www.MaterialsViews.

com

Cancer Treatment

Combined Near Infrared Photothermal Therapy and

Chemotherapy Using Gold Nanoshells Coated Liposomes
to Enhance Antitumor Effect
Liyao Luo, Yanhong Bian, Yanping Liu, Xuwu Zhang, Meili Wang, Shanshan Xing,
Lei Li, and Dawei Gao*

Novel antitumor system based on the targeting photothermal and pH-responsive

nanocarriers, gold nanoshells coated oleanolic acid liposomes mediating by chitosan
(GNOLs), is designed and synthesized for the first time. The GNOLs present
spherical and uniform size (172.03 nm) with zeta potential (20.7 ± 0.4 mV), which
are more easily accumulated in tumor. Meanwhile, the GNOLs exhibit a slow and
controlled release of oleanolic acid at pH 7.4, as well as a rapid release at pH 5.5,
which is beneficial for tumor-targeting drug release. Under near infrared (NIR)
irradiation, hyperthermia can be generated by activated gold nanoshells to perform
photothermal therapy effect, which triggers drug release from the carriers by
activating the gel to liquid crystalline phase transition of the liposomes. Moreover, the
NIR assisting drug release can be easily and selectively activated locally due to the
spatially and real-timely controllable property of light. The experimental results also
verify that the GNOLs with NIR irradiation achieve more ideal antitumor effects
than other oleanolic acid formulations in vitro and in vivo. Hence, the drug delivery
system exhibits a great potential in chemo-photothermal antitumor therapy.

1. Introduction organic.[2,3] However, the potential application of conven-

Multifunctional liposomes have provoked an upsurge of
interest, mainly due to their similarity to cell membranes and
remarkable ability of loading and transporting drug.[1] They
are frequently used to improve therapeutic effects of various
water soluble/insoluble drugs by enhancing bioavailability,
solubility, retention time, or reducing toxicity to normal
L. Luo, Y. Bian, Dr. Y. Liu, Dr. X. Zhang, M. Wang,
S. Xing, L. Li, Prof. D. Gao
Chemical Key Lab of Hebei Province
Department of Biological Engineering
Yanshan University
No. 438 Hebei Street, Qinhuangdao 066004, China
E-mail: dwgao@ysu.edu.cn
Prof. D. Gao
State Key Laboratory of Metastable Materials Science and Technology
Yanshan University
Qinhuangdao 066004, P. R. China
DOI: 10.1002/smll.201503961
tional liposomes is limited due to their inherent physical and
chemical instability, poor targeting, uncontrolled release of
the encapsulated drugs, short circulation lifetime in vivo, and
easily captured by mononuclear phagocytic system.[4]
To implement targeting treatment and controlling release
of drugs in liposomes, various types of stimuli-responsive
materials have been studied, along with external stimuli
such as light, magnetic field, ultrasound, and electric field.[5–7]
Phototherapy, as an emerging treatment technology, can
strongly absorb near infrared (NIR) light and convert it into
cytotoxic heat to destroy hyperthermia sensitive tumor cells
with minimal effects on the surrounding normal cells, which
provides an alternative way for tumor therapy.[8] Gold nano-
materials have been paid much attention due to their excel-
lent photothermal properties.[7] It was reported that different
gold nanostructures with tunable surface plasmon resonance
(SPR) had various light-absorbing ranges.[9] The SPR wave-
length of the gold nanoparticles (AuNPs) is 520 nm in the
visible region.[10] When the gold nanoshells are formed, the

small 2016, © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.small-journal.com 1
DOI: 10.1002/smll.201503961
 full papers
SPR wavelength can redshift to NIR region (650–900 nm).[11]
As NIR light is relatively transparent for biological tissues, it
can penetrate the tissues with little energy lost and reach the
diseased region.[12] Subsequently, with high optical absorp-
tion in the NIR region, gold nanoshells’ material can convert
NIR light energy to heat energy with high efficiency to induce
localized hyperthermia in irradiation tissue.[13,14] Due to the
heat tolerance of tumor cells being lower than normal tissue,
the temperature in the region 42–47 °C can ablate tumor
cells.[15] So the tumor cells are selectively destroyed by loos-
ening cell membranes and denaturing proteins with the local-
ized hyperthermia.[16,17] Meanwhile, the heat activates gel
phase to liquid crystalline phase transition for bio-membrane,
which leads the encapsulated drugs release to kill the tumor
cells.[18] Gold nanoshells-coated liposomes are sensitive to
NIR irradiation, which can induce cytotoxic hyperthermia.
Oleanolic acid (OA, Figure 1), a naturally pentacyclic trit-
erpenoid compound widely distributed in many traditional
Chinese medicines, has varieties of biological actions, such
as antioxidant, antiinflammatory, antifungal properties, and
especially antitumor effect.[19] However, the poor aqueous
solubility and low dissolution rate of OA in the gastrointes-
tinal tract greatly limit its effective absorption and bioavail-
ability in body. In our previous study, the chitosan-coated OA
liposomes (CS-OA-Lips) were prepared and achieved signifi-
cant antitumor effect through pH-responsive target action.[20]
The chitosan can form a hydrophilic shell on the liposome
surface to increase the physical stability of liposomes and
provide steric protection to escape adsorption of opsonins
and phagocytosis of macrophages. Meanwhile, the chitosan
also possesses amino groups in the molecules, which is
capable of combining with protons and make chitosan car-
rying the positive charges.[21] In our study, the CS-OA-Lips
loaded more positive charges, it was reported that the tumor
cells loaded abundant negative charges, so the CS-OA-Lips
could combine with the tumor cells easily.[20,22] In addition,
since the tumors have leaky blood vessels, liposomes with
diameters less than 200 nm can escape into tumor tissues
and accumulate by the enhanced permeability and retention
(EPR) effect.[23,24]
The novel concept advanced here, of using gold
nanoshells-coated CS-OA-Lips for targeted delivery of
Figure 1.  Structure of OA.
2 www.small-journal.com © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
OA, took advantage of the mechanism of chitosan and gold
nanoshells for controlling drug release and photothermal
therapy. The carriers not only had pH-sensitivity to achieve
drug-controlled release in tumor tissue, but also could con-
vert the absorbed NIR light into heat to trigger OA release
in tumor tissues. Through pH-response and active targeting
effect, the gold nanoshells coated oleanolic acid liposomes
(GNOLs) could selectively target the tumor site and be taken
up by tumor cells. Under NIR irradiation, hyperthermia
would be generated by the activated gold nanoshells to per-
form photothermal therapy effect, thus triggering release
of OA from the carriers by activating the gel to liquid crys-
talline phase transition of the liposomes. The drug delivery
system may open up a new avenue for future tumor therapy
design.
2. Results and Discussion
2.1. Preparation of GNOLs
In the study, a reproducible and facile strategy to generate
uniform GNOLs was presented. Briefly, the CS-OA-Lips were
prepared by the method that was previously reported.[20] In
this process, the amount of chitosan could affect the coating
of OA liposomes (OA-Lips). The orthogonal experiment was
done, and confirmed that the optimal concentration of chi-
tosan was 0.1% and the volume ratio of CS to OA liposome
was 1:1 on preparation of the CS-OA-Lips, and the results
were shown in Table S1 and Table S2 (Supporting Informa-
tion). Subsequently, the GNOLs were synthesized by the
seeds growth method.[25] The small-sized gold seeds were
generated by NaBH4 reduction and then incubated with CS-
OA-Lips to obtain gold nanoparticles-coated CS-OA-Lips
(AuNPs-CS-OA-Lips). Fratoddi et al. found that complexes
with thiol groups or amine groups could be easily combined
with gold nanoparticles taking advantage of the formation of
Au S or Au N bonds.[26] Due to amine groups in the mole-
cules, chitosan played a crucial role for forming nanoshells by
connecting gold nanoparticles with the liposomes. The gold
seeds were attached selectively on the surface of CS-OA-Lips
by forming Au N bond. After that, a growth solution con-
taining AuCl3 and NaBH4 was added to the AuNPs-CS-OA-
Lips solution. Li et al. and Topete et al. researches found that
the seeds growth would undergo two steps. In the first stage,
small gold seeds are prepared. In the second step, the seeds
are added to a growth solution containing Au precursor and
reducing agents, and the newly reduced Au0 then grows on
the seed surface to form large AuNPs. In this process, the
newly reduced Au0 can only assemble on the surface of the
gold seeds to grow into gold nanoshells, and no nucleation of
new particles occurs in solution.[27,28] In our study, the gold
seeds were formed on the surface of CS-OA-Lips at first, and
then the AuCl3 and reducing agents were added into the gold
seeds solution to grow into nanoshells. The obtained GNOLs
integrated NIR light and pH dual-stimuli responsive proper-
ties, chemotherapy, photothermal effect, and tumor-targeting
into one system. Upon NIR irradiation, the GNOLs could
efficiently convert light energy into heat with temperature
small 2016,
DOI: 10.1002/smll.201503961
www.MaterialsViews.com

Scheme 1. Schematic illustration of the GNOLs formation and the antitumor effect under NIR irradiation.

increased, which induced the instability of liposome mem-
brane and activated drug release simultaneously. A typical
schematic of synthesis and photothermal therapy of GNOLs
was illuminated in Scheme 1.
2.2. Characterization of GNOLs
The encapsulation efficiency was determined by gel chroma-
tography separation and high performance liquid chromatog-
raphy (HPLC). The encapsulation efficiency of GNOLs was
93.4% and almost all of OA was encapsulated into the lipo-
some vesicles. At the same time, the encapsulation efficien-
cies of OA were 94.3% and 94.7% for conventional OA-Lips
and CS-OA-Lips, respectively. There were no statistical dif-
ferences among the three OA vehicles in encapsulation effi-
ciency. The transmission electron microscopy (TEM) images
in Figure 2 revealed vividly the process of forming GNOLs.
Figure 2a showed the morphologies of AuNPs-CS-OA-Lips,
and it could be seen that AuNPs were already coated on
the surface of CS-OA-Lips. At the second reduction phase,
the gold seeds on the surface of AuNPs-CS-OA-Lips con-
nected with a lot of AuNPs to form gold nanoshells gradually
(Figure 2b). The GNOLs presented spherical shape, uniform
size, and excellent dispersability. Besides, the gold nanoshell-
coated OA liposomes without chitosan modification were
designed and prepared, but the gold nanoparticles could not
combine with the liposomes in the absence of chitosan. The
TEM image was shown in Figure S1 (Supporting Informa-
tion). The results verified that chitosan played a crucial role
for forming nanoshells by connecting gold nanoparticles with
the liposomes.
The GNOLs had a positive zeta potential (20.7 ± 0.4 mV)
that was similar with the CS-OA-Lips (19 ± 0.814 mV). It indi-
cated that the gold nanoshells didn’t change the property of
positive charges of CS-OA-Lips, which was beneficial to com-
bine with the negative charges on the surface of the tumor
cells.[22] Therefore, the GNOLs were liable to accumulate in
the tumor tissue, and the tumor-targeting accumulation of
GNOLs could decrease the dosage of drugs and reduce the
damage to normal tissue at the same time.[29] The particle
size was detected by dynamic light scattering (DLS), and
the size distribution of GNOLs was shown in Figure 3. The
Figure 2. a) TEM of AuNPs-CS-OA-Lips; b) TEM of GNOLs. The inset
showed a magnified TEM image.

small 2016, © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.small-journal.com 3
DOI: 10.1002/smll.201503961
 full papers
Figure 3.  Size distribution of the GNOLs.
average diameter of the liposomes was about 172.03 nm and
the particle size was relatively uniform without contact and
fusion phenomena, which was attributed to the repulsion of
surface positive charges of the carriers. The size of liposomes
might be an important factor for biomedical applications,
as the tumors had leaky blood vessels, and the GNOLs pre-
pared with mean size of 150–200 nm could exactly escape
into tumor tissues and accumulate by EPR effect.[22] On the
contrary, owing to the relatively large sizes, the GNOLs could
not go through the tight junctions between endothelial cells
on normal vascular linings.[29] In addition, the particles with
the diameter less than 200 nm could circulate in the blood
stream for a long time since they were hardly captured by the
reticuloendothelial system.[30] Hence, the GNOLs were more
conducive to tumor therapy.
The elements of GNOLs were determined by energy dis-
persive X-ray spectroscopy (EDS) on the TEM as shown in
Figure 4. Cu and part of C belonged to carbon-coated grid.
The remaining part of C, P, and O elements derived from
compounds of liposomes. Most importantly, the existence of
Au element indicated that the gold nanoparticles successfully
coated on the surface of CS-OA-Lips.
Due to the plasmon resonance of gold nanoparticles, the
AuNPs-CS-OA-Lips exhibited a strong absorbance at 520 nm
Figure 4.  EDS of the GNOLs.
4 www.small-journal.com © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
Figure 5.  UV–vis spectra of CS-OA-Lips, AuNPs-CS-OA-Lips, and GNOLs.
in comparison with the CS-OA-Lips (Figure 5). When the
gold nanoparticles aggregated to form gold nanoshells, the
surface plasmon absorption peak was redshifted to NIR
region.[31] As shown in Figure 5, the GNOLs exhibited a
broad absorption band between 600 and 850 nm. Therefore,
the GNOLs could rapidly convert the light absorbed at the
NIR wavelengths into thermal energy to ablate the sur-
rounding cells. Meanwhile, the thermal responsive liposomes
were composed of phospholipids, which were also activated
from gel to liquid crystalline phase transition by NIR light.[32]
So the porous membranes would release drugs encapsulated
in the liposomes.
The Fourier Transform Infrared Spectroscopy (FTIR)
spectra of pure chitosan, OA-Lips, CS-OA-Lips, and
GNOLs, were determined to investigate the interaction
among liposomes, chitosan, and gold nanoshells (Figure 6).
As shown in Figure 6A, the peak of 3440 cm−1 was attrib-
uted to axial stretching vibration of O H superimposed to
the N H symmetrical stretching vibration and the inter-
hydrogen bonds of the polysaccharide. Moreover, the peaks
at 2924, 1631, and 1049 cm−1 were attributing to C H, N H,
and C O stretching vibration, respectively. In comparison to
FTIR spectrum of OA liposomes (B) and pure chitosan (A),
the peaks at 3388 and 1635 cm−1 could be observed for the
Figure 6. FTIR of A) pure chitosan, B) OA-Lips, C) CS-OA-Lips,
and D) GNOLs.
small 2016,
DOI: 10.1002/smll.201503961
www.MaterialsViews.com
Figure 7. Encapsulation efficiencies of CS-OA-Lips and GNOLs in
different storage conditions for 30 d. All values were expressed as mean
± standard deviation (SD) (n = 3).
CS-OA-Lips (C) arising from N H band, which indicated
that chitosan had been modified on the OA liposomes. Fur-
thermore, the same characteristic absorption peaks of the
chitosan appeared in the spectrum of GNOLs. Compared
the spectra of GNOLs and CS-OA-Lips, there were some
changes on the peaks (3000–3700 cm−1) due to the interaction
of gold nanoparticles with N H stretching band.
2.3. Stability Study
The influences of temperature and time were explored on
the stability of drug delivery system, and the drug encap-
sulation efficiencies were analyzed (Figure 7). And the sta-
bility of CS-OA-Lips was shown in the previous study.[20] In
this study, the GNOLs were stored at 4, 25, and 37 °C for a
period of 0, 5, 10, 15, 20, 25, and 30 d. It was observed that the
encapsulation efficiency of GNOLs at 4 °C was significantly
higher than those at 25 and 37 °C on the same time point. For
example, the encapsulation efficiencies of the GNOLs were
41.38 ± 1.59%, 68.28 ± 2.21%, and 77.52 ± 1.23% stored at 37,
25, and 4 °C for 30 d, respectively. The results directly vali-
dated that lower temperature (4 °C) was more conductive to
preserving the liposomal integrity because high temperature
Figure 8. a) Drug release profiles of CS-OA-Lips in vitro under different pH values. b) Drug release profiles of GNOLs in vitro under different pH
values. All values were expressed as mean ± SD (n = 3).
small 2016, © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/smll.201503961
could accelerate the phase transitions of the lipid mem-
branes.[18] The similar results were obtained on CS-OA-Lips
with 37.14 ± 1.37%, 67.31 ± 1.58%, and 76.97 ± 1.18% stored
at 37, 25, and 4 °C for 30 d, respectively. Obviously, the encap-
sulation efficiencies of GNOLs were higher than those of CS-
OA-Lips at any temperature. This main reason was that the
pyknotic gold nanoshells on the surface of chitosan-modified
liposomes hindered the drug release from the liposomes and
enhanced the stability of the carrier.
2.4. pH-Responsive Drug Releases
In the previous study, the drug release effect of CS-OA-Lips
was determined using a dynamic dialysis method at different
pH conditions (Figure 8a).[20] The result showed that the CS-
OA-Lips presented two drug release stages, which included
a relatively large burst effect in the initial stage and subse-
quently a slow release phase. Compared with pH 7.4 condi-
tions, the drug release rate of CS-OA-Lips was significantly
enhanced under acidic conditions (pH 5.5). It proved that the
CS-OA-Lips had pH-sensitivity to achieve drug-controlled
release.
In this study, the same method was used to determine the
drug release effects of GNOLs. In addition, the influence of
NIR light irradiation on drug release was studied, as shown
in Figure 8b. The GNOLs presented a similar phenomenon
like CS-OA-Lips without NIR light. The burst release was
mainly ascribed to drug detachment from the outer surface
of the liposomes, while the later slow release was due to
the fact that the diffusion of the drug from the lipid bilayer
entered the release media.[33] Therefore, the GNOLs were
beneficial to the delay of the drug release from the capsules.
We also examined the drug release performance of the mix-
ture containing AuNPs and OA liposomes under different
pH conditions. The results indicated that the mixture had no
pH-responsive ability on the drug release (Figure S2, Sup-
porting Information).
In the absence of NIR light irradiation, the drug release
rates of GNOLs were lower than those of CS-OA-Lips under
the conditions of pH 7.4 or pH 5.5. In the medium of pH
7.4, the release rates of CS-OA-Lips and GNOLs reached
50 ± 4% and 42 ± 1% 12 h later, respectively. The main
www.small-journal.com 5
 full papers
www.MaterialsViews.com

reason was that the gold nanoshells could prevent the drug
diffusion from liposomes, and confine them well inside the
nanocarriers. However, the GNOLs were pH-sensitive, and
their drug release rate was significantly enhanced in acidic
condition. For example, after 8 h, 53 ± 1% of encapsulated
OA had been released at pH 5.5, while only 39 ± 2% of the
drug had been released under pH 7.4. The GNOLs exhib-
ited a slow, controlled release of OA under normal physi-
ological conditions and a rapid release in acidic environment.
Therefore, the GNOLs had the potential of controlling drug
targeting release effectively in tumor sites, in view of the neg-
ative charges property of solid tumors.[34]
After 8 h, the samples of NIR-treated groups were being
exposed to laser irradiation with a wavelength of 808 nm
(1 W cm−2, 4 min). The results showed that the release rates
of OA upon NIR light activation were increased whether at
Figure 9. Antitumor effects of the OA formulations with or without
pH 5.5 or pH 7.4. After 24 h, the drug release rate of the NIR NIR laser. Notes: Free OA (−) was OA solution without laser; Free OA
group reached 92 ± 1%, while the non-NIR group was only (+) was OA solution with laser; CS-OA-Lips (−) were chitosan-coated
69 ± 1%. The main reason was that the GNOLs could absorb OA liposomes without laser; CS-OA-Lips (+) were chitosan-coated
NIR light and convert the absorbed energy into heat to cause OA liposomes with laser; GNOLs (−) were gold nanoshells-coated OA
phase conversion of liposome membranes, thereby providing liposomes without laser; GNOLs (+) were gold nanoshells-coated
on-demand release of encapsulated drugs.[35] In view of the OA liposomes with laser. ** indicates P <0.01 compared with free OA
solution. All values were expressed as mean ± SD (n = 3).
spatially and temporally controllable property of light, the
NIR-induced drug release could be easily and selectively
activated locally and achieving on-demand drug release. could also destroy the tumor cells because of the low toler-
The gold nanoshells and chitosan combined on-demand ance of tumor tissues, along with minimal effects on the sur-
drug release system could eliminate the adverse side effects rounding normal cells.[18]
caused by drug leakage or non-specific drug release, there- Compared with all containing OA liposomes, the blank
fore reducing side effects in healthy tissues. liposomes were also investigated to evaluate the photo-
thermal effect. As shown in Figure 10, the gold nanoshells-
coated blank liposomes (AuNS-CS-Lips) with NIR
2.5. Antitumor Effect of GNOLs In Vitro irradiation showed the highest inhibition rates (49.78 ±
1.56%), which stated that NIR laser irradiation through gold
In this study, the antitumor effects of GNOLs with or without nanoshells would cause a local increase in temperature to
NIR irradiation were investigated on 143B cells. As shown in realize the antitumor effect. The other liposomes also had
Figure 9, a concentration-dependent cell-killing effect could inhibitory effect on the tumor growth to some extent, which
be observed on all of the OA formulations against 143B cells.
It was also obvious that the free OA had the minimum tumor
inhibition rate (42.81 ± 2.09% at 100 μg mL−1). The 143B cells
treated with CS-OA-Lips and GNOLs exhibited tumor inhi-
bition rates of 71.52 ± 1.21% and 73.74 ± 1.32%, respectively,
which indicated the antitumor effect of OA was improved
by encapsulating in liposomes. Furthermore, the antitumor
effects were significantly different between NIR irradia-
tion and non-irradiation. The inhibition rates of CS-OA-
Lips with and without laser were 73.11 ± 2.46% and 71.51 ±
1.21%, respectively (100 μg mL−1 OA), and it could be seen
the laser was ineffective on CS-OA-Lips. On the contrary,
the results of GNOLs were obviously different, their inhibi-
tion rates were 73.74 ± 1.32% at non-irradiation, and 86.91 ±
2.53% at NIR irradiation, respectively. The phenomena could
be due to the several reasons. First, the laser-caused hyper-
thermia increased permeability and fluidity of the liposomes
membrane, therefore promoting drug release and enhancing Figure 10. Antitumor effects of several kinds of blank liposomes.
drug accumulation inside the tumor tissue.[36,37] Meanwhile, Notes: Lips mean blank liposomes; CS-Lips were chitosan-coated
blank liposomes; AuNS-CS-Lips were gold nanoshells-coated blank
NIR-induced hyperthermia promoted phase conversion from
liposomes without laser; AuNS-CS-Lips (+) were gold nanoshells-
gel-to-liquid crystalline of cells membrane, and dramati- coated blank liposomes with laser. * indicates P <0.05, ** indicates
cally enhanced the uptake of OA by 143B cells leading to P <0.001 compared with blank liposomes. All values were expressed as
the tumor cells apoptosis.[38] Furthermore, the hyperthermia mean ± SD (n = 3).

6 www.small-journal.com © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim small 2016,
DOI: 10.1002/smll.201503961
www.MaterialsViews.com
Figure 11. Fluorescence images of 143B tumor cells treated with
samples incubating for 24 h later. a) Control group, b) GNOLs (without
laser), c) PBS solution (with laser), d) AuNS-CS-Lips (with laser),
and e) GNOLs (with laser).
might be due to the composition of liposomes and antitumor
effect of chitosan, but the effect was low.[39]
To analyze the photothermal effect of GNOLs with NIR
irradiation against tumor cells, the 143B cells treated with
drug were dyed by fluorescein diacetate (FDA). The FDA
could penetrate cytomembrane and accumulate in the living
cells, which exhibited green fluorescence under fluorescent
microscope. On the contrary, the dead cells could not be
dyed by FDA and were left without color.[40,41] As shown in
Figure 11a,c, no difference was observed between the groups
of phosphate buffers (PBS) solution with or without laser, so
only NIR irradiation could not affect the state of cells. For
the group of GNOLs without laser irradiation, there were
slightly less living cells compared to PBS group. In the AuNS-
CS-Lips group (Figure 11d), the area with laser irradiation
(Left) showed a small number of living cells, but more living
cells could be observed at the area of non-NIR irradiation
(Right). This phenomenon disclosed there was inhibitory
Figure 12. a) Tumor growth curves of tumor-bearing mice treated with different formulations. b) Body weight changes of the mice after treatment.
c) Images of typical tumors at the end of the experiment. ** indicates P <0.01 compared with OA group. All values were expressed as mean ±
SD (n = 6).
small 2016, © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/smll.201503961
effect on 143B cells for AuNS-CS-Lips plus NIR irradiation
to some extent. It was notable that the group of GNOLs
with laser irradiation had the least living cells (Figure 11e) at
the irradiated region. Therefore, the combination of photo-
thermal therapy and chemotherapy was more efficient to
induce cells’ death compared with solo treatment.
2.6. Antitumor Effect In Vivo
To evaluate the therapeutic effect of GNOLs on the com-
bined photothermal ablation and chemotherapy for solid
tumors, the in vivo antitumor examination was carried out
using U14 tumor-bearing mice as animal model. Figure 12
showed the tumor volumes and weights of mice that received
different treatments.
It was notable that the saline and laser groups showed
an uninhibited tumor growth with no statistical differences
(Figure 12a), therefore, provided the solo laser irradiation
could not inhibit the tumor growth efficiently. The tumor size
of free OA group was about 50% smaller than that of the
saline control group, so it performed an antitumor effect in
some degree. The GNOLs without NIR irradiation (GNOLs
(−) group) displayed a slightly antitumor efficiency than the
CS-OA-Lips, and both of them were more efficient in tumor
therapy than the free OA group. More noticeably, the tumor
size in the group treated with GNOLs plus NIR irradiation
(GNOLs (+) group) showed the highest antitumor efficacy
with an inhibition rate of 79.65% during the experiment.
These results demonstrated that chemo-photothermal com-
bined treatment based on GNOLs was superior to chemo-
therapy or photothermal treatment alone. The body weight
was also monitored as an indicator for evaluation of drug tox-
icity (Figure 12b). There was no significant reduction in body
weight of the mice treated with the drug vehicles, implying
the low toxicity of drug delivery system. The photographs of
the typical tumors at the end of the treatment period offered
www.small-journal.com 7
 full papers
an intuitive proof for the antitumor efficacy of different OA
formulations (Figure 12c). The tumor size of the GNOLs (+)
was the smallest in all the groups, so it could be seen that the
GNOLs as chemo-photothermal combined treatment dem-
onstrated a good drug delivery system for effective antitumor
therapy. The reason was that GNOLs with NIR light irradia-
tion might be attributed to the localized hyperthermia induced
by laser irradiation. The GNOLs could accumulate into tumor
tissues by the EPR effect; with the NIR irradiation, the gold
nanoshells absorbed the light and converted it to hyperthermia,
further inducing more OA released rapidly from the GNOLs
to realize the most efficient anticancer effect; meanwhile, the
tumor tissues were more sensitive to heat than normal tissues,
the tumors were easy to be killed directly by the NIR-induced
hyperthermia. Therefore, the chemo-photothermal combined
treatment could obtain an excellent antitumor effect.
3. Conclusion
In this study, gold nanoshells-coated OA liposomes were
prepared which exhibited a photothermal and pH-respon-
sive drug release and superior antitumor property. So a
smart drug delivery system for improving the efficiency of
antitumor therapeutic drug was developed based on gold
nanoshells coated liposomes. By taking advantage of the
photothermal effect of Au nanoshells and thermal-sensitivity
of lipid bilayers, the lipid coat was substantially disrupted
after NIR irradiation and a boost of drug release could be
achieved. Because of pH-responsive of chitosan, the drug
system could easily realize drug targeting transport and pH-
responsive release for the tumor tissue. The novel concept
of gold nanoshells-coated OA liposomes mediating tumor
therapy here can render a great potential for chemo-photo-
thermal antitumor therapy.
4. Experimental Section
Materials: OA was purchased from Sichuan Shifang Huakang
Pharmaceutical Raw Material Plant. Soya lecithin was purchased
from Shenyang Tianfeng Biological Pharmaceutical Co., Ltd.
(Shenyang, China). Chitosan (degree of deacetylation: 92%) was
purchased from Sinopharm Chemical Reagent Co., Ltd (Shanghai,
China). Cholesterol, surfactant Tween-80, and anhydrous ethanol
were purchased from Tianjin Guangfu Fine Chemical Research
Institute (Tianjin, China). AuCl3 was purchased from Chengdu
Xiya Chemical Industry Co., Ltd (Chengdu, China). NaBH4 was
purchased from Tianjin Guangfu Fine Chemicals Co., Ltd (Tianjin,
China). Sephadex G-75 was obtained from Beijing BioDee Bio Tech
Corporation Ltd (Beijing, China). Phosphate-buffered saline was
sourced from Sigma Chemical Company (Henan, China). Acetic
acid glacial was obtained from Tianjin Fengchuan Chemical Rea-
gent Technologies Co., Ltd. (Tianjin, China). Methanol and acetoni-
trile were purchased from Tianjin Four Friends Fine Chemicals Co.,
Ltd. and were chromatographic grade. All other chemical agents
used were of analytical grade.
Chitosan-Modified OA Liposomes Preparation: The OA liposomes
(OA-Lips) were synthesized by the ethanol injection method that was
8 www.small-journal.com © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
[20]
previously reported. In the study, the orthogonal experiment was
done to obtain the best condition to prepare the chitosan-coated OA
liposomes. For preparing chitosan-modified OA liposomes (CS-OA-
Lips), briefly, the chitosan (10 mg) was dissolved in 10 mL of acetic
acid glacial solution (0.1 m, pH 3.5). Then the chitosan solution was
added dropwise into the above OA liposomal suspension under
magnetic stirrer. The mixture was stirred for 2 h at room temperature.
Eventually, the CS-OA-Lips were prepared successfully.
Preparation of Gold Nanoshells Coated OA Liposomes: Gold
seeds were synthesized according to the method previously
reported by Jana et al.[25] Briefly, fresh NaBH4 solution (253 ×
10–3 m) and AuCl3 solution (0.5 × 10–3 m) were prepared and they
were cooled down in an ice bath condition for 5 min. Then 10 μL
of NaBH4 solution was added into the AuCl3 solution (1 mL) with
vigorous shock and the gold seeds were formed immediately.
Gold seeds and CS-OA-Lips (1:3 in volume ratio) were mixed
uniformly followed by being incubated on a rocking incubator for
20 h at room temperature. Then the AuNPs-CS-OA-Lips were syn-
thesized. Subsequently, 25 μL of AuCl3 solution (20 × 10–3 m) was
mixed with 975 μL of AuNPs-CS-OA-Lips solution. After that, the
resulting solution and fresh NaBH4 solution were placed in ice
bath for 5 min. Then 10 μL of NaBH4 solution (253 × 10–3 m) was
added into the mixture following mingling evenly. The solution was
incubated for 8 h at room temperature and aged overnight at 4 °C.
Finally, the GNOLs were obtained. Additionally, the gold nanoshell
coated OA liposomes without chitosan modification were designed
and prepared in the same way.
Characterization of GNOLs: The encapsulation efficiency of
GNOLs was determined by dextran gel column chromatography
and HPLC. Drug encapsulation efficiency was calculated from the
following equation
EE(%) = ( Wen /Wtotal ) × 100%

Where Wen was the amount of OA encapsulated in liposomes,
and Wtotal was the initial amount of OA added in the preparation,
respectively.
Morphology of the GNOLs was observed by transmission elec-
tron microscopy (TEM, HT 7700, Japan). Malvern Zetasizer ZS (Mal-
vern Instruments, UK) was used to measure the sizes and surface
zeta potentials of the liposomes. First, the sample was diluted
ten times with deionized water. Then diameter and zeta potential
of the GNOLs were detected by DLS and electrophoretic mobility
measurement, respectively. Each sample was measured three
times at room temperature. The composition of the samples was
also determined by EDS analysis using Energy Dispersive X-Ray
Spectroscopy on the TEM (HT7700, Japan).
The UV–vis absorption spectra were recorded on a spectro-
photometer (UV2550, Shimadzu, Japan). FTIR analyses of pure chi-
tosan, OA-Lips, CS-OA-Lips, and GNOLs were recorded in the range
of 400–4000 cm−1 with speed of 2 mm s−1 by an FTIR spectrom-
eter. The samples were prepared by processing compressed KBr
disks before detection and analysis at room temperature.
Stability Study: The stability of liposomes is a key factor for the
therapy effect. In the study, the stability of GNOLs and CS-OA-Lips
was evaluated after storage at 4, 25, and 37 °C for a period of 0, 5,
10, 15, 20, 25, and 30 d, respectively. At the different time points,
the stability of the liposomes was evaluated by drug encapsulation
efficiency. All experiments were performed in triplicate.
small 2016,
DOI: 10.1002/smll.201503961
www.MaterialsViews.com

Drug Release Assay: The released performances of OA from  OD sample 

GNOLs were examined using a dynamic dialysis method at pH 5.5
and pH 7.4 conditions. The PBS media (pH 5.5 and 7.4) were used
to simulate the tumor and the normal body environments, respec-
tively, which was further divided into two groups (NIR group and
non-NIR groups). An aliquot of each sample (1 mL) was loaded in
the pretreated dialysis bag (cutoff 12–14 kDa) and was immersed
in 200 mL of PBS with a stirring of 100 rpm at 37 °C. After 8 h,
the samples in NIR groups were irradiated with NIR light (808 nm,
1 W cm−2) for 4 min. At the designated time points (12, 16, 20, and
24 h), the samples in NIR groups were treated in the same way.
During the experiment, eight samples were prepared in each group
(NIR group and non-NIR group) and the dialysis bags were with-
drawn from the release media at the predetermined time intervals
(1, 2, 4, 8, 12, 16, 20, and 24 h). Similarly, the drug release perfor-
mance of the mixture containing AuNPs and OA liposomes under a
different pH condition was examined in the same way. Concentra-
tion of OA was determined by HPLC after demulsification, and all
experiments were carried out in triplicates.
Antitumor Effect of GNOLs In Vitro: The antitumor effects of
free OA, blank liposomes (Lips), chitosan-coated blank liposomes
(CS-Lips), CS-OA-Lips, AuNS-CS-Lips, and GNOLs were evaluated
on 143B osteosarcoma cells using 3-(4,5-Dimethylthiazol-2-yl)-
2,5-diphenyltetrazolium bromide (MTT) assay. The 143B cells were
cultured in Dulbecco’s modified eagle medium with 1% penicillin/
streptomycin and 10% fetal bovine serum at 37 °C in a humidified
incubator. PBS was used as the negative control. Each group was
dealt with or without NIR light irradiation.
Briefly, the 143B cells were seeded in 96-well flat-bottomed
plates at a density of 5000 cells per well. Then the cells were cul-
tured at 37 °C for 24 h in a 5% CO2 atmosphere. After that, the
above samples were filtered through sterile 0.22 μm filter mem-
brane and diluted to different concentrations (100, 50, 25, 12.5,
6.25, 3.13, 1.56, and 0.78 μg mL−1). Thereafter, each sample was
added into the appropriate well in triplicate (100 μL per well), and
then the cells were attached and cultured. After 4 h, the NIR groups
were exposed in NIR light (808 nm, 1 W cm−2) condition for 30 sec
and incubated continuously for 20 h. The other groups were incu-
bated directly with the samples for 24 h. After the above treatment,
the culture medium was discarded and replaced by 200 μL MTT
solution(0.5 mg mL−1), and the cells were incubated for another
4 h. After removing the MTT solution, 150 μL of dimethyl sulfoxide
was added to each well to dissolve the blue formazan crystals pro-
duced by live cells. Then the absorbance of each well at 490 nm
was detected by a microplate reader (SpectraMax M2e). The cell
inhibition rate was calculated as follows
Inhibition rate (%) =  1 − × 100%
 OD control 
ODsample and ODcontrol represented the absorbance of the
sample groups and PBS group, respectively.
The photothermal effect of GNOLs with NIR irradiation on
tumor cells was evaluated. Briefly, the 143B cells were seeded in
6-well flat-bottomed plates with a density of 5000 cells per well,
and incubated at 37 °C in a humidified atmosphere with 5%
CO2. After the cells were cultured for 24 h, 1 mL of filter-sterilized
sample (PBS, AuNS-CS-Lips or GNOLs) was added in appropriate
well in triplicate. To be cultured for 4 h, the cells were irradiated
for 30 s by NIR light (808 nm, 1 W cm−2). Continued culturing 20 h
later, the cells were stained by FDA followed by observing cells’
survival status. The cells incubated with free OA, CS-OA-Lips, or
GNOLs in the absence of laser irradiation were as control. The cells
were imaged through a light microscope (Nikon, Japan).
Antitumor Effect In Vivo Study: Kunming female mice (29 ± 1 g)
were purchased from Vitalriver Laboratory Animal Center (Beijing,
China). They were allowed to be free to drink and eat at the feeding
conditions (20 °C, 60–65% relative humidity). U14 cells were pur-
chased from Chinese academy of medical sciences tumor cells
bank (Beijing, China). All the animal experiments were carried
out in compliance with the Guide for Care and Use of Laboratory
Animals.
For the establishment of animal tumor model, the 200 μL of
U14 cells suspension (1 × 107 cells) was subcutaneously injected
into the right armpit of each mouse. When the tumors reached
≈100 mm3, the mice were randomly divided into six groups
(Table 1). Normal mice without any treatment were as control.
During the experiment, the mice were treated by the intratumoral
injection at a dosage of 1.8 mg OA per kg every day for a total of
15 d. Meanwhile, the mice were subjected to NIR laser irradiation
(808 nm, 1 W cm−2) for 4 min every day in the laser irradiation
groups.
Tumor volumes and body weights of mice were recorded
every other day since the first administration. The tumor volume
was obtained according to the equation (a × b2) / 2, where a
and b were the longest and shortest diameters, respectively.
All of the mice were sacrificed on the fifteenth day. The tumors
were photographed and weighed to obtain the tumor inhibition
rates of different drugs, which were calculated according to the
formula
 Tumor volume in test group 
Inhibition rate(%) =  1 − × 100%
 Tumor volume in control group 

Table 1 . Animal groups in experiments.

Groups Number Treatment methods

Normal group 6 Normal mice without injection
Negative group 6 Physiological saline (200 μL) was injected every day
Laser group 6 Physiological saline (200 μL) was injected plus NIR irradiation the tumor for 4 min every day
Free OA group 6 Free OA suspension (200 μL) was injected every day
CS-OA-Lips group 6 CS-OA-Lips suspension (200 μL) was injected every day
GNOLs (−) group 6 GNOLs suspension (200 μL) was injected every day
GNOLs (+) group 6 GNOLs suspension (200 μL) was injected with NIR light irradiated the tumor for 4 min every day

small 2016, © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.small-journal.com 9
DOI: 10.1002/smll.201503961
 full papers
Statistical Analysis: Quantitative data were presented as a
mean value with its standard deviation indicated (mean ± SD). Sta-
tistical significance (*P <0.05 or **P <0.01) was determined using
one-way analysis of variance with SPSS 19.0 for Windows (SPSS
Inc., Chicago, IL, USA).
Supporting Information
Supporting Information is available from the Wiley Online Library
or from the author.
Acknowledgements
L.L., Y.B., and Y.L. contributed equally to this work. This work was
supported by the National Natural Science Foundation (Grant No.
21476190) the Hebei province key basic research Foundation
(Grant No.15961301D), the Hebei province science and technology
project (Grant No. 142777118D), and fund from Qinhuangdao sci-
ence and technology research and development project (Grant No.
201202B029).
[1] R. Chandrawati, F. Caruso, Langmuir 2012, 28, 13798.
[2] T. Yang, M. K. Choi, F. D. Cui, J. S. Kim, S. J. Chung, C. K. Shim,
D. D. Kim, J. Control. Release 2007, 120, 169.
[3] B. B. Sharma, S. K. Jain, S. P. Vyas, J. Microencapsul. 1994, 11,
279.
[4] L. P. Zhao, H. Xiong, H. Peng, Q. Wang, D. Han, C. Q. Bai, Y. Z. Liu,
S. H. Shi, B. Deng, Eur. Food Res. Technol. 2011, 232, 647.
[5] P. Bawa, V. Pillay, Y. E. Choonara, T. L. Du, Biomed. Mater. 2009,
4, 139.
[6] J. Ge, E. Neofytou, C. T. Rd, R. E. Beygui, R. N. Zare, ACS Nano
2012, 6, 227.
[7] K. C. Hribar, M. H. Lee, D. Lee, J. A. Burdick, ACS Nano 2011, 5,
2948.
[8] P. Huang, J. Lin, W. W. Li, P. F. Rong, Z. Wang, S. J. Wang,
X. P. Wang, X. L. Sun, M. Aronova, G. Niu, R. D. Leapman, Z. H. Nie,
X. Y. Chen, Angew. Chem. 2013, 52, 13958.
[9] X. H. Huang, I. H. El-Sayed, M. A. El-Sayed, Methods Mol. Biol.
2010, 624, 343.
[10] K. Imura, T. Nagahara, H. Okamoto, J. Am. Chem. Soc. 2004, 126,
12730.
[11] J. K. Young, E. R. Figueroa, R. A. Drezek, Ann. Biomed. Eng. 2012,
40, 438.
[12] R. Weissleder, Nat. Biotechnol. 2001, 19, 316.
[13] X. Huang, M. A. El-Sayed, J. Adv. Res. 2010, 1, 13.
10 www.small-journal.com © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
[14] A. M. Gobin, M. L. Lee, N. J. Halas, W. D. James, R. A. Drezek,
J. L. West, Nano Lett. 2007, 7, 1929.
[15] L. O. Svaasand, C. J. Gomer, E. Morinelli, Laser. Med. Sci. 1990, 5,
121.
[16] W. C. Jeong, S. H. Kim, S. M. Yang, ACS Appl. Mater. Inter. 2014,
6, 826.
[17] W. I. Choi, A. Sahu, Y. H. Kim, G. Tae, Ann. Biomed. Eng. 2012, 40,
534.
[18] T. Ta, T. M. Porter, J. Control. Release 2013, 169, 112.
[19] Y. N. Zhang, W. Zhang, D. Hong, L. Shi, Q. Shen, J. Y. Li, J. Li,
L. H. Hu, Bioorgan. Med. Chem. 2008, 16, 8697.
[20] Y. H. Bian, D. W. Gao, Y. P. Liu, N. Li, X. W. Zhang, R. Y. Zheng,
Q. Q. Wang, L. Y. Luo, K. Dai, RSC Adv. 2015, 5, 18725.
[21] G. F. Marcos, M. J. Alonso, J. Control. Release 2012, 161, 496.
[22] N. G. Schipper, Pharm. Res.-Dordr. 1997, 14, 923.
[23] D. Peer, J. M. Karp, S. Hong, O. C. Farokhzad, R. Margalit,
R. Langer, Nat. Nanotechnol. 2007, 2, 751.
[24] X. P. Duan, Y. P. Li, Small 2013, 9, 1521.
[25] N. R. Jana, L. Gearheart, C. J. Murphy, J. Phys. Chem. B. 2001, 105,
4065.
[26] I. Fratoddi, I. Venditti, C. Cametti, M. V. Russo, J. Mater. Chem. B
2014, 2, 4204.
[27] N. Li, P. X. Zhao, D. Astruc, Angew. Chem. 2014, 53, 1756.
[28] A. Topete, M. Alatorre-Meda, E. M. Villar-Álvarez, A. Cambón,
S. Barbosa, P. Taboada, V. Mosquera, ACS Appl. Mater. & Inter.
2014, 6, 11142.
[29] T. M. Sun, Y. S. Zhang, P. Bo, H. D. Choon, M. X. Yang, Y. N. Xia,
Angew. Chem. 2014, 53, 12320.
[30] K. Maruyama, Adv. Drug Delivery Rev. 2011, 63, 161.
[31] J. Song, L. Cheng, A. Liu, J. Yin, M. Kuang, H. Duan, J. Am. Chem.
Soc. 2011, 133, 10760.
[32] A. K. Rengan, M. Jagtap, A. De, R. Banerjee, R. Srivastava,
Nanoscale 2014, 6, 916.
[33] P. Panwar, B. Pandey, P. C. Lakhera, K. P. Singh, Int. J. Nanomed.
2010, 5, 101.
[34] X. S. Liu, Y. J. Chen, H. Li, N. Huang, Q. Jin, K. F. Ren, J. Ji, ACS Nano
2013, 7, 6244.
[35] A. K. Rengan, M. Jagtap, A. De, R. Banerjee, R. Srivastava,
Nanoscale 2013, 6, 916.
[36] Y. G. Sun, Y. N. Xia, Anal. Chem. 2002, 74, 5297.
[37] C. M. Cobley, J. Chen, E. C. Cho, L. V. Wang, Y. N. Xia, Chem. Soc.
Rev. 2011, 40, 44.
[38] X. H. Zheng, X. Da, F. F. Zhou, B. Wu, W. R. Chen, Mol. Pharmaceut.
2011, 8, 447.
[39] T. L. Torzsas, C. Kendall, M. Sugano, Y. Iwamoto, A. V. Rao, Food
Chem. Toxicol. 1996, 34, 73.
[40] S. A. Altman, L. Randers, G. Rao, Biotechnol. Progr. 1993, 9,
671.
[41] K. W. Chan, G. Liu, X. Song, H. Kim, T. Yu, D. R. Arifin, A. A. Gilad,
J. Hanes, P. Walczak, P. C. van Zijl, J. W. Bulte, M. T. McMahon, Nat.
Mater. 2013, 12, 268.
Received: December 30, 2015
Revised: April 28, 2016
Published online:
small 2016,
DOI: 10.1002/smll.201503961