Anal Bioanal Chem (2011) 400:535–545
DOI 10.1007/s00216-011-4770-4
ORIGINAL PAPER
PEGylated polyethyleneimine grafted silica nanoparticles:
enhanced cellular uptake and efficient siRNA delivery
Haisung Lee & Dongkyung Sung &
Murugan Veerapandian & Kyusik Yun & Soo-Won Seo
Received: 12 November 2010 / Revised: 24 January 2011 / Accepted: 3 February 2011 / Published online: 22 February 2011
# Springer-Verlag 2011
Abstract The present paper reports the utilization of hybrid
nanocomposite particles consisting of PEI25k-PEG5k copoly-
mer grafted silica nanoparticles (SiO2NPs) for enhanced cellular
uptake and siRNA delivery. High-resolution transmission
electron microscopy and dynamic light scattering measure-
ments ensured the average particle size of the final hybrid
component as 45 nm (core SiO2, 28–30 nm and shell PEI25k-
PEG5k, 12–15 nm). Surface morphology from atomic force
microscopy analysis showed the significant relationship
between the particle size and shape. 29Si and 13C cross-
polarization–magic angle spinning solid state nuclear magnetic
resonance (NMR), 1H-NMR, and Fourier transform infrared
spectroscopy were used to obtain the relevant structural
information (such as Q3, silanol; Q4, siloxane functional
Electronic supplementary material The online version of this article
(doi:10.1007/s00216-011-4770-4) contains supplementary material,
which is available to authorized users.
H. Lee and D. Sung contributed equally to this work.
H. Lee : S.-W. Seo (*)
Interdisciplinary Program of Biomedical Engineering,
School of Medicine, Sungkyunkwan University,
50 Irwon-Dong, Gangnam-Gu,
Seoul 135-710, Republic of Korea
e-mail: sswbme@skku.edu
D. Sung
Department of Life Science,
The Graduate School of Korea University,
Anam-dong, Seongbuk-Gu,
Seoul 136-701, Republic of Korea
M. Veerapandian : K. Yun (*)
College of Bionanotechnology,
Gachon BioNano Research Institute, Kyungwon University,
San 65 Bokjeong-dong, Sujeong-gu,
Seongnam City 460-701, Republic of Korea
e-mail: ykyusik@kyungwon.ac.kr
groups of SiO2NPs; resonance shifts and bending vibrations of
PEI25k, –CH2–CH2–NH–; and PEG5k, –CH2–CH2–O–) from
copolymer nanoparticle. Stable complexation of siRNA and
nanocomposite particle (wt.%:wt.%) was achieved from 1:5 to
1:15 ratio. Nanocomposite particle (N/P) ratio and siRNA
concentration determine the stability and knockdown efficiency
of the PEI25k-PEG5k-graft-SiO2NPs–siRNA complexes. It
was shown that highly positively charged (zeta potential,
+66 mV) PEI25k-PEG5k-graft-SiO2NPs result in strong
affinity with negatively charged siRNA. Confocal microscopy
showed intensified cellular uptake of siRNA into cytoplasm of
A549 cancer cell utilized for in vitro study. In conclusion, the
coherence, graft density of copolymer-SiO2NPs, and siRNA
concentration were found to strongly influence the stability,
and hence determine the knockdown efficiency, of PEI25k-
PEG5k-graft-SiO2NPs–siRNA complexes.
Keywords siRNA . PEI25k-PEG5k-graft-SiO2NPs .
Cellular transfection . Low cytotoxicity
Introduction
The novel design of a nanoparticle delivery system for
nucleic acids such as small interference RNAs (siRNAs)
has been studied due to their potential applications for in
vitro and in vivo experiments [1–4]. The protected and
successful intracellular delivery of siRNAs is the final goal
to an extensive application, using siRNAs as therapeutic
agents [5, 6]. A variety of nonviral delivery systems such as
liposomes, polymers, cationic charged peptides, and gold
nanoparticles has been utilized for enhanced intracellular
siRNAs delivery [7, 8]. There are many different accessible
methods for making the complexes, using cationic carriers
such as polymers, liposomes, cationic peptides, and hard-
536
sphere nanoparticles such as gold nanoparticles, quantum
dots, and silica nanoparticles [9–15]. Functionalized nano-
particles can be placed in the core of the complex, and
siRNAs make a position on the surface of cationic nano-
particles. This method has been used for multifunctional
nanoparticles such as siRNA delivery and other diagnostics
methods: magnetic resonance (MR) imaging and fluores-
cent imaging analysis [16–27]. Core-shell nanoparticles
with optimized size tuning and maintenance of charge
intensity are the key factors to achieve successful delivery
of siRNAs [28–30]. Recently, the functionalized silica
nanoparticles via surface modification were released upon
invention of multi-modal and smart nanoparticles capable
of searching for specific cancer cells or receptor and
delivery to targeted location [31–33]. Even though, there
are several promising results that were obtained from the
functionalized nanoparticles (such as SiO2NPs) for drug
delivery and other biological application. The choice of
better delivery system with more specific stabilizing agent
and biocompatible modification procedures is expected to
acquire the complete pharmaceutical and pharmacological
value with minimal/no toxic effects. To do so, currently, the
pharmaceutical and bioengineering scientists were involved
in finding new and innovative research schemes. Here, we
present a simple and fast method to fabricate core-shell
silica nanoparticles for siRNA delivery system. This
research is based on block-copolymer grafting method
using polyethyleneimine (PEI), polyethyleneglycol (PEG),
and epoxy functionalized silane to conjugate onto the
surface of silica nanoparticles. This hydrophilic and highly
positively charged surface modified silica nanoparticle can be
used as a drug delivery system for anti-cancer drug, anti-
biotics, and therapeutic siRNAs. Techniques such as cross-
polarization–magic angle spinning (CP-MAS) solid nuclear
magnetic resonance (NMR), 1H-NMR, high-resolution trans-
mission electron microscopy (HR-TEM), Bio-atomic force
microscopy (AFM), and dynamic light scattering (DLS) have
been used to investigate the different properties of the
nanostructures. Furthermore, the significance of enhanced
cellular uptake, effective siRNA delivery with low cytotox-
icity, and higher knockdown efficiency of core-shell silica
nanoparticles were compared with the PEI cationic polymer.
Materials and methods
General
Dichloromethane (DCM), NaHCO3, trimethoxy (3-(oxiran-2-
ylmethoxy)propyl)silane (3-GPTMS), silica nanoparticles
(LUDOX AS-40 Colloidal Silica), sodium sulfate (Na2SO4),
dimethylsulfoxide (DMSO), rhodamine isothiocyanate (RITC),
DSS [3-(trimethylsilyl)-1-propane sulfonic acid], sodium salt
H. Lee et al.
97%, ethyl acetate (EA), hexane, and branched polyethylenei-
mine 25,000 (PEI25k) were obtained from Sigma. Mono-N-
hydroxysuccinyl-polyethyleneglycol 5,000 (NHS-PEG5k) was
purchased from Sun Bio Chemical company (South Korea).
All chemicals were purchased from commercial sources.
Synthesis of derivative 1 (NHS-PEG5k and PEI25k)
A solution of NHS-PEG5k (50 mg, 10 mM) in anhydrous
DCM (80 ml) was added to a solution of PEI25k (250 mg,
10 mM) in anhydrous DCM (80 ml) under 0 °C using ice-
bath for cooling down. The mixture was stirred at room
temperature then naturally increased from 0 °C for 12 h.
For checking of reaction progress, the TLC was used with
eluent condition MeOH: DCM=1:1 (Rf value of NHS-
PEG5k (0.3∼0.4), PEI25k (0.5∼0.6), PEI25k-PEG5k mix-
ture (0.0∼0.1), and co-mixture of three (0.5∼0.6) samples
were spotted and stained with iodine after 12 h of reaction
period). All the compounds showed a tailed spot because of
their polar nature. It was finally washed with DCM three
times. The solvent was evaporated with Speed-Vac and
freeze-dried, and the crude compound was kept in −20 °C.
The crude solid material was obtained after the evaporation
of the solvent. It was finally purified by Sephadex LH20.
Synthesis of derivative 2 (PEI25k-PEG5k and 3-GPTMS)
A solution of PEI25k-PEG5k (100 mg) in DCM (80 mL) was
added to a solution of 3-GPTMS (0.80 mL, 4.6 mM) in the
same solvent (20 mL) under 0 °C using ice-bath for cooling
down. The mixture was stirred at room temperature for 12 h.
Synthesis of derivative 3 (PEI25k-PEG5k-graft-SiO2NPs)
Fifty-three milligrams of derivative 2 was placed in a round
bottom flask and then dissolved with 1.25 mL of ethanol,
1.25 mL of water, and 1.25 mL of acetic acid. One hundred
twenty-five microliters of an aqueous suspension of
SiO2NPs was added. The reaction mixture was warmed at
80 °C under stirring for 48 h. After this time, ethanol was
evaporated under reduced pressure, and solid NaHCO3 was
added to the suspension to reach a pH value between 7 and
8. The precipitate was filtered and washed with borate
buffer (pH 9.5, 5×2 mL) and with water (5×2 mL). The
solid was dried under vacuum and redissolved with 100 mL
of DCM. The organic solution was washed with water
(100 mL) and dried over Na2SO4.
Synthesis of derivative 4 (RITC-labeled
PEI25k-PEG5k-graft-SiO2NPs)
One hundred micrograms per milliliter colloidal solution of
PEI25k-PEG5k-graft-SiO2NPs was placed in a V-shaped
PEGylated polyethyleneimine grafted silica nanoparticles 537

small volume vial and then added with 5 mM RITC
solution in DMSO under pH 7.4∼8.0. The reaction mixture
was stirred at room temperature for 12 h under dark room
condition. After this time, the mixture was washed with
DCM twice, and aqueous colloidal solution was separated
by dialysis for 2 days.
Characterization of size and zeta potential of silica
nanoparticles
Size and zeta potential were analyzed using Malvern
Zetasizer Nano ZS series with a laser at 633 nm with a
90° scattering angle. The concentration of analytical
samples was 0.1 mg/mL.
AFM analysis
The AFM images were obtained by Nanowizard II (JPK;
Germany). It was equipped with atomic lattice resolution on
inverted microscope (<0.055 nm RMS z noise level). The
cantilever used in our experiment was Soft Tapping Mode
probes.
Solid-state NMR analysis
Solid-state 29Si CP-MAS NMR analysis was performed
using a Bruker Avance 400 MHz spectrometer operating at
79.545 MHz. The samples were located in 4-mm ZrO2
rotors. Magic angle spinning was carried out at a spinning
rate of 7 kHz. The spectrum width was 23.8 kHz; delay or
repetition time was 5.0 s. Contact time was performed for
5 ms. The 29Si spectrum was calibrated with the signal from
using external standard DSS 0.00 ppm. In case of 13C solid-
state NMR experiments, it has been done with the same
NMR instrument such as 29Si CP-MAS NMR analysis
operating at 100.690 MHz; 4-mm ZrO2 rotors were used,
and magic angle spinning was carried out at a spinning rate
of 7 kHz. Two milliseconds contact time, 30.3 kHz spectrum
width, and 5.0 s delay or repetition time were used.
Cytotoxicity of silica nanoparticles
Human alveolar basal epithelial cell (A549 cells) was
provided by the Korea Cell Line Bank. The A549 cells
were used for the in vitro study of the nanoparticles
toxicology. At day 1, 1.0×104 cells were placed in each
well of a 96-well plate in 100 μL of Dulbecco's modified
Eagle's medium containing 10% fetal bovine serum (FBS,
purchased from GIBCO) and cultured for 24 h at 37 °C.
Then, 100 μL of the colloidal solution of nanoparticles in
pH 7.0 PBS was added to the cell culture plate for
incubation. The final concentrations of nanoparticles ranged
from 25×10−3 to 0.10 mg/mL. The nanoparticles were pure
PEI25k-PEG5k-graft-SiO2NPs containing no fluorescent
dye molecules to stay away from interferences between
the nanoparticles and the tetrazolium salt of a detection
reagent. A prepared dye XTT solution was added to the
plate wells, and then, the cells were incubated at 37 °C for

H O
N
NH2 N CH3
H O HO OH
n
N N
H
N HO SiO2 OH
N N HO OH
H n

N O
O Si CH3
EtOH, H 2O
H2N N
H
CH3
OH O O AcOH @ 80 Celcuis for 48hrs

CH3

PEG5K-PEI 25K-silane

O H
N
H3C N H2N
O H
n
H
N N N
N N
n H

O N
2 Si
O N NH2 PEG5K-PEI 25K layer
H
O
OH

PEG5K-PEI 25K-silica nanoparticle 2

Fig. 1 Synthetic scheme of PEG5k-PEI25k-graft-SiO2NPs

538 H. Lee et al.

4 h. In this period, the living cells converted the tetrazolium
component of the dye solution into a formazan product that
showed absorption at 470 nm in a UV–Vis spectrum. The
formazan product was solubilized, and the absorbance at
470 nm was measured using a 96-well plate reader. The
experiment was repeated three times to get the average results.
Agarose gel electrophoresis
The complexes of siRNA and PEI25k-PEG5k-graft-
SiO2NPs were formulated with applicable N/P ratios.
Afterwards, the complexes were applied to 1% (w/v)
agarose gel contacting 0.5 μg/mL ethidium bromide (EtBr).

Fig. 2 a, b High-resolution
TEM images of PEI25k-
PEG5k-graft-SiO2NPs core-
shell nanoparticles. c DLS size
analysis and d DLS zeta poten-
tial analysis
PEGylated polyethyleneimine grafted silica nanoparticles 539

Table 1 Energy-dispersive X-ray spectroscopy (EDS) analysis of PEI25k-PEG5k-graft-SiO2NPs

Element Peak area Area sigma K factor Abs corrn. wt.% wt.% sigma Atomic %

Ck 355 54 2.208 1.000 15.28 2.07 23.11

Nk 177 54 2.965 1.000 10.23 2.83 13.27
Ok 893 59 1.810 1.000 31.54 1.95 35.83
Sik 2,202 82 1.000 1.000 42.95 2.11 27.79
Total 100
14
Si and 8 O atomic composition contributed more than half of the total atomic composition of PEI25k-PEG5k-graft-SiO2NPs. Then, 6 C and 7 N
contribution in total atomic composition was about half a portion. 6 C fraction in EDS was bigger than 7 N for the reason that hydrocarbon was included in
PEI as well as PEG. Furthermore, 8 O atomic composition was similar with 14 Si atomic composition because 8 O atomic composition was located in
SiO2NPs and PEG5k as well

RNase protection assay Cells were routinely cultivated at 37 °C in a humidified

siRNA (2 μg) and its complexes with PEI25k-PEG5k-graft-
SiO2NPs were prepared just before the experiments. After
the preparation of the complexes, each sample was incubated
with 10 U of RNase ONE ribonuclease (Promega) for 1 h at
37 °C. Nucleic acid degradation was detected by gel
electrophoresis. The levels of the remaining siRNAs were
quantified by Quantity One software (Bio-Rad, USA).
Confocal laser scanning microscopic observation
Human alveolar basal epithelial cell (A549 cells) was
provided by the Korea Cell Line Bank. A549 cells were
cultured in RPMI 1640 medium (Invitrogen, USA) supple-
mented with 10% heat-inactivated FBS and 1% antibiotics.
atmosphere of 5% CO2. A549 cells were seeded in a 12-
well culture plate at a density of 2×105 cells per well and
allowed to attach overnight. Then, incubation medium was
replaced with fresh media. The complex was prepared with
PEI25k-PEG5k-graft-SiO2NPs and FITC-labeled siRNA
via different conditions. The nucleus was stained by DAPI
(blue), and the complex was represented with green through
FAM-labeled RNA-PEI25k-PEG5k-graft-SiO2NPs, and
then the membrane was colored with wheat germ aggluti-
nin; transfection was carried out by adding the prepared
complex to the wells in a drop-wise manner. After 4 h
incubation, the medium was exchanged again with fresh
medium, and transfected cells were incubated for 24 h. A
series of images were obtained by using a confocal laser
scanning microscope (Bio-Rad, USA).

Fig. 3 29Si CP MAS NMR

spectrum of bare SiO2NPs, ami-
nopropyl functionalized
SiO2NPs, and PEI25k-PEG5k-
graft-SiO2NPs (a); 13C CP MAS
NMR spectrum of aminopropyl
functionalized SiO2NPs and
PEI25k-PEG5k-graft-SiO2NPs,
as well as SiO2NPs (b); and
600-MHz 1H-NMR NMR spec-
trum of PEI25k, PEG5k, as well
as PEI25k-PEG5k
copolymer (c)
540
Fig. 4 Cytotoxicity of PEI25k-PEG5k-graft-SiO2NPs in A549 lung
cancer cell with XTT assay
Reverse transcription-PCR method
Twenty-four-hour post-transfection with GAPDH, total RNA
in the sample was extracted by using a RNA Trizol according
to the manufacturer's protocol (Invitrogen). Total RNA
concentration was measured by spectrophotometer (Nano-
drop) at 260 nm. RNA was used to produce cDNA with an
iScriptTM cDNA synthesis kit (Bio-Rad). Polymerase chain
reaction (PCR) primers human glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) and actin were synthesized from
Bioneer Inc. The sequence of primers used was as follows:
human GAPDH (sense, GGTCGGAGTCAACGGATTTG;
antisense, CCTCCGACGCCTGCTTCACC) and actin (sense,
AGGGAAATCGTGCGTGACAT; antisense, CATCTGCTG-
GAAGGTGGACA). For each reaction, 1 μL of cDNA was
placed in a 19-μL reaction mixture PCR Master Mix
(Bioneer, Korea) and 0.5 μM of each primer. All amplification
reactions were performed in T1 thermocycler. PCR products
were run on a 1.0% agarose gel electrophoresis and were
visualized by ethidium bromide and scanned by a Gel Doc
Fig. 5 Gel retardation assay of the different N/P ratio complexes and
RNase protection assay. a Agarose gel electrophoresis of PEI25k-
PEG5k-graft-SiO2NPs–RNA complex at different N/P ratio. Com-
plexes were prepared at N/P ratios range from 1 to 30. b After 1 h
H. Lee et al.
2000 analyzer (Bio-Rad, USA). The level of each gene
expression was semi-quantified by densitometric analysis
using the software. Relative expression levels were estimated
by the density ratio of human GAPDH to actin.
Results
The general procedure to fabricate core-shell SiO2NPs has
been performed via three simple steps. Initially, we
synthesized the block-copolymer consisting of PEI25k
and NHS-PEG5k (see Electronic Supplementary Material
Fig. S1); later, this copolymer is modified with the 3-
GPTMS (see Electronic Supplementary Material Fig. S2).
Then, the final PEI25k-PEG5k-graft-SiO2NPs have been
fabricated by conjugation of the copolymer (PEI25k-
PEG5k), which has been modified with a silane from 3-
GPTMS, and this silane reacts with silica nanoparticles (see
Fig. 1). The synthesized PEI25k-PEG5k-graft-SiO2NPs
showed an average particle size of 45 nm by TEM and
DLS (Fig. 2a–c). The core part of the nanoparticles showed
an average particle size of 28–30 nm, and the shell part
(PEI25k-PEG5k block-copolymer) of the particle was 12–
15-nm thick as measured from TEM analysis (Fig. 2b).
PEGylated-PEI-SiO2NPs were analyzed with Nano ZS zeta
potential analyzer in neutral pH (7.0). The zeta potential
became more positive such as +66 mV in pH 7 (Fig. 2d). For
the detailed atomic analysis of the PEGylated-PEI polymer
shell layer and silica core particle, energy-dispersive X-ray
spectroscopy atomic analysis has been done. Table 1
demonstrated the atomic composition (wt.%) of 14Si, 8O,
and 6C in the part of core-shell silica nanoparticles. The
morphology of PEI25k-PEG5k-graft-SiO2NPs was further
observed from AFM. Figure S3 in the Electronic Supple-
mentary Material shows the two- and three-dimensional
topography of PEI25k-PEG5k-graft-SiO2NPs. From the
AFM study, the size of the particles was measured to be
around 40–50 nm; this is in good relation with the other
measurements.
incubation with RNase and RNA-PEI25k-PEG5k-graft-SiO2NPs, low
N/P ratio siRNA complex was degraded by RNase, but high N/P ratio
siRNA was protected by RNase
PEGylated polyethyleneimine grafted silica nanoparticles 541

c 100 d
110
90
100
80

Intensity of GAPDH
Intensity of GAPDH

90
70
80
60
70

50 60

40 50

30 40

0 50 100 150 200 250 300 NC RNA 1 5 7 10

Dose of siRNA (pmol) N/P ratio

Fig. 6 Semiquantitative RT-PCR analysis for the GAPDH mRNA knockdown. a, c A549 cells transfected naked RNA and complex prepared with
different siRNAs concentrations (25, 50, 100, 200, and 300 pmol). b, d Complexes were prepared at N/P ratios range from 1 to 10

Figure 3 shows relevant structural information with
solid-state NMR spectrum analysis in 29Si, 13C with CP/
MAS NMR (Fig. 3a, b), and 1H-NMR (Fig. 3c). Solid-
state NMR has been used as one of the key analytical
methods to investigate the molecular structures and
dynamics of the surface-bound and intercalated organic
species [34, 35]. Owing to its easy interpretation of
spectra, 29Si CP-MAS NMR is widely used to study the
surface analysis of structural information from silica. The
relation between Si chemical shifts and corresponding
assigned structures onto the surface of silica is simple to
analyze. Figure 3a shows that 29Si CP-MAS NMR
spectrum of the different organosilanes and organic
functional groups in PEI25k-PEG5k copolymer materials
Fig. 7 Confocal laser microsco-
py images of A549 cells trans-
fected with PEI25k-PEG5k-
graft-SiO2NPs. a Transfection
after 2 h incubation. b Trans-
fection after 4 h incubation
grafted onto the surface of SiO2NPs is evaluated to that of
the unmodified SiO2NPs and 3-aminopropyltriethoxysilane
(3-APTES)-modified SiO2NPs. The 29Si CP-MAS NMR
spectrum of the unmodified silica nanoparticles shows three
peaks at −100.2 and −108.9 ppm, which are assigned to be
Q3, free silanols, and Q4, siloxane functional groups. In
the 29Si CP-MAS NMR spectra of the modified silica
nanoparticles, two functional groups of peaks are detected
at all times. Comparing to the spectrum of unmodified
SiO2NPs, Q3 peak is enhanced, and Q4 peaks are reduced.
The reaction of SiO2NPs surface with a triethoxysilyl
functional group is able to show the way to three available
types of new chemical structures onto the SiO2NPs. The
T3 and T4 structures are the product of each reaction
542
individually, two and three ethoxy groups of the same
silane molecule with the silanols at the surface of
SiO2NPs. It was noticed that a nice correspondence is
located between 29Si CP-MAS NMR analysis and 13C CP-
MAS NMR analysis (Fig. 3a, b).
1
H-NMR spectroscopy was used to elucidate the
structural information from the PEI25k, PEG5k, and
PEGylated-PEI. The sharp peak at around 3.6 ppm is
assigned to be PEG (–CH2–CH2–O–), and proton shift
located from 2.0 to 2.6 ppm is found to be associated with
PEI (–CH2–CH2–NH2–). From the proton signals identi-
fied from the current experiment, it is clear that the
successful copolymerization exists between the
PEGylated-PEI groups (Fig. 3c). The analysis of Fourier
transform infrared spectrum about unmodified SiO2NPs,
amine functionalized SiO2NPs, and PEI25k-PEG5k-graft-
SiO2NPs was performed to obtain the stretching and
bending vibrations before and after surface modification
of SiO 2NPs (see Electronic Supplementary Material
Fig. S4). Our PEI25k-PEG5k-graft-SiO2NPs have been
represented with the competitive low cytotoxicity in XTT
assay with A549 lung cancer cell. We have performed the
Fig. 8 Confocal laser microsco-
py images of A549 cells trans-
fected with PEI25k-PEG5k-
graft-SiO2NPs-siRNA complex.
Complexes were prepared with
200 nM FAM-oligo at N/P ratio
of 1:10 and transfected into the
A549 cells for 4 h. Blue nucleus
(DAPI), green FAM-labeled
siRNA-PEI25k-PEG5k-graft-
SiO2NPs complexes, red mem-
brane (wheat germ agglutinin,
Alexa Fluor 594 conjugate).
Magnification, ×1,000
H. Lee et al.
XTT assay with different concentrations from 25 to 100 μg/
mL. As we can see in Fig. 4, the PEI25k-PEG5k-graft-
SiO2NPs showed a significant low cytotoxicity in XTT
assay. Thus, the SiO2NPs could be used as a possible
delivery system for siRNAs. In order exploit the cellular
uptake and its efficiency towards the siRNA delivery, an in
vitro bioanalysis is done. The cell transfection efficiency has
been tested through conjugated RITC red fluorescence dye
on the PEI25k-PEG5k-graft-SiO2NPs with further purifica-
tion process: dialysis (MWCO, 50 K) for 2 days. For
confirmation of the efficient transfection and delivery, we
tested the gel retardation assay with different siRNA/particle
ratios in weight percentage. The ratio of complex was
increased gradually from 1 to 30 (Fig. 5a). The ratio 1:1=
siRNA:particles (wt.%/wt.%) seems to be an unstable
complex in solution and has no significant protection of
siRNA. In the case of 1:3 ratio complexes, it was not clearly
produced, which was characterized with tailed band in
agarose gel. The complex was formulated strongly from
the 1:5 to 1:15 ratio. Figure 5b shows the low ratio complex
which seems to be not protecting the RNase. In parallel, 1:5
ratio complexes could protect from RNase.
PEGylated polyethyleneimine grafted silica nanoparticles 543

In semiquantitative RT-PCR analysis for the GAPDH

mRNA knockdown, A549 cells were transfected with
naked RNA, and the complex was prepared with different
siRNA concentrations (25, 50, 100, 200, and 300 pM;
Fig. 6a, c). The knockdown efficiency was increased
through the change of siRNAs concentrations. In Fig. 6a, c,
there is not much difference between 0- and 50-pM
ranges. It was the high knockdown efficiency in the
300 pM of siRNA. Moreover, Fig. 6b, d showed that the
1:10 ratio complex worked properly to knockdown
GAPDH comparing beta-actin. Therefore, we optimized
the suitable N/P ratio (wt.%/wt.%) and siRNA concentra-
tion. Figure 7 ensures that the transfection of the particle
itself with red fluorescence and the nucleus was stained
with DAPI (blue). The silica particles during the 4-h incubation
were transfected better than that of the 2-h incubation. We
confirmed that our PEI25k-PEG5k-graft-SiO2NPs did have the
ability of transfection into A549 lung cancer cell without
difficulty.
Figure 8 describes that the cell transfection experiment
was performed via confocal laser scanning microscopy. The
cell transfection experiment has been performed with A549
cell by DAPI staining in nucleus into blue color and the
complex between FAM-labeled siRNA and PEI25k-
PEG5k-graft-SiO2NPs into green color; the membrane with
wheat germ agglutinin, Alexa Fluor 594 conjugate, was
colored red. The complex of FAM-labeled siRNA and
PEI25k-PEG5k-graft-SiO2NPs was transfected suitably in
the A549 lung cancer cell. The PEI25k-PEG5k block-
copolymer had been present that provided a significant
protection from RNase and knockdown efficiency in a few
research papers [36–40]. In relation to that, our PEI25k-
PEG5k-graft-SiO2NPs had shown a considerable efficiency
of protection and knockdown. From Fig. 9, the cytotoxicity
of PEI25k and PEI25k-PEG5k-graft-SiO2NPs provided the
merit of our core-shell SiO2NPs for siRNA delivery system Fig. 9 Cell cytotoxicity and knockdown efficiency between PEI25k
(Fig. 9a). In knockdown efficiency, our PEI25k-PEG5k- polymer and PEI25k-PEG5k-graft-SiO2NPs. a Cytotoxicity experi-
ment has been performed in fixed concentration (100 μg/mL of each
graft-SiO2NPs could have a priority to deliver siRNA under
delivery system). b GAPDH knockdown experiment has been done in
serum condition comparing PEI cationic polymer with 200 pM siRNA concentration, 1:10 N/P ratio condition. Each symbol
itself. The PEI25k-PEG5k-graft-SiO2NPs did knock down (triangle, rectangle, and circle, and so on) indicates each experiment.
GAPDH, about half of them; PEI25k only did knock down Experiment A (cytotoxicity) has been repeated seven times, and
experiment B (comparison of GADPH knockdown efficiency) has
GAPDH with about 20% in same condition (Fig. 9b). been repeated three times

Discussion MR imaging through paramagnetic oxide nanoparticles for

We show that the SiO2NPs can be modified to deliver
siRNA into cancer cell line with a significant low
cytotoxicity. The functionalization of SiO2NP surface has
been demonstrated using a covalent bonding between
PEGlyated-PEI copolymer and epoxy group. This inexpen-
sive, simple fabrication method can be utilized for multi-
functionality, which may conjugate with a fluorescence dye,
diagnostics applications under in vitro or in vivo conditions
as reported before individually [41].
The highly positive-charged PEGylated-PEI copolymer
grafted SiO2NPs can provide a chance for enhancing the
interaction with negative charge surface of cell membrane.
The advantages of using cationic PEI polymer for gene
delivery stay on the simple experimental condition which
PEI can be combined with siRNA or DNA using charge–
544
charge interaction [42–45]. There were many mechanisms
explained on the release of siRNA from delivery materials.
In the present study, it is probably released from the
nanocomposite based on the phenomena of ligand exchange
effect/charge effect [45].
The high molecular weight and charged polymer
normally have a high cytotoxicity. The researches of
reducing the cytotoxicity of PEI polymer have been studied
using anhydride, aldehyde moieties to neutralize primary
amine groups, and other conjugation of low molecular
weight polymers. For example, partial acetylation of PEI
amino groups was shown to enhance gene transfer activity
[46]. N-Acetylation of PEI lowered its toxicity [47]. Also,
conjugation of acetate, butanoate, and hexanoate to
branched PEI at low degree of substitution (below 25%)
resulted in moderate improvement of transfection activity
as demonstrated by Putnam et al. [48]. Petersen et al.
reported that the larger molecular weight of PEG polymers
(>5 kDa) can influence the surface charge, cytotoxicity, and
aggregation of PEI polymers [49]. Similar to other study,
we have used simple PEGylation onto PEI polymer using
NHS-PEG 5 kDa, and then PEGylated PEI copolymer has
been grafted on the SiO2NPs using epoxy silane compound.
The two combinations significantly reduced the cytotox-
icity of PEI25k polymer, which is observed from the XTT
cytotoxicity assay comparing to only PEI polymer. For
instance, it is reported that the hypothetical “proton sponge”
mechanism of PEI on modification with PEG synergisti-
cally improves the biodistribution and pharmacokinetics of
their complexes [50]. Neu et al. demonstrated that usage of
endosome-disrupting polymers such as PEG grafted and
hyperbranched polyethyleneimine could allow moving
across cell membranes via endocytosis and disrupting
intracellular organelles using a proton-sponge effect [51].
Additionally, this endosomolytic effect is believed to arise
from the large number of amine groups on each PEI
polymer, leading to proton absorption in acid organelles
and an osmotic pressure buildup across the organelle
membrane. This osmotic pressure causes a disruption of
the acidic endosomes and a release of the trapped materials
into the cytoplasm [52]. Finally, the PEI-PEG-graft-
SiO2NPs could allow the high cellular transfection with a
low cytotoxicity, safe protection, and delivery of siRNA
using a proton-sponge effect, steric interaction, electrostatic
interaction, and polymer bridging interactions.
Conclusion
From the current study, we have examined the simple and
facile approach to hybridize the inorganic SiO2NPs and
organic copolymer PEI25k-PEG5k. As obtained, nano-
hybrid was successfully evaluated for siRNA delivery.
H. Lee et al.
Optimal nanoparticle size (∼45 nm), highly positive-charged
(+66 mV) surface, and RNAs protection ability directed the
fabricated nano-hybrid for enhanced cellular uptake and
effective siRNA delivery in A549 cancer cell, as studied
under in vitro condition. In the complexation between
PEI25k-PEG5k-graft-SiO2NPs and siRNAs, it was shown
that N/P ratio could verify the complexation efficiency of
PEI25k-PEG5k-graft-SiO2NPs. The complete complexation
was reached at N/P ratio (1:10). Moreover, N/P ratio 1:10
complex did protect RNase well than other N/P ratio
complexes. In conclusion, the optimized PEI25k-PEG5k-
graft-SiO2NPs can have the ability to modify and enhance
the stability of siRNA. Furthermore, our PEI25k-PEG5k-
graft-SiO2NPs promise to apply as an effective siRNA
delivery system with low cytotoxicity compared with
PEI25k cationic polymer. The additional optimization is also
promising to enhance the biocompatibility of modified-
PEI25k-PEG5k-graft-SiO2NPs for in vivo applications. A
detailed study is in progress to unravel the underlying
mechanism of dual function of this hybrid nanoparticle for
both MR imaging detection of single cell and siRNA
transfection using synthesized T1 weight contrast agent.
Acknowledgements This study was supported by a grant of the
Ministry of Health and Welfare (A040041) and Samsung Biomedical
Research Institute, Republic of Korea (PB00021). We thank Ms.
Yunhee Kim for solid NMR spectroscopic analysis in NICEM, SNU,
and Ms. Youngshin Yoo for HR-TEM analysis in SKKU.
References
1. Ulbrich W, Lamprecht A (2010) J R Soc Interface 7:55–66
2. Green JJ, Zhou BY, Mitalipova MM, Beard C, Langer R, Jaenisch
R (2008) Nano Lett 8:3126–3130
3. Allouche J, Boissiere M, Helary C, Livage J, Coradin T (2006) J
Mater Chem 16:3120–3125
4. Yu YY, Chen CY, Chen WC (2003) Polymer 44:593–601
5. Borrego T, Andrade M, Pinto ML, Silva AR, Carvalho AP, Rocha
J, Cristina F, Joao P (2010) J Colloid Interf Sci 344:603–610
6. Kleemann E, Neu M, Jekel N, Fink L, Schmehl T, Gessler T,
Seeger W, Kissel T (2005) J Control Release 109:299–316
7. Burke RS, Pun SH (2008) Bioconjugate Chem 19:693–704
8. Ashtari P, He X, Wang K, Gong P (2005) Talanta 67:548–554
9. Jere D, Jiang HL, Arote R, Kim YK, Choi YJ, Cho MH, Akaike
T, Cho CS (2009) Expert Opin Drug Deliv 6:827–834
10. Fuller JE, Zugates GT, Ferreira LS, Ow HS, Nguyen NN, Wiesner
UB, Robert SL (2008) Biomaterials 29:1526–1532
11. Choi SJ, Oh JM, Choy JH (2009) J Inorg Biochem 103:463–471
12. Wang H, Yang R, Yang L, Tan W (2009) ACS Nano 3:2451–2460
13. Elbakry A, Zaky A, Liebl R, Rachel R, Goepferich A, Breunig M
(2009) Nano Lett 9:2059–2064
14. Bartlett DW, Su H, Hildebrandt IJ, Weber WA, Davis ME (2007)
Proc Natl Acad Sci USA 104:15549–15554
15. Merkel OM, Livrizzi D, Pfestroff A, Schurrat T, Behe M, Kissel T
(2009) Bioconjugate Chem 20:174–182
16. Christensen LV, Chang CW, Yockman JW, Conners R, Jackson H,
Zhong Z, Jan F, David AB, Sung WK (2007) J Control Release
118:254–261
PEGylated polyethyleneimine grafted silica nanoparticles
17. Meade BR, Dowdy SF (2008) Adv Drug Deliv Rev 60:530–536
18. Yoon TJ, Kim JS, Kim BG, Yu KN, Cho MH, Lee JK (2005)
Angew Chem Int Ed 44:1068–1071
19. Bartlett DW, Davis ME (2008) Biotechnol Bioeng 99:975–985
20. Akhtar S, Benter IF (2007) J Clin Invest 117:3623–3632
21. Davis ME (2009) Mol Pharmaceut 6:659–668
22. Lipski AM, Pino CJ, Haselton FR, Chen IW, Shastri VP (2008)
Biomaterials 29:3836–3846
23. Xu Y, Li Q (2007) Clin Chem 53:1503–1510
24. Feng L, Wang Y, Wang N, Ma Y (2009) Polym Bull 63:313–327
25. Joubert M, Delaite C, Bourgeat-Lami B, Dumas P (2005)
Macromol Rapid Commun 26:602–607
26. Estévez MC, O’donoghue MB, Chen X, Tan W (2009) Nano Res
2:448–461
27. Graf C, Vossen DLJ, Imhof A, Blaaderen AV (2003) Langmuir
19:6693–6700
28. An Y, Chen M, Xue Q, Liu W (2007) J Colloid Interf Sci
311:507–513
29. Park KM, Kang HC, Cho JK, Chung IJ, Cho SH, Bae YH, Kun N
(2009) Biomaterials 30:2642–2652
30. Salon MC, Gerbaud G, Abdelmouleh M, Bruzzese C, Boufi S,
Belgacem MN (2007) Magn Reson Chem 45:473–483
31. Brannon-Peppas L, Blanchette JO (2004) Adv Drug Deliv Rev
56:1649–1659
32. Kobayashi S, Nakase I, Kawabata N, Yu HH, Pujals S, Imanishi
M, Ernest G, Shiroh F (2009) Bioconjugate Chem 20:953–959
33. Lessard-Viger M, Rioux M, Rainville L, Boudreau D (2009) Nano
Lett 9(8):3066–3071
34. Mijatovic WH, Binder HG (2000) Mikrochim Acta 133:175–181
35. Takayama K, Tadokoro A, Pujals S, Nakase I, Giralt E, Futaki S
(2009) Bioconjugate Chem 20:249–257
36. Mao S, Neu M, Germershaus O, Merkel O, Sitterberg J, Bakowsky
U, Thomas K (2006) Bioconjugate Chem 17:1209–1218
545
37. Petersen H, Petra MF, Dagmar F, Thomas K (2002) Macro-
molecules 35:6867–6874
38. Nguyen HK, Lemieux P, Vinogradov SV, Gebhart CL, Guérin L,
Paradis G, Bronich TK, Alakhov VY, Kabanov AV (2000) Gene
Ther 7:126–138
39. Thomas M, Lu JJ, Ge Q, Zhang C, Chen J, Klibanov AM (2005)
Proc Natl Acad Sci U S A 102:5679–5684
40. Urban-Klein B, Werth S, Abuharbeid S, Czubayko F, Aigner A
(2005) Gene Ther 12:461–466
41. Liong M, Lu J, Kovochich M, Xia T, Ruehm SG, Nel AE,
Tamanoi F, Zink JI (2008) ACS Nano 2:889–896
42. Brus C, Petersen H, Aigner A, Czubayko F, Kissel T (2004)
Bioconjugate Chem 15:677–684
43. Merdan T, Callahan J, Petersen H, Kunath K, Bakowsky U,
Kopecková P, Thomas K, Jindřich K (2003) Bioconjugate Chem
14:989–996
44. Brus C, Petersen H, Aigner A, Czubayko F, Kissel T (2004) Eur J
Pharm Biopharm 57:427–430
45. Glodde M, Sirsi SR, Lutz GJ (2006) Biomacromolecules 7:347–
356
46. Forrest ML, Meister GE, Koerber JT, Pack DW (2004) Pharm Res
21:365–371
47. Thomas M, Klibanov AM (2002) Proc Natl Acad Sci U S A
99:14640–14645
48. Doody AM, Korley JN, Dang KP, Zawaneh PN, Putnam D (2006)
J Control Release 116:227–237
49. Petersen H, Fechner PM, Martin AL, Kunath K, Stolnik S,
Roberts CJ, Dagmar F, Martyn CD, Thomas K (2002) Bioconjugate
Chem 13:845–854
50. Kunath K, von Harpe A, Petersen H, Fischer D, Voigt K, Kissel T,
Ulrich B (2002) Pharm Res 19:810–817
51. Neu M, Fischer D, Kissel T (2005) J Gene Med 7:992–1009
52. Duan H, Nie S (2007) J Am Chem Soc 129:3333–3338