Content uploaded by Timo ten Hagen
Author content
All content in this area was uploaded by Timo ten Hagen on Nov 10, 2021
Content may be subject to copyright.
Cetuximab-oxaliplatin-liposomes for epidermal growth factor receptor
targeted chemotherapy of colorectal cancer
Sara Zalba
a,b
, Ana M. Contreras
a
, Azadeh Haeri
b
, Timo L.M. ten Hagen
b
, Iñigo Navarro
c
,
Gerben Koning
b,
⁎, María J. Garrido
a,
⁎
a
Department of Pharmacy and Pharmaceutical Technology, University of Navarra, 31008 Pamplona, Spain
b
Laboratory Experimental Surgical Oncology, Department of Surgery, Erasmus MC Cancer Institute, Rotterdam 3015 GE, The Netherland
c
Department of Chemistry and Edaphology, University of Navarra, 31008 Pamplona, Spain
abstractarticle info
Article history:
Received 23 February 2015
Received in revised form 8 May 2015
Accepted 11 May 2015
Available online 19 May 2015
Chemical compounds studied in this article:
Oxaliplatin (PubChem CID: 5310940)
3-(2-Pyridyldithio)propionic-acid-N-
hydroxysuccinimide-ester (PubChem CID:
100682)
Tris(2-carboxyethyl)-phosphine-hydrochloride
(PubChem CID: 2734570)
1,1′-Dioctadecyl-3,3,3′,3′-
tetramethylindocarbocyanine perchlorate
(PubChem CID: 16212735)
β-Mercaptoethanol (PubChem CID: 1567)
Hydrogenatedphosphatidylcholine (PubChem
CID: 94190)
1,2-distearoyl-sn-glycero-3-
phosphoethanolamine-N-[Methoxy
(polyethylenglycol)-2000] (PubChem CID:
406952)
Cholesterol (PubChem CID:5997)
1,2-distearoyl-sn-glycero-3-
phosphoethanolamine-N-[Maleimide
(polyethylene glycol) 2000] (PubChem
CID:406950)
Keywords:
EGFR
Targeted liposomes
Oxaliplatin
Cetuximab
Fab′
Colorectal cancer
Oxaliplatin (L-OH), a platinum derivative with good tolerability is currently combined with Cetuximab (CTX), a
monoclonal antibody (mAb), for the treatment of certain (wild-type KRAS) metastatic colorectal cancer (CRC)
expressing epidermal growth factor receptor (EGFR).
Improvement of L-OH pharmacokinetics (PK) can be provided by its encapsulation into liposomes, allowing a
more selective accumulation and delivery to the tumor. Here, we aim to associate both agents in a novel liposo-
mal targeted therapy by linking CTX to the drug-loaded liposomes. These EGFR-targeted liposomes potentially
combine the therapeutic activity and selectivity of CTX with tumor-cell delivery of L-OH in a single therapeutic
approach.
L-OH liposomes carrying whole CTX or CTX-Fab′fragments on their surface were designed and characterized.
Their functionality was tested in vitro using four human CRC cell lines, expressing different levels of EGFR to
investigate the role of CTX-EGFR interactions in the cellular binding and uptake of the nanocarriers and encapsu-
lated drug. Next, those formulations were evaluatedin vivo in a colorectalcancer xenograft model with regard to
tumor drug accumulation, toxicity and therapeutic activity.
In EGFR-overexpressing cell lines, intracellular drug delivery by targeted liposomes increased with receptor
density reaching up to 3-fold higher levels than with non-targeted liposomes. Receptor specific uptake was
demonstrated by competition with free CTX, which reduced internalization to levels similar to non-targeted
liposomes.In a CRC xenograft model, drug delivery was strongly enhanced upon treatment with targetedformu-
lations. Liposomes conjugated with monovalent CTX-Fab′fragments showed superior drug accumulation in
tumor tissue (2916.0 ± 507.84 ng/g) compared to CTXliposomes (1546.02 ± 362.41 ng/g) or non-targeted lipo-
somes (891.06 ± 155.1 ng/g). Concomitantly, CTX-Fab′targeted L-OH liposomes outperformed CTX-liposomes,
which on its turn was still more efficacious than non-targeted liposomes and free drug treatment in CRC bearing
mice.
These results show thatsite-directed conjugationof monovalent CTX-Fab′provides targetedL-OH liposomes that
display an increased tumor drug delivery and efficacy over a formulation with CTXand non-targeted liposomes.
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
Platinum derivatives are widely used in cancer chemotherapy. They
are used in the chemotherapeutic treatment of approximately 70% of
solid tumors, including non-small andsmall cell lung, breast, colorectal,
gastric, pancreatic, esophageal, testicular, cervical, ovarian cancers and
Journal of Controlled Release 210 (2015) 26–38
⁎Corresponding authors.
E-mail addresses: g.koning@erasmusmc.nl (G. Koning), mgarrido@unav.es
(M.J. Garrido).
http://dx.doi.org/10.1016/j.jconrel.2015.05.271
0168-3659/© 2015 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Journal of Controlled Release
journal homepage: www.elsevier.com/locate/jconrel
non-Hodgkin's lymphoma [1,2]. L-OH is a third generation platinum
derivative used in first line treatment of colorectal cancer (CRC) [3].
It is one of the platinum derivatives with lower toxicity in patients
[4]. However, L-OH pharmacokinetics is characterized by a high
volume of distribution due to its rapid and irreversible binding to
erythrocytes and to plasma and tissue proteins. This causes serious
limitations to reach desirable unbound drug concentrations in
the target tissue as tumor and subsequent therapeutic efficacy [4].To
overcome this, novel strategies aim at modifying L-OH biodistribution,
for instance by drug encapsulation in nanocarriers, particularly in
liposomes [5].
Liposomal encapsulation causes a strong reduction in the volume
of distribution of the loaded drug, a decrease in drug toxicity,
prolonged presence of drug in circulation and enhanced tumor accu-
mulation [6].
Two important aspects of liposomes that contribute significantly to
tumor accumulation of entrapped drugs are their small size (usually
around 100 nm) and surface pegylation. Both contribute to increase
the circulation half-life of this formulation in blood stream delaying
its caption by the reticulo-endothelial system and thereby, increasing
the probability to reach the tumor [7]. The latter is promoted by the
enhanced permeability and retention effect (EPR) in tumors, caused
by an increased vascular permeability in this tissue and a lack of
lymphatic flow [8], allowing preferential nanoparticle extravasation
and retention at the tumor site.
This selective retention of nanoparticles, together with a specific
intracellular drug delivery into tumor cells, can be promoted by cell-
specific targeting [9]. For this approach, one needs the presence of a spe-
cific molecule or ligand in the surface of the liposome able to recognize a
receptor or antigen on the cancer cell, promoting the internalization of
the complex to reach the intracellular targets. Different types of ligands
are in use for nanoparticle targeting, such as vitamins, carbohydrates,
growth factors, peptides, and monoclonal antibodies [10,11].Incurrent
cancer therapy, the use of targeted agents is increasing; for example,
growth factor receptor inhibitors, molecules that regulate cell-signaling
and gene expression or inhibitors of angiogenesis [12,13].
In recent years, L-OH combined with Cetuximab (CTX), a chimeric
human-mouse antibody specific for the epidermal growth factor receptor,
is the chemotherapeutic treatment of certain (wild-type KRAS) metasta-
tic CRC expressing EGFR [14,15].
Binding of CTX to EGFR leads to the blockade of the receptor
followed by a rapid internalization, causing not only a down regula-
tion of receptors at the surface of the cells, but also an inhibition
of EGFR, cell proliferation signals triggering of apoptosis and the
activation of the immune system to destroy tumor cells [16]. This
receptor is over-expressed in many solid tumors including 65–70%
of CRC [17,18].
In this study, we aim to combine L-OH chemotherapy and CTX
targeted therapy of CRC by designing a L-OH-loaded liposomal formula-
tion targeted to EGFR on CRC cells. Such an approach combines the L-OH
drug delivery advantages of PEG-liposomal encapsulation, with tumor
cell-specific delivery by CTX on the liposomal surface.
Linking whole mAb usually results in random orientation of this
molecule on the liposomal surface, which limits antigen or receptor
binding site exposure and also yields exposure of Fc′domains. The latter
can promote recognition by Fc′-receptors on macrophages and faster
liposome clearance from circulation as well as, induce unwanted
immune reactions [19]. To overcome this, smaller targeting molecules
able to recognize the receptor are coupled but they do not display the
unwanted immune activity [20]. Therefore, CTX-Fab′fragments, as a se-
lective ligand for EGFR, were coupled site-specifically to the liposome.
CTX-mAb and CTX-Fab′L-OH-liposomes were prepared and character-
ized. Of these formulations, we investigated binding and uptake by
human CRC cell lines and intracellular drug delivery. Next, we studied
in vivo drug delivery to the tumor and therapeutic activity in CRC
xenograft bearing mice.
2. Materials and methods
2.1. Drugs and chemicals
3-(2-Pyridyldithio)propionic-acid-N-hydroxysuccinimide-ester
(SPDP), Tris(2-carboxyethyl)-phosphine–hydrochloride (TCEP),1,1′-
Dioctadecyl-3,3,3′,3′-tetramethyl-indocarbocyanine perchlorate (Dil),
Vivaspin tubes (300,000 MWCO) and β-Mercaptoethanol were
purchased from Sigma (Barcelona, Spain). The mAb CTX (Erbitux®)
and L-OH (Eloxatin®) were provided by the University Clinic of
Navarre (Pamplona, Spain). Hydrogenated-phosphatidylcholine
(HSPC), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[Methoxy
(polyethylenglycol)-2000] (DSPE-PEG2000) and 1,2-distearoyl-sn-
glycero-3-phosphoethanolamine-N-[Maleimide (polyethylene glycol)
2000] (DSPE-PEG2000-Mal), werefromAvantiPolarLipidsInc.
(Alabaster, Alabama, USA) and Lipoid (Ludwigshafen, Germany)
and were kindly provided by the Laboratory Experimental Surgical
Oncology from the Erasmus MC in Rotterdam (The Netherlands).
Cholesterol (CH) and bovine serum albumin (BSA) were purchased
from Sigma (Barcelona, Spain). Amicon tubes with membranes
of 10,000, 30,000 and 50,000 MWCO were obtained from Millipore
Corporation (Billerica, MA, USA).
2.2. Oxaliplatin liposomes
Liposomes of L-OH were developed using the film hydration tech-
nique following the methodology previously described by Zalba et al.
[21]. Briefly, the lipids HSPC:CH:DSPE-PEG
2000
:DSPE-PEG
2000
-Mal, in
a molar ratio of 1.85:1:0.12:0.03, were dissolved in a solution of
chloroform:methanol [9:1 (v/v)]. The solvent was evaporated using a
rotary evaporator (Büchi-R144, Switzerland) at 65 °C to obtain a lipid
film, which was further dried under vacuum. The film was hydrated
with a solution of L-OH (5 mg/ml) in glucose 5%, resulting a final solu-
tion of 10 mM of lipids. The same procedure was followed to prepare
the non-loaded or blank liposomes but adding glucose 5%. The solution
was extruded through a 100 nm polycarbonate membrane (Mini-
Extruder Set, Avanti Polar Lipids Inc., Alabaster, Alabama,USA) to obtain
a homogeneous liposome population (LP-N).
Non-encapsulated L-OH was removed by ultrafiltration at 2200 g for
60 min using the Amicon system (10,000 MWCO). Liposomes were
washed three times with Hepes saline solution (pH 6.7) and the final
formulation was stored at 4 °C until use.
In order to prepare labeled liposomes, the fluorescent probe Dil was
added to the lipid mixture [1% (w/w)] and dissolved in the solution of
chloroform:methanol [9:1 (v/v)] together with lipids. The Dil-LP-N
preparation procedure was similar to non-fluorescent liposomes.
These liposomes were also used to formulate the EGFR targeted lipo-
somes with CTX and Fab′fragment.
2.3. Development of immunoliposomes
2.3.1. Liposomes coupled to Cetuximab
The method followed to develop the immunoliposomes has been
previously described by Songs and coworkers [22].Briefly, CTX was
mixed by orbital shaking with SPDP at 1:12 molar ratio for 1 h at
room temperature (RT). Afterwards the excess of SPDP was removed
from the mixture by ultrafiltration at 2200 g for 30 min using the
Amicon system (50,000 MWCO). The ultrafiltered solution was incubated
under agitation for 1 h with TCEP (1:625 molar ratio) at RT [23].These
two reactions provided thiolation of CTX, which was next co-incubated
with liposomes overnight at 10 °C. The thiol groups behave as crosslinker
with the maleimide groups of the DSPE-PEG
2000
-Mal lipid, forming a
thioether bond, as is shown in Fig. 1.Thefinal formulation was purified
by ultrafiltration at 2200 g for 60 min using the Vivaspin system
(300,000 MWCO), followed by a washout process with Hepes saline
to remove the non-bound ligand. To prevent the particle aggregation,
27S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
liposomes were incubated with 1 mM of L-cysteine that quenches the free
maleimide radicals avoiding the formation of disulfide bond [24]. The
immunoliposome (LP-CTX) was stored at 4 °C until use.
2.3.2. Liposomes coupled to Fab′fragment of Cetuximab
To obtain the Fab′fragment from CTX, several steps were followed.
First, the mAb was hydrolized using a pepsin solution (1:20 w/w) pre-
pared in sodium acetate (100 mM, pH 3.7). This process was carried
out at 37 °C for 2 h obtaining the (Fab′)
2
and the crystallizable fragment,
Fc′.ThisFc′was removed by ultrafiltration at 2200 g for 30 min using
the Amicon system (50,000 MWCO). Afterwards the collected (Fab′)
2
fragments were incubated at 37 °C for 30 min with a solution of
15 mM of β-Mercaptoethanol to obtain single Fab′fragments, as is
shown in Fig. 2. These molecules were purified by several cycles of ultra -
filtration (30,000 MWCO) and washing. To check the isolation process
of Fab′fragments and to verify the presence of the different molecules
in each of the steps mentioned (whole mAb, the non-reduced (Fab′)
2
fragments and the Fab′fragment) an 8% gel SDS page was used.
The induced hinge-region free thiol group in the structure of Fab′
allowed the conjugation with DSPE-PEG
2000
-Mal of the liposome by
incubation at 10 °C overnight. The elimination of the non-coupled
fragments was carried out by three cycles of ultrafiltration in Amicon
tubes (50,000 MWCO) and washing with Hepes saline solution. To
prevent the instability of the formulation (LP-Fab′), it was incubated
with 1 mM of L-cysteine and stored at 4 °C until use [24].
2.3.3. Optimization of the ligand/lipid ratio
In order to select the most adequate ligand to lipid ratio in eachtype
of formulation, different quantities of ligand per μmol of lipid (5, 10, 20,
30, 40 and 50 μg) were tested for both ligands, CTX and Fab′.Inthefinal
NH2
SPDP
TCEP
Pyridine-
2-thione
+
Maleimide
LP-CTX
Thioethe r bond
+
Fig. 1. Schematic representation of the method used for LP-CTX production.
CTX
Maleimide
Thioether bond
pepsin
(Fab’)
2
fragment
β- Mercaptoethanol
+
Fab’ fragments
LP-Fab’
Fig. 2. Schematic representation of the method used for LP-Fab′production.
28 S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
formulation, the efficiency of the coupling was measured using the
MicroBCA™kit (Thermo Fisher Scientific Inc, Waltham, MA USA). The
effiency was calculated as the relation between the initial and the final
amount of the ligand incorporated in the liposomes.
In a parallel study, fluorescent liposomes (Dil-LP), previously formu-
lated with different ligand/lipid ratios, were assayed for cell uptake in
HCT-116 cell line. This human colorectal cancer cell line was purchased
from ATCC and maintained at 37 °C, 5% CO
2
in RPMI containing 10% (v/v)
fetal bovine serum and 1% (v/v) of Penicillin–Streptomycin.
For this study, cells were harvested by trypsinization, counted
and seeded in 96-well black plates at a density of 15 × 10
3
cells/well.
One day later, the culture medium was replaced by a fresh medium con-
taining a lipid concentration of 100 μM/well of each type of fluorescent
formulations, Dil-LP-CTX and Dil-LP-Fab′. After 24 h of treatment, the
plates were washed twice with PBS and the fluorescence was measured
(Tecan group Ltd, Maennedorf, Switzerland). The signal corresponding
to the internalization with the different formulations was compared to
the control basal signal.
2.4. Characterization of liposomes
The particle size and zeta potential of the formulations were ana-
lyzed by laser diffractometry using a Zetasizer Nano Series (Malvern
Instruments, UK). Formulations were diluted 1:50 (v/v) in deionized
water in order to ensure a convenient scattered intensity on the
detector. The efficiency of L-OH encapsulation was measured by atomic
absorption spectrometry. The concentration of the lipid in each formu-
lation was quantified using the phosphate assay method [25].
The efficiency of conjugation for the two ligands was quantified
using the MicroBCA™kit following the manufacturing instructions.
2.5. Evaluation of L-OH release from the liposomal formulations
The release rate of L-OH from liposomes was characterized for the
three types of formulations. Aliquots of 1 mL corresponding to 60 μg
of L-OH loaded in liposomes were incubated at 37 °C in a dialysis
tube (12,000 MWCO). The receptor medium consisted of 500 mL of
isoosmotic Hepes saline solution under orbital shaking to maintain sink
conditions during the experiment [26]. Several samples of the receptor
medium were collected at 0, 10, 30, 60 min and 4, 7 and 24 h by triplicate
and the L-OH quantified by atomic absorption spectrometry.
The influence of serum proteins in the stability of liposomes was
assayed by their incubation in presence of 50% FBS (v/v).
The accumulative release curve was built as the percentage of
release (% R) at each assayed time applying the following formula:
%R¼Qa
Qt
100
where, Qa represents the amount of L-OH measured in the collected
sample, and Qt, the total amount of the drug at the beginning of the
experiment.
2.6. In vitro studies
2.6.1. Cytotoxicity study
Four human colon cancer cell lines, HCT-116, HT-29 and SW-480,
SW-620 were purchased from ATCC and routinely maintained at stan-
dard conditions in RPMI and DMEM respectively, containing 10% (v/v)
FBS and a 1% (v/v) of Penicillin–Streptomycin.
For this study, cells were detached by trypsinization and seeded in
96 well microtiter plates at a density of 5x10
3
cells/well. Twenty-four
hours later, the cells were exposed for 4 h at different concentrations
of L-OH, (from 0.1 to 100 μM) free and encapsulated in LP-N, LP-CTX
and LP-Fab′. After drug exposure times, cells were rinsed with PBS and
new fresh medium was added. Cell viability was measured at 72 h
post-treatment using the Neutral Red Assay [27].Datawereexpressed
as concentration of L-OH that gives a 50% inhibition of cell growth
compared to untreated or control cells (IC
50
). Blank liposomes were
also tested under the same conditions.
2.6.2. Influence of liposomes in the EGFR phosphorylation status
In order to evaluate the influence of the ligands CTXand Fab′,freeor
coupled to liposomes in the EGFR phosphorylation status, all formula-
tions were assayed in HCT-116 and SW-480 cell lines.
Cells at a density of 8 × 10
5
cells/well were seeded in 6 well plates.
Twenty-four hours later, cells were exposed for 15 min at the following
treatments: 20 ng/ml of EGF, 27 μg/ml of CTX and Fab′,and1mMofLP-
N, LP-CTX and LP-Fab′. Afterwards cells, washed twice with ice-cold
PBS, were treated with 100 μl of RIPA buffer supplemented with sodium
orthovanadate (1 mM), sodium fluoride (10 mM), β-glycerophosphate
(100 mM) and Phosphatase inhibitor cocktail tablets (Roche®,
Indianapolis, USA), harvested with a cell scrapper and incubated at
4 °C for 30 min. Lysates were centrifuged at 12,500 g for 30 min at
4 °C to collect the supernatant. The content of protein in those samples
was quantified using the Micro BCA™kit. Aliquots of 40 μg of protein
were loaded on the SDS PAGE (8%) and transferred to a nitrocellulose
membrane by wet blotting. This membrane, incubated overnight at
4 °C in Tris buffer solution (TBS) with 1% Tween-20 and 5% of BSA and
the anti-EGFR [phospho Y1068 from abcam®(Cambridge, UK)], was
washed and incubated for 1 h at RT with the secondary antibody, an
anti-rabbit peroxidase-labeled (Cell-signaling, Danvers, USA). Blots
were developed using the ECL system Super Signal ULTRA kit (Thermo
Fisher Scientific Inc, Waltham, MA USA), and immunoreactive proteins
were visualized on the high-performance chemiluminescence film
(HyperfilmTM, Amersham Bioscience, Piscataway, NJ, USA). β-actin
wasusedascontrol.
2.6.3. Cellular uptake of liposomes
Cell uptake in four cell lines using fluorescent liposomal formula-
tions LP-N, LP-CTX and LP-Fab′was characterized at two temperatures,
37 and 4 °C respectively.
For this study, cells were seeded in 96 well microtiter black plates at
adensityof15×10
3
cells/well. After 24 h, 100 μM lipid of each formu-
lation was added to the culture medium. Cells collected at different time
points, from 0.5 to 24 h, were washed twice with PBS to measure the
incorporated fluorescence.
In order to investigate the role of the EGFR in the uptake, a parallel
study following the above protocol was carried out, but here, cells
were pretreated with 100 μg/mL of CTX for 1 h at 37 °C before the expo-
sure to liposome treatment. Additionally, cell uptake with and without
CTX pretreatment was also explored by images captured by fluores-
cence microscopy at 40× magnification. In this experiment, 60,000
cells were seeded in culture slides chambers (BD Falcon, Bedford,
USA) and incubated for 24 h with 100 μM lipid of each type of formula-
tion. After treatments, cells were fixed in formaldehyde 4% by incubation
for 10 min at RT and preserved at 4 °C until analysis. Cell membrane was
visualized by adding anti-alpha 1 sodium potassium ATPase antibody
(abcam®, Cambridge, UK), and DNA in nucleus was stained with DAPI.
Images were acquired using the Axio Cam MR3 video camera connected
to the Zeiss Imager M1 microscope (Carl Zeiss AG, Oberkochen,
Germany) equipped with epifluorescence optics and Axiovison software
(4.6.3.0 version).
2.7. In vivo study
Fifty-four female athymic nude mice weighing 20–25 g (Harlan,
Barcelona, Spain) were housed in plastic cages under standard and
sterile conditions (25 °C, 50% relative humidity, 12 h dark/light), with
water and food ad libitum. All experiments were performed according
to European animal care regulations and the protocol was approved
by Ethical committee of the University of Navarra (075/07).
29S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
A subcutaneous tumor was induced by the inoculation of 5 × 10
6
SW-480 cells in 100 μl of PBS, in the right flank of the mice. When the
tumors reached approximately a volume of 200 mm
3
, mice were ran-
domly divided into different groups: control, free L-OH, LP-N, LP-Fab′,
and LP-CTX. Mice were intravenously (i.v.) injected with 2.5 mg/kg
dose of L-OH every three days during three cycles, as is shown in
Fig. 3. The corresponding empty formulations were also evaluated in
three extra-groups by i.v. administration of the equivalent amount of
lipids.
Tumor size, measured by an electronic caliper, and the body weight
were recorded every two days. Tumor volume was calculated according
to the following formula: V(mm
3
) = 4/3π(d
2
xD/
2
), where d and D are
respectively the smallest and the largest tumor diameters.
At the end of the experiment, day 24, all mice were sacrificed to
remove the liver, spleen and the tumor to quantify the L-OH levels
reached in those organs. They were homogenized in 1 mL of a mixture
NaCl (150 mM)/HCl (470 mM) using the Mini-Beadbeater device
(Bioespec Products, Bartlesville, UK), and mixed with 3 mL of HNO
3
(35%). Samples were heated for 2 h in a water bath at 65 °C and stored
until use. Platinum levels were measured by atomic absorption
spectrometry technique.
2.8. Statistical methods
All data were expressed as the mean and standard deviation (SD).
The statistical analysis was performed using a non-parametric test,
Kruskall–Wallis to compared all treatments followed by the U of
Mann–Whitney test to compare two by two groups. The significance
level was set at p b0.05.
3. Results
3.1. Preparation of CTX-Fab′fragments
Fig. 4 shows the SDS-PAGE results from the CTX fragmentation
(panel A) and the conjugated liposomal formulations (panel B). The
molecules obtained during the process of antibody digestion and Fab′
isolation were identified in the different bands observed in each lane.
In lane 3, full CTX antibody is applied resulting in a typical CTX band
found at 150 kDa. CTX treated with pepsin resulted in (Fab′)
2
fragments
observed at around 100 kDa and the protease itself, pepsin, at around
40 kDa (corresponding to lane 2). Successful production of monovalent
Fab′fragments after 2 h reduction with β-mercaptoethanol was demon-
strated by theband at 50 kDa, observed in lane 5. Formulations obtained
by covalent fusion of the modified ligands, CTX and CTX-Fab′,toprevi-
ous prepared liposomes are represented in lanes 2 and 3 respectively,
in panel B. These bands correspond to liposomes purified, suggesting
the non-presence of unbound ligands in the samples.
These results suggest that, the protocol based on an enzymatic
digestion by pepsin followed by the β-mercaptoethanol treatment to
reduce the disulfide bonds in the hinge region, was adequate to obtain
the monovalent Fab′fragments suitable for conjugation to liposomes.
3.2. Preparation and characterization of targeted L-OH liposomes
Liposomes containing L-OH were prepared and conjugated to CTX or
Fab′fragments using thiol-ether linkage.
The physicochemical characterization of nanoparticles in terms of
particle size, polydispersity index (PDI) and zeta potential of the three
types of liposomes developed in this study are listed in Table 1.
All liposomal formulations (targeted and control or non-targeted)
showed a particle size around 120 nm associated with a very low PDI,
suggesting a homogenous nanoparticle population, with a similar
value for zeta potential. Protein conjugation did not affect the main
particle characteristics as none of the assayed parameters reached a
statistical significant difference among formulations. Liposomes were
stable at 4 °C for, at least, one month.
The three formulations showed similar efficiency of encapsulation
with an average value of 32.07 ± 4.9% and a loading capacity of
65.2 ± 7.2 μg/mg lipid, these values are similar to those previously
reported for non-targeted L-OH liposomes [21]. Therefore, there is no
influence of the coupling process on the amount of the encapsulated
drug.
3.3. Protein coupling efficiency at different ligand/lipid ratios and effect on
cell binding
Coupling of protein to the liposomes was evaluated by protein and
phospholipid assays at different ligand/lipid ratios. The coupling effi-
ciency represented in Fig. 5 shows similar outcome for both ligands
CTX and Fab′.Thisefficiency was calculated as the relationship between
the initial and final amount of ligands conjugated to the surface of the
liposomes. A linear increase between 5 and 30 μg of protein/μmol of
lipid was followed by a plateau in the range of 30–40 μgofprotein/
μmol of lipid.
The highest efficiency was found at the lowest protein concentra-
tion, 93.3 ± 4.73% at 5 μg CTX or Fab′while, at higher concentrations
(30 μg) the efficiency was statistically lower (p b0.05), reaching values
1
st
dose
Day 0
Tum or
inoculation
Day 12
Day 18
Day 15
2
nd
dose
3
rd
dose
End of
experiment
Day 24
Oxaliplatin
quantification
Fig. 3. Scheme of the in vivo experimental design.
30 S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
between 66.66 ± 2.9% for Fab′and 63.40 ± 1.9% for CTX. Taking
into account this result, further studies were carried out using the
ratio 30 μgofprotein/μmol of lipid. Results in Table 1 correspond to
liposomes formulated with this ratio.
The effectof ligand-density on binding to EGFR positive HCT-116 co-
lorectal cancer cells was investigated in a cell association study with
fluorescent liposomes. Fig. 6 indicates that the binding, representing
the sum of binding and uptake, for both formulations increased statisti-
cally (p b0.01) as increasing the ligand density from 10 to 30 μg/μmol,
reaching a plateau at 30 μg/μmol. But, no further increase in binding
was observed at 50 μg/μmol.
3.4. In-vitro L-OH release
Kinetics release of L-OH from LP-N, LP-CTX and LP-Fab′in Hepes
saline buffer and 50% serum were investigated during the first 24 h.
Release of L-OH from liposomal formulations was similar among
them, as is observed in Fig. 7. Release increased up to 5 h to approxi-
mately 25–35% and remained constant up to 24 h. In order to study
the influence of serum proteins on the stability of liposomes, formula-
tions were incubated with 50% (v/v) of FBS in Hepes saline. Presence
of serum caused an increase in L-OH release corresponding to 10–20%,
which mainly occurred in the initial phase.
3.5. Effect of ligand binding on EGFR activation
To investigate the effect of ligand binding on receptor activity, EGFR
phosphorylation status was studied in SW-480 and HCT-116 after
15 min of exposure to different treatments. Stimulation of cells with
EGF increased the receptor phosphorylation compared to basal levels
in both cell lines; while treatments with free ligands, CTX and Fab′,in-
duced a total inhibition of basal phosphorylation (Fig. 8). Treatments
with LP-N, LP-CTX or LP-Fab′in both cell lines did not change the
basal receptor status, indicating that the binding of targeted liposomes
and possible multivalent interactions did not cause EGFR receptor
cross linking and activation.
3.6. Targeted liposomes bind selectively to overexpressed EGFR in CRC cell
lines
Cellular association of targeted liposomes was quantified for several
CRC cell lines with different EGFR expression levels at 37 °C and 4 °C
[28–31]. This process was time-dependent in the four cell lines
(Fig. 9). For targeted liposomes, LP-Fab′and LP-CTX, at 37 °C cellular
association was more rapid and higher in HCT-116 and SW-480 than
in HT-29 and SW-620; being more favorable for LP-Fab′, especially at
the first times of incubation 3 and 5 h. The difference in the signal
among celllines is compatible with the low and moderate EGFR expres-
sion levels described for SW-620 and HT-29 respectively, compared to
the elevated expression in the other two [28–31].
Fig. 4. Electrophoresis on polyacrylamide gel (8%) to check: A) the molecules obtained during different steps of CTX fragmentation. Lanes represent: 1, molecular weight size marker; 2,
pepsin; 3, monoclonal antibody, CTX; 4, (Fab′)
2
fragments; 5, monovalent Fab′fragment; B) the purified targeted liposomes: lane 2, LP-CTX and 3, LP-Fab.
Table 1
Characterization of the three types of liposomes developed. Datarepresent the averageof
three ind ependent studies and th eir SD.
LP-N LP-Fab′LP-CTX
Size (nm) 119.6 (0.36) 123.0 (1.59) 120.0 (1.19)
PDI 0.065 (0.018) 0. 057 (0.022) 0.053 (0.024)
Zeta potential (mV) −22.4 (1.08) −20.8 (2.19) −24.4 (1.45)
Fig. 5. Relationship between the initial and the final amount of ligands per μmol of lipid
in the developed formulations. Symbols represent the means and vertical lines thei r
corresponding SD of three independent studies. Lines represent the interpolation of the
experimental data. ** p b0.01. (U of Mann–Whitney test).
31S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
Blocking the receptor by CTX, strongly reduced the association of
targeted-liposomes by receptor positive cells, reaching similar levels
to those observed for non-targeted formulations. The role of the recep-
tor in the specific binding and internalization of the targeted liposomes
is endorsed by these results and the fact, that the internalization of
targeted and non-targeted liposomes was comparable in SW-620, a
cell line with low EGFR expression.
On the other hand, the active and specific binding to the receptor
was supported by the results found at 4 °C, where the signal
corresponded to non-specific binding [32]. The relative values of the
fluorescence at this temperature were very similar in all cell lines and
formulations. The observed differences in cellular association of
targeted liposomes at 4 °C and 37 °C are suggestive of rapid andefficient
internalization of liposomes and their contents.
In positive EGFR cell lines, the association of LP-Fab′was higher than
LP-CTX, while both are considerably higher than for non-targeted
liposomes.
Increased cell-specific association of targeted liposomes by receptor-
positive cells was confirmed by fluorescence microscopy (Fig. 10).
Targeted liposomes (red) bound to SW-480 cells in a specific manner
as proven by competition with free CTX (lower panels) and not to
receptor-negative SW-620. HT-29 and HCT-116 incubations had an
outcome similar to SW-480 (data not shown).
Therefore, these images support that targeted liposome uptake
occurs through the EGFR.
3.7. Targeted L-OH liposomes improve cytotoxicity over the free drug and
non-targeted liposomal L-OH
Cytotoxicity of targeted L-OH liposomes was compared to free and
non-targeted liposomal L-OH in four different colorectal cancer cell
lines. IC
50
values were calculated from the dose–response curves. Both
targeted liposomal L-OH formulations demonstrated increased efficacy
toward all tested EGFR receptor-positive colorectal cancer cell lines,
when compared to free and non-targeted liposomal L-OH, as evidenced
by the lower IC
50
values (Table 2). Remarkably, targeted liposomes de-
creased IC
50
values 2–4 fold in HT-29 and SW-480, which demonstrated
significant resistance to free L-OH, reaching nearly similar efficacy as in
the more sensitive HCT-116 cells.
Efficacy of targeted liposomes and non-targeted liposomes wassim-
ilar in the negative EGFR cell line, supporting again the importance of
receptor-binding for intracellular drug delivery and efficacy No cytotoxic
effect was found for empty targeted liposomes (data not shown).
3.8. Targeting L-OH liposomes improves therapeutic efficacy in colon cancer
bearing mice
Therapeutic activity of the L-OH targeted liposomes was studied in
SW-480 tumor bearing mice then, compared to treatments with the
free and non-targeted liposomes. Tumor growth was individually moni-
tored in all animals over time. Three 2.5 mg/kg i.v. doses of L-OH, free and
encapsulated, were administered on days: 12, 15 and 18 post-cancer cell
implantation.
Fig. 6. Effect of ligand density on association of fluorescent targeted liposomes (100 μM)
with 10, 30 or 50 μgofligand/μmol of lipid with HCT-116 cells. Data,expressed as relative
light units (R.L.U.), represent the means of three independent studies with their SD.
** p b0.01. (U of Mann–Whitney test).
Fig. 7. Time profiles of the accumulated drug release for the three formulations: stealth liposomes (LP-N) and targetedliposomes (LP-Fab′and LP-CTX) in Hepes buffer (panel A) and in
presence of FBS (panel B). Each value represents the average of three independent studies and their corresponding SD.
1234567
12 345 6 7
SW-480
HCT-116
EGFr-PActin
EGFr-PActin
Fig. 8. Bands represent the phosphorylation of EGFR after different treatments. 1: control;
2: EGF; 3: free CTX; 4: free Fab′fragment;5:LP-N;6:LP-CTX;7:LP-Fab′.
32 S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
Tumor growth in the control group increased statistically (p b0.01)
in comparison with treated groups just after second cycle administra-
tion on day 15, as is shown in Fig. 11. In addition, the antitumor effect
for liposomal formulations was found statistically (p b0.01) higher
than for the free drug after third cycle.
Results show that L-OH encapsulated was more efficient inducing a
delay in the tumor growth than the free drug, but LP-Fab′was able to
control that delay even one week after the last dose administration,
reaching a statistical significant difference (p b0.05) with LP-CTX. The
administration of empty targeted liposomes and free ligands, at the
same concentration that was coupled in the formulations, did not
show any statistically significant influence on tumor growth (see
supplementary material). It is worthy to confirm that no changes in
body weight was observer in any of the groups over the duration
of the study supporting then, the lack of toxicity of the dose used for
L-OH (see supplementary material).
3.9. Targeting L-OH liposomes improves tumor drug delivery
To study intra-tumor drug delivery, tumors from all animals were
collected at the end of the in vivo experiment, fixed on day 24 post-
cells implantation, to measure platinum levels.
The treatment with targeted liposomes led to a higher tumor accu-
mulation of L-OH compared to non-targeted and the free drug,
as shown in Fig. 12. In addition, the drug concentration in tumor for
LP-Fab′, 2916.6 ± 507.84 ng/g of tissue, was almost double than for
LP-CTX, 1546.5 ± 362.41 ng/g of tissue. This result likely explains the
higher delay in tumor growth found in mice treated with L-OH LP-Fab
′formulation shown in Fig. 11.
Accumulation of targeted liposomes showed higher affinity to
the liver compared to non-targeted, due to the presence of EGFR in
hepatocytes while in the spleen, LP-N and LP-Fab′were found in higher
concentration than LP-CTX and L-OH. Both results together led to
suggest that whole antibody may induce a more rapid clearance than
Fab′-fragment.
4. Discussion
To our knowledge, this is the first study describing the development
and the in vitro/in vivo evaluation of L-OH loaded EGFR targeted lipo-
somes using two specific ligands, CTX or its Fab′fragment. This strategy
represents an advanced step to reach specific delivery of L-OH into
colon cancer cells.
Previously, we have reported the development of a pegylated liposo-
mal formulation of L-OH [21]. This formulation formed the basis to fur-
ther advance L-OH delivery by coupling two specific ligands for EGFR
targeting to those liposomes. The selection of the ligands was based
on the therapeutic combination applied in patients with CRC, CTX plus
L-OH. This strategy seeks the improvement of the pharmacodynamics
of L-OH previously found with non-targeted liposomes [21],andthere-
duction of the adverse effects observed in patients after repeated doses
of the free CTX [33]. Moreover, this novel combination treatment aims
to specifically deliver L-OH at high levels into tumor cells. Here, we
could demonstrate that the targeting to EGFR on CRC cells increases
Fig. 9. Relative association rate obtained at several times during the first 24 h of the treatment with targeted and non-targeted liposomes in different conditions: at 37 °C, with a pretreatment
with free CTX and at 4 °C. Bars represent the average of three replicates with their SD.*p b0.05, (Kruskall–Wallis followed by the U of Mann–Whitney test).
33S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
cell binding of liposomes and intracellular drug delivery and thereby,
could reverse L-OH resistance.
Targeted liposomes have been described to deliver higher quantities
of drugs to tumors, which resulted in an enhanced therapeutic efficacy
over free drug and liposomal formulations [10–13,22,32,34].For
targeting of liposomes, antibodies have been used extensively [10–13].
However, the presence of Fc′fragments can induce the uptake of
those antibodies by the mononuclear phagocyte system, leading to a
rapid clearance of liposomes from the circulation and developing in
some cases, anti-drug antibodies (ATA) that may contribute to immu-
nogenic reactions [33]. Although the avidity by the receptors is lower
than for the whole mAb, Fab′fragments abolish the uptake by the
phagocytic system and reduce that immunogenicity improving in vivo
efficacy [19,20,32].
In this study, Fab′was obtained by enzymatic fragmentation of CTX
followed by a posterior reduction of (Fab′)
2
with β-mercaptoethanol.
This methodology provided similar results to previous described in
the literature, obtaining monovalent fragments Fab′from mAb but
here, the purification of Fab′was done by ultrafiltration using Amicon
systems instead of the sepharose columm used by Sapra et al. [32].
This purification process was simple and rapid, allowing the isolation
and identification of various fragments obtained during CTX fragmenta-
tion, as was observed in the analysis by electrophoresis shown in Fig. 4
(SDS-PAGE).
The optimization of the ligand/lipid ratio in targeted liposomes, for
both ligands, was explored from two different angles. First, we focused
on obtaining a high ligand density with highly efficient coupling quan-
tified as the relationship between the initial and final amount of the
SW-480
Membrane Nuclei Liposomes Merge
LP-CTX LP-Fab’ LP-N Control
Membrane Nuclei Liposomes Merge
LP-CTX LP- Fab ’ LP-N Control
CTX PRETREATMENT
Fig. 10. Fluorescencemicroscopy of the uptake of liposomesafter 24 h of exposure in EGFR-positive, SW-480 and EGFR-negative, SW-620 celllines at 40×. Lower panels represent results
from a CTX pretreatment previous to the liposome exposure. Different dyes have been used to stain the membrane (green), the nuclei (blue) and the liposomes (red).
34 S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
ligand attached to the liposome. Secondly, we studied the effect of
ligand density on tumor cell targeting, and assessed as cell association
of formulations with different ligand/lipid ratios using a cell line with
a high EGFR expression [28–31]. Ligand coupling to the liposomes in-
creased linearly with increasing ligand/lipid ratio and reached a plateau
at 30 μgofprotein/μmol lipid. At this density, coupling efficiencies
between 60and 70% were achieved for both CTX and Fab′. This percent-
age is in line with data previously published, describing a relatively
wide range in efficiencies, as this process depends on the formulation
and the ligand [24,32]. In this case, because of the molecular weight of
both molecules are different, the number of ligands attached to the
surface of liposomes were approximately 10 for CTX vs. 31 for Fab′.
Membrane Nuclei Liposomes Merge
LP-CTX LP-Fab ’ LP-N Control
SW-620
CTX PRETREATMENT
Membrane Nuclei Liposomes Merge
LP-CTX LP-Fab ’ LP-N Control
Fig. 10 (continued).
Table 2
IC
50
(μM) values measured at 72 h after 4 h of drug exposure. Data represent the average and SD of three independent studies.
HCT-116 HT-29 SW-480 SW-620
L-OH 28.67 (5.17)*
a,b,c
57.87 (7.16) *
a,b,c
79.22 (8.11) *
a,b,c
40.49 (0.28) *
a,b,c
LP-N 22.20 (1.73)*
a,d
26.00 (3.91) *
a,d
39.30 (4.46) *
a,d
31.99 (2.12) *
a
LP-CTX 16.64 (2.71)*
a
21.26 (1.26) *
a
26.01 (2.61) *
a
30.65 (2.69) *
a
LP-Fab′17.19 (1.12) *
a
23.82 (0.89) *
a
28.12 (0.40) *
a
35.86 (3.51) *
a
*p b0.05; a, difference amongcell lines; b, difference between free andnon-targeted; c, difference between free and targeted; d, difference between non-targeted and targeted, Kruskall–
Wallis followed by the U of Mann–Whitney test, were applied for statistical significance.
35S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
The ligandcoupling did not alter the physicochemical characteristics
of the liposomes such as particle size, PDI and zetapotential. The particle
size, around 120 nm, associated with a very low PDI support a good
standardization of the developed method to formulate the different
types of liposomes, which are small enough to suffer the EPR effect
after their i.v. administration to animals [35].
Additionally, the presence of the ligand did not affect the drug
release rate. Release kinetics suggests a certain depot effect that could
be achieved with these liposomes in an in vivo system. The presence
of serum in the dialysis tube increased slightly the drug release from
liposomes (10–20%). This may be due to the opsonization of serum
proteins that affects the stability of the liposome, although in this case
that influence has a low impact [26]. This effect has been also reported
by Abu-Lilla and coworkers (2009) for L-OH liposomes, with similar
impact on the release rate [36].
Next to optimizing ligand density with respectto coupling efficiency,
optimal tumor cell binding is also of major importance. Cell binding to
EGFR overexpressing HCT-116 cells increased with increasing ligand
density up to 30 μg/μmol lipid and remained comparably high, when
further increasing liposomal ligand density (Fig. 6). This is expected as
coupling efficiencies do not increase much above 30 μg/μmol (Figs. 5–
6). A clear relationship between ligand density and tumor cell binding
has been described before for whole mAb [19] and for antibody
fragments [37].
It is interesting to point out that although the Fab′fragment presents
only one binding site compared to the two sites present in the whole an-
tibody, similar cell association level was obtained for liposomes
equipped with either one of the ligands. This may well be explained
by the slightly higher number of actual Fab′ligands at the liposomal
surface compared to the larger mAb, in combination with the random
orientation of the whole antibody, which for a portion of the mAb will
physically hinder binding to the receptors [38].Apparentlyboth
compensate for the lower number of antigen binding sites per ligand
and suggest that Fab′coupling to liposomes for this ligand restores the
multivalent binding [32]. The association studies carried out in cell
lines with different levels of EGFR expression, showed an increase in
the association of targeted liposomes at 37 °C with increasing receptor
density. This result together with the inhibition of the internalization
by a pretreatment with free mAb, reveal that this process takes place
through selective binding to the receptor. The behavior of the negative
control cell line, SW-620, confirmed those results as no differences in
the internalization between targeted and non-targeted liposomes
were found. The high cell association of targeted liposomes at 37 °C
compared to 4 °C strongly indicates cellular uptake of targeted
liposomes via an active process [32].
Regarding the difference between both ligand-targeted liposomes,
Fab′fragment seemed to influence positively the uptake, probably due
to the random orientation of the whole antibody coupled in the lipo-
some as has been commented above. This limitation has also led to
some authors to use the Fab′fragment, as a selective ligand for coupling
to nanoparticles, in order to enhance the cell uptake and improve the
in vivo efficacy [24,32,34].
Cellular association was confirmed by fluorescence microscopy. The
highest fluorescence intensity associated with SW-480 cells was found
for LP-Fab′, in comparison with other cell lines and LP-CTX, which
supports the relationship between EGFR expression and the degree of
internalization and the higher receptor binding capacity of LP-Fab′.
Cytotoxicity studies confirmed efficacious intracellular L-OH delivery
by targeted liposomes. Targeted liposomal L-OH outperformed liposo-
mal and free L-OH in all EGFR overexpressing cell lines, and was even
able to reverse L-OH resistance in the two cell lines which displayed
decreased L-OH sensitivity. This increase in the drug sensitivity may
be supported by a modification in the mechanisms involved in this cel-
lular uptake. Thus, L-OH resistant mechanisms are related to the copper
receptor which mediates the cell entry of the free L-OH and the efflux
pumps that decrease the cytoplasmic drug concentration [39,40]. The
advantage of the drug encapsulation in targeted liposomes lies in the
selectivity of the uptake by a receptor up-regulated enhancing thus,
the intracellular drug concentration. This phenomenon has been de-
scribed even for non-targeted formulations like Lipodox, which reduces
the exchange of doxorubicin by the P-glycoprotein [41].
Differences in cellular uptake observed between LP-Fab′and LP-CTX
did not markedly affect IC
50
values in the cytotoxicity studies. Whereas
differences in binding were observed at shorter incubations, in this
Fig. 11. Time profile of the relative tumor growth for the different treatments. Each point
represents the average of six mice and the bars are the SD. Arrows represent the dose
administration at days 12, 15 and 18 after tumor inoculation.* p b0.05, Kruskall–Wallis
followed by the U of Mann–Whitney.
Fig. 12. L-OH levels measured in spleen and liver (panel A) and tumor (panel B), 144 h
after last dose administration following the different tr eatments. Bars represent the
average of six mice, and the vertical lines their corresponding SD (*, p b0.05; **, p b0.01
provided by the comparison between targeted and non-targeted liposomes using U of
Mann–Whitney).
36 S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
study the IC
50
was measured at 72 h after 4 h of drug exposure. Other
studies report similar observations [42,43]. Iden and Allen found that
IC
50
of CD19-targeted Dox liposomes decreased with increasing drug
exposure time or the time to measure the cytotoxicity after treatment
[42]. Importantly, as shown in Table 2, a targeted liposomal L-OH
increased sensitivity of colorectal cancer cell lines more than three
times compared to the free or liposomal L-OH.
The in vivo efficacy study was carried out using a very low dose of
L-OH every three days [44]. This regimen was chosen in order to iden-
tify the importance of targeted delivery during several cycles. L-OH
(2.5 mg/kg i.v. dose) was able to inhibit the growth of the tumor in
mice after the second dose. Liposomal encapsulation together with
the targeting strategy clearly benefited tumor response. One week
after the last cycle, at day 24 post cell implantation, the reduction of
tumor growth for liposomes was much higher than for the free drug
when compared to the control group. These differences, expressed as
% inhibition was ranged from 52%, for LP-N, to 66% for LP-Fab′, while
for free L-OH that effect was 34%.
These results show that the antitumor activity of L-OH liposomal
encapsulation was more effective than the free drug and even more
importantly, that targeting using the Fab′ligand was more efficacious
than CTX. Although the dose of L-OH is lower than the dose used by
other authors, the aim of this work was to elucidate the effectiveness
of the targeted liposomes. Yang et al. [45] found an inhibition of 26%
after several doses of 5 mg/kg administered every four days of non-
targeted L-OH pegylated liposomes, which is slightly lower than the
inhibition found for LP-N in this study. A washout of 72 h was chosen
because very low levels of liposomes were detected by Yang and
coworkers in tumor at that time after liposome administration and be-
cause, similar washout has also been reported by Abu-Lila et al. [44],
which represents approximately five times the half-life of L-OH in
nude mice [45]. However, we found relatively high levels of drug in
tumor 144 h after the third administration, especially in the case of
targeted liposomes, as shown in Fig. 10. This observation supports a se-
lective accumulation of targeted formulations that correlate to tumor
growth inhibition. LP-Fab′presented the higher L-OH accumulation in
tumor tissue followed by LP-CTX and LP-N. This result is according to
data previously reported by several authors, where a higher efficiency
in the selective targeting by Fab′ligands than for the whole antibody,
was described [32,46]. This may well be explained by the presence of
Fc′fragment on whole mAb carrying liposomes, which can be recog-
nized by macrophages leading to a more rapid removal from circulation
than the Fab′fragment [19,20,32], in combination with a lower ability to
bind to receptor due to the orientation of the attached antibodies
[10–13].
In addition, data from Fig. 10 can also be supported by previous re-
sults from a pilot study carried out with similar protocol but for only
two doses. L-OH levels increased from 4 until 24 h post-dosing for the
formulations, especially for LP-Fab′, and decreased for the free drug, as
is shown in supplementary material (Fig. 1S). These results correlated
with L-OH plasma levels at 24 and 72 h listed in table 1S, where those
levels were maintained for LP-Fab′compared to LP-CTX and LP-N.
Note that animals from the in vivo study did not show significant
toxicity, loss body weight or cachexia (Fig. 3S), even in groups treated
with targeted formulation, where L-OH tissue accumulation was not
associated to an enhancement of toxicity.
5. Conclusion
In conclusion, we demonstrated that EGFR targeted L-OH liposomes
were more efficient in tumor drug accumulation than free L-OH or non-
targeted liposomes, which improved efficacy in mice inoculated with a
CRC cell line overexpressing this receptor. In vitro results supported
these results by proving that targeted liposomes, particularly LP-Fab′
that induced cell-specific binding, uptake, intracellular drug delivery
and subsequent cytotoxicity to EGFR-positive CRC cells. Remarkably,
targeted delivery was also effective in L-OH resistant cell lines, suggest-
ing that this approach could reverse drug resistance by inducing active
cellular drug uptake through the EGFR. Therefore, EGFR targeted L-OH-
liposomes seem to be a promising novel nanomedicine to improve CRC
treatment.
Acknowledgements
Funding for this project came from the Spanish Government
(Instituto de Salud Carlos III, ref: PS09/02512-FISS), Government of
Navarra (ref: IIQ14334.RI1) and from the University of Navarra. Sara
Zalba was supported by a grant from the Government of Navarra. The
authors want to thank the Imaging Unit from CIMA (Pamplona, Spain)
for their support to this work.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.jconrel.2015.05.271.
References
[1] B.W. Harper, et al., Advances in platinum chemotherapeutics, Chemistry 16 (24)
(2010) 7064–7077.
[2] D. Liu, et al., Application of liposomal technologies for delivery of platinum analogs
in oncology, Int. J. Nanomedicine 8 (2013) 3309–3319.
[3] A. Stein, D. Arnold, Oxaliplatin: a review of approved uses, Expert. Opin.
Pharmacother. 13 (1) (2012) 125–137.
[4] F. Levi, et al., Oxaliplatin: pharmacokinetics and chronopharmacological aspects,
Clin. Pharmacokinet. 38 (1) (2000) 1–21.
[5] S. Zalba, M.J. Garrido, Liposomes, a promising strategy for clinical application of
platinum derivatives, Expert Opin. Drug Deliv. 10 (6) (2013) 829–844.
[6] A. Gabizon, Emerging role of liposomal drug carrier systems in cancer chemothera-
py, J. Liposome Res. 13 (1) (2003) 17–20.
[7] S.M. Moghimi, J. Szebeni, Stealth liposo mes and long circulating nanoparticles:
critical issues in pharmacokinetics, opsonization and protein-binding properties,
Prog. Lipid Res. 42 (6) (2003) 463–478.
[8] H. Maeda, The enhanced permeability and retention (EPR) effect in tumor vascula-
ture: the key role of tumor-selective macromolecular drug targeting, Adv. Enzym.
Regul. 41 (2001) 189–207.
[9] T.M. Allen, P.R. Cullis, Liposomal drug delivery systems: from concept to clinical
applications, Adv. Drug Deliv. Rev. 65 (1) (2013) 36–48.
[10] E. Ruoslahti, S.N. Bhatia,M.J. Sailor, Targeting of drugs and nanoparticles to tumors, J.
Cell Biol. 188 (6) (2010) 759–768.
[11] V.P. Torchilin, Drug targeting, Eur. J. Pharm. Sci. 11 (Suppl. 2) (2000) S81–S91.
[12] V.P. Torchilin, Targeted pharmaceutical nanocarriers for cancer therapy and
imaging, AAPS J. 9 (2) (2007) E128–E147.
[13] A.S. Manjappa, et al., Antibodyderivatization and conjugation strategies: application
in preparation of stealth immunoliposome to target chemotherapeutics to tumor, J.
Control. Release 150 (1) (2011) 2–22.
[14] Y.L. Lin, et al., Oxaliplatin-based chemotherapy is more beneficial in KRAS mutant
than in KRAS wild-type metastatic col orectal cancer patients, PLoS One 9 (2)
(2014) e86789.
[15] B. Vincenzi, et al., The biological properties of cetuximab, Crit. Rev. Oncol. Hematol.
68 (2) (2008) 93–106.
[16] N. Yarom, D.J. Jonke r, The role of the epidermal growth factor receptor in the
mechanism and trea tment of colorectal cancer, Discov. Med. 11 (57) (2011)
95–105.
[17] V. Deschoolmeester, et al., A review of the most promising biomarkers in colorectal
cancer: one step closer to targeted therapy, Oncologist 15 (7) (2010) 699–731.
[18] C.M.Rocha-Lima, et al.,EGFR targeting of solidtumors, Cancer Control 14 (3) (2007)
295–304.
[19] G.A. Koning, et al., Interaction of differently designed immunoliposomes with colon
cancer cells and Kupffer cells. An in vitro comparison, Pharm. Res. 20 (8) (2003)
1249–1257.
[20] P. Holliger, P.J. Hu dson, Engineered antibody fragments and the rise of single
domains, Nat. Biotechnol. 23 (9) (2005) 1126–1136.
[21] S. Zalba, et al., Application of different methods to formulate PEG-liposomes of
oxaliplatin: evaluation in vitro and in vivo, Eur. J. Pharm. Biopharm. 81 (2) (2012)
273–280.
[22] S. Song, et al., Peptide ligand-mediated liposome distribution and targeting to EGFR
expressing tumor in vivo, Int. J. Pharm. 363 (1-2) (2008) 155–161.
[23] M. Rothdiener, et al., Antibody targeting of nanoparticles to tumor-specific
receptors: immunoliposomes, Methods Mol. Biol. 624 (2010) 295–308.
[24] J. Beuttler, et al., Targeting of epidermal growth factor receptor (EGFR)-expressing
tumor cells with sterically stabilized affibody liposomes (SAL), Bioconjug. Chem.
20 (6) (2009) 1201–1208.
[25] G. Rouser, S. Fkeischer, A. Yamamoto, Two dimensional then layer chromatographic
separation of pol ar lipids and determination of phospholipids by phosphorus
analysis of spots, Lipids 5 (5) (1970) 494–496.
37S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
[26] A. Hioki, et al., Development of an in vitro drug release assay of PEGylated liposome
using bovine serum albumin and high temperature, Biol. Pharm. Bull. 33 (9) (2010)
1466–1470.
[27] C.W. Lowik, et al., Quantification of adherent and nonadherent cells cultured in 96-
well plates using the supravital stain neutral red, Anal. Biochem. 213 (2) (1993)
426–433.
[28] R. Wild,et al., Cetuximab preclinical antitumor activity (monotherapyand combina-
tion based) is not predicted by relative total or activated epidermal growth factor
receptor tumor expression levels, Mol. Cancer Ther. 5 (1) (2006) 104–113.
[29] S. Skvortsov, et al., Cetuximab inhibits thymidylate synthase in colorectal cells
expressing epidermal growth factor receptor, Proteomics Clin. Appl. 2 (6) (2008)
908–914.
[30] M. Shimizu, et al., (-)-Epigallocatechin gallate and polyphenon E inhibit growth and
activation of the epidermal growth factor receptor and human epidermal growth
factor receptor-2 signaling pathways in human colon cancer cells, Clin. Cancer Res.
11 (7) (2005) 2735–2746.
[31] P.C. Lee,et al., Targeting colorectal cancer cells with single-walled carbon nanotubes
conjugated to anticancer agent SN-38 and EGFR antibody, Biom aterials 34 (34)
(2013) 8756–8765.
[32] P. Sapra, et al., Improved therapeutic responses in a xenograft model of human B
lymphoma (Namalwa) for liposomal vincristine versus liposomal doxorubicin
targeted via anti-CD19 IgG2a or Fab′fragments, Clin. Cancer Res. 10 (3) (2004)
1100–1111.
[33] B.A. Baldo, Adverse events to monoclonal antibodies used for cancer therapy: focus
on hypersensitivity responses, Oncoimmunology 2 (10) (2013) e26333.
[34] J.W. Park, et al., Anti-HER2 immunoliposomes: enhanced efficacy attributable to
targeted delivery, Clin. Cancer Res. 8 (4) (2002) 1172–1181.
[35] C.L. Waite, C.M. Roth, Nanoscale drug delivery systems for enhanced drug penetra-
tion intosolid tumors: current progress and opportunities, Crit. Rev. Biomed. Eng. 40
(1) (2012) 21–41.
[36] A. Abu-Lila, et al., Oxaliplatin targeting to angiogenic vessels by PEGylated cationic
liposomes suppresses the angiogenesis in a dorsal air sac mouse model, J. Control.
Release134(1)(2009)18–25.
[37] B.P. Gray, M.J. McGuire, K.C. Brown, A liposomal drug platform overrides peptide
ligand targeting to a cancer biomarker, irrespective of ligand affinity or density,
PLoS One 8 (8) (2013) e72938.
[38] A. Beduneau, et al., Design of targeted lipid nanocapsules by conjugation of whole
antibodies and antibody Fab′fragments, Biomaterials 28 (33) (2007) 4978–4990.
[39] S.B. Howell, et al., Copper transpor ters and the cellula r pharmacology of th e
platinum-containing cancer drugs, Mol. Pharmacol. 77 (6) (2010) 887–894.
[40] D. Wang, S.J. Lippard, Cellular processing of platinum anticancer drugs, Nat. Rev.
Drug Discov. 4 (4) (2005) 307–320.
[41] C. Riganti, et al., Liposome-encapsulated doxorubicin reverses drug resistance by
inhibiting P-glycoprotein in human cancer cells, Mol. Pharm. 8 (3) (2011) 683–700.
[42] D.L. Iden, T.M. Allen, In vitro and in vivo comparison of immunoliposomes made by
conventional cou pling techniques with th ose made by a new post-insertion
approach, Biochim. Biophys. Acta 1513 (2) (2001) 207–216.
[43] D. Goren, et al., Targeting of stealth liposomes to erbB-2 (Her/2) receptor: in vitro
and in vivo studies, Br. J. Cancer 74 (11) (1996) 1749–1756.
[44] A.S. Abu Lila, et al., Sequential administration with oxaliplatin-containing PEG-
coated cationic liposomes promotes a significant delivery of subsequent dose into
murine solid tumor, J. Control. Release 142 (2) (2010) 167–173.
[45] C. Yang, et al., Oxaliplatin long-circulating liposomes improved therapeutic index of
colorectal carcinoma, BMC Biotechnol. 11 (2011) 21.
[46] P. Simard, J.C. Leroux, In vivo evaluation of pH-sensitiv e polymer-based
immunoliposomes targeting the CD33 antigen, Mol. Pharm. 7 (4) (2010) 1098–1107.
38 S. Zalba et al. / Journal of Controlled Release 210 (2015) 26–38
