Drug delivery to tumours using a novel 5-FU derivative encapsulated into lipid nanocapsules

Article abstract of DOI:10.1080/1061186X.2018.1547733

In this work, a novel lipophilic 5-fluorouracil (5-FU) derivative was synthesised and encapsulated into lipid nanocapsules (LNC). 5-FU was modified with lauric acid to give a lipophilic mono-lauroyl-derivative (5-FU-C12, MW of about 342 g/mol, yield of reaction 70%). 5-FU-C12 obtained was efficiently encapsulated into LNC (encapsulation efficiency above 90%) without altering the physico-chemical characteristics of LNC. The encapsulation of 5-FU-C12 led to an increased stability of the drug when in contact with plasma being the drug detectable until 3 h following incubation. Cytotoxicity assay carried out using MTS on 2D cell culture showed that 5-FU-C12-loaded LNC had an enhanced cytotoxic effect on glioma (9L) and human colorectal (HTC-116) cancer cell line in comparison with 5-FU or 5-FU-C12. Then, HCT-116 tumour spheroids were cultivated and the reduction of spheroid volume was measured following treatment with drug-loaded LNC and drugs alone. Similar reduction on spheroids volume was observed following the treatment with drug-loaded LNC, 5-FU-C12 and 5-FU alone, while blank LNC displayed a reduction in cell viability only at high concentration. Globally, our data suggest that the encapsulation increased the activity of the 5-FU-C12. However, in-depth evaluations of LNC permeability into spheroids are needed to disclose the potential of these nanosystems for cancer treatment.

Full text of DOI:10.1080/1061186X.2018.1547733

Journal of Drug Targeting
ISSN: 1061-186X (Print) 1029-2330 (Online) Journal homepage: http://www.tandfonline.com/loi/idrt20
Drug delivery to tumors using a novel 5-FU
derivative encapsulated into lipid nanocapsules
Giovanna Lollo, Kevin Matha, Martina Bocchiardo, Jérôme Bejaud, Ilaria
Marigo, Angelique Virgone-Carlotta, Thomas Dehoux, Charlotte Rivière, Jean
Paul Rieu, Stephanie Briançon, Thomas Perrier, Olivier Meyer & Jean-Pierre
Benoit
To cite this article: Giovanna Lollo, Kevin Matha, Martina Bocchiardo, Jérôme Bejaud, Ilaria
Marigo, Angelique Virgone-Carlotta, Thomas Dehoux, Charlotte Rivière, Jean Paul Rieu, Stephanie
Briançon, Thomas Perrier, Olivier Meyer & Jean-Pierre Benoit (2018): Drug delivery to tumors using
a novel 5-FU derivative encapsulated into lipid nanocapsules, Journal of Drug Targeting, DOI:
10.1080/1061186X.2018.1547733
To link to this article: https://doi.org/10.1080/1061186X.2018.1547733
Accepted author version posted online: 21
Nov 2018.
Submit your article to this journal
Article views: 5
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=idrt20

Drug delivery to tumors using a novel 5-FU derivative encapsulated
into lipid nanocapsules
Giovanna Lollo1,2*, Kevin Matha2,3*, Martina Bocchiardo2, Jérôme Bejaud2, Ilaria Marigo4, Angelique
Virgone-Carlotta5, Thomas Dehoux5, Rivière Charlotte5, Rieu Jean Paul5, Stephanie
Briançon1,Thomas Perrier6, Olivier Meyer6, and Jean-Pierre Benoit2,3
1, Université de Lyon, Université Claude Bernard Lyon 1, CNRS UMR 5007, Laboratoire
d'Automatique et de Génie des Procédés (LAGEP), 43 bd 11 Novembre, 69622 Villeurbanne, France;
Institut des Sciences Pharmaceutiques et Biologiques, 8 av Rockefeller, 69373 Lyon, France.
2, MINT, Université d’Angers, INSERM U1066, CNRS UMR 6021, Angers 49933, France.
3, Pharmacy Department, Angers University Hospital, 4 rue Larrey, Angers 49933, France
4, Veneto Institute of Oncology IOV-IRCCS, 35128 Padova, Italy
5, Univ Lyon, Université Claude Bernard Lyon 1, CNRS, Institut Lumière Matière, Villeurbanne,
France
6, Carlina Technologies, 22 rue Roger Amsler, 49100 Angers, France
*Both authors equally contributed to the work

Abstract
In this work, a novel lipophilic 5-FU derivative was synthetized and encapsulated into lipid
nanocapsules (LNC). 5-FU was modified with lauric acid to give a lipophilic mono-lauroyl-derivative
(5-FU-C12, MW of about 342 g/mol, yield of reaction 70%). 5-FU-C12 obtained was efficiently
encapsulated into LNC (encapsulation efficiency above 90%) without altering the physico-chemical
characteristics of LNC. The encapsulation of 5-FU-C12 led to an increased stability of the drug when
in contact with plasma being the drug detectable until 3h following incubation. Cytotoxicity assay
carried out using MTS on 2D cell culture showed that 5-FU-C12-loaded LNC had an enhanced
cytotoxic effect on glioma (9L) and human colorectal (HTC-116) cancer cell line in comparison with
5-FU or 5-FU-C12. Then, HCT-116 tumor spheroids were cultivated and the reduction of spheroid
volume was measured following treatment with drug-loaded LNC and drugs alone. Similar
reduction on spheroids volume was observed following the treatment with drug-loaded LNC, 5-FU-
C12 and 5-FU alone, while blank LNC displayed a reduction in cell viability only at high
concentration. Globally, our data suggest that the encapsulation increased the activity of the 5-FU-
C12. However, in depth evaluations of LNC permeability into spheroids are needed to disclose the
potential of these nanosystems for cancer treatment.

1. Introduction
5-Fluorouracil (5-FU, 5-fluoro-1H-pyrimidine-2,4-dione) is an antineoplasic agent used against a
wide range of solid tumors (such as breast, head and neck, colon, pancreas and stomach tumors)
[1]. In addition to its direct cytotoxic effect on tumor cells, the administration of low doses of 5-FU
is able to induce a selective depletion of immunosuppressive myeloid cell population, namely
myeloid-derived suppressor cells (MDSCs), which hampers tumor growth by enhancing antitumor
T-cell response [2, 3].
Intracellularly 5-FU is converted into different cytotoxic metabolites (fluorodeoxyuridine
monophosphate (FdUMP), fluorodeoxyuridine triphosphate (FdUTP) and fluorouridine triphosphate
(FUTP). These active metabolites disrupt RNA synthesis and inhibit the action of thymidylate
synthase (TS). The rate-limiting enzyme in 5-FU catabolism is dihydropyrimidine dehydrogenase
(DPD), which converts 5-FU to its inactive metabolite dihydrofluorouracil (DHFU). More than 80% of
administered 5-FU is normally catabolized primarily in the liver, where DPD is abundantly
expressed, and as a consequence lower amounts of drug are able to reach the tumor target site [1].
Moreover, lacks in drug efficiency are caused by a non-favorable pharmacokinetic profile (i.e. poor
distribution to the tumor tissue, short plasma half-life (15-20 min), rapid catabolism, schedule-
dependent toxicity profile) and phenomenon of drug resistance (accelerated efflux of the active
form by P-gp protein at the surface of cells, variation in DPD activity or gene amplification of TS)
that occur frequently [4]. Additionally, 10–20% of patients treated with standard 5-FU usually show
severe toxicities. DPD deficiency, a pharmacogenetic syndrome leading to limited detoxification
capabilities, makes patients overexposed and prone to toxicities, thus often hampering treatment
completion, when not directly life-threatening [5].
In this respect, to improve toxicity/efficacy balance [6], numerous modifications of the 5-FU
structure have been performed and novel derivatives of 5-FU have been reported [7, 8]. Among
them, tegafur, carmofur and floxuridine, 5-FU prodrugs, have proven their clinical efficacy with low
toxicity and enhanced metabolic stability [9]. Also, capecitabine (Xeloda®), an oral fluoropyrimidine
carbamate, that is activated selectively by the thymidine phosphorylase (TP) to form 5-FU, has been
developed to increase tumor selectivity [10]. However, the results obtained using these derivatives
are still marginal and 5-FU biodistribution and toxicity remain a challenging issue in oncology.
Strategies based on nano and micro medicines appear as a novel therapeutic approach to optimize
drug biodistribution and antitumor effect [11-13]. The final aim is to design systems with a high
drug loading and an optimal release of chemotherapeutic agents into the tumor tissue, which
reduce drug accumulation and toxicity in healthy tissues [14, 15]. A liposomal formulation named
LipoFufol® made of 5-FU combined to 2′-deoxyinosine and folic acid to improve its efficacy-toxicity
balance has been developed [16]. Besides, solid lipid nanoparticles or PLGA nanoparticles have also

been described for targeted delivery of 5-FU [17, 18]. Even if the encapsulation in nanosystems
results in an increased therapeutic efficacy of the drug, its high hydrophilicity represents a limiting
step for the loading of such nanosystems as well as for the premature release of 5-FU from the
nanocarriers when injected in blood.
In the present work, we combined the synthesis of a novel lipophilic 5-FU derivative made of 5-FU
conjugated to lauric acid and a nanotechnology approach based on lipid nanocapsules (LNC). The
rationale behind this strategy is that, increasing the lipophilicity of the drug, it will be possible to
obtain a higher drug loading and a better controlled release of the drug once encapsulated into the
lipid nanocarriers. The feasibility and transposability of the system were evaluated and the batch
formulation was scaled up 20-fold. In vitro studies on glioma (9L) and colon (HCT-116) cancer cell
lines to assess the efficacy of the derivative both in its free form or encapsulated into LNC were
performed. Besides, three-dimensional (3D) spheroids made of HCT-116 cells were generated and
the effect of 5-FU derivative alone and loaded into LNC was studied.
Globally, the approach presented in this paper was focused on the development of a novel 5-FU
derivative-loaded LNC and on the development of a more predictive in vitro models to highlight the
added value of nano-therapeutic strategies.
1. Materials and Methods
1.1 Chemicals
5-Fluorouracil
(5-FU),
formaldehyde,
4-dimethylaminopyridine
(DMAP),
N,N'
dicyclohexylcarbodiimide (DCC) and lauric acid were purchased from Sigma-Aldrich (St. Quentin-
Fallavier, France). Acetonitrile, methanol and ethyl acetate were purchased from Fisher Scientific
(Illkirch, France) and silica gel from Merck (Fontenay-sous-Bois, France). Labrafac® (glyceryl
tricaprylate), Span 80® (sorbitane monooleate) and Kolliphor® HS 15 (polyethyleneglycol mono-and
di-esters of 12-hydroxystearic acid and about 30% polyethylene glycol) were obtained from Abitec
Corp. (Colombus, Ohio, USA), Fluka-Sigma-Aldrich and BASF (Ludwigshafen, Germany), respectively;
NaCl was purchased from Prolabo (Fontenay-sous-Bois, France) and water was obtained from a
MilliQ system (Millipore, Paris, France).
1.2 Synthesis and characterization of 5-FU-C12 derivative
1.2.1 Synthesis of 5-FU-C12
The synthetic procedure for the preparation of 5-FU-C12 (5-fluoro-2,4-dioxo-3,4-
dihydropyrimidine-1(2H)-yl)methyl hexadecanoate) involved two steps and was performed
according to the scheme below (Figure 1).

Figure 1: Scheme of 5-FU-C12 synthesis
Briefly, 5-FU (500 mg, 3.8 mmol) reacted with formaldehyde (37% wt.) in aqueous solution (50:50
v/v; 2.5 mL of formaldehyde/2.5 mL of MilliQ water (Millipore, Paris, France)) in a round-bottom
flask immersed in a water bath (60 °C). The reaction was conducted under magnetic stirring for 1 h
and gave as products a mixture of N-1-hydroxymethyl-5-fluorouracil, N-3-hydroxymethyl-5-
fluorouracil and N,N'-1,3-bis(hydroxymethyl-5-fluorouracil) [19, 20].
Then, the round-bottom flask was cooled in an ice-bath (0-5 °C) and acetonitrile was added (15 mL).
After, DCC (1.27 g, 6.08 mmol), DMAP (34 mg, 0.266 mmol) and lauric acid (1.23 g, 6.14 mmol)
were added to achieve the esterification. The reaction mixture was stirred for 1 h at 0 °C, then 24 h
at room temperature and monitored by thin-layer chromatography. The secondary product
dicyclohexyl urea (DCU) was separated by filtration and eliminated after several washings with
acetonitrile. The product was recovered by solvent evaporation under reduced pressure and
purification on
a
silica gel chromatographic column (isocratic elution; eluent:
dichloromethane/ethyl acetate 95:5 v/v).
1.2.2 Characterization of 5-FU-C12: 1H NMR
1
Confirmation of the structure and purity of 5-FU-C12 were performed by H NMR. NMR spectra
were obtained on a Bruker 500 MHz spectrometer (Bruker France SAS, Wissembourg, France) using
deuterated dimethyl sulfoxide (DMSO-d6) as solvent.
1.2.3 Pre-formulation studies: solubility of 5-FU-C12 derivative
The saturation solubility of the 5-FU-C12 derivative in different oils or solvents was determined by
adding an excess amount of drug in 1 mL of various liquids in small vials. Vials were placed at room
temperature or in a water bath at 60°C during 3h, and continuously stirred to reach equilibrium (48

h at 25°C). After that, drug mixtures were centrifuged at 1,700 g for 20 min at 20 °C (Centrifuge
Eppendorf 5810R, Montesson, Paris, France). The supernatant was separated and added to
acetonitrile and solubility was quantified by HPLC at 215 nm. The solubility studies were carried out
in triplicate and results were reported as mean ± SD. LogP was also calculated through the program
ACD/ChemSketch (Advanced Chemistry Development (ACD/Labs), Strasbourg, France).
1.2.4 HPLC determination of 5-FU-C12
Chromatography was performed using a Waters 717 Plus Autosampler and a Waters 600 Pump
Controller (Waters S.A., Saint-Quentin-en-Yvelynes, France) with an XTerra® C18-RP18 5µm 150
mm x 4.6 mm column (Waters, Milford, Ireland) with precolumn and a Waters 2487 Dual
Absorbance Photodiode Array Detector set at λ = 263 nm. A 40 μL aliquot of each filtrate was
injected in duplicate into the HPLC column. The column was eluted at 1 mL/min flow rate using a
gradient obtained by mixing amounts of MilliQ water (A) and acetonitrile (B). The initial mobile-
phase composition was 90% A and 10% B; a first linear gradient was applied to reach a composition
of 70% A 30% B after 5 min, a second linear gradient was applied to reach a composition of 100% B
after 10 min, maintained for 5 min and then returned to 90% A and 10% B. The peak of 5-FU-C12
appears at 16.5 min. Data acquisition, analysis and reporting were performed using Empower
chromatography software (Milford, Massachusetts, USA).
1.3 Development and characterization of blank and 5-FU-C12 loaded LNC
1.3.1 Preparation of blank and 5-FU-C12-loaded LNC
LNC were formulated using a phase inversion-based process previously described [21, 22]. Briefly,
different amounts of 5-FU-C12 were firstly stirred in a mixture of Labrafac® and Span® 80. Then,
0.967 g of Kolliphor® were added together with NaCl (45 mg) and water (1.02 mL). Three
temperature cycles (between 45 and 70 °C) were performed to obtain the phase inversion of the
emulsion. Between these temperatures, there was a phase inversion zone (PIZ) at around 55-60 °C.
At 1-3 °C from the beginning of the PIZ of the last temperature cycle, a rapid cooling and dilution
with purified ice cooled water (2.15 g) led to LNC formation. The nanocapsules were then stored at
4 °C. Sterile 5-FU-C12 LNC batches were obtained by filtration through 0.22 µm Millipore®
Stericup™ filter units (Merck, Darmstadt, Germany). Sterility of the formulation was assessed using
soybean casein digest medium at 20-25°C and thioglycolate medium at 30-35°C incubated during
14 days and 5 days of subculture according to the European Pharmacopoeia (Ph Eur), 9 Ed
(Confarma, Hombourg, France). Endotoxin content was assessed using chromogenic kinetic method
with kinetic QCL Limulus Amebocyte Lysate in GMP conditions (Confarma, Hombourg, France).

1.3.2 Physico-chemical characterization
Particle size and  potential of colloidal systems were determined using a Zetasizer Nano ZS
(Malvern Instruments, Worcestershire, UK). For size measurement, samples were diluted to an
appropriate concentration in deionized water and each analysis was carried out at 25ºC with a
detection angle of 173º. The  potential values were calculated from the mean electrophoretic
mobility values, as determined by Laser Doppler Anemometry (LDA). For LDA measurements,
samples were diluted in in 0.1mM KCl.
1.3.3 Cryogenic-transmission electron microscopy (cryoTEM)
To evaluate the morphology of blank and loaded LNC, diluted samples were dropped onto 300
Mesh holey carbon films (Quantifoil R2/1) and quench-frozen in liquid ethane using a cryo-plunge
workstation (made at Laboratoire de Physique des Solides-LPS Orsay, France). The specimens were
then mounted on a precooled Gatan 626 specimen holder, transferred in the microscope (Phillips
CM120) and observed at an accelerating voltage of 120 kV (Centre Technologique des
Microstructures (CTμ), platform of the University Claude Bernard Lyon 1, Villeurbanne, France).
1.3.4 Determination of drug encapsulation efficiency
Firstly, the mixtures of Labrafac® and Span®80, in which 5-FU-C12 was solubilized, were collected
and analyzed by HPLC (after dilution 1:1000 in acetonitrile) to calculate the total amount of 5-FU-
C12 dissolved. Once the LNC were formulated, they were filtered through a 0.22 μm filter to
eliminate the free drug. Samples of drug-loaded LNC were prepared by dissolving an aliquot of LNC
dispersion in acetonitrile (dilution 1:200) and 40 μL were injected in the HPLC according to the
protocol previously described. The encapsulation efficiency (%) was determined according to the
following formula:
푚푒푎푠푢푟푒푑 푑푟푢ꢀ 푝푎푦푙표푎푑
( )
퐸퐸 % =
× 100
푡ℎ푒표푟푒푡ꢁ푐푎푙 푑푟푢ꢀ 푝푎푦푙표푎푑
Linear titration curves, with freshly made 5-FU-C12 acetonitrile solutions at a concentration range
from 0.5 to 25 μg/mL, were used to extrapolate results (r²>0.999).
1.3.5 Storage stability studies in colloidal suspension of 5-FU-C12-loaded LNC
The stability of 5-FU-C12-loaded LNC was evaluated after storage at +4 °C during 1 month. The
macroscopic aspect, particle size, polydispersity and leakage of the drug were assessed at fixed
time intervals and after filtration using a Minisart® 0.22 μm filter (Merck, Fontenay-sous-Bois,

France). In addition, particle stability was also assessed following incubation during 48 h at 37°C
with cell culture medium.
1.3.6
Stability study in plasma of 5-FU-C12-loaded LNC
The stability in a relevant physiological medium was evaluated using human plasma provided by
Etablissement Français du Sang (EFS, Pays de la Loire, Nantes, France).
Samples of plasma were defrosted and diluted (80:20) with filtered PBS (0.22 µm filter). Thirty
samples were prepared mixing 50 µL of 5-FU-C12-loaded LNC or 5-FU-C12 acetone solution and
kept at 37°C in a water bath. At different time points (0, 5, 10, 15, 30, 60, 180, 360 min and 24 and
48 h), samples were taken and diluted with 800 µL of acetonitrile in order to precipitate plasma
proteins and centrifuged at 12,290 g during 15 min at 4°C. Supernatants were recovered, filtered
through 0.22 µm filter and analyzed by HPLC as previously described. Calibration curve of 5-FU-C12
in plasma from 1 to 22.5 μg/mL was performed (r²>0.999).
1.3.7 In vitro release studies of 5-FU-C12-loaded LNC
The release rate of 5-FU-C12 from the loaded LNC was determined using an in vitro dialysis
technique [23]. Briefly, a known suspension of 5-FU-C12-loaded LNC was dispersed in a solution of
freshly prepared phosphate buffered saline (PBS), pH 7.4, containing Tween® 80 (0.1%, w/v). The
mixture was incubated at 37°C and subjected to continuous shaking at 100 rpm, using magnetic
stirring. The released 5-FU-C12 was sampled at defined time periods (0.5h, 1h, 3h, 5h, 8h, 12h, 24h
and 48h). An aliquot was recovered and immediately replaced with an equal volume of fresh
solution. The concentration of the aliquot was measured using the HPLC method previously
described.
1.4 Scale-up formulation study
Batches of 5-FU-C12-loaded LNC were scaled up to 112 g (twenty times bigger than the lab scale).
The experimental conditions were adapted in order to prepare up to 100 mL of LNC in a single step
process. An intermediate volume (50 mL, 56g) was used in order to evaluate the robustness of the
formulation process. Blank and 5-FU-C12-loaded LNC were characterized in terms of mean size,
zeta potential, and encapsulation efficiency using the methods previously described.
1.5 In vitro experiments
1.5.1 Cell culture
Two cell lines have been cultured to carry out the experiments: rat 9L gliosarcoma cells and human
HCT-116 colorectal carcinoma cancer cell.

9L cells were obtained from the European 127 Collection of Cell Culture (Sigma, Saint-Quentin
Fallavier, France). HCT-116 colorectal carcinoma (CCL247) cell line was purchased from the
American Type Culture Collection (ATCC, Manassas, Virginia, USA).
9L cells were grown in EMEM (Lonza, Verviers, Belgium) supplemented with 10% fetal bovine
serum (Lonza), 1% antibiotics (10,000 units penicillin, 10 mg streptomycin, 25 μg amphotericin
B/mL solubilized in appropriate citrate buffer (Sigma-Aldrich)) and 1% non-essential amino acids
(Lonza).
HCT-116 line were cultured in DMEM-Glutamax (Lonza) supplemented with 10% of heat-inactivated
fetal bovine serum (FBS; Sigma-Aldrich), 100 units/100 μg of penicillin/streptomycin. Cell lines (9L
and HCT-116) were thawed and cultured in T75 flasks with filter caps (Thermo Scientific™ Nunc™,
Villebon-sur-Yvette, France), maintained at 37 °C in a humidified atmosphere with 5% of CO2.
1.5.2 In vitro cell viability
In vitro cytotoxicity assays were performed using CellTiter 96® AQueous One Solution cell
proliferation assay kit (Promega, Charbonnières-Les-Bains, France) containing a tetrazolium
compound
(3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-
tetrazolium, inner salt; MTS). Briefly, 9L (6,950×103 per well) and HT-116 cells (3,500×103 per well)
were plated into 96-well plates and incubated at 37°C during 48h in air controlled atmosphere (5%
CO2). The medium was removed and cells were treated during 24 or 48h with increasing
concentrations 5-FU-C12 LNC diluted with serum free DMEM medium supplemented with 1%
antibiotics (10,000 units penicillin, 10 mg streptomycin, 25 μg amphotericin B/mL solubilized in
appropriate citrate buffer), 1% HEPES buffer, 1% NEAA (Non Essential Amino Acid) 100X (Lonza), 1%
sodium pyruvate (Lonza), 1% N1 medium supplement 100X (0.5 mg/mL recombinant human
insulin, 0.5 mg/mL, human transferrin partially iron-saturated, 0.5 μg/mL sodium selenite, 1.6
mg/mL putrescine, and 0.73 μg/mL progesterone (Sigma-Aldrich)). After incubation, cell survival
percentage was determined using the CellTiter 96® AQueous One Solution cell proliferation assay
kit. According to the procedure described by the supplier, medium with treatments was removed
and 200 μL of a 1:5 diluted MTS solution were added in each well. After 2 h at 37 °C, the
absorbance at 492 nm was recorded using a Microplater Reader (Multiskan Ascent®, Labsystem,
Cergy Pontoise, France).
Cell viability (CV) percentage was evaluated through the following formula:
퐴푏푠표푟푏푎푛푐푒 푡푟푒푎푡푒푑 푤푒푙푙
( )
퐶푉 % =
× 100
퐴푏푠표푟푏푎푛푐푒 푐표푛푡푟표푙 푤푒푙푙

with Absorbance control well, the absorbance value of untreated cells (incubated only with fresh
medium).
1.5.3 Dose response curves of in vitro cell viability data and IC50
Dose (concentration of drug treatments, μM) response (cellular viability, %) curves were plotted for
the test after correction by subtracting the background (medium) absorbance. Percentage of viable
cells was calculated based on the absorbance values (λ = 492 nm) in cells treated with media only
(assumed as 100% viable). A linear model was used to estimate the regression parameters. In
particular, the Log transformation of concentration used was calculated in order to have a normal
distribution for this variable. This model allowed us to estimate the concentration of the drug
required to reduce cell viability by 50% (IC50) and CI (confidence interval) stated at the 95%
confidence level.
1.5.4 3D cell model: MCTS formation and treatment
Multi cellular tumor spheroids (MCTS) were formed according to a previous published method [24].
Briefly, MCTS were formed using HTC-116 cell line in Ultra Low Attachment (ULA) 96 wells Round-
Bottom plate (Greiner bio-one) to avoid cell-substrate attachment. The cells were trypsinized and
were counted using a Malassez grid in order to obtain 2,400 cells per milliliter. This concentration
of cells (i.e., 480 cells per well in a volume of 200 μL) was chosen in order to obtain a single
spheroid per well, with a spheroid diameter at the end of the experimentation not exceeding 500
μm.
The plate was centrifuged for 5 minutes at 1,200 g at room temperature to initiate the formation of
spheroids. The plate was placed in the incubator under agitation at 37˚C and 5% CO2 during the
whole experiment. At the end of the first day after seeding, 100 μL of culture medium was added to
ensure proper 3D growth. After two days after seeding, MCTS were treated with 5-FU aqueous
solution, 5-FU-C12 diluted in acetone and 5-FU-C12-loaded LNC at various drug concentrations: 2,
10 and 50 μM. Blank LNC were also used to assess nonspecific toxicity that could arise from the
system. MCTS were monitored at 24 and 48 h post treatment. Eight spheroids (n=8) were probed at
each concentration. A ring of detaching cells appeared spontaneously after one day of treatment.
The spheroids were transferred into new well plates to eliminate mechanically this uncohesive
peripheral cell layer and to renew the drug and culture medium. Therefore, the reduction in
volume that we monitor during treatment arises from a loss of viability as well as from a loss of
cohesiveness.

1.6.5 Phase contrast follow up of MCTS volume
Photographs of MCTS were taken with an inverted microscope (Leica DMIRB) in phase contrast
inside the 96-well plates at 0, 24 and 48h time points after 5-FU exposure. We performed edge
detection using a sobel threshold for each spheroid using the Image J software. The resulting binary
images were fitted to an ellipse of major (퐿푀) and minor (퐿푚) axes using the ImageJ “Analyse
Particles” plugins. From this, a mean diameter was calculated, 퐷 = (퐿푀 + 퐿푚)/2. The volume 푉
was then determined assuming that the spheroids are spherical 푉 = 휋퐷3/6.
1.6 Statistical analysis
For the results of in vitro cytotoxic activity evaluation, statistical differences were determined using
non parametric tests, such as Wilcoxon-Mann-Whitney test and Kruskal-Wallis test. The type one
error rate was taken as alpha = 5% (p < 0.05).
2. Results
2.1 Synthesis and characterization of 5-FU-C12 derivative
2.1.1 Synthesis of 5-FU-C12
In order to obtain the lipophilic 5-FU derivative named 5-FU-C12 made of 5-FU conjugated to lauric
acid, a two-step process was performed. In the first step of the synthesis, pure 5-FU reacted with
aqueous formaldehyde to give a mixture of N-N1-bis- (hydroxymethyl-5-fluorouracil) and mono-
hydroxymethyl substituted 5-FU (N-1- hydroxymethyl-5-fluorouracil, N-3-hydroxymethyl-5-
fluorouracil). As previously described by Liu et al., [19] following this step, the 1H NMR spectrum
showed that N-1-hydroxymethyl-5-fluorouracil was obtained in higher amount than the others [19].
Then, the preferential conjugation of lauric acid in the N1 position by means of DCC/DMAP
chemistry was achieved. The alkyl chain was attached in preference to the functionalized nitrogen
in position 1 because the amount of intermediate N-1-hydroxymethyl-5-fluorouracil was higher
than the others [20] and also because this was the sterically less hindered and more nucleophilic
site (due to its higher pKa value). After purification by silica gel column chromatography, 5-FU-C12
(molecular weight (MW) 342 g/mol) was recovered as the main product in the form of white
powder with a yield of 70%.
2.1.2 Characterization of 5-FU-C12: 1H NMR
5-FU-C12 was characterized by 1H NMR analysis and the spectrum obtained is reported in Figure 2.
5-FU-C12 1H NMR spectrum clearly indicated that the peak at δ = 8.9 ppm was due to the aromatic
-CH proton of 5-FU; peak at δ = 6.3 ppm was due to -CH2- of the bridging methylenic group

between 5-FU and lauric acid. Further, peaks at δ = 3.1, 2.3, 2.1 and 1.7 ppm may be assigned to
the methylenic groups and terminal methyl group of the lauric acid chain.
Figure 2: 1H NMR spectrum of 5-FU-C12.
2.1.3 Pre-formulation: water solubility of 5-FU-C12
Water solubility of 5-FU-C12 was tested at two different temperature conditions (room
temperature and 60°C) as showed in Table 1. The 5-FU derivative was practically insoluble in water
as reported in Table 1. This finding confirmed the increased hydrophobic behavior of 5-FU-C12
compared to 5-FU, which is normally soluble in water (12 g/L, data from Ph Eur, 9 Ed).
Table 1: Solubility in water of 5-FU-C12 at different conditions (room temperature and 60°C).
Condition
Initial conc. 5-FU-C12
(mg/mL)
5-FU-C12 dissolved
Solubility (mg/mL)
(%)
Room Temperature
60 °C
1.38
1.50
1.02
5.62
0.01
0.09
Practically
insoluble
Subsequently, with the purpose of identifying the most adequate solubilizing agent for 5-FU-C12 to
formulate 5-FU-C12-loaded LNC, we tested the solubility in different oils and organic solvents
(ethanol and acetone) (see Table 2). In particular, we evaluated a series of excipients that are

generally used for the formulation of LNC [22, 25]. The solubility of 5-FU-C12 was also tested in the
mixture of oil and surfactant (Labrafac® and Span® 80), currently used in other formulations to
obtain loaded LNC [15]. Solubilization of 5-FU-C12 in oils was not possible after simple vortex
passage (the unbundling of the bottom required a lot of time and was not effective). Samples were
then maintained under magnetic stirring for 24 h, and plunged in a water bath set at 60 °C. After 48
h, samples were centrifuged and macroscopic observation did not reveal presence of sediment.
Assays to assess the amount of solubilized product were performed using HPLC according to the
method described in the experimental section. The results are shown in Table 2.
Table 2: Solubility of 5-FU-C12 in different oils (a) and class 3 solvents (b).
Solubilizing agent
5-FU-C12 Solubility(mg/mL)
Acetone
30.00
3.81
14.21
10.34
13.60
15.05
Class 3 solvent
Ethanol
Labrafac®
Captex® 8000
Span® 80
Oil and surfactant
Mixture (Labrafac® + Span® 80)
As further proof of the increased lipophilic behavior of the derivative, we calculated the cLogP
value : it was 4.48 for 5-FU-C12, while the value for 5-FU was -0.78.
2.2 Development and characterization of blank and 5-FU-loaded LNC
2.2.1 Physico-chemical characterization of blank and 5-FU-C12-loaded LNC
Blank and loaded LNC were obtained following the Phase Inversion Technique (PIT) previously
described [15]. To load 5-FU-C12 into LNC, Labrafac® and Span® 80 were chosen for the first step of
preparation (preliminary dissolution of the derivative before adding other components). The
solubility of the active in the mixture of selected excipients (50/50 w/w) was confirmed using HPLC.
The results of the physico-chemical characterization of blank and 5-FU-C12-loaded LNC obtained
with this process are showed on Table 3. The average size for all the systems developed was
approximately 65 nm. The polydispersity index was below 0.1, indicating unimodal and narrow size
distribution. The zeta potential values were neutral or slightly negative, ranging from -4 to -7 mV
and corresponded to classical values obtained for LNC.
To evaluate the encapsulation efficiency of the drug into the LNC, samples of 5-FU-C12-loaded LNC
were prepared and purified through 0.22 µm filtration prior to HPLC analysis. As indicated, the
encapsulation efficiency at different payloads was high, around 98%. This result was due to the high
hydrophobicity of the derivative, which was dissolved in the oily hydrophobic core of LNC. Finally,
the endotoxin content was assessed and the value was under 5 EU/mL for every batch (data not

shown) according to the Ph Eur 9 and under GMP conditions. The value obtained for the endotoxin
content was consistent with an IV administration of the LNC.
Table 3: Physico-chemical characterization of blank and 5-FU-C12-loaded LNC.
5-FU-C12
payload*
(mg/g)
Size*
(nm)
ζ-potential*
Formulation
Blank LNC
PDI
(mV)
-
65 ± 3
<0.1
-6 ± 1
1.7 ± 0.1
2.4 ± 0.2
4.5 ± 0.1
65 ± 2
64 ± 3
64 ± 1
<0.1
<0.1
<0.1
-4 ± 2
-2 ± 1
-4 ± 2
5-FU-C12-loaded LNC
*Data represent average ± S.D.; PDI: polydispersity index, 5-FU-derivative payload = mg of 5-FU-
derivative/g of LNC dispersion
A cryoTEM analysis of blank and 5-FU-C12-loaded LNC diluted twice in water, are shown in Figure 3.
This observation was chosen in addition to the DLS technique because it allowed to discriminate
the contribution of small versus big particles. Besides, DLS presented the size as a mean
hydrodynamic radius and could therefore be biased by extreme sizes of nanoparticles [26]. Both
LNC formulations presented almost monodispersed population with spherical shape and no
differences were found between empty and loaded systems.
Figure 3: Microstructure obtained by cryoTEM of blank (A and B) and LNC loaded with 1.7 mg/g of
5-FU-C12 (C and D).

2.2.2 Storage stability in aqueous suspension
The different batches of 5-FU-C12-loaded LNC were stable over one month, indeed, no differences
were observed in the average size, while a small increase in the PDI value, which remains below the
value of 0.1, was observed after 1 month as shown in Figure 4.
Stability studies of the drug-loaded nanosystems were performed to evaluate the leakage of the
derivative from the LNC at different payloads after 1 month of storage in water suspension
(measurement of variation in the drug payload and, consequently, in the rate of encapsulation). No
leakage of the derivative from the nanocapsules was detected over the period studied.
In addition, upon incubation in cell culture medium at 37°C during 48h, LNC still maintained their
integrity in terms of size and polydispersity index.
100
90
80
70
60
50
40
30
20
10
0
0.200
0.180
0.160
0.140
0.120
0.100
0.080
0.060
0.040
0.020
0.000
Day 0
Day 7
Day 14
Day 21
Day 28
PDI
5-FU-C12-LNC (1.7 mg/g)
Figure 4: Size and PDI evolution over time of 5-FU-C12-loaded (1.7 mg/g) LNC in suspension at 4°C
(n=3).
An in vitro release study to check the release of the drug from the nanocarriers, for the formulation
with 5-FU-C12-loaded in the LNC, was carried out in PBS/Tween® 80. This experiment showed that
5-FU-C12 was not released from the LNC over 24h of incubation. Indeed, thanks to the
hydrophobicity of the derivative, the drug was completely retained into the nanosystem.
2.2.3 Stability in plasma
Subsequently, the stability of encapsulated 5-FU derivative in LNC was investigated in human
plasma by measuring with HPLC method the percentage of remaining intact drug over time. 5-FU-
C12 acetone solution was also tested. As shown in Figure 5, the drug integrity of 5-FU-C12 was

maintained in its free form during the first 15 min, then the concentration decreased and only 50%
of the drug was detected after 1h of incubation. In the case of encapsulated 5-FU-C12, almost 100
% of the drug was detected following 3h of incubation and 50% of the initial content was stable
until 24 h of incubation. The improvement in drug stability detected at least during the first 3h of
incubation was ascribed to LNC encapsulation which increased drug solubility and protect the drug
from the rapid degradation.
5-FU-C12 solution
120
100
80
60
40
20
0
5-FU-C12-loaded LNC
0
0.08 0.17 0.25
0.5
1
3
8
24
48
time (h)
Figure 5: Stability of 5-FU-C12 following incubation of drug solution or encapsulated in LNC in
human plasma.
2.3 Scale up of blank and 5-FU-C12 loadedLNC
In order to scale up the LNC formulation using the phase inversion method, two large
volume batches were prepared in order to evaluate the robustness of the formulation process.
Firstly, an intermediate batch corresponding to a batch ten times bigger than the laboratory one
was produced. Then, a second batch twenty-fold bigger was obtained. The physico-chemical
characteristics of both batches in terms of size, polydispersity index and zeta potential were
assessed and the results of the characterization are reported in Table 4. No significant differences
were found comparing the laboratory batch (scale x1) and the scale up batches (scale x10
corresponding to 56g of formulation and scale x20 corresponding to 112g of formulation). The LNC
maintained their initial properties in terms of size and polydispersity indicating that the process was
transposable to industrial settings.

Table 4: Physico-chemical characterization of blank LNC at different batch scales
Scale x1 = 5.6g
Scale x10 = 56g
Scale x20 = 112g
Size
(nm)
PDI
Size
(nm)
PDI
Size
(nm)
PDI
ζ-
ζ-
ζ-
potential
potential
potential
(mV)
-10.2±2
(mV)
-
(mV)
-
1 month
65±2
73±3
<0.1
<0.1
70±2
66±2
<0.1
<0.1
70±2
73±2
<0.1
<0.1
2 months
-6.68±2
-5.84
-
2.4 In vitro cell studies
2.4.1 In vitro cytotoxicity on 2D cell growth
In order to analyze the biological effect of the preparations, we tested the viability of glioma (9L)
and colon cancer (HCT-116) cells following 24h and 48h of treatment with escalating doses of
different formulations by using the MTS assay. 5-FU derivative in the free form (5-FU-C12
solubilized in acetone) or encapsulated into LNC compared to native 5-FU aqueous solution were
tested. Blank LNC were also tested as control to exclude toxic effects of the nanocarriers. In the
case of 9L cell line, cell viability was not affected upon 24h of incubation, therefore the effect
following 48h of incubation was showed (Figure 6 A). While, HCT-116 cell viability was strongly
reduced following 24h of exposure to treatments and no significant differences were observed
upon 48h of incubation (Figure 6 B).

Figure 6: Cell viability of 9L (A) and HCT-116 (B) cells vs drug concentration after 48h incubation (37
°C) with blank and 5-FU-C12-loaded LNC, compared to free 5-FU (solution in water) and 5-FU-C12
(solution in acetone). Drug concentration is presented in log scale (n=5; mean ± S.D.).
After 48h of exposure to 5-FU-C12-loaded LNC, viability decreased as a function of the drug
concentration exposure for both cell lines tested (Figure 6). IC50 values for drug (concentration μM)
and LNC treatments (concentration mg/mL) and the respective confidence interval (µM or mg/mL),
are shown in Table 5.

Table 5. IC50 (μM) and (mg/mL of LNC) of 5-FU-C12-loaded LNC compared to 5-FU-C12 derivative,
5-FU aqueous solution and blank LNC on 9L and HCT-116 cell line.
Formulation
IC50 (5-FU µM) and CI
HCT-116
IC50 (mg/mL LNC) and CI
9L
-
9L
HCT-116
0.77
0.193
Blank LNC
-
(0.53, 0.97)
0.22
(0.17, 0.29)
(0.082, 0.457)
0.058
(0.033, 0.098)
2.24
(1.75, 2.86)
16.91
2.48
(1.45, 4.22)
20.72
5-FU-C12 LNC
5-FU-C12 solution
5-FU water solution
-
-
-
-
(15.50, 22.88)
9.09
(3.48, 23.78)
(11.20, 38.35)
4.08
(1.056, 21.86)
A moderate cytotoxic effect on 9L cell line of 5-FU compared to 5-FU-C12 solution was found, the
IC50 values being 9.09 and 16.91 µM, respectively. However, when the 5-FU-C12 was encapsulated
into LNC, the cytotoxic effect was more pronounced with an IC50 value of 2.24 µM. Cytotoxicity on
9L cell line was observed for blank LNC (non-loaded nanocarrier dispersions diluted in culture
medium at the same excipients concentration than that needed for loaded ones) only at high
concentration around 0.77 mg/mL.
The viability of HCT-116 cells exposed to different concentrations of 5-FU alone or encapsulated
into LNC was determined following 48 h of treatment. Pure 5-FU (water solution) was more
effective than 5-FU-C12 solution being the values of IC50 of 4.8 and 20.72 µM, respectively. 5-FU-
C12-loaded LNC showed a toxic effect more pronounced as compared with free drug being the IC50
value of 2.48 µM. Blank LNC were 3 times less toxic as compared to the loaded system being the
values 0.193 and 0.053 mg/mL, respectively (Table 5).
2.4.2 In vitro cytotoxicity on 3D cell growth
To establish the impact on tumor viability of our formulations on a more lifelike in vitro culture
system, we tested the dose response of 5-FU solution in comparison with the modified 5-FU-C12-
loaded LNC during a 48 h treatment on a MultiCellular Tumor Spheroids (MCTS) derived from the
HCT-116 cell line. The volume of MCTS was evaluated from phase contrast microscopy images as a
readout of cytotoxic effect. The data reported in Figure 7 represent normalized MCTS volume for
different drug treatments with respect to the control volume (MCTS not treated). Three different
drug concentrations were tested: 2, 10 and 50 µM of 5-FU, 5-FU-C12 and 5-FU-C12 –loaded LNC.
As showed in the Figure 7, loaded-LNC with modified 5-FU-C12 and 5-FU-C12 or 5-FU in solution
were able to reduce the volume of the spheroids after 2 days of treatment (D2) even at drug doses
of 2 µM. The cytotoxic effect was more pronounced at higher doses (10 and 50 µM).

Similar to the non-specific toxicity observed in 2D cell culture, blank LNC induced a slight reduction
of MCTS volume only at very high drug concentration.
Figure 7: Normalized volume of multicellular HT116 spheroids after treatment with blank LNC, LNC
loaded with 5-FU-C12, modified 5-FU-C12 dissolved in acetone and 5-FU clinical formulation.
Spheroids were treated with different drug concentrations (2, 10 and 50 µM) at different time
points (Day 0, Day 1 and Day 2).
3. Discussion
5-FU is a common chemotherapeutic agent used for the treatment of different cancers [27]. Its
rapid catabolism, short half-life (15-20 min), indiscriminate biodistribution, aspecific cytotoxicity
and myelosuppression impose the need to develop an alternative formulation and delivery system
for this anticancer drug [27]. When administered intravenously, approximately 90% of an injected
dose of 5-FU is metabolized to inactive 5-FUH2 by DPD in the liver, peripheral blood mononuclear
cells, intestinal mucosa, pancreas, lungs and kidneys, thus limiting its efficacy. Thus, the
maintenance of a therapeutic serum concentration requires continuous administration of high
doses of this drug with subsequent severe toxic effects [6].To solve the limitations related with 5-
FU administration, two main strategies have been investigated in the present work: i) the synthesis
of a novel 5-FU derivative and ii) the encapsulation of this active into lipid nanocapsules. As
compared to other lipid-based nanocarriers like liposomes, LNC display some convenient features;
in fact, they are prepared by an organic solvent free and soft-energy procedure and present great
storage stability [15]. In addition, physico-chemical properties of optimal LNC (i.e. size range,
polydispersity index) can be monitored and designed ad hoc, since they are strongly dependent on

the proportions of different components (oily phase, hydrophilic surfactant, aqueous phase - water
plus sodium chloride).
Different studies report attempts to improve the in vivo performances of 5-FU, through the
development of prodrugs and encapsulation in drug delivery nanosystems [8, 16, 28]. However, the
low encapsulation efficiency due to the inherent hydrophilic nature of 5-FU was observed for
different nanocarriers [16]. In our work, 5-FU was modified with a biocompatible moiety, lauric
acid, to obtain a lipid6drug conjugate having more affinity for the hydrophobic core of the LNC
(production yield of 70%). The easy synthesis and higher degree of purification ensured the
complete elimination of formaldehyde in the first step of the synthesis (Figure 1). 5-FU-C12 was
1
characterized by nuclear magnetic resonance (NMR) and the H NMR (Figure 2) indicates that only
one chain of lauric acid was conjugated to the N1 position. Previously, a derivative of 5-FU named
5-FUDIPAL obtained by conjugating palmitic acid to the drug was published. In this compound, two
lipid tails were attached to the lipid–5-FU conjugate (in N1 and N3 position) which imparted
hydrophobic characteristics to 5-FU [23]. In our case, the conjugation of only a single chain of lauric
acid to N1 position strongly affects the hydrophobicity of the novel compound, 5-FU-C12. Indeed,
pre-formulation solubility studies carried out in a polar solvent (water), non-polar solvents (acetone
and ethanol), oils and surfactants demonstrated that 5-FU-C12 was not soluble in water while
freely soluble in polar solvents and oils. Its hydrophobicity makes 5-FU-C12 a suitable candidate for
LNC encapsulation. In Figure 3, we demonstrated that LNC loaded with 5-FU derivative with a
hydrodynamic size of around 65 nm and a neutral surface charge. Using the PIT method, spherical
and monodispersed systems, also confirmed by cryoTEM images, were obtained. The conjugated
lauric acid tail was expected to elevate the non-covalent interaction between the drug and the oily
core of the system resulting in a high amount of drug loading and absence of fast release.
Moreover, the encapsulation provides protection to the drug from plasmatic degradation. 5-FU-
C12-loaded LNC were incubated with human plasma and the drug was detected until 3h following
incubation, while 5-FU-C12 was eliminated earlier being detected only 1h following incubation
(Figure 5). Considering that the plasmatic half-life of 5-FU is of around 30 min [29], the LNC here
developed led to an important increase in drug detection when incubated with plasma.
The biological evaluation of 5-FU-C12 on 2D cell culture measured by MTS showed that 5-FU-C12-
loaded LNC had an enhanced cytotoxic effect on 9L and HTC-116 cell line in comparison with
modified drug alone, being the IC50 values eight and ten times lower compared to the 5-FU-12
alone. Also, loaded-LNC were more cytotoxic with respect to 5-FU. It was reported that following 48
h of treatment on HCT-116 cells, the IC50 value of PLGA nanoparticles loaded with 5-FUDIPAL, a
similar palmitic acid-conjugate, was around 23 µM [23], twenty times higher in comparison to the
IC50 value of 5-FU-C12-loaded LNC (IC50 around 2 µM). Besides, 5-FU-loaded pH-sensitive

liposomes having different lipid compositions had been also described and their activity tested on
HCT-116 cell line. The authors found differences in cell sensitivity when treated with the
formulation obtained and ascribed the differences and the resulting low anti-cancer activity of
liposomes to the low entrapment efficiency of 5-FU [30]. Low encapsulation efficiency and
premature release of the drug strongly affect the efficacy of the nanosystems encapsulating 5-FU in
these nanosystems. Comparing the results on HCT-116 cell line to other nanosystems loaded with
5-FU derivative, the encapsulation into LNC was more effective in reducing cell viability. 5-FU-C12
loaded LNC here developed were able to encapsulate in a high amount the novel hydrophobic
compound and to retain the drug in the oil core even when diluted in simulated physiological
media, therefore promising to increase the half-life of the drug.
In a 2D cell model free drug or nanosystems had to enter into a cell monolayer to exert their
cytotoxic activity. The difference in the cytotoxic effect is related to the stability, solubility, release
and internalization ability of drug when entering cells. From our result is evident that the
encapsulation of the 5-FU-C12 enhanced the stability and internalization of the drug, being the
effect of drug loaded LNC more pronounced in comparison to free drug.
Once demonstrated the enhanced cytotoxic effect of drug loaded-LNC on 2D cell culture, HCT-116
cells were used to form 3D spheroids and the effect of 2 day-treatment, with 5-FU-C12 loaded LNC,
5-FU or 5-FU-C12 alone, was investigated. The aim was to test the LNC developed in a more
complex system that mimics the 3D chemical and physical gradients that occur in in vivo tumors.
The volume of the spheroids was measured as a readout for cytotoxicity after 24 and 48 h of
treatment. As showed in Figure 7, for all the treatments we observed a reduction of MCTS volume
as time increases, demonstrating the inhibition of MCTS growth. Interestingly, the effect of drugs
alone and 5-FU-C12 loaded LNC had a similar effect on MCTS volume, suggesting a comparable
efficacy in 3D systems. Moreover, the effect of blank LNC on MCTS growth is less pronounced than
that of loaded LNC. This observation is particularly evident at day 2 at the highest concentration
tested (0.875 mg/mL). This result contrasts with the observations made on the 2D cell model,
where the non-specific toxicity of blank LNC appears at lower concentration (0.193 mg/mL). This
difference should be analyzed considering that we used different readouts for the efficacy in 2D
and 3D cell model [31]. In 2D we evaluated cell death through a metabolic essay (MTS) while in 3D
we evaluated MCTS growth, which results from the equilibrium between cell death and
proliferation. The difference we observe thus suggests that blank LNC causes cell death, which is
the origin of their toxicity, but does not affect cell proliferation. In future works, there are many
parameters that should be investigated to confirm this analysis, such as permeability of the
spheroids to the treatment [32, 33].

Globally, from the preliminary results, we observed that the encapsulation into LNC does not
hamper the penetration into the spheroid, and enhance cytotoxic effect respect to the other
treatments. Further in depth studies taking into account permeability, distribution of the drug and
mechanism of internalization of drug-loaded LNC into MCTS has to be performed to fully disclose
the efficacy of this system.
4. Conclusion
In the present work, we reported evidence about the feasibility of an oily-core nanoformulation for
5-FU-derivative encapsulation. Using a simple synthesis process, we developed with a high yield of
production a novel lipophilic 5-FU derivative namely 5-FU-C12, which was successfully encapsulated
into LNC. The novel formulation obtained enhanced the solubility and the stability of the drug when
in contact with plasma. Further, in vitro studies were carried out in different tumor cell lines (9L and
HCT-116) to assess the cytotoxic activity of the encapsulated drug with respect to the drug in
solution or the commercial one. 5-FU-C12-loaded LNC resulted in a higher cytotoxic effect
compared with 5-FU-C12 alone or 5-FU while blank LNC showed cytotoxic ability only at very high
concentrations indicating possible limited toxic effects of the carriers. Then, to obtain reliable and
detailed information about LNC-tumor interaction in a more physiologic situation, we analyzed the
effects of the treatments on HCT-116 3D-spheroid cultures. The results evidenced that the effect on
spheroids survival was at least similar between drug-loaded LNC, 5-FU-C12 and 5-FU alone. We can
therefore state that 5-FU-C12 encapsulation in a complex system does not reduce the penetration
capacity into the spheroids. Further research will be needed to make these nanosystems not only
comparable to solution treatments in their effect on 3D systems but to make them more effective.
Acknowledgments
Authors are thankful to Pierre-Yves Dugas from the Univ Lyon, Université Claude Bernard Lyon 1,
CNRS, C2P2 UMR 5265, for his kind help and expertise for cryoTEM observation.
This work has been carried out within the research program of NICHE, financially supported by
EuroNanoMed-II (4th call), RESOLVE, financially supported by EuroNanoMed-III (8th call) and ITMO
Cancer within the frame of the Programme Plan Cancer 2014-2019 (no. PC201507).

Bibliography
1.
2.
Longley DB, Harkin DP, Johnston PG. 5-Fluorouracil: mechanisms of action and
clinical strategies [Review Article]. Nature Reviews Cancer. 2003;3:330. doi:
10.1038/nrc1074.
Vincent J, Mignot G, Chalmin F, et al. 5-Fluorouracil Selectively Kills Tumor-
Associated Myeloid-Derived Suppressor Cells Resulting in Enhanced T Cell–
Dependent Antitumor Immunity. Cancer Research. 2010;70(8):3052-3061. doi:
10.1158/0008-5472.can-09-3690.
3.
4.
Ugel S, Delpozzo F, Desantis G, et al. Therapeutic targeting of myeloid-derived
suppressor cells. Current Opinion in Pharmacology. 2009 2009/08/01/;9(4):470-481.
doi: https://doi.org/10.1016/j.coph.2009.06.014.
Neto OV, Raymundo S, Franzoi MA, et al. DPD functional tests in plasma, fresh
saliva and dried saliva samples as predictors of 5-fluorouracil exposure and
occurrence of drug-related severe toxicity. Clinical Biochemistry. 2018 2018/04/04/.
doi: https://doi.org/10.1016/j.clinbiochem.2018.04.001.
5.
Yang CG, Ciccolini J, Blesius A, et al. DPD-based adaptive dosing of 5-FU in patients
with head and neck cancer: impact on treatment efficacy and toxicity. Cancer
Chemotherapy and Pharmacology. 2011 2011/01/01;67(1):49-56. doi:
10.1007/s00280-010-1282-4.
6.
7.
8.
9.
Lee JJ, Beumer JH, Chu E. Therapeutic drug monitoring of 5-fluorouracil. Cancer
Chemotherapy and Pharmacology. 2016 2016/09/01;78(3):447-464. doi:
10.1007/s00280-016-3054-2.
S E Ward EK, J Cowan,M Marples, B Orr, and M T Seymour. The clinical and
economic benefits of capecitabine and tegafur with uracil in metastatic colorectal
cancer. Br J Cancer. 2006;3(95):1.
Álvarez P, Marchal JA, Boulaiz H, et al. 5-Fluorouracil derivatives: a patent review.
Expert Opinion on Therapeutic Patents. 2012 2012/02/01;22(2):107-123. doi:
10.1517/13543776.2012.661413.
P Cellier BL, L Martin, B Vié, C Chevelle, V Vendrely, A Salemkour, C Carrie, G Calais,
P Burtin, L Campion, M Boisdron-Celle, A Morel, V Berger, E Gamelin. Phase II study
of preoperative radiation plus concurrent daily tegafur-uracil (UFT) with leucovorin
for locally advanced rectal cancer. BMC Cancer. 2011;11:98.
10.
Zhang M WW, Liu J, Yang H, Jiang Y, Tang W, Li Q, Liao X. Comparison of the
effectiveness and toxicity of neoadjuvant chemotherapy regimens,
capecitabine/epirubicin/cyclophosphamide
fluorouracil/epirubicin/cyclophosphamide,
vs
5-
followed
by
adjuvant,
capecitabine/docetaxel vs docetaxel, in patients with operable breast cancer. Onco
Targets Ther. 2016 8(9):3443-50.
Henderson LA, Shankar LK. Clinical Translation of the National Institutes of Health’s
Investments in Nanodrug Products and Devices. The AAPS Journal. 2017
2017/03/01;19(2):343-359. doi: 10.1208/s12248-016-9995-x.
Grabbe S, Haas H, Diken M, et al. Translating nanoparticulate-personalized cancer
vaccines into clinical applications: case study with RNA-lipoplexes for the treatment
of melanoma. Nanomedicine. 2016;11(20):2723-2734. doi: 10.2217/nnm-2016-
0275. PubMed PMID: 27700619.
11.
12.

13.
14.
Sykes EA, Dai Q, Sarsons CD, et al. Tailoring nanoparticle designs to target cancer
based on tumor pathophysiology. Proceedings of the National Academy of Sciences.
2016;113(9):E1142-E1151. doi: 10.1073/pnas.1521265113.
Lollo G, Hervella P, Calvo P, et al. Enhanced in vivo therapeutic efficacy of
plitidepsin-loaded nanocapsules decorated with a new poly-aminoacid-PEG
derivative. International Journal of Pharmaceutics. 2015 2015/04/10/;483(1):212-
219. doi: https://doi.org/10.1016/j.ijpharm.2015.02.028.
15.
16.
17.
18.
Sasso MS, Lollo G, Pitorre M, et al. Low dose gemcitabine-loaded lipid nanocapsules
target monocytic myeloid-derived suppressor cells and potentiate cancer
immunotherapy.
Biomaterials.
2016
2016/07/01/;96:47-62.
doi:
https://doi.org/10.1016/j.biomaterials.2016.04.010.
Fanciullino R, Mollard S, Giacometti S, et al. In Vitro and In Vivo Evaluation of
Lipofufol, a New Triple Stealth Liposomal Formulation of Modulated 5-Fu: Impact
on Efficacy and Toxicity. Pharmaceutical Research. 2013 2013/05/01;30(5):1281-
1290. doi: 10.1007/s11095-012-0967-2.
Roullin V-G, Deverre J-R, Lemaire L, et al. Anti-cancer drug diffusion within living rat
brain tissue: an experimental study using [3H](6)-5-fluorouracil-loaded PLGA
microspheres. European Journal of Pharmaceutics and Biopharmaceutics. 2002
2002/05/01/;53(3):293-299. doi: https://doi.org/10.1016/S0939-6411(02)00011-5.
Patel MN, Lakkadwala S, Majrad MS, et al. Characterization and Evaluation of 5-
Fluorouracil-Loaded Solid Lipid Nanoparticles Prepared via a Temperature-
Modulated
Solidification
Technique.
AAPS
PharmSciTech.
2014
2014/12/01;15(6):1498-1508. doi: 10.1208/s12249-014-0168-x.
19.
20.
Liu Z, Fullwood N, Rimmer S. Synthesis of allyloxycarbonyloxymethyl-5-fluorouracil
and copolymerizations with -vinylpyrrolidinone [10.1039/B002092N]. Journal of
Materials Chemistry. 2000;10(8):1771-1775. doi: 10.1039/b002092n.
Liu Z, Rimmer S. Synthesis and release of 5-fluorouracil from poly(N-
vinylpyrrolidinone) bearing 5-fluorouracil derivatives. Journal of Controlled Release.
2002 2002/05/17/;81(1):91-99. doi: https://doi.org/10.1016/S0168-3659(02)00048-
2.
21.
22.
B. Heurtault PS, B. Pech, J. E. Proust, and J. P. Benoit. A novel phase inversion -
based process for the preparation of lipid nanocarrier. Pharm Res. 2002;19(6):875–
880.
Moysan E, Gonzalez-Fernandez Y, Lautram N, et al. An innovative hydrogel of
gemcitabine-loaded lipid nanocapsules: when the drug is a key player of the
nanomedicine structure [10.1039/C3SM52781F]. Soft Matter. 2014;10(11):1767-
1777. doi: 10.1039/c3sm52781f.
23.
Ashwanikumar N, Kumar NA, Asha Nair S, et al. 5-Fluorouracil–lipid conjugate:
Potential candidate for drug delivery through encapsulation in hydrophobic
polyester-based nanoparticles. Acta Biomaterialia. 2014 2014/11/01/;10(11):4685-
4694. doi: https://doi.org/10.1016/j.actbio.2014.07.032.
24.
25.
Virgone-Carlotta A, Lemasson M, Mertani HC, et al. In-depth phenotypic
characterization of multicellular tumor spheroids: Effects of 5-Fluorouracil. PLOS
ONE. 2017;12(11):e0188100. doi: 10.1371/journal.pone.0188100.
Lollo G, Vincent M, Ullio-Gamboa G, et al. Development of multifunctional lipid
nanocapsules for the co-delivery of paclitaxel and CpG-ODN in the treatment of

glioblastoma.
International
Journal
of
Pharmaceutics.
2015
2015/11/30/;495(2):972-980. doi: https://doi.org/10.1016/j.ijpharm.2015.09.062.
Wibroe PP, Ahmadvand D, Oghabian MA, et al. An integrated assessment of
morphology, size, and complement activation of the PEGylated liposomal
doxorubicin products Doxil®, Caelyx®, DOXOrubicin, and SinaDoxosome. Journal of
26.
Controlled
Release.
2016
2016/01/10/;221:1-8.
doi:
https://doi.org/10.1016/j.jconrel.2015.11.021.
27.
28.
29.
Gusella M, Crepaldi G, Barile C, et al. Pharmacokinetic and demographic markers of
5-fluorouracil toxicity in 181 patients on adjuvant therapy for colorectal cancer.
Annals of Oncology. 2006;17(11):1656-1660. doi: 10.1093/annonc/mdl284.
Carrillo E, Navarro SA, Ramírez A, et al. 5-Fluorouracil derivatives: a patent review
(2012
–
2014). Expert Opinion on Therapeutic Patents. 2015
2015/10/03;25(10):1131-1144. doi: 10.1517/13543776.2015.1056736.
Bocci G, Danesi R, Di Paolo A, et al. Comparative Pharmacokinetic Analysis of 5-
Fluorouracil and Its Major Metabolite 5-Fluoro-5,6-dihydrouracil after Conventional
and Reduced Test Dose in Cancer Patients. Clinical Cancer Research.
2000;6(8):3032-3037.
30.
31.
Udofot O AK, Israel B, Agyare E. Cytotoxicity of 5-fluorouracil-loaded pH-sensitive
liposomal nanoparticles in colorectal cancer cell lines. Integr Cancer Sci Ther.
2015;2(5):245-252.
Solomon MA, Lemera J, D’Souza GGM. Development of an in vitro tumor spheroid
culture model amenable to high-throughput testing of potential anticancer
nanotherapeutics. Journal of Liposome Research. 2016 2016/07/02;26(3):246-260.
doi: 10.3109/08982104.2015.1105820.
32.
33.
Ho WY, Yeap SK, Ho CL, et al. Development of Multicellular Tumor Spheroid (MCTS)
Culture from Breast Cancer Cell and a High Throughput Screening Method Using the
MTT Assay. PLOS ONE. 2012;7(9):e44640. doi: 10.1371/journal.pone.0044640.
PubMed PMID: PMC3435305.
Takechi-Haraya Y, Goda Y, Sakai-Kato K. Control of Liposomal Penetration into
Three-Dimensional Multicellular Tumor Spheroids by Modulating Liposomal
Membrane Rigidity. Molecular Pharmaceutics. 2017 2017/06/05;14(6):2158-2165.
doi: 10.1021/acs.molpharmaceut.7b00051.