Biomaterials 32 (2011) 3459e3470

Contents lists available at ScienceDirect

Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials

The promotion of siRNA delivery to breast cancer overexpressing epidermal

growth factor receptor through anti-EGFR antibody conjugation by
immunoliposomes
Jie Gao a, c,1, Wei Liu b,1, Yu Xia a,1, Wei Li a, c, Jing Sun b, Huaiwen Chen a, c, Bohua Li a, c, d, Dapeng Zhang a, c,
Weizhu Qian a, c, Yanchun Meng a, c, Li Deng b, Hao Wang a, c, d, Jianming Chen b, **, Yajun Guo a, c, d, *
a

International Joint Cancer Institute, The Second Military Medical University, 800 Xiang Yin Road, Shanghai 200433, PR China
Department of Pharmaceutical Science, College of Pharmacy, The Second Military Medical University, 325 Guo He Road, Shanghai 200433, PR China
National Engineering Research Center for Antibody Medicine, State Key Laboratory of Antibody Medicine & Targeting Therapy and Shanghai Key Laboratory of Cell Engineering,
399 Libing Road, Shanghai 201203, PR China
d
PLA General Hospital Cancer Center, PLA Graduate School of Medicine, Beijing 100853, PR China
b
c

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 27 December 2010
Accepted 13 January 2011
Available online 5 February 2011

The LPD (liposome-polycation-DNA complex) is an effective nanovector for systemically small interfering
RNA (siRNA) delivery which was well characterized previously. However, little effort was spend on the
development of targeted LPD conjugated with tumor specic antibody (TLPD) which would be potent in
promoting siRNA delivery in tumor. Here, we prepared TLPD through a self-assembling process followed
by anti-EGFR antibody conjugation. The effect of antibody type, conjugation strategy and amount on the
physicochemical and biological properties of TLPD was investigated. We obtained optimized TLPD
conjugated with anti-EGFR Fab by conventional conjugation (TLPD-FCC), which possessed a small size
around 150 nm and superior in vitro stability. Compared with nontargeted LPD (NTLPD), TLPD-FCC
showed signicantly enhanced binding afnity and luciferase gene silencing activity in EGFR overexpressing MDA-MB-231 breast cancer cells in vitro. Moreover, the in vivo accumulation of TLPD-FCC was
obviously higher than that of NTLPD in MDA-MB-231 tumor 24 h post intravenous injection. The
promoted uptake of TLPD-FCC in MDA-MB-231 tumor was further conrmed by confocal microscopy.
Notably, three intravenous injections of siRNA in TLPD-FCC signicantly silenced luciferase expression by
w20%, whereas NTLPD showed little effect. All these results suggested that our TLPD-FCC have a great
potential in delivering siRNA to EGFR overexpressing breast cancers.
2011 Elsevier Ltd. All rights reserved.

Keywords:
Antibody
EGFR
Gene silencing
PEG
siRNA delivery

1. Introduction
Gene therapy using small interfering RNA (siRNA) represents
a potent and specic method in tumor targeted therapy [1].
Although siRNA has the potential to be a powerful therapeutic
drug, the efcient siRNA delivery to target tissues following
systemic administration is the most important hurdle for widespread use of siRNA in vivo [2,3]. Therefore, the major challenge for
siRNA applications is developing an efcient siRNA systemic
delivery system [4,5]. Although there is still much to improve,

* Corresponding author. International Joint Cancer Institute, The Second Military

Medical University, 800 Xiang Yin Road, Shanghai 200433, PR China. Tel./fax: 86
21 81870801.
** Corresponding author. Tel./fax: 86 21 81871291.
E-mail addresses: yjcjm@163.com (J. Chen), yjguo@smmu.edu.cn (Y. Guo).
1
Note: Jie Gao, Wei Liu and Yu Xia contributed equally to this paper.
0142-9612/$ e see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2011.01.034

lipid-based delivery system has been successfully used in siRNA

systemic delivery [5]. Huang et al. developed a well-established
LPD (liposome-polycation-DNA complex) and it was shown to
deliver siRNA efciently to tumor cells overexpressing sigma
receptor [6,7]. The small molecule anisamide was adopted as the
targeting ligand of LPD. However, antibodies, including monoclonal antibodies (mAbs) or mAb fragments, are among the most
frequently used ligands for targeted nanoparticles [8]. Compared
with other targeting ligands, antibodies have many unique
advantages. Firstly, antibodies have much higher specicity and
afnity than the small molecule ligands. Secondly, antibodies own
huge amounts of amino, carboxyl and thiol groups, providing
exible conjugation sites for crosslinking of nanoparticles [9].
Thirdly, many mAbs are approved by US FDA (Food and Drug
Administration) for cancer therapy, whereas none of other ligands
(such as RGD peptides) are approved. Among them the anti-EGFR
(epidermal growth factor receptor) mAbs or its derivates are

3460

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

widely used for selective delivery of nanoparticles to cancer cells

overexpressing EGFR [10,11].
Although numerous reports demonstrated antibodies can
signicantly enhance the targeting efciency of liposomes in
delivering chemotherapeutics systemically, there have been relatively quite few reports in the literature that demonstrate antibodies could enhance siRNA systemic delivery [12e15]. Most
antibodies are protein macromolecules of electronegativity, having
opposite charges against cationic liposomes for siRNA delivery,
might have a tendency to induce instability, aggregation and
content leakage for liposomes delivering siRNA, leading to impaired
targeting efciency. For example, antibody conjugation may
signicantly affect the size of liposomes. However, the exact role of
antibody conjugation on size of liposomes remains uncharacterized. The increase in size of antibody-conjugated nanoparticles
should be avoided or alleviated, because nanoparticles of small size
have a much longer circulation time in vivo than larger ones [16,17].
Moreover, antibody conjugation may alter the surface charge,
hydrophobic or hydrophilic properties of liposomes. All these
physicochemical properties greatly contribute to the stability and
biologic activity of liposomes. Considering antibodies play a vital
role in targeted liposomes, a practical strategy of investigating the
effect of antibodies type, conjugation strategy and amount on the
essential physicochemical properties (such as size, charge, et al.)
and targeting efciency of liposomes for siRNA delivery should be
initiated. Although little of such work has been found presently
from the literature, this investigation represents an important
aspect at the cutting edge in the design of antibodies targeted
liposomes for siRNA delivery. If these problems were made clear,
a better application would be achieved for siRNA systemic delivery.
Here, we investigated the effect of antibody type, conjugation
strategy and amount on the essential physicochemical properties
and targeting efciency of targeted LPD conjugated with anti-EGFR
antibody (TLPD). The optimized TLPD conjugated with anti-EGFR
Fab by conventional conjugation strategy (TLPD-FCC) was obtained.
2. Materials and methods
2.1. Materials
1,2-Dioleoyl-3-trimethylammonium-propane (chloride salt) (DOTAP), cholesterol
(Chol), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene
glycol)-2000] (ammonium salt) (DSPE-mPEG) and 1,2-distearoyl-sn-glycero-3phosphoethanolamine-N-[maleimide(polyethylene glycol)-2000] (ammonium salt)
(DSPE-PEG-Mal) were purchased from Avanti Polar Lipids (Alabaster, AL). Protamine
sulfate salt from salmon (Grade X, amorphous powder) and deoxyribonucleic acid
solution from calf thymus (calf thymus DNA, sonicated DNA for hybridization) were
obtained from SigmaeAldrich (St. Louis, MO). All organic reagents are of analytical
grade and purchased from Sinopharm (Shanghai, China). Human-mouse chimeric
anti-EGFR mAb was kindly provided by National Engineering Research Center for
Antibody Medicine (Shanghai, China) and Fab of anti-EGFR mAb (anti-EGFR Fab) was
prepared as we described previously [18]. FITC labeled goat anti-human IgG (H L)
was purchased from Zymed (San Francisco, CA). 2-iminothiolane (Trauts Reagent)
was obtained from Pierce (Oud Beijerland, NL). Anti-luciferase siRNA (sense: 50 CUUACGCUGAGUACUUCGATT-30 , antisense: 50 -UCGAAGUACUCAGCGUAAGTT-30 ),
negative control (NC) siRNA and FAM-labeled NC siRNA (FAM-siRNA) were synthesized by GenePharma Co., Ltd (Shanghai, China). Cy5-labeled NC siRNA (Cy5-siRNA)
synthesized by Ribobio Co., Ltd (Guangzhou, China). Dulbeccos modied Eagles
medium (DMEM) and fetal bovine serum (FBS) were purchased from Invitrogen
(Carlsbag, CA, USA).
2.2. Cell lines
The human breast cancer cell lines MDA-MB-231, SK-BR3 and MCF-7 were
purchased from ATCC (American Type Culture Collection, VA). The MDA-MB-231
cells were stably transfected with luciferase gene using pcDNA3.1 vector. The cells
were maintained in DMEM supplemented with 10%FBS, 25 mM HEPES buffer,
100 U/ml penicillin and 100 mg/ml streptomycin in a humidied atmosphere of 5%
CO2 at 37  C. Supplementary Fig. 1 shows that MDA-MB-231 cells overexpressed
EGFR, SK-BR3 cells expressed EGFR moderately, while MCF-7 cells expressed EGFR
lowly.

2.3. Liposome preparation

Liposomes were prepared as described previously with slight modications
(Fig. 1) [6,19]. Briey, cationic liposomes composed of DOTAP and Chol (1:1 M ratio,
10 mM) were prepared by the thin lm hydration method, followed by membrane
extrusion with a hand held extruder (Avestin, Ottawa). The extrusion was performed
in a stepwise manner using progressively decreasing pore-sized membranes (from
400, 200, 80 and 50 nm) (Nucleopore, Whatman), with 10e20 cycles per pore-size. To
prepared naked LPD, 124 ml cationic liposomes, 15 ml protamine (2 mg/ml) and 11 ml
DEPC water were mixed as solution A and kept at room temperature (RT) for 10 min.
Meanwhile, 90 ml siRNA (24 mg, 20 mM), 2.4 ml calf thymus DNA (24 mg, 10 mg/ml) and
57.6 ml DEPC water were also mixed as solution B and kept at RT for 10 min. The
solution A and B were mixed quickly to form naked LPD and incubated at RT for
15 min. For antibody conjugation, antibody (anti-EGFR mAb and Fab) was thiolated
by 2-iminothiolane at a molar ratio of 50:1. The thiolated antibody was dialyzed in
distilled water for 48 h to remove excess 2-iminothiolane. The sulfhydryl groups of
thiolated antibody were determined using Ellmans assay [20]. For conventional
conjugation, naked LPD was mixed with 37.8 ml micelle solution of DSPE-PEG-Mal
(10 mg/ml) to form LPD at 50  C for 10 min. A different amount (120, 300, 600 and
1200 mg) of thiolated antibody was mixed with LPD and the mixture was incubated at
16  C overnight to form targeted LPD conjugated with anti-EGFR antibody (TLPD). For
post-insertion, a different amount (120, 300, 600 and 1200 mg) of thiolated antibody
was rst incubated with 37.8 ml micelle solution of DSPE-PEG-Mal (10 mg/ml) at 16  C
overnight. Then naked LPD was mixed with the antibody-conjugated micelles to form
TLPD at 50  C for 10 min.
Nontargeted LPD (NTLPD) was prepared in the same way as TLPD except that
DSPE-PEG-Mal was replaced by DSPE-mPEG. All the reagents and equipment used in
preparing liposomes were sterile. The prepared naked LPD, TLPD and NTLPD were
stored at 4  C. The following designations are used: TLPD-FC (TLPD conjugated with
anti-EGFR Fab by conventional conjugation), TLPD-FP (TLPD conjugated with antiEGFR Fab by post-insertion), TLPD-MC (TLPD conjugated with anti-EGFR mAb by
conventional conjugation) and TLPD-MP (TLPD conjugated with anti-EGFR mAb by
post-insertion).
2.4. Characterization of liposomes
2.4.1. Particle size and zeta potential
After the liposomes were dispersed in deionized water, their size and zeta
potential were analyzed using Zeta sizer Nano S (Malvern instruments, UK).
2.4.2. Transmission electron microscopy (TEM)
The morphological examination of liposomes was performed by TEM. Briey,
samples were prepared by dropping one drop of liposomes dispersion onto a copper
grid coated with a carbon membrane. Then the samples were stained by 2% phosphotungstic acid (PTA) and dried. The liposomes were visualized under the Hitachi
H-600 TEM (accelerating voltage of 75 kV).
2.4.3. Gel retardation assay
The siRNA binding afnity of liposomes was evaluated by gel retardation assay.
The liposomes (untreated or destroyed by Triton-X100) were loaded into individual
wells of agarose gel, electrophoresed and visualized with Goldenview dye
staining.
2.4.4. Evaluation of siRNA encapsulation efciency (EE)
The siRNA EE of the liposomes was determined by ultra-ltrating the liposomes
entrapping FAM-siRNA with DEPC water using Amicon Ultra-4 centrifugal lter
devices (100,000 NMWL, Millipore, Massachusetts). After thoroughly ultra ltration,
unencapsulated FAM-siRNA was collected and quantied using a calibration line
constructed by a number of FAM-siRNA solutions with known concentrations. The
uorescence of FAM-siRNA was measured with the spectrouorometer (Synergy
4, Biotek, Wi, USA) using excitation and emission wavelengths of 495 and 525 nm,
respectively. The siRNA EE was calculated from the following formula: (Mi  MU)/
Mi  100%. MU and Mi were denoted as the mass of unencapsulated siRNA and
initially added siRNA, respectively.
2.4.5. siRNA serum stability
Serum stability of siRNA in aqueous solution versus in the liposomes was
characterized using agarose gel electrophoresis. Samples of siRNA either in aqueous
solution or liposomes were mixed in a 1:1 volume ratio with FBS to give 50% serum
concentration and incubated at 37  C. At different time points, aliquots containing
0.25 mg siRNA of each sample (the liposomes were rst destroyed by Triton-X100)
were loaded onto a gel and electrophoresis was performed to visualize intact siRNA.
2.4.6. Dynamic light scattering (DLS)
The interaction of the liposomes with bovine serum albumin (BSA) was examined
by dynamic light scattering (DLS) using Zeta sizer Nano S. Firstly, the size and size
distribution of the liposomes and BSA solution (Concentration (C) 1 mg/ml) was
measured by DLS. Secondly, we prepared a liposome solution with C 28 mg/ml (total
mass/volume) by diluting the stock solution and a BSA solution with C 58 mg/ml.

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

3461

Fig. 1. The procedures involved in developing naked LPD, LPD and TLPD. (A) Naked LPD composed of DOTAP/Chol liposomes, protamine, calf thymus DNA and siRNA was prepared
by a self-assembling process. Then LPD was formed by mixing naked LPD with DSPE-PEG-Mal micelles. For conventional conjugation, thiolated antibody was incubated with LPD to
form TLPD; for post-insertion, thiolated antibody was rst incubated DSPE-PEG-Mal micelles to form antibody-conjugated micelles, which was then mixed with naked LPD to form
TLPD. (B) Thiolated antibody preparation. Fab was prepared by papain digestion of mAb. Antibody (including Fab and mAb) was thiolated by 2-iminothiolane to obtain antibody
with sulfhydryl groups.

Then, 1 ml diluted liposomes solution was sequentially mixed with 240 ml BSA
solution. The mixture was tuned by 1 ml distilled water and 200 ml PBS (pH 7.4) to
keep the mass ratio of liposomes and BSA was about 1/500 in the DLS experiments. At
last, the size and size distribution of the mixture was measured by DLS.

(Promega). The protein concentrations of the samples were determined by using

a MicroBCA protein assay kit (Beyotime Biotechnology. Haimen, China). Luciferase
activity of a sample was normalized with the protein concentration and expressed as
percent luminescence intensity compared to the untreated control.

2.5. In vitro transfection efciency

2.7. Animal studies

For transfection efciency analysis, MDA-MB-231, SK-BR3 or MCF-7 cells were

seeded in 48-well plates at a density of 7.5  104 cells per well overnight. Cells were
treated with the liposomes at a concentration of 250 nM or 500 nM FAM-siRNA for
24 h. The cells were trypsinized, washed and the mean uorescence intensity (MFI)
was analyzed by FCM (ow cytometry). FCM was performed using a FACScan ow
cytometer (Becton Dickinson, San Jose, CA).

All animals were purchased from the Shanghai Experimental Animal Center of
Chinese Academic of Sciences (Shanghai, P. R. China). Mice were placed in a pathogen-free environment and allowed to acclimate for a week before being used in
studies. All procedures were performed in accordance with guideline of the
Committee on Animals of the Second Military Medical University, Shanghai, P. R.
China.

2.6. In vitro gene silencing

2.8. Tissue distribution

The in vitro gene silencing assay was taken in MDA-MB-231 cells or MDA-MB231 cells mixed with SK-BR3 cells or MDA-MB-231 cells mixed with MCF-7 cells in
48-well plates. For pure cells, MDA-MB-231 cells were seeded at a density of
1.5  104 cells per well overnight. For mixed cells, MDA-MB-231 cells were seeded at
a density of 7.5  103 cells per well, mixing with SK-BR3 or MCF-7 cells (3  104 cells
per well). Cells were treated with liposomes entrapping NC or anti-luciferase siRNA
(the nal siRNA concentration was 250 nM or 500 nM) for 24 h until fresh culture
medium was changed. 48 h later, the luciferase activity of the cells was measured
according to the protocol provided by Promega with slight modication. Briey, the
cells were washed and incubated with lysis buffer at RT for 20 min. Five ml cell lysate
was mixed with 25 ml substrate (Luciferase Assay System, Promega Co., Madison, WI)
and the luminescence was measured by GloMax 96 Microplate Luminometer

Balb/c nude mice (female, 4e6 weeks, w20 g) were inoculated s.c. on the right
back with 5  106 MDA-MB-231 cells. Once tumors reached about 300e500 mm3,
successfully establishment of tumor xenograft model was demonstrated by tumor
luminescent images. Briey, mice were given an intraperitoneal injection of luciferin
(Promega) at a dose of 150 mg/kg. The successfully establishment of tumor xenograft model was demonstrated by IVIS Lumina II Imaging System (Xenogen), which
was taken to capture the visible light photograph and luminescent image
(Supplementary Fig. 2). Immunouorescence staining was taken to examine EGFR
expression of MDA-MB-231 tumor xenograft. As shown in Supplementary Fig. 3,
immunouorescence revealed a high EGFR expression in MDA-MB-231 tumor
tissues, indicating that MDA-MB-231 cells did not lost EGFR expression in vivo. For
tissue distribution study, mice were randomly assigned to treatment groups: 3 mice

3462

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

per group  3 groups. Liposomes entrapping Cy5-siRNA were injected i.v. as a single
dose (1.2 mg siRNA/kg) via tail vein. The mice were anaesthetized by inhalation. At
different time points after injection, the in vivo images were observed with IVIS
imaging system (excitation 640 nm) and recorded by a built-in CCD camera. After
24 h, the mice were killed, and the excised tumors and major organs were also
imaged.
For in vivo uptake study, tumors were collected from sacriced mice, immediately placed in OCT medium, and rapidly frozen in liquid nitrogen. Frozen sections
were cut at 10 mm sections and xed with acetone at 20  C. After washing with
PBS, sections were counterstained with 40 ,6-Diamidino-2-phenylindole dihydrochloride (DAPI; Sigma, Fluka Chemie, Buchs, Israel) and visualized with Leica TCS
SP5 Confocal Microscope.
2.9. In vivo gene silencing
Balb/c nude mice (female, 4e6 weeks, w20 g) were inoculated s.c. on the right
and left back with 5  106 MDA-MB-231 cells. Once tumors reached about
w50 mm3, mice were randomly assigned to treatment groups: 4e5 mice per
group  5 groups. Mice were injected via the tail vein with different formulations
(1.2 mg siRNA/kg, one injection per day for 3 days). One day after the third injection,
mice were euthanized and the tumors were excised. The tumors was weighed and
homogenized in lysis buffer (1 ml lysis buffer per 100 mg) followed by centrifugation at 12,000 g for 10 min. The luciferase activity of the samples was determined as
described in Section 2.6.
2.10. Statistical analysis
Direct comparison between two groups was conducted by Students unpaired t
test. P value of <0.05 was considered statistically signicant.

3. Results
3.1. Particle size and zeta potential
Naked LPD, LPD and NTLPD had a similar particle size of
140e150 nm. We investigated the correlation of the conjugation
strategy types, antibody types and amount with the size and zeta
potentials of TLPD (Fig. 2). Different amounts of mAb and Fab were
conjugated to LPD by conventional conjugation and post-insertion
strategies. The size of TLPD varied markedly (Fig. 2A). Nearly all
prepared TLPD-MC and TLPD-MP had a size larger than 200 nm,
indicating that mAb conjugation have a signicant impact on size of
TLPD. At the same amount of conjugated antibody, the size of TLPDMC was smaller than that of TLPD-MP, and the size of TLPD-FC was
smaller than that of TLPD-FP, suggesting that post-insertion

strategy is more likely to increase the size of TLPD than conventional conjugation strategy. It has to be noted that the general size
of TLPD-FC was between 150 and 160 nm. The liposomes of such
a small size may be desired for longer circulation time and high
cellular uptake efciency than liposomes of size larger than 200 nm
[16,17]. As a result, TLPD-FC was used in the subsequent experiments. For convenience, TLPD-FC conjugated with 120, 300, 600
and 1200 mg Fab was denoted as TLPD-FCA, TLPD-FCB, TLPD-FCC
and TLPD-FCD respectively.
The zeta potential of liposomes was also investigated. The zeta
potential of naked LPD was 48.2  3.5 mV (mean  SD, n 4). After
PEGylation, the zeta potential of LPD and NTLPD was 21.7  2.1 mV
and 22.9  2.5 mV, respectively (mean  SD, n 4), indicating
PEGylation can signicantly reduce the zeta potential of the liposomes. After antibody conjugation, the zeta potential of TLPD
further decreased to a value below 12 mV (Fig. 2B). Accompanying
with increased amount of conjugated antibody, the zeta potential of
TLPD gradually decreased, and was even measured as negative
values.
3.2. Characterization of TLPD-FC
3.2.1. Gel retardation assay
Gel retardation assay was commonly used to examine the
siRNA binding afnity of liposomes. Fig. 3A shows no siRNA
migration was observed for intact naked LPD, NTLPD and TLPD-FC,
and the encapsulated siRNA was completely retarded in the wells
of gel. However, the siRNA was released and showed bright bands
after destroyed by the detergent. These results conrmed that
naked LPD, NTLPD and TLPD-FC all had powerful binding afnity to
siRNA.
3.2.2. siRNA encapsulation efciency (EE)
The siRNA EE of liposomes was precisely calculated by subtracting unencapsulated siRNA from total siRNA. Consistent with
results obtained in the gel retardation assay, the siRNA EE of all
liposomes was very high (>90%) (Fig. 3B), indicating that naked
LPD, NTLPD and TLPD-FC have potent encapsulation capacity to
siRNA, and PEGylation and antibody conjugation have little adverse
impact on siRNA EE for NTLPD and TLPD-FC.

Fig. 2. The size (A) and zeta potential (B) of TLPD. TLPD conjugated with a different amount (120 mg, 300 mg, 600 mg and 1200 mg) of antibody was prepared by conventional
conjugation or post-insertion strategies. The implications of TLPD-FC, TLPD-MC, TLPD-FP and TLPD-MP were seen in Section 2.3. Their particle size and zeta potential were analyzed
using Zeta sizer Nano S. Data are expressed as mean  SD for 4e6 samples.

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

3463

Fig. 3. Gel retardation assay (A) and siRNA EE (B). (A) Gel retardation assay. Naked LPD (lane 1, 2), NTLPD (lane 3, 4), TLPD-FCA (lane 5, 6), TLPD-FCB (lane 7, 8), TLPD-FCC (lane 9, 10)
and TLPD-FCD (lane 11, 12). The latter and former lane indicated samples untreated or destroyed by Triton-X100, respectively. Untreated siRNA was run in lane 13. Each lane contains
0.4 mg siRNA. The siRNA was visualized with Goldenview dye staining. The arrows indicated the free siRNA. (B) siRNA EE determined by ultra ltration method. The siRNA EE was
calculated from the following formula: (Mi  MU)/Mi  100%. MU and Mi were denoted as the mass of unencapsulated siRNA and initially added siRNA, respectively. Data are
expressed as mean  SD (n 3).

3.2.3. Determination of conjugated antibody on the surface of

liposomes
The presentation and integrity of anti-EGFR Fab on TLPD-FC
were conrmed by SDS-PAGE (Supplementary Fig. 4). Compared
with the native form (Lanes 1e3), anti-EGFR Fab conjugated to
TLPD-FC was intact and had a slightly larger molecular weight
(Lanes 4e7). The result conrmed that the anti-EGFR Fab was
conjugated to TLPD-FC and had a slower swimming motility in the
gels. No conjugated antibody was detected in naked LPD and NTLPD
(data not shown).

The morphology of all liposomes appeared spherical in shape. It is

noticed that naked LPD showed a relatively monodispersed size of
about 100 nm and a nely disseminated pattern (Fig. 5A). After
PEGlyation, however, NTLPD and TLPD-FCC showed a signicant
increase in the population of small particles, which consist of
independent particles and sprouts (about 20e30 nm) on the
100 nm-particles (Fig. 5B, C). This phenomenon was in accordance
with results previously reported [21]. The small particles were
made up of DSPE-PEG micelles, DSPE-PEG containing particles and

3.2.4. In vitro gene silencing of TLPD-FC

Luciferase has been widely used as a target in gene silencing
experiments, because it could be easily detected by a fast and
sensitive method. We examined the luciferase gene silencing
activity of TLPD-FC (TLPD-FCA, TLPD-FCB, TLPD-FCC and TLPD-FCD)
in MDA-MB-231 cells. As shown in Fig. 4, accompanying with
increased amount of conjugated antibody, the gene silencing
activity of TLPD-FC gradually increased, reaching a culmination
when the conjugated antibody was 600 mg (TLPD-FCC). However,
the gene silencing activity decreased in TLPD-FC when the amount
of conjugated antibody was 1200 mg (TLPD-FCD). In conclusion,
TLPD-FCC possessed the best gene silencing activity among TLPDFC. As a result, TLPD-FCC was used in the subsequent experiments.

3.3. Characterization of TLPD-FCC

3.3.1. TEM
The size and morphology of TLPD-FCC, NTLPD and naked LPD
were observed by TEM. Fig. 5 shows the liposomes stained by PTA.
The color of the stained region (i.e. liposomes) is shown as grey
white on the TEM photograph, as PTA is a negative staining reagent.

Fig. 4. In vitro gene silencing of TLPD-FC. MDA-MB-231 cells were incubated with
TLPD-FC (including TLPD-FCA, TLPD-FCB, TLPD-FCC and TLPD-FCD) at 37  C for 24 h.
After 48 h, luciferase activities of the cells were analyzed, normalized with the protein
concentration and expressed as percent luminescence intensity compared to the
untreated control. Data are expressed as mean  SD (n 3).

3464

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

Fig. 5. TEM. Naked LPD (A), NTLPD (B) and TLPD-FCC (C) were prepared by dropping one drop of liposomes dispersion onto a copper grid coated with a carbon membrane, then the
samples were stained by 2% PTA and dried. The liposomes were visualized under the TEM. Bar represents 100 nm.

particles stripped off. However, these small particles cannot be

discriminated, since they all showed similar size of 20e30 nm.
Furthermore, there is no difference between NTLPD and TLPD-FCC
in shape and size distribution, suggesting that antibody conjugation have little impact on the structure of liposomes.
Notably, the size of liposomes observed by TEM is smaller than
that obtained from the DLS experiment (about one-third
decrease). The reason is that the size of liposomes obtained from
the DLS experiment reects the hydrodynamic size of liposomes,
whereas the size of liposomes observed by TEM shows that of
dried liposomes [22]. As a result, the observed decrease size was
expected since the hydration of the shell portion of liposomes
disappeared in TEM.
3.3.2. siRNA degradation in serum
Serum stability assay was routinely performed to characterize
liability of the liposomes entrapping siRNA in aqueous solution of
50% serum [19,23]. All samples were incubated with an equal
volume of serum and incubated at 37  C. At xed time points,
aliquots were taken to visualize intact siRNA. Fig. 6 shows the
results of siRNA degradation of all samples in serum. A positive
control sample containing intact siRNA (lane 5) was run along with
all samples. As expected, pure siRNA displayed only faint bands
after 3 h and became completely degraded after 6 h (lane 4), suggesting that siRNA is prone to degradation caused by serum. In
contrast, siRNA in naked LPD, NTLPD and TLPD-FCC was protected

well in serum, and remained relatively intact even after 48 h (lane

1, 2 and 3). The entire length of the siRNA degradation study was
144 h, because serum precipitated after 144 h. It has to be noted
that after 36 h, siRNA in naked LPD exhibited fainter bands than
that in NTLPD and TLPD-FCC, suggesting that NTLPD and TLPD-FCC
possess lower siRNA serum degrading rate than naked LPD.
3.3.3. Dynamic light scattering (DLS)
Fig. 7 shows the interaction of naked LPD, NTLPD and TLPD-FCC
with BSA measured by DLS. As shown in Fig. 7A, just one single
peak for BSA, naked LPD, NTLPD and TLPD-FCC indicated their
relative narrow distribution in the experimental condition. But for
the naked LPD and BSA mixture, their corresponding single peak
showed in Fig. 7B disappeared and merged into one broad
distributed peak. On the contrary, for the solutions of NTLPD or
TLPD-FCC mixing with BSA, the peaks of all samples are same as
their characteristic single peaks measured alone (Fig. 7B). When
mixed with BSA, the peak coupling appeared in naked LPD solution
but not in NTLPD and TLPD-FCC, indicating that the strong interaction exists between naked LPD and BSA.
3.4. Transfection efciency of TLPD-FCC
We chose three breast cancer cell lines with different EGFR
expression levels to evaluate the transfection efciency of TLPDFCC and NTLPD. Fig. 8 shows that TLPD-FCC had signicantly

Fig. 6. siRNA serum stability. Samples of siRNA either in aqueous solution or lipsomes were mixed in a 1:1 ratio with fresh serum to give 50% serum concentration and incubated at
37  C. At different incubation times, aliquots containing 0.25 mg siRNA of each sample (the liposomes were rst destroyed by Triton-X100) were loaded onto a gel and electrophoresis was performed to visualize siRNA. Lane 1, naked LPD; lane 2, NTLPD; lane 3, TLPD-FCC; lane 4, siRNA; lane 5, intact siRNA.

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

3465

Fig. 8. In vitro transfection efciency. MDA-MB-231, SK-BR3 or MCF-7 cells were

seeded in 48-well plates with a density of 7.5  104 cells per well overnight. TLPD-FCC
or NTLPD entrapping FAM-siRNA were incubated with the cells (nal siRNA concentration was 250 nM or 500 nM) for 24 h. The cells were trypsinized, washed and the
MFI was analyzed by FCM.

Fig. 7. The interaction of liposomes and BSA was examined by DLS. (A) The size and
size distribution of BSA solution (C 1 mg/ml), naked LPD, NTLPD and TLPD-FCC was
measured by DLS respectively. (B) The size and size distribution of naked LPD, NTLPD
and TLPD-FCC mixing with BSA was measured by DLS respectively. The mixture was
tuned by 1 ml distilled water and 200 ml PBS (pH 7.4) to keep the mass ratio of liposomes and BSA was about 1/500 in the DLS experiments.

increased transfection efciency in MDA-MB-231 cells compared

with NTLPD (1.5 or 1.36 fold higher in 250 nM or 500 nM siRNA). In
SK-BR3 cells, the increased transfection efciency was moderate
(1.25 or 1.13 fold higher in 250 nM or 500 nM siRNA). However, no
increased transfection efciency was observed in the MCF-7 cells,
which had the lowest EGFR expression. These results rmly
demonstrated that anti-EGFR Fab conjugation could specically
enhance the binding afnity of TLPD-FCC in breast cancer cells of
high or moderate EGFR expression, but not in cells with low EGFR
expression. These results were consistent with our previous results
about the in vitro targeting efciency of targeted nanoparticles in
hepatocellular carcinoma cell lines with different tumor specic
antigen expression [24].
3.5. In vitro gene silencing of TLPD-FCC
We examined the luciferase gene silencing activity of TLPD-FCC
and NTLPD in MDA-MB-231 cells. As shown in Fig. 9A, TLPD-FCC
showed signicantly higher gene silencing activity than NTLPD at
250 nM and 500 nM anti-luciferase siRNA concentration (P < 0.05;
P < 0.001). As it should be, NC siRNA formulated in TLPD-FCC and
NTLPD showed inferior activity. Our results agree with that
obtained in the in vitro transfection efciency, suggesting that

TLPD-FCC show signicantly enhanced gene silencing activity

compared to NTLPD.
The in vivo environment consists of various cells with different
EGFR expression levels, and the specic gene silencing activity of
TLPD-FCC to the cells overexpressing EGFR should be investigated.
We employed MDA-MB-231 cells mixed with SK-BR3 cells or MDAMB-231 cells mixed with MCF-7 cells to examine the specic gene
silencing of TLPD-FCC. Fig. 9B shows the gene silencing of TLPD-FCC
in MDA-MB-231 cells mixed with SK-BR3 cells. Similar to the
results obtained in MDA-MB-231 cells, TLPD-FCC showed signicantly higher gene silencing activity than NTLPD at 250 nM and
500 nM anti-luciferase siRNA concentration (P < 0.01; P < 0.05). In
MDA-MB-231 cells mixed with MCF-7 cells, TLPD-FCC also displayed a superior gene silencing activity compared with NTLPD at
250 nM and 500 nM anti-luciferase siRNA concentration (P < 0.01;
P < 0.001) (Fig. 9C). These results conrmed the specic gene
silencing activity of TLPD-FCC to breast cancer cells overexpressing
EGFR.
3.6. Tissue distribution
Cy5 uorophore was employed in tissue distribution owing to
its superior stability and long-wavelength excitation and emission,
which is suitable for in vivo imaging. The images of Cy5-siRNA and
NTLPD entrapping Cy5-siRNA were taken in vitro to visualize its
uorescence, using IVIS imaging system. Supplementary Fig. 5
conrmed that Cy5-siRNA of high siRNA concentration (80 mg/ml)
showed bright yellow uorescence signals and changed to be red
accompanying with decreased siRNA concentration (line I). NTLPD
entrapping Cy5-siRNA exhibited even brighter uorescence signals
than unencapsulated Cy5-siRNA (line II), whereas NTLPD entrapping unlabelled siRNA was found to have no uorescence signals
(line III). Similar results were obtained in TLPD-FCC (data not
shown). These results suggested that the uorescence signals of
Cy5-siRNA was not quenched but even slightly enhanced by
encapsulation in NTLPD and TLPD-FCC.
We adopted IVIS imaging system to visualize the liposomes
entrapping Cy5-siRNA in vivo distribution at different times after
intravenous injections. As shown in Fig. 10A, NTLPD and TLPD-FCC

3466

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

accumulated in tumors within 1 h, and this accumulation gradually

increased, reaching the ultimate at 12 h. At 24 h, there was still
a distinct accumulation in the tumor for TLPD-FCC, whereas little
accumulation was observed for NTLPD. No uorescence signals
were detected in the untreated tumors during the entire length of
the study.
To clearly observe the uorescence signals, the tumors and
major organs were excised and collected at 24 h post injection. As
shown in Fig. 10B, the tumors possessing the strongest signals
among all groups were found in the mice treated with TLPD-FCC,
whereas the hearts, lungs, spleens and livers showed only background to moderate signals. It has to be noted that kidneys from
mice treated with NTLPD and TLPD-FCC also displayed strong
signals, indicating that NTLPD and TLPD-FCC were undergoing
renal excretion at w24 h. Fluorescence signals were hardly detected in all the tissues collected from untreated mice.
3.7. In vivo uptake study
To investigate the binding and internalizing of TLPD-FCC and
NTLPD in vivo, confocal microscopy was used to observe the in vivo
uptake of TLPD-FCC and NTLPD in tumor sections from MDA-MB231 tumors. Fig. 11 shows that TLPD-FCC accumulated profusely
throughout the tumors tissues (including intercellular substance
and cytoplasm) in a pattern consistent with receptor-mediated
endocytosis, reecting by a strong and extensive cytosolic delivery
of Cy5-siRNA. In contrast, NTLPD showed only minimal binding or
uptake, which did not seem to be signicantly different than
untreated controls, indicating that it was less efcient in intracellular delivery.
3.8. In vivo gene silencing
The in vivo activity of TLPD-FCC and NTLPD was assessed by the
luciferase gene silencing in MDA-MB-231 tumor xenograft. After
three consecutive injections, tumors were excised, homogenized
and the luciferase activity was determined. Fig. 12 shows the in vivo
gene silencing activity of siRNA in different formulations. The
luciferase activity was compared with the untreated control.
Remarkably, TLPD-FCC entrapping anti-luciferase siRNA showed
a signicant enhanced in vivo gene silencing activity compared
with NTLPD entrapping NC siRNA, TLPD-FCC entrapping NC siRNA
and NTLPD entrapping anti-luciferase siRNA (P < 0.001, P < 0.05
and P < 0.01 respectively). The results suggested that TLPD-FCC
have a signicant enhanced gene silencing activity in vivo
compared with NTLPD.
4. Discussion

Fig. 9. In vitro gene silencing. The gene silencing assay was taken in (A) MDA-MB-231
cells or (B) MDA-MB-231 cells mixed with SK-BR3 cells or (C) MDA-MB-231 cells mixed
with MCF-7 cells. After 48 h, luciferase activities of the cells were analyzed, normalized
with the protein concentration and expressed as percent luminescence intensity
compared to the untreated control. Data are expressed as mean  SD (n 3e4).

A major limitation to the use of RNAi (RNA interfering) is the

effective delivery of siRNA to the target cells [1,2]. The development
of cationic liposomes is a practical way to deliver siRNA systemically [3]. For example, LPD was shown to deliver siRNA efciently to
tumor cells overexpressing sigma receptor [6,7]. Although the small
molecule anisamide has been implicated in promoting LPD uptake,
it is not clear whether antibodies, which are amongst the most
frequently used ligands for targeted nanoparticles, contribute
directly to promote the development of LPD. Herein, we have
investigated extensively the effect of antibody type, conjugation
strategy and amount on the physicochemical properties and targeting efciency of TLPD. Our optimized immunoliposomes for
siRNA delivery, which we termed TLPD-FCC, demonstrated to
possess a small size around 150 nm, superior serum stability,
specic cell targeting and gene silencing efciency, and thus could

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

3467

Fig. 10. Tissue distribution. Balb/c nude mice bearing MDA-MB-231 tumor xenograft on the right back (white arrows) were injected via tail vein as a single dose with LPD and TLPD
entrapping Cy5-siRNA (1.2 mg siRNA/kg). At different time points (1 h, 6 h, 12 h and 24 h) after injection, the in vivo images were observed with IVIS imaging system (A). Mice were
killed 24 h after injection. Tumors and various organs (heart, liver, spleen, lung and kidney) were collected and imaged with IVIS imaging system (B).

Fig. 11. In vivo uptake study. Tumors were collected from sacriced mice, frozen and sectioned. Sections were counterstained with DAPI and visualized with Leica TCS SP5 Confocal
Microscope. The red uorescence of Cy5 and blue uorescence of DAPI were examined. (A) Bar represents 75 mm, (B) Bar represents 25 mm.

3468

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

Fig. 12. In vivo gene silencing. The luciferase silencing activity of TLPD-FCC and NTLPD
was assessed in MDA-MB-231 tumor xenograft. After three consecutive injections,
tumors were excised and homogenized. Luciferase activities of the tissue lysate were
analyzed, normalized with the protein concentration and expressed as percent luminescence intensity compared to the untreated control. Data are expressed as
mean  SD (n 16e20).

be applied to potentially increase the feasibility of RNAi therapy in

clinical application.
To investigate the effect of antibody conjugation on the physicochemical properties of TLPD, we prepared a series of TLPD
conjugated with anti-EGFR mAb or Fab by conventional conjugation or post-insertion strategy and tested their size and zeta
potential. Fig. 2B shows that the zeta potential of TLPD was
signicantly reduced compared with LPD, indicating that antibody
can signicantly reduce the zeta potential of TLPD. This is not
surprising because antibodies are usually negatively charged,
whereas LPD possesses positive charges. It has to be noted that the
zeta potential of TLPD correlated inversely with the size of TLPD
(Fig. 2). When the zeta potential decreased, the size of TLPD
increased gradually, reaching up to a very large size if the value was
near or below 0 mV, suggesting that TLPD of low zeta potential have
a tendency to aggregate. The results indicated that maintaining
optimal zeta potential (w10 mV) was very critical to the size and
stability of TLPD. Specically, we have shown that mAb and Fab
have different effects on size and zeta potential of TLPD. Compared
with mAb, Fab had a much weaker effect on the size and zeta
potential of TLPD. Similarly, conventional conjugation strategy had
a much weaker effect on the size and zeta potential of TLDP
compared with post-insertion strategy. Herein, we have shown, for
the rst time, the effect of antibody type, conjugation strategy and
amount on the physicochemical properties of TLPD. Our novel
formulation, TLPD conjugated with anti-EGFR Fab by conventional
conjugation, which we termed TLPD-FC, offered the optimum size
(w150 nm) and zeta potential (w10 mV) compared to other
strategies.
We subsequently examined the siRNA EE of TLPD-FC. Conventional siRNA loading method that is mixing siRNA with preformed
PEGylated liposomes (pre-PEGylation) usually leads to low siRNA
EE [25,26]. In contrast, we prepared NTLPD and TLPD-FC by postPEGylation and found that they all had high and similar siRNA EE
(>90%) as naked LPD, suggesting that the post-PEGylation method,
as well as the antibody conjugation, have little adverse impact on
siRNA EE for NTLPD and TLPD-FC (Fig. 3). Using serum stability
assay, we further characterized the in vitro stability of TLPD-FCC.

Fig. 6 showed that naked LPD could protect siRNA from serum
degradation to a great extent, and PEGylation further strengthened
this protection. These results demonstrated that post-PEGylation
method provided a favorable protection to siRNA degradation in
serum, in contrast to the pre-PEGylation method, which usually
leads to the results that most of the siRNA were only bound to the
outer surface of the liposomes, and would immediately release and
be degraded by serum nuclease.
The mechanism of serum stability of naked LPD, NTLPD and
TLPD-FCC was explored by DLS. In DLS experiments, the scattering
light intensity (I) is proportional to concentration (C) and molecular
weight (MW) as I w CMW. This I is sensitive to the changes of MW at
the given conditions. If BSA was absorbed on the liposomes, the MW
of liposome will increase, resulting in the peak position shift or
peak area changes, that is, the size and size distribution will change.
Fig. 7 shows that when mixed with BSA, the peak coupling
appeared in naked LPD solution but not in NTLPD and TLPD-FCC,
indicating the strong interaction exists between naked LPD and
BSA. Thus, comparing with naked LPD, the well separated two
peaks appeared for NTLPD and TLPD-FCC clearly indicated the
strong in vitro stability of our NTLPD and TLPD-FCC, suggesting
PEGylation could reduce protein adsorption on nanoparticles and
improve their stability [27]. The zeta potential of NTLPD was
22.9  2.5 mV, whereas the zeta potential of TLPD-FCC was further
reduced to a value of 10.2  1.6 mV (Fig. 2B). However, as
demonstrated by DLS and serum stability assay, this decreased zeta
potential of TLPD-FCC does not mean its instability. The TEM
further conrmed that there was no difference between NTLPD and
TLPD-FCC in shape and size distribution, suggesting that antibody
conjugation have little impact on the structure of liposomes (Fig. 5).
The result that TLPD-FCC possesses superior in vitro stability is
interesting. Because antibodies are negatively charged hydrophilic
macromolecules, they might break the Hydrophile-Lipophile
Balance (HLB) of the amphipathic molecule DSPE-PEG and induce
the instability like aggregation and content leakage of TLPD-FCC.
However, it seems that antibodies did not induce the instability like
aggregation and content leakage of TLPD-FCC. The superior in vitro
stability of TLPD-FCC might benet from optimal antibody type and
conjugation strategy.
Although SDS-PAGE demonstrated the anti-EGFR Fab was
indeed conjugated to all TLPD-FC, including TLPD-FCA, TLPD-FCB,
TLPD-FCC and TLPD-FCD (Supplementary Fig. 4), they showed
different gene silencing activity in MDA-MB-231 cells. It is interesting to note that TLPD-FCC possessed the best gene silencing
activity among TLPD-FC, while the gene silencing activity
decreased in TLPD-FCD. To conrm that TLPD-FCC possesses
enhanced specic binding afnity in EGFR overexpressing breast
cancer cells, we performed in vitro transfection efciency assay in
three breast cancer cell lines with different EGFR expression levels.
The results showed that TLPD-FCC possessed much higher transfection efciency than NTLPD in breast cancer cells of high (MDAMB-231) or moderate EGFR expression (SK-BR3), but not in cells
with low EGFR expression (MCF-7) (Fig. 8). It has to be noted that
the zeta potential of NTLPD was much higher than that of TLPD-FCC
(22.9  2.5 mV versus 10.2  1.6 mV), and the high zeta potential of
NTLPD would contribute to its strong non-specic electrostatic
interactions. Therefore, it would be expected that the TLPD-FCC
possess much more enhanced specic binding afnity than NTLPD
if the non-specic electrostatic interactions of TLPD-FCC and NTLPD
were the same.
The gene silencing activity of TLPD-FCC highly correlated with
the transfection efciency (Fig. 9). When anti-luciferase siRNA was
formulated into TLPD-FCC, the gene silencing activity was signicantly improved than NTLPD due to its enhanced transfection
efciency (Fig. 8). However, the NC siRNA only showed weak or

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

moderate activity, which was probably due to an off-target effect.

Notably, enhanced gene silencing activity for TLPD-FCC was not
only found in MDA-MB-231 cells, but also in MDA-MB-231 cells
mixed with SK-BR3 cells or MDA-MB-231 cells mixed with MCF-7
cells. This result is fairly important because this strategy of MDAMB-231 cells mixing with various cells with different EGFR
expression levels mimics the in vivo cellular environment to
a higher degree. These results rmly conrmed that the gene
silencing activity of TLPD-FCC was highly sequence dependent and
TLPD-FCC showed signicantly enhanced gene silencing activity
compared to NTLPD.
In vivo distribution assay showed that NTLPD and TLPD-FCC had
different distribution patterns (Fig. 9). Although both of NTLPD and
TLPD-FCC accumulated in tumors within 1 h, reaching the ultimate
at 12 h, little accumulation was observed for NTLPD after 24 h. In
contrast, there was still a distinct accumulation in the tumor for
TLPD-FCC after 24 h. Thus, TLPD-FCC showed superior EGFR overexpressing tumor targeting ability in vivo compared to NTLPD. No
obvious siRNA uptake was observed in the liver and spleen for
NTLPD and TLPD-FCC (Fig. 9B). The reduced uptake of NTLPD and
TLPD-FCC in the liver and spleen was due to the PEG on the NTLPD
and TLPD-FCC surface arranged in the brush mode, which prevented serum opsonization and abolished the non-specic RES
(reticular endothelial system) uptake [21]. As a result, the fully
protected NTLPD and TLPD-FC of low RES uptake might have an
enhanced trend to reach the tumor via the enhanced permeability
and retention (EPR) effect. Furthermore, in vivo uptake study
showed that TLPD-FCC accumulated profusely throughout the
tumors tissues in a pattern consistent with receptor-mediated
endocytosis. However, NTLPD showed only minimal binding or
uptake. These data are consistent with the mechanism for long
circulating immunoliposomes delivery involving two phases
[18,28]. In the rst or tissue phase, long circulating TLPD-FCC and
NTLPD slowly accumulate in tumor tissue, ultimately reaching high
tumor levels due to EPR effect. In the second or cellular phase,
NTLPD remains in the interstitial or extracellular space and
undergoes degradation, non-specic endocytosis and eventual
drug release. In contrast, TLPD-FCC binds to and internalize in
tumor cells via ligandereceptor interactions. At present, there is
still considerable debate about the in vivo distribution of active
versus passive targeting in nanoparticle delivery. Some groups
reported that the use of tumor targeting ligands did not inuence
the total nanoparticle accumulation in tumor, but increased
receptor mediated internalization [29,30]. In contrast, several
groups showed that active ligand targeting increased both intracellular nanoparticle uptake and total tumor accumulation (our
data support this viewpoint) [31,32]. This discrepancy is likely due
to the differences in the tumor models, detecting time points,
nanoparticle types and targeting ligands types.
Consistent with the in vitro gene silencing assay, TLPD-FCC
showed signicantly higher gene silencing activity in vivo than
other formulations (Fig. 12). The enhanced activity of TLPD-FCC
was mainly due to the signicantly improved tumor uptake as
shown in Fig. 11. Other formulations, including NTLPD entrapping
NC siRNA, TLPD-FCC entrapping NC siRNA and NTLPD entrapping
anti-luciferase siRNA, had little gene silencing activity. It is notable
that the in vivo gene silencing of TLPD-FCC (about w21% gene
silencing) was not very sufcient, which was probably due to
MDA-MB-231 tumor is subcutaneous inoculated and not euangiotic. It would be expected that TLPD-FCC would achieve better
gene silencing activity in euangiotic orthotopic xenografts of lung
or liver tumor. However, the results still conrmed that TLPD-FCC
showed signicant gene silencing activity and this activity was
highly dependent on anti-EGFR Fab targeting and sufcient tumor
delivery.

3469

5. Conclusion
In this study, we investigated extensively the effect of antibody
type, conjugation strategy and amount on the physicochemical
properties and targeting efciency of TLPD. As a result, we obtained
TLPD-FCC which possessed a small size around 150 nm, high siRNA
EE, superior serum stability and very weak interaction with BSA.
Compared with NTLPD, TLPD-FCC showed a signicantly enhanced
EGFR targeting efciency both in vitro and in vivo. Furthermore,
TLPD-FCC exhibited superior gene silencing activity in EGFR overexpressing tumors.
Acknowledgements
This work was nancially supported by National Natural Science
Foundation of China, Shanghai Commission of Science & Technology, Ministry of Science and Technology of China (973 & 863
program projects), Pudong Commission of Science and Technology
of Shanghai and a special grant from Ministry of Education of China
(Key Laboratory) and Shanghai Commission of Education and
National Special Projects for New Drug Development and Infectious
Diseases. We thank Ms. Yang Yang, Ms. Jing Xu for their technical
assistance.
Appendix
Figure with essential color discrimination. Figs. 1, 10 and 11 in
this article is difcult to interpret in black and white. The full color
images can be found in the on-line version, at doi:10.1016/j.
biomaterials.2011.01.034.
Appendix. Supplementary data
Supplementary data associated with this article can be found in
the on-line version, at doi:10.1016/j.biomaterials.2011.01.034.
References
[1] de Fougerolles A, Vornlocher HP, Maraganore J, Lieberman J. Interfering with
disease: a progress report on siRNA-based therapeutics. Nat Rev Drug Discov
2007;6:443e53.
[2] Baker M. RNA interference: homing in on delivery. Nature 2010;22(464):
1225e8.
[3] Morille M, Passirani C, Vonarbourg A, Clavreul A, Benoit JP. Progress in
developing cationic vectors for non-viral systemic gene therapy against
cancer. Biomaterials 2008;29:3477e96.
[4] Oh YK, Park TG. siRNA delivery systems for cancer treatment. Adv Drug Deliv
Rev 2009;61:850e62.
[5] Tseng YC, Mozumdar S, Huang L. Lipid-based systemic delivery of siRNA. Adv
Drug Deliv Rev 2009;61:721e31.
[6] Li SD, Chono S, Huang L. Efcient gene silencing in metastatic tumor by siRNA
formulated in surface-modied nanoparticles. J Control Release 2008;126:
77e84.
[7] Li SD, Chen YC, Hackett MJ, Huang L. Tumor-targeted delivery of siRNA by selfassembled nanoparticles. Mol Ther 2008;16:163e9.
[8] Gao J, Feng SS, Guo Y. Antibody engineering promotes nanomedicine for
cancer treatment. Nanomedicine (Lond) 2010;5:1141e5.
[9] Liu Y, Li K, Liu B, Feng SS. A strategy for precision engineering of nanoparticles
of biodegradable copolymers for quantitative control of targeted drug
delivery. Biomaterials 2010;31:9145e55.
[10] Mamot C, Drummond DC, Greiser U, Hong K, Kirpotin DB, Marks JD, et al.
Epidermal growth factor receptor (EGFR)-targeted immunoliposomes
mediate specic and efcient drug delivery to EGFR- and EGFRvIII-overexpressing tumor cells. Cancer Res 2003;63:3154e61.
[11] Kim IY, Kang YS, Lee DS, Park HJ, Choi EK, Oh YK, et al. Antitumor activity of
EGFR targeted pH-sensitive immunoliposomes encapsulating gemcitabine in
A549 xenograft nude mice. J Control Release 2009;140:55e60.
[12] Chen Y, Zhu X, Zhang X, Liu B, Huang L. Nanoparticles modied with tumortargeting scFv deliver siRNA and miRNA for cancer therapy. Mol Ther
2010;18:1650e6.
[13] Pirollo KF, Rait A, Zhou Q, Hwang SH, Dagata JA, Zon G, et al. Materializing the
potential of small interfering RNA via a tumor-targeting nanodelivery system.
Cancer Res 2007;67:938e43.

3470

J. Gao et al. / Biomaterials 32 (2011) 3459e3470

[14] Peer D, Park EJ, Morishita Y, Carman CV, Shimaoka M. Systemic leukocytedirected siRNA delivery revealing cyclin D1 as an anti-inammatory target.
Science 2008;319:627e30.
[15] Zheng X, Vladau C, Zhang X, Suzuki M, Ichim TE, Zhang ZX, et al. A novel in
vivo siRNA delivery system specically targeting dendritic cells and silencing
CD40 genes for immunomodulation. Blood 2009;113:2646e54.
[16] Buse J, El-Aneed A. Properties, engineering and applications of lipid-based
nanoparticle drug-delivery systems: current research and advances. Nanomedicine (Lond) 2010;5:1237e60.
[17] Peer D, Karp JM, Hong S, Farokhzad OC, Margalit R, Langer R. Nanocarriers as
an emerging platform for cancer therapy. Nat Nanotechnol 2007;2:751e60.
[18] Gao J, Kou G, Wang H, Chen H, Li B, Lu Y, et al. PE38KDEL-loaded anti-HER2
nanoparticles inhibit breast tumor progression with reduced toxicity and
immunogenicity. Breast Cancer Res Treat 2009;115:29e41.
[19] Gao J, Sun J, Li H, Liu W, Zhang Y, Li B, et al. Lyophilized HER2-specic
PEGylated immunoliposomes for active siRNA gene silencing. Biomaterials
2010;31:2655e64.
[20] Mao HQ, Roy K, Troung-Le VL, Janes KA, Lin KY, Wang Y, et al. Chitosan-DNA
nanoparticles as gene carriers: synthesis, characterization and transfection
efciency. J Control Release 2001;70:399e421.
[21] Li SD, Huang L. Nanoparticles evading the reticuloendothelial system: role of
the supported bilayer. Biochim Biophys Acta 2009;1788:2259e66.
[22] Lee J, Cho EC, Cho K. Incorporation and release behavior of hydrophobic drug
in functionalized poly(d, l-lactide)-block-poly(ethylene oxide) micelles.
J Control Release 2004;94:323e35.
[23] Schiffelers RM, Ansari A, Xu J, Zhou Q, Tang Q, Storm G, et al. Cancer siRNA
therapy by tumor selective delivery with ligand-targeted sterically stabilized
nanoparticle. Nucleic Acids Res 2004;32:e149.

[24] Gao J, Kou G, Chen H, Wang H, Li B, Lu Y, et al. Treatment of hepatocellular

carcinoma in mice with PE38KDEL type I mutant-loaded poly(lactic-co-glycolic acid) nanoparticles conjugated with humanized SM5-1 F(ab) fragments.
Mol Cancer Ther 2008;7:3399e407.
[25] Buyens K, Demeester J, De Smedt SS, Sanders NN. Elucidating the encapsulation of short interfering RNA in PEGylated cationic liposomes. Langmuir
2009;25:4886e91.
[26] Buyens K, Lucas B, Raemdonck K, Braeckmans K, Vercammen J, Hendrix J, et al.
A fast and sensitive method for measuring the integrity of siRNA-carrier
complexes in full human serum. J Control Release 2008;126:67e76.
[27] Aggarwal P, Hall JB, McLeland CB, Dobrovolskaia MA, McNeil SE. Nanoparticle
interaction with plasma proteins as it relates to particle biodistribution,
biocompatibility and therapeutic efcacy. Adv Drug Deliv Rev 2009;61:428e37.
[28] Mamot C, Drummond DC, Noble CO, Kallab V, Guo Z, Hong K, et al. Epidermal
growth factor receptor-targeted immunoliposomes signicantly enhance the
efcacy of multiple anticancer drugs in vivo. Cancer Res 2005;65:11631e8.
[29] Bartlett DW, Su H, Hildebrandt IJ, Weber WA, Davis ME. Impact of tumorspecic targeting on the biodistribution and efcacy of siRNA nanoparticles
measured by multimodality in vivo imaging. Proc Natl Acad Sci 2007;104:
15549e54.
[30] Kirpotin DB, Drummond DC, Shao Y, Shalaby MR, Hong K, Nielsen UB, et al.
Antibody targeting of long-circulating lipidic nanoparticles does not increase
tumor localization but does increase internalization in animal models. Cancer
Res 2006;66:6732e40.
[31] Dass CR, Choong PF. Selective gene delivery for cancer therapy using cationic
liposomes: in vivo proof of applicability. J Control Release 2006;113:155e63.
[32] Hilgenbrink AR, Low PS. Folate receptor-mediated drug targeting: from
therapeutics to diagnostics. J Pharm Sci 2005;94:2135e46.