Synthesis of octadecylamine-retinoic acid conjugate for enhanced cytotoxic effects of 5-FU using LDL targeted nanostructured lipid carriers

Article abstract of DOI:10.1016/j.ejmech.2012.05.024

The aim of the present study was to reduce 5-FU side effects by targeted nanostructured lipid carriers (NLCs) to LDL receptors that are over expressed in colorectal carcinoma and also use of a new synthesized conjugate of retinoic acid as a cytotoxic agent. Fatty acyl amide derivative of retinoic acid was synthesized by its conjugation to octadecylamine with the expectation to improve its loading capacity in NLCs of 5-FU. The NLCs were prepared by an emulsification-solvent evaporation method using cholesterol and cholesteryl stearate. Physical properties and drug release were studied in NLCs. The cytotoxicity of NLCs loaded with 5-FU and retinoic acid conjugate was studied on colon cancer cells (HT29) using MTT assay. To confirm that drug targeting has been done through LDL receptors, APO-E was omitted from the cell culture and the MTT assay was repeated. FTIR and ¹H NMR spectra confirmed successful production of the conjugate. Results showed the IC₅₀ of free 5-FU was about 7.6 μM while in comparable concentration, the cytotoxicity of 5-FU loaded in NLCs containing the retinoic acid conjugate was nearly 2 fold of NLCs just loaded with 5-FU and more than 5 fold of free 5-FU. The retinoic acid conjugate loaded NLCs prepared by cholesterol can target LDL receptors of HT29 cells and seems promising in reducing 5-FU dose in colorectal cancer.

Full text of DOI:10.1016/j.ejmech.2012.05.024

European Journal of Medicinal Chemistry 54 (2012) 429e438
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Synthesis of octadecylamine-retinoic acid conjugate for enhanced cytotoxic
effects of 5-FU using LDL targeted nanostructured lipid carriers
,
b
b
a
Jaleh Varshosaz a , Farshid Hassanzadeh , Hojjat Sadeghi , Sare Andalib
*
a Department of Pharmaceutics, Faculty of Pharmacy and Novel Drug Delivery Systems Research Center, Isfahan University of Medical Sciences, Isfahan, Iran
b Department of Medicinal Chemistry, Faculty of Pharmacy and Isfahan Pharmaceutical Sciences Research Center, Isfahan University of Medical Sciences, Isfahan, Iran
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
< Retinoic acid was conjugated to
octadecylamine.
< 5-FU and retinoic acid conjugate
were loaded in cholesterol NLCs.
< NLCs targeted LDL receptors of HT29
cells.
< IC50 of NLCs of retinoic acid conju-
gate and 5-FU was more than 5 fold
of free 5-FU.
< NLCs of 5-FU and retinoic acid
conjugate seems promising in colo-
rectal cancer.
a r t i c l e i n f o
a b s t r a c t
Article history:
The aim of the present study was to reduce 5-FU side effects by targeted nanostructured lipid carriers
(NLCs) to LDL receptors that are over expressed in colorectal carcinoma and also use of a new synthesized
conjugate of retinoic acid as a cytotoxic agent. Fatty acyl amide derivative of retinoic acid was synthe-
sized by its conjugation to octadecylamine with the expectation to improve its loading capacity in NLCs
of 5-FU. The NLCs were prepared by an emulsification-solvent evaporation method using cholesterol and
cholesteryl stearate. Physical properties and drug release were studied in NLCs. The cytotoxicity of NLCs
loaded with 5-FU and retinoic acid conjugate was studied on colon cancer cells (HT29) using MTT assay.
To confirm that drug targeting has been done through LDL receptors, APO-E was omitted from the cell
culture and the MTT assay was repeated. FTIR and 1H NMR spectra confirmed successful production of
Received 27 January 2012
Received in revised form
17 May 2012
Accepted 18 May 2012
Available online 28 May 2012
Keywords:
5-FU
Retinoic acid
Targeted NLCs
Colorectal cancer
Cytotoxicity
HT29 cell line
Octadecylamine
the conjugate. Results showed the IC50 of free 5-FU was about 7.6 mM while in comparable concentration,
the cytotoxicity of 5-FU loaded in NLCs containing the retinoic acid conjugate was nearly 2 fold of NLCs
just loaded with 5-FU and more than 5 fold of free 5-FU. The retinoic acid conjugate loaded NLCs
prepared by cholesterol can target LDL receptors of HT29 cells and seems promising in reducing 5-FU
dose in colorectal cancer.
Ó 2012 Elsevier Masson SAS. All rights reserved.
* Corresponding author. Department of Pharmaceutics, Faculty of Pharmacy and
Novel Drug Delivery Systems Research Center, Isfahan University of Medical
Sciences, Isfahan, PO Box 81745-359, Iran. Tel.: þ98 311 7922579; fax: þ98
311668001.
1. Introduction
Cancer is the second cause of death after cardiac disease in
America [1]. Colorectal cancer is the third most common form of
E-mail address: varshosaz@pharm.mui.ac.ir (J. Varshosaz).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2012.05.024

430
J. Varshosaz et al. / European Journal of Medicinal Chemistry 54 (2012) 429e438
cancer and cause of death due to cancer in the United States.
Colorectal cancer by incidence of 1 million new cases in each year
affects men and women in world wide. Adenomatus polyps and
malignant cells that are formed in colon replicate so fast and
become tumors that can spread to other sites [2] indicating an
urgent need for more effective colon cancer chemotherapy. Tar-
geted therapy for colorectal cancer allows for the local high
concentration of chemotherapeutic drugs and reduces their side
effects [2].
Nanostructured lipid carriers (NLCs) are a new generation of
colloidal drug carriers. These nanostructures contain liquid lipid
that improves drug loading and stability of nanoemulsion [3,4] due
to an unjustified, imperfect lipid matrix having the space for the
molecules of drug.
5-FU is a pyrimidine analog that is used in different solid tumors
of colon, liver, neck, head and breast cancers. 5-FU is that inhibits
the activity of thymidylate synthetase. Unfortunately, 5-FU shows
many disadvantages such as: inactivation by dihydropyrimidine
dehydrogenase that results in inadequate absorption by gastroin-
testinal tract, very short half life and the toxic effects on bone
marrow and un-selectivity on normal cells. For these reasons
scientists try to enhance the efficacy of this drug by increasing the
time of circulation and reducing side effects by localizing the drug
to the infected cells by target therapy [1,2].
enhance its solubility in the mixture of solvents used for the
production of NLCs, so higher entrapment of retinoic acid caused
better uptake of retinoic acid by the cells.
2. Materials and methods
2.1. Materials
5-FU was provided from Sigma (USA), Retinoic acid was
purchased from Solmag Chemical Company (Italy), cholesterol,
oleic acid, octanol, Tween 20, Tween 80, chloroform and silica gel
F254 60 were from Merck Chemical Company (Germany). Soy leci-
thin S100 was from Lipoid (Germany), Dicyclohexyl carbodiimide
(DCC) from Fluka (USA), cholesteryl stearate (CS) from Aldrich (US),
N-hydroxysuccinimide (NHS), dimethyl sulfoxide (DMSO), meth-
anol, thriethyl amine, thiazolyl blue tetrazolium bromide (MTT)
and octadecylamine were from Sigma (USA). HT29 cell line from
Pasture institute (Iran), Tripsin/EDTA, and RPMI 1640 from PAA
Company (Austria). Steptomycin/Pencilline from Gibko (US), 96
well plate from Nunck Company (Denmark).
2.2. Conjugation of retinoic acid and octadecylamine
Retinoids are a family of chemicals containing vitamin A and its
natural and synthetic derivatives that are some important
chemotherapeutic and chemo preventive agents for a lot of cancers
like leukemia and some forms of skin cancer [5,6]. Between retinoic
Fig. 1 shows the chemical structure of retinoic acid and octa-
decylamine. For preparation of this conjugate 500 mg of retinoic
acid, 330 mg of DCC and 400 mg of NHS were added to 15 mL of
DMSO. This solution was shaken at room temperature for 72 h until
retinoic acid was activated. After 72 h the precipitate of dicyclo-
hexyl urea was separated by filtration. Then this solution was
dripped into the chloroform solution of octadecylamine (425 mg/
17 mL). After 24 h shaking, the solvents were evaporated under
vacuum distillation.
acid receptors that are types of
sensitivity to retinoic acid in cell lines and mediates retinoid
activity [7]. It is shown that the expression rate of RAR- in tumor
tissues is significantly lower than those of both normal and adja-
cent tissues. The expression rate of RAR- decreases in human
colorectal tumor tissues, suggesting RAR- may be involved in the
formation of human colorectal cancer. RAR- may become a new
a, b and g, the type b is necessary for
b
b
b
b
prognostic indicator of human colorectal cancer [8].
2.3. Characterization of synthesized derivative of retinoic acid
Unfortunately, chemotherapy with retinoic acid results in
undesirable side effects and retinoic acid-resistant tumors are very
common. Retinol (vitamin A) is thought to exert its effects through
the actions of its metabolite, all-trans-retinoic acid (ATRA). The
ability of retinol to inhibit the growth of ATRA-sensitive and ATRA-
resistant human colon cancer cell lines was examined by Park et al.
[9]. Retinol inhibited cell growth in a dose-responsive manner.
Conjugation of some drugs to fatty acids has been reported to
enhance their stability, solubility, short half life, and their cellular
uptake. For example the conjugate of paclitaxel to a natural fatty
acid has enhanced its tumor accumulation [10] or fatty acyl amide
derivative of doxorubicin i.e., dodecanoyl-doxorubicin is consis-
tently the most effective derivative which inhibited the prolifera-
tion of colon and ovarian cancer cells [11]. Also fatty acyl derivatives
of cytarabine improved cellular uptake with a longer duration of
action [12].
For the purification and identification of the product a prepara-
tive TLC (silica gel f254 60) was used. Fifty mg of the solid mixture
collected from the previous stage was dissolved in the minimum
amount of ether and was implanted on the preparative TLC with the
mobile phase of chloroform. The middle spot with the Rf ¼ 0.5 was
carved and the silica gel was removed.
FTIR and 1H NMR spectra were used for analysis of synthesized
retinoic acid-octadecylamine. The samples were measured at 298ꢀk
with about 5 wt% CDCl3 or (CD3)2SO solution using NMR Spec-
trometer (AC-80, Bruker Biospin, Germany).
a
CH3
CH3 O
Expression of LDL receptor is higher in some cancer cells like
leukemia, breast cancer and human lung cancer tissues [13]. Also in
human colon carcinoma LDL receptor mRNA expression is signifi-
cantly increased so using the cholesterol as a drug carrier for
treatment of colorectal carcinomas seems logic [14]. Apolipoprotein
E (Apo-E) is a class of apolipoprotein found in the chylomicron, that
binds to LDL receptors on liver cells and peripheral cells. It is
essential for the normal catabolism of triglyceride-rich lipoprotein
constituents [15]. For these reasons in the present study the LDL
receptors of colorectal cancer cells were targeted by cholesterol.
RAR-independent inhibitory effect of retinoic acid on cell growth
on colon cancer was also used to enhance the cytotoxic effect of 5-
FU. To do so retinoic acid was conjugated to octadecylamine first to
H3C CH3
OH
CH3
b
NH2
Fig. 1. Chemical structure of a) retinoic acid and b) octadecylamine.

J. Varshosaz et al. / European Journal of Medicinal Chemistry 54 (2012) 429e438
431
Table 1
59.5% ꢁ 5% of lipid (cholesteryl stearate or cholesterol), 0.5% of lecithin,
10% of PEG 40 stearate (as the pegylated lipid to help escaping NLCs
from reticuloendothelial system),10% of the synthesized derivative i.e.,
retinoic acid-octadecylamine, 20% ꢁ 5% of oil (oleic acid or octanol) and
20 mg of 5-FU were dissolved in 5 mL of the mixture of acetone/ethanol
by the ratio of 3:1. Then this organic phasewasslowlyadded to 25 mL of
distilled water containing 0.5% of Tween 80 while stirred on a magnetic
stirrer during 15 min. The organic solvent was allowed to evaporate for
1 h using a magnetic stirrer with a minimum speed. The effects of oil
type, lipid type and their contents were investigated and nanoparticles
with the greatest drug loading efficiency, the least particle size, the
greatest absolute value of zeta potential and the greatest drug release
were considered as the optimum formulation.
Different formulations of prepared NLCs (containing 20 mg 5-FU in 25 mL of NLCs
dispersion).
Formulation
code
Octanol
(OC) %
Oleic acid
(OA) %
Cholesterol
(C) %
Cholesteryl
stearate (CS) %
CS64.5OA15
CS59.5OA20
CS54.5OA25
e
15
20
25
15
20
25
e
e
64.5
59.5
54.5
e
e
e
e
e
C
64.5OA15
e
64.5
59.5
54.5
e
C59.5OA20
C54.5OA25
CS64.5OC15
CS59.5OC20
CS54.5OC25
C64.5OC15
C59.5OC20
e
e
e
e
15
20
25
15
20
25
64.5
59.5
54.5
e
e
e
e
e
e
64.5
59.5
54.5
e
e
C
54.5OC25
e
e
2.5. Particle size and zeta potential measurements
The meanparticle size and zetapotential of NLCs were measured by
photon correlation spectroscopy (PCS) at a fixed angle of 90ꢀ (Zetasizer
3000HS, Malvern Instrument, UK). Nanodispersion was suitably
diluted to measure mean particle size and polydispersity index.
For this purpose the samples were diluted 10 fold with de-ionized
water before measurements of particle size and zeta potential.
2.4. Preparation of NLCs
Among all liquid lipids selected for solubility screening, 5-FU had
the highest solubility in octanol and oleic acid. Therefore, these two
liquid lipids were selected for preparation of NLCs. In a preliminary test
seven factors each in 2 levels including: lipid type and content, oil type
and content, lecithin and PEG 40 stearate contents and the ratio of
acetone/ethanol were first screened by the Taguchi design using an L8
orthogonal array to find the most effective factors on 3 responses of
particle size, drug loading and release efficiency. This screening test
showed the most important effective factors were the lipid and oil type
and their content (data are not shown here). Therefore, they were
selected for further optimization studies. The content of 5-FU was fixed
at 20 mg as higher amount was not soluble in the organic phase. Ret-
inoic acid-octadecylamine conjugate concentrationwas also fixed at 10
weight percent of the NLCs as this study was a preliminary study to see
whether it can increase cytotoxicity of 5-FU or not and of course further
studies are needed to optimize its cytotoxic concentrations which has
synergistic effectwith 5-FU. Accordingtothe formulation type (Table 1)
2.6. Entrapment efficiency and drug loading
For the quantitative determination of 5-FU a spectrophotometric
method was used. The amount of the loaded 5-FU into the NLCs was
determined indirectly as follows: 600
mL of sample emulsion and
600 L of blank emulsion were centrifuged (Microcentrifuge Sigma
m
30k, UK) at 10000 rpm for 5 min to separate nanoparticles. The
supernatant containing the free drug was diluted 40 times and
analyzed using a spectrophotometer (RF-5301 PC, Shimadzu, Kyoto,
Japan) at lmax ¼ 267 nm. The difference between the total and free
drug shows the amount of the encapsulated drug. The encapsulation
efficiency (EE) of 5-FU in NLCs was determined as the ratio between
the actual and theoretical loading by using the following equation:
CH3
CH3 O
H3C
CH3
OH
CH3
CH3
O
O
H3C
CH3
CH3
DCC
O
N
CH3
O
O
HN
O
CH3
CH3
O
O
H3C CH3
O
NH2
CH3
N
O
CH3
CH3 O
H3C
CH3
N
H
O
HN
O
CH3
Fig. 2. Schematic representation of chemical reaction for preparation of retinoic acid-octadecylamine conjugate.

432
J. Varshosaz et al. / European Journal of Medicinal Chemistry 54 (2012) 429e438
ꢀ ꢁ
a mixture of acetone/ethanol and the content of loaded 5-FU in NLCs
was measured. It was in accordance to the indirect method. For this
reason the indirect method was used throughout the study due to its
easiness.
Entrapped drug in NLCs
Total amount of drug added
EE %
¼
ꢂ 100
(1)
Drug loading capacity (DL) was calculated according to the Eq.
(2). The total amount of the drug, lipid and oil excipients added
during preparation were considered as the total weight of NLCs:
2.7. Drug release studies
ꢀ ꢁ
EntrappeddruginNLCs
To study the drug release from nanoparticles, 1 mL of dispersion
was transferred to a dialysis tube (molecular weight cutoff 12000,
Membra-CelÒ, Viskase, USA), then the sealed tube was put into
a beaker of 70 mL of phosphate buffer solution (pH 7.4) containing
2% Tween 20. Samples were shaken horizontally in a shaker bath
DL %
¼
Weightof NLCsðamountof drugþamountof lipidÞ
ꢂ100
(2)
(Lab tech, Korea) at 37 ꢁ 1 ꢀC with 40 strokes per minute. 600
mL of
The validity of this method was checked by directly measuring the
entrapped drug in the separated NLCs. So that they were dissolved in
the medium was taken at predetermined time intervals and the
Fig. 3. FTIR spectra of a) octadecylamine, b) retinoic acid and c) octadecylamine-retinoic acid conjugate.

J. Varshosaz et al. / European Journal of Medicinal Chemistry 54 (2012) 429e438
433
absorbance of free 5-FU was measured at lmax ¼ 268 nm. The
2.8. Cell culture
samples were returned to the test medium again. The parameter of
release efficiency within 20 h (RE20%) was used to compare the
release profiles [16]:
Human colon cancer cell line HT29 was used in this study. The cells
were cultured on RPMI 1640 containing 10% FBS (Fetal Bovine Serum)
and 1% antibiotics mixture of penicillin (10000 U/mL) and streptomy-
cine (10000 m
g/mL) at 37 ꢀC and in 5% CO2. HT29 cells are from colon
t
Z
epithelial like cells in culture medium. First 180 mL of the suspension of
ydt
cells at a density of 5 ꢂ104 cells/ml were seeded into each well of a 96-
well culture plate (SPL Lifescience, Korea) and incubated for 24 h at
37 ꢀC in 5% CO2 and 100% humidity before cell viability test.
0
RE% ¼
ꢂ 100
100t
(3)
y
Fig. 4. 1H NMR spectra of a) octadecylamine, b) retinoic acid and c) octadecylamine-retinoic acid conjugate.

434
J. Varshosaz et al. / European Journal of Medicinal Chemistry 54 (2012) 429e438
Table 2
Physical properties of different NLCs.
Formulation code
Intensity Z-average
PdI
Zeta potential (mv)
Drug loading efficiency %
RE20 %
particle size (nm)
CS64.5OA15
CS59.5OA20
CS54.5OA25
C64.5OA15
226.5 ꢁ 2.1
130.9 ꢁ 2.4
135.3 ꢁ 3.9
119.0 ꢁ 0.9
102.0 ꢁ 1.3
105.8 ꢁ 1.9
257.4 ꢁ 27.9
150.6 ꢁ 8.8
284.1 ꢁ 16.8
105.5 ꢁ 1.0
80.6 ꢁ 0.5
0.28
0.42
0.30
0.12
0.18
0.23
0.39
0.44
0.48
0.24
0.21
0.40
ꢃ14.2 ꢁ 2.3
ꢃ26.0 ꢁ 1.9
ꢃ9.1 ꢁ 0.6
ꢃ20.8 ꢁ 1.1
ꢃ29.0 ꢁ 3.0
ꢃ25.1 ꢁ 0.7
ꢃ19.3 ꢁ 1.4
ꢃ18.3 ꢁ 1.4
ꢃ7.7 ꢁ 1.6
ꢃ20.1 ꢁ 0.7
ꢃ8.6 ꢁ 1.0
ꢃ14.5 ꢁ 1.0
48 ꢁ 2.3
30 ꢁ 1.7
34 ꢁ 1.0
30 ꢁ 1.3
30 ꢁ 1.7
38 ꢁ 3.1
38 ꢁ 1.3
51 ꢁ 2.9
29 ꢁ 1.9
38 ꢁ 2.3
22 ꢁ 1.8
36 ꢁ 2.6
51.53 ꢁ 3.20
48.37 ꢁ 1.03
48.35 ꢁ 0.78
35.33 ꢁ 0.89
32.69 ꢁ 2.00
37.98 ꢁ 1.98
40.59 ꢁ 1.77
44.46 ꢁ 2.11
51.53 ꢁ 2.82
32.94 ꢁ 2.26
10.21 ꢁ 1.31
48.37 ꢁ 0.87
C
59.5OA20
C54.5OA25
CS64.5OC15
CS59.5OC20
CS54.5OC25
C64.5OC15
C59.5OC20
C54.5OC25
110.1 ꢁ 1.5
Considering that cholesterol is fitted to the LDL receptors via Apo-E
protein, the cellular viability test was carried out when Apo-E was
removed from the cell culture medium to study if the targeting of
nanoparticles is through the LDL receptors or just passive penetration
of NLCs causes the cellular uptake of drug. Actually we expected that if
Apo-E peptide is omitted from the culture medium the intermediate
ligand for linking cholesterol NLCs is absent and consequently they are
nottakenup bythe LDL receptors. Toshow this thecells weretreated by
the RPMI medium that was without serum, for 48 h.
Mean of each group ꢃ mean of blank
Cell survival ¼
mean of negative control ꢃ mean of blank
(4)
2.10. Atomic force microscopy (AFM)
Observation of the morphology and particle size of NLCs was per-
formed by an Atomic Force Microscope (Bruker, Nanos 1.1, Germany).
70
2.9. Cytotoxicity assessment-MTT assay
a
60
C64.5OC15
After cells were seeded on 96 well plate each raw was treated
50
with: 20
or without retinoic acid-derivative, 20
2, 1/4, 1/8) of dialyzed NLCs loaded with 5-FU with or without reti-
noic acid-derivative, 20 L of 8.2 M solution of free 5-FU (as positive
control) to reach the same concentration of 5-FU loaded in NLCs, and
20 L of culture medium (as negative control). Then the plate was
incubated for 72 h and after that 20 L of MTT was added. After 3 h
incubation the cell medium was removed cautiously while not
allowing the produced Formazan crystals to pour. Then 180 L of
m
L of three concentrations (1/2,1/4,1/8) of blank NLCs with
C59.5OC20
mL of three concentrations (1/
C54.5OC25
C64.5OA15
40
30
m
m
C59.5OA20
20
m
C54.5OA25
m
10
0
m
0
5
10
15
20
25
DMSO was added to the crystals until they were dissolved. Imme-
diately after pippeting each raw was separately analyzed by ELISA
method. Cell viability for each sample was calculated using Eq. (4):
Time (hr)
70
60
50
40
30
20
10
0
600
C64.5OC15
b
C59.5OC20
500
C54.5OC25
CS64.5OC15
CS59.5OA20
CS54.5OA25
CS64.5OA15
CS64.5OC15
CS54.5OC25
CS59.5OC20
400
CS59.5OC20
CS54.5OC25
300
C64.5OA15
C59.5OA20
200
C54.5OA25
CS64.5OA15
100
CS59.5OA20
CS54.5OA25
0
0
5
10
15
20
25
>300
150-200 100-150 50-100
20-50
<20
Time (hr
)
Particle size range (nm)
Fig. 5. Stacked percent of volume distribution of different particle size ranges of NLC
formulations.
Fig. 6. Release profiles of 5-FU from NLC formulations containing a) cholesterol, b)
cholesteryl stearate and different amounts of oils (n ¼ 3).

J. Varshosaz et al. / European Journal of Medicinal Chemistry 54 (2012) 429e438
435
AFM images were obtained by measurement of the interaction forces
between the tip and the sample surface. The experiments were done in
air at room temperature (25 ꢀC) operating in contact mode. Droplets of
size of different NLCs while the results of stacked percent of volume
distribution in each particle size range for different NLCs are shown
in Fig. 5. These data are according to volume and not intensity
measurements. As this figure shows the most popular volume
distribution for all formulations is seen between 50 and 100 nm.
CS64.5OA15 NLCs have the highest volume distribution at size range
of more than 300 nm. Considering this figure it may be concluded
that the higher range of volume particle size distribution in NLCs
prepared bycombination of CS and OA is higher than those prepared
by cholesterol probably due to the higher molecular weight of
cholestryl stearate compared to cholesterol (653.12 g/mol versus
386.7 g/mol for cholesterol) [19]. The other reason is that cholestryl
stearate has no surfactant characteristic compared to cholesterol
[20] which reduces the surface tension and consequently decreases
theparticle size distribution [21]. AsTable 2indicates almost all NLCs
have polydispersity index lower than 0.3 which is quite acceptable
and just NLCs of CS64.5OC15, CS59.5OC20, CS54.5OC25, CS59.5OA20 and
C54.5OC25 have PdI>0.3. It seems that the combination of octanol and
cholesteryl stearate causes the higher PdI than the mixture of oleic
acid and cholesterol. This may be due to higher viscosity of the
the final suspension (20 mL) were deposited onto a small mica disk.
After the drop was dried, the contact mode was used at room
temperature. The measurements were performed in different sample
locations. The mean size of NLCs was obtained by processing the
topographical AFM images with the AFM Nanos 1.1 software.
2.11. Statistical analysis
SPSS software version 11.5 was used for all statistical analysis.
Univariate analysis of data by a full factorial design was used for
comparison between particle size, zeta potential, loading and
release efficiency percent of 5-FU in different NLC formulations. The
cell culture data were expressed as mean ꢁ SEM and were compared
by analysis of variance (ANOVA) test followed by the post hoc test of
LSD. A significant level of p < 0.05 was considered in all cases.
3. Results and discussion
Considering the cytotoxic effects reported for 5-FU especially on
the colorectal carcinoma and other types of malignancies [17,18] by
directly promoting apoptosis, we tried to see if it is possible to
increase cytotoxic effect of this drug by using retinoic acid con-
taining NLCs targeted to LDL receptors of HT29 cell line.
3.1. Conjugation of retinoic acid and octadecylamine
Chemical reaction of retinoic acid and DCC, NHS and octade-
cylamine are shown in Fig. 2.
Conjugation of retinoic acid and octadecylamine was confirmed
by FTIR spectra as seen in Fig. 3. In this figure the peak of amine
group of NH2 in octadecylamine is seen in 3332 cmꢃ1 which is
shifted to 3311 cmꢃ1 in the product and this means the free NH2
group is not present in the product. The peak of C]O in retinoic
acid that was seen in 1687 cmꢃ1 is not present in the product but is
seen in 1628 cmꢃ1. This shift in the peak location means the
production of an amide bond.
Fig. 4 shows the 1H NMR spectra of octadecylamine, retinoic acid
and their conjugate. 1H NMR (CDCl3) results were as follows:
d
(ppm): 6.92 (1H, dd, Hb) 6.25 (2H, m, Hc, Hd), 6.12 (2H, m, Hc, Hf),
5.66 (1H, s, Hg), 5.45 (1H, br5, eNHe), 2.4 (3H, s, Hh), 2.05 (5H, m, Hi,
Hj),1.72 (3H, s, Hk),1.65 (2H, m, HL),1.55 (2H, CH2eCH2eNHeC¼O),
1.45 (2H, m, Hm),1.32 (2H, m, CH2eCH2eNHeC¼O),1.25e1.31 (30H,
m, CH2 Stearyl amine), 2.04 (2H, t, J ¼ 6, Hj), 2 (3H, s, Hi), 1.05 (6H, s,
Hn, Ho), 0.9 (3H, m, CH2eCH3).
3.2. Physical properties of NLCs
After the preliminary study which showed the most effective
variables on the physicochemical properties of NLCs, different
formulations were prepared from 5-FU by two types of lipids and
two types of oils each in three different percentages by emulsifi-
cation-solvent evaporation method containing retinoic acid-
octadecylamine conjugate. Table 2 shows the physical properties
of NLCs.
Different particle sizing techniques report primary results based
on number, volume, weight, surface area, or intensity. As a general
rule specifications should be based in the format of the primary
result for a given technique. Laser diffraction generates results based
on volume distributions and any specification should be volume
based. Likewise, an intensity basis should be used for DLS specifi-
cations, volume for acoustic spectroscopy, and number for image
analysis. Table 2 shows the intensity Z-average values of particles
Fig. 7. Atomic force microscopy of NLCs of 5-FU.

436
J. Varshosaz et al. / European Journal of Medicinal Chemistry 54 (2012) 429e438
120
100
80
60
40
20
0
Groups
Fig. 8. Percent of viable cells of HT29 cells after treatment with different concentrations of 5-FU loaded NLCs with or without retinoic acid conjugate in comparison with free 5-FU
and also in presence or absence of Apo-E by MTT assay (n ¼ 3).
combination of CS and OC than the mixtureof C and OAwhich causes
continuous increase of the surface tension and production of more
particles with different sizes, their aggregation, size growth and
making heterogeneous systems [22,23].
drug from the core of NLCs. The diffusion of drug out of the core of
NLCs is affected by the amount of the drug loading capacity. Table 2
shows that changing the lipid type from cholestryl stearate to
cholesterol decreases the RE20 which may be because of the
hydrogen bond between the fluorine atom of 5-FU and OH of the
cholesterol. This bond is more probable than hydrogen bond
between the fluorine of 5-FU and O atom of ketone group of cho-
lestryl stearate. Octanol reduces the RE20 comparing to oleic acid
probably because of the enlargement of particle size of NLCs and
consequently reduction of their surface area. Increasing the oil %
from 20 to 25% increased the RE20 due to reduction of the viscosity
of the NLCs and so making them leaky.
Zeta potential is a key factor in stability of nanoparticles. Zeta
potential of NLCs changed between ꢃ29.0 ꢁ 3.0 mV (C59.5OA20
)
and ꢃ7.7 ꢁ 1.6 mV (CS54.5OC25) (Table 2). All NLCs had negative
charge. Concerning the effect of lipid type on zeta potential, it was
noticed that the absolute value of zeta potential was increased
when the lipid type changed from cholestryl stearate to cholesterol
possibly due to reduction in particle size which increases the
negative charge density of charges on NLCs [24]. However, the
differences were not significant between CS59.5OA20, C59.5OA20 and
C54.5OA25 (P > 0.05).
The optimum formulation chosen for further studies was NLCs
of C54.5OA25 due to its logical particle size (105.8 nm), good stability
due to a relative high zeta potential (ꢃ25 mV), an acceptable drug
loading efficiency and release efficiency of 38%.
Changing the oil type from oleic acid to octanol had a reverse
effect on zeta potential may be because of negative zeta potential of
oleic acid which has a carboxylic acid group compared to octanol
[25]. However, the differences between CS64.5OC15, CS59.5OC20 and
C64.5OC15 were not significant (P > 0.05).
3.4. AFM
The greatest loading efficiency of 5-FU was observed in
CS59.5Oc20 (Table 2) which may be due to higher partitioning of
drug to this lipid/oil mixture. It was noticed that by changing
cholestryl stearate to cholesterol and octanol to oleic acid the
loading efficiency percent reduces due to the reduction of the
particle size of NLCs. Increasing the oil content of NLCs from 20 to
25% caused many imperfections offering space to accommodate the
drug (Table 2).
Fig. 7 shows the AFM contact mode atomic force microscopy
carried out on optimized NLCs. Imaging shows that the NLCs are
spherical and round shaped and are dispersed within the disper-
sion as discrete particles of about 100 nm in diameter and show
little or no aggregation. Photon correlation spectroscopy similarly
demonstrates the particle size of NLCs.
3.5. Cytotoxicity assessment-MTT assay
3.3. In vitro drug release
The anti-proliferative activity of cholesterol NLCs containing 5-
FU and retinoic acid conjugate (IC50 ¼ 4.5
m
M) was smaller than
5-FU release profiles from different NLCs are seen in Fig. 6. As
this figure shows after 20 h of release test about 35e45% of the drug
is released by a sustained behavior.
As it can be seen from 5-FU release profile from NLCs, there is an
initial burst diffusion of drug which is attributed to the drug
molecules adsorbed onto the surface of NLCs which can be released
instantaneously by contacting to the release medium. This follows
by a sustained release profile which is attributed to the release of
free 5-FU in comparable concentration of 8.2
m
M (IC50 ¼ 7.6 M).
m
Cholesterol NLCs containing 5-FU and the conjugate of retinoic acid
showed cell survival of about one fifth of free 5-FU and the NLCs
without retinoic acid conjugate showed about half cytotoxicity in
respect to free 5-FU that showed 46% cell viability.
Fig. 8 shows the results of cell viability by MTT assay. As this
figure shows free 5-FU in concentration of 8.2
mM has caused just
about 64% cytotoxicity compared to the same concentration of 5-FU

J. Varshosaz et al. / European Journal of Medicinal Chemistry 54 (2012) 429e438
437
loaded in NLCs containing the conjugate of retinoic acid-
octadecylamine that showed about 91% cytotoxicity and NLCs
without the conjugate but the same concentration of free drug that
showed about 84.5% cytotoxicity. The results show that the blank
NLCs containing retinoic acid conjugate but without drug has
shown greater cytotoxicity even compared to free 5-FU (p < 0.05).
This may indicate the cytotoxicity of retinoic acid conjugate on
HT29 cell line in cholesterol containing NLCs which is well targeted
to the LDL receptors of these cells. Lower concentrations of 5-FU in
test groups showed less cytotoxicity than the concentration of
improve the treatment outcome of human colorectal carcinoma.
Further studies are needed to optimize the ATRA conjugate
concentration in presence of 5-FU for obtaining maximum syner-
gistic effect in reducing needed dose of 5-FU. The results should be
checked in vivo to confirm the promising results on the cell culture.
Declaration of interest
The authors report no conflicts of interest.
Acknowledgments
8.2
8.2
m
m
M (p < 0.05) but the same trend as the groups treated with
M is seen in 4.1 and 1.025 M concentration.
Apolipoprotein E (Apo-E) is one of protein constituents of
m
The authors acknowledge the financial support from Isfahan
plasma lipoproteins that serve various functions, including regu-
lation of the metabolism of several different lipoproteins [15,26]. It
is a constituent of liver-synthesized very low density lipoproteins
(VLDL), which function primarily in the transport of triglycerides
from the liver to peripheral tissues, and of a subclass of high density
lipoproteins (HDL), which participate in cholesterol redistribution
among cells. In addition, Apo-E becomes a major protein constit-
uent of intestinally synthesized chylomicrons, which transports
dietary triglyceride and cholesterol. A major of physiological role
for Apo-E in lipoprotein metabolism is its ability to mediate high-
affinity binding of Apo-E-containing lipoproteins to the low
density lipoprotein (LDL) receptor [26e28]. Lipoprotein binding to
the receptors initiates the cellular uptake and degradation of the
lipoproteins, which leads to the use of the lipoprotein cholesterol in
the regulation of intracellular cholesterol metabolism. The normal
plasma concentration of this protein is 5 mg/dL. Omitting the
serum from the cell culture of HT29 cells caused cell viability to
University of Medical Sciences.
References
[1] A.E.B. Yassin, M.K. Anwer, H.A. Mowafy, I.M. El-Bagory, M.A. Bayomi,
I.A. Alsarra, Optimization of 5-flurouracil solid-lipid nanoparticles: a prelimi-
nary study to treat colon cancer, Int. J. Med. Sci. 7 (2010) 398e408.
[2] S. Li, A. Wang, W. Jiang, Z. Guan, Pharmacokinetic characteristics and anticancer
effects of 5-Fluorouracil loaded nanoparticles, BMC Cancer 8 (2008) 103e121.
[3] M. Radtk, E.B. Souto, R.H. Müller, Nanostructured lipid carriers:
a novel
generation of solid lipid drug carriers, Pharm. Technol. Eur. 17 (2005) 45e50.
[4] C.W. How, R. Abdullah, R. Abbasalipourkabir, Physicochemical properties of
nanostructured lipid carriers as colloidal carrier system stabilized with poly-
sorbate 20 and polysorbate 80, Afr. J. Biotechnol. 10 (2011) 1684e1689.
[5] Y. Zheng, P.M. Kramer, R.A. Lubet, V.E. Steele, G. Kelloff, M.A. Pereira, Effect of
retinoids on AOM-induced colon cancer in rats: modulation of cell prolifera-
tion, apoptosis and aberrant crypt foci, Carcinogenesis 20 (1999) 255e260.
[6] P.J. O’Dwyer, T.S. Ravikumar, D.P. McCabe, G. Steele Jr., Effect of 13-cis-retinoic
acid on tumor prevention, tumor growth, and metastasis in experimental
colon cancer, J. Surg. Res. 43 (1987) 550e557.
[7] B.K. Murray, B. Brown, P.M. Scherer, D.P. Tomer, K.R. Garvin, B.G. Hughes,
K.L. O’Neill, Induction of apoptosis in HT-29 human colon adenocarcinoma
cells by 13-cis-retinoic acid and vitamin E succinate, Nutr. Res. 26 (2006)
186e192.
increase from 9.3% to 75.3% in NLCs containing 8.2 mM 5-FU and the
conjugate of retinoic acid-octadecylamine. Therefore, when APO-E
that is necessary for transferring cholesterol to LDL receptors is
absent the cell viability increases near 8 fold that means targeting
of cholesterol NLCs to colon cancer cells has been done by LDL
receptors and when APO-E is removed from the serum in the
medium of cell culture the targeting to LDL receptors is blocked and
cell viability is increased.
Hegg et al. [29] also used a cholesterol-rich microemulsion that
binds to LDL receptors and is taken up by malignant cells over-
expressing these receptors. This system was further used in
Hodgkin’s and non-Hodgkin’s lymphoma by Pinheiro et al. [30]
who resulted that the cholesterol-rich microemulsion loaded
with etoposide is suitable for patient use.
[8] J. Fang, H. Wei, Z. Zheng, J. Liu, B. Hu, J. Huang, Expression of retinoic acid
receptor
b in human colorectal cancer, Afr. J. Biotechnol. 9 (2010) 6783e6786.
[9] E.Y. Park, A. Dillard, E.A. Williams, E.T. Wilder, M.R. Pepper, M.A. Lane, Retinol
inhibits the growth of all-trans-retinoic acidesensitive and all-trans-retinoic
acideresistant colon cancer cells through
a
retinoic acid recep-
toreindependent mechanism, Cancer Res. 65 (2005) 9923e9933.
[10] M.O. Bradley, N.L. Webb, F.H. Anthony, P. Devanesan, P.A. Witman,
S. Hemamalini, M.C. Chander, S.D. Baker, L. He, S.B. Horwitz, C.S. Swindell,
Tumor targeting by covalent conjugation of a natural fatty acid to paclitaxel,
Clin. Cancer Res. 7 (2001) 3229e3238.
[11] B.S. Chhikara, N. St Jean, D. Mandal, A. Kumar, K. Parang, Fatty acyl amide
derivatives of doxorubicin: synthesis and in vitro anticancer activities, Eur. J.
Med. Chem. 46 (2011) 2037e2042.
[12] B.S. Chhikara, D. Mandal, K. Parang, Synthesis and evaluation of fatty acyl ester
derivatives of cytarabine as anti-leukemia agents, Eur. J. Med. Chem. 45 (2010)
4601e4608.
[13] S. Vitols, G. Gahrton, M. Bjorkholm, C. Peterson, Hypocholesterolemia in
malignancy due to elevated LDL receptor activity in tumor cells: evidence
from studies in patients with leukaemia, Lancet 2 (1985) 1150e1154.
[14] A. Niendorf, H. Ngele, D. Gerding, U. Meyer-Pannwitt, A. Gebhardt, Increased
LDL receptor mRNA expression in colon cancer is correlated with a rise in
plasma cholesterol levels after curative surgery, Int. J. Cancer 61 (1995)
461e464.
[15] R.W. Mahley, Apolipoprotein E: cholesterol transport protein with expanding
role in cell biology, Science 240 (1988) 622e630.
[16] P. Costa, J. Manuel, S. Lobo, Modeling and comparison of dissolution profiles,
Eur. J. Pharm. Sci. 13 (2001) 123e133.
[17] M.E. Nita, H. Nagawa, O. Tominaga, N. Tsuno, S. Fujii, S. Sasaki, C.G. Fu,
T. Takenoue, T. Tsuruo, T. Muto, 5-fluorouracil induces apoptosis in human
colon cancer cell lines with modulation of Bcl-2 family proteins, Br. J. Cancer
78 (1998) 986e992.
[18] H.H.J. Backus, D.F. Dukers, C.J. van Groeningen, W. Vos, E. Bloemena,
D. Wouters, J.M.G.H. van Reil, K. Smid, G. Giaccone, H.M. Pinedo, G.J. Peters, 5-
fluorouracil induced fas upregulation associated with apoptosis in liver
metastases of colorectal cancer patients, Ann. Oncol. 12 (2001) 209e216.
[19] H.A.E. Benson, V. Sarveiya, S. Risk, M.S. Roberts, Influence of anatomical site
and topical formulation on skin penetration of sunscreens, Ther. Clin. Risk
Manag. 1 (2005) 209e218.
[20] R. Rai, C.R. Srinivas, Photoprotection, Indian J. Dermatol. Venereol. Leprol. 73
(2007) 73e79.
[21] M. Patel, S.K. Jain, A.K. Yadav, D. Gogna, G.P. Agrawal, Preparation and char-
acterization of oxybenzone-loaded gelatin microspheres for enhancement of
sun screening efficacy, Drug Deliv. 13 (2006) 323e330.
4. Conclusions
DCC and NHS can mediate ATRA conjugation to octadecylamine
by making an amide linkage. NLC formulations of 5-FU and the new
conjugate were prepared in a cholesterol base to fit with LDL
receptors of HT29 cells of colon carcinoma. The formulation of
C54.5OA25 consisting of 54.5% of cholesterol and 25% of oleic acid,
20 mg 5-FU, 0.5% of lecithin, 10% of PEG 40 stearate was selected as
the optimum NLC formulation due to its logical particle size
(105.8 nm), low polydispersity index (PdI ¼ 0.23), good stability
due to relatively high zeta potential (ꢃ25 mV), an acceptable drug
loading efficiency and release efficiency of 38%. The presence of
ATRA conjugate along with 5-FU loaded in cholesterol NLCs tar-
geted to LDL receptors can enhance cellular cytotoxicity of 5-FU. So
that when 5-FU was loaded in the NLCs containing retinoic acid
conjugate, the cytotoxicity was nearly 2 fold of NLCs just loaded
with 5-FU and more than 5 fold of free 5-FU in human colon cancer
cell line of HT29 significantly. We postulate that 5-FU and ATRA
conjugated octadecylamine loaded in cholesterol NLCs that are
targeted to LDL receptors may serve as a potential strategy to

438
J. Varshosaz et al. / European Journal of Medicinal Chemistry 54 (2012) 429e438
[22] S.A. Wissing, R.H. Muller, The development of an improved carrier system for
sunscreen formulations based on crystalline lipid nanoparticles, Int. J. Pharm.
242 (2002) 373e375.
[23] J. Pardeike, A. Hommoss, R.H. Muller, Lipid nanoparticles (SLN, NLC) in
cosmetic and pharmaceutical dermal products, Int. J. Pharm. 366 (2009)
170e184.
[24] V. Jenning, A.F. Thunemann, S.A. Gohla, Characterisation of a novel solid lipid
nanoparticle carrier system based on binary mixtures of liquid and solid
lipids, Int. J. Pharm. 199 (2000) 167e177.
[25] R.H. Muller, M. Radtke, S.A. Wissing, Nanostructured lipid matrices for
improved microencapsulation of drugs, Int. J. Pharm. 242 (2002) 121e128.
[26] R.W. Mahley, T.L. Inncrarity, S.C. Rail, K.H. Weisgraber, Plasma lipoproteins:
apolipoprotein structure and function, J. Lipid Res. 25 (1984) 1277e1294.
[27] M.S. Brown, J.L. Goldstein, A receptor-mediated pathway for cholesterol
homeostasis, Science 232 (1986) 34e47.
[28] M.S. Brown, J.L. Goldstein, Lipoprotein receptors in the liver. Control signals
for plasma cholesterol traffic, J. Clin. Invet. 72 (1983) 743e747.
[29] R. Hegg, J. Carvalho, S.R. Graziani, C.J. Valduga, D.C. Fernandes, C.H.M. Azevedo,
A.C. Loprete, D.G. Rodrigues, R.C. Maranhao, Use of a cholesterol-rich micro-
emulsion that binds to low-density lipoprotein (LDL) receptors as a vehicle for
etoposide, Proc. Am. Soc. Clin. Oncol. 22 (2003) 584.
[30] K.V. Pinheiro, V.T. Hungria, E.S. Ficker, C.J. Valduga, C.H. Mesquita,
R.C. Maranhão, Plasma kinetics of a cholesterol-rich microemulsion (LDE) in
patients with Hodgkin’s and non-Hodgkin’s lymphoma and a preliminary
study on the toxicity of etoposide associated with LDE, Cancer Chemother.
Pharmacol. 57 (2006) 624e630.