Rational Design of Cholesterol Derivative for Improved Stability of Paclitaxel Cationic Liposomes

Article abstract of DOI:10.1007/s11095-018-2367-8

Purpose: This work explores synthesis of novel cholesterol derivative for the preparation of cationic liposomes and its interaction with Paclitaxel (PTX) within liposome membrane using molecular dynamic (MD) simulation and in-vitro studies. Methods: Chole

Full text of DOI:10.1007/s11095-018-2367-8

Pharm Res (2018) 35:90
https://doi.org/10.1007/s11095-018-2367-8
RESEARCH PAPER
Rational Design of Cholesterol Derivative for Improved Stability
of Paclitaxel Cationic Liposomes
Jasmin Monpara1 & Chryso Kanthou2 & Gillian M. Tozer2 & Pradeep R. Vavia1
Received: 14 September 2017 /Accepted: 10 February 2018
#
Springer Science+Business Media, LLC, part of Springer Nature 2018
ABSTRACT
showed that CAE has more pronounced membrane thicken-
ing effect as compared to cholesterol. The cationic liposomes
showed slightly improved cytotoxicity in three different cell
lines and improved endothelial cell migration inhibition com-
pared to conventional liposomes. Furthermore, the COMET
assay showed that CAE alone does not show any genotoxicity.
Conclusions The novel cationic ligand (CAE) retains pacli-
taxel within the phospholipid bilayer and helps in improved
drug loading and physical stability.
Purpose This work explores synthesis of novel cholesterol de-
rivative for the preparation of cationic liposomes and its inter-
action with Paclitaxel (PTX) within liposome membrane using
molecular dynamic (MD) simulation and in-vitro studies.
Methods Cholesteryl Arginine Ethylester (CAE) was synthe-
sized and characterized. Cationic liposomes were prepared using
Soy PC (SPC) at a molar ratio of 77.5:15:7.5 of SPC/CAE/
PTX. Conventional liposomes were composed of SPC/choles-
terol/PTX (92:5:3 M ratio). The interaction between paclitaxel,
ligand and the membrane was studied using 10 ns MD simula-
tion. The interactions were studied using Differential Scanning
Calorimetry (DSC) and Small Angle Neutron Scattering analy-
sis. The efficacy of liposomes was evaluated by MTT assay and
endothelial cell migration assay on different cell lines. The safety
of the ligand was determined using the Comet Assay.
.
KEY WORDS COMETassay molecular dynamic
.
.
simulation paclitaxel-loaded cationic liposomes trans-well
migration assay
ABBREVIATIONS
1HNMR
CAE
DMEM
DPX
H5V
HDMEC
Proton nuclear magnetic resonance
Cholesteryl arginine ethylester
Dulbecco’s modified eagle’s medium
Disterene plasticizer xylene
Mouse endothelial cell line
Human dermal microvascular
endothelial cells
Results The cationic liposomes improved loading efficiency
and stability compared to conventional liposomes. The in-
creased PTX loading could be attributed to the hydrogen
bond between CAE and PTX and deeper penetration of
PTX in the bilayer. The DSC study suggested that inclusion
of CAE in the DPPC bilayer eliminates Tg. SANS data
IC50
Concentration at which 50%
inhibition seen
Intramolecular hydrogen bonds
Low melting point
Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s11095-018-2367-8) contains supplementary
material, which is available to authorized users.
IntraHB
LMP
MD Simulation Molecular dynamic simulation
MDA-MB 231 Human breast cancer adenocarcinoma
cell line
* Pradeep R. Vavia
pr.vavia@ictmumbai.edu.in
1
Department of Pharmaceutical Sciences and Technology, University
MolSA
MTT
Molecular surface area
under Section 3 of UGC Act – 1956, Elite Status and Center of
Excellence – Govt. of Maharashtra, TEQIP Phase II Funded, Institute of
Chemical Technology
3-(4,5-dimethylthiazol-2yl)-2,
5-diphenyltetrazolium bromide
Optical density
OD
Mumbai 400019, India
OPLS3
POPC
Optimized potentials for liquid simulations
1-palmitoyl-2-oleoyl-sn-
glycero-3-phosphocholine
2
Tumor Microcirculation Group, Department of Oncology & Metabolism
School of Medicine, The University of Sheffield
Sheffield, UK

90
Page 2 of 17
Monpara, Kanthou, Tozer and Vavia (2018) 35:90
PSA
PTX
RMSD
SASA
SPC
Polar surface area
Paclitaxel
cationic liposomes and the negatively charged glycocalyx of
endothelial cells (14). PTX is a suitable drug for vascular
targeted delivery systems as, apart from its use as first-line
chemotherapy in several cancers, it also has anti-angiogenic
properties (15). Metronomic doses of PTX have been shown
to downregulate vascular endothelial growth factor (VEGF)
and angiopoetin-1, and inhibit several endothelial cell func-
tions in-vitro such as proliferation, migration, metalloprotease
production and morphogenesis (16). Despite the potential ad-
vantages of liposomal PTX, its development has witnessed
slow progress mainly due to instability of PTX within the
liposomal bilayer (17). Therefore, pharmaceutical develop-
ment is still required to develop a PTX formulation with
higher loading efficiency and sufficient colloidal stability (18).
Molecular dynamic (MD) simulation is a force-field based
simulation that is used to calculate the behavior of a molecule
in given surroundings, taking into account the molecular mo-
tions such as vibration, bond stretching, angle bendings, and
bond formation (19). MD simulation is a tool to simulate the
interactions of drugs and proteins with the cellular membrane.
MD simulation is being increasingly used to study the interac-
tions of drugs within liposome membrane because of the close
resemblance of liposomes with cellular membranes (20). For
instance, Stepniewski et al. studied the surface structure of
PEGylated liposomes and its interaction with an ionic envi-
ronment (21). Kang and Loverde studied the molecular inter-
action of hydrophobic drugs with a phospholipid bilayer (22).
Jämbeck et al. reported MD studies of liposomes as a carrier
for hypericin using coarse grain models (23).
In the present study, we have designed and evaluated an
arginine conjugated cholesterol derivative; cholesteryl arginine
ethyl ester (CAE) in cell-free and cell systems. Arginine is con-
sidered the most hydrophilic of the natural amino-acids. The
guanidinium group of arginine has the capacity for up to six
hydrogen bonds (H-bonds). The resonance stabilized group has
pKa over 12, and thus arginine is protonated and positively
charged in all biological conditions (24). Cholesterol, due to its
effect on membrane fluidity, provides stability to the liposome
bilayer. Additionally, metabolism of CAE should follow normal
biological metabolic pathways for cholesterol and arginine and
therefore CAE can be considered biodegradable. Thus, CAE
has potential as a suitable ligand to provide stability and positive
charge to PTX liposomes.
Relative mean square deviation
Solvent accessible surface area
Soy phosphatidylcholine
Transferable intermolecular potential with
3 points
TIP3P
TLC
Thin layer chromatography
INTRODUCTION
Paclitaxel (PTX) is one of the most widely used anticancer drugs
(1). It was first approved as Taxol® for use in ovarian, lung and
breast cancer, as well as AIDS-related Kaposi’s sarcoma.
Taxol® is formulated in a (50:50 v/v) mixture of
Cremophor® EL (polyethoxylated castor oil) and dehydrated
alcohol and given as an infusion of a diluted 0.3–1.2 mg/ml
solution in 0.9% sodium chloride or similar, over one to three
hours (2). However, there are several drawbacks to this formu-
lation, which include development of severe hypersensitivity re-
action to Cremophor® EL and alcohol-induced phlebitis at the
injection site (3). Microprecipitation of PTX in the infusion ve-
hicle and leaching of plasticizers from the infusion can also oc-
cur. An alternative PTX formulation, free of Cremophor® EL,
was approved by the FDA in 2005 as Abraxane® for the treat-
ment of metastatic breast cancer (4). Abraxane® is a freeze dried
formulation of albumin-bound PTX, which upon reconstitution
produces nanoparticles of 130 nm diameter average size (5).
Although Abraxane® is associated with improved tumor re-
sponse, it has poor colloidal stability in the blood circulation,
resulting in release of free PTX into the bloodstream from dis-
assembly of the Abraxane® nanoparticles (6).
There are numerous reports advocating liposomes as an
alternative to Taxol® for improved PTX delivery to tumours
(7). For instance, liposomes can alter the pharmacokinetics
and distribution profile of PTX favoring improved clinical
outcomes (8). Currently, there are several liposomal PTX for-
mulations in development, but Lipusu® is the only one ap-
proved so far (9). In addition, LEP-ETU® developed by
NeoPharm (10) has completed Phase II clinical studies and
EndoTag-1® developed by Medigene AG (11) has also com-
pleted Phase II clinical studies, as a tumor endothelial cell
targeted PTX delivery system. Cationic liposome based
targeting of an anti-angiogenic agent to tumor endothelial
cells has gained tremendous attention in the last decade due
to preferential distribution to angiogenic blood vessels (12).
Hiroshi Kiwada et al. has reviewed the cationic liposomes for
targeting anticancer drug to the tumor vasculature, their in-
vivo pharmacokinetics and toxicity and their clinical applica-
tions (13). The mechanism by which this distribution occurs is
still unclear, although it is commonly believed to relate to an
electrostatic interaction between the positive charge of
MATERIALS AND METHODS
Materials
Unsaturated Soy phosphatidylcholine (SPC) and 1,2-
dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) were obtained
from Lipoid GmbH (Ludwigshafen, Germany). Cholesteryl
Arginine Ethyl Ester (CAE) was synthesized and purified in-

Paclitaxel Cationic Liposomes (2018) 35:90
Page 3 of 17 90
house following established procedures (25,26). PTX was obtain-
ed from Dr. Reddy’s laboratories Ltd. (Hyderabad, India).
Cholesterol was purchased from Merck India Pvt. Ltd.
(Mumbai, India). 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetra-
zolium bromide (MTT) was purchased from Sigma-Aldrich
Co Ltd. (Dorset, UK). Comet Assay kit was purchased from
Trivigen Inc. (AMS Biotechnology (Europe) Limited,
Abingdon, UK). All other reagents were of analytical grade
and purchased from Merck India Pvt. Ltd. (Mumbai, India).
tautomers, isomers or ionization states from flat the geometry
of .sdf file using Epik followed by geometry optimization based
on the OPLS3 force field. pKa prediction of CAE was per-
formed using empirical Epik® workflow. Other chemical pa-
rameters were calculated using QikProp®. Further calcula-
tions were carried out using MacroModel® module and the
OPLS3 force field as implemented in MacroModel®. The
energy of the PTX and CAE was minimized using
MacroModel-minimization step. Pre-equilibration of lipid bi-
layer composed of 1-palmitoyl-2-oleoyl-sn-glycero-3-
phosphocholine (POPC) phospholipid and embedding PTX
and CAE or PTX and Cholesterol was performed using the
Desmond® program. The predefine transferable intermolec-
ular potential functions (TIP3P) water model was used to as-
similate water molecules. Appropriate Na+ and Cl− were
added to neutralize the systems. The total number of atoms
in the investigated systems was approximately 10, 430 includ-
ing about 2000 water molecules and 40 POPC phospholipids.
The orthorhombic periodic box dimensions were set to 10 Å
X 10 Å X 10 Å. After building the solvated system, minimi-
zation and relaxation of the membrane were performed using
the OPLS3 force field in Desmond®. MD simulation was
performed using Desmond® at constant temperature and
pressure (NPT ensemble; 300 K, 1.01325 bars) for 10 ns.
Cell Lines
The immortalised mouse Endothelial Cell line H5V was a
kind gift from Dr. Annunciata Vecchi (27). The human breast
cancer adenocarcinoma cell line (MDA-MB 231) was from
ATCC (ATCC HTB-26). Both H5V and MDA MB 231 cells
were maintained in Dulbecco’s modified Eagle’s medium
(DMEM, Lonza, UK) supplemented with 10% heat-
inactivated FBS, 10 mM L-glutamine, 100 U/ml penicillin
and 100 μg/ml streptomycin in a 5% CO2 /air incubator at
37°C. Human Dermal Microvascular Endothelial Cells
(HDMEC) were purchased from PromoCell GmbH and
maintained in Endothelial Cell Growth Medium MV2
(PromoCell GmbH) supplemented with Supplement Mix C-
39226 (PromoCell GmbH), 100 U/ml penicillin and 100 μg/
ml streptomycin in a 5% CO2 /air incubator at 37°C.
Synthesis of Cholesteryl Arginine Ethyl Ester
Computational Methods
The designed Cholesteryl Arginine Ethyl Ester was synthe-
sized by esterification of L-arginine with ethanol to form L-
Arginine ethyl ester 2HCl (AEE) followed by N-amidation
using Cholesteryl chloroformate to form the final product,
Cholesteryl Arginine Ethyl Ester (CAE), as shown in Fig. 1.
Briefly, L-Arginine (base) (2 g) was dispersed in ethanol
(100 ml) in a round bottom flask and cooled to 10°C.
Thionyl chloride (1.5 Mole equivalents; 1.2 ml) was added
The structures of PTX, CAE, Arginine ethyl ester (AEE), and
Cholesterol for initial studies were generated using
ChemBioOffice® V.16 and saved as structure-data files
(.sdf). Structures were imported into Maestro® 10.2 module
of Schrödinger® Discovery Suite Release 2016–2. LigPrep®
module was used to generate 3D structures of possible
Fig. 1 Schematics for the synthesis
of Cholesteryl arginine ethyl ester.

90
Page 4 of 17
Monpara, Kanthou, Tozer and Vavia (2018) 35:90
dropwise to the mixture. The mixture was refluxed for 4 h at
60°C to carry out esterification. The reaction was monitored
by TLC using Butanol-Acetic Acid-Water (3:1:1) as mobile
phase and ninhydrin solution for detection of the product.
After completion of the reaction, excess ethanol was removed
by vacuum drying to obtain a yellowish liquid, AEE. AEE was
solubilized into 50 ml Tetrahydrofuran: water (50:50) mixture
and pH was adjusted to 9.0 with 1 M NaOH solution.
Further, amidation was carried out by Schotten-Baumann
reaction using Cholesteryl chloroformate (25). Cholesteryl
chloroformate (0.9 Mole equivalent; 4.6 g) was solubilized in
30 ml Tetrahydrofuran and added drop-wise to L-Arginine
Ethyl Ester solution under continuous stirring with pH of the
reaction mixture was maintained between 7.5–7.8. The whole
reaction was carried out at 10–15°C for 6–7 h. The reaction
was monitored by TLC as mentioned above. After completion
of the reaction, pH was neutralized by adding dilute HCl
solution. Tetrahydrofuran was removed by vacuum evapora-
tion. The final product was extracted with chloroform and
dried using rotary evaporator. The product was eluted with
CH2Cl2/CH3OH (9:1 v/v) from a silica gel (70–230 mesh),
and the solvent was removed under reduced pressure to give a
light yellowish solid (5.0 g, 70% yield), Rf = 0.6 (CH2Cl2/
CH3OH 4:1 V/V). The elute were analyzed using
Anisaldehyde as a developer in TLC to confirm cholesterol
large multilamellar liposomes was extruded through a poly-
carbonate membrane of pore size 200 nm using lipid extruder
(Lipex Biomembrane Inc., Canada) to form small unilamellar
liposomes of uniform particle size. For Differential Scanning
Calorimetry, DPPC was used instead of SPC to prepare lipo-
somes. The particle size and zeta potential of liposomes were
measured by dynamic light scattering and electrophoretic
light scattering, respectively, using Malvern Zetasizer
(Malvern Instrument Ltd., UK).
Drug Loading Efficiency
PTX loading was determined using a reported method (28).
Briefly, Sephadex G-25 gel was autoclaved in ultra-purified
water to allow hydration of Sephadex G-25 particles. The
column was formed in 5 ml syringe by placing glass wool at
the bottom of the syringe and centrifuging at 350 g for 5 min.
The dry column was loaded with 1 ml blank liposomes to
saturate the column. The loaded column was then centrifuged
at 1520 g for 5 min to remove the blank liposomes.
Subsequently, PTX liposomes were added to the column
and centrifuged at 1520Xg for 5 min. PTX was extracted
from one ml of elute using tert-butyl methyl ether on a vortex
mixer for 60 s. Upon centrifugation at 1520Xg for 10 min, the
separated organic layer was then transferred to clean centri-
fuge tube and dried using nitrogen at 40°C. PTX content was
determined using a reverse phase HPLC system (Waters,
USA) unit, Hypersil C-18 column (250 × 4.6 mm, 5 μ). The
mobile phase consisted of acetonitrile: water (60:40). The
analysis was performed at a flow rate of 1.0 ml/min with
UV detection at 227 nm (29). Loading efficiency was deter-
mined using following equation:
1
backbone. The HNMR spectrum was recorded using a
Brucker® 800 MHz spectrophotometer using spectroscopic
grade CDCl3. Molecular mass determination of the synthe-
sized compounds was carried out using LC-MS (Finnigan
LCQ Advantage Max, Thermo electron corporation, USA).
Anal. Calcd. for C36H62N4O4 (MW 614.92): C, 70.32; H,
10.16; N, 9.11; O, 10.41. Found: C, 70.20; H, 10.20; N,
9.30; O, 10.30. MS (positive/negative ES): m/z 615.27
1
[M + H]+, H NMR (800 MHz, CDCl3, 20°C, TMS)
Loading Efficiency ¼ ðAmount of drug in liposomes Ä Amount of lipid in liposomesÞ X 100
0.854–0.868 (6H, m, 1), 0.9–3(aliphatic protons of cholesteryl
and arginine moiety, 40H, m, unlabelled), 4.189–4.211(2H,
triplet, f), 4.43(1H, s, e), 5.32(1H, s, d), 5.829–5.837(1H, dou-
blet, c), 6.972(4H, broad s, b), 7.71(1H, broad s, a).
Stability of Liposomes
Stability of liposomes was observed at 2–8°C for three
months. After the preparation of liposomes, 2 ml aliquots were
removed from the stability samples and subjected to particle
size determination, zeta potential measurement, and drug
loading measurement.
Preparation of Liposomes
The liposomes were prepared by the thin film hydration
method. Briefly, lipids and PTX were dissolved in 3 ml of
chloroform. The thin film was prepared using a 250-ml round
bottom flask and vacuum evaporator at 60°C under reduced
pressure. The film was hydrated using 0.9%w/v NaCl solution
at 60°C. Cationic liposomes were composed of SPC/CAE/
PTX (77.5:15:7.5 M ratio). The conventional liposomes were
composed of SPC/cholesterol/PTX (92: 5: 3 M ratio). For the
preparation of blank cationic liposomes and blank conven-
tional liposomes, PTX was omitted from the formulation.
The total solid content of the liposomal dispersion was kept
constant to 10 mM/10 ml in each case. The suspension of
Differential Scanning Caorimetry (DSC)
For DSC study, DPPC liposomes were prepared using DPPC
instead of SPC in same molar ratios as described above.
Briefly, a thin film of the liposome constituents was hydrated
using water at 60°C for few hours. The final lipid concentra-
tion was adjusted to 30 wt%. The mixture was resuspended
using vigorous vortex mixing and subsequently using
ultrasonication to yield homogenous multilamellar vesicles.

Paclitaxel Cationic Liposomes (2018) 35:90
Page 5 of 17 90
The dispersions (30 μL) were filled into DSC pans made of
aluminum, hermetically sealed and were stored overnight at
4°C. DSC measurements were performed by using Perkin
Elmer Pyris 6 DSC (USA) with empty aluminum pans as
reference. The heating thermograms were obtained between
20°C and 55°C at a scan rate of 2°C/min. Baseline correction
was performed using Pyris DSC software version 8.0.
For a spherical particle of radius R, F(Q) is given by
"
#
sinðQRÞ−QRcosðQRÞ
FðQ Þ ¼ 3
:
3
ðQRÞ
For a system consisting of monodisperse unilamellar vesi-
cles (33), dΣ/dΩ can be expressed as
ꢄ
ꢅ
2
dΣ
dΩ
4
3
3J1ðQRÞ
4
3
3 3J1½Q ðR þ tފ
Q ðR þ tÞ
2
ðQ ; RÞ ¼ nðρv−ρsÞ
πR3
−
πðR þ tÞ
QR
Small Angle Neutron Scattering (SANS)
where n denotes the number density of the vesicles, ρv and ρs
are the scattering length densities of the vesicle bilayer and the
solvent, respectively. R is the radius of the vesicle and t is the
thickness of the bilayer. J1(x) is the first order Bessel function
and is given by
To evaluate the effect of CAE and Cholesterol on SPC lipo-
somes, SANS study was carried out. The study was performed
using SANS instrument in DHRUVA reactor at the Bhabha
Atomic Research Centre (BARC) in Trombay, India (30).
The samples were poured in a 0.5 cm path-length UV grade
quartz sample holder with tight-fitting Teflon® stoppers and
sealed with parafilm. The mean wavelength of the
monochromatized beam is 5.2 Å with a spread of Δλ/λ ~
15%. The angular distribution of neutrons scattered by the
sample is recorded using a 1 m long one-dimensional He3
position sensitive detector. The instrument covers a Q-range
of 0.015–0.35 Å−1. The temperature in all the measurements
was kept fixed at 30°C.
sinx−xcosx
J1ðxÞ ¼
x2
The polydispersity in size distribution of particle is incor-
porated using the following integration
dΣ
dΩ
dΣ
dΩ
ðQ Þ ¼ ∫
ðQ ; RÞf ðRÞdR þ B;
where f(R) is the particle size distribution and usually
accounted by a log-normal distribution as given by
Scattering intensities from the sample solutions were
corrected for detector background and sensitivity, empty cell
scattering, and sample transmission and solvent intensity was
subtracted from that of the sample and the resulting corrected
intensities were normalized to absolute cross-section units.
"
#
ꢀ
ꢁ
2
1
1
2σ2
R
Rmed
pffiffiffiffiffi
f ðRÞ ¼
exp −
ln
;
2πRσ
Where, Rmed is the median value and σ is the standard devia-
tion (polydispersity) of the distribution. The mean radius (Rm)
is given by Rm = Rmed exp.(σ2/2) (34).
Small-Angle Neutron Scattering Analysis
The data have been analyzed by comparing the scattering
from different models to the experimental data. Throughout
the data analysis, corrections were made for instrumental
smearing, where the calculated scattering profiles smeared
by the appropriate resolution function to compare with the
measured data (35). The fitted parameters in the analysis were
optimized using nonlinear least-square fitting program to the
model scattering (36).
The differential scattering cross-section per unit volume (dΣ/
dΩ) as measured for a system of monodisperse particles in a
medium can be expressed as
ꢀ
ꢁ
ꢂ
ꢃ
ðQ Þ ¼ nV 2 ρp−ρs 2PðQ ÞSðQ Þ þ B;
dΣ
dΩ
where n denotes the number density of particles, ρp and ρs are,
respectively, the scattering length densities of particle and sol-
vent and V is the volume of the particle. P(Q) is the
intraparticle structure factor and S(Q) is the interparticle struc-
ture factor. B is a constant term representing incoherent back-
ground, which is mainly due to the hydrogen present in the
sample (31,32).
Intraparticle structure factor P(Q) is decided by the shape
and size of the particle and is the square of single-particle form
factor F(Q) as determined by
In-Vitro Cytotoxicity Study
In-vitro cytotoxicity of PTX solution (in methanol), cationic
liposomes, conventional liposomes and blank liposomes de-
void of PTX (all in 0.9% saline) was determined using the
MTT assay. The studies were carried out on three different
cell lines - MDA MB 231, H5V and HDMEC to determine
the sensitivity of PTX formulations on tumour cells, immor-
talized mouse endothelial cells and human primary endothe-
lial cells respectively. In separate experiments, MDA MB 231,
H5V and HDMEC cells were seeded at 1*104 cells/well,
D
E
2
PðQ Þ ¼ jFðQ Þj

90
Page 6 of 17
Monpara, Kanthou, Tozer and Vavia (2018) 35:90
5*103 cells/well and 2*103 cells/well, respectively, in 96-well
flat-bottom tissue culture plates 24 h prior to assay. For MDA
MB 231 and H5V, PTX solution was tested at a concentra-
tion range of 0.1 nM to 30 nM and the liposomal formulations
were tested at a PTX concentration range of 10 nM to
150 nM. Because of the greater sensitivity of HDMEC to
PTX, PTX solution was tested at a concentration range
0.1 nM to 10 nM and the liposomes were tested at a PTX
concentration range of 0.5 nM to 30 nM for this cell type. The
stock solution (1 μM) of PTX was prepared in methanol and
the stocks solutions ((1 μM PTX) of liposomes were prepared
in PBS. Further dilutions were prepared in the cell culture
media for all samples. The negative control cells were incu-
bated with the appropriate drug vehicles only.
For the MTT assay, MTT (3–4,5-dimethylthialzol-20yl)-
2,5-diphenyl tetrazolium bromide) was dissolved in sterile
PBS at a concentration of 5 mg/ml. After 72 h of incubation
of cells with drugs at 37°C, as described above, 20 μl of MTT
solution was added to each well and incubated for 5 h. The
unreduced MTT and medium were discarded from the wells.
Then, 100 μl of acidified isopropanol (1 ml 2 M HCl mixed
with 50 ml isopropanol) was added to each well to dissolve the
MTT formazan crystals. Absorbance was read at 540 nm in a
microplate reader. Percentage viability was determined using
the following equation:
untreated cells. The data obtained were expressed as mean
standard deviation from the mean (SD) and analyzed using a
one-way ANOVA within Graph Pad Prism® 7.03 statistical
analysis software. P < 0.05 was considered to be statistically
significant. All the experiments were carried out in triplicate.
Comet Assay
The comet assay (single cell gel electrophoresis) measures de-
oxyribonucleic acid (DNA) strand breaks in eukaryotic cells.
Briefly, the cells are embedded in agarose gel followed by lysis
by detergent and high salt concentration to form nucleoids. If
the nucleoids have strand breaks, they form comet-like struc-
tures after electrophoresis at high pH that can be observed by
fluorescence microscopy. The intensity of the comet tail rela-
tive to the head reflects the number of DNA breaks. The
comet assay was performed to determine potential
genotoxicity of the synthesized compound-CAE. CAE was
dissolved in dimethyl sulphoxide (DMSO) at the level of
10 mg/ml and sufficient volume was added in the media to
achieve a final concentration of 5 μg/ml, which corresponds
to the amount needed to deliver 175 mg/m2 dose of PTX. A
ten times higher concentration of CAE (i.e. 50 μg/ml) was also
tested to confirm the safety of the compound. The final con-
centration of DMSO achieved in each well was below 1% v/v
which is considered non-harmful to the cells (38). Nonetheless,
control cells were treated with an equivalent concentration of
DMSO to nullify the effect, if any.
Briefly, in two independent experiments, HDMEC and
MDA MB 231 cells were seeded into six-well plates at a den-
sity of 5*104 cells/well and allowed to grow for 24 h. The test
solutions were added to achieve the desired concentrations
and incubated for a further 24 h. Cells treated with 200 μM
hydrogen peroxide for 20 min served as a positive control.
The negative control cells were treated with an equivalent
volume of DMSO mixed with media. After the incubation,
cells were trypsinized and washed with PBS. Cell density was
adjusted to 1*104 cells/ml in PBS. The comet assay was per-
formed using a Comet assay kit (Trivigen Inc.). The cells were
suspended in LMP agarose at 37°C and allowed to set on the
comet slides at room temperature. Once the cell suspension
set on the slide, cell lysis was performed using the lysis buffer
included in the kit for one hour at 4°C. Cell lysis was followed
by DNA unwinding using alkaline unwinding solution for one
hour at 4°C. After unwinding, electrophoresis was carried out
at 21 V, 300mAmp for 30 min. The agarose matrix was
washed twice with ultrapure water followed by 70%V/V eth-
anol for five minutes each and allowed to dry. The DNA was
stained using SYBR® Gold dye and images were captured
using fluorescence microscopy. Images were analyzed using
Comet Score IV software (39). Tail moment and Olive mo-
ment were used to determine the extent of DNA damage.
Each parameter gives a value comparing the amount of
%Cell Viability ¼ ½ OD ðTest SampleÞ Ä OD ðUntreated Control SampleފX100
The IC50 values for different PTX formulations were cal-
culated by non-linear regression curve fitting on % cell viabil-
ity vs PTX concentration plots.
Trans-Well Cell Migration Assay
Inhibition of the migratory function of HDMEC cells by PTX
solution and the PTX liposome formulation was determined
using the Trans-well Migration Assay (37). Briefly, 7.5*104
cells were suspended in endothelial cell medium and placed
in the upper compartment of an 8-μm pore Trans-well
(Costar, Corning). After one hour, drug or liposomes were
added to both the upper and lower compartments. Three
different concentrations (0.1 nM, 0.5 nM and 1 nM) of PTX
solution and 0.1 nM, 1 nM and 2.5 nM of PTX liposomes
were tested in the assay. The cells were allowed to migrate to
the lower compartment for 6 h at 37°C. After the incubation,
media were removed from the Trans-wells and cells were
fixed using ice-cold methanol. The cells were stained using
hematoxylin solution and the cells at the proximal side of the
membrane were removed using cotton swabs. The membrane
was removed from the inserts and mounted to frosted slides
using DPX (Distrene 80). The cells that migrated to the lower
side of the membrane were counted using a microscope fitted
with a 40X objective and % migration was normalized against

Paclitaxel Cationic Liposomes (2018) 35:90
Page 7 of 17 90
DNA in the tail compared to the amount of DNA in the head
calculated using fluorescence intensity. The Olive moment is
used to take account of inter-sample variability. The tail mo-
ment and Olive moment were calculated as follows:
Atomistic Molecular Dynamic Simulation
In our study, MD simulation followed by trajectory analysis
showed H-bond interaction between the guanidine group of
CAE and the carbonyl group of the taxane ring of PTX, as
seen in Figs. 2a1 and 3a. There was no H-bond interaction
between PTX and cholesterol, rather PTX has a higher affin-
ity towards the aqueous compartment mainly due to hydrogen
bonding of carbonyl and hydroxyl groups of PTX with water
molecules as shown in Figs. 2b1 and 3a. The interactions
between PTX and water were not observed when cholesterol
was replaced by CAE in the membrane. As shown in Fig. 2a2,
b2, the MD simulation of two systems (i.e. PTX + CAE in
POPC membrane and PTX + Cholesterol in POPC mem-
brane), the sterol ring of CAE was primarily centered in the
lipid bilayer core. However, the sterol ring of cholesterol cen-
tered towards the outer part of the phospholipid membrane.
Apart from that, PTX was found in the vicinity of the sterol
ring system of either CAE or cholesterol in both the simula-
tions. The interaction of PTX with CAE could be explained
by a higher probability of H-bond between PTX and CAE.
Although H-bond was not seen between PTX and cholesterol,
there might be other hydrophobic interactions strong enough
to bring the taxane ring of PTX near the cholesterol molecule.
The relative mean square deviation (RMSD) plot gives a
deviation of the ligand with respect to the reference confor-
mation (0 ns). The RMSD plots of ligands (PTX: CAE or
PTX: Cholesterol) (Figs. 3b and 4) show only a small change
(about 2 Å) in both the simulations. Though the change in
RMSD is very small, there is a noticeable difference between
both the simulations. The PTX-CAE system showed higher
mobility in the bilayer evident from RMSD plot referenced to
time 0 ns but stabilized at 1.3 Å. The PTX-cholesterol system
showed slightly lower mobility but stabilized at 1.95 Å. The
Intramolecular Hydrogen Bonds (IntraHB) plot shows that
PTX could frequently form two H-bonds during the simula-
tion i.e. one intramolecular H-bond and one H-bond with
CAE. However, only one intramolecular H-bond was formed
in the PTX-Cholesterol system (Fig. 3a and 4). The Molecular
Surface Area (MolSA) is equivalent to the van der Waals sur-
face area. The MolSA for PTX was increased from 675 Å2 to
690 Å2 for in PTX: CAE liposomes but decreased from
680 Å2 to 660 Å2 in PTX: cholesterol liposome, which indi-
cates an increase in van der Waals interaction between PTX
Tail Moment ¼ ðTail LengthÞ X ð%DNA in the tailÞ
Olive Moment ¼ ðCentre of mass of tail−Centere of mass of headÞXð%DNA in tailÞ
The data obtained was expressed as mean standard devia-
tion from the mean (SD). The data was analyzed using a one-
way ANOVA within Graph Pads Prism® 7.03 statistical analysis
software. P < 0.05 was considered to be statistically significant.
RESULTS AND DISCUSSION
Chemical Parameters
The chemical parameters predicted using QikProp® are
listed in Table I. The H-bond acceptor in PTX outnumbers
H-bond donors. Thus, we can expect H-bond interactions of
PTX with water through oxygen atoms and with CAE
through guanidine nitrogen atoms (40). Generally, H-bonds
of a molecule with water results in an increase in its aqueous
solubility. However, the hydrogen bond interaction of PTX
with water is very weak and insufficient to solubilize the mol-
ecule due to the phenomenon called Bhydrophobic
clustering^ and higher lipophilicity (log p = 4.9) of PTX.
Altogether, these characteristics lead to precipitation of PTX
in an aqueous environment (41). CAE has a higher probability
of forming H-bonds compared to cholesterol due to the pres-
ence of guanidine groups. We reasoned that this would result
in a greater interaction of CAE with PTX and higher stability
of PTX within the liposome bilayer.
The hydrophilic and hydrophobic surface areas (SA) pro-
vide information regarding structural features responsible for
hydrophilic and hydrophobic intermolecular interactions (42).
The higher hydrophobic SA of CAE and cholesterol com-
pared to their hydrophilic SA indicates that CAE and choles-
terol may have higher interaction with the lipophilic tails rath-
er than the hydrophilic head group of the phospholipid bilay-
er. The pKa prediction of CAE revealed that the guanidine
group of CAE has pKa more than 11 and hence is expected to
be positively charged in all biological conditions.
Tab le I Chemical Parameters for PTX, CAE, Cholesterol, and AEE Calculated Using QikProp®
Molecule
Mol.wt.
Volume
Total SA
Hydrophilic SA
Hydrophobic SA
H-bond donor
H-bond acceptor
log p
PTX
853.9
620.9
202.2
386.6
2271.2
2136.0
794.3
1046.7
1141.1
507.6
215.8
202.0
245.7
49.1
350.0
939.0
261.9
658.0
3.0
7.0
6
17.6
9.4
5
4.983
4.541
−0.915
7.03
CAE
AEE
Cholesterol
1390.0
736.6
1
1.7

90
Page 8 of 17
Monpara, Kanthou, Tozer and Vavia (2018) 35:90
Fig. 2 (a1)-(b1) Representation of H-bond interaction within the lipid bilayer. (a2)-(b2) Trajectories of PTX at the end of 10 ns MD simulation of (a2) PTX+
CAE system and (b2) PTX + Cholesterol system. The PTX molecule is represented in blue color. CAE molecule is represented in maroon color. The simulation
box is represented in pink color. The water molecule is represented in tube model and the phospholipids are represented in wire model. H-bonds are
represented as green and red dashed lines. (a1)-(a2) The H-bond interaction between CAE and PTX in cationic liposomes (b1)-(b2) in conventional liposomes,
Cholesterol could not form H-bonds with PTX. Moreover, due to cholesterol, the membrane showed higher water permeability and thus, water molecules
could form H-bonds with PTX.
with CAE and surrounding phospholipid molecules com-
pared to PTX-cholesterol interaction. Comparable results
were found with respect to Solvent Accessible Surface Area
(SASA), which indicates an increase in solvent accessible sur-
face area (phospholipid accessible surface area in this case) in
PTX-CAE system compared to decrease in SASA in PTX-
Cholesterol system. This might be an additional factor
resulting in higher PTX stability in bilayer when CAE is in-
corporated in the bilayer. The Polar Surface Area (PSA),
which is the solvent accessible surface area in the molecule,
contributes only by oxygen and nitrogen atoms, which
remained approximately the same for both systems.
Additionally, in the PTX-Cholesterol system, the membrane
has higher water permeability and PTX could form H-bonds
with water molecules (Fig. 2b1, b2). Kang et al. has reported
MD simulation PTX incorporated in POPC membrane.
They observed higher water penetration in the membrane
and ‘water fingers’ in the proximity of PTX aggregates when
the PTX concentration is higher than 12 mol% (22). Previous
studies by Balasubramanian et al. and Belsito et al. reported
that the preferential site of PTX deposition is the surface of
phospholipid vesicles (43,44). The PTX molecule cannot force
lipid molecules away for inserting in the bilayer and has lim-
ited access to the hydrophobic core of the bilayer (43). The
self-aggregation of PTX in cholesterol containing liposomes
can be explained by higher water permeability of membrane
and H-bonds between PTX and water. This may in-turn ex-
pel the PTX from lipid bilayer resulting in self-aggregation
due to hydrophobicity overcoming H-bonds. The results of
MD simulation indicated that CAE could be a suitable cat-
ionic ligand to prepare PTX loaded cationic liposomes.
Formulation Development
Despite offering several advantages over conventional chemo-
therapy and other nano delivery systems of PTX, liposomal
PTX has witnessed slow development mainly due to limitations
on PTX loading and stability in the lipid bilayer (17). Several
studies indicate that PTX is extremely unstable within the bi-
layer and self-aggregates at concentration ≥ 3 mol%. Previous
studies of the interactions of PTX with a phospholipid bilayer
suggested that, at concentrations less than 3 mol%, PTX favors
the fluid state of the membrane (i.e. lowering Tm) mainly by
interaction with the surface of the phospholipid vesicle.
However, at higher concentration, PTX penetration in the
membrane is less due to bulkiness of the molecule and limited
access to the hydrophobic region of the membrane (43).
Cholesterol has been used traditionally to stabilize the lipid bi-
layer due to its effects on bilayer fluidity and rigidity mainly
through the elimination of phase transition and decreasing
membrane permeability at higher concentration. However, at
low cholesterol concentration, the membrane has higher perme-
ability due to increased membrane fluidity (45). In the presence
of cholesterol, a phospholipid membrane has poor ability to
incorporate PTX due to competition between the two for load-
ing pockets in the lipid bilayer (46).

Paclitaxel Cationic Liposomes (2018) 35:90
Page 9 of 17 90
Fig. 3 (a) H-bond plot for the MD simulation of cationic liposomes having PTX: CAE and conventional liposomes having PTX: Cholesterol. There is a H-bond
interaction between PTX and CAE but no H-bonds were observed between PTX and cholesterol (b) RMSD plot for the MD simulation of cationic liposomes
having PTX: CAE and conventional liposomes having PTX: Cholesterol. The ligands had slightly higher mobility in the bilayer in cationic liposome membrane.
As shown in Fig. 5a, increasing cholesterol concentration
results in a decrease in PTX loading in an SPC bilayer. J.
Allen Zhang et al. also observed a similar effect of cholesterol
in a phospholipid bilayer (28). In contrast, increasing CAE
concentration in the lipid bilayer did not reduce PTX loading
up to 15 mol%. Further increase in CAE concentration re-
duced PTX loading in the lipid bilayer. This may be attribut-
ed to a reduced number of phospholipid molecules to form an
ordered bilayer which can incorporate PTX and CAE at the
same time. The cationic liposomes were composed of SPC/
Fig. 4 Simulation interaction diagrams both simulations (PTX: CAE system and PTX: Cholesterol system).

90
Page 10 of 17
Monpara, Kanthou, Tozer and Vavia (2018) 35:90
Fig. 5 (a) PTX loading in liposome bilayer. As the concentration of cholesterol increases in the lipid bilayer, the loading of PTX decreases. However, CAE
showed stabilizing effect and PTX loading did not decrease up to 15 mol%. The error bars represent the mean SD of each value (n = 3) (b) stability of
liposomes over three months. The cationic liposomes retained the loaded PTX while stored at 2–8°C for three months. However, conventional liposomes
showed a consistent decrease in PTX loading. The error bars represent the mean SD of each value (n = 3).
CAE/PTX (77.5:15:7.5 M ratio) with particle size 135.4
1.3 nm and PTX loading 7.5 mol%. The cationic liposomes
had a surface charge of +40 mV. The conventional liposomes
were composed of SPC/Cholesterol/PTX (92: 5: 3 M ratio)
with particle size 145.4 1.4 nm and −11.7 mV surface charge
with 3 mol% PTX loading. Figure 5b shows stability testing
over three months for cationic liposomes and conventional lipo-
somes at 2–8°C. Cationic liposomes could retain PTX for the
study period (approximately 98% at the end of three months).
However, conventional liposomes could not retain PTX in the
lipid bilayer (approximately 55% at the end of three months).
The possible explanation for higher PTX loading in the pres-
ence of CAE could be deeper penetration of CAE into the
hydrophobic core of the membrane. This may result in greater
interaction between PTX and hydrophobic acyl chains mainly
due to van der Waals interactions (22).
characteristic parameters can be used to describe the changes
in the lipid membrane such as pretransition (Tp), main tran-
sition (Tg), width of transition at half peak height (ΔT1/2) and
the changes of enthalpy (ΔH). Upon addition of thermal en-
ergy, the pretransition Tp (Lβ’ to Pβ’) occurs between gel phase
rippled phase and the main transition (Tm) of phospholipids
bilayer occurs between gel (ordered) to fluid (disordered) la-
mella state (Pβ to Lα) affecting the van der waals interactions
between hydrocarbon chains, increasing their mobility. The
interaction of the drug molecule within the bilayer can be
determined by the co-operativity of the molecules during the
transition process (47).
The natural lecithins such as Egg PC and Soya PC have
unsaturation in the fatty acid chains therefore, the Tg of lipo-
somes consisting such lipids is extremely low (−5°C to −15°C
The results of MD simulation are in good agreement with
these observations and with current research work. The MD
simulation shows that in presence of cholesterol (1) PTX is
primarily located at the outer hydrophobic core of the lipo-
some (2) there is increased water permeability of the bilayer (3)
increased probability of H-bond between PTX and water.
These factors could be responsible for the expulsion of PTX
from the bilayer, with the use of cholesterol. On the other
hand, in the presence of CAE, greater PTX loading and
stability might be resulting from (1) H-bond between PTX
and CAE (2) deeper penetration of PTX to the hydrophobic
core of the membrane and (3) decreased probability of inter-
action with water as shown in the MD simulation.
Differential Scanning Calorimetry
Differential Scanning Calorimetry is most frequently used
technique to determine the drug- biomembrane interactions
within the formed liposomal membrane. Several
Fig. 6 Differential Scanning Calorimetry: The DSC thermograms of DPPC
liposomes consisting of different concentrations of cholesterol, PTX and CAE.

Paclitaxel Cationic Liposomes (2018) 35:90
Table II Differential Scanning
Page 11 of 17 90
ΔH (J/g)
Sr. No.
Liposome formulation
Tm (°C)
ΔT1/2 (°C)
Calorimetry Parameters
1
DPPC
40.49 0.07
40.20 0.09
40.05 0.08
40.31 0.09
40.29 0.12
40.15 0.14
40.00 0.13
40.33 0.10
40.29 0.14
–
0.76 0.03
0.92 0.02
1.12 0.02
1.08 0.03
1.08 0.04
1.12 0.02
1.28 0.04
1.08 0.03
0.58 0.01
–
12.443 0.97
2
DPPC +3% PTX
11.336 0.92
9.213 0.95
11.236 0.93
7.709 1.12
6.062 0.97
7.536 1.11
4.556 0.94
0.832 0.91
–
3
DPPC +5% PTX
4
DPPC +5% Cholesterol
DPPC +10% Cholesterol
DPPC +15% Cholesterol
DPPC +10% Cholesterol +5% PTX
DPPC +5% CAE
5
6
7
8
9
DPPC +10% CAE
10
11
DPPC +15% CAE
DPPC +10% CAE + 5%PTX
–
–
–
for Egg PC and −20°C to −30°C for Soya PC). Additionally,
the peak corresponding to Tg is well defined only after
prolonged storage of the liposomes below Tg. Nevertheless,
the peak obtained is not as sharp as natural lecithins consisting
of mixture of fatty acids. Synthetic lipids on the other hand
consist of pure component and thus give sharp peak at their
respective Tg (48,49). Thus, DPPC was selected due to its
suitability of studying thermal behavior (Tg~ 41°C) and be-
cause it belongs to choline family of phospholipids.
The interactions of PTX and/or cholesterol with DPPC
membrane has been widely studied in the past (43,50–52).
The addition of PTX or cholesterol is known to abolish Tp
in the DPPC membrane. Tp is highly sensitive to the presence
of other molecules in the polar head region of phospholipids
and can not be ascribed to any specific molecular changes.
With increasing concentration, PTX shifts the Tm towards
slightly lower temperature with broadening of the main tran-
sition with no significant changes in ΔH. The broadening of
main transition can be described by ΔT1/2 which is a measure
of the phospholipid assemblies indicating a decrease in size of
the cooperative unit. PTX induced changes in Tm indicates
that paclitaxel should be located in outer hydrophobic region
of the membrane (43,50,51,53). Cholesterol slightly reduced
the Tm with broadening of the transition peak but markedly
reduced the ΔH with increasing cholesterol concentration.
Cholesterol is assumed to disrupt the ordered array of hydro-
carbon chains in gel phase and thereby increasing the fluidity
at low concentration and at high concentration, modulating
its rigidity by restricting the movement of hydrocarbon chains
above Tm which results in elimination of pretransition and
broadening of Tg (48,52,54,55).
We have studies thermal properties of DPPC liposomes con-
taining different concentration of PTX and cholesterol (Fig. 6
and Table II). As shown in Fig. 6, pure liposomes consisting of
only DPPC showed endothermic peak at 35°C corresponding
to pretransition (Tp) and at 40.49°C corresponding to the main
transition (Tm). Addition of PTX slightly lowered the Tm and
ΔH with broadening of the main transition evident from in-
crease in ΔT1/2. This indicates that PTX destabilizes the phos-
pholipid assembly with reduction in cooperativity of the main
transition leading to more flexible membrane. Further, the
changes in Tm indicates PTX should be located in outer hydro-
phobic co-operative zone of the membrane (51). Addition of
cholesterol produced very slight reduction in Tm. Conversely,
Tab le I II Bilayer Thickness Determined by SANS
Sr. No.
Liposome composition
Bilayer thickness t (Å)
1
2
3
4
5
6
SPC
24.3
27.2
27.1
28.2
29.9
30.1
SPC:PTX (97:3 M ratio)
SPC:Cholesterol (95:5 M ratio)
SPC:Cholesterol:PTX (92:5:3 M ratio)
SPC:CAE (85:15 M ratio)
SPC:CAE:PTX (77.5:15:5 M ratio)
Fig. 7 Small Angle Neutron Scattering Study: (a) SPC liposomes (b)
SPC:PTX (97:3) liposomes (c) SPC:Cholesterol (95:5) liposomes (d)
SPC:Cholesterol:PTX (92:5:3) liposomes (e) SPC:CAE (85:15) liposomes
and (f) SPC:CAE:PTX (77.5:15:5) liposomes.

90
Page 12 of 17
Monpara, Kanthou, Tozer and Vavia (2018) 35:90
Fig. 8 Cytotoxicity of PTX solution, conventional liposomes and cationic liposomes after 72 h on (a) MDA-MB 231 cells, (b) H5V cells (c) HDMEC cells. The
lines represent non-linear curve fits for each sample. The error bars represent the mean SD of each value (n = 3).
the ΔH decreased markedly with slight increase in ΔT1/2
(Table II). The changes in ΔH of main transition depends upon
the localization of the molecule within the hydrophobic interior
of the membrane rather than the vicinity of the polar head
group. In ternary system containing PTX, Cholesterol and
DPPC, ΔH was reduced significantly compared to DPPC lipo-
somes indicating perturbation of PTX and cholesterol within
the hydrophobic core of the membrane.
Addition of CAE in the DPPC liposomes produced pro-
nounced decrease in ΔH compared to similar concentrations
of cholesterol (Table II) indicating deeper penetration of CAE
molecule within the DPPC bilayer. At 15 mol% the main tran-
sition and hence enthalpy of transition were completely elimi-
nated. At 10 mol% concentration the ΔH was only 0.8 J/g
which was not detectable after addition of PTX. This implies
that PTX and CAE should be located deeper in the DPPC
bilayer which is consistent with the findings of MD simulation.
Small Angle Neutron Scattering (SANS)
Small Angle Neutron Scattering study was performed to eval-
uate the vesicular structure and bilayer thickness of the lipo-
somes. Figure 7 shows scattering profiles of liposomes made of
different compositions of SPC, PTX, cholesterol and CAE.
The liposomes were prepared in the deuterated water as it
provides a good contrast between the liposomes and the dis-
persion medium. The scattering intensity in the low-Q region
of the data decreases in a straight line with a slope of (−2) [as
1/Q2], thus indicating the formation of vesicles in these sys-
tems. As shown in the Fig. 7, the scattering curves showed
typical scattering profiles exhibited by a vesicular system
(56). Further the absence of Braggs peaks which occur in the
SANS spectra because of multilamellarity in the vesicle was
absent in all formulations which confirmed the formation of
small unilamellar vesicles.
Table IV IC50 value of PTX
Formulations on Different Cell Lines
Formulation
MDA-MB 231 cell line
H5V cell line
HDMEC cell line
Conventional liposomes
Cationic liposome
PTX solution
96.41 nM
81.76 nM
16.67 nM
93.47 nM
70.43 nM
17.47 nM
19.13 nM
10.17 nM
2.3 nM

Paclitaxel Cationic Liposomes (2018) 35:90
Page 13 of 17 90
These vesicles have been characterized by the bilayer thick-
ness as the measurement of the size of vesicles is limited by the
liposomes in terms of IC 50 value is small, cationic liposomes
may nevertheless provide higher clinical benefits compared to
conventional liposomes due to their preferential distribution
to the endothelial cells.
Qmin of the SANS instrument. The bilayer thickness obtained
from SANS results are given in Table III. In single component
bilayer, the hydrocarbon chains of the phospholipids are high-
ly interdigitated above Tg (57). The Tg of SPC is well below
the room temperature at which SANS was carried out and
hence, it was expected that the hydrocarbon chains should be
in interdigitated phase and the thickness of the bilayer should
be minimum amongst all the measured samples. The SANS
data showed that SPC liposomes had lowest bilayer thickness
amongst all the samples. The effect of cholesterol on the bi-
layer thickness has been studied extensively. It is reported that
cholesterol increases bilayer thickness by increasing lipid acyl
chains ordering in egg lecithin and bilayers consisting of sev-
eral saturated lipids (58,59). As shown in Table III, the bilayer
thickness increased after addition of cholesterol and it is be-
lieved to be due to reduction in allowed conformations of the
hydrocarbon chain at liquid crystalline state (59). Inclusion of
PTX in liposomes increased the bilayer thickness to similar
extent. This can be attributed to the area condensing effect
similar to that of cholesterol (53). CAE has a similar but more
pronounced effect on the membrane thickness. The molecular
size of CAE is larger than cholesterol and hence the observed
effect of CAE on membrane thickness is more significant than
that of cholesterol. We assume that when CAE is incorporated
in the bilayer membrane, it extends the bilayer across its hy-
drophobic chain region thereby making more space for ac-
commodation of PTX molecules within hydrophobic pockets
in between the bilayer membrane. This accommodation of
PTX along with the size of CAE has a pronounced effect on
the membrane thickness as compared to cholesterol contain-
ing liposomes which could be possible reason behind higher
drug loading and stability of CAE containing PTX liposomes.
Endothelial Cell Migration Assay
The potential of PTX formulations to inhibit migration of en-
dothelial cells was determined using the Trans-well migration
assay. Figure 9 shows the comparison of different PTX formu-
lations and blank liposomes with the control. The control group
consisted of cells without any treatment and the blank liposomes
group consisted of the liposomes without PTX. Statistical anal-
ysis was performed using GraphPad Prism 7.02 software by one-
way ANOVA. The blank liposomes showed no effect on migra-
tion ability of the endothelial cells. PTX solution at 0.1 nM
concentration did not show any significant inhibition (p =
0.0985) while at 1 nM concentration it showed very significant
inhibition (p = 0.0001). On the other hand, conventional lipo-
somes and cationic liposomes showed significant inhibition at
0.1 nM concentration having p-value 0.0142 and 0.0094, respec-
tively. At higher concentration, i.e. 1 nM, both the formulation
showed highly significant inhibition of migration (p = 0.0001).
The endothelial cell migration assay was performed at well
below the cytotoxic concentration of each formulation. The
liposomal vehicles did not show any significant migration
In-Vitro Cytotoxicity Study
The cytotoxicity of PTX in solution was compared with PTX
in conventional liposomes and cationic liposomes on three
different cell lines, MDA-MB 231, H5V and HDMEC cells.
Figure 8 shows % viability of the cell populations at different
concentrations. Table IV summarizes IC50 value of PTX
solution and PTX liposomes at 72 h incubation. The IC50
value for the solution and the liposomes was comparatively
higher in MDA-MB 231 and H5V cells than HDMEC cells,
most likely because they are transformed cells, whereas the
HDMECs are a primary cell line. The PTX solution was
more potent (showing lower IC50 values) as compared to
the liposomes in all cell lines. This might be attributed to the
slower release of PTX from the liposomal formulation (17).
Cationic liposomes showed slightly lower IC50 values com-
pared to the values for conventional liposomes in all cell lines.
Though the difference in conventional liposomes and cationic
Fig. 9 Trans-well Migration Assay: After six hours of incubation with test
samples, the %migration of treated HDMEC cells was calculated in relation
to the untreated (control) HDMEC cells. Blank liposomes (without PTX) did
not inhibit the migration of cells. At 0.1 nM PTX concentration, the liposomes
and the solution inhibited the cell migration to similar extent. At 1 nm PTX
concentration, the solution had slightly higher inhibition potential compared to
the liposomes. There was no significant difference in % migration between
the solution and the liposomes or between the liposomes. The error bars
represent the mean SD of each value (n = 3). (* represents p < 0.05, **
represents p < 0.01, **** represents p < 0.0001).

90
Page 14 of 17
Monpara, Kanthou, Tozer and Vavia (2018) 35:90
inhibition even at the higher concentration used, which indicates
that the migration inhibition obtained with PTX encapsulated
liposomes was due to the drug only. The liposomal formulation
showed significant inhibition at the lower concentration used,
which was unexpected, as PTX solution did not show significant
inhibition at this concentration compared to the control.
However, when analyzed using one-way ANOVA, there was
no significant difference between PTX solution and liposome
formulation at low concentration. All the formulations could
significantly inhibit the migration of endothelial cells at the
higher concentration used, which was not cytotoxic to the cells.
These results suggest that PTX encapsulated in liposomes main-
tained efficacy as an inhibitor of endothelial migration.
compound cholesteryl arginine ethyl ester (CAE) and
so to establish its safety for pharmaceutical applications.
HDMEC and MDA-MB 231 cells were exposed to two
levels of CAE; with 5 μg/ml corresponding to the
amount needed to deliver 175 mg/m2 dose of PTX
and with 50 μg/ml corresponding to ten times the
amount needed for drug delivery (60). The treated cells
were compared with control untreated cells. Hydrogen peroxide
(200 nM) treatment for 20 min was used as a positive control.
Figure 10 (a to h) show representative images of HDMEC and
MDA MB-231 cell nuclei stained using SYBR® Gold dye after
electrophoresis to separate intact DNA from damaged frag-
ments. The positive control showed the migration of broken
strands of DNA forming a tail appearing like a comet
(Fig. 10b, f). The negative control group showed cells with a
circular appearance (Fig. 10a, e). None of the treatments
showed the formation of a tail and the cells appeared circular,
like untreated control cells (Fig. 10c, d, g, h).
Alkaline Comet Assay
The Alkaline Comet assay was performed to determine
whether there was any genotoxicity of the synthesized
Fig. 10 Comet Assay and Comet Score: Fluorescence Microscopy Image (20X objective) of HDMEC cells [(a) Negative Control (Untreated) (b) Positive
Control (treated with 1 ml, 200 μM H2O2 for 20 min (c) CAE (5 μg/ml) treated cells (d) CAE (50 μg/ml) treated cells] and MDA-MB231 cells [(e) Negative
Control (Untreated) (f) Positive Control (treated with 1 ml, 200 μM H2O2 for 20 min (g) CAE (5 μg/ml) treated cells (h) CAE (50 μg/ml) treated cells]. Comet
score: In HDMEC) and MDA-MB231 cells) CAE did not cause any significant differences in tail moment and olive moment compared to the untreated (negative
control) cells at 5 μg/ml or 50 μg/ml concentration of CAE. The 5 μg/ml concentration of CAE corresponds to the amount needed to deliver a clinical dose PTX in
cationic liposomes. The 50 μg/ml concentration was also tested to ensure safety of CAE. The data are expressed as mean SD (n = 3). The triple asterisks
indicate p < 0.001.

Paclitaxel Cationic Liposomes (2018) 35:90
Page 15 of 17 90
The results were analyzed using one-way ANOVA with
GraphPad Prism 7.02 software. As shown in Fig. 10, both con-
centrations of CAE tested (5 μg/ml, 50 μg/ml) did not cause
any significant difference in the tail moment or olive moment in
either HDMEC or MDA-MB 231 cells. This initial analysis by
the comet assay indicated that the synthesized compound does
not have any genotoxicity. With the help of other techniques,
exhaustive safety profile can be generated for establishing the
use of the compound in pharmaceutical applications.
in-vitro cell studies. Authors are thankful to University Grant
Commission (UGC), Government of India, for financial assis-
tance and AICTE-NAFETIC for providing facilities to per-
form the experimental work. The authors are thankful to Dr.
V. K. Aswal and Dr. Debes Ray at the Bhabha Atomic
Research Centre, Mumbia, India for help with carrying out
the Small Angle Neutron Scattering Study. The authors are
thankful to Dr. Spencer Collis and Ms. Caroline Daye, the
department of Oncology and Metabolism, The Medical
School, The University of Sheffield for help with carrying
out the Comet Assay. Authors declare no conflict of interest.
CONCLUSION
It can be concluded that CAE can be used as a membrane
stabilizing lipid for the preparation of PTX loaded cationic
liposomes. The MD simulation of CAE and PTX incorporat-
ed liposomes revealed the possibility of hydrogen bonding
between PTX and CAE. In addition to hydrogen bonding,
PTX could penetrate deeper in the liposome bilayer, when
CAE was used. These factors might be responsible for in-
creased loading and stability of PTX within the bilayer. The
DSC results showed that inclusion of CAE in the liposome
membrane eliminated Tg with increasing concentration which
could result in increased rigidity of the membrane and hence
improved stability. The SANS study showed that inclusion of
CAE has a similar effect on the increase in thickness of the
liposome membrane as compared to cholesterol. However,
the effect is more pronounced than cholesterol suggesting dif-
ference in the location and alignment within the membrane.
In addition to stability, CAE also provides positive charge to
the liposomes in all biological conditions due to guanidinium
groups and thus may provide a further advantage over tertiary
or quaternary ammonium group containing lipids. The cat-
ionic liposomes prepared with CAE and PTX have slightly
improved cytotoxicity and ability to inhibit endothelial migra-
tion compared to conventional liposomes. The metabolism of
CAE should result in formation of cholesterol and arginine
through breakdown of carbamate bond and both the compo-
nent should follow normal metabolic pathways in a biological
system. Thus, CAE can be considered as a biocompatible
cationic ligand. The absence of genotoxicity supports this hy-
pothesis. The in-vitro results presented here are encouraging
and supportive of using CAE as a cationic ligand. Further in-
vivo studies are now needed to establish potential utility of
CAE in pharmaceutical preparations.
REFERENCES
1. Surapaneni MS, Das SK, Das NG. Designing paclitaxel drug de-
livery systems aimed at improved patient outcomes: current status
and challenges. ISRN pharmacology. 2012;2012:1–15.
2. Singla AK, Garg A, Aggarwal D. Paclitaxel and its formulations.
Int J Pharm. 2002;235(1–2):179–92.
3. Gelderblom H, Verweij J, Nooter K, Sparreboom A, Cremophor
EL. The drawbacks and advantages of vehicle selection for drug
formulation. Eur J Cancer. 2001;37(13):1590–8.
4. US Food and Drug Administration. Drugs@ FDA: FDA approved
drug products. 2017. Available from: https://www.accessdata.fda.
gov/scripts/cder/daf.html.
5. Rizvi NA, Riely GJ, Azzoli CG, Miller VA, Ng KK, Fiore J, et al.
Phase I/II trial of weekly intravenous 130-nm albumin-bound pac-
litaxel as initial chemotherapy in patients with stage IV non–small-
cell lung cancer. J Clin Oncol. 2008;26(4):639–43.
6. Ruttala HB, Ko YT. Liposome encapsulated albumin-paclitaxel
nanoparticle for enhanced antitumor efficacy. Pharm Res.
2015;32(3):1002–16.
7. Bernabeu E, Helguera G, Legaspi MJ, Gonzalez L, Hocht C, Taira
C, et al. Paclitaxel-loaded PCL–TPGS nanoparticles: in vitro and
in vivo performance compared with Abraxane®. Colloids Surf B:
Biointerfaces. 2014;113:43–50.
8. Yoshizawa Y, Kono Y, K-i O, Kimura T, Higaki K. PEG
liposomalization of paclitaxel improved its in vivo disposition and
anti-tumor efficacy. Int J Pharm. 2011;412(1–2):132–41.
9. Koudelka Š, Liposomal TJ. Paclitaxel formulations. J Control
Release. 2012;163(3):322–34.
10. Slingerland M, Guchelaar HJ, Rosing H, Scheulen ME, van
Warmerdam LJ, Beijnen JH, et al. Bioequivalence of liposome-
entrapped paclitaxel easy-to-use (LEP-ETU) formulation and pac-
litaxel in polyethoxylated castor oil: a randomized, two-period
crossover study in patients with advanced cancer. Clin Ther.
2013;35(12):1946–54.
11. clinicaltrials.gov. A trial evaluating the efficacy and safety of
EndoTAG®-1 in combination with paclitaxel and gemcitabine
compared with paclitaxel and gemcitabine as first-line therapy in
patients with visceral metastatic triple-negative breast cancer.
https://clinicaltrials.gov/ct2/show/NCT03002103, 2016.
12. Campbell RB, Fukumura D, Brown EB, Mazzola LM, Izumi Y,
Jain RK, et al. Charge determines the distribution of liposomes
between the vascular and extravascular compartments of tumors.
Cancer Res. 2002;62(23):6831–6.
ACKNOWLEDGMENTS AND DISCLOSURES
The authors are extremely thankful to The British Council,
UK and Department of Biotechnology, India for providing a
fellowship under the Newton-Bhabha Ph.D. placement pro-
gram 2015–16 for Mr. Jasmin Monpara, which facilitated the
13. Lila ASA, Ishida T, Kiwada H. Targeting anticancer drugs to tu-
mor vasculature using cationic liposomes. Pharm Res. 2010;27(7):
1171–83.

90
Page 16 of 17
Monpara, Kanthou, Tozer and Vavia (2018) 35:90
14. Thurston G, McLean JW, Rizen M, Baluk P, Haskell A, Murphy
TJ, et al. Liposomes target angiogenic endothelial cells in tumors
and chronic inflammation in mice. J Clin Invest. 1998;101(7):1401–
13.
15. Schmitt-Sody M, Strieth S, Krasnici S, Sauer B, Schulze B, Teifel
M, et al. Targeting therapy: paclitaxel encapsulated in cationic
liposomes improves antitumoral efficacy. Clin Cancer Res.
2003;9(6):2335–41.
16. Bocci G, Di Paolo A, Danesi R. The pharmacological bases of the
antiangiogenic activity of paclitaxel. Angiogenesis. 2013;16(3):481–
92.
17. Yang T, Cui F-D, Choi M-K, Cho J-W, Chung S-J, Shim C-K,
et al. Enhanced solubility and stability of PEGylated liposomal pac-
litaxel: in vitro and in vivo evaluation. Int J Pharm. 2007;338(1–2):
317–26.
18. Sofias AM, Dunne M, Storm G, Allen C. The battle of BNano^
paclitaxel. Adv Drug Deliv Rev. 2017;122:20–30.
19. Huynh L, Grant J, Leroux J-C, Delmas P, Allen C. Predicting the
solubility of the anti-cancer agent docetaxel in small molecule ex-
cipients using computational methods. Pharm Res. 2008;25(1):
147–57.
20. Xiang T-X, Anderson BD. Liposomal drug transport: a molecular
perspective from molecular dynamics simulations in lipid bilayers.
Adv Drug Deliv Rev. 2006;58(12–13):1357–78.
21. Stepniewski M, Pasenkiewicz-Gierula M, Róg T, Danne R,
Orlowski A, Karttunen M, et al. Study of PEGylated lipid layers
as a model for PEGylated liposome surfaces: molecular dynamics
simulation and Langmuir monolayer studies. Langmuir.
2011;27(12):7788–98.
22. Kang M, Loverde SM. Molecular simulation of the concentration-
dependent interaction of hydrophobic drugs with model cellular
membranes. J Phys Chem B. 2014;118(41):11965–72.
23. Jämbeck JP, Eriksson ES, Laaksonen A, Lyubartsev AP, Eriksson
LA. Molecular dynamics studies of liposomes as carriers for
photosensitizing drugs: development, validation, and simulations
with a coarse-grained model. J Chem Theory Comput.
2013;10(1):5–13.
24. Hristova K, Wimley WC. A look at arginine in membranes. J
Membr Biol. 2011;239(1–2):49–56.
25. Wang Z. Schotten-Baumann Reaction. In: Comprehensive
Organic Name Reactions and Reagents. John Wiley & Sons,
Inc.; 2010. p. 2536–2539.
33. Chen S-H, Lin T-L. 16. Colloidal solutions. In: Price DL, Sköld K,
editors. Methods in experimental physics: Elsevier; 1987. p. 489–
543.
34. Pedersen JS. Analysis of small-angle scattering data from colloids
and polymer solutions: modeling and least-squares fitting. Adv
Colloid Interf Sci. 1997;70:171–210.
335. Pedersen JS, Riekel C. Resolution function and flux at the sample
for small-angle X-ray scattering calculated in position–angle–
wavelength space. J Appl Crystallogr. 1991;24(5):893–909.
36. Bevington PR, Robinson DK, Blair JM, Mallinckrodt AJ, McKay
S. Data reduction and error analysis for the physical sciences.
Comput Phys. 1993;7(4):415–6.
37. Valster A, Tran NL, Nakada M, Berens ME, Chan AY, Symons M.
Cell migration and invasion assays. Methods. 2005;37(2):208–15.
38. Lovelock J, Bishop M. Prevention of freezing damage to living cells
by dimethyl sulphoxide. Nature. 1959;183(4672):1394–5.
39. Instruments P. http://www.perceptive.co.uk/products/comet-
assay-iv/.
40. Li Y-C, Rissanen S, Stepniewski M, Cramariuc O, Róg T, Mirza S,
et al. Study of interaction between PEG carrier and three relevant
drug molecules: piroxicam, paclitaxel, and hematoporphyrin. J
Phys Chem B. 2012;116(24):7334–41.
41. Vander Velde DG, Georg GI, Grunewald GL, Gunn CW,
Mitscher LA. " hydrophobic collapse" of taxol and Taxotere solu-
tion conformations in mixtures of water and organic solvent. J Am
Chem Soc. 1993;115(24):11650–1.
42. Stanton DT, Mattioni BE, Knittel JJ, Jurs PC. Development and
use of hydrophobic surface area (hsa) descriptors for computer-
assisted quantitative structure− activity and structure− property
relationship studies. J Chem Inf Comput Sci. 2004;44(3):1010–23.
43. Balasubramanian SV, Straubinger RM. Taxol-lipid interactions:
taxol-dependent effects on the physical properties of model mem-
branes. Biochemistry. 1994;33(30):8941–7.
44. Belsito S, Bartucci R, Sportelli L. Paclitaxel interaction with phos-
pholipid bilayers: high-sensitivity differential scanning calorimetric
study. Thermochim Acta. 2005;427(1–2):175–80.
45. Lian T, Ho RJ. Trends and developments in liposome drug delivery
systems. J Pharm Sci. 2001;90(6):667–80.
46. Zhao L, Feng S-S. Effects of cholesterol component on molecular
interactions between paclitaxel and phospholipid within the lipid
monolayer at the air–water interface. J Colloid Interface Sci.
2006;300(1):314–26.
26. Gao X, Huang L. A novel cationic liposome reagent for efficient
transfection of mammalian cells. Biochem Biophys Res Commun.
1991;179(1):280–5.
47. Demetzos C. Differential scanning calorimetry (DSC): a tool to
study the thermal behavior of lipid bilayers and liposomal stability.
Journal of liposome research. 2008;18(3):159–73.
27. Garlanda C, Parravicini C, Sironi M, De Rossi M, De Calmanovici
RW, Carozzi F, et al. Progressive growth in immunodeficient mice
and host cell recruitment by mouse endothelial cells transformed by
polyoma middle-sized T antigen: implications for the pathogenesis
of opportunistic vascular tumors. Proc Natl Acad Sci. 1994;91(15):
7291–5.
28. Zhang JA, Anyarambhatla G, Ma L, Ugwu S, Xuan T, Sardone T,
et al. Development and characterization of a novel Cremophor®
EL free liposome-based paclitaxel (LEP-ETU) formulation. Eur J
Pharm Biopharm. 2005;59(1):177–87.
29. Chen D-B, T-z Y, Lu W-L, ZHANG Q. In vitro and in vivo study of
two types of long-circulating solid lipid nanoparticles containing
paclitaxel. Chem Pharm Bull. 2001;49(11):1444–7.
30. Aswal V, Small-angle GP. Neutron scattering diffractometer at
Dhruva reactor. Curr Sci. 2000;79(7):947–53.
48. Ladbrooke B, Chapman D. Thermal analysis of lipids, proteins and
biological membranes a review and summary of some recent stud-
ies. Chem Phys Lipids. 1969;3(4):304–56.
49. Taylor KM, Morris RM. Thermal analysis of phase transition be-
haviour in liposomes. Thermochim Acta. 1995;248:289–301.
50. Bernsdorff C, Reszka R, Winter R. Interaction of the anticancer
agent Taxol™(paclitaxel) with phospholipid bilayers. J Biomed
Mater Res. 1999;46(2):141–9.
51. Zhao L, Feng S-S, Kocherginsky N, Kostetski I. DSC and EPR
investigations on effects of cholesterol component on molecular
interactions between paclitaxel and phospholipid within lipid bilay-
er membrane. Int J Pharm. 2007;338(1–2):258–66.
52. McMullen TP, McElhaney RN. New aspects of the interaction of
cholesterol with dipalmitoylphosphatidylcholine bilayers as re-
vealed by high-sensitivity differential scanning calorimetry.
Biochimica et Biophysica Acta (BBA)-Biomembranes.
1995;1234(1):90–8.
31. Hayter J, Penfold J. Determination of micelle structure and charge
by neutron small-angle scattering. Colloid Polym Sci. 1983;261(12):
1022–30.
32. Kaler E. Small-angle scattering from colloidal dispersions. J Appl
Crystallogr. 1988;21(6):729–36.
53. Zhao L, Feng S-S. Effects of lipid chain length on molecular inter-
actions between paclitaxel and phospholipid within model
biomembranes. J Colloid Interface Sci. 2004;274(1):55–68.

Paclitaxel Cationic Liposomes (2018) 35:90
Page 17 of 17 90
54. Ladbrooke B, Williams RM, Chapman D. Studies on lecithin-
cholesterol-water interactions by differential scanning calorimetry
and X-ray diffraction. Biochimica et Biophysica Acta (BBA)-
Biomembranes. 1968;150(3):333–40.
55. Estep T, Mountcastle D, Biltonen R, Thompson T. Studies on the
anomalous thermotropic behavior of aqueous dispersions of
dipalmitoylphosphatidylcholine-cholesterol mixtures.
Biochemistry. 1978;17(10):1984–9.
57. Braganza LF, Worcester DL. Hydrostatic pressure induces hydro-
carbon chain interdigitation in single-component phospholipid bi-
layers. Biochemistry. 1986;25(9):2591–6.
58. Levine Y, Wilkins M. Structure of oriented lipid bilayers. Nature
New Biology. 1971;230(11):69–72.
59. McIntosh TJ. The effect of cholesterol on the structure of phospha-
tidylcholine bilayers. Biochimica et Biophysica Acta (BBA)-
Biomembranes. 1978;513(1):43–58.
60. Squibb B-M. Taxol®(paclitaxel) injection package insert. 2000. In:
2008. US Food and Drug Administration. Drugs@ FDA: FDA
approved drug products. 2017. Available from: https://www.
accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.
process&ApplNo=020262.html.
56. Yue B, Huang C-Y, Nieh M-P, Glinka CJ, Katsaras J. Highly stable
phospholipid unilamellar vesicles from spontaneous vesiculation: a
DLS and SANS study. J Phys Chem B. 2005;109(1):609–16.