Gene silencing via RNAi and siRNA quantification in tumor tissue using MEND, a liposomal siRNA delivery system

Abstract

Small interfering RNA (siRNA) would be predicted to function as a cancer drug, but an efficient siRNA delivery system is required for clinical development. To address this issue, we developed a liposomal siRNA carrier, a multifunctional envelope-type nanodevice (MEND). We previously reported that a MEND composed of a pH-sensitive cationic lipid, YSK05, showed significant knockdown in both in vitro and in tumor tissue by intratumoral injection. Here, we report on the development of an in vivo siRNA delivery system that is delivered by systemic injection and an analysis of the pharmacokinetics of an intravenously administered siRNA molecule in tumor tissue. Tumor delivery of siRNA was quantified by means of stem-loop primer quantitative reverse transcriptase PCR (qRT-PCR) method. PEGylation of the YSK-MEND results in the increase in the accumulation of siRNA in tumor tissue from 0.0079% ID/g tumor to 1.9% ID/g tumor. The Administration of the MEND (3 mg siRNA/kg body weight) showed about a 50% reduction in the target gene mRNA and protein. Moreover, we verified the induction of RNA interference by 5' RACE-PCR method. The collective results reported here indicate that an siRNA carrier was developed that can deliver siRNA to a target cell in tumor tissue through an improved siRNA bioavailability.Molecular Therapy (2013); doi:10.1038/mt.2013.57.

Full text

original article
© The American Society of Gene & Cell Therapy
Small interfering RNA (siRNA) would be predicted to
function as a cancer drug, but an efficient siRNA delivery
system is required for clinical development. To address
this issue, we developed a liposomal siRNA carrier, a
multifunctional envelope-type nanodevice (MEND). We
previously reported that a MEND composed of a pH-
sensitive cationic lipid, YSK05, showed significant knock-
down in both in vitro and in tumor tissue by intratumoral
injection. Here, we report on the development of an
in vivo siRNA delivery system that is delivered by systemic
injection and an analysis of the pharmacokinetics of an
intravenously administered siRNA molecule in tumor tis-
sue. Tumor delivery of siRNA was quantified by means
of stem-loop primer quantitative reverse transcriptase
PCR (qRT-PCR) method. PEGylation of the YSK-MEND
results in the increase in the accumulation of siRNA in
tumor tissue from 0.0079% ID/g tumor to 1.9% ID/g
tumor. The Administration of the MEND (3 mg siRNA/kg
body weight) showed about a 50% reduction in the tar-
get gene mRNA and protein. Moreover, we verified the
induction of RNA interference by 5′ RACE-PCR method.
The collective results reported here indicate that an
siRNA carrier was developed that can deliver siRNA to
a target cell in tumor tissue through an improved siRNA
bioavailability.
Received 22 November 2012; accepted 5 March 2013; advance online
publication 9 April 2013. doi:10.1038/mt.2013.57
INTRODUCTION
A small interfering RNA (siRNA) is thought to be a new class of
medicines for cancer treatment because siRNA can, in principle,
inhibit the expression of any genes of interest.1,2 However, naked
siRNA is easily degraded in the blood stream by ribonuclease
and cannot pass through a cellular membrane because of its large
molecular size and high hydrophilicity. ese properties clearly
suggest that an ecient delivery system for siRNA that specically
targets cancer cells is required for developing an RNAi medicine.3
Many groups have reported on the development of siRNA delivery
system for tumor targeting using a variety of formulations, such as
polyplexes, lipoplexes, micelles, and liposomes.4 Toaddress this
issue, we developed a multifunctional envelope-type nanodevice
(MEND) containing macromolecules such as oligonucleotides
and proteins in free forms or condensed forms with a polycation
that are encapsulated within a lipid envelope.5
Macromolecules that have a prolonged retention time in the
blood stream can passively accumulate in tumor tissue. e pas-
sive accumulation in the tumor results from its abnormal vas-
culature, namely an aberrant vascular architecture and lack of
lymphatic drainage, which is referred to as the enhanced perme-
ability and retention (EPR) eect.6 Generally, materials that are
delivered are highly positive charged that permits them to elec-
trostatically interact with the negatively charged siRNA. Although
attempts have been made to use the EPR eect to deliver siRNA
to tumor tissue in many siRNA delivery systems, the highly posi-
tively charged carriers are rapidly eliminated from the circulation,
which results in a poor EPR eect.7 is is because a positively
charged material is immediately recognized by serum proteins
and immune cells in the blood.3 erefore, carriers with a high
cationic charge must be coated with a large amount of polyeth-
ylene glycol (PEG) to prevent their recognition by immune cells
or serum proteins in the blood stream. Modifying an siRNA car-
rier with a heavy coating of PEG results in a low cellular uptake
and attenuated endosomal escape, and hence, a decrease in siRNA
delivery ability.8 We have attempted to overcome this discrepancy
by using cleavable PEG in response to an enzyme that is highly
expressed in cancer cells and a pH-sensitive fusogenic peptide,
and found that these systems could induce gene silencing in
tumor tissue.9,10 ese systems circumvented the issue by remov-
ing PEG or assisting endosomal escape with the peptide. However,
as these carriers were composed of a conventional cationic lipid,
1,2-dioleoyltrimethylammoniumpropane (DOTAP), a 10–15
mol% of PEG modication to mask the positively charged surface
of the MEND was needed for a prolonged circulation. To avoid
such an extensive PEG modication, we recently approached this
dilemma by designing a new pH-sensitive cationic lipid, YSK05.11
We reported that YSK05 had a high pH-responsive fusogenic
ability derived from the unique head group and fatty acid in the
preparation. A MEND that contained YSK05 has a higher gene-
knockdown ability than the commercially available transfection
reagent (Lipofectamine 2000) and demonstrated a signicant gene
reduction by intratumoral injection in tumor-bearing mice. We
Gene Silencing via RNAi and siRNA Quantification
in Tumor Tissue Using MEND, a Liposomal siRNA
Delivery System
Yu Sakurai1, Hiroto Hatakeyama1, Yusuke Sato1, Mamoru Hyodo1, Hidetaka Akita1 and
Hideyoshi Harashima1
1Faculty of Pharmaceutical Sciences, Hokkaido University, Sapporo, Japan
1195
1203
Gene Silencing via RNAi and siRNA Quantication
Molecular erapy
10.1038/mt.2013.57
original article
9April2013
21
6
22November2012
5March2013
Correspondence: Hideyoshi Harashima, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan. E-mail: harasima@pharm.hokudai.ac.jp
© e American Society of Gene & Cell erapy
Molecular Therapy vol. 21 no. 6, 1195–1203 june 2013 1195

© The American Society of Gene & Cell Therapy
Gene Silencing via RNAi and siRNA Quantification
also already claried that a MEND composed of YSK05 required
a relatively small amount of PEG modication to confer adequate
pharmacokinetics characteristics, because YSK05 was neutral in
the blood because of its appropriate pKa. is result suggests that
a MEND composed of YSK05 would be suitable for delivering
siRNA to tumor tissue. In this study, we rst attempted to identify
an appropriate lipid composition for the in vivo circumstance with
liver as a model organ. We then used the post-PEGylated YSK05-
MEND for targeting cancer, aimed at the gene-silencing eect in
tumor tissue via its systemic administration.
To develop a more ecient siRNA carrier, it is important
to analyze the biodistribution of siRNA aer systemic injection
and to determine the precise mechanism responsible for gene
reduction. A well-validated method for quantifying the amount
of siRNA in blood or in organs was recently reported.12 In the
report, the amount of siRNA, even if chemically modied or for-
mulated into a lipid nanoparticle, in the mouse liver aer systemic
injection could be measured with high sensitivity and selectiv-
ity. We applied this method in this study for the quantication
of siRNA in subcutaneously inoculated tumor tissue and evalu-
ated the eect on the eciency of siRNA delivery. In addition,
the o-target eect of siRNA can lead to articial phenotypes and
a complicated understanding of the therapeutic eects of siRNA.
e causes of o-target eects can be roughly classied into three
types: microRNA-like eect, immunostimulatory eect via toll-
like receptors (TLRs), and competition with the RNAi machinery
with endogenous microRNAs.13 For example, the immune stimu-
lation eect of siRNA can show a therapeutic eect via the produc-
tion of inammatory cytokines by inhibiting neovascularization.14
erefore, for development of an siRNA medicine, it is necessary
to distinguish between siRNA-induced gene silencing from non-
specic eects. We herein report that a MEND composed of our
unique cationic lipid, YSK05, can induce gene silencing via RNAi.
RESULTS
Construction of the YSK05-MEND
As we previously reported on the optimized lipid composition of the
YSK05-MEND (please refer to Supplementary Figure S1 on the
structural information of YSK05) was YSK05/POPE/chol/PEG2,000-
DMG=50/25/25/3 (POPE; 1-palmitoyl-2- oleoyl-sn-glycero-3-
phosphoethanolamine, PEG2,000-DMG; 1,2- dimyristoyl-sn-glycerol,
PEG) for in vitro transfection activity,11 we rst investigated the
issue of whether the optimized MEND resulted in a reduction in
the target gene in the liver, using the scavenger receptor B1 (Srb1) as
a target gene. e liver, in which liposomal carriers tended to accu-
mulate, was used as a model organ for a verication of the in vivo
potential of the MEND containing YSK05 for the sake of simplicity.
However, the MEND containing POPE, as a helper lipid, failed to
silence SrbI mRNA (Supplementary Figure S2a). We next tested
the gene-silencing eect of the POPE-based MEND in tumor tis-
sue aer systemic injection. We chose polo-like kinase 1 (PLK1) as
an anticancer gene, because it has been validated as a well-known
target for cancer therapy.15 When PEGylation was performed by
coincubating micelles of PEG2,000-DSG (PEG2,000-DSG; 1,2-distear-
oyl-sn-glycerol, PEG) with the MEND at 45 °C for 45 minutes in a
10% ethanol solution for using the EPR eect, the post-PEGylated
POPE-based MEND also showed no silencing eect in tumor tissue
(Supplementary Figure S2b). Phosphoethanolamine was judged
to be the inappropriate head group of the helper lipid, as previous
reports showed that the phosphoethanolamine head group was
more readily recognized by immune cells rather than hepatocytes16
Accordingly, we replaced POPE with 1,2-distearoyl-sn-glycero-
3-phosphocholine (DSPC) and improved the lipid ratio in accor-
dance with ndings reported from other investigators.17 As a result,
in the liver, DSPC-based MEND showed a signicant knockdown
(Supplementary Figure S2a). In summary, the basal lipid compo-
sition for in vivo applications was determined to be YSK05/choles-
terol/DSPC/PEG2,000-DMG at a ratio of 50/40/10/3.
As plasma a concentration of >5 mol% of PEGylation on the
surface of MEND was saturated, the amount of PEGylation was
determined to be 5 mol% (Supplementary Figure S3). e aver-
age z-average diameter and ζ-potential of the MENDs are shown
in Tab le 1. With (PEG-MEND) or without (MEND) 5 mol% PEG
insertion, their z-average diameters were nearly 100 nm and their
ζ-potentials were almost neutral. e particle size distributions of
both MENDs were homogenous, judged from the polydispersity
index (PdI). To investigate whether siRNA was assembled into the
lipid envelope, the siRNA encapsulated in MENDs was exposed
to 90% mouse serum that contained abundant levels of RNase for
periods of up to 48 hours (Figure 1). Both MENDs prevented the
degradation of the siRNA by RNase in plasma for periods of up to
48 hours. However, in the presence of Triton X-100, the formu-
lated siRNA was as rapidly degraded as free siRNA.
Table 1 Characteristics of the MENDs
Lipid
MEND PEG-MEND
YSK05/DSPC/
chol/PEG2,000-DMG
YSK05/DSPC/
chol/PEG2,000-DMG/
PEG2,000-DSG
50/10/40/3 50/10/40/3/5
z-average (nm) 104 ± 6 101 ± 8
Polydispersity index 0.16 ± 0.02 0.18 ± 0.03
ζ-potential (mV) 3.3 ± 1.5 2.6 ± 2.0
Data are represented by mean ± SD.
Figure 1 Stability of siRNA encapsulated into MENDs in mouse
serum. Naked and formulated siRNAs were coincubated with mouse
serum for 48 hours, and then analyzed the integrity of siRNA by TBE-
PAGE. MEND, multifunctional envelope-type nanodevice.
Tr iton +
Tr iton −
PEG-MEND
Tr iton +
Tr iton −
Free siRNA
036
Incubation time (hours)
24 48
MEND
1196 www.moleculartherapy.org vol. 21 no. 6 june 2013

© The American Society of Gene & Cell Therapy Gene Silencing via RNAi and siRNA Quantification
Pharmacokinetics analysis of systemically injected
MENDs and free siRNA
We previously reported on a method for evaluating the blood con-
centrations of both siRNA and a lipid envelope, using two radio
isotopes (RIs), [3H] and [32P] as a tracer for the lipid envelope
and siRNA, respectively. [32P]-labeled siRNA was synthesized by
conjugating 5′ end of siRNA with [32P]-phosphate, as previously
reported.10 In this study, we compared the siRNA concentration in
plasma between MEND and PEG-MEND aer intravenous injec-
tion into normal mice (Figure 2). Although the siRNA encapsu-
lated in the MEND was as rapidly eliminated from the blood stream
as free siRNA, the siRNA formulated in the PEG-MEND showed
a higher blood concentration (30% ID/ml plasma, 24 hours aer
injection). Moreover, the concentration of the lipid component
was equal to that of the siRNA in the PEG-MEND. e half-life in
β phase of free siRNA and the siRNA encapsulated in the MEND
or PEG-MEND calculated by tting to a 2-compartment model
were 1.5, 1.1, and 16.9 hours, respectively. Additional pharmacoki-
netics parameters are shown in Supplementary Table S1.
Tumor accumulation of encapsulated siRNA and lipid
envelope
To measure the amount of siRNA and lipid envelope that was accu-
mulated in tumor tissue, we prepared the tumor-bearing mice by
innoculating nude mice with OS-RC-2 cells (renal cell cancer)
on the right ank. A higher accumulation of lipid marker (4.0%
ID/g tumor) in PEG-MEND was detected than in the MEND
(Figure3a). Concerning the siRNA, 3.0% ID/g of tumor and 5.1%
ID/g of tumor had accumulated in the case of the MEND and
PEG-MEND, respectively (Supplementary Figure S5). To verify
the high accumulation in the MEND, we attempted to quantify the
siRNA accumulation by means of a stem-loop qRT-PCR method.
In the case of the PEG-MEND, 1.9% ID/g of tumor of siRNA had
accumulated, which was 240-times higher than that for the MEND
(Figure3b). To further investigate the distribution of formulated
siRNAs, tumor sections aer encapsulating uorescence-labeled
siRNA in the MEND were observed by confocal laser scanning
microscopy. Numerous red dots, indicative of siRNA, were obser ved
only in the PEG-MEND–treated group (Figure 3c–e).
Gene silencing in tumor tissue after the systemic
injection of MENDs
We compared the MEND and PEG-MEND with reference to their
in vivo gene-knockdown eects. Each MEND was injected into
the tail veins of OS-RC-2–bearing mice, and the mRNA expres-
sion of the target gene was measured by the qRT-PCR method. At
a dosage of 4 mg siRNA/kg of body weight, the clear reduction in
the target gene was observed only in the case of the PEG-MEND,
whereas the MEND failed to suppress PLK1 mRNA expression
(Figure 4). Next, a dose-response curve for mRNA expression was
constructed (Figure 5a). PLK1 expression was decreased in a dose-
dependent manner, and up to 3 mg/kg (a signicant gene silencing)
was observed, whereas a 5 mg/kg injection of control siRNA (anti-
luciferase, si-luc) assembled in the PEG-MEND failed to induce
any detectable target gene expression. A western blotting analysis
revealed that PLK1 protein levels were also suppressed aer an anti-
PLK1 siRNA (si-PLK1) intravenous administration (Figure 5b).
5′ RACE-PCR method for the evaluation of RNAi-
induced silencing
To exclude the possibility that the reduction in PLK1 expres-
sion was nonspecic, we carried out the RNA-ligase–mediated
rapid amplication of 5′ cDNA ends by the PCR (5′ RACE-PCR)
method. In the method, only the mRNA cleaved by siRNA, not
uncleaved mRNA, was detected. As shown in Figure 6a, nested
PCR was performed for a specic and substantial amplication
of a small amount of cleaved mRNA fragment. Two sequential
PCRs of cleaved mRNA would generate the 307 bp PCR product
that exhibits RNAi-induced silencing. Only in the group treated
with si-PLK1, about a 300 bp fragment was detected (Figure 6b).
To conrm that the fragment showed the cleaved mRNA, the
sequence of the PCR products was determined by DNA sequenc-
ing (Figure 6c). e predicted sequence was observed in each
sample of the si-PLK1–treated group.
Toxicological analyses
Finally, we tested the toxicity of the MENDs. As the liver is one of
the main clearance organs for liposomes, liver toxicity sometimes
becomes a limitation to therapeutic usage. Aer the intravenous
injection of both MENDs, the activities of asparate aminotrans-
ferase and alanine aminotransferase, which are typically used as
a marker of liver injury, were measured. In all treated groups, no
signicant change was observed compared with untreated mice
at 24 hours aer administration (Figure 7a). In addition, IL-6
was not produced by the injection of either MEND, whereas poly
I:C administration at a dose of 4 mg/kg induced abundant levels
of IL-6 (Figure 7b). ese results suggest that the PEG-MEND
shows no acute side eects.
DISCUSSION
As it has been reported that non-specic eects, such as the immune
stimulation of injected siRNA, could aect the therapeutic eect,18
Figure 2 Blood concentration profile of systemically injected
MENDs. The lipid envelope and siRNA were labeled with RIs [3H] and
[32P], respectively. MENDs and free siRNA were injected via the tail vein
of ICR mice, and then at 0.17, 1, 3, 6, and 24 hours after injection, the
radio activity in plasma was measured by liquid scintillation counting.
The data are represented as the mean ± SD (n = 3). MEND, multifunc-
tional envelope-type nanodevice; RI, radio isotope.
0.1
04812
Time after injection (hours)
16 20 24
1
10
Plasma concentration
(%ID/ml plasma)
100
Free siRNA
MEND–envelope [3H]
MEND–siRNA [32P]
PEG-MEND–envelope [3H]
PEG-MEND–siRNA [32P]
Molecular Therapy vol. 21 no. 6 june 2013 1197

© The American Society of Gene & Cell Therapy
Gene Silencing via RNAi and siRNA Quantification
it is important to conrm RNAi-mediated silencing in the targeted
tissue. Hence, we veried whether a MEND composed of a cationic
lipid YSK05 could deliver siRNA to tumor tissue and subsequently
silence the target gene expression via RNAi machinery.
We attempted to overcome the adverse eect of PEG modica-
tion on the ability to deliver siRNA in a subcellular level by using
a functional device, such as a PEG-peptide-DOPE (PPD; DOPE,
1,2-dioleoyl-sn-grycelophophoethanolamine) ternary conjugate
and a short GALA (shGALA).9,10,19 In the PPD structure, PEG is
conjugated with a lipid anchor via a peptide targeted by matrix
metalloproteinases, which are highly expressed in a broad range of
cancers. e PEG chain is removed in tumor tissue in response to
matrix metalloproteinases whereas PPD behaves like normal PEG
in the blood. Conversely, shGALA is a pH-sensitive fusogenic
peptide, whose sequence is optimized for in vivo tumor target-
ing. As shGALA has a characteristic repeat sequence, including a
protonable glutamic acid, the structure of shGALA changes from
a random coil to an α-helix in an acidic compartment of a cell,
such as endosomes and lysosomes. erefore, the fusogenic ability
of shGALA caused by the hydrophobicity conferred by the α-helix
form is able to overcome the inhibitory eect of PEG on endo-
somal escape. As these systems contained DOTAP that is a popu-
lar cationic lipid used for in vitro transfections,20 a large degree of
PEG modication was necessary for targeting a tumor via the EPR
eect. In fact, these devices improve the endosomal escape ability
of a PEG-modied MEND. To propose an alternate approach for
overcoming the issue caused by PEGylation instead of a peptide-
based solution, we reported on the possibility of using a newly
designed cationic lipid, YSK05. e eect of PEGylation on lipo-
somes composed of the conventional lipid, DOTAP, or YSK05
on fusion activity was evaluated to verify that YSK05 had suit-
able properties for tumor targeting, when intravenously injected.
When liposomes were modied with a sucient level of PEG to
stably circulate in the blood stream, a 5% PEG-modied YSK05
liposome had a higher membrane fusion ability than a 15% PEG-
modied DOTAP liposome (Supplementary Figure S4).
Despite the high transfection ability of the previously optimized
YSK05-MEND for intratumoral injection, no silencing activity
was observed in the liver aer systemic injection (Supplementary
Figure S2a). e injection of liposomes containing phosphoetha-
nolamine can activate a complement cascade via alternative path-
ways, and consequently the binding of complement protein to
the surface of liposomes should enhance their incorporation into
macrophages.16,21 ese reports suggest that an optimized POPE-
based MEND would be taken up by liver macrophages, kuper
cells, and hence, would not be able to deliver siRNA to hepatocytes.
Similarly, tumor tissue contains high levels of tumor-associated
macrophages.22 e POPE-based MEND failed to decrease the tar-
get gene expression in tumor tissue, possibly due to the fact that
large amounts were incorporated into these macrophages.
We previously reported that when tumor-bearing mice were
injected with 4 mg/kg of shGALA-modied, DOTAP-based
Figure 3 Tumor accumulation of MEND components. (a) Lipid envelope accumulation was evaluated by RI. MENDs were labeled with
[3H]-cholesteryl-hexadecyl ether (CHE), and then systemically injected into the tail vein of ICR mice. The amount of lipid envelope accumulation
was calculated from the count of [3H] in tumor lysate 24 hours after injection. (b) siRNA accumulation was determined by the stem-loop qRT-PCR
method. The amount of intact siRNA molecules was determined from a standard curve plotted by external siRNA in non-treatment tumor lysate.
**P < 0.01 (by two-tail unpaired t-test). The data are represented as the mean ± SD (n = 3). (c–e) CLSM observation of tumor tissue. Free and for-
mulated Cy5-labeled siRNA into MENDs were systemically administered into the tumor-bearing mice at a dose of 4 mg/kg, and 6 hours and tumor
tissue was collected after the injection. Thick tumor sections of 16 µm were prepared with a cryostat, and tumor sections derived from the mice
(c) untreated, (c) treated with MEND and (e) treated with PEG-MEND were observed by CLSM. Red and green correspond to Cy5-labeled siRNA
and fluorescein-isolectin B4 (endothelial cell), respectively. Scale bars: 40 µm. CLSM, confocal laser scanning microscopy; MEND, multifunctional
envelope-type nanodevice; PEG, polyethylene glycol; qRT-PCR, quantitative reverse transcriptase PCR; RI, radio isotope.
0
PEG-MENDMEND 0
0.5
1
1.5
2
2.5
3**
**
PEG-MENDMEND
1
2
3
Lipid accumulation (%ID/g tumor)
siRNA accumulation (%ID/g tumor)
4
5
a
cde
b
1198 www.moleculartherapy.org vol. 21 no. 6 june 2013

© The American Society of Gene & Cell Therapy Gene Silencing via RNAi and siRNA Quantification
MEND four times via the tail vein, a signicant silencing of the
target gene in tumor tissue somehow occurred. Conversely, just
once injection of the YSK05-based MEND at 4 mg/kg demon-
strated a signicant target gene reduction. Taken together, these
ndings suggest that the DSPC-based YSK05-MEND has more
appropriate properties for targeting tumors aer systemic injec-
tion through the EPR eect than the DOTAP-based MEND.
First, the in vitro and in vivo properties of the MEND and
PEG-MEND were evaluated. e siRNA formulated in MENDs
was stable in fresh mouse serum for a period of up to 48 hours,
but not in the presence of Triton X-100. e siRNA and lipid
envelope formulated in the PEG-MEND circulated in the blood
stream even 24 hours aer the administration in RI experiments,
whereas those formulated in the MEND did not, despite the sta-
bility of siRNA formulated in MEND against mice serum. ese
results show that only the PEG-MEND was able to exploit the EPR
eect for the delivery of siRNA to tumor tissue. When a similar
RI experiment was performed to measure the amounts of siRNA
that accumulated in tumor tissue, unexpectedly high amounts
of siRNA were detected in spite of the low lipid content of the
MEND (Figure 3a and Supplementary Figure S4). As it has been
previously reported that naked siRNA was unstable in blood,23
and consequently, should not be delivered to tumor tissue, this
result with RIs is likely to be articial. us, to clarify the cause of
the discrepancy in lipid and siRNA accumulation, we investigated
the biodistribution of [32P]-labeled naked siRNA aer systemic
administration (Supplementary Figure S5). A high [32P] count
was detected in liver, lungs, and tumor, unlike a previous report
(Supplementary Figure S6).23 In this latter report, in contrast to
our results, the authors reported that siRNA did not accumulate in
those organs in a full length, as evidenced by a northern blotting
analysis. erefore, there is a strong possibility that a catabolite of
the labeled siRNA was detected in our RI experiment on siRNA
biodistribution study.
en, to precisely determine the amount of siRNA that accu-
mulated in tumors, we performed stem-loop primer-mediated
qRT-PCR, because this method selectively detects full-length
siRNA.12 e results show that the PEG-MEND delivered 240-
fold more siRNA than the MEND (Figure 3b). Furthermore, in
the confocal laser scanning microscopy data, a high degree of u-
orescent signal of siRNA was detected in the tumor tissue treated
with the PEG-MEND, but not in the mice treated with the MEND
(Figure 3c–e). In the case of normal organs, the use of stem-loop
primer qRT-PCR showed that free siRNA did not accumulate in
the liver, spleen, kidney, and lung at all, which is consistent with
a previous report (Supplementary Figure S7). In addition, the
siRNA distribution with stem-loop primer data was more similar
to the biodistribution of the lipid envelop distribution than that
with the radio activity. ese results suggest that experiments on
organ distribution of siRNA with RIs include artifacts, and there-
fore, the stem-loop qRT-PCR is a more valid procedure for the
quantitative evaluation of siRNA accumulation not only in liver
but also in tumors and other organs.
In terms of mRNA knockdown, the PEG-MEND resulted
in about a 60% reduction in the target gene mRNA expression,
whereas MEND failed to show any gene-knockdown eect. is
result indicates that the improvement of siRNA accumulation in
tumor tissue was directly involved with the gene-silencing activity
in tumors. en, we quantitatively analyzed the eciency of siRNA
knockdown in tumor tissue. As 1 g of tumor tissue was estimated
to contain 1.0 × 108 − 109 cells,24 the number of siRNA molecules
in a single cancer cell is calculated to be 7.0 × 104 − 105 molecule/
cell. is estimated value was far higher than the value reported by
Landesman et al. to achieve a 50% knockdown of the target mRNA
in rat liver (~500 molecules/cell). e dierence in knockdown
ecacy must be the result of the eciency of the siRNA, the type
of target gene, lipid composition, and related factors. One of the
Figure 4 Comparison between the MEND and PEG-MEND on gene-
silencing activity. PLK1 mRNA expression was quantified 24 hours
after the systemic injection of the MENDs into the tumor-bearing mice
at a dose of 4 mg siRNA/kg bodyweight. The data are represented as
the mean ± SD (n = 3). **P < 0.01 (by one-way nrANOVA, followed by
Student Newman-Keuls correction). ANOVA, analysis of variance; MEND,
multifunctional envelope-type nanodevice; N.T., nontreatment; PEG,
polyethylene glycol.
0N.T. MEND PEG-MEND
**
**
0.2
0.4
0.6
0.8
Relative mRNA expression
PLK1/GAPDH
1
1.2
1.4
1.6
Figure 5 Gene-silencing effect of PEG-MEND. (a) mRNA level were
measured with qRT-PCR and (b) PLK1 protein expression was determined
by western blotting. PEG-MEND was injected into the tumor-bearing
mice at various dosages. *P < 0.05 (by one-way nrANOVA, followed by
Bonferroni correction, versus N.T.). ANOVA, analysis of variance; MEND,
multifunctional envelope-type nanodevice; N.T., nontreatment; PEG,
polyethylene glycol; qRT-PCR, quantitative reverse transcriptase PCR.
N.T.
N.T.
1.0 2.0 3.0 4.0 5.0 5.0 Dose
(mg/kg)
si-luc
si-luc
4 mg/kg
si-PLK1
4 mg/kg
PLK1
ACTB
si-PLK1
0.2
0.4
0.6
0.8
Relative mRNA expression
PLK1/GAPDH
1
1.2 *
a
b
Molecular Therapy vol. 21 no. 6 june 2013 1199

© The American Society of Gene & Cell Therapy
Gene Silencing via RNAi and siRNA Quantification
major factors contributing to this dierence might be the acyl chain
moiety of the PEG-lipid. It was reported that the fusion ability of
PEGylated liposomes decreased with the increasing length of the
acyl chain moiety, because the higher was the rate of elimination
out of the liposome membrane, the shorter was the acyl chain.25,26
In the delivery of nucleic acids, the length of the acyl chain is also
important. For example, Mok et al. reported 30-fold higher trans-
fection levels when a plasmid DNA encapsulated cationic lipo-
some with a C8 PEG-lipid was achieved, compared with that with
C14 PEG-lipid.27 e endosome escape ability of the PEG-MEND
should be relatively low, because the C18 PEG-lipid was used in
our study. In contrast, the C14 PEG-lipid was used in Landesman’s
study. is issue is specic to passive tumor targeting system via
EPR eect requiring a prolonged circulation of the carrier, and
consequently, must make siRNA delivery more dicult than liver.
Concerning the dose dependency, the gene-knockdown e-
cacy reached a plateau at 3 mg/kg. is saturation might be caused
by a restricted distribution of siRNA in tumor tissue. Actually, the
tumor distribution of the siRNA formulated in the PEG-MEND
appeared to be limited from the point of view of the confocal laser
scanning microscopy results (Figure 3e). e diusion of anti-
cancer medicine in tumor tissue is more dicult than that for
a normal organ because of the abundant extracellular matrix, a
long distance from vessels and cancer cells, and a high interstitial
uid pressure.28 e limitation in the tumor distribution is also
applicable to nano-sized carriers. Cabral et al. reported that small
micelles (30 nm in diameter) containing (1,2-diaminocyclo hex-
ane)platinum(II) are spread over the entire tumor tissue, whereas
large micelles (100 nm in diameter) remained near the tumor
vessels.29 erefore, the size of the PEG-MEND would be a rate-
determining step in gene reduction by siRNA in tumor tissue, and
consequently, a decrease in PEG-MEND diameter can overcome
the saturation in the RNAi eect.
Inhibition of the PLK1 gene is known to have the potential to
suppress the growth of cells in a wide range of cancers.15 In renal
cell carcinomas, a low-molecule weight inhibitor of PLK1 likewise
also delayed the growth, both in vitro and in vivo,30 and hence, it is
logical to regard si-PLK1 as a therapeutic gene against one of renal
cell carcinoma cell lines, namely, OS-RC-2 cells. Unexpectedly,
the continuous inhibition of PLK1 for a 2-week period failed to
inhibit OS-RC 2 tumor growth (data not shown). It is true that
PLK1 is a well-known representative anticancer gene, but PLK1
knockdown by Lipofectamine 2000 showed a weaker eect on
cell viability in OS-RC-2 cells (IC50 of viability; 100 nmol/l, IC50
of PLK1 mRNA; 0.1 nmol/l) than in another cell line (IC50 of
viability; 10 nmol/l, IC50 of PLK1 mRNA; 2 nmol/l in HeLa cell).
Further studies directed toward the elucidation of an appropriate
gene for killing renal cell carcinomas will be required for renal
Figure 6 Confirming RNA interference. (a) The outline of 5′ RACE-PCR method for PLK1 mRNA cleavage. Cleaved PLK1 mRNA resulted in the
307 bp nested PCR products. (b) An actual result of electrophoresis of nested PCR products. (c) Predicted sequence of nested PCR product around
siRNA cleavage site and the actual result of sequencing the PCR fragment. N.T., nontreatment.
Ligated RNA oligo
PLK1 mRNA
PLK1 mRNA GeneRacer RNA adaptor
siRNA cleavage cite
5′
5′3′
Sample
3′
- 5′3′ -
PLK1 siRNA
antisense strand
Predicted siRNA cleavage site
10 bp
Reverse transcription
307 bp RACE PCR product
1st PCR
2nd PCR
Reverse transcription
using gene specific primer
GeneRacer RNA adaptor
ligation
siRNA cleavage site
PLK1 mRNA
Cap
1 1,486-7 2,204
400 bp
300 bp
5.0 mg/kg
si-PLK1
5.0 mg/kg
si-luc
N.T.
a
b
c
1200 www.moleculartherapy.org vol. 21 no. 6 june 2013

© The American Society of Gene & Cell Therapy Gene Silencing via RNAi and siRNA Quantification
cell cancer treatment. In terms of the toxicity of MENDs, in both
groups treated with MENDs, a much lower amount of IL-6 was
detected than in the poly I:C group (Figure 7b) though inam-
matory cytokine production was investigated because it has been
reported that siRNA has an immunostimulatory eect, as it is rec-
ognized by the TLRs 3, 7, and 8 or RIG-I.31 e weak eect on an
innate immune system might be attributed to the low accumula-
tion in spleen and the high fusion ability of YSK05. Both MENDs
were mostly distributed in liver, little in spleen (Supplementary
Figure S8). Splenocyte captures liposome and removes it from
blood stream, and consequently produces inammatory cyto-
kines by recognizing extrinsic substances, such as nucleic acids.32
Followed by accumulation in immune cell, endosomal acidi-
cation and maturation are required for TLR expression. In fact,
chroloquin treatment that caused a destruction of endosomes
could decrease cytokine production in murine dendric cells.33
erefore, YSK05 might enable siRNA to escape before TLR mat-
uration from endosomes and avoid being recognized by TLR, and
consequently, could result in a low reduction of IL-6. is result
suggests that both MENDs are likely to be safe siRNA carriers in
terms of the absence of liver toxicity but also immune stimulation.
In conclusion, a MEND composed of YSK05 was used to
deliver siRNA to tumor tissue by intravenous injection and to
inhibit mRNA and protein expression of the target gene at a suf-
cient dose to permit its experimental use. Moreover, no adverse
eects, such as liver toxicity and immunostimulation, were
observed for the PEG-MEND. Collectively, an appropriately pre-
pared MEND can be a novel tool for the investigations of the in
vivo molecular biology of cancer and remains a viable basic tech-
nology for serving as an siRNA medicine in the future.
MATERIALS AND METHODS
Materials. DSPC; 1,2-dimyristoyl-sn-glycerol, methoxypolyethylene glycol
(PEG2000-DMG); and 1,2-Distearoyl-sn-glycerol, methoxypolyethylene gly-
col (PEG2000-DSG) were purchased from NOF corporation (Tokyo, Japan).
Cholesterol was obtained by SIGMA (St Louis, MO). [3H]-cholesteryl-
hexadecyl ether was purchased from PerkinElmer Life Sciences (Tokyo,
Japan). YSK05 was synthesized and puried as previously reported.11 siR-
NAs were purchased from Hokkaido System Science (Sapporo, Japan), and
the sequences of siRNAs in the reports were shown in Supplementary
Table S1. OS-RC-2 human renal cell carcinoma cells were kindly provided
by K Hida (Hokkaido University, Sapporo, Hokkaido, Japan).
MEND preparation and characterization. e MEND was prepared as
previously reported.11 Briey, an siRNA solution was added to an alco-
hol solution, in which YSK05, cholesterol, DSPC, and PEG2000-DMG
(50/40/10/0.03; mol% of total lipid) were dissolved. e alcohol was then
removed by ultraltration through an Amicon system (MWCO 50,000;
Millipore, Billerica, MA). e MEND was next incubated with 5 mol%
of PEG2000-DSG in a 10% ethanol solution for PEGylation. Ethanol was
again removed using the Amicon system (Millipore). e z-average and
the ζ-potential were determined using a Zetasizer Nano ZA ZEN3600
(MALVERN Instrument, Worchestershire, UK).
Serum resistance assay. Free or lipid-formulated siRNA (0.2 mg/ml) were
incubated in 90% (v/v) mouse serum volume at 37 °C up to 48 hours in the
presence or the absence of 0.1% Triton X-100. siRNA was extracted from
serum using acid phenol/chloroform method. Aliquots containing 22 ng
siRNA of each sample were loaded onto a 20% polyacrylamide gel, and
electrophoresis was performed to visualize the intact siRNA. Mouse serum
obtained within 6 hours was used for this assay.
Animal study and preparing tumor-bearing mice. Male ICR mice (4
weeks old) were obtained from Japan SLC (Shizuoka, Japan). For pre-
paring tumor-bearing mice, OS-RC-2 cells (Riken Cell Bank, Tsukuba,
Japan) were cultured in RPMI-1640 supplemented with 10% fetal bovine
serum, penicillin (100 U/ml), and streptomycin (100 µg/ml) at 37 °C in a
5% CO2 atmosphere. OS-RC-2 cells (1 × 106 cells) in 70 µl phosphate-bu-
ered saline were subcutaneously injected into male BALB/cAJcl-nu/nu
mice (CLEA Japan, Tokyo, Japan) on the right ank, and then grown
until the tumor volume was 80–150 mm3. e experimental protocols
were reviewed and approved by the Hokkaido University Animal Care
Committee in accordance with the guidelines for the care and use of labo-
ratory animals.
Analysis of plasma concentration profiles and biodistribution of siRNA
and the lipid envelope. e biodistribution of lipid and siRNA were mea-
sured with RIs, [3H] and [32P], respectively. e lipid envelope was labeled
with [3H]-cholesteryl-hexadecyl ether and [32P]-phosphorylated siRNA
was prepared, as previously reported.10 RI-labeled MENDs were formu-
lated by partially replacing components with RI-labeled ones in the prepa-
ration. e RI-labeled MEND was injected into ICR mice intravenously,
and blood, liver, spleen, kidney, lung, and tumor were then collected.
Serum was obtained by centrifugation. Approximately 0.1 g of each organ
Figure 7 Toxicological analyses of injected MENDs. (a) Liver toxicity
was evaluated by measuring AST and ALT activities in serum 6 hours
after injection of each sample. (b) Immunostimulatory effect of injected
MEND was evaluated by measuring IL-6 in serum. Poly I:C of 4 mg/kg of
was used as an immunostimulatory oligonucleotide. IL-6 concentration
was determined by ELISA at each time point. ALT, alanine aminotransfer-
ase; AST, asparate aminotransferase; MEND, multifunctional envelope-
type nanodevice; N.T., nontreatment; PEG, polyethylene glycol.
0
004812
Time after injection (hours)
Poly I:C
MEND
PEG-MEND
16 20 24
1,000
2,000
3,000
4,000
IL-6 concentration (pg/ml)
5,000
6,000
7,000
8,000
N.T. MEND PEG-MEND
AST ALT
PBS
5
10
15
20
25
Serum AST or ALT activity
(IU/I)
30
35
40
a
b
Molecular Therapy vol. 21 no. 6 june 2013 1201

© The American Society of Gene & Cell Therapy
Gene Silencing via RNAi and siRNA Quantification
was lysed in Soluene-350 (PerkinElmer) at 50 °C over night, and 10 ml of
Hionic Fluor (PerkinElmer) was added to lysate. RI counts of these sam-
ples were measured using an LSC-6100 (ALOKA, Tokyo, Japan).
siRNA quantification with stem-loop PCR. siRNA quantication using a
previously reported method.12,34 Frozen tumor tissue that was weighed in
advance was directly added into 500 µl of 0.25% Triton X-100 at 95 °C. e
tumor tissue was then homogenized with PreCellys (Bertin Technologies,
Montigny-le-Bretonneux, France). For preparing standard curves, known
amounts of serially diluted siRNA were added to the tissue homogeneate
from the non-treatment group.
Reverse transcription reactions were performed with TaqMan
MicroRNA Reverse Transcription kit (Applied Biosystems, Carlsbad,
CA). Tumor tissue homogeneates were heated at 95 °C for 10 minutes.
en, 5 µl of each sample was directly added into 10 µl of the reverse
transcription mixture which was placed at 4 °C beforehand, and the
reverse transcription program was started (16 °C × 30 minutes, 42 °C × 30
minutes, and 85°C × 30 minutes). cDNA of 2 µl of was mixed with 7 µl of
deionized distilled water, 1 µl of 20× Probe and Primer set, and 10 µl of 2×
TaqMan PCR Master Mix, No AmpErase UNG (Applied Biosystems). e
PCR parameters consisted of a primary denaturing at 95 °C × 10 minutes,
followed by 40 cycles of PCR at 95 °C × 15 seconds, 60 °C × 1 minutes with
LightCycler 480 II System (Roche Diagnostics GmbH, Germany).
Confocal laser scanning microscopy study. MENDs encapsulating Cy5-
labeled siRNA were injected into tumor-bearing mice. Isolectin B4 of
50µg of was intravenously injected into the tumor-bearing mice to visual-
ize endothelial cells 10 minutes before collection, and then tumor tissue
was excised. ick tumor sections of 16 µm were prepared using a cryo-
stat (CM3000; Leica, Tubingen, Germany), and then tissue images were
obtained by FV10i (Olympus, Tokyo, Japan).
Gene expression analysis in tumor tissue. Tumor tissue was bro-
ken with Precellys and total RNA was puried with TRIzol (Ambion,
Austin, TX) according to the manufacturer’s protocol. A total RNA of 1
µg was reverse-transcribed with a RNA-to-cDNA Reverse Transcription
Kit (Applied Biosystems) and 5 µl of 50 times diluted cDNA was then
subjected to PCR amplication with 2× Fast SYBR Green Master Mix
(Applied Biosystems) and 500 nmol/l of forward and reverse primer
sets, whose sequences were shown in Supplementary Table S2. Relative
PLK1 mRNA amount was calculated with ddCt method and normalized
to GAPDH mRNA amount.
Western blot analysis for protein expression in tumor tissue. Tumor tissue
was lysed in RIPA buer with Halt Protease Inhibitor Single-Use Cocktail
(ermo Fisher Scientic, Waltham, MA) on ice. e protein concentra-
tion in the tumor lysate was measured with BCA Protein Assay Reagent
(ermo Fisher Scientic), and then 2 µg aliquots of protein were subjected
to SDS-PAGE. Mouse monoclonal anti-PLK1 antibody (WH0005347M1-
100UG; SIGMA Aldrich) and anti-ACTB antibody (sc-130301; Santa Cruz
Biotechnology, Santa Cruz, CA) were used as the rst antibody, and ECL
Antimouse IgG, Horseradish peroxidase-linked species-specic F(ab’)2
fragment was used as the second antibody.
RNAi confirmation with 5′ RLM RACE-PCR method. Rapid amplication
of cDNA 5′ ends (5′ RACE)-PCR was performed referring to Gene Racer
protocol and previous reports.17,35 First, GeneRacer Adaptor, in which the
5′ ends were aminated to avoid self-ligation, was ligated into all RNA spe-
cies in total RNA solution extracted from tumor homogenate with T4 RNA
ligase (Ambion). en PLK1 mRNA was subjected to reverse transcribing
using PLK1 specic reverse transcription primer. PCR primer sets were
designed against the ligated RNA oligo and the mRNA sequence lower
than the expected siRNA cleavage site (Ad5 outer primer and PLK1 outer
primer). To obtain the PCR amplicon derived from cleaved target mRNA
specically, a second PCR was performed with primer sets against the
inner site rather than the rst PCR (nested PCR; Ad5 inner primer and
PLK1 inner primer). Nested PCR products were subjected to 10% TBE-
polyacrylamide gel electrophoresis. e nested PCR amplicon band cor-
responding to mRNA cleavage was extracted from the polyacrylamide
gel using QIAquick Gel Extraction Kit (QIAGEN, Hilden, Germany), and
then sequenced with ABI 3130 system using the PLK1 sequencing primer.
e sequences of oligonucleotides used in this experiment were shown in
Supplementary Table S2.
Toxicological test. Mice were intravenously injected with siRNA formu-
lated in MENDs at a dose of 4 mg siRNA/kg body weight once. Blood
samples were collected at various time points and allowed to stand at
4°C for coagulation. Serum was obtained by centrifuging the coagulated
blood at 3,000 rpm at 4 °C for 30 minutes. Cytokine levels were determined
using an Enzyme linked Immunosorbent assay (ELISA) kit for IL-6 (R&D,
Minneapolis, MN). Serum alanine transaminase and aspirate amino trans-
ferase activities were determined using a commercially available kit (Wako
Chemicals, Osaka, Japan).
Statistical analysis. Comparisons between multiple treatments were made
using one-way analysis of variance, followed by the Bonferroni test. Pair-
wise comparisons between treatments were made using the Student’s t-test.
A P value of <0.05 was considered to be signicant.
SUPPLEMENTARY MATERIAL
Figure S1. Structural information for YSK05. Structural formula of
the pH-sensitive cationic lipid, YSK05, whose IUPAC name is 1- methyl-
4,4-bis[[9Z,12Z]-ocatadeca-9,12-dien-1-yloxy]piperidine.
Figure S2. Gene-silencing effect of MENDs in liver and tumor.
Figure S3. Comparison between YSK05 and conventional cationic
lipid, DOTAP.
Figure S4. The fusogenic ability of YSK liposome and DOTAP lipo-
some was determined by means of a hemolysis assay.
Figure S5. [32P] accumulation in tumor tissue.
Figure S6. The biodistribution of systemically injected free [32P]-siRNA.
Figure S7. Comparison of the methodology between radio isotope
and stem-loop qRT-PCR methods.
Figure S8. The biodistribution of systemically injected [3H]-
cholesteryl-hexadecyl ether–labeled MEND and PEG-MEND.
Table S1. Pharamacokinetics parameters of siRNA encapsulated in
MENDs.
Table S2. Sequences of oligonucleotides.
ACKNOWLEDGMENTS
This study was supported in part by a grant-in-aid for Young Scientists
(B) from the Japan Society for the Promotion of Science (JSPS), the
Special Education and Research Expenses of the Ministry of Education,
Culture, Sports, Science, and Technology (MEXT) of Japan, a grant-
in-aid for Scientific Research on Innovative Areas “Nanomedicine
Molecular Science” (No. 2306) from MEXT of Japan, and by a grant
for Industrial Technology Research from New Energy and Industrial
Technology Development Organization (NEDO). The authors declare
no conflict of interest.
REFERENCES
1. Burnett, JC and Rossi, JJ (2012). RNA-based therapeutics: current progress and future
prospects. Chem Biol 19: 60–71.
2. Castanotto, D and Rossi, JJ (2009). The promises and pitfalls of RNA-interference-
based therapeutics. Nature 457: 426–433.
3. Whitehead, KA, Langer, R and Anderson, DG (2009). Knocking down barriers:
advances in siRNA delivery. Nat Rev Drug Discov 8: 129–138.
4. Shim, MS and Kwon, YJ (2010). Efficient and targeted delivery of siRNA in vivo. FEBS J
277: 4814–4827.
5. Kogure, K, Akita, H, Yamada, Y and Harashima, H (2008). Multifunctional envelope-
type nano device (MEND) as a non-viral gene delivery system. Adv Drug Deliv Rev 60:
559–571.
6. Maeda, H, Matsumura, Y and Kato, H (1988). Purification and identification of
[hydroxyprolyl3]bradykinin in ascitic fluid from a patient with gastric cancer. J Biol
Chem 263: 16051–16054.
1202 www.moleculartherapy.org vol. 21 no. 6 june 2013

© The American Society of Gene & Cell Therapy Gene Silencing via RNAi and siRNA Quantification
7. Ozpolat, B, Sood, AK and Lopez-Berestein, G (2010). Nanomedicine based
approaches for the delivery of siRNA in cancer. J Intern Med 267: 44–53.
8. Hatakeyama, H, Akita, H and Harashima, H (2011). A multifunctional envelope type
nano device (MEND) for gene delivery to tumours based on the EPR effect: a strategy
for overcoming the PEG dilemma. Adv Drug Deliv Rev 63: 152–160.
9. Hatakeyama, H, Akita, H, Ito, E, Hayashi, Y, Oishi, M, Nagasaki, Y et al. (2011).
Systemic delivery of siRNA to tumors using a lipid nanoparticle containing a tumor-
specific cleavable PEG-lipid. Biomaterials 32: 4306–4316.
10. Sakurai, Y, Hatakeyama, H, Sato, Y, Akita, H, Takayama, K, Kobayashi, S et al. (2011).
Endosomal escape and the knockdown efficiency of liposomal-siRNA by the fusogenic
peptide shGALA. Biomaterials 32: 5733–5742.
11. Sato, Y, Hatakeyama, H, Sakurai, Y, Hyodo, M, Akita, H and Harashima, H (2012). A
pH-sensitive cationic lipid facilitates the delivery of liposomal siRNA and gene silencing
activity in vitro and in vivo. J Control Release 163: 267–276.
12. Landesman, Y, Svrzikapa, N, Cognetta, A 3rd, Zhang, X, Bettencourt, BR,
Kuchimanchi, S et al. (2010). In vivo quantification of formulated and chemically
modified small interfering RNA by heating-in-Triton quantitative reverse transcription
polymerase chain reaction (HIT qRT-PCR). Silence 1: 16.
13. Jackson, AL and Linsley, PS (2010). Recognizing and avoiding siRNA off-target effects
for target identification and therapeutic application. Nat Rev Drug Discov 9: 57–67.
14. Kleinman, ME, Yamada, K, Takeda, A, Chandrasekaran, V, Nozaki, M, Baffi, JZ et al.
(2008). Sequence- and target-independent angiogenesis suppression by siRNA via
TLR3. Nature 452: 591–597.
15. Lens, SM, Voest, EE and Medema, RH (2010). Shared and separate functions of polo-
like kinases and aurora kinases in cancer. Nat Rev Cancer 10: 825–841.
16. Patel, HM and Moghimi, SM (1998). Serum-mediated recognition of liposomes by
phagocytic cells of the reticuloendothelial system - The concept of tissue specificity.
Adv Drug Deliv Rev 32: 45–60.
17. Judge, AD, Robbins, M, Tavakoli, I, Levi, J, Hu, L, Fronda, A et al. (2009). Confirming
the RNAi-mediated mechanism of action of siRNA-based cancer therapeutics in mice. J
Clin Invest 119: 661–673.
18. Robbins, M, Judge, A, Ambegia, E, Choi, C, Yaworski, E, Palmer, L et al. (2008).
Misinterpreting the therapeutic effects of small interfering RNA caused by immune
stimulation. Hum Gene Ther 19: 991–999.
19. Hatakeyama, H, Akita, H, Kogure, K, Oishi, M, Nagasaki, Y, Kihira, Y et al. (2007).
Development of a novel systemic gene delivery system for cancer therapy with a
tumor-specific cleavable PEG-lipid. Gene Ther 14: 68–77.
20. Simberg, D, Weisman, S, Talmon, Y and Barenholz, Y (2004). DOTAP (and other
cationic lipids): chemistry, biophysics, and transfection. Crit Rev Ther Drug Carrier Syst
21: 257–317.
21. Moghimi, SM and Hunter, AC (2001). Recognition by macrophages and liver cells
of opsonized phospholipid vesicles and phospholipid headgroups. Pharm Res 18:
1–8.
22. Joyce, JA and Pollard, JW (2009). Microenvironmental regulation of metastasis. Nat
Rev Cancer 9: 239–252.
23. Gao, S, Dagnaes-Hansen, F, Nielsen, EJ, Wengel, J, Besenbacher, F, Howard, KA et al.
(2009). The effect of chemical modification and nanoparticle formulation on stability
and biodistribution of siRNA in mice. Mol Ther 17: 1225–1233.
24. Del Monte, U (2009). Does the cell number 10(9) still really fit one gram of tumor
tissue? Cell Cycle 8: 505–506.
25. Holland, JW, Hui, C, Cullis, PR and Madden, TD (1996). Poly(ethylene glycol)–lipid
conjugates regulate the calcium-induced fusion of liposomes composed of
phosphatidylethanolamine and phosphatidylserine. Biochemistry 35: 2618–2624.
26. Silvius, JR and Zuckermann, MJ (1993). Interbilayer transfer of phospholipid-anchored
macromolecules via monomer diffusion. Biochemistry 32: 3153–3161.
27. Mok, KW, Lam, AM and Cullis, PR (1999). Stabilized plasmid-lipid particles: factors
influencing plasmid entrapment and transfection properties. Biochim Biophys Acta
1419: 137–150.
28. Heldin, CH, Rubin, K, Pietras, K and Ostman, A (2004). High interstitial fluid pressure -
an obstacle in cancer therapy. Nat Rev Cancer 4: 806–813.
29. Cabral, H, Matsumoto, Y, Mizuno, K, Chen, Q, Murakami, M, Kimura, M et al. (2011).
Accumulation of sub-100 nm polymeric micelles in poorly permeable tumours
depends on size. Nat Nanotechnol 6: 815–823.
30. Ding, Y, Huang, D, Zhang, Z, Smith, J, Petillo, D, Looyenga, BD et al. (2011).
Combined gene expression profiling and RNAi screening in clear cell renal cell
carcinoma identify PLK1 and other therapeutic kinase targets. Cancer Res 71:
5225–5234.
31. Robbins, M, Judge, A and MacLachlan, I (2009). siRNA and innate immunity.
Oligonucleotides 19: 89–102.
32. Wilson, KD, de Jong, SD and Tam, YK (2009). Lipid-based delivery of CpG
oligonucleotides enhances immunotherapeutic efficacy. Adv Drug Deliv Rev 61:
233–242.
33. Diebold, SS, Kaisho, T, Hemmi, H, Akira, S and Reis e Sousa, C (2004). Innate antiviral
responses by means of TLR7-mediated recognition of single-stranded RNA. Science
303: 1529–1531.
34. Cheng, A, Li, M, Liang, Y, Wang, Y, Wong, L, Chen, C et al. (2009). Stem-loop RT-PCR
quantification of siRNAs in vitro and in vivo. Oligonucleotides 19: 203–208.
35. Davis, ME, Zuckerman, JE, Choi, CH, Seligson, D, Tolcher, A, Alabi, CA et al. (2010).
Evidence of RNAi in humans from systemically administered siRNA via targeted
nanoparticles. Nature 464: 1067–1070.
Molecular Therapy vol. 21 no. 6 june 2013 1203