Introduction of stearoyl moieties into a biocompatible cationic polyaspartamide derivative, PAsp(DET), with endosomal escaping function for enhanced siRNA-mediated gene knockdown

Article abstract of DOI:10.1016/j.jconrel.2010.03.019

Applications of siRNA for cancer therapy have been spotlighted in recent years, but the rational design of efficient siRNA delivery carriers is still controversial, especially because of possible toxicity of the carrier components. Previously, a cationic polyaspartamide derivative, poly{. N-[. N-(2-aminoethyl)-2-aminoethyl]aspartamide} (PAsp(DET)), was reported to exert high transfection efficacy for plasmid DNA with negligible cytotoxicity. However, its direct application for siRNA delivery was fairly limited due to the unstable polymer/siRNA complex formation. In this study, to overcome such instability, stearic acid as a hydrophobic moiety was conjugated to the side chain of PAsp(DET) with various substitution degrees. The stearoyl introduction contributed not only to siRNA complex formation with higher association numbers but also to complex stabilization. The obtained stearoyl PAsp(DET)/siRNA complex significantly accomplished more efficient endogenous gene (BCL-2 and VEGF) knockdown in vitro against the human pancreatic adenocarcinoma (Panc-1) cells than did the unmodified PAsp(DET) complex and commercially available reagents, probably due to the facilitated cellular internalization. This finding suggests that the hydrophobic PAsp(DET)-mediated siRNA delivery is a promising platform for in vivo siRNA delivery.

Full text of DOI:10.1016/j.jconrel.2010.03.019

Journal of Controlled Release 145 (2010) 141–148
Contents lists available at ScienceDirect
Journal of Controlled Release
journal homepage: www.elsevier.com/locate/jconrel
Introduction of stearoyl moieties into a biocompatible cationic polyaspartamide
derivative, PAsp(DET), with endosomal escaping function for enhanced
siRNA-mediated gene knockdown
Hyun Jin Kim a, Atsushi Ishii b, Kanjiro Miyata c,d, Yan Lee c,d, Shourong Wu a, Makoto Oba e,
Nobuhiro Nishiyama c,d, Kazunori Kataoka a,c,d,
⁎
a
Department of Materials Engineering, Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
b
NanoCarrier Co., Ltd., 5-4-19 Kashiwanoha, Kashiwa, Chiba 277-0882, Japan
c
Center for NanoBio Integration, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
Center for Disease Biology and Integrative Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
Department of Clinical Vascular Regeneration, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
d
e
a r t i c l e i n f o
a b s t r a c t
Article history:
Applications of siRNA for cancer therapy have been spotlighted in recent years, but the rational design of
efficient siRNA delivery carriers is still controversial, especially because of possible toxicity of the carrier
components. Previously, a cationic polyaspartamide derivative, poly{N-[N-(2-aminoethyl)-2-aminoethyl]
aspartamide} (PAsp(DET)), was reported to exert high transfection efficacy for plasmid DNA with negligible
cytotoxicity. However, its direct application for siRNA delivery was fairly limited due to the unstable polymer/
siRNA complex formation. In this study, to overcome such instability, stearic acid as a hydrophobic moiety
was conjugated to the side chain of PAsp(DET) with various substitution degrees. The stearoyl introduction
contributed not only to siRNA complex formation with higher association numbers but also to complex
stabilization. The obtained stearoyl PAsp(DET)/siRNA complex significantly accomplished more efficient
endogenous gene (BCL-2 and VEGF) knockdown in vitro against the human pancreatic adenocarcinoma (Panc-
1) cells than did the unmodified PAsp(DET) complex and commercially available reagents, probably due to the
facilitated cellular internalization. This finding suggests that the hydrophobic PAsp(DET)-mediated siRNA
delivery is a promising platform for in vivo siRNA delivery.
Received 12 January 2010
Accepted 24 March 2010
Available online 30 March 2010
Keywords:
RNAi
siRNA delivery
Polyplex
Stearoylation
Polyaspartamide
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
could be internalized into cells, the internalized siRNA would be de-
graded in lysosome. Hence, a carrier system to protect siRNA from
RNA interference (RNAi) is a powerful regulatory mechanism of
the target mRNA degradation induced by the complementary RNA
strand. Because of the selectivity of the targeted mRNA degradation as
well as the potential of silencing almost all the endogenous genes, small
interfering RNA (siRNA) has been highlighted as an attractive choice
for future therapeutics. The advent of siRNA opened a new era of therapy
for cancers, autoimmune diseases, and dominant genetic disorders by
the selective control of disease-related gene expression [1].
In spite of its promise, the in vivo therapeutic application of siRNA
needs to overcome several biological hurdles [2]. Naked siRNA is easily
degraded by nucleases in the blood-stream and also is cleared by the
kidneys within several minutes [3]. The negative charge of the siRNA
inhibits its interaction with negatively charged plasma membrane
for the internalization into cells. Even though some portions of siRNAs
these external circumstances and to allow it to escape the endosome
efficiently before the lysosomal degradation is required for effective
siRNA therapeutics.
Various siRNA carriers such as lipid-based carriers (lipoplexes) [4]
and polycation-based carriers (polyplexes) [5] have been developed
to fulfill such requirements, including cellular uptake and endosomal
escape of siRNA. For example, polyethylenimine (PEI) has been widely
studied for siRNA delivery as well as plasmid DNA delivery due to its
powerful endosomal escape ability, based on the proton sponge hy-
pothesis [5,6]. Despite the partial success of such delivery vehicles,
toxicity issues derived from the carrier components limit practical
applications. Thus, the design of a carrier for efficient siRNA delivery
with limited cytotoxicity is of great importance. We previously
reported that the cationic polyaspartamide derivative, poly{N-[N-(2-
aminoethyl)-2-aminoethyl]aspartamide} (PAsp(DET)), possessed
pH-sensitive endosome destabilizing activity [7]. PAsp(DET) can de-
stabilize the cellular membrane only at the endosomal pH, probably
because the mono-protonated form of 1, 2-diaminoethane moiety
in the PAsp(DET) side chain at neutral pH is converted to the di-
⁎
Corresponding author. Department of Materials Engineering, Graduate School of
Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan.
Tel.: +81 3 5841 7138; fax: +81 3 5841 7139.
E-mail address: kataoka@bmw.t.u-tokyo.ac.jp (K. Kataoka).
0168-3659/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jconrel.2010.03.019

142
H.J. Kim et al. / Journal of Controlled Release 145 (2010) 141–148
protonated form at the acidic pH condition in the endosome. PAsp
(DET)-based polyplexes with plasmid DNA have remarkable trans-
fection efficiency without marked cytotoxicity in various cultured
cells, including primary cells [8,9]. Also, our recent studies have
revealed the in vivo utilities of PAsp(DET) polyplexes through several
animal experiments: (i) local transfection into a rabbit's clamped
carotid artery with neointima via intra-arterial injection [10] and into
mouse lungs via intra-tracheal injection [11], (ii) systemic delivery
into subcutaneous pancreatic tumors via intravenous injection into a
mouse tail vein [12], and (iii) transfection into a mouse skull via the
regulated release from a calcium phosphate cement scaffold [13].
Although PAsp(DET) polyplexes showed significant success in de-
livering plasmid DNA into cells, their direct application to siRNA de-
livery was appreciably limited due to the instability of the polymer/
siRNA complexes under physiological conditions. One of the reasons
for such lower stability of the PAsp(DET)/siRNA complex is explained
by the short and rigid structure of siRNA (21 base pairs long) compared
to plasmid DNA (over 3000 base pairs long). The condensation of a
single plasmid DNA by polycations provides the polyplex with a stable
core, whereas such condensation does not occur with much shorter
siRNAs [14]. Therefore, an additional association force is needed for
stable siRNA complex formation.
As for an additional association force, hydrophobic interaction is
a promising candidate for stabilization of siRNA complex based on
the polyion complex formation. Indeed, hydrophobic group-modified
polycations, such as poly(N-methyldietheneamine sebacate) (PMDS),
PEI, and oligo-arginine, were tested for siRNA delivery, resulting in the
formation of stable complexes [15–17]. Here, we report the devel-
opment of hydrophobic polycations with high complex stability as
well as low cytotoxicity by using PAsp(DET) as the backbone poly-
cation and stearoyl groups as a hydrophobic moiety. To optimize the
interaction between hydrophobic polycations and siRNA, stearoyl
PAsp(DET) was synthesized with varying substitution degrees and
characterized from the view-point of siRNA complex stability and
RNAi activity in cultured cells. Furthermore, the efficient RNAi ob-
tained from stearoyl PAsp(DET)/siRNA complexes motivated us to
investigate the transfection mechanism, which revealed cellular in-
ternalization and intracellular trafficking to be key steps.
(sense: 5′-GCA GCA CGA CUU CUU CAA GdTdT-3′; antisense: 5′-CUU GAA
GAA GUC GUG CUG CdTdT-3′), BCL-2 (human BCL-2, M13994) siRNA
(sense: 5′-CAG GAC CUC GCC GCU GCA GAC-3′; antisense: 3′-CGG UCC
UGG AGC GGC GAC GUC UG-5′ [19]), and VEGF (human VEGF, NM_
001025366) siRNA (sense: 5′-GGA GUA CCC UGA UGA GAU CdTdT-3′;
antisense: 5′-GAU CUC AUC AGG GUA CUC CdTdT-3′) were synthesized
by Hokkaido System Science Co., Ltd. (Hokkaido, Japan).
2.2. Synthesis
Synthesis methods are available in Supplementary data.
2.3. Preparation and characterization of siRNA complex with stearoyl
PAsp(DET)
Polycations were dissolved in 10 mM HEPES buffer (pH 7.3) or 50%
ethanol solution (ethanol/10 mM HEPES buffer, 1:1 v/v) and then
mixed with 20 µM siRNA solution (10 mM HEPES buffer, pH 7.3) to
form siRNA complexes (5 µM of siRNA) at the desired N/P ratio.
Complex size and zeta potential were determined using a Zetasizer
(Malvern Instruments, Worcestershire, U.K.) with a He–Ne Laser (λ=
633 nm) for the incident beam at a detection angle of 173° and a
temperature of 25 °C. The size measurement was performed in a low-
volume quartz cuvette (ZEN2112, Malvern Instruments, volume 12 µL).
The data obtained from the rate of decay in the photon correlation
function were analyzed by the cumulant method and the corresponding
hydrodynamic diameter of the complexes was then calculated by
the Stokes–Einstein equation. For zeta potential measurements, each
complex solution was placed in a folded capillary cell (Malvern In-
struments). Zeta potential was calculated from the measured electro-
phoretic mobility using the Smoluchowski equation.
2.4. Diffusion coefficient measurement by fluorescence correlation
spectroscopy (FCS)
FCS experiments were performed using a Confocor3 module (Carl
Zeiss, Jena, Germany) equipped with a Zeiss C-Apochromat 40× water
objective. A HeNe laser (543 nm) was used for Cy3-labeled siRNA
excitation and emission was filtered through a 560–615 nm band pass
filter. Samples were placed into 8-well Lab-Tek chambered borosil-
icate cover-glass (Nalge Nunc International, Rochester, NY) and
measured at room temperature. siRNA stock was prepared to contain
1% Cy3-labeled siRNA concentration, and each analysis of naked Cy3-
siRNA, PAsp(DET)/Cy3-siRNA, stearoyl PAsp(DET)/Cy3-siRNA com-
plexes (5 µM siRNA, N/P 5.0), and Rhodamine 6G as a reference in
10 mM HEPES buffer (pH 7.3) consisted of 10 measurements with a
sampling time of 20 s. The measured autocorrelation curves were
fitted with the Zeiss Confocor3 software package to obtain the dif-
fusion coefficient, D.
Stability of siRNA complexes in the cell culture condition was
evaluated in DMEM without L-glutamine and phenol red containing
10% FBS (DMEM/FBS). siRNA stock was prepared to contain 2% Cy3-
labeled siRNA concentration, and each sample of naked Cy3-siRNA,
PAsp(DET)/Cy3-siRNA, stearoyl PAsp(DET)/Cy3-siRNA complexes
(5 µM siRNA, N/P 5.0) was diluted 10 times with DMEM/FBS and
incubated at designated period before measurements.
2. Materials and methods
2.1. Materials
β-Benzyl-L-aspartate N-carboxy-anhydride (BLA-NCA) was synthe-
sized according to Fuchs's method [18]. N,N-Dimethylformamide (DMF),
dichloromethane (DCM), n-butylamine, diethylenetriamine (DET),
methanol (MeOH), N-hydroxysuccinimide (NHS), and N-methyl-2-
pyrrolidone (NMP) were purchased from Wako Pure Chemical Indus-
tries, Ltd. (Osaka, Japan). Poly(L-lysine) hydrobromide (Mw=15,000–
30,000), Dulbecco's modified Eagle's medium (DMEM), DMEM without
L-glutamine and phenol red, diisopropylethylamine (DIPEA), stearic acid,
and 0.4% trypan blue solution were purchased from Sigma-Aldrich Co.
(St. Louis, Mo). 1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide, hy-
drochloride (EDC) was purchased from Dojindo (Kumamoto, Japan).
DMF, n-butylamine, DET, and DIPEA were distilled with the conven-
tional methods before use. The human pancreatic adenocarcinoma cell
line, Panc-1, was obtained from the American Type Culture Collection
(Manassas, VA). The luciferase-expressing mouse melanoma cell line,
B16F10-Luc was purchased from Caliper LifeScience (Hopkinton, MA).
Fetal bovine serum (FBS) was purchased from Dainippon Sumitomo
Parma Co., Ltd. (Osaka, Japan). Lipofectamine 2000 and ExGen500
were purchased from Invitrogen (Carlsbad, CA) and Fermentas (Ontario,
Canada), respectively. Firefly luciferase siRNA (sense: 5′-CUU ACG CUG
AGU ACU UCG AdTdT-3′; antisense: 5′-UCG AAG UAC UCA GCG UAA
GdTdT-3′), Cy5-labeled firefly luciferase siRNA, Cy3-labeled firefly
luciferase siRNA, Enhanced Green Fluorescence Protein (EGFP) siRNA
2.5. Endogenous luciferase gene knockdown in B16F10-Luc
Luciferase-expressing B16F10 cells were seeded into a 96-well
plate at a density of 5000 cells/well in DMEM containing 10% FBS.
Firefly luciferase and EGFP siRNA complexes were transfected at
100 nM siRNA. After 48 h incubation, the media was exchanged and
the cells were incubated for another 24 h. Luciferase gene knock-
down was measured using the Luciferase Assay System (Promega) in

H.J. Kim et al. / Journal of Controlled Release 145 (2010) 141–148
143
a luminescence microplate reader (Mithras LB 940, Berthold technol-
ogies, Bad Wildbad, Germany).
rank correlation coefficient can remove the effects of brightness dif-
ference between two images, it is suitable for excluding the parameter
for the amount of intracellular uptake of Cy3-labeled siRNA.
2.6. BCL-2 endogenous gene knockdown by real-time PCR
2.10. Cell viability assay
Panc-1 cells were seeded into a 6-well plate at a density of 100,000
cells/well in DMEM containing 10% FBS. BCL-2 siRNA complexes were
transfected at 100 nM siRNA. After 48-hour incubation, total RNA was
collected with an RNeasy Mini kit (Qiagen). Reverse transcription PCR
was carried out using a TAKARA PrimeScript RT reagent kit and real-
time PCR was measured using a QuantiTect SYBR Green PCR kit
(Qiagen) with an ABI 7500 Fast Real-Time PCR System (Applied
Biosystems, Foster City, CA). GAPDH was used as an internal control
for the real-time PCR amplification. The mRNAs of BCL-2 and GAPDH
were amplified with TAKARA BCL-2 and GAPDH primers (Primer Set
ID HA032557 and HA067812, respectively). All procedures followed
the manufacturer's protocol.
Cell viability was determined using a Cell Counting Kit-8 (Dojindo,
Japan). Panc-1 cells were seeded into a 24-well plate at a density of
20,000 cells/well in DMEM containing 10% FBS. After overnight in-
cubation, siRNA complexes were introduced and cells were then
incubated for another 48 h. The growth medium was exchanged with
fresh media (500 µl) containing the manufacturer's reagent (50 µl)
and the cells were then incubated for 1.5 h. The absorbance of the
media was measured at 450 nm using a microplate reader (Biorad).
3. Results and discussion
3.1. Synthesis of stearoyl PAsp(DET) and stearoyl PLL
2.7. VEGF endogenous gene knockdown in real-time PCR
Stearoyl PAsp(DET) was synthesized by amide bond formation be-
tween N-succinimidyl octadecanoate and the primary amine of PAsp
(DET) (Scheme 1). The feed ratio was defined as the relative molar ratio
of N-succinimidyl octadecanoate to primary amines of PAsp(DET) in
the reaction mixture. The resulting introduction rate of the stearoyl
amide was calculated from the peak intensity ratio between the methyl
protons of stearoyl moieties (CH3–, δ=1.19 ppm) and β-protons of
aspartate groups in PAsp(DET) (–CH2–, δ=3.16 ppm) in the 1H NMR
spectra (data not shown). Three different substituted stearoyl PAsp
(DET)s (12%, 19%, and 32%) were prepared by varying the feed ratio.
As a control polycation without endosomal disrupting ability,
23% stearoyl PLL was synthesized in the same fashion by amide bond
formation between N-succinimidyl octadecanoate and primary
amines of PLL. The feed ratio was calculated from the relative molar
ratio of N-succinimidyl octadecanoate to primary amines of PLL.
The resulting stearoyl substitution rate was obtained from the peak
intensity ratio between the methyl protons of stearoyl moieties and
ε-protons of the lysine groups (–CH2–, δ=3.04–3.17 ppm) in the 1H
NMR spectra (data not shown). A series of the hydrophobic poly-
cations were abbreviated as PAsp(DET)-ST12, PAsp(DET)–ST19,
PAsp(DET)-ST32 for 12%, 19%, 32% stearoyl PAsp(DET), respectively,
and as PLL-ST23 for 23% stearoyl PLL.
Endogenous VEGF gene suppression in Panc-1 cells was evaluated
using the same procedure as BCL-2 gene knockdown with actin used
as the internal control. VEGF was amplified using 5′-AGT GGT CCC
AGG CTG CAC-3′ as the forward primer and 5′-TCC ATG AAC TTC ACC
ACT TCG T-3′ as the reverse primer. Actin was amplified using 5′-CCT
GGC ACC CAG CAC AAT G-3′ as the forward primer and 5′-CGC CGA
TCC ACA CGG AGT A-3′ as the reverse primer. All experiments were
performed in triplicate and data was normalized to actin expression.
2.8. Flow cytometer measurement
B16F10 cells were seeded into a 6-well plate at a density of
100,000 cells/well in DMEM containing 10% FBS. Cy5-labeled siRNA
complexes were prepared for each polymer (N/P 5.0) and transfected
at 100 nM siRNA. After 3-hour incubation, the media was removed
and the cells were washed with 0.5 mL of PBS. The cells were treated
with a trypsin-EDTA solution for 2 min and suspended in PBS. The
intracellular uptake of the siRNA complexes was measured using a
BDTM LSR II flow cytometer (BD Biosciences).
2.9. Confocal laser scanning microscope (CLSM)
Panc-1 cells were seeded into a 35-mm glass-based dish (Iwaki,
Tokyo, Japan) at a density of 50,000 cells/well in DMEM containing
10% FBS. The Cy3-labeled siRNA complex of each polymer (N/P 5.0)
was transfected at 100 nM siRNA. After the designated incubation
period, each dish was observed using a CLSM (ZEISS LSM 510, Carl
Zeiss, Oberlochen, Germarny) equipped with a C-Apochromat 63×
objective (Carl Zeiss). Late-endosomes/lysosomes were stained by
LysoTracker Green (Molecular Probes, Eugene, OR) for 1.5 h and nuclei
were stained with Hoechst 33342 (Dojindo, Japan) for 10 min before
each observation. 0.4% trypan blue solution was added to media just
before observation to quench extracellular-bound Cy3-labeled siRNA
[20]. The excitation wavelengths used were 488 nm (Ar laser) for
LysoTracker, 543 nm (He–Ne laser) for Cy3-labeled siRNA, and 710 nm
(MaiTai laser; two photon excitation; Spectra Physics, Mountain View,
CA) for Hoechst 33342.
3.2. Formation of siRNA complex with stearoyl PAsp(DET)
Formation of siRNA complexes with stearoyl PAsp(DET) was
evaluated by gel electrophoresis at various N/P ratios. The N/P ratio
is defined as the molar ratio of the total amines in PAsp(DET) (N) to
the phosphates in siRNA (P). Note that stearoyl introduction con-
currently reduces the number of primary amines (N) of PAsp(DET)
with increasing degree of substitution. Thus, the total amines in
stearoyl PAsp(DET) do not include the primary amines of PAsp(DET)
which were converted to amide bonds after stearoyl introduction.
Because the 1, 2-diaminoethane side chain in PAsp(DET) takes a mono-
protonated form at pH 7.3 [7], The free siRNA band disappears above
the stoichiometric N/P ratio of complete siRNA formation (N/P=2)
(Fig. S1). However, smear bands were detected at higher molecular
weight positions at N/P=2.5–4.5. Thus N/P≥5 was selected for further
experiments because all of the PAsp(DET) and stearoyl PAsp(DET)
derivatives showed the disappearance of the free siRNA and smear
band, indicating that all the siRNA molecules were associated with
polycations.
Statistical colocalization was calculated from Spearman's rank
correlation coefficient [21];
di2
À
Á
ρS = 1−6∑
;
n n2−1
The size and zeta potential of siRNA complexes with stearoyl PAsp
(DET) in 10 mM HEPES buffer (pH 7.3) were determined by dynamic
light scattering (DLS) and electrophoretic mobility measurements,
respectively, and the results are summarized in Table 1. Stearoyl
PAsp(DET) complexes (N/P 5–N/P 10) ranged in size from 120 nm to
where a set of n pairs of numbers (xi, yi) means the values recorded
from the red and green channels of a image, and di is the difference
between the rank position in the (xi, yi) data pair. Because Spearman's

144
H.J. Kim et al. / Journal of Controlled Release 145 (2010) 141–148
Scheme 1. Synthetic routes of N-succinimidyl octadecanoate (2), stearoyl PAsp(DET) (4), and stearoyl poly(L-lysine) (6).
200 nm with a polydispersity index (PDI) of less than 0.2. Zeta
potentials of the siRNA complexes showed relatively constant values
(+30 mV to +40 mV) irrespective of the stearoylation degree. On
the other hand, the siRNA complex formed with unmodified PAsp
(DET) exhibited appreciably low scattered light intensity close to the
background of the solvent. In this regard, complex formation was
further evaluated by FCS with Cy3-labeled siRNA, which allows de-
termination of the diffusion coefficient of a complex in a very small
volume (∼femto L) [22,23]. All the stearoyl PAsp(DET) complexes
had a one order of magnitude lower diffusion coefficient than naked
siRNA and unmodified PAsp(DET) complex (Table 2). Considering
that the diffusion coefficient is reversely correlated to the hydrody-
namic diameter of the complexes, with the assumption that siRNA
complexes are spherical, the obtained result confirmed that stearoyl
introduction would contribute to the formation of larger complexes
with higher association numbers, compared to the unmodified PAsp
(DET). It is worth noting that the diffusion coefficients of the PAsp
(DET)-ST12, ST19, and ST32 complexes measured by FCS are roughly
correlated to those measured by DLS (3.27 0.13, 2.79 0.32, and
2.59 0.26 µm2/s).
The stability of stearoyl PAsp(DET)/siRNA complex was investigated
in cell culture conditions. The diffusion coefficient was monitored over
time by FCS to estimate the dissociation tendency of siRNA complexes
in DMEM containing 10% FBS (Fig. 1). The diffusion coefficient of
unmodified PAsp(DET)/siRNA complex was increased to that of naked
siRNA with increased incubation period, indicating that the unmodified
PAsp(DET)/siRNA complex was completely dissociated in a 3-hour
incubation period. In contrast, stearoyl PAsp(DET)/siRNA complexes
maintained their initial diffusion coefficient for a 3-hour incubation
period. This result indicates the higher stability of stearoyl PAsp(DET)
complexes than the unmodified PAsp(DET) complex in the serum-
containing media. These results reveal that the stearoyl PAsp(DET) com-
plex tolerates the cell culture condition better than does the unmodified
PAsp(DET) complex, allowing for improved cellular internalization.
3.3. Gene expression inhibition by stearoyl PAsp(DET)/siRNA complexes
To determine the optimal stearoylation degree for efficient gene
knockdown, the luciferase gene knockdown ability of stearoyl PAsp
(DET)/siRNA complexes was measured in luciferase-expressing B16F10
Table 1
Size and zeta potential of stearoyl PAsp(DET)/siRNA complexes.
Stearoyl introduction rate N/P
Size (nm) S.D. Zeta potential (mV) S.D.
12%
5
150
6
2
5
1
4
1
1
39.7 8.0
39.0 8.5
37.4 9.7
29.5 6.1
36.8 5.2
36.0 5.3
38.2 6.6
32.0 5.0
34.5 5.1
35.5 5.0
Table 2
7.5 136
10
5
7.5 134
10
15
5
Diffusion coefficient of stearoyl PAsp(DET)/Cy3-siRNA complexes.
122
196
19%
Diffusion coefficient (µm2/s) S.D.
129
141
187 24
Naked Cy3-siRNA
Unmodified PAsp(DET)
PAsp(DET)–ST12
PAsp(DET)–ST19
PAsp(DET)–ST32
56.26 0.18
33.88 0.14
2.17 0.03
1.89 0.03
1.46 0.02
32%
7.5 177 45
10 192 23

H.J. Kim et al. / Journal of Controlled Release 145 (2010) 141–148
145
for siRNA knockdown [24]. The endogenous VEGF gene knock-
down efficacy of the PAsp(DET)–ST19 complex in Panc-1 cells was
compared with two commercially available transfection reagents,
Lipofectamine2000 and the linear polyethylenimine-based reagent
ExGen500. VEGF mRNA levels were evaluated by real-time PCR
(Fig. 3A). The PAsp(DET)–ST19/siRNA complex exhibited around 60%
VEGF mRNA reduction, similar to Lipofectamine2000 and much higher
than ExGen500 (15% reduction). Furthermore, the amount of VEGF
protein secreted from Panc-1 cells was also measured by im-
munoassay techniques. The VEGF protein production in cells treated
with the PAsp(DET)–ST19 complex was also reduced to approximately
50% (Fig. S2).
Another therapeutic gene, anti-apoptotic BCL-2, was also selected
to evaluate the potential for future siRNA therapeutics in cancer
treatments. The BCL-2 protein has been a common cause of tumor
genesis and chemotherapy resistance in various types of tumors [25].
Anti-cancer drug treatment with the reduction of BCL-2 protein ex-
pression has been one of the approaches to induce apoptosis in cancer
cells [26]. The BCL-2 siRNA sequence used here was selected from a
previously published work that indicated this sequence showed high
Fig. 1. Time-dependent change in diffusion coefficients of siRNA complexes in the cell
culture medium measured by FCS. Curves show naked siRNA (●), unmodified PAsp
(DET)/siRNA (○), PAsp(DET)–ST12 (▼), PAsp(DET)–ST19 (△), and PAsp(DET)–ST32
(■)complexes.
cells (Fig. 2). siRNA complexes were prepared with stearoyl PAsp(DET)
comprising different stearoyl introduction ratios at various N/P ratios.
The luminescence intensity of non-treated cells was set as 1.0, and EGFP
siRNA was used as a negative control. Unmodified PAsp(DET) complex
showed almost no luciferase gene knockdown at the tested N/Ps.
However, stearoyl introduction significantly enhanced the gene sup-
pression efficacy. Especially, the PAsp(DET)–ST19 showed 60% gene
suppression at the N/P ratios of 7.5 and 10. Apparently, stearoyl in-
troduction had an optimal stearoylation ratio in the gene suppression
tendency. Considering that all the stearoyl PAsp(DET) (12%, 19%, and
32%) had almost the same cellular uptake in 24-hour incubation by flow
cytometer measurement (data not shown), 19% stearoyl introduction
was the optimal ratio, possibly because of the excellent balance between
siRNA release via the intracellular dissociation and sufficient stability
of the siRNA complex in the serum-containing media (Fig. 1). It is
worth mentioning that none of the tested complexes showed signifi-
cant gene knockdown when prepared containing EGFP siRNA, suggest-
ing low cytotoxicity and minimal off-target effects of PAsp(DET)-based
complexes.
Based on this result, the PAsp(DET)–ST19 complex was further
examined for the knockdown ability of a therapeutic gene. VEGF, a
well-known regulator of angiogenesis, was selected as the target gene
Fig. 3. Endogenous gene knockdown by various siRNA complexes. (A) Comparison of
stearoyl PAsp(DET) with ExGen500 and Lipofectamine2000 for endogenous VEGF
knockdown by real-time PCR. Each sample was transfected in the Panc-1 cells at
100 nM siRNA concentration for 48 h. The VEGF mRNA amount was normalized with
actin mRNA as an internal control. (B) Comparison of stearoyl PAsp(DET) with stearoyl
PLL for anti-apoptotic BCL-2 endogenous gene knockdown measured by real-time PCR.
Each sample was transfected in the Panc-1 cells for 48 h. The BCL-2 mRNA amount was
normalized with GAPDH mRNA as an internal control. In both experiments, luciferase
siRNA was used as a control. Results are expressed as the mean S.D. (n=4).
Fig. 2. Luciferase gene knockdown by siRNA complexes of stearoyl PAsp(DET) with
different substitution degrees in the B16F10-Luc cell. EGFP siRNA (black) was used as a
control for luciferase siRNA (gray). The siRNA complexes were treated at 100 nM
concentration for 72-hour incubation. Results are expressed as the mean S.D. (n=4).

146
H.J. Kim et al. / Journal of Controlled Release 145 (2010) 141–148
The PLL-ST23 showed almost no endogenous gene knockdown or
slightly increased BCL-2 expression at the examined N/Ps. On the
other hand, gene knockdown was significantly improved with the
PAsp(DET)–ST19 complex, with mRNA expression reduced over 60%.
The obtained result suggests that the endosomal escaping ability of
PAsp(DET) might be crucial for efficient gene knockdown.
3.4. Cellular uptake and intracellular trafficking of stearoyl PAsp(DET)
The transfection results revealed that stearoyl introduction into
a PAsp(DET) polymer backbone led to improved siRNA delivery. To
elucidate the reason for this improvement, the cellular uptake and
intracellular trafficking of the stearoyl PAsp(DET)/siRNA complex
were monitored by flow cytometry and CLSM using fluorescence dye-
labeled siRNA. B16F10 cells were incubated with naked Cy5-labeled
siRNA, Cy5-labeled siRNA complexes of unmodified PAsp(DET) and
PAsp(DET)–ST19 for 3 h and analyzed by flow cytometer (Fig. 4).
Stearoyl introduction clearly improved the intracellular uptake of
siRNA complexes formed with PAsp(DET)–ST19. The uptake of PAsp
(DET)–ST19 complex (mean fluorescence 2.3×103) was significantly
higher than those of unmodified PAsp(DET) complex (8.7×102) and
naked siRNA (7.4×102). This result is probably due to the higher
stability of siRNA complexes with stearoyl groups in the cell culture
condition (Fig. 1). The uptake of unmodified PAsp(DET) complex was
shown to be similar to that of naked siRNA, indicating that cellular
Fig. 4. Cellular uptake of the siRNA complexes in B16F10 cells. Each peak represents the
cells treated by naked siRNA (A), unmodified PAsp(DET) (B), and PAsp(DET)–ST19 (C)
siRNA complexes (N/P 5.0) from the left.
in vivo gene knockdown efficacy [19]. In this experiment, the PAsp
(DET)–ST19 was compared with 23% stearoyl PLL (PLL-ST23) to
confirm the backbone superiority for siRNA delivery because PLL is
a polycation with poor endosomal escape ability (Fig. 3B). After 48 h
of incubation, the BCL-2 mRNA level was evaluated by real-time PCR.
Fig. 5. CLSM images showing intracellular distribution of the siRNA complexes in Panc-1 cells. Panels A and B represent the cells incubated with PLL–ST23 complexes (N/P 5.0) after
3-hour and 24-hour incubations, respectively. Panels C and D represent the cells incubated with PAsp(DET)–ST19 complexes (N/P 5.0) after 3-hour and 24-hour incubations,
respectively. Cy3-labeled siRNA and LysoTracker (a late-endosome and lysosome marker) are shown in red and green, respectively. The nucleus was stained with Hoechst 33342 in

H.J. Kim et al. / Journal of Controlled Release 145 (2010) 141–148
147
uptake of unstable siRNA complex is not efficient. On the other hand,
the cellular uptake of PAsp(DET)–ST19 and PLL–ST23 was compared
in Panc-1 cells. Although the PLL–ST23 complex showed no gene
knockdown in Fig. 3B, it (1.3×104) was internalized more efficiently
than the PAsp(DET)–ST19 complex (1.0×104) (data not shown),
indicating that steps following internalization, such as endosome
escape, are crucial for siRNA-mediated gene knockdown.
The endosome escape behavior of the PAsp(DET)–ST19 complex
was compared with that of the PLL–ST23 complex by CLSM ob-
servation in Panc-1 cells (Fig. 5). The siRNA was labeled with Cy3 (red),
and late-endosomes and lysosomes were stained with LysoTracker
Green (green). Nuclei were stained with Hoechst 33342 (blue). Cy3-
labeled siRNA colocalized in late-endosomes and lysosomes should
be shown as yellow, and endosomal escape of the siRNA would induce
the recovery of red color from yellow over time. After 3-hour in-
cubation, Cy3-labeled siRNA of the PLL–ST23 complex was observed
mainly in the cell interior with yellow fluorescence and localized in
small confined spaces (Fig. 5A), indicating that the majority of Cy3-
labeled siRNA was trapped in late-endosomes/lysosomes and only a
small portion of the complex escaped from the organelles. Even after
24-hour incubation (Fig. 5B), very little red fluorescence was observed
in the cytoplasm, whereas large yellow regions were still observed,
indicating that the major stearoyl PLL complexes were still trapped
in the late-endosome/lysosome compartments. In contrast, the PAsp
(DET)–ST19 complex exhibited some red fluorescence surrounding
yellow late-endosomes/lysosomes even after only 3-hour incubation
(Fig. 5C). The observation in 24-hour incubation (Fig. 5D) showed
that red and green fluorescence were separated in the cytoplasm
with almost no yellow fluorescence, demonstrating the ability of
the stearoyl PAsp(DET) to decrease the amount of siRNA molecules
trapped in the late-endosomes/lysosomes.
The detailed colocalization profile was further analyzed using
ImageJ software, which calculates Spearman's rank correlation coef-
ficient from the red and green channels of the confocal images [21].
The coefficient value ranges from −1 to 1, where 0 indicates that there
is no discernable correlation and −1 and +1 mean strong negative
and positive correlations, respectively. The correlation coefficient
value for the PLL–ST23 complex of the 24-hour incubation was 0.72,
which means there was a strong positive correlation between the
distribution of red and green fluorescence. In contrast, the correlation
coefficient value of the PAsp(DET)–ST19 complex after 24-hour
incubation was calculated to be 0.23, indicating little discernable
correlation between Cy3-siRNA and late-endosomes/lysosomes. The
obtained correlation values quantitatively confirm that stearoyl PAsp
(DET) significantly decreased the amount of siRNA complexes trapped
in late-endosomes/lysosomes. This stronger endosome escape ability
is likely responsible for the enhanced siRNA transfection compared
to stearoyl PLL. The outstanding endosome escape ability of stearoyl
PAsp(DET) is considered to be mainly due to the pH-dependent pro-
tonation of N-(2-aminoethyl)-2-aminoethyl groups in the PAsp(DET)
side chain [7]. Consequently, the enhanced gene suppression efficacy
of the PAsp(DET)–ST19 appears to be based on the higher cellular
uptake and efficient endosomal escape (Fig. 4).
Fig. 6. Viability of Panc-1 cells treated with each siRNA complex under the same
experimental condition as the real-time PCR. Results are expressed as the mean S.D.
(n=4).
4. Conclusion
In this study, we demonstrated that stearoyl introduction onto
PAsp(DET) side chains led to the formation of a siRNA complex with
improved in vitro RNAi activity. Stearoyl introduction significantly
enhanced the cellular uptake of siRNA due to complex stabilization
via hydrophobic interaction. Additionally, the PAsp(DET) polycation
backbone contributed to the excellent endosomal escape of siRNA
complexes. Ultimately, the PAsp(DET)–ST19 complexes formed at
N/Ps of 7.5–10 exhibited the best combination of high RNAi activity
with low cytotoxicity of the tested samples, including commercially
available reagents. The stearoyl PAsp(DET)/siRNA complex is a
promising candidate for therapeutic siRNA delivery, a result of rational
carrier design aimed to overcome specific subcellular barriers.
Acknowledgements
This work was financially supported by a Core Research for
Evolutional Science and Technology (CREST) grant from the Japan
Science and Technology Agency (JST) as well as Takeda Science
Foundation. HJ Kim is thankful to Mr. Kazuya Suma for his help with
the polymer synthesis.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi: 10.1016/j.jconrel.2010.03.019.
References
[1] L. Aagaard, J.J. Rossi, RNAi therapeutics: principles, prospects and challenges, Adv.
Drug Deliv. Rev. 59 (2007) 75–86.
[2] D.R. Corey, Chemical modification: the key to clinical application of RNA interference?
J. Clin. Invest. 117 (2007) 3615–3622.
[3] D.A. Braasch, Z. Paroo, A. Constantinescu, G. Ren, O.K. Őz, R.P. Mason, D.R. Corey,
Biodistribution of phosphodiester and phosphorothioate siRNA, Bioorg. Med.
Chem. Lett. 14 (2004) 1139–1143.
[4] A. Akinc, A. Zumbuehl, M. Goldberg, E.S. Leshchiner, V. Busini, N. Hossain, S.A.
Bacallado, D.N. Nguyen, J. Fuller, R. Alvarez, A. Borodovsky, T. Borland, R. Constien, A.
de Fougerolles, J.R. Dorkin, K.N. Jayaprakash, M. Jayaraman, M. John, V. Koteliansky,
M. Manoharan, L. Nechev, J. Qin, T. Racie, D. Raitcheva, K.G. Rajeev, D.W.Y. Sah, J.
Soutschek, I. Toudjarska, H.-P. Vornlocher, T.S. Zimmermann, R. Langer, D.G.
Anderson, A combinatorial library of lipid-like materials for delivery of RNAi
therapeutics, Nat. Biotechnol. 26 (2008) 561–569.
[5] H. de Martimprey, C. Vauthier, C. Malvy, P. Couvreur, Polymer nanocarriers for the
delivery of small fragments of nucleic acids: oligonucleotides and siRNA, Eur. J.
Pharm. Biopharm. 71 (2009) 490–504.
[6] M. Neu, D. Fischer, T. Kissel, Recent advances in rational gene transfer vector
design based on poly(ethylene imine) and its derivatives, J. Gene Med. 7 (2005)
992–1009.
[7] K. Miyata, M. Oba, M. Nakanishi, S. Fukushima, Y. Yamasaki, H. Koyama, N. Nishiyama,
K. Kataoka, Polyplexes from poly(aspartamide) bearing 1, 2-diaminoethane side
3.5. Cell viability of stearoyl PAsp(DET)
Low cytotoxicity as well as high knockdown efficacy is critical
factors for the development of an effective siRNA carrier. Therefore,
cell viability after introduction of stearoyl PAsp(DET)/siRNA com-
plexes was measured by the MTT assay in Panc-1 cells (Fig. 6). The
obtained cell viability was over 90% for each stearoyl PAsp(DET)
complex, similar to ExGen500 and higher than Lipofectamine 2000
(approximately 80% cell viability), indicating the minimal cytotox-
icity of the stearoyl PAsp(DET) complexes in this experimental
condition.

148
H.J. Kim et al. / Journal of Controlled Release 145 (2010) 141–148
chains induce pH-selective, endosomal membrane destabilization with amplified
transfection and negligible cytotoxicity, J. Am. Chem. Soc. 130 (2008) 16287–16294.
[17] W.J. Kim, L.V. Christensen, S. Jo, J.W. Yockman, J.H. Jeong, Y. -H. Kim, S. W. Kim,
Cholesteryl oligoarginine delivering vascular endothelial growth factor siRNA
effectively inhibits tumor growth in colon adenocarcinoma, Mol. Ther. 14 (2006)
343–350.
[18] A. Harada, K. Kataoka, Formation of polyion complex micelles in an aqueous
milieu from a pair of oppositely-charged block copolymers with poly(ethylene
glycol) segments, Macromol. 28 (1995) 5294–5299.
[19] M. Ocker, D. Neureiter, M. Lueders, S. Zopf, M. Ganslmayer, E.G. Hahn, C. Herold, D.
Schuppan, Variants of bcl-2 specific siRNA for silencing antiapoptotic bcl-2 in
pancreatic cancer, Gut 54 (2005) 1298–1308.
[20] K. de Bruin, N. Ruthardt, K. von Gersdorff, R. Bausinger, E. Wagner, M. Ogris, C.
Bräuchle, Cellular dynamics of EGF receptor-targeted synthetic viruses, Mol. Ther.
15 (2007) 1297–1305.
[8] N. Kanayama, S. Fukushima, N. Nishiyama, K. Itaka, W.-D. Jang, K. Miyata, Y.
Yamasaki, U.-I. Chung, K. Kataoka, A PEG-based biocompatible block catiomer
with high buffering capacity for the construction of polyplex micelles showing
efficient gene transfer toward primary cells, ChemMedChem 1 (2006) 439–444.
[9] K. Masago, K. Itaka, N. Nishiyama, U.-I. Chung, K. Kataoka, Gene delivery with
biocompatible cationic polymer: pharmacogenomic analysis on cell bioactivity,
Biomaterials 28 (2007) 5169–5179.
[10] D. Akagi, M. Oba, H. Koyama, N. Nishiyama, S. Fukushima, T. Miyata, H. Nagawa, K.
Kayaoka, Biocompatible micellar nanovectors achieve efficient gene transfer to
vascular lesions without cytotoxicity and thrombus formation, Gene Ther. 14
(2007) 1029–1038.
[11] M. Harada-Shiba, I. Takamisawa, K. Miyata, T. Ishii, N. Nishiyama, K. Itaka, K.
Kangawa, F. Yoshihara, Y. Asada, K. Hatakeyama, N. Nagaya, K. Kataoka, Intratracheal
gene transfer of adrenomedullin using polyplex nanomicelles attenuates mono-
crotaline-induced pulmonary hypertension in rats, Mol. Ther. 17 (2009) 1180–1186.
[12] K. Miyata, M. Oba, M.R. Kano, S. Fukushima, Y. Vachutinsky, M. Han, H. Koyama, K.
Miyazono, N. Nishiyama, K. Kataoka, Polyplex micelles from triblock copolymers
composed of tandemly aligned segments with biocompatible, endosomal escaping,
and DNA-condensing functions for systemic gene delivery to pancreatic tumor
tissue, Pharm. Res. 25 (2008) 2924–2936.
[21] A.P. French, S. Mills, R. Swarup, M.J. Bennett, T.P. Pridmore, Colocalization of
fluorescent markers in confocal microscope images of plant cells, Nat. Protoc. 3
(2008) 619–628.
[22] M. Meyer, A. Philipp, R. Oskuee, C. Schmidt, E. Wagner, Breathing life into polycations:
functionalization with pH-responsive endosomolytic peptides and polyethylene
glycol enables siRNA delivery, J. Am. Chem. Soc. 130 (2008) 3272–3273.
[23] J. DeRouchey, C. Schmidt, G.F. Walker, C. Koch, C. Plank, E. Wagner, J.O. Rädler,
Monomolecular assembly of siRNA and poly(ethylene glycol)-peptide copoly-
mers, Biomacromolecules 9 (2008) 724–732.
[13] K. Itaka, S. Ohba, K. Miyata, H. Kawaguchi, K. Nakamura, T. Takato, U.-I. Chung, K.
Kataoka, Bone regeneration by regulated in vivo gene transfer using biocompat-
ible polyplex nanomicelles, Mol. Ther. 15 (2007) 1655–1662.
[14] S.C. de Smedt, J. Demeester, W.E. Hennink, Cationic polymer based gene delivery
systems, Pharm. Res. 17 (2000) 113–126.
[15] Y. Wang, S. Gao, W.-H. Ye, H.S. Yoon, Y.-Y. Yang, Co-delivery of drugs and DNA from
cationic core-shell nanoparticles self-assembled from a biodegradable copolymer,
Nat. Mater. 5 (2006) 791–796.
[24] L.M. Ellis, D.J. Hicklin, VEGF-targeted therapy: mechanisms of anti-tumour activity,
Nat. Rev. Cancer 8 (2008) 579–591.
[25] D.S. Ziegler, A.L. Kung, Therapeutic targeting of apoptosis pathways in cancer,
Curr. Opin. Oncol. 20 (2008) 97–103.
[26] J. George, N.L. Banik, S.K. Ray, Bcl-2 siRNA augments taxol mediated apoptotic
death in human glioblastoma U138MG and U251MG cells, Neurochem. Res. 34
(2009) 66–78.
[16] A. Alshamsan, A. Haddadi, V. Incani, J. Samuel, A. Lavasanifar, H. Uludag, Formulation
and delivery of siRNA by oleic acid and stearic acid modified polyethylenimine, Mol.
Pharm. 6 (2009) 121–133.