International Journal of Pharmaceutics 450 (2013) 323–330

Contents lists available at SciVerse ScienceDirect

International Journal of Pharmaceutics

journal homepage: www.elsevier.com/locate/ijpharm

Pharmaceutical nanotechnology

Preparation and characterization of “dextran-magnetic layered

double hydroxide-fluorouracil” targeted liposomes
Jie Huang a,b , Guojing Gou a,b,∗ , Bing Xue a,b , Qianshun Yan a,b , Yue Sun a,b , Li-E. Dong a,b
a
Department of Chemistry, Ningxia Medical University, Yinchuan 750004, China
b
Research Center of Medical Science and Technology, Ningxia Medical University, Yinchuan 750004, China

a r t i c l e i n f o a b s t r a c t

Article history: This work was aimed at assessing the preparation and characteristics of “dextran-magnetic layered dou-
Received 26 November 2012 ble hydroxide-fluorouracil” liposomes (DMFL). DMFL was prepared by the optimized reverse evaporation
Received in revised form 29 March 2013 method, which concerned the entrapment efficiency and slow-released effect. The factors affecting the
Accepted 9 April 2013
entrapment efficiency of DMFL were studied using orthogonal design, and the optimum conditions are:
Available online 13 April 2013
weight ratio of lecithin to cholesterol (2:1), weight ratio of lecithin to DMF (7:1), emulsification time
(30 min) and temperature (50 ◦ C). The characteristics of optimized DMFL on encapsulation efficiency,
Keywords:
mean diameter and pH value were 85.47 ± 0.83, 160.4 ± 0.55 nm and 6.58 ± 0.05, respectively. In vitro
“Dextran-magnetic layered double
hydroxide-fluorouracil” delivery system
drug release profile of DMFL followed the Higuchi release model equation Q = 9.2338t1/2 + 22.821. The
Liposome magnetic targeting results showed that DMFL had sensitive magnetic targeted responsibility. The results
Magnetic targeted of XRD, FT-IR and TEM indicated that the structure and property of DMF were not destroyed during the
Sustained release process of forming DMFL, and the phospholipid bilayer and the hexagonal skeleton DMF were obvious
and complete after being lyophilized powder. This lyophilized method could be used to store the DMFL
easily. These results suggested that DMFL had the potential for developing as a practical preparation for
administration.
© 2013 Elsevier B.V. All rights reserved.

1. Introduction intercalated tetracycline to alendronate LDH film. It was shown that

Layered double hydroxides (LDHs), an anion exchange-
able intercalated assembly functional materials, also were
called hydrotalcite. The general formula of LDHs is [M1−x II
Mx III (OH)2 ]x+ [An− x/n ]x− ·mH2 O, where MII represents the divalent
and MIII represents the trivalent metal cations and An− is an
hydrated exchangeable anion (Parello et al., 2010). In recent years,
LDHs have been the hotspot (Nalawade et al., 2009; Karan and Ay,
2012) in material, chemistry, biology and medicine fields due to
their advantages, such as ion exchangeability, thermal property,
memory effect, absorbability, biocompatibility and degradability.
LDHs were proposed as carriers for various drugs to prolong the
release time and improve the stability of drugs. Comparing the drug
release rate of fenbufen intercalated in Mg/Al-LDH and Li/Al-LDH
to that of fenbufen itself, fenbufen intercalated in LDH released
more slowly than itself, proving that LDH-drugs were effective
sustained release drug carrier system (Li et al., 2004). Michelle
Chakraborti had conducted experiments showing that LDH dis-
played the slow-release and carriage actions for tetracycline by
∗ Corresponding author at: Department of Chemistry, Ningxia Medical University,
Yinchuan 750004, China. Tel.: +86 13895652708; fax: +86 951 6980110.
E-mail addresses: ggj64@yahoo.com.cn, ggj64@aliyun.com (G. Gou).
the activity on alkaline phosphatase increased and the bone nodule
formed when contacting osteoblasts after five weeks (Chakraborti
et al., 2011). The successful synthesis of magnetic layered double
hydroxide (MLDH) (Gou et al., 2009a) provided a new magnetic
targeting and controlled-release carrier for modern drug deliv-
ery system, by combining the paramagnetism of Fe(OH)x with the
drugs control release of LDH. “Dextran-magnetic layered double
hydroxide-fluorouracil” (DMF) magnetic targeted delivery system
(Gou et al., 2009b) was synthetized in our group by intercalating
the anticancer drug fluorouracil into the galleries of MLDH with
organic composition method. DMF could decrease the toxicity sig-
nificantly (Gou et al., 2012a) and prolong the release of fluorouracil
in vivo (Gou et al., 2011), separately.
Although the drug release properties of LDH-drugs had been
studied widely in the field of drug delivery, but few studies focus
on the targeting and dosage forms of LDH-drugs system. The
structures of liposomes, a new drug delivery system, are simi-
lar to plasmalemma, in which both hydrophilous and lipophilic
drugs are encapsulated. It had been reported that liposomes
can significantly decrease drug toxicity, magnify biocompatibility
(Liu et al., 2011), increase dissolubility of drug and prolong the
release time (Sun et al., 2008). Our study was aimed at preparing
“dextran-magnetic layered double hydroxide-fluorouracil” tar-
geted liposomes (DMFL), exploring the prepared methods and

0378-5173/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijpharm.2013.04.010
324 J. Huang et al. / International Journal of Pharmaceutics 450 (2013) 323–330
appropriate conditions, so as to provide a possible injection deliv-
ery approach for DMF magnetic targeted deliver system. Compared
with the traditional fluorouracil dosage forms, such as tablets,
injections and implants, the successful preparation of DMFL would
provide a new way for targeted and sustained release drug deliver
system, in particular a new type formulation for fluorouracil, which
has both excellent controlled release property and magnetic target-
ing effect.
2. Materials and methods
2.1. Materials
Injectable soya lecithin was provided by Shanghai Taiwei Phar-
maceutical Co. Ltd. China. Cholesterol was obtained from Anhui
Tianqi Chemical Company Co. Ltd. China. Fluorouracil was pur-
chased from Japan TCI Company Co. Ltd. Japan. Methanol was a
gift from Guangdong Guanghua Chemical Factory Co. Ltd. China.
All the other chemicals and reagents used were of analytical purity
grade or higher, obtained commercially.
2.2. Synthesis of DMF
DMF was synthesized by coprecipitation method (Gou et al.,
2004, 2012b). The reaction system consisted of pH meter, stirring
magneton, precipitant drip system and 1000 mL nitrogen protec-
ting reactor. In detail, 39.76 g ferrous chloride and 27.03 g ferric
chloride were dissolved by 200 mL double distilled water and
vigorously stirred to get a homogeneous system. Sixty-nine mL
intercalator FU solution (concentration was 0.145 mol L−1 ) was
added into above system and completely stirred. Three hundred
mL sodium hydroxide solution (concentration was 2 mol L−1 ) was
added dropwise to the reactor at room temperature in the nitro-
gen protection conditions. The slurry was diluted by 100 mL double
distilled water until the final pH value of 9.240. Then the alkalify
system was closed, other reaction conditions were retained. The
slurry was crystallized for 30 min until the pH value decreased
to 8.855. Afterwards, 84 g dextran was dissolved by 200 mL dou-
ble distilled water and reacted with MLDH-FU slurry for 40 min
while the pH value reached to 8.387. After that the third order
compound was placed into 1000 mL dehydrated alcohol quickly
which had been frozen 4 h at −80 ◦ C and stirred fleetly to carry
out the solvent transformation. The system obtained above was
aged for 14 h and centrifuged for 15 min with the rotation speed
of 5000 r min–1 . Then the hygric solid phase from centrifugation
was dried under vacuum. Finally DMF was obtained as the dried
sample.
2.3. Selection of prepared methods for DMFL
2.3.1. Film dispersion method
Firstly, DMFL was prepared according to the film dispersion
method as reported previously (Kuznetsov et al., 2001). Lecithin
and cholesterol in weight ratios of 3:1 were dissolved in chloro-
form. The powder of DMF and phospholipids with weight ratios of
1:5 was dissolved by double distilled water with water bath ultra-
sound for 20 min. The organic solvent was vacuum evaporated in
water bath at 45 ◦ C for 1 h until a dry thin lipid film was formed.
After that a sprinkle of chloroform was used to dissolve the film.
Then the film was hydrated for 2 h by the addition of DMF solu-
tion (volume ratio of chloroform to double distilled water, 1:2) and
traces of chloroform were removed. In order to reduce the dimen-
sions of DMFL, DMFL suspension was dispersed by using sonifier
cell disrupter for 5 min. Finally, the DMFL was completed.
2.3.2. Reverse evaporation method
Secondly, DMFL was prepared by reverse evaporation method
as reported previously (Wu et al., 2009). Lecithin and cholesterol
in weight ratios of 3:1 were dissolved in chloroform. The organic
solvent was vacuum evaporated for 1 h until a dry thin lipid film
was formed. The powder of DMF with lecithin at weight ratios of
1:5 was dissolved by double distilled water with ultrasonication. A
sprinkle of chloroform was used to dissolve the film. Then the film
was emulsionized by the dropwise of DMF solution with a mechan-
ical stirrer for 30 min with a water bath ultrasound (90w) (volume
ratio of oil phase to water phase, 1:2), following hydrated with same
time and ways as above. Same ultrasound condition was used to
disperse particle diameter and obtain DMFL.
2.3.3. Alcohol injection method
Thirdly, DMFL was prepared with alcohol injection method as
reported in literature (Li et al., 2009). The weight ratios of lecithin
to cholesterol, lecithin to DMF and the volume ratio of oil to water
ratio were same as above. Lecithin and cholesterol were dissolved in
alcohol. The powder of DMF was dissolved by double distilled water
through ultrasonication for 20 min. Then the lipid solution was
evaporated by the dropwise of DMF solution for 2 h and magnetism
stirred in stirring hot plate until the organic solvent was absolutely
volatilized. Finally, we dealt the liposomes with ultrasound as what
we mentioned to reduce particle diameter, and DMFL was prepared.
2.4. Orthogonal test
Encapsulation efficiency is affected by several factors, includ-
ing the ratio of lecithin to cholesterol, lecithin to drug, oil phase
to water phase, emulsification time and temperature, etc. (Liang,
2010). Optimal formulation can be chosen by orthogonal test in
the case of more experiments. Finally, four factors included the
weight ratios of lecithin to cholesterol and lecithin to drug, emulsi-
fication time and temperature were determined as the orthogonal
factors after single factor tests. Therefore an orthogonal experiment
with four factors and three levers L9 (34 ) was designed to shift out
the optimized plan with encapsulation efficiency as the evaluation
index.
2.5. Determination of the encapsulation efficiency
Dialysis method was applied to determine the encapsulation
efficiency. Four mL DMFL was pipetted into the dialysis bag. Then
both ends of dialysis bag were tied up by thread and bound to
screw propeller of dissolution tester. Two hundred mL double dis-
tilled water was used as dissolution medium. Dialysis bag was
immerged at 37 ± 0.5 ◦ C with rotation speed of 85 r min−1 for 12 h
until unloaded drug was depleted. Then the determined method of
the unencapsulated FU in DMFL was described as following. Four
mL dialyzed dissolution solution was pipetted to 10 mL volumet-
ric flask and brought to volume by 0.1 mol L−1 hydrochloric acid,
following handling with ultrasound water bath to dissolve DMF
completely. The total FU in DMFL was determined with the follow-
ing steps. Firstly, 0.5 mL DMFL were pipetted to 25 mL volumetric
flask and brought to volume by methanol in order to demulsifica-
tion. Then the handled way was same as the determined method of
unencapsulated FU in DMFL. The absorbance of FU in the DMF was
measured by ultraviolet spectrophotometry at 265 nm. The work-
ing curve of FU was A = 0.0094 + 0.0414C (␮g mL−1 ), r = 0.9994. The
drug entrapment efficiency (EE) was calculated from Eq. (1).
Wfree FU × DLFU − Wfree FU × DLFU
EE% = × 100
Wtotal FU × DLFU
1 − W 
= free FU
× 100 (1)
Wtotal FU
 J. Huang et al. / International Journal of Pharmaceutics 450 (2013) 323–330 325

where Wfree FU is the content of unencapsulated FU in DMFL, DLFU

is drug loaded rate of FU in DMF, and Wtotal FU is the content of total
FU in DMFL.

2.6. The drug release experiment in vitro

DMFL and FU-loaded liposomes (FUL) were prepared with the

same prescription came from orthogonal test. The drug release
experiment in vitro was determined by dialysis method. Four mL
DMFL and FUL were placed to dialysis bags, respectively. Dial-
ysis bags were immerged at 37 ± 0.5 ◦ C with rotation speed at
85 r min−1 . Two hundred mL phosphate buffer (PBS) with pH of
7.4 was applied as dissolution medium. Aliquots of the dissolution
medium (5 mL) were withdrawn at each time interval and the same
volume of fresh PBS was added to maintain the constant volume.
The absorbance of FU in DMFL was determined as described in Sec-
tion 2.5. The accumulative release (AR) rate of FU in DMFL was
calculated from Eq. (2). Fig. 1. XRD pattern of DMF (A), freeze-dried DMFL without cryoprotectant (B) and
freeze-dried DMFL with mannitol (C).
C
AR% = release FU × 100 (2)
Ctotal FU
3. Results and discussion
where Crelease FU is the absorbance of released FU at each time inter-
val and Ctotal FU is the absorbance of total FU in DMFL. 3.1. Characterization of DMF and DMFL

Fig. 1 shows the XRD patterns of DMF(A), DMFL(B) and

2.7. Characterization of DMFL
2.7.1. X-ray diffractomer (XRD) of DMFL
According to solid phase analysis requirement, the DMFL were
lyophilized without cryoprotectant and lyophilized with mannitol
(10% aqueous solution), respectively. DMF and freeze-dried lipo-
somes were analyzed by XRD, respectively. Powder XRD patterns
of the samples were recorded using a Rigaku D/max-rB XRD-6000
diffractometer under the following conditions: 40 kV, 30 mA, Cu
K␣ ( = 0.15406 nm) radiation. The samples as unoriented powders
were step-scanned in steps of 2◦ in the range from 5 to 80◦ using a
count time of 4 s per step.
2.7.2. Fourier transform infrared (FT-IR) of DMFL
FT-IR was applied to identify the structure of DMF, freeze-dried
liposomes and synthetic materials including lecithin, cholesterol
and mannitol, respectively. FT-IR spectra were recorded on a Bruker
TENEOR27 instrument. The samples were finely ground for 1 min,
combined with oven dried spectroscopic grade KBr and pressed
into a disc. The spectrum of each sample was recorded in triplicate
by accumulating 20 scans at 2 cm−1 resolution between 400 and
4000 cm−1 .
2.7.3. Morphology of DMF and DMFL
The morphology of DMF and DMFL was examined by HITACHI
H-7560B transmission electron microscopy. Little DMF and DMFL
suspension were dropped on to copper grid and air-dried. Both
suspensions were deposited on to copper grid and observed by
transmission electron microscopy (TEM).
2.7.4. Characterization of other properties of DMFL
Appropriate amounts of dilute solution of DMFL and FUL
in clean sample cell were determined by the United States
PSS NICOMPTM 380 submicron particle size analyzer. pH value
was measured by PHSJ-4A pH meter. Equal amounts of DMFL
and FUL in glass bottles were set beside magnets composed
of ˚17 mm × 3 mm neodymium iron boron permanent magnets
(field strength 2000 Gs/unit) and their morphological changes were
observed.
lyophilized DMFL(C). The characteristic diffraction peaks of the tar-
geted delivery system DMF, the nuclear skeleton of DMFL, as shown
in Fig. 1A, appearing in (11.039◦ , 0.8008 nm), (22.200◦ , 0.4001 nm),
(33.080◦ , 0.2706 nm) and (57.580◦ , 0.1599 nm), correspond to the
crystal diffraction of R-hexagonal lattice, which indicates that the
DMF lattice can be indexed as Fe3.6 Fe0.9 (O,OH,Cl)9 magnetic layered
compound hydroxide (MLDH) (Gou et al., 2012a). The diffraction
peaks at (35.391◦ , 0.2634 nm) are homologous with the peak of
magnetic iron oxide (3 1 1) (35.442◦ , 0.2632 nm) in the standard
card PDF 19-0629, clarifying that a part of DMF is oxidized to tiny
magnetic iron oxide. But the magnetic iron oxide is not the main lat-
tice due to the content of magnetic iron oxide is only 0.1, attributed
to the protection of dextrane for MLDH and anti-oxidant of the
MLDH-FU (Gou et al., 2011). The weak diffraction peak of DMF at
11.012◦ , the strong peaks of the magnetic iron oxide and the organic
amorphous peaks at (18.019◦ , 24.231◦ ) in Fig. 1B suggest that the
lyophilized DMFL without mannitol are not helpful to protect the
lattice of DMF due to the damage of LDH and the raised oxidation
level. In contrast, the diffraction peaks of the lyophilized liposomes
with mannitol at 11.038◦ , 22.670◦ , 32.998◦ demonstrate the lattice
of DMF is retained completely, as depicted in Fig. 1C. Moreover, the
oxidation intensity of (3 1 1) is similar to DMF, indicating that the
addition of mannitol is beneficial to protect the DMF lattice.
The results of XRD indicate that DMF is the compound lattice
consisting of the staple magnetic layered compound hydroxide and
the tiny magnetic iron oxide. The magnetism and controlled action
of the MLDH layers build a structure foundation, the parties hereto
do hereby the slow release and magnetic targeted function are inte-
grated by the DMF delivery system. In addition, the lyophilized
DMFL with mannitol have better effect on protection DMF, which
is significant to the storage and formed pulver for DMFL.
Fig. 2 shows the FT-IR spectra patterns of DMF, cholesterol,
lecithin, mannitol and lyophilized DMFL, respectively. The markers
absorption of 1, 3, 4, 5, 8 in Fig. 2A reflect the IR spectra patterns
of carrier MLDH of the delivery system DMF (Gou et al., 2012b),
assigning to the O–H stretching vibration of water hydroxyl and
ıO–H bending vibration of hydroxy in MLDH layers, ıMO–H rocking
vibration, ıO–H deformation vibration in hydroxy hydrogen of layer
surface and Fe–O stretching vibration of lattice oxygen within
MLDH layers, respectively. Peaks marked with 2, 6, 7 are IR spectra
326 J. Huang et al. / International Journal of Pharmaceutics 450 (2013) 323–330
Fig. 2. FT-IR spectra of DMF (A), cholesterol (B), lecithin (C), mannitol (D) and freeze-
dried DMFL with cryoprotectant (E).
patterns of DET in the surface layer of DMF, attributing to the
␯C–H stretching vibration of carbon skeleton and CH2 methylene,
carbonyl vibration of FU in interlayer and C–H rocking vibration
in multiple carbocyclic of DET, respectively. IR absorption of
intercalated FU in DMF is shown in ␣, ␤, and ␥ marked locations.
In detail, high-frequency ␯C O vibration red shifted 28 cm−1 and
affiliated to absorption peak 3, and two ␯C–N stretching vibrations
(1427.71, 1346.88 cm−1 ) are observed in the ␣ (1457 cm−1 ) peak
and the merged peak (1340 cm−1 ), respectively. Moreover, the
␦N–H deformation vibration blue shifted 25 cm−1 is located in
the marked place ␤ (1271 cm−1 ), and ␦C–H vibration (638 cm−1 )
of unsaturated hydrogen atoms in plane ring of FU is reflected
in valley peak ␥. Fig. A manifests that DMF is a supramolecular
compound consisting of DET, MLDH and intercalated FU.
Fig. 2B–E is the IR spectra patterns of freeze-dried of DMFL,
cholesterol, lecithin and mannitol, respectively. Absorption band
(3750–3152 cm−1 ) of E-1 include the ␯O–H stretching vibration
of hydroxy in DMF (A, 3417 cm−1 ), cholesterol (B, 3634 cm−1 ),
lecithin (C, 628 cm−1 ) and mannitol (D, 3688 cm−1 ), the stretch-
ing vibration of amido in FU (A-1, 3131.92 cm−1 ) and lecithin
(C, 3304 cm−1 ). The absorption of E-2 (2941 cm−1 ) reflects the
C–H vibration of DET and intercalated FU (2924 cm−1 ), lecithin
and mannitol (2924 cm−1 ). In detail, C-2 (2925 cm−1 , 2855 cm−1 )
are attributed to the antisymmetric and symmetric stretching
vibration of CH2 in phospholipid hydrophobic chain, where the
antisymmetric vibration frequency changes of CH2 are the index
for phospholipid molecular conformation and reflect the physi-
cal state change of bilayer. Since the intermolecular forces are
weak with increases in CH2 symmetric and antisymmetric vibra-
tion frequency (Nie et al., 1989), the absorption (2941 cm−1 ) and
blue shifting (16 cm−1 ) of CH2 asymmetric stretching vibration of
lecithin (C-2, 2925 cm−1 ) illustrate that the addition of cholesterol
can decrease the intermolecular forces of phospholipid aliphatic
chain, stabilize and protect the bilayer structure, therefore caus-
ing the increase of ␯ (CH2 ) vibration frequency of DMFL. The
E-3 absorption band (1662–1626 cm−1 ) is a combined absorp-
tion constituted with the similar vibration frequency, such as
the ␯C O vibration of carbonyl (A-3, 1635 cm−1 ) of intercalated
FU in DMF, the C CH vibration of unsaturated hydrocarbon (B-3
1658–1624 cm−1 ) in cholesterol and the ıH–O–H vibration of water
in mannitol. Absorption band E-4 (1421 cm−1 ) combines the high-
frequency ␯C–N vibration of amide (A-␣, 1457 cm−1 ) of FU in DMF
with the ıC–O rocking vibration of hydroxy (D-4, 1403 cm−1 ) in
Table 1
Choice of preparation method.
Preparation method Film dispersion Reverse Ethanol
evaporation injection
Encapsulation efficiency (%) 25.25 58.70 40.65
mannitol together. Absorption band E-5, 6 (1082–1023 cm−1 ) are
synthetical absorption of all components of DMFL, near to the
␯C–OH vibration of carboxyhemoglobin of hydroxy in DET (A-5 to
6, 1153–1016 cm−1 ), cholesterol (B-5, 1100 cm−1 ), mannitol (D-5 -
6, 1079–1030 cm−1 ) and the ␯P–O–C vibration of phosphoester bond
(C-5, 1089 cm−1 ) in lecithin. Absorption band E-7 (881–708 cm−1 )
contains the vibration of carbon skeleton of DET in DMF and the
rocking vibration of hydrocarbon bond in mannitol.
In a conclusion, DMF is proved to be the main lattice in
lyophilized DMFL, which the characteristic absorption peaks (3,
␣, 5–8, ␥) of DMF and the core skeleton of DMFL are obviously
reflected in Fig. E. Lecithin as the key material to prepare lipo-
somes, is proved to exist in DMFL by these characteristic absorption
peaks, including the ␯C O stretching vibration of ester carbonyl
(C, 1734 cm−1 ) which appears at the shoulder peak of DMFL
(1730 cm−1 ), the ı (CH2 ) scissoring vibration of methylene (C,
1464 cm−1 ) in hydrophobic chain appeared at the peak of DMFL
(1458 cm−1 ) and the ␯C–O–C stretching vibration of carbon oxy-
gen bond (1238 cm−1 ) of DMFL located to range from 1240 to
1213 cm−1 . Moreover, it has been proved by the blue shift of ␦
(CH2 )antisymmetric stretching vibration that cholesterol is able to
stabilize and protect the lecithin bilayer. In summary, the complete
structures of lecithin bilayer and its core skeleton DMF have built
the structural foundation of magnetic target, control release and
cell biology advantage for DMFL.
3.2. Choice of prepared methods and formulation optimization of
DMFL
In general, the encapsulation efficiency and physicochemical
properties of liposomes prepared with different methods with a
same prescription are different. Film dispersion method, reverse
evaporation method and alcohol injection method are chosen to
select the suitable prepared method in this study, respectively. The
results in Table 1 show that the encapsulation efficiency of reverse
evaporation method is higher than that of other methods. Film
dispersion method is suit to envelope liposoluble drug, just like
artemisinin (Isacchi et al., 2011) and paclitaxel (Kim et al., 2011),
attributing to the formation of small inner aqueous phase. Con-
versely, reverse evaporation method and alcohol injection method
are suitable to envelope hydrophilia drug (Zhu et al., 2006; Zhao
et al., 2007). But the liposomes encapsulated by alcohol injection
method are leaked easily than reverse evaporation method and
lead to the decrease in entrapment efficiency. For these reasons,
reverse evaporation method is selected to prepare DMFL. The suc-
cessful preparation of DMFL is important for developing prolong
controlled release formulations to realize the special therapeutic
effect of the DMF drug delivery system, as the DMFL can signifi-
cantly improve the sustained release and targeting effect compared
with the conventional magnetic targeted liposomes.
The drug release curves of DMFL prepared with different meth-
ods in PBS medium are displayed in Fig. 3. The initial concentration
of FU (0.3768 ␮g mL−1 ) in DMFL prepared by reverse phase evap-
oration method is minimum, while the initial concentration of FU
(1.9952 ␮g mL−1 ) prepared by ethanol injection method is highest.
The drug release of DMFL prepared by ethanol injection method
appears to be fastest and tends to equilibrium at 8 h, but the drug
release rate of DMFL prepared by reverse evaporation method is
slowest. The drug release rate of DMFL prepared by film dispersion
 J. Huang et al. / International Journal of Pharmaceutics 450 (2013) 323–330 327

the four factors is A2 B3 C3 D2 . Therefore the optimum preparation

conditions is: weight ratio of lecithin to cholesterol (3:1), emulsifi-
cation (30 min), temperature (50 ◦ C) and weight ratio of lecithin to
DMF (7:1). The average encapsulation efficiency of DMF in DMFL
prepared with the optimal formulation is 85.47 ± 0.83 (n = 4).

3.3. Properties of DMFL

3.3.1. Diameter and morphology of DMFL

The mean particle size of DMFL and FUL prepared with the
same optimum prescription from orthogonal test is determined
using submicron particle size analyzer. The average particle size
of DMFL is 160.4 ± 0.55 nm, which is close to the particle size esti-
mated from TEM in Fig. 4, and the average particle size of FUL is
120 ± 0.45 nm. However, parts of particle size of both DMFL and
FUL reach to 500 nm, this is most likely due to the aggregation
and sedimentation of liposomes. DMF nanoparticles are three-level
Fig. 3. Release plots of DMFL by different preparation methods in PBS. supramolecular complexes in which fluorouracil is encapsulated
by magnetic layered hydroxides and dextran, so the particle size of
DMF is larger than original drug fluorouracil, following the particle
Table 2
Three levels of four factor in orthogonal experiments. size of DMFL is much larger than FUL. Therefore, the diameter deter-
mination results demonstrate that the particle size of liposomes
Level Factor
is probably related to the particle size of encapsulated drugs. In
A: L:CH (w/w) B: DMF:L (w/w) C: Time (min) D: Temperature (◦ C) addition, the mean pH value of DMFL is 6.58 ± 0.05 (n = 4), indicat-
1 1:1 1:3 10 45 ing the DMFL are suited for injection in vessel due to their neutral
2 3:1 1:5 20 50 character.
3 5:1 1:7 30 55 Fig. 4 shows the morphology results of DMF and DMFL in
TEM. The morphology of DMF prepared by “intercalation assembly
MLDH-FU with co-precipitation – in situ composite of DET – sol-
method is in the middle compared with that of the former above. vent conversion” method are shown in Fig. 4A-a, which appears to
Although the DMFL prepared with ethanol injection method and be spherical symmetry, and the diameter of the larger DMF is nearly
dispersion method are earlier to reach equilibrium than reverse 68.9 nm. Most hexagonal rough sketches of MLDH-FU are indistinct
evaporation method, the slow-release process of DMFL prepared in to observe due to the composition and encapsulation of polymer
three methods can be sustained for more than 36 h. In brief, utiliza- DET, but the hexagonal rough sketches of MLDH-FU encapsulated
tion of reverse phase evaporation method may be able to improve with DET are clear and the edge length is approximately 29.8 nm
not only the encapsulation efficiency of DMFL, but also the property in Fig. 4A-b. The morphology of DMFL is shown in Fig. 4B, and the
of sustained release of DMFL significantly. core skeleton of DMFL, i.e. DMF is clearly observed. The hexagon of
The weight ratios of lecithin to cholesterol and lecithin to skeleton of DMF is nearly exposed and the edge length is approxi-
drug, emulsification time and temperature are chosen as major mately 30.8 nm in Fig. 4B-b, which is similar to the diameter of DMF
factors determining the optimal DMFL, based on the single factor in Fig. 4A-b. In addition, the morphology of phospholipid bilayer in
experiments. The L9 (34 ) orthogonal test is applied to explore DMFL is observed explicitly and the diameter of DMFL is nearly
optimal prescription with encapsulation efficiency as the index. 110.5 nm while the thickness of phospholipid bilayer is approxi-
The detailed orthogonal experiment and related results are listed mately 21 nm in Fig. 4B-a. Furthermore, the diameter of hexagon
in Table 2 and Table 3, respectively. By variance R analysis, the skeleton of DMF in Fig. 4B-a is 68.5 nm, corresponding to the diam-
order of influence for encapsulation efficiency is weight ratio eter of DMF in Fig. 4A-a. However, other phospholipid bilayers
of lecithin to cholesterol, emulsification time, temperature, and appear to be indistinct due to their smaller thickness. Fig. 4B proves
weight ratio of lecithin to drug, and the best combination level of that the DMFL have been successfully prepared and their structure
and geometry feature appear very prominent. What is more, the
Table 3 dispersity of DMF is significantly improved through transforming
Results of orthogonal experiment. into DMFL following decrease in the aggregation and distributed
density of DMF. Otherwise, the diameter of DMF in TEM is smaller
No. A B C D E%
than the average particle size determined by submicron particle
1 1 1 1 1 68.64 size analyzer, illustrating that the diameter of TEM is at the area
2 1 2 2 2 61.84
3 1 3 3 3 75.78
of skew normal distribution, mainly distributing the bigger par-
4 2 1 2 3 66.39 ticles caused by the higher concentration and the aggregation of
5 2 2 3 1 68.91 metastable DMFL particles.
6 2 3 1 2 75.68 The structure model of DMFL is shown in Fig. 5 according to the
7 3 1 3 2 24.35
results of XRD, IR and TEM. The results of XRD indicate that the
8 3 2 1 3 19.23
9 3 3 2 1 75.68 structure and crystal feature of DMF are not affected by the informa-
tion of DMFL. With nanoparticles DMF as the core skeleton, adipoid
k1 a 68.75 53.13 54.52 49.57
envelope layers are formed by the affinity among the hydrophilic
k2 70.32 50.00 46.45 53.96
k3 18.23 54.19 56.35 53.80 group of lecithin and cholesterol and the hydroxy of DET on the
Rb 52.09 4.190 9.900 4.390 surface of DMF. The lipid bilayer built with self-assembly of lecithin
a
ki : the average value of encapsulation efficiency of DMFL in different levels of and cholesterol is able to spontaneously form spherical micelle par-
every factor. ticles by the surface action between the medium and the lipophilic
b
R: the range of ki in different levels of every factor. group on the surface of DMFL, leading to the surface tension of the
328 J. Huang et al. / International Journal of Pharmaceutics 450 (2013) 323–330

Fig. 4. TEM morphology of DET-MLDH-FU tri-polymer (A) and DMFL (B).

Fig. 5. DMFL structure model figure.

DMFL system decrease to the minimum. This structure model is
indirectly supported by the IR characterization results, which show
that the hydrophilic group and lipophilic group in the phospholipid
bilayer of liposomes and the core skeleton DMF coexist in the
DMFL. In addition, the results of TEM offer the visualized evidence
for the model of DMFL, in which the shape of DMFL is spherical,
and the phospholipid bilayer and the hexagon skeleton of DMF
are obviously seen in Fig. 4B-a. The magnetic skeleton DMF in the
core of DMFL would enable this drug delivery system to display
magnetic targeting. The diphasic action between the DMF in lipo-
somes and the inhibition of phospholipid bilayer, together with the
inherent controlled-release function of DMF should significantly
strengthen slow-release performance of the DMFL. Moreover, the
dispersion and liquefaction of DMF with DMFL could provide a way
for intravascular injection medicining, improving the stability of
drugs and remaining the cell biology advantage of MLDH delivery
system.
3.3.3. The drug release of DMFL in vitro
Fig. 7 shows the release kinetics curve of DMFL, FUL prepared
with the same optimal prescription and FU solution. The release
of FU solution reaches equilibrium in 10 h which is much earlier
than DMFL and FUL. Both DMFL and FUL appear two kinetics pro-
cesses, including fast release in 0–2 h and slow release in 2–4 h. In
the period of 0–2 h, the release rate of both DMFL and FUL reaches
to nearly 40%. The fast release process may be caused by the factors
as follows. Firstly, the leakage of liposomes and un-encapsulated
drugs can affect the initial release (Yu et al., 2003). Secondly, the
liposomes in metastable state have a change process during the

3.3.2. Stability and magnetism of DMFL

The magnetic experiments prove that the DMFL prepared by the
orthogonal prescription are quite stable and sensitive to exterior
magnetic field. Both the FUL and DMFL present the opalescence
and dispersion stability as shown in Fig. 6A. There are no demixing
and sedimentation after 45 days in 4 ◦ C, which illustrates that
both FUL and DMFL are extraordinarily stable. The FUL and DMFL
in glass bottle intervened with magnetic field inclusion sixteen
neodymium iron boron permanent magnets (2000 Gs/unit) for
72 h, FUL have no change, but in the case of DMFL, the laminated
hyperchromic substances are gathered to the place corresponding
to the outside magnetics, and the color of other aqueous phase
becomes pallid, as displayed in Fig. 6B. In addition, the phenomena
in Fig. 6C are similar to those seen in Fig. 6B, after being intervened
for 5 min with the same magnetic field in vertical direction.
Our research indicates that the magnetic interference cannot
Fig. 6. Results of magnetic response of DMFL (A) the stabilize dispersion state of
destroy the structure and stability of liposomes and the DMFL
FUL and DMFL after standing for 45 days in 4 ◦ C; (B) the change of DMFL after 72 h
show the sensitive response and definite targeting for external by the beside of sixteen magnets; (C) DMFL interfered after 5 min by the beside of
magnetic. sixteen magnets.
 J. Huang et al. / International Journal of Pharmaceutics 450 (2013) 323–330
Fig. 7. Release curve of DMFL, FUL and FU in 24 h.
Table 4
Equation for release curve of DMFL.
Model Equation
Zero level dynamic equation Q = 1.7221t + 31.167
First order kinetic equation ln(100 − Q) = −0.0325t + 4.2353
Higuchi equation Q = 9.2338t1/2 + 22.821
dynamic action of the dissolution system, following the new oil
and water allocation formed with the released material. In 2–24 h,
the release curve of FUL almost becomes a straight line and con-
sists of the zero level dynamics, illustrating that the liposomes
can control the drug release ideally. The release tendency of DMFL
close to FUL, which indicates that the controlling level of the drug
release affected by the same lipidic structure is similar. However,
the release curve of DMFL is lower than that of FUL with equivalence
drugs due to the additional control of DMF, the core skeleton of
DMFL. Replacement of dissociative FU from “phospholipid bilayer
– DMF” system is easier than replacing DMF from DMFL. Along with
the consumption of free FU and the hysteresis release of remaining
drugs encapsulated by DET, the delayed release of DMFL starts to
appear at 15 h, releasing subsequent drugs come from the interlayer
galleries of MLDH-FU within DMF through ions exchange. At 24 h
the release rate of DMFL reaches only to 64%, while FUL reaches to
74%. It is clear that the release process of DMFL is more complicated
than FUL. The released drugs of DMFL come from the dissociative
FU packed between the phospholipid bilayer and DMF, insider FU
encapsulated by DET and intercalated FU between interlayers of
MLDH within DMF, and their release are controlled by the conjunct
limit of the phospholipid bilayer of liposomes, the core skeleton
DMF and the ions exchange course of MLDH-FU/PBS, respectively.
Furthermore, with the prolongation of time, the delayed release
character of DMFL will get revealed gradually attributed chiefly to
the control function of the core skeleton DMF.
As shown in Table 4, the Higuchi equation is the most suitable
model to describe the drug release of DMFL in 0–24 h. When we
take a look on that concretely, according to the curve change ten-
dency and piecewise fitting results of mechanism model, it might be
deduced that there exist actually two zero-order kinetic processes
corresponding to the intervals of 2–15 h and 18–24 h, respectively.
4. Conclusion
We had prepared “dextran-magnetic layered double hydroxide-
fluorouracil” magnetic targeted liposomes (DMFL) with reverse
evaporation method, according to the optimal prescription from
329
orthogonal experiment. In the optimal prescription, the weight
ratio of lecithin to cholesterol is 3:1, the weight ratio of drug to
lecithin is 1:7 and the emulsification time and temperature are set
respectively to 30 min and 50 ◦ C. The mean entrapment rate, pH
value and mean diameter of DMFL prepared by optimal prescription
were 85.47 ± 0.83, 6.58 ± 0.05 and 160.4 ± 0.55 nm, respectively;
and the drug release of DMFL in PBS was suitable for Higuchi equa-
tion Q = 9.2338t1/2 + 22.821. We have shown that DMFL possess
an outstanding controlled release features superior to the original
drug liposomes FUL, provided with magnetic targeting. In addi-
tion, we have found that the phospholipid bilayer and the core
skeleton DMF of DMFL could remain intact in lyophilized powder,
which is important for storing DMFL. Our findings may have impor-
tant implications for development of a liquid state formulation and
injection delivery approach to realize the special therapeutic action
of the DMF targeted delivery system.
Acknowledgment
This work was supported by the National Natural Science Foun-
dation of China (No. 81260483 and No. 20961008).
r References
0.8802 Chakraborti, M., Jackson, J.K., Plackett, D., Brunette, D.M., Burt, H.M., 2011. Drug
0.9354 intercalation in layered double hydroxide clay application in the development
0.9542 of a nanocomposite film for guided tissue regeneration. Int. J. Pharm. 416,
305–313.
Gou, G.J., Bao, F.J., Jiang, Y.X., 2009a. Synthesis method of dextran-magnetic layered
double hydroxide-fluorouracil. Patent CN 101607091B, issued 12/23.
Gou, G.J., Bao, F.J., Yang, J.H., 2009b. A synthesis method of magnetic layered double
hydroxide. Patent CN 101607088B, issued 12/23.
Gou, G.J., Wang, Z.Y., Liu, Y.H., Xue, B., Huang, J., Sun, Y., 2012a. Supra-molecular
assembly characterization and urgent toxic level of dextran-magnetic lay-
ered double hydroxide-fluorouracil drug delivery system. Acta Chim. Sinica 70,
161–169.
Gou, G.J., Liu, Y.H., Sun, Y., Huang, J., Xue, B., 2012b. Ion exchange reaction kinetics
of magnetic layered double hydroxides with 5-fluorouracil. Chem. J. Chinese U.
33, 119–127.
Gou, G.J., Liu, Y.H., Sun, Y., Huang, J., Xue, B., Dong, L.E., 2011. Super-molecular
assembly and magnetic targeted slow-release effect of dextran-magnetic lay-
ered double hydroxide-fluorouracil drug delivery system. Acta Pharm. Sinica
46, 1390–1398.
Gou, G.J., Ma, P.H., Chu, M.X., 2004. Preparation and basic properties of lay-
ered double hydroxides intercalated by chloride anion. Acta Chim. Sinica 62,
2150–2160.
Isacchi, B., Arrigucci, S., Marca, G.L., Bergonzi, M.C., Vannucchi, M.G., Novelli, A., Bilia,
A.R., 2011. Conventional and long-circulating liposomes of artemisinin: prepa-
ration, characterization, and pharmacokinetic profile in mice. J. Liposome Res.
221, 237–244.
Karan, B.Z., Ay, A.N., 2012. Layered double hydroxides – multifunctional nanoma-
terials. Chem. Pap. 66, 1–10.
Kim, M.J., Jang, D.H., Lee, Y.I., Jung, H.S., Lee, H.J., Choa, Y.H., 2011. Preparation,
characterization, cytotoxicity and drug release behavior of liposome-enveloped
paclitaxel/Fe3 O4 nanoparticl. J. Nanosci. Nanotechnol. 11, 889–893.
Kuznetsov, A.A., Filippov, V.I., Alyautdin, R.N., Torshina, N.L., Kuznetsov, O.A., 2001.
Application of magnetic liposomes for magnetically guided transport of mus-
cle relaxants and anti-cancer photodynamic drugs. J. Mang. Mang. Mater. 255,
95–100.
Li, B.G., He, J., Evans, D.G., Duan, X., 2004. Inorganic layered double hydroxides as a
drug delivery system—intercalation and in vitro release of fenbufen. Appl. Clay
Sci. 27, 199–207.
Li, L.P., Wang, Q., 2009. Preparation techniques of curcumin liposome. J. Fourth Mil.
Med. Univ. 30, 3166–3168.
Liang, C.L. 2010. The study on preparation and pharmacodynamic of 5-fluorouracil
magnetic solid lipid nanoparticles. Zhenzhou.
Liu, R.Z., Gan, L., Yang, X.L., Xu, H.B., 2011. Chitosan as a condensing agent induces
high gene transfection efficiency and low cytotoxicity of liposome. J. Biosci.
Bioeng. 11, 98–103.
Nalawade, P., Aware, B., Kadam, V.J., Hirlekar, R.S., 2009. Layered double hydroxides:
a review. J. Sci. Ind. Res. India 68, 267–272.
Nie, S.Q., Zen, K., Pang, G.H., Lin, K.C., 1989. Studyes on the interaction of
BPP with DPPC liposome by FT-IR spectroscopy. Acta Bioch. Bioph. Sin. 5,
208–211.
Parello, M.L., Rojas, R., Giacomelli, C.E., 2010. Dissolution kinetics and mechanism of
Mg–Al layered double hydroxides: a simple approach to describe drug release
in acid media. J. Colloid Interface Sci. 351, 134–139.
Sun, W.T., Zhang, N., Li Zou, W.W., Xu, W.F., 2008. Preparation and evaluation of
N3 -O-toluyl-fluorouracil-loaded liposomes. Int. J. Pharm. 353, 243–250.
330 J. Huang et al. / International Journal of Pharmaceutics 450 (2013) 323–330

Wu, M.C., Zhu Li Zhao Yi Pan, L.J., Wang, C., 2009. Study on preparation of 5 -
palmitoyl-5-fluorouridine liposome and its cell toxicity. Chin. Pharm. J. 44,
520–523.
Yu, B.T., Sun, X., Zhang, Z.R., 2003. Enhanced liver targeting by synthesis of N1-
stearyl-FU and incorporation into solid lipid nanoparticles. Arch. Pharm. Res.
26, 1096–1101.
Zhao, X., Yang, H., Chen, E.G., Yu, G.L., Hao, W.G., 2007. Preparation of polygu-
luronate sulfate liposomes and determination of its encapsulation efficiency.
Chin. Pharm. J. 42, 1476–1479.
Zhu, L., Wang, L.L., NI, K.Y., Tong, X., Dai, L., Tu, J., 2006. Preparation of stable magne-
toliposomes with high encapsulation efficiency of methotrexate. Chin. Pharm. J.
41, 1324–1327.