1 was investi-
gated in concentrations of 0.5 and 1% (v/v) by resazurin reduction
assay on Caco-2 cell-line. Cells were cultured in 24-well plates at a con-
centration of 2.5 × 104 cells per well. Minimum Essential Medium
(MEM) was used to grow culture of the Caco-2 cells at 37 °C maintaining
95% humidity and 5% CO2 atmosphere for 14 days [20]. MEM was
employed as a negative control and Triton X100 (1% w/v) as a positive
control, respectively. SEDDS were dispersed in MEM without phenol
red in order to obtain a final concentration of 0.5% and 1%. 500 μL of
emulsified SEDDS were added to the cell layer and incubated for 4 h.
After washing cells with 500 μL of 10 mM phosphate buffered saline
pH 7.4, 500 μL of resazurin solution was used to incubate the cells for
3 h. Fluorescence of supernatant was analyzed using an excitation
wavelength of 540 nm and an emission wavelength of 590 nm. Cell

A. Jalil et al. / Journal of Molecular Liquids 295 (2019) 111649
3
viability was calculated as percentage of viable cells considering viabil-
ity of cells 100% treated with MEM (negative control) using following
equation [21].
Experimental values−Negative control
Positive control−Negative control
Viable cell ð%Þ ¼
X 100
2.8. Chelation studies with alkyl-EDTA and alky-EDTA SEDDS
2.8.1. Metal binding capacity studies with alkyl-EDTA
EDTA and alkyl-EDTA were evaluated for their binding capacity to-
wards divalent metal ions by complexometric titration. In this study cal-
cium and magnesium chelation capacity was determined according to a
method previously described by our research group with some modifi-
cations [22]. For this purpose, ammonia buffer was prepared by dissolv-
ing 3.50 g of ammonium chloride in 28.5 mL of ammonia solution
(35.6%) and diluted with demineralized water to a final volume of
50 mL. 100 mM solutions of EDTA and alkyl-EDTA were prepared in am-
monia buffer at pH 10.5. Solution of Eriochrome Black T (EBT) was pre-
pared by dissolving 20 mg in 1.5 mL of ammonia solution and 500 μL of
ethanol. Briefly, 10 μL of EBT solution was added to EDTA and alkyl-
EDTA solutions and titrated with 100 mM solution of calcium chloride
as well as magnesium chloride till change of color.
2.8.2. Metal binding capacity studies with alkyl-EDTA SEDDS
SEDDS loaded with alkyl-EDTA, were evaluated for their capacity to
bind calcium and magnesium ions. For this purpose, FA and FB as listed
in Table 1 were diluted from 1:10 up to 1:50 in ammonia buffer
pH 10.5. Afterwards, 5 μL of EBT solution was added to 500 μL of
SEDDS dilutions and titrated with 10 mM calcium chloride as well as
magnesium chloride till change of color indicated the end point.
2.9. Time dependent antimicrobial activity of SEDDS and CHG
In this study, the antimicrobial activity of CX loaded SEDDS and
chlorhexidine digluconate (CHG) was compared over a time period of
8 h. For this purpose, Gram negative E. coli was used as model germ in
LB medium with
a bacterial density of approximately 3.0
× 108 CFU/mL (1 McFarland). Growth medium was prepared by homog-
enizing 250 μL of the 1 McFarland bacterial suspension with 250 μL of LB
medium/water mixture (1:1) resulting in a final bacterial density of ap-
proximately 1.5 × 108 CFU/mL. On the one hand, SEDDS FA-CX1% were
diluted 1:40 and on the other hand 111 μg of free CHG were dissolved
in 500 μL of growth medium to obtain CX in a final concentration of
125 μg/mL. SEDDS without CX and growth medium without CHG
were used as control. LB medium without drug and bacteria was taken
as blank. Samples were placed in thermomixer at 600 rpm, 37 °C. Ali-
quots (100 μL) were transferred to 96 well plate at predetermined
time points and measured at 600 nm using microplate reader for optical
density. Bacterial population in LB medium without drug was taken as
100% at each time point.
2.10. Growth inhibition and comparative antimicrobial studies of SEDDS
Firstly, growth inhibition studies with alkyl-EDTA loaded SEDDS (FA
and FB) were carried out using E. coli as model germ. E. coli grown over-
night in LB medium were taken and diluted with sterile water to obtain
a bacterial concentration of 0.5 McFarland. 100 μL of this bacterial sus-
pension was diluted with 1.9 mL of LB medium to prepare inoculum.
SEDDS with and without alkyl-EDTA were diluted in different ratios
ranging from 1:10 to 1:100 in 450 μL of growth medium (50% LB me-
dium) and 50 μL of diluted bacterial suspension was added. Secondly,
for comparative antimicrobial studies, 1:20, 1:30 and 1:40 dilution of
SEDDS FA-CX1%, FA-ED1.5% and FA-CX1%ED1.5% in a growth medium and
50 μL of 0.5 McFarland bacterial suspensions without dilution was

4
A. Jalil et al. / Journal of Molecular Liquids 295 (2019) 111649
added. Diluted SEDDS without CX and alkyl-EDTA served as positive
control whereas solutions of CX and alkyl-EDTA in growth medium
were used as negative control. Afterwards, samples were incubated
using thermomixer at 37 °C, 600 rpm for 20 h. Aliquots (100 μL) from
each dilution were transferred to 96 well plate and measured at
600 nm for optical density using microplate reader. 500 μL of growth
medium without SEDDS containing inoculum was taken as 100% for cal-
culating percentage growth while growth medium without E. coli
served as blank [23].
of EDTA monohydride and dianhydride affecting both yield and purity
of the target compound. Under optimized conditions 70% EDTA
monoanhydride was obtained in precipitated form.
In the second step, dodecyl amine was attached to EDTA
monoanhydride in order to form alkyl-EDTA. The synthesis scheme is
shown in Fig. 1. The target compound was obtained in precipitated
form and was purified from unreacted components as described
above. The yield of alkyl-EDTA was 55%. Purity of target compound
was confirmed by TLC as shown in Fig. 2. TLC 1 developed in iodine
chamber resulted in three yellow spots that were sufficiently separated
from each other regarding their RF values. TLC 2 developed with ninhy-
drin showed only one pink spot of dodecyl amine whereas no color was
spotted at point ‘b’ endorsing absence of amine in alkyl-EDTA and con-
sequently confirming its purity.
2.11. Statistical analysis
The results are compiled and expressed as means ( SD) of three ex-
periments. The data obtained from the experiments were analyzed by
using Student's t-test with 95% confidence interval (p value ≤0.05).
The 1H NMR spectra of EDTA monoanhydride and alkyl-EDTA were
taken in DMSO-d6 as solvent. The 1H NMR spectra can be seen in supple-
mentary figures. The 1H NMR of EDTA monoanhydride recorded peaks
at ppm 3.71 (4H, 2 x CH2CO), 3.42 (s, 4H, 2 x CH2CO), 2.79 (t, J =
6.2 Hz, 2H, CH2N) and 2.58 (t, J = 6.2 Hz, 2H, CH2N) as can be seen in
S.Fig. 1. The 1H NMR of alkyl-EDTA showed peaks at ppm 8.00 (t, J =
6.2 Hz, 1H, NH), 3.42 (s, 4H, 2 x CH2CO2H), 3.35 (s, 2H, CH2CO), 3.18
(s, 2H, CH2CO), 3.05 (m, 2H, CH2NH), 2.74–2.70 (m, 4H, NCH2CH2N),
1.38 (m, 2H CH2), 1.22 (s, 18H, (CH2)9) and 0.84 (t, J = 6.6 Hz, 3H,
CH3) as illustrated in S.Fig. 2.
3. Results and discussion
3.1. Synthesis and characterization of alky-EDTA
Synthesis of alkyl-EDTA from EDTA dianhydride led to the formation
of mono- and di-alkylamide-EDTA that were difficult to separate from
each other. In order to sort out this problem EDTA monoanhydride
was formed from EDTA dianhydride by the addition of water. Too fast
addition of water led to the formation of EDTA instead of EDTA
monoanhydride while too less and slow addition resulted in a mixture
Furthermore, negative ESI mode under optimized conditions using
liquid chromatography-mass spectrometry provided a selective method
Fig. 1. Synthetic pathway of N-[2-[bis(carboxymethyl)amino]ethyl]-N-[2-(dodecylamino)-2-oxoethyl]-glycine (monododecylamide-EDTA).

A. Jalil et al. / Journal of Molecular Liquids 295 (2019) 111649
5
SEDDS was 2% (m/v) and 3% (m/v), respectively. The calculated Log D
of CX was 2.9 [19]. In order to evaluate their homogeneity
SEDDS/RM
and stability, all SEDDS pre-concentrates were emulsified in 10 mM
phosphate buffer as well as in MEM and measured for droplet size and
polydispersity index (PI) as shown in Table 1.
As cytotoxicity assays with SEDDS are frequently performed on
Caco-2 cells, this cell line was chosen for comparison reasons. As
depicted in Fig. 4 increasing concentrations of SEDDS from 0.5 to 1%
(v/v) did not reduce cell viability below 85% over a period of 4 h. In par-
ticular, 1% (v/v) SEDDS FA-CX1% containing 100 μg/mL of CX also showed
no toxic effect on cell viability suggesting that insertion of CX into
SEDDS droplets eliminates the toxic effect derived from cationic charge
of naked CX. In addition, pronounced changes in viability of cells be-
tween 0.5% and 1% (v/v) of SEDDS formulations have not been observed.
3.3. Chelation studies
3.3.1. Chelation studies of alkyl-EDTA
Experimental data showed that the ability of alkyl-EDTA to com-
plex calcium and magnesium is lower than that of EDTA. The capacity
of EDTA and alkyl-EDTA to bind calcium and magnesium is shown in
Fig. 5(a). EDTA bound 3409 70 μmol of calcium per g and 3450
36 μmol of magnesium per g whereas alkyl-EDTA bound only 1958
24 μmol of calcium per g and 1916
40 μmol of magnesium per
g. The likely reason for a lower binding capacity is the amide bond
formation at one of the carboxyl groups being not anymore available
Fig. 2. Elution of (a) EDTA (b) alkyl-EDTA and (c) dodecylamine on thin layer
chromatography using methanol:ethylacetate:acetic acid in 3:1:2 as mobile phase. TLC 1
was developed in iodine chamber while TLC 2 was developed by spraying ninhydrin to
detect amino groups.
for determination of molecular mass of alkyl- EDTA. To obtain most ac-
curate mass spectrometry conditions alkyl-EDTA was injected using di-
rect infusion through a syringe pump. The spectrum of alkyl-EDTA
showed base peak ions at mass-to-charge ratio (m/z) at 460.3 Da as
shown in Fig. 3 confirming the chemical structure of the target
compound.
3.2. SEDDS development and basic characterization
Lipids, surfactant and co-solvents were investigated for their emulsi-
fying properties as well as ability to form SEDDS containing CX and
alkyl-EDTA. DMSO was identified as most appropriate solvent for both
CX and alkyl-EDTA. The most suitable pre-concentrates are shown in
Table 1. Cremophor EL and Tween 80 were chosen because of their
high HLB value as well as high emulsifying properties [24]. The maxi-
mum concentration of CX and alkyl-EDTA that can be loaded into
Fig. 4. Biocompatibility assay performed with indicated SEDDS formulations in
concentrations of 0.5% (black bars) and 1% (white bars) obtained after an incubation
period of 4 h. Indicated values are means SD of three experiments.
Fig. 3. Molecular mass determination of alkyl-EDTA using methanol as mobile phase via
liquid chromatography mass spectroscopy technique.

6
A. Jalil et al. / Journal of Molecular Liquids 295 (2019) 111649
Fig. 5. (a) Metal binding capacity per g of EDTA (black bars) and alkyl-EDTA (white bars) for calcium chloride and magnesium chloride using eriochrome black T as indicator. Indicated
values are means SD of three experiments. (b) Complex formation of alkyl-EDTA SEDDS formulation FA-ED3% and FB-ED3% with calcium chloride and magnesium chloride using
different SEDDS dilution 1:10 to 1:50 employing eriochrome black T as indicator. SEDDS without alkyl-EDTA were used as control. All values are mean SD of three experimental value.
for metal ion binding. Moreover, the 12 carbon alkyl chain intro-
duces a lipophilic character providing likely a steric hindrance for
metal ions because of micelle formation. These findings suggest on
the one hand lower metal binding affinity of alkyl-EDTA but on the
other hand demonstrate the ability of alkyl-EDTA to from metal com-
plexes similar to EDTA.
3.4. Time dependent antimicrobial activity of SEDDS and CHG
In this study, the antimicrobial effect of CX in SEDDS was compared
with water soluble CHG. The results revealed that CX containing SEDDS
FA-CX1% express increased antimicrobial activity in comparison to CHG
over a time period of 8 h. The viability of bacterial cells was reduced to
one third in case of SEDDS FA-CX1% within the first 2 h while CHG
showed 59% viability as displayed in Fig. 6. After 8 h, SEDDS exhibited
a 2-fold increased antimicrobial effect in comparison to CHG. The in-
creased antimicrobial effect might be explained by lipophilic character
of SEDDS that effectively interact with the bacterial cell membrane to
enhance the interaction of CX in comparison to water soluble CHG. It
is well established that nanocarrier systems with lipophilic character
provide higher ability for cell membrane permeation compared to hy-
drophilic carriers [25]. Shazely et al. showed that solid lipid
3.3.2. Chelation studies of alkyl-EDTA SEDDS
Alkyl-EDTA SEDDS exhibited high affinity to divalent ions as shown
in Fig. 5(b). Just a minor difference in binding capacity of the two SEDDS
formulations was found. This difference might be related to the instant
release of PEG 300 in aqueous media that decreases the surface tension
bringing more alkyl-EDTA on the surface resulting in enhanced metal
binding capacity.

A. Jalil et al. / Journal of Molecular Liquids 295 (2019) 111649
7
Fig. 6. Percentage viability of E. coli over time in presence of FA-CX1% and CHG in
concentration of 125 μg/mL. All values are mean SD of three experimental values.
nanoparticles (SLN) of ciprofloxacin (CIP) exhibit superior antibacterial
activity than free CIP. The authors explained this observation with the
lipophilicity of SLN interacting more intensely with the bacterial mem-
brane in turn boosting the entrance of CIP into microbial cells [26]. Sim-
ilarly, Zaichik et al. reported that vancomycin loaded SEDDS show
higher activity against S. aureus than free vancomycin [27].
3.5. Growth inhibition and comparative antimicrobial studies
Magnesium and calcium cations are necessary for many cellular
functions including ion homeostasis, signaling pathways and morpho-
genesis in microbial cells. However, specific function of divalent cations
and their effect on cellular mechanisms are likely different depending
on each microorganism. Generally, the role of calcium is well-defined
in the synthesis of bacterial exoskeleton for bacterial growth [28]. The
bacterial cell wall is mainly composed of tubulin-like proteins that spe-
cifically require calcium in order to initiate their polymerization to gen-
erate the outer cover [29]. The absence of calcium seems to terminates
the division of bacteria eventually resulting in spherical, swelled large
vacuoles ultimately causing cell lyses [30]. Likewise, magnesium has a
vital role in enzymatic activity as well as maintaining integrity of bacte-
rial cell membrane [31]. Deprivation of magnesium causes the degrada-
tion of bacterial ribosomes halting enzymatic activity [32]. Suzuki and
Hayashi conducted experiments on magnesium starvation in E. coli
and observed massive ribosomal degeneration resulting in ceasing cell
growth making them susceptible to external environment due to in-
creased transport across the membrane [33]. The role of EDTA as a che-
lator is well-established by binding divalent cations and therefore
depriving microorganisms from essential growth factors such as cal-
cium and magnesium resulting in their growth inhibition [34]. Finch
and Brown revealed in their findings that EDTA has greater affinity for
divalent cations than the components of cell membrane resulting in
its destabilization [35]. This destabilization by EDTA was explained by
Vaara et al. in terms of change in permeability of the bacterial mem-
brane due to release of lipopolysaccharides (LPS) [36]. Further investi-
gations unclosed that change in microbial permeability is due to stress
on the lipid organization by intercalation of EDTA inducing membrane
curvature and protrusion creating space for free movement of mole-
cules across the outer membrane barrier [37]. In present study, SEDDS
formulations FA and FB containing alkyl-EDTA expressed significant in-
hibition of E. coli growth as shown in Fig. 7. The difference in growth
Fig. 7. Percentage growth of E. coli after 20 h incubation with alkyl-EDTA loaded SEDDS
representing (white bars) and unloaded SEDDS in (black bars) as control in different
ratios containing alkyl-EDTA from 1:10 (3000 μg/mL), 1:20 (1500 μg/mL), 1:30 (1000
μg/mL), 1:40 (750 μg/mL), 1:50 (600 μg/mL), 1:60 (500 μg/mL), 1:70 (428.6 μg/mL),
1:80 (375 μg/mL), 1:90 (333.4 μg/mL) and 1:100 (300 μg/mL). All values are mean SD
of three experiments.
inhibition of E. coli was in accordance with the variable concentrations
of alkyl-EDTA in SEDDS. Both formulations inhibited the growth of E.
coli in dilution range of 1:10, 1:20 and 1:30. Moreover, even higher dilu-
tions of SEDDS containing lower concentrations of alkyl-EDTA slowed
down the growth of E. coli. Furthermore, SEDDS dilution of 1:40 of
both formulations inhibited growth between 80 and 85% whereas dilu-
tions from 1:50 up to 1:100 provided growth inhibition up to 30 to 50%.
Results affirm that growth inhibition of E. coli is directly dependent on
the concentration of alkyl-EDTA in SEDDS as also previously reported
in the literature by different drug delivery systems containing EDTA.
Valenta et al. conjugated EDTA to chitosan via carbodiimide reaction
and confirmed the potential of this conjugate to inhibit growth of E.

8
A. Jalil et al. / Journal of Molecular Liquids 295 (2019) 111649
Acknowledgements
The authors are thankful and would like to extend acknowledge-
ment to Higher Education Commission of Pakistan (HEC) and the
Austrian Agency for International Cooperation in Education and Re-
search (ÖAD) for their support.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.molliq.2019.111649.
References
[1] S.B. Zaman, M.A. Hussain, R. Nye, V. Mehta, K.T. Mamun, N. Hossain, Cureus 9
(2017).
[2] K.K. Kumarasamy, M.A. Toleman, T.R. Walsh, J. Bagaria, F. Butt, R. Balakrishnan, U.
Chaudhary, M. Doumith, C.G. Giske, S. Irfan, Lancet Infect. Dis. 10 (2010) 597.
[3] A. Al-Bakri, G. Othman, Y. Bustanji, J. Appl. Microbiol. 107 (2009) 280.
[4] E. Dellera, M.C. Bonferoni, G. Sandri, S. Rossi, F. Ferrari, C. Del Fante, C. Perotti, P.
Grisoli, C. Caramella, Eur. J. Pharm. Biopharm. 88 (2014) 643.
[5] N.-O. Hübner, A. Kramer, Skin Pharmacol. Physiol. 23 (2010) 17.
[6] M. Bonferoni, G. Sandri, S. Rossi, E. Dellera, A. Invernizzi, C. Boselli, A. Cornaglia, C.
Del Fante, C. Perotti, B. Vigani, Mar. Drugs 16 (2018) 56.
[7] A. Umerska, M. Strandh, V. Cassisa, N. Matougui, M. Eveillard, P. Saulnier, Biomole-
cules 8 (2018) 122.
[8] S. Prior, B. Gander, J.M. Irache, C. Gamazo, J. Antimicrob. Chemother. 55 (2005) 1032.
[9] J.J. Dynes, J.R. Lawrence, D.R. Korber, G.D. Swerhone, G.G. Leppard, A.P. Hitchcock,
Sci. Total Environ. 369 (2006) 369.
[10] M.R. Brown, R. Richards, Nature 207 (1965) 1391.
Fig. 8. Percentage E. coli growth in presence of SEDDS dilution1:20, 1:30 and 1:40 of FA-
ED1.5% containing alkyl-EDTA (750 μg/mL, 500 μg/mL and 375 μg/mL) and FA-CX1%
having CX (500 μg/mL, 334 μg/mL and 250 μg/mL), respectively while FA-CX1%ED1.5%
contains combination of similar concentrations depicting augmented anti-microbial
effect. All values are mean SD of three experimental values.
[11] S. Finnegan, S.L. Percival, Adv. Wound Care 4 (2015) 415.
[12] R. Zhang, M. Chen, Y. Lu, X. Guo, F. Qiao, L. Wu, Sci. Rep. 5 (2015) 12944.
[13] L. Drago, E. De Vecchi, L. Nicola, M.R. Gismondo, BMC Infect. Dis. 7 (2007) 111.
[14] S. González-López, D. Camejo-Aguilar, P. Sanchez-Sanchez, V. Bolaños-Carmona, J.
Endod. 32 (2006) 781.
[15] B.J. Rasimick, M. Nekich, M.M. Hladek, B.L. Musikant, A.S. Deutsch, J. Endod. 34
(2008) 1521.
[16] R.N. Gursoy, S. Benita, Biomed. Pharmacother. 58 (2004) 173.
[17] J. Griesser, G. Hetényi, H. Kadas, F. Demarne, V. Jannin, A. Bernkop-Schnürch, Int. J.
Pharm. 538 (2018) 159.
[18] T. Takeshita, T.-A. Shimohara, S. Maeda, J. Am. Oil Chem. Soc. 59 (1982) 104.
[19] A. Bernkop-Schnürch, A. Jalil, J. Control. Release 271 (2018) 55.
[20] P. Sabaeifard, A. Abdi-Ali, M.R. Soudi, C. Gamazo, J.M. Irache, Eur. J. Pharm. Sci. 93
(2016) 392.
[21] S. Zaichik, C. Steinbring, C. Menzel, L. Knabl, D. Orth-Höller, H. Ellemunter, K.
Niedermayr, A. Bernkop-Schnürch, Int. J. Pharm. 547 (2018) 282–290.
[22] K. Netsomboon, W. Suchaoin, F. Laffleur, F. Prüfert, A. Bernkop-Schnürch, Eur. J.
Pharm. Biopharm. 111 (2017) 26.
[23] R. Wahab, S.T. Khan, S. Dwivedi, M. Ahamed, J. Musarrat, A.A. Al-Khedhairy, Colloids
Surf. B: Biointerfaces 111 (2013) 211.
[24] M.A. Rahman, A. Hussain, M.S. Hussain, M.A. Mirza, Z. Iqbal, Drug Dev. Ind. Pharm.
39 (2013) 1.
[25] S. Kalepu, M. Manthina, V. Padavala, Acta Pharm. Sin. B 3 (2013) 361.
[26] G.A. Shazly, Biomed. Res. Int. (2017) 2017.
[27] S. Zaichik, C. Steinbring, C. Caliskan, A. Bernkop-Schnürch, Int. J. Pharm. 554 (2019)
125.
[28] V. Norris, S. Grant, P. Freestone, J. Canvin, F. Sheikh, I. Toth, M. Trinei, K. Modha, R.I.
Norman, J. Bacteriol. 178 (1996) 3677.
[29] X.C. Yu, W. Margolin, EMBO J. 16 (1997) 5455.
[30] T. Onoda, J. Enokizono, H. Kaya, A. Oshima, P. Freestone, V. Norris, J. Bacteriol. 182
(2000) 1419.
[31] A. Tissieres, J.D. Watson, D. Schlessinger, B. Hollingworth, J. Mol. Biol. 1 (1959) 221.
[32] J.E. Lusk, R. Williams, E.P. Kennedy, J. Biol. Chem. 243 (1968) 2618.
[33] H. Suzuki, Y. Hayashi, Biochim. Biophys. Acta Special. Sec. Nucl. Acid Relat. Subj. 87
(1964) 610.
[34] B. Loretz, A. Bernkop-Schnürch, AAPS J. 8 (2006) E756.
[35] J. Finch, M. Brown, J. Antimicrob. Chemother. 1 (1975) 379.
[36] M. Vaara, Microbiol. Mol. Biol. Rev. 56 (1992) 395.
[37] V. Prachayasittikul, C. Isarankura-Na-Ayudhya, T. Tantimongcolwat, C.
Nantasenamat, H.-J. Galla, Acta Biochim. Biophys. Sin. 39 (2007) 901.
[38] C. Valenta, B. Christen, A. Bernkop-Schnürch, J. Pharm. Pharmacol. 50 (1998) 445.
[39] K. Ramalingam, V.A. Lee, Artificial Cells, Nanomedicine, and Biotechnology, 2018 1.
[40] S.L. Percival, P. Kite, K. Eastwood, R. Murga, J. Carr, M.J. Arduino, R.M. Donlan, Infect.
Control Hosp. Epidemiol. 26 (2005) 515.
[41] R.J. Sherertz, M.S. Boger, C.A. Collins, L. Mason, I.I. Raad, Antimicrob. Agents
Chemother. 50 (2006) 1865.
[42] L. Leive, Biochem. Biophys. Res. Commun. 18 (1965) 13.
[43] I. Stojanov, A. Milovanovic, D. Ruzic-Muslic, R. Ratajac, M.Z. Balos, N. Maksimovic, J.
Apic, Turk. J. Vet. Anim. Sci. 42 (2018) 91.
coli [38]. A similar antimicrobial strategy against pathogenic microbes
had been developed by incorporating EDTA in the aqueous phase of a
nanoemulsion that was obtained with immiscible oil phase under ex-
treme shear forces [39]. All these EDTA based drug delivery systems
have in common to halt bacterial growth by depriving bacteria from cal-
cium and magnesium resulting in a defective cellular bacterial structure.
In addition to these studies, it is reported that EDTA also increases cellu-
lar uptake of antibiotics and antiseptics. Therefore, besides growth inhi-
bition effect of EDTA, improved efficiency of antimicrobials in presence
of EDTA has been reported in various studies [40,41]. According to Leive
et al. the susceptibility of E. coli colonies towards antimicrobial agents
increases after pretreatment with EDTA [42]. Therefore combination of
CX and alkyl-EDTA in SEDDS was also studied and indeed a pronounced
improvement in antimicrobial activity shown in Fig. 8 was observed. FA-
CX1%ED1.5% showed 34.3- and 12.9-fold higher microbicidal effect than
FA-ED1.5% and FA-CX1%, respectively. Such enhanced antimicrobial activ-
ity of CX was also observed in a previous study backing the augmented
antimicrobial activity in combination with EDTA [43]. Stojanov et al. also
reported elevated antimicrobial efficiency of CX in presence of EDTA
[43].
4. Conclusion
Within the scope of this study novel SEDDS were developed contain-
ing combination of CX with alkyl-EDTA and characterized for cytotoxic-
ity, metal ion binding capacity, growth inhibition and enhanced
antimicrobial effect. The study has demonstrated ability of alkyl-EDTA
loaded SEDDS to bind calcium and magnesium. Furthermore, alkyl-
EDTA loaded SEDDS show concentration dependent growth inhibition
of E. coli. CX loaded SEDDS displayed 2-fold higher antimicrobial activity
than CHG. Moreover, combination of CX and alkyl-EDTA in SEDDS ex-
hibited improved antimicrobial properties. According to these results,
the combined use of antiseptics with alkyl-EDTA in SEDDS might be a
promising strategy in the global battle against bacterial resistance.