Dynamic Article LinksC<Nanoscale

Cite this: Nanoscale, 2011, 3, 3923

www.rsc.org/nanoscale PAPER

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3FView Article Online / Journal Homepage / Table of Contents for this issue

Inhibiting metastasis of breast cancer cells in vitro using gold nanorod-siRNAdelivery system†

Weiqi Zhang,‡a Jie Meng,‡a Yinglu Ji,b Xiaojin Li,a Hua Kong,a Xiaochun Wu*b and Haiyan Xu*a

Received 3rd June 2011, Accepted 19th July 2011

DOI: 10.1039/c1nr10573f

Breast cancer is the most common malignant disease in women, and it is not the primary tumor but its

metastasis kills most patients with breast cancer. Anti-metastasis therapy based on RNA interference

(RNAi) is emerging as one of promising strategies in tumor therapy. However, construction of an

efficient delivery system for siRNA is still one of the major challenges. In this work, siRNA against

protease-activated receptor-1 (PAR-1) which is a pivotal gene involved in tumor metastasis was

conjugated to gold nanorods (AuNRs) via electrostatic interaction and delivered to highly metastatic

human breast cancer cells. It was demonstrated that the siRNA oligos were successfully delivered into

the cancer cells and mainly located in vesicle-like structures including lysosome. After transfected with

the complex of AuNRs and PAR-1 siRNA (AuNRs@PAR-1 siRNA), expression of PAR-1 at both

mRNA and protein levels were efficiently down regulated, as evidenced by quantitative real time PCR

and flow cytometry analysis, respectively. Transwell migration assay confirmed the decrease in

metastatic ability of the cancer cells. The silencing efficiency of the complex was in-between that of

TurboFect and Lipofectamine, however, the cytotoxicity of the AuNRs was lower than that of the

latter two. Taken together, AuNRs with PAR-1 siRNA are suited for RNAi based anti-metastasis

therapy.

1. Introduction

Breast cancer is the most common malignant disease in women.

Despite exciting progress in the understanding of breast cancer

development and progression, and in the development of novel

therapeutic strategies, breast cancer remains the top leading

cause of cancer-related death in women. It has been proved that

breast cancer-related deaths are mainly due to the ‘‘incurable’’

nature of metastatic breast cancer (MBC) at the current time. It is

estimated that about 6% of patients have metastatic disease at the

time of diagnosis and 20% to 50% patients first diagnosed with

primary breast cancer will eventually develop metastatic disease.

The current treatment strategies for breast cancer metastasis

largely rely on the use of systemic cytotoxic agents, which

frequently deteriorate the patient’s life quality due to severe side

effects and, in many cases, have limited long-term success.

Therefore, MBC remains the most challenging task facing both

cancer researcher and oncologist.1–4

aInstitute of BasicMedical Sciences, Chinese Academy ofMedical Sciencesand Peking Union Medical College, Beijing, 100005, P. R. China. E-mail:xuhy@pumc.edu.cnbCAS Key Laboratory of Standardization and Measurement forNanotechnology, National Center for Nanoscience and Technology,Beijing, 100190, P. R. China. E-mail: wuxc@nanoctr.cn

† Electronic supplementary information (ESI) available: See DOI:10.1039/c1nr10573f

‡ These authors contributed equally to this work.

This journal is ª The Royal Society of Chemistry 2011

The protease-activated receptor-1 (PAR-1) is a seven-pass

transmembrane G protein that is overexpressed in various highly

metastatic tumor cells, and its expression level is found to be

positively correlated with levels of invasion and metastasis for

various kinds of cancer cells. It has been proved that activation of

PAR-1 protein leads to elevated expression of genes associated

with adhesion, migration and invasion of breast cancer and

melanoma cancers.5–7 Boire et al. found that 80% down-regula-

tion of PAR-1 expression by RNAi in breast cancer cells could

result in 60–70% loss in migration and invasion in vitro.5 These

research results indicate the feasibility of PAR-1 gene silencing in

treatment against breast cancer metastasis. With the rapid

development of biotechnology, RNAi therapy has become one of

the most promising therapeutic modules in anti-tumor fights. For

example, Villares et al. applied neutral liposome as transfection

vector to deliver PAR-1 siRNA into melanoma cells and

inhibited the tumor growth and metastasis significantly.8 Quite

recently, Davis and co-workers have provided first clinical

evidence of anti-tumor therapy based on RNAi via targeted

nanoparticles.9 Hence RNA interference (RNAi) based anti-

metastasis therapy represents one of promising strategies.

Fabricating efficient delivery systems for siRNA plays

a crucial role in RNAi technique as well as finding optimum

target genes and designing siRNA sequences.10 Viral or nonviral

(such as lipid) vectors have shown their efficacy in gene delivery

for years;11,12 however, the existing transfection vectors still have

their limitations and obstacles. For instance, virus vectors have

Nanoscale, 2011, 3, 3923–3932 | 3923

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3F

View Article Online

risks of mutagenesis and may induce unexpected immune

responses; transfection efficiency of liposome in primary cells still

requires improvement and their potential toxicity remains

concerns.11,13 In recent years, nanoscale materials have exhibited

attractive potentials as a new kind of nonviral vector in RNAi

due to their excellent ability in entering cells, multifunctionality

and better biocompatibility,14,15 to name a few.16–19 For example,

gold nanoparticles capped with polyethyleneimine (PEI) as

siRNA vector had superior silencing effects than PEI alone while

exhibiting no obvious cytotoxicity,16 and oligonucleotides-

modified gold nanoparticles could be internalized by scavenger

receptor-mediated endocytosis.19

Gold nanorods (AuNRs), rod-shaped gold nanoparticles, have

unique optical properties. They have two surface plasmon reso-

nance(SPR) bands across visible and near infrared spectral

region, a transverse surface plasmon resonance (TSPR) band

around 520 nm and a longitudinal surface plasmon resonance

(LSPR) band at longer wavelength with tunable maximum

according to aspect ratio (length versus width). The enhanced

SPR endows the AuNRs more potentials in optical imaging

(such as two-photon luminescence and surface enhanced Raman

scattering),20 photothermal therapy, diseases diagnosis, bio-

sensing, and gene delivery.21,22 For example, upon light stimu-

lation at the nanorods’ LSPR peaks, the DNA sequence coupled

with the AuNRs could be selectively released.23–25 In addition,

the facile modification and large surface area-to-volume ratio of

the AuNRs confer the ability to efficiently bind and deliver

nucleic acids into cells.21–27 Interestingly, AuNRs have been

demonstrated to be able to penetrate across blood brain barrier

(BBB) and silence the expression of DARPP-32 by delivering

specific siRNAs into the neuron cells.26

Considering the significance of anti-metastasis therapy and the

unique properties of the AuNRs, we investigated the feasibility

of RNAi via AuNRs vectors in inhibiting breast cancer metas-

tasis. In this work, the cell line of MDA-MB-231 is taken as

Scheme 1 Schematic illustration of inhibiting metastasis of breast cancer ce

conjugated with cationic AuNRs by electrostatic interaction. (b) The MDA-M

and can efficiently migrate through the transwell membrane. (c) The complex

by the cells. After that, the PAR-1 siRNA can be released in the cytoplasm u

silencing complex (RISC), PAR-1 siRNA recognizes the PAR-1 mRNA and in

protein down-regulated and result in the inhibition of the breast cancer cells’

3924 | Nanoscale, 2011, 3, 3923–3932

a human breast cancer model in vitro because it is highly meta-

static with overexpression of PAR-1.5 We demonstrated that

cationic polyelectrolyte coated AuNRs efficiently bind the

siRNA oligos (which are specifically targeted to PAR-1) and

form a complex of AuNRs and PAR-1 siRNA (AuNRs@PAR-1

siRNA). The complex could easily enter the cells and effectively

silence PAR-1 expression both at mRNA and protein levels. A

marked inhibition of MDA-MB-231 cells migration was further

proved by transwell migration assay. The whole design is shown

in Scheme 1.

2. Materials and methods

Materials

Cetyltrimethylammonium bromide (CTAB), hydrogen tetra-

chloroaurate (III) trihydrate (HAuCl4$3H2O), silver nitrate

(AgNO3), L-ascorbic acid, glutaraldehyde (50% aqueous solu-

tion), and sodium borohydride (NaBH4) were purchased from

Alfa Aesar. Poly (sodium-p-styrenesulfate) (PSS, molecular

weight: 70000) and poly (diallyldimethyl ammoniumchloride)

(PDDAC, 20%) were purchased from Aldrich.

PAR-1 siRNA, NC siRNA oligos and fluorescently labeled

siRNA (PAR-1 siRNA-FAM) were synthesized by GenePharma

(Shanghai, China). The sequences are as follows: PAR-1 siRNA:

50-AGAUUAGUCUCCAUCAAUA-30, NC siRNA (which had

no target interior the cells): 50-UUCUCCGAACGUG

UCACGU-30.8 Lipofectamine� 2000 (Lipofectamine) and Tur-

boFect� siRNA transfection reagent (TurboFect) was

purchased from Invitrogen and Fermentas respectively. Deion-

ized water (18.2 M U cm�1) produced by Milli-Q Synthesis

(Millipore Co., USA) was used in all the experiments. Stock

solutions of sodium borohydride were freshly prepared for each

experiment. Serum-free medium (opti-MEM) was purchased

from Invitrogen. All chemicals were used as received.

lls using AuNRs@PAR-1 siRNA. (a) The PAR-1 siRNA oligos can be

B-231 cells with over-expression of PAR-1 protein are highly metastatic

of PAR-1 siRNA and AuNRs (AuNRs@PAR-1 siRNA) are internalized

nder the intracellular environment. With the assistance of RNA induced

duces the degradation of them. These make the expression level of PAR-1

migration.

This journal is ª The Royal Society of Chemistry 2011

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3F

View Article Online

Preparation of AuNRs

The synthesis of AuNRs coated with PDDAC was described in

previous work.28,29 The PDDAC-coated AuNRs were centri-

fuged and re-dispersed in RNase free water as final stock solu-

tion. The concentration of AuNRs was 0.6 nM in number of

nanorods not in gold atoms.

Optical spectroscopy

The UV–vis–NIR spectroscopy of AuNRs was performed on

Perkin Elmer UV–vis/near-infrared spectrophotometer (Lamdba

950). The bandwidth of the light source was 1 nm. 20 ml (0.5 mg

ml�1) of PAR-1 siRNA oligos was added into 1 ml of 0.6 nM

AuNRs and mixed by rapid pipetting. Before loading into the

quartz cell for measurements, the complex was left undisturbed

for about 30 min to allow complete complexation.

Scanning electron microscopy

Scanning electron microscopy (SEM) was performed on

Hitachi S-5200. Briefly, a 10 ml drop of AuNRs or AuNR-

s@PAR-1 siRNA complex were deposited on a conducting

silicon wafer and left dried. Images were recorded at various

magnifications.

Dynamic light scattering (DLS) measurements

Zeta potential and size of AuNRs were measured with a Zeta-

sizer Nano ZS90 (Malvern instruments) at room temperature.

Various amounts of PAR-1 siRNA oligos (0.5 mg ml�1) was mixed

with 1 ml AuNRs (0.6 nM) by pipetting several times. The ratio

of PAR-1 siRNA to AuNRs (siRNA/AuNRs) was defined as

mass of siRNA (mg) to amounts of AuNRs (pM) and the resul-

tant complex was defined as AuNRs@PAR-1 siRNA (siRNA/

AuNRs ratio number). The complexes were stayed at room

temperature for about 30 min before DLS measurements were

performed.

Agarose gel electrophoresis

A 0.5 mg of PAR-1 siRNA oligo was mixed with 50 ml AuNRs at

different siRNA/AuNRs ratio. After 30 min incubation under

the room temperature, solution of the complex was centrifuged

at 12000 rpm for 5 min. Then 20 ml of supernatants was loaded

onto 1% agarose gel containing ethidium bromide (0.5 mg ml�1)

in 1 � TAE buffer (Tris-acetate-EDTA buffer). The gel was run

for 10 min at 150 V and visualized under UV light using Binta

2020D imaging system (Binta, Beijing, China).

Cells culture

Highly metastatic human breast cancer cell line MDA-MB-231

was obtained from Cell Resource Center, IBMS, CAMS and

cultivated in Leibovitz’s L-15 Medium (Gibco Invitrogen, CA,

USA) supplemented with 10% fetal bovine serum (FBS), 100 U

ml�1 penicillin, 100 U ml�1 streptomycin and maintained in

a 37 �C humidified incubator with a low-CO2 environment.

This journal is ª The Royal Society of Chemistry 2011

Confocal microscopy

MDA-MB-231 cells were seeded and grown on coverslips in 24-

well plates 4 � 104 cells well�1). The cells were transfected with

naked PAR-1 siRNA-FAM or AuNRs conjugated with PAR-1

siRNA-FAM (AuNRs@PAR-1 siRNA-FAM) for 6 h in Lei-

bovitz’s L-15 Medium containing 10% FBS. The ratio of siRNA/

AuNRs in AuNRs@PAR-1 siRNA-FAM complex is 17. The

cells were washed twice with cold PBS buffer solution and fixed

in 1% formaldehyde for 15 min at room temperature. After

washing for 3 times with cold PBS buffer solution, the coverslips

were mounted using an aqueous mounting medium with DAPI

(Zhongshan Goldenbridge biotechnology Co, Beijing, China).

The cells were visualized with laser confocal microscope (Ultra-

VIEW VoX, Perkin Elmer) and the images were analyzed using

Volocity-5 software (Perkin Elmer).

Subcellular localization of AuNRs by TEM

The subcellular localization of AuNRs was observed on

a transmission electron microscope (TEM, JEM-1010). MDA-

MB-231 cells of 1 � 106 were seeded into a 100 mm dish and left

overnight to allow cells attachment. AuNRs@PAR-1 siRNA

was added to the dish at AuNRs concentration of 60 pM. After

incubation for 10 min or 48 h, the cells were washed with PBS

buffer solution for several times and scraped gently from the

dish. The collected cells were centrifuged into a small pellet

which was fixed in 2.5% glutaraldehyde for 1 h. The resulting

pellets were dehydrated gradually by alcohol and embedded in

Epon. Ultrathin sections were cut and placed on a copper

meshwork.

Transfection procedure of PAR-1 siRNA

MDA-MB-231 cells (1.6 � 105 cells/well) were seeded into 6-well

plates and left overnight to adhere and reach a confluence of about

70%. Themediumwas supplementedwith antibiotics except for cells

treated with Lipofectamine that was antibiotics-free. AuNRs of 0.12

pM or 0.2 pMwas mixed with 2 mg siRNA oligos in eppendorf tube

to formAuNRs@PAR-1siRNAwith ratio17or10 respectively.The

mixed solutions were incubated at room temperature for 30min and

gentlydripped intowells and thefinalmediumvolumewas2ml.Cells

with no transfection reagents and siRNA molecules were untreated

controls. Cells transfected with 2 mg NC siRNA conjugated with

AuNRs (AuNRs@NC siRNA) or 2 mg naked siRNA were set as

controls. The transfection procedure mediated by TurboFect and

Lipofectamine were performed according to the manufacturer’s

instruction. Briefly, 2 mg siRNA diluted in opti-MEM was mixed

with 4 ml TurboFect or 5 ml Lipofectamine diluted in 250 ml opti-

MEMrespectively. The solutionswere incubated for 20min at room

temperature, and then added into the plates. After 2 days of incu-

bation, the expression of PAR-1 at mRNA and protein levels was

analyzed by real time PCR and flow cytometry respectively.

RNA extraction, reverse transcription and real time PCR

Total RNA was extracted using Trizol reagent (Invitrogen). A

0.5 mg total RNA from each sample was reverse transcribed to

cDNA in a final volume of 20 ml using M-MLV reverse tran-

scriptase (Takara, Otsu, Japan) primed by oligo (dT). Real time

Nanoscale, 2011, 3, 3923–3932 | 3925

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3F

View Article Online

PCR analysis was carried out in a total reaction volume of 25 ml

containing 1 ml cDNA as template, 12.5 ml of 2 � SYBR Green I

Master Mix (Takara, Otsu, Japan), and 200 nM each of sense

and antisense primers. All reactions were run in triplicate on an

iQ5 thermocycler (Bio-rad, USA), after a pre-denaturation step

at 95 �C for 30s, 40 cycles were performed at 95 �C for 5s and

then 60 �C for 40s. Expression was normalized to GAPDH and

following primers were used: PAR-1 sense 50-GGGCTTCCT

TCACTTGTCT-30 and antisense 50-ACTTCTTGCTGCG

GTTGG-30;30 GAPDH sense 50-GGTCACCAGGGCTGCTTT

TA-30 and antisense 50-GAGGGATCTCGCTCCTGGA-30.31

Flow cytometry assay

For analysis of cell surface expression of PAR-1, flow cytometry

analysis was performed by using a Beckman Coulter Elite ESP

flow cytometer. Cultured cells in 6-wells plates were washed twice

with D’Hanks solution and harvested using 0.2% (m/v) EDTA in

PBS for 3 min. To remove the residual EDTA, the collected cells

were further washed once and then re-suspended with 1% bovine

serum albumin in PBS (PBS-BSA). Cells of 5 � 105 were incu-

bated on a rotating shaker with 1 mg anti-human PAR-1 anti-

bodies (ATAP2, Santa Cruz Biotechnology) at room

temperature for 1 h. As a negative control, cells were incubated

without primary antibody. Cells were washed once with PBS-

BSA and then incubated on a rotating shaker with FITC-

conjugated second antibody (Jackson Immunoresearch, USA) at

room temperature for 1 h. The cells were then washed and fixed

in 1% paraformaldehyde at 4 �C before performing flow

cytometry analysis.

Transwell migration assay

Tomeasure the migration activity ofMDA-MB-231 cells in vitro,

transwell migration assay was conducted in 24-well Millicell

hanging cell culture inserts (pore size is 8 mm, Millipore). Briefly,

after transfected by various siRNA formulations, cells were

harvested and suspended in serum-free medium (opti-MEM). A

200 ml of cells (1 � 105 cells) with different treatment was added

into the upper chamber and 1.3 ml complete L-15 medium was

added to the lower chamber as conditional medium. After 24-

hour of incubation, cells on upper surface of the filter were

removed by wiping with a cotton swab. The migrated cells were

fixed with 4% paraformaldehyde and then stained by crystal

violet. The numbers of the migrated cells were counted in five

randomly selected fields under microscope, and cell migration

rate was calculated by formula as follow:

Relative rate of migration (%) ¼ migrating cells with treatment/

migrating cells without treatment � 100

The transwell assay was carried out at least three times with

each transfection formula, and two repeat inserts were counted

for each transfection formula.

Cell viability assay

MTS assay. MTS assay kit (CellTiter 96 @ AQueous Non-

Radioactive Cell Proliferation Assay, Promega) was used to

3926 | Nanoscale, 2011, 3, 3923–3932

evaluate the viability of MDA-MB-231 cells according to the

manufacturer’s protocol. In brief, the cells were seeded onto the

96-well plate at a rate of about 8000 cells per well and incubated

overnight to allow cells attachment. After 24 h or 48 h of incu-

bation, the cells were washed twice with D-Hank’s solution and

then incubated with 100 ml fresh medium combined with 20 ml

MTS at 37 �C for 90 min. The absorbance of the plates was

recorded at 490 nm of wavelength using BioTek Synergy� 4

Hybrid Multi-Mode Microplate Reader (BioTek Instruments,

USA). To eliminate the interference from background contrib-

uted by cell debris and other nonspecific absorbance, the

absorbance at 630 nm was set as reference wavelength.

Measurements were conducted in triplicate and the viability of

cells incubated without transfection reagent was denoted as

100%.

Trypan blue exclusion. The viability of cells after an incubation

time of 24 h and 48 h with different concentration of AuNRs was

directly assessed by cell counting. Briefly, 4 � 104 cells/well were

seeded into 24-well plates and incubated overnight. After

a simple wash with D-hank’s solution, fresh medium containing

different concentration of AuNRs (0, 60, 75, 100, 120 pM) was

added into the wells. At the end of incubation period, the cells

were washed twice with 1 ml D-hank’s solution and detached

using 100 ml trypsin solutions for 3 min. Complete medium of

400 ml was added to each well to stop trypsin reaction. The cells

were then stained with 0.4% Trypan blue dye and loaded onto the

hemocytometer. By light microscopy, numbers of viable cells

were recorded. The cell viability of AuNRs treatment was

expressed as percentage of the number of the cells incubated with

AuNRs versus the total number of untreated cells. Each treat-

ment was conducted in triplicate.

Statistical analysis

Data were expressed as means � SD where indicated. Statistical

differences were analyzed using the Student’s t test, and value of

P < 0.05 was defined as statistically significant.

3. Results and discussion

Physicochemical characterization of AuNRs andAuNRs@PAR-1

siRNA

The surface of the as-prepared AuNRs was assembled by

a CTAB bilayer (serving as templates during synthesis). Due to

the cytotoxicity of CTAB molecules, two other polymers (PSS

and PDDAC) were further assembled to the AuNRs via layer by

layer technique. This coating approach provided AuNRs with

positive charges to bind nucleic acids and a high stability in

biological buffers and cell culture media.15,22

The optical image of the AuNRs solution (inserted in Fig. 1a)

is a pink red transparent and homogenous solution. The stock

solution of the AuNRs was stable for one year at 4 �C as no

change in UV-Vis-NIR spectra was observed over time. When

dispersed in serum containing medium (SCM), no obvious

agglomeration was observed which suggested good stability of

the AuNRs under the cell culture conditions (Fig. S1).† It is

considered that the cationic AuNRs was coated by serum

This journal is ª The Royal Society of Chemistry 2011

Fig. 1 Characterization of AuNRs before and after interaction with

siRNA oligos. (a) UV-vis-NIR absorption spectra of AuNRs (black line)

and AuNRs@PAR-1 siRNA complex (gray line). The inserted picture is

a stock solution of the AuNRs. (b, c) SEM images of as-prepared AuNRs

(b) and AuNRs@PAR-1 siRNA (c). Inserted images present the AuNRs

with lower magnification.

Fig. 2 Physicochemical characterization of AuNRs@PAR-1 siRNA.

(a–b) DLS analysis of the complexes at different PAR-1 siRNA/AuNRs

ratios: (a) Zeta potential, (b) hydrodynamic diameter. (c–d) agarose gel

electrophoresis of AuNRs@PAR-1 siRNA: (c) the complexes of

AuNRs@PAR-1 siRNA were centrifuged, and the supernatants were

electrophoresized. The siRNA to AuNRs ratio was expressed as mg

siRNA/pM AuNRs.

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3F

View Article Online

proteins in the SCM, which resulted in a stable dispersion of

AuNRs in the medium.32,33

The AuNRs have a TSPR peak at 512 nm and LSPR peak at

812 nm (Fig. 1a). From the UV–vis–NIR spectrum of the

AuNRs incubated with PAR-1 siRNA, a 10 nm red shift was

observed in the localized LSPR peak, indicative of the

complexation between AuNRs and PAR-1 siRNA (AuNR-

s@PAR-1 siRNA). The conjugated PAR-1 siRNA changed the

local refractive index of the AuNRs and induced a peak shift of

LSPR.26 XPS analysis of AuNRs@PAR-1 siRNA showed that

there was an marked increase in N/Au atomic ratio for the

complex in reference to that for the AuNRs alone, providing

a side evidence of complex formation in the solution (Table S1 in

supporting information). Fig. 1b shows a SEM image of AuNRs

deposited from the solution. From the image, the AuNRs were

well-dispersed without aggregation as well as in a narrow size

distribution. The aspect ratio was estimated to be 4.2 according

to the mean length of 57 nm and the mean diameter of 13.5 nm.

As Qiu et al. evidenced recently, the PDDAC-coated AuNRs

with an aspect ratio of 4 could be efficiently internalized by cells

with better biocompatibility which was believed to be beneficial

in siRNA delivery.29 Besides, the aspect ratio is well matched

with that calculated from the peak wavelength of the LSPR

band.20 Fig. 1c provided a SEM image of the complex, indicating

that the PAR-1 siRNA did not influence the dispersion status of

AuNRs.

Zeta potential analysis was performed to examine charge

variance of AuNRs after complexing with PAR-1 siRNA at

different ratios (Fig. 2a). The ratio of PAR-1 siRNA oligos to

AuNRs was defined as siRNA/AuNRs (mg/pM). As presented,

the AuNRs alone was positively charged and had a Zeta

This journal is ª The Royal Society of Chemistry 2011

potential value of +44.1 mV. Increasing the amount of siRNA,

Zeta potential of AuNRs@PAR-1 siRNA decreased gradually

and reversed to negative when the siRNA/AuNRs ratio reached

13.3. It was noted that when the ratio was 11.7, Zeta potential of

the complex was +4.3 mV, a value approaching to zero which

suggested instability of the complex under this condition.16,34

This was resulted from the decrease of electrostatic repulsion

between nanorods which made the AuNRs tend to aggregate.

Indeed, results obtained from dynamic light scattering analysis

(DLS) were consistent with the Zeta potential measurement. The

complex with ratio of 11.7 had a dramatic size increase in

reference to the AuNRs alone and the AuNRs@PAR-1 siRNA

with other ratios (Fig. 2b). Hence the ratio of 11.7 was not

suitable for RNAi trial. As DLS analysis assumes that particles

are spherical, size measurements for rod-shaped AuNRs were

mainly used to evaluate the stability of the AuNRs@PAR-1

siRNA.15,35

Nanoscale, 2011, 3, 3923–3932 | 3927

Fig. 3 Confocal and optical merged images of MDA-MB-231 cells after 6 h incubation with siRNA oligos. Nucleus were stained blue with DAPI and

siRNA were labeled with green fluorescent FAM. (a) cells with no treatment, (b) cells treated with naked PAR-1 siRNA-FAM, and (c) cells transfected

with AuNRs@PAR-1 siRNA-FAM. The Ratio of siRNA to AuNRs is 17. The arrows point the dot-like aggregates of AuNRs inside the cells.

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3F

View Article Online

To further examine the binding between AuNRs and PAR-1

siRNAs, gel electrophoresis retardation assay was applied. The

solution of AuNRs@PAR-1 siRNA with different ratios was

centrifuged and then the supernatants were subjected to the

electrophoresis. As the siRNA absorbed by AuNRs would

centrifuged down with the AuNRs, the siRNA molecules in the

supernatants were deemed as siRNA molecules out of AuNR-

s@PAR-1 siRNA. Fig. 2c showed that by adding more AuNRs

(the ratio of siRNA/AuNRs was decreased), free siRNA mole-

cules in the solution became less as the weak siRNA band

appeared to be weaker in the gel and finally fainted when the

ratio of siRNA/AuNRs was #10. According to above charac-

terizations, complexes with ratio of 10 (AuNRs@PAR-1 siRNA

Fig. 4 TEM images of MDA-MB-231cells after 48h incubation with

AuNRs@PAR-1 siRNA(17), in which the black arrow head points the

cytoplasmic membrane, the white arrow head points mitochondria, the

white tailed arrow points lysosome, and the black tailed arrow points

vesicle: (a) Cells with no treatment as control. (b) A whole-cell view of

cells treated with AuNRs@PAR-1 siRNA shows the cytoplasmic loca-

tion of AuNRs. (c) A representative image of AuNRs mainly located in

lysosome and vesicles. The circled area points that a perinuclear location

of AuNRs. (d) An image with higher magnification. The ellipse demon-

strates the AuNRs are near the mitochondria, and the boxed area shows

a forming endocytotic pit with AuNRs.

3928 | Nanoscale, 2011, 3, 3923–3932

(10)) and 17 (AuNRs@PAR-1 siRNA(17)) were selected to be

used in subsequent experiments to conduct RNAi with the cells

because these two complexes had similar dispersion status and

opposite apparent charges: the AuNRs@PAR-1 siRNA(10) was

positively charged with a Zeta potential of +21.2 mV, and the

AuNRs@PAR-1 siRNA(17) was negatively charged with a Zeta

potential of �22.0 mV.

Cellular uptake and intracellular location of AuNRs

The ability of delivery systems to ferry nucleic acids across cellular

membrane is very crucial for successful gene silencing. In order to

track whether the siRNA molecules were delivered into the cells,

fluorescently labeled PAR-1 siRNA (PAR-1 siRNA-FAM)

was used in confocal fluorescence microscopy assay. Images

acquired from white light and fluorescent confocal microscopy

were merged and shown in Fig. 3. It was clearly seen that cellular

Fig. 5 The mRNA level of PAR-1 expression in MDA-MB-231 cells by

real time PCR. The GAPDH gene was set as inner control and the

expression of control group was set as 100%. The data was representative

of at least four experiments performed. Results are mean � SD and

‘‘*’’represents significantly different from control, p < 0.05(*) and

p < 0.01(**) by Student’s t test.

This journal is ª The Royal Society of Chemistry 2011

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3F

View Article Online

uptake of the naked siRNA and the AuNRs@PAR-1 siRNA-

FAM was remarkably different. As pointed by arrows in Fig. 3c,

the small round black dots in cytoplasm under white light channel

indicated the entry of AuNRs. This suggests that the AuNRs can

act as siRNA vector to bring siRNA into the cells. Moreover,

there was more strong fluorescence in cells incubated with

AuNRs@PAR-1 siRNA-FAM which was evidenced by the

intensive fluorescent signals diffused among the cytoplasm. On

the contrary, the entry of naked siRNA into the cells was quite

limited, the fluorescence wasmainly located on the cell membrane

and there were no visible fluorescence inside the cells. The cyto-

plasmic location of siRNA is one of the crucial factors that

guarantee successful RNAi as the gene silencing machinery is

located in cytoplasm.17,36 The poor ability of naked siRNA to

enter the cells was resulted from their negative charge and

Fig. 6 Flow cytometric analysis of cell surface expression of PAR-1. (a–g) R

MDA-MB-231 treated with different siRNA formulas. Gray shaded histogram

control staining only with FITC-conjugated second antibody. (h) Histogram o

representative of two independent experiments. Results are mean � SD and ‘‘

(**) by Student’s t test.

This journal is ª The Royal Society of Chemistry 2011

vulnerability towards nuclease.With assistance ofAuNRs, a large

amount of the PAR-1 siRNA could be carried to the cytoplasm

and this was a prerequisite for effective gene silencing.

In order to understand the subcellular localization of

AuNRs@PAR-1 siRNA, transmission electron microscopy

(TEM) was further conducted. After only 10-minutes incubation,

the AuNRs were observed in the small vesicles in cells (Fig. S2)†.

This phenomenon validated the outstanding capability of

AuNRs in entering cancer cells. Besides, the vesicle location of

AuNRs in only 10 min suggested that the AuNRs would be

internalized through the endocytosis pathway.37 As incubation

time over, more AuNRs were found inside the cells. Fig. 4 pre-

sented representative TEM images of the cells after 48 h of

culture, which showed that the AuNRs were mainly located in

organelles; they could be seen clearly in the lysosome and vesicle-

epresentative flow cytometric diagrams present the PAR-1 expression in

s represent the PAR-1 expression; histograms in the black line represent

f relative expression of PAR-1 protein in MDA-MB-231 cells. Results are

*’’represents significantly different from control, p < 0.05(*) and p < 0.01

Nanoscale, 2011, 3, 3923–3932 | 3929

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3F

View Article Online

like structures. It was also observed that the AuNRs retained in

vesicles were located in perinuclear regions and near mitochon-

dria, however, no AuNRs were observed in nucleus during the 48

h incubation. As indicated above, the AuNRs@PAR-1 siRNA

were rapidly internalized by the cell and further located in the

lysosome and vesicles, and formed small AuNRs aggregates. The

environmental conditions including pH and redox status in

lysosome and vesicles would induce the detachment of the

siRNA which made siRNA molecules gain the chance to enter

cytoplasm which was expected to induce the RNAi effects.

Silencing effect of PAR-1 expression

In the experiment, MDA-MB-231 cells in 6-wells plate were

incubated with AuNRs@PAR-1 siRNA(17) and AuNRs@PAR-

1 siRNA(10) respectively for 48 h, while naked PAR-1 siRNA,

AuNRs@NC siRNA and commercial Lipofectamine and

TurboFect were taken as controls. To confirm the knocking

down effects, quantitative real time PCR was performed to

evaluate the mRNA variance of PAR-1 expression (Fig. 5), and

flow cytometry was employed to analyze the protein expression

level of PAR-1 (Fig. 6). It could be seen that PAR-1 expression of

the cells incubated with the naked siRNA almost paralleled the

untreated cells both in mRNA and protein levels. As the naked

siRNA itself was negative charged and vulnerable to nuclease in

media, it was hard for them to cross the cells membrane (shown

in Fig. 3) and would not induce an effective gene silencing.

Compared with untreated cells, AuNRs@PAR-1 siRNA(17)

exhibited a high silencing efficiency of 61.8% in mRNA and

56.8% in protein level, while AuNRs@PAR-1 siRNA(10)

exhibited a medium silencing degree of 42.9% in mRNA level and

Fig. 7 Transwell migration assay ofMDA-MB-231 cells transfected with diffe

transwell inserts of MDA-MB-231 cells transfected with different siRNA for

(10), d-AuNRs@PAR-1 siRNA(17), e-Lipofectamine, f-TurboFect). The ins

violet. (g) Histogram of relative cell migration of MDA-MB-231 with differe

migrated cells with treatment compared to those without treatment and result

mean � SD and ‘‘*’’represents significantly different from control, p < 0.05(*

3930 | Nanoscale, 2011, 3, 3923–3932

50.0% in protein level. No significant silencing effects were

observed in the cells treated with AuNRs@NC siRNA, which

validated the specificity of PAR-1 gene silencing. The cationic

lipid Lipofectamine showed silencing effect of 79.7% in mRNA

level and 81.1% in protein level, and the cationic polymer Tur-

boFect presented 39.0% and 50.5% respectively. The optimum

silencing efficiency of AuNRs@PAR-1 siRNA(17) was between

the two commercial controls. The silencing efficiency of

AuNRs@PAR-1 siRNA(10) were slightly lower than that of

AuNRs@PAR-1 siRNA(17) both at mRNA and protein level,

however, it is interesting to notice that more AuNRs@PAR-1

siRNA(10) were internalized into the cells as shown in Fig. S3.

Because AuNRs@PAR-1 siRNA(10) complex was positively-

charged, it is rational to consider that the siRNA molecules were

absorbed more compactly than those in negatively-charged

AuNRs@PAR-1 siRNA(17), which may hinder the release of

siRNA molecules and result in a lower gene silencing

efficiency.18,38

Inhibition effects of RNAi on migrating function of MDA-MB-

231 cells

As mentioned above, metastasis of breast cancer cells is posi-

tively associated with the expression level of PAR-1. Here

transwell migration assay was used to examine the motility of

MDA-MB-231 cells after incubated with AuNRs@PAR-1

siRNA (Scheme 1). As shown in Fig. 7, a decrease in cell

migration was obvious and proportional to the gene silencing

degree. The relative migration inhibition for Lipofectamine,

TurboFect, AuNRs@PAR-1 siRNA(17) and AuNRs@PAR-1

siRNA(10) were 77.4%, 42.5%, 56.7% and 45.2% respectively,

rent siRNA formulations. (a–f) Representative microscopic images of the

mulations (a-control, b-AuNRs@NC siRNA, c-AuNRs@PAR-1 siRNA

erts show the macroscopic images of transwell inserts stained by crystal

nt treatment. Relative cell migration was expressed as the percentage of

s are representative of at least three independent experiments. Results are

) and p < 0.01(**) by Student’s t test.

This journal is ª The Royal Society of Chemistry 2011

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3F

View Article Online

while the negative control of AuNRs@NC siRNA had only

minimal effect on the cellular migration. These provide strong

supportive evidence for successful PAR-1 gene silencing from the

perspective of cellular function: AuNRs delivered PAR-1 siRNA

into the breast cancer cells and down-regulated the PAR-1

expression effectively and finally inhibited the migration function

of the cells.

Although the RNAi efficiency using AuNRs was not as high as

that of Lipofectamine, the multifunctional potentials and low

cytotoxicity of AuNRs make them attractive and suitable as

RNAi vector in anti-metastasis therapy. Moreover, by further

surface modification, there is still enough room to enhance their

performance in delivering siRNA and down-regulating the gene

expression.

Fig. 8 Cytotoxic evaluation of AuNRs towards MDA-MB-231 cells by

MTS assay and trypan blue exclusion assay. (a–b) MTS assay evaluated

the cellular viability of cells treated with different concentration of

AuNRs after 24h (a) and 48h (b) incubation. The concentrations

of AuNRs, Lipofectamine and TurboFect were expressed as folds of

optimum concentration and the 0, 0.5, 1, 2, 5, 10 folds of optimum

concentration for AuNRs were corresponding to 0, 30, 60, 120, 300,

600 pM respectively. The dot line labeled the IC50 value. (c) Trypan blue

exclusion assay directly assessed the cell numbers after incubation with

AuNRs of different concentration for 24 and 48 h.

Cytotoxicity evaluation of AuNRs

Low cytotoxicity is a prerequisite for an ideal transfection vector.

We examined cytotoxicity of the AuNRs and made comparisons

with that of two commercial transfection vectors, Lipofectamine

and TurboFect. In order to compare cytotoxic effects with

concentration variation, an optimum concentration for each

vector was used as starting one followed by folds increase. For

the two commercial vectors, recommended concentrations in the

transfection protocols by manufacturers were used as optimum

concentrations (0.2 ml and 0.25 ml per well of a 96-well plate for

TurboFect and Lipofectamine respectively). For AuNRs, 60 pM

was used as starting concentration, as it is the lowest concen-

tration of AuNRs in our RNAi assay. Data obtained from MTS

assay showed that cellular viability was decreased to different

percentages with increasing concentrations for AuNRs, Lip-

ofectamine, and TurboFect, indicating that the cytotoxic effects

of the three vectors were dose-dependent (Fig. 8a–b). When the

concentration was less than 2 folds of the optimum concentra-

tions for the three vectors, no significant toxic effects were

observed as the cells viability was higher than 85% after 24 h and

about 80% after 48h of the culture. However, as the concentra-

tion is increased to the 5 or 10 folds of the optimum concentra-

tions, the survival proportion of the cells incubated with

Lipofectamine and TurboFect for 24 h and 48 h dropped

dramatically while that with AuNRs was still above 79%. The

value of half maximal inhibitory concentration (IC50) was

determined to assess the cyotoxicity of the three vectors; the

results were presented in Fig. 8a–b. After 24h and 48h of incu-

bation, IC50 for AuNRs was still >10-fold of 60 pM while for

Lipofectamine and TurboFect, IC50 values were 6.5- and 8.1-fold

of the optimum concentration, respectively after 24 h of culti-

vation, and 5.0- and 8.4- fold of the optimum concentration,

respectively after 48 h of cultivation. Recently, Rayavarapu et al.

and Alkilany et al. reported an IC50 value of modified AuNRs

which was in the same order of magnitude with our

observation.32,39

Trypan blue exclusion assay was taken to double check the

cytotoxicity of AuNRs, with a special focus on the AuNRs

concentration employed in the siRNA transfection. Fig. 8c

showed that viable cells was more than 86% after incubated with

AuNRs under a concentration < 2 folds of optimum dose

(120 pM) for 24 h and 48 h. The concentrations of 60 pM and

100 pM of AuNRs, which were used in RNAi experiments for

This journal is ª The Royal Society of Chemistry 2011

AuNRs@PAR-1 siRNA(17) and AuNRs@PAR-1 siRNA(10),

respectively showed minimal cytotoxic effects to the MDA-MB-

231 cells. Clearly, the cationic AuNRs were more biocompatible

than the two commercial transfection vectors. El-Sayed et al.

Nanoscale, 2011, 3, 3923–3932 | 3931

Dow

nloa

ded

by G

eorg

etow

n U

nive

rsity

Lib

rary

on

06 M

arch

201

3Pu

blis

hed

on 1

5 A

ugus

t 201

1 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1NR

1057

3F

View Article Online

reported that gold nanoparticles targeted into nucleus induced

DNA damage and cellular apoptosis while the gold nanoparticles

localized in cytoplasm had less influence on cellular viability.40

According to our TEM observation, no AuNRs entered into

nucleus. Hence the absence of AuNRs in nucleus may be one

reason for the low cytotoxicity towards MDA-MB-231 cells.

4. Conclusion

In summary the gold nanorods coated with PDDAC can carry

PAR-1 siRNA into cytoplasm of metastatic breast cancer cells

and result in an efficient down-regulation of PAR-1 expression

both at mRNA and protein levels. The gene silencing led to

marked migration inhibition of the metastasis breast cancer cells.

Our results demonstrate that the AuNRs with PAR-1 siRNA are

suited for RNAi based anti-metastasis therapy.

Acknowledgements

Authors thank for financial support fromNational Key Program

of China (973 program 2010CB934002, 2011CB933504,

2011CB932802), and Natural Science Foundation of China

(NSFC 81000665). Authors also thank Ms Chaoying Wang,

Institute of Physics, Chinese Academy of Sciences for her kind

help in the experiment of scanning electron microscopy.

Reference

1 O. Pagani, E. Senkus, W. Wood, M. Colleoni, T. Cufer,S. Kyriakides, A. Costa, E. P. Winer, F. Cardoso and on behalf ofthe ESO–MBC Task Force, J. Natl. Cancer Inst., 2010, 102, 456–463.

2 P. S. Steeg, Nat. Med., 2006, 12, 895–904.3 A. Jemal, R. Siegel, J. Xu and E. Ward, Ca-Cancer J. Clin., 2010, 60,277–300.

4 B. Weigelt, J. L. Peterse and L. J. van’t Veer, Nat. Rev. Cancer, 2005,5, 591–602.

5 A. Boire, L. Covic, A. Agarwal, S. Jacques, S. Sherifi andA. Kuliopulos, Cell, 2005, 120, 303–13.

6 N. Nguyen, A. Kuliopulos, R. A. Graham and L. Covic, Cancer Res.,2006, 66, 2658–65.

7 Z. Salah, M. Maoz, E. Pokroy, M. Lotem, R. Bar-Shavit andB. Uziely, Mol. Cancer Res., 2007, 5, 229–40.

8 G. J. Villares, M. Zigler, H. Wang, V. O. Melnikova, H. Wu,R. Friedman, M. C. Leslie, P. E. Vivas-Mejia, G. Lopez-Berestein,A. K. Sood and M. Bar-Eli1, Cancer Res., 2008, 68, 9078–86.

9 M. E. Davis, J. E. Zuckerman, C. H. Choi, D. Seligson, A. Tolcher,C. A. Alabi, Y. Yen, J. D. Heidel and A. Ribas, Nature, 2010, 464,1067–70.

10 K. A. Whitehead, R. Langer and D. G. Anderson, Nat. Rev. DrugDiscovery, 2009, 8, 129–38.

11 I. M. Verma and N. Somia, Nature, 1997, 389, 239–42.12 V. V. Kumar, R. S. Singh and A. Chaudhuri,Curr. Med. Chem., 2003,

10, 1297–306.13 J. S. Zhang, F. Liu and L. Huang, Adv. Drug Delivery Rev., 2005, 57,

689–98.

3932 | Nanoscale, 2011, 3, 3923–3932

14 Y. Pan, S. Neuss, A. Leifert, M. Fischler, F. Wen, U. Simon,G. Schmid, W. Brandau and W. Jahnen-Dechent, Small, 2007, 3,1941–9.

15 A. M. Alkilany, P. K. Nagaria, M. D. Wyatt and C. J. Murphy,Langmuir, 2010, 26, 9328–33.

16 W. J. Song, J. Z. Du, T. M. Sun, P. Z. Zhang and J. Wang, Small,2010, 6, 239–46.

17 A. K. Varkouhi, R. J. Verheul, R. M. Schiffelers, T. Lammers,G. Storm and W. E. Hennink, Bioconjugate Chem., 2010, 21, 2339–46.

18 A. P. Perez, E. L. Romero and M. J. Morilla, Int. J. Pharm., 2009,380, 189–200.

19 P. C. Patel, D. A. Giljohann, W. L. Daniel, D. Zheng, A. E. Prigodichand C. A. Mirkin, Bioconjugate Chem., 2010, 21, 2250–6.

20 X. Huang, S. Neretina and M. A. El-Sayed, Adv. Mater., 2009, 21,4880–910.

21 A. K. Salem, P. C. Searson and K. W. Leong, Nat. Mater., 2003, 2,668–71.

22 H. C. Huang, S. Barua, D. B. Kay and K. Rege, ACS Nano, 2009, 3,2941–52.

23 A. Wijaya, S. B. Schaffer, I. G. Pallares and K. Hamad-Schifferli,ACS Nano, 2009, 3, 80–86.

24 H. Takahashi, Y. Niidome and S. Yamada, Chem. Commun., 2005,(17), 2247–9, DOI: 10.1039/b500337g.

25 C. C. Chen, Y. P. Lin, C. W. Wang, H. C. Tzeng, C. H. Wu,Y. C. Chen, C. P. Chen, L. C. Chen and Y. C. Wu, J. Am. Chem.Soc., 2006, 128, 3709–15.

26 A. C. Bonoiu, S. D. Mahajan, H. Ding, I. Roy, K. T. Yong,R. Kumar, R. Hu, E. J. Bergeya, S. A. Schwartzb andP. N. Prasada, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 5546–50.

27 K. V. Chakravarthy, A. C. Bonoiu, W. G. Davis, P. Ranjan, H. Ding,R. Hu, J. B. Bowzardb, E. J. Bergeyd, J. M. Katzb, P. R. Knight,S. Sambharab and P. N. Prasad, Proc. Natl. Acad. Sci. U. S. A.,2010, 107, 10172–7.

28 L. Feng, X. Wu, L. Ren, Y. Xiang, W. He, K. Zhang, W. Zhou andS. Xie, Chem.–Eur. J., 2008, 14, 9764–71.

29 Y. Qiu, Y. Liu, L. Wang, L. Xu, R. Bai, Y. Ji, X. Wu, Y. Zhao, Y. Liand C. Chen, Biomaterials, 2010, 31, 7606–19.

30 Q. D. Nguyen, O. De Wever, E. Bruyneel, A. Hendrix, W. Z. Xie,A. Lombet, M. Leibl, M. Mareel, F. Gieseler, M. Bracke andC. Gespach, Oncogene, 2005, 24, 8240–51.

31 Y. Zhu, Z. Sun, Q. Han, L. Liao, J. Wang, C. Bian, J. Li, X. Yan,Y. Liu, C. Shao and R. C. Zhao, Leukemia, 2009, 23, 925–33.

32 A. M. Alkilany, P. K. Nagaria, C. R. Hexel, T. J. Shaw, C. J. Murphyand M. D. Wyatt, Small, 2009, 5, 701–8.

33 B. D. Chithrani, A. A. Ghazani and W. C. Chan, Nano Lett., 2006, 6,662–8.

34 S. Chono, S. D. Li, C. C. Conwell and L. Huang, J. ControlledRelease, 2008, 131, 64–9.

35 H. Ding, K. T. Yong, I. Roy, H. E. Pudavar, W. C. Law, E. J. Bergeyand P. N. Prasad, J. Phys. Chem. C, 2007, 111, 12552–7.

36 S. Oliveira, G. Storm and R. M. Schiffelers, J. Biomed. Biotechnol.,2006, 2006, 63675.

37 L. Wang, Y. Liu, W. Li, X. Jiang, Y. Ji, X. Wu, L. Xu, Y. Qiu,K. Zhao, T. Wei, Y. Li, Y. Zhao and C. Chen, Nano Lett., 2011,11, 772–80.

38 W. Liu, S. Sun, Z. Cao, X. Zhang, K. Yao, W. W. Lu and K. D. Luk,Biomaterials, 2005, 26, 2705–11.

39 R. G. Rayavarapu, W. Petersen, L. Hartsuiker, P. Chin, H. Janssen,F. W. van Leeuwen, C. Otto, S. Manohar and T. G. van Leeuwen,Nanotechnology, 2010, 21, 145101.

40 B. Kang, M. A. Mackey and M. A. El-Sayed, J. Am. Chem. Soc.,2010, 132, 1517–9.

This journal is ª The Royal Society of Chemistry 2011