differences ☆ ⁎ Donald D. Rao

Advanced Drug Delivery Reviews 61 (2009) 746–759
Contents lists available at ScienceDirect
Advanced Drug Delivery Reviews
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / a d d r
siRNA vs. shRNA: Similarities and differences☆
Donald D. Rao a, John S. Vorhies a, Neil Senzer a,b,c,d, John Nemunaitis a,b,c,d,⁎
a
Gradalis, Inc., Dallas, TX, USA
Mary Crowley Cancer Research Centers, Dallas, TX, USA
c
Texas Oncology PA, USA
d
Baylor Sammons Cancer Center, Dallas, TX, USA
b
a r t i c l e
i n f o
Article history:
Received 23 January 2009
Accepted 13 April 2009
Available online 20 April 2009
Keywords:
RNA interference
Bi-functional
Cancer
Personalized
a b s t r a c t
RNA interference (RNAi) is a natural process through which expression of a targeted gene can be knocked
down with high specificity and selectivity. Using available technology and bioinformatics investigators will
soon be able to identify relevant bio molecular tumor network hubs as potential key targets for knockdown
approaches. Methods of mediating the RNAi effect involve small interfering RNA (siRNA), short hairpin RNA
(shRNA) and bi-functional shRNA. The simplicity of siRNA manufacturing and transient nature of the effect
per dose are optimally suited for certain medical disorders (i.e. viral injections). However, using the
endogenous processing machinery, optimized shRNA constructs allow for high potency and sustainable
effects using low copy numbers resulting in less off-target effects, particularly if embedded in a miRNA
scaffold. Bi-functional design may further enhance potency and safety of RNAi-based therapeutics. Remaining
challenges include tumor selective delivery vehicles and more complete evaluation of the scope and scale of
off-target effects. This review will compare siRNA, shRNA and bi-functional shRNA.
© 2009 Elsevier B.V. All rights reserved.
Contents
1.
2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . .
Targeted cancer gene therapy and RNA interference. . . . . .
2.1.
Personalized approach for cancer therapy . . . . . . .
2.2.
RNA interference for cancer. . . . . . . . . . . . . .
3.
Small interfering RNA (siRNA) and short hairpin RNA (shRNA)
3.1.
siRNA . . . . . . . . . . . . . . . . . . . . . . . .
3.2.
shRNA. . . . . . . . . . . . . . . . . . . . . . . .
3.3.
Bi-functional shRNA . . . . . . . . . . . . . . . . .
3.4.
Summary of si/sh/bi . . . . . . . . . . . . . . . . .
4.
Delivery . . . . . . . . . . . . . . . . . . . . . . . . . .
5.
Off-target effects . . . . . . . . . . . . . . . . . . . . . .
5.1.
Specific off-target effects . . . . . . . . . . . . . . .
5.2.
Nonspecific off-target effects . . . . . . . . . . . . .
6.
The future outlook . . . . . . . . . . . . . . . . . . . . .
7.
Conclusions . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
746
747
747
747
748
748
750
751
752
753
753
753
754
755
755
756
756
1. Introduction
☆ This review is part of the Advanced Drug Delivery Reviews theme issue on “Towards
Therapeutic Application of RNA-mediated Gene Regulation”.
⁎ Corresponding author. 1700 Pacific, Suite 1100, Dallas, Texas 75201, USA. Tel.: +1 214
658 1964; fax: +1 214 658 1992.
E-mail addresses: [email protected] (D.D. Rao), [email protected]
(J.S. Vorhies), [email protected] (N. Senzer), [email protected]
(J. Nemunaitis).
0169-409X/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.addr.2009.04.004
Cancer is a disease of genes, whether based on aberrant changes in
sequence or expression (epigenomics). The constellation of genetic
and epigenetic abnormalities characterizing cancer cells present new
and more specific targets for cancer treatment and, hopefully,
prevention. Over the last decade, the mapping of the human genome,
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
along with improved understanding of signal transduction and the
pathways responsible for tumor survival, have been transforming
therapeutic oncology from a more promiscuously targeted chemotherapeutic approach towards a highly selective targeted therapeusis. Indeed, in 1997, antibody-based Herceptin (trastuzumab)
became the first targeted therapy for breast cancer, specifically for
HER2-positive metastatic breast cancer. In 2001, the small molecule
Gleevec (imatinib mesylate), became the first approved kinase
inhibitor for cancer targeting bcr-abl in chronic myeloid leukemia
(CML), and it has since been approved for the treatment of
gastrointestinal stromal tumors (GIST) targeting c-kit. Over the last
few years, several other targeted cancer therapies have been
approved. Targeted therapeutics directed against amplified genes
and/or over-expressed proteins in malignant cells have proven to be
powerful tools for cancer treatment. A growing understanding and use
of proteomic, genetic, and pharmacogenomic tools are actualizing a
long desired concept of personalized cancer therapy. Genetic
abnormalities of each patient's tumor can be analyzed through a
variety of established means to quantitatively determine both gene
and protein over- and under-expression. Moreover, functional pathways can be determined and integrated within the cancer network
allowing for the identification of key molecular relays enabling
experimental testing of target specific therapeutics. Such information
can potentially allow medical care takers to prioritize, if not yet
optimize, treatment for cancer patients, and to uncover surrogate
biomarkers for prognosis, prediction, and therapy assessment.
The recent discovery of RNA interference (RNAi), a natural process
through which the expression of a targeted gene can be knocked down
with high specificity and selectivity, presents an invaluable tool for
personalized cancer therapy. Target specific RNAi agents have the
potential to selectively knockdown key abnormally over- or constitutively expressed molecular targets that are essential for the survival of
each patient's tumor for effective personalized cancer treatment.
Conceptually, target specific RNAi agents can also be applied in
combination with immune modulating agents or small molecules to
improve the efficacy of cancer treatment. Like other new therapeutic
paradigms, there are a multitude of issues which need to be addressed
in order for us to translate RNA interference technology (siRNA,
shRNA, bi-functional RNA) from laboratory to bedside and from
concept to reality. These issues include comparison of each of the RNAi
technologies with respect to effective delivery, possible off-target
effects and the pharmacokinetics and pharmacodynamics. This review
will focus on several issues currently confronting clinical development
of RNAi therapeutics. We will discuss the role of RNAi technology in
personalized cancer gene therapy and address clinical considerations
of appropriate RNAi-based therapeutic agents for cancer.
2. Targeted cancer gene therapy and RNA interference
2.1. Personalized approach for cancer therapy
Most human tumors manifest gene expression patterns that differ
not only from their normal counterparts but, to a lesser extent, even
from each other based on both intrinsic gene modifications and
modulated cancer cell–matrix interactions. Such variability in genetic
patterns found between histologically identical tumors arising in
different patients may well explain the widely divergent responses to
the standard treatment regimens most often prescribed for a particular
tumor type. This is supported by recent studies demonstrating that
certain patterns of genetic expression (i.e. expression signatures)
identified in tumor samples from patients with breast cancer are not
only strongly correlated with prognosis [1–4] but can actually be subclassified into differing prognostic categories [5]. The presence of
functional redundancy in a robust, predominantly scale-free network
such as cancer “buffers” the effect of any single gene/target modification
on the malignant process, with rare exception (e.g., CML). The hierarchy
747
of cancer scale-free networks does not have a threshold for single target
disintegration insofar as random pathway component failure predominantly affects targets with low connectivity within the network, thereby
having limited functional impact. However, highly connected information-transfer targets do allow for “attack vulnerability.” In other words,
the disordered circuitry characteristic of malignancy results in a change,
such that the otherwise robust oncogenic process can become, almost
paradoxically, more highly dependent on a specific rewired pathway
(i.e., “pathway addiction”). Conceptually, knockout of those “rewired”
tumor specific oncogenic pathways should produce a lethal effect on
cancer cells yet not significantly perturb normal cell functionality.
Genetic diversity of cancer, pathway addiction and targeted therapy are
not the subject of this review and have been extensively reviewed
elsewhere [6–8]; here, we discuss how RNAi-based therapeutics can be
best applied in light of these mechanisms.
For example, we harvested tumor and normal cells from the lymph
nodes of a melanoma cancer patient for molecular profiling by
microarray and proteomic analysis [9]. Expression profiles for
malignant versus normal tissue were compared at the mRNA and
protein levels. The goal was to identify a group of gene and protein
doublets differentially over-expressed in that individual's malignancy.
Fig. 1 is a comparative analysis of protein expression profile by the
two-dimensional difference gel electrophoresis (2D-DIGE) analysis.
2D-DIGE can very effectively identify several over-expressed proteins
in the patient's tumor. Correlated DNA/RNA over-expression can then
be confirmed with microarray data. The resulting data is further
analyzed by a modeling and simulation computational system
developed by our team specifically for clinical application including,
but not limited to, gene set enrichment analysis and network
inference modeling platforms (Fig. 2). Grouped gene expression
patterns that are highly correlated with the pathway phenotype in an
individual patient allow target genes to be selected, and prioritized
based on connectivity and vector-driven criteria developed analytic
network algorithms. Once this individual “cancer fingerprint” is
created, it serves as the template for the design, synthesis, and
subsequent validation of individualized therapeutic RNAi molecules
with knockdown activity against these “high-degree hub” genes for a
personalized and targeted cancer gene therapy.
2.2. RNA interference for cancer
The concept of antisense oligodeoxynucleotides as modulators of
gene expression and their application in targeted cancer gene therapy
was developed more than 25 years ago (for review, [10]). By processes
still unknown, the antisense nucleic acid (ASNA) strand and the mRNA
target come into proximity leading to the destruction of the mRNA
target either by endogenous nucleases, such as RNase H [11,12] that
are recruited into the mRNA–ASNA duplex or by intrinsic enzymatic
activity engineered into the ASNA sequence, as is the case with
ribozymes [13,14] and DNAzymes [15,16]. The discovery of an
evolutionarily conserved gene silencing mechanism whereby small
sequences of extrinsic dsRNA or intrinsic microRNA inhibit complementary post-transcriptional mRNA (siRNA) or suppress translation
(miRNA), respectively, ignited strong hope that the natural gene
silencing process would be specific and robust. The silencing process
occurs following interaction of the RNA effector precursors with the
RNase III enzymes Drosha (for miRNA) and Dicer (for miRNA and
siRNA) and subsequent formation of the RNA-interfering silencing
complex (RISC) [17]. Endonucleolytic cleavage of the target mRNA
occurs at a single site ~10 nucleotides from the 5′ end of the guide
(antisense) siRNA sequence [18,19]. RNAi offers several advantages
over antisense and ribozyme approaches, including ease of synthesis
[20] and greater activity [18,21–24].
Preclinical studies confirm that RNAi techniques can be used to
silence cancer-related targets [25–35]. In vivo studies have also shown
favorable outcomes by RNAi targeting of components critical for
748
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
Fig. 1. Analysis of the 2-D DIGE images by DeCyder Software and mass spectrometry protein identification. The upper panel shows on the left the protein expression pattern of a
normal lymph node from patient RW following 2-D gel electrophoresis. In the middle of the upper panel the protein expression pattern of a malignant lymph node from patient RW is
shown and to the right is an overlaid image with proteins from the normal lymph node labeled with Cy3 (green) and proteins from the malignant lymph node labeled with Cy5 (red).
Circles indicate protein spots with significant expression level changes. The upper right panel shows the fold-of-change distribution curve and the level of change for the spots of
interest. The middle panel shows the 3D view of one protein spot change between the normal and malignant lymph nodes. Mass spectrometry (lower panel) subsequently (following
robotic spot picking) identified this protein as RACK1.
tumor cell growth [26,36–39], metastasis [40–42], angiogenesis
[43,44], and chemoresistance [45–47].
3. Small interfering RNA (siRNA) and short hairpin RNA (shRNA)
The applications of RNAi can be mediated through two types of
molecules; the chemically synthesized double-stranded small interfering RNA (siRNA) or vector based short hairpin RNA (shRNA). Effective
RNAi was initially demonstrated by the application of synthetic siRNA
[48]; later, siRNA produced in vitro by T7 RNA polymerase was found to
be active and it was soon demonstrated that active siRNA consists of a
hairpin structure can be transcribed in cells from an RNA polymerase III
promoter on a plasmid construct [49,50]. Although siRNA and shRNA
can be applied to achieve similar functional outcomes, siRNA and shRNA
are intrinsically different molecules. Therefore, the molecular mechanisms of action, the RNA interference pathways, the off-target effects and
the applications can also be different.
3.1. siRNA
Fluorescent labeled siRNA has been used to trace the fate of
delivered siRNA. A fluorescent label can either be tagged onto the 5′
end or the 3′ end of the siRNA. Using a fluorescence resonance energy
transfer (FRET)-based visualization method, the intact siRNA can be
observed to be translocated into the nucleus within 15 min of the
delivery and then disseminated into the cytoplasm within the next 4 h
both in intact and dissociated form [51]. The initial accumulation of
siRNA in the nuclei is similar to observations made on the behavior of
antisense olignucleotides [52]. The translocation of antisense oligonucleotide into the nuclei was not dependent on either the ATP pool or
temperature and thus may not involve the active import transport
system of the nuclear pore [52]; it is not clear whether siRNA
translocate into the nuclei using the same mechanism as antisense
olignucleotides. Using HeLa cells and targeting 7SK snRNA which is
exclusively located in the nucleus, Robb showed the efficiency of
siRNA mediated silencing to be greater than that effected by antisense
7SK DNA [53]. Berezhna et al. observed nuclear localization of siRNA
targeted against small nuclear RNA (snRNA) and cytoplasmic
localization of siRNA targeted against viral mRNA suggesting selective
localization and compartmentalization of siRNA based on its intended
target [54]. Ago1 and Ago2 containing RISC were found both in the
cytoplasm and nucleus [55–57]. A recent study using fluorescence
correlation spectroscopy and fluorescence cross-correlation spectroscopy (FCS/FCCS) to correlate the presence of siRNA with Ago2
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
749
Fig. 2. (A) Nearest neighbor protein–protein (first order) interactions of the 6 prioritized proteins in VisualCell. Second order interactions for Stathmin1 (SDCBP) (B) and RACK1
(GNB2L1) (C).
protein, indicated a shuttling of RISC between nucleus and cytoplasm
[58]. Importin 8 (Imp8) binds to all Ago proteins in a Ran-dependent
manner, but independently of RNA [59]. Knockdown of Imp8 results in
a shift of Ago2 from the nucleus to the cytoplasm without affecting the
total quantity of Ago2. However, although Imp8 is not required for
target mRNA cleavage it is necessary for Ago2 binding to miRNA
targets. In Caenorhabditis elegans, the argonaute protein NRDE-3 is
essential for binding nuclear RNAs and appears to interact with
cytoplasmic siRNAs generated by RNA-dependent RNA polymerase
(RdRP) followed by redistribution to the nucleus [60]. The nuclear
RISC (nRISC) is a complex that is 20× smaller in size than the cytoplasmic RISC (cRISC). The nucleus may be the check point controlling
distribution as either the nuclear acting siRNA or the cytoplasmic
acting siRNA.
Dynamically, siRNA steadily increases its accumulation in cells for
4 h before plateau [61]. The steady-state nuclear distribution of siRNA
was mainly found in the nucleolus region and was excluded from the
nucleoplasm [62]. The cytoplasmic distribution of siRNA appears to be
in the perinuclear region forming a ring-like pattern around the
nucleus [63]. The nucleolus and perinuclear regions are possibly the
main site for RNAi. However, Ohrt et al. labeled siRNA with fluorescent
dye at the 3′ end of either strand of siRNA and did not find
accumulation of siRNA at the perinuclear region, but rather evenly
distributed throughout the cytoplasm [62]. This discrepancy may be
the result of the fluorescent tagging process. At 48 h post injection, the
majority of siRNA appears to have been degraded with only 1%
fluorescence remaining in the cell. The spatial and temporal distribution of siRNA within the cell is in accord with the observed kinetics of
siRNA mediated RNA interference activity which peaks around 24 h
post delivery and diminishes within 48 h.
The life-cycle of siRNA inside transfected cells is diagrammatically
illustrated in Fig. 3. In Drosophila, double-stranded RNA-binding
proteins (dsRBPs), such as R2D2 and Loquacious (Loqs), function in
tandem with Dicer (Dcr) enzymes in RNA interference (RNAi) [64–
66]. Dcr-1/Loqs and Dcr-2/R2D2 complexes generate microRNAs
(miRNAs) and small interfering RNAs (siRNAs), respectively. Thus,
Loqs and R2D2 represent two distinct functional modes for dsRBPs in
the RNAi pathways [67]. In mammalian cells, only one Dicer gene has
thus far been identified [68]. Human Dicer is an integral component of
the RNA interference pathway. Dicer processes pre-microRNA and
double-strand RNA (dsRNA) to mature miRNA and siRNA, respectively, and transfers the processed products to the RISC [69,70]. Since
there is only one Dicer in the human, the RNA-interfering pathway for
siRNA and for miRNA may not be as compartmentalized as for
Drosophila. Dicer is a multi-domain RNase III-related endonuclease
responsible for processing dsRNA to siRNAs [71]. Dicer preferentially
binds to the 5′ phosphate of 2 nt 3′ overhang and cleaves dsRNAs into
21 to 22 nucleotide siRNAs [72,73]. Mammalian Dicer interacts with
the double-stranded Tat–RNA-binding protein (TRBP) or PACT (PKR
activating protein) to mediate RNA interference and miRNA processing. TRBP and PACT are structurally related but exert opposite
regulatory activities on RNA-dependent protein kinase (PKR). Knockdown of both TRBP and PACT in cultured cells leads to significant
inhibition of gene silencing mediated by short hairpin RNA but not by
siRNA, suggesting that TRBP and PACT function primarily at the step of
siRNA production [74]. Human TRBP and PACT directly interact with
750
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
Fig. 3. Schematic of the siRNA mediated RNA interference pathway. After entry into the cytoplasm, siRNA is either loaded onto RISC directly or utilize a Dicer mediated process. After
RISC loading, the passenger strand departs, thereby commencing the RNA interference process via target mRNA cleavage and degradation. siRNA loaded RISCs are also found to be
associated with nucleolus region and maybe shuttled in and out of nucleus through an yet unidentified process.
each other and associate with Dicer to stimulate the cleavage of
double-stranded or short hairpin RNA to siRNA [74]. Dicer knockout ES
cells can effectively load processed siRNA onto RISC and carry out RNA
interference as efficiently as Dicer+ ES cells [68]. So, it appears that in
mammalian cells, a perfectly processed siRNA can be effectively
loaded onto RISC for RNAi without the help of the TRBP/PACT/Dicer
complex. The TRBP/PACT/Dicer complex, however, is required to
process either shRNA or long dsRNA to appropriate size and form for
their loading onto RISC.
Duplex siRNA in association with holo-RISC, composed of at least
Ago-2, Dicer and TRBP, is identified as the RISC loading complex (RLC)
[75]. In the RLC, the two strands of the duplex are separated, resulting in
the departure of the passenger strand [76–78]. The passenger strand is
cleaved by the RNase-H like activity of Ago-2, provided there are
thermodynamically favorable conditions for passenger strand departure. This is referred to as the cleavage-dependent pathway [79]. There is
also a cleavage-independent by-pass pathway, in which the passenger
strand with mismatches is induced to unwind and depart by an ATP
dependent helicase activity [76,79,80]. The RISC with single-stranded
guide strand siRNA is then able to execute multiple rounds of RNA
interference. ATP is not required for shRNA processing, RISC assembly,
cleavage-dependent pathway, or multiple rounds of target-RNA
cleavage [81–83]. Single-stranded siRNA (containing 5′-phosphates)
and pre-miRNA can be loaded on RISC, but not duplex siRNA [84].
3.2. shRNA
shRNAs, as opposed to siRNAs, are synthesized in the nucleus of
cells, further processed and transported to the cytoplasm, and then
incorporated into the RISC for activity [85]. The life-cycle of shRNA
inside of transfected cells is diagrammatically illustrated in Fig. 4. To
be effective, the shRNA are designed to follow the rules predicated by
the specifics of the cellular machinery and are presumably processed
similar to the microRNA maturation pathways. Thus, studies on the
synthesis and maturation of miRNAs have provided the groundwork
for the synthesis of shRNA [86], particularly the miR-30 based shRNAs
[87].
shRNA can be transcribed by either RNA polymerase II or III
through RNA polymerase II or III promoters on the expression cassette.
The primary transcript generated from RNA polymerase II promoter
contains a hairpin like stem-loop structure that is processed in the
nucleus by a complex containing the RNase III enzyme Drosha and
the double-stranded RNA-binding domain protein DGCR8 [88]. The
complex measures the hairpin and allows precise processing of
the long primary transcripts into individual shRNAs with a 2 nt 3′
overhang [89]. The processed primary transcript is the pre-shRNA
molecule. It is transported to the cytoplasm by exportin 5, a Ran-GTPdependent mechanism [90,91]. In the cytoplasm the pre-shRNA is
loaded onto another RNase III complex containing the RNase III
enzyme Dicer and TRBP/PACT where the loop of the hairpin is
processed off to form a double-stranded siRNA with 2 nt 3′ overhangs
[92–94]. The Dicer containing complex then coordinates loading onto
the Ago2 protein containing RISC as described earlier for siRNA. PreshRNA has been found to be part of the RLC; thus, pre-shRNA may
potentially directly associate with RLC rather than through a two steps
process via a different Dicer/TRBP/PACT complex [95].
After loading onto RLC and passenger strand departure; both siRNA
and shRNA in the RISC, in principle, should behave the same. The
argonaute family of proteins is the major component of RISC [96,97].
Within the Argonaute family of proteins, only Ago2 contains the
endonuclease activity necessary to cleave and release the passenger
strand of the double-stranded stem [76,77,79]. The remaining three
members of Argonaute family, Ago1, Ago3 and Ago4, which do not
have identifiable endonuclease activity, are also assembled into RISC
and presumably function through a cleavage-independent manner.
Thus, RISC can be further classified as cleavage-dependent and
cleavage-independent [79].
The argonaute family of proteins in RISCs are not only involved in the
loading of siRNA or miRNA, but also implicated in both transcriptional
(targeting heterochromatin) and post-transcriptional gene silencing.
Ago protein complexes loaded with passenger strandless siRNA or
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
751
Fig. 4. Schematic of the shRNA mediated RNA interference pathway. After delivery of the shRNA expression vector into the cytoplasm, the vector needs to be transported into the
nucleus for transcription. The primary transcripts (pre-shRNA) follow a similar route as discovered for the primary transcripts of microRNA. The primary transcripts are processed by
the Drosha/DGCR8 complex and form pre shRNAs. Pre-shRNAs are transported to the cytoplasm via exportin 5, to be loaded onto the Dicer/TRBP/PACT complex where they are
further processed to mature shRNA. Mature shRNA in the Dicer/TRBP/PACT complex are associated with Argonaute protein containing RISC and provide RNA interference function
either through mRNA cleavage and degradation, or through translational suppression via p-bodies.
miRNA seeks out complementary target sites in mRNAs, where
endonucleolytically active Ago-2 cleaves mRNA to initiate mRNA
degradation [98,99]. Other Ago protein containing complexes without
endonucleolytic activity predominantly bind to partially complementary target sites located at the 3′ UTR for translation repression through
mRNA sequestration in processing bodies (p-bodies) [100–102]. The
detailed mechanism of mRNA sequestration in p-bodies and later
release from p-bodies is still a debated issue; deadenylation of the target
mRNA which leads to destabilization of the mRNA was also observed to
occur in p-bodies [103,104]. Coimmunoprecipitation experiments
determined that RISCs are also strongly associated with polyribosomes
or the small subunit ribosomes [95] and Ago-2 (actually identified as
elF2c2), strongly suggesting that RISC surveillance is compartmentalized with translational machinery of the cell. Details of the mechanism
involving mRNA scanning and target mRNA identification are still
largely unknown. Whatever the scanning or surveillance mechanism
may be, once the target mRNA is identified, the target mRNA is either
cleaved or conformationally changed following which both types of
structures are routed to the p-body for either sequestration or
degradation [103,104]. The active siRNA or miRNA loaded complex is
then released for additional rounds of gene silencing activity.
3.3. Bi-functional shRNA
There is, however, a third unique RNAi option in development
called bi-functional shRNA. shRNA can potentially be manipulated to
take advantage of the gene silencing machinery within the cells to
improve its efficiency and durability of action. Conceptually, targeted
shRNAs can be designed so as to effectively load shRNA onto both the
cleavage-dependent and the cleavage-independent RISCs. This differential processing is mediated by two pathways primarily dependent
on strand complementarity and/or access to RNase-H cleavage and,
presumably, for final target effect, on interaction with Imp8.
Simultaneous expression of both types of shRNAs (i.e. the bi-
functional shRNA) in cells should achieve a higher level of efficacy,
greater durability compared to siRNA, and a more rapid onset of gene
expression silencing (the rate dependent on mRNA turnover and
protein kinetics) compared to shRNA as illustrated on Fig. 5. The “bifunctional” shRNA, by virtue of loading onto multiple types of RISCs, is
thus able to simultaneously induce degradation of target mRNA and
also inhibit translation through mRNA sequestration. This bi-functional design should be, in principal, much more efficient for two
reasons; first, the bi-functional will promote loading the guide strand
onto at least two types of RISCs to increase activity; second, by loading
onto both cleavage-dependent RISC and cleavage-independent RISC,
target mRNA can be silenced both through mRNA degradation and
translational inhibition or sequestration.
The design of the bi-functional shRNA expression unit consists of two
stem-loop shRNA structures; one stem-loop structure composed of fully
matched passenger and guide strand for cleavage-dependent RISC
loading, the second stem-loop structure composed of mis-matched
passenger strand (at the positions 9–12) for cleavage-independent RISC
loading.
There are several experimental observations that support this
approach. In Drosophila, Ago1 preferentially binds to miRNAs that
have been excised from imperfectly paired hairpin precursors, whereas
those miRNAs that have near-perfectly paired hairpin precursors are
bound by Ago2 [105–108]. In HEK293 cells transfected with tagged-Ago
proteins, coimmunoprecipitation found similar sets of about 600
transcripts to be bound to Ago1, 2, 3 or 4 [95], suggest all four
mammalian Ago protein containing RISCs are involved in the RNAi
function. Insofar as most mRNA have multiple miRNA target sites (with
distance constraints) at their 3′ UTR, the miRNA mediated RNAi system
appears to be redundant for the targeted mRNAs allowing for
cooperative downregulation to ensure target mRNA knockdown. The
bi-functional shRNA approach mimics the natural process by mediating
target mRNA knockdown through multiple RNAi pathways and
complexes.
752
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
Fig. 5. Schematic of the bi-functional shRNA concept. The bi-functional concept is to design two shRNAs for each targeted mRNA; one with perfect match, one with mismatches at the
central location (bases 9–12). The purpose of the bi-functional design is to promote loading of mature shRNAs onto both cleavage-dependent and cleavage-independent RISCs, so that
the expression of target mRNA can be more effectively and efficiently shut down both through target mRNA degradation and through translational repression.
In C. elegans, structural features of small RNA precursors determine
Argonaute loading [109]. Recently, Azuma-Mukai et al. observed
miRNAs associated with hAgo-2 and hAgo-3 have some overlaps;
however, some are discriminately loaded onto hAgo-2 or hAgo-3
[110]. Further work is needed to resolve the specificity of miRNA
loading onto different Ago containing RISCs. Although most miRNA
target sites have been identified to be located at the 3′-UTR region,
recent systemic identification of mRNAs recruited to hAgo-2 have
identified many mRNAs with target sites located at the coding region
and some at the 5′-UTR [111]. hAgo-2 could initiate the target mRNA
degradation with its slicing activity in the coding region. Tay et al.
recently found many naturally occurring miRNA targets are located in
the coding region of embryonic regulated genes to modulate
embryonic stem cell differentiation [112], further support that
miRNA can act through mRNA regions other than 3′-UTR.
3.4. Summary of si/sh/bi
In summary, exogenously introduced siRNA with appropriate
length (19–21 nt) and 2 nt 3′ overhang, can be loaded onto RISC for
RNAi function without interacting with either Dicer, TRBP or PACT;
however, the loading process is10× less efficient than shRNA.
Increasing the length of the siRNA duplex to 29–30 nt with a 2 nt 3′
overhang only at one end of the duplex (specifically antisense [113])
appears to improve efficacy [114]. If so, it is possibly because
increasing the length of an siRNA duplex with an unprocessed end
forces directionality as a result of imposed thermodynamic instability
determining the guide strand motif and thereby enhancing its
association with Dicer/TRBP/PACT complex for more efficient loading
onto RLC [114]. shRNA, on the other hand, assimilates into the
endogenous miRNA pathway and in so doing is significantly more
efficient [115–117]. Additionally, fluorescent tagged siRNA tracing
indicated high degradation and turnover of exogenously introduced
siRNA. Less than 1% of the introduced duplex remains in the cell 48 h
after administration. shRNA can be continuously synthesized by the
host cell, therefore, its effect should be much more durable.
Concentrations necessary for effective knockdown are usually in the
low nM range for most siRNAs, while less than 5 copies of shRNA
integrated in the host genome is sufficient to provide continual gene
knockdown effect (Cleary M, personal communication). The higher
dose required for siRNA can further contribute to the off-target effects
to be discussed later.
Understanding the mechanism and the dynamics of siRNA and
shRNA at the cellular and molecular level greatly enhanced the effort
in developing therapeutic siRNAs for various diseases. As a result,
modifications can be made to improve the efficacy and stability of
RNAi agents. Chemically synthesized siRNA is easier to modify
through chemistry; however, bulk manufacturing of complex structures such as modified siRNA is more expensive. Vector based shRNA
relies on the host machinery for expression, but, on the other hand, is
more difficult to modify. Modification can only be achieved through
manipulating expression strategy (e.g. bi-functional shRNA), redesigning shRNA structure, or by varying promoter regulation. The
shRNA expression units can be incorporated into varieties of plasmids
and viral vectors for delivery and integration. In addition, vector based
shRNA expression can also be regulated or induced [118–120].
Numerous siRNAs have been demonstrated to be effective for invivo tumor growth modulations via intratumoral, ex-vivo, or systemic
routes of application (For review see [121–123]. Vector based shRNA
has, likewise, demonstrated in-vivo effectiveness (For review see
[121,123,124]. In-vivo studies employ a variety of delivery methods
and may not ensure equivalency of strand biasing; therefore, it is hard
to perform direct comparison between siRNA and shRNA. McAnuff et
al., using a luciferase expression system, compared the potency of
siRNA versus shRNA mediated knockdown in vivo; they found that
siRNA and shRNA are equivalent in potency at 10 mg dose; however,
on a molar basis, the shRNA was 250 fold more effective than the
siRNA [115]. In an effort to assess the potential of RNAi as a therapeutic
for hepatitis C (HCV), siRNANS5B was targeted against the nonstructural protein 5B viral polymerase coding region fused with
luciferase gene [125]. Luciferase expression in vivo was reduced by
75%. Using a cognate shRNANS5B produced a 92.8% average inhibition
over 3 experiments. McCleary et al. incubated shRNA, directed against
firefly luciferase and containing 29 mer stems and 2-nt 3′ overhangs,
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
with recombinant human Dicer [116]. The resulting 22-nt shRNA
products, with predictable guide strands, were compared to identically targeted 21 mer siRNA in HeLa cells. More effective inhibition
was seen with the shRNA. In another study of the feasibility of RNAi for
treatment of HCV, 19 and 25 bp shRNAs were compared with 19 and
25 bp siRNA directed against the HCV IRES (internal ribosomal entry
site) using a luciferase reporter in the AVA5 cell line with stable
expression of the HCV subgenomic genotype 1b replicon [117]. The
19 bp shRNAs were more potent than either the 19- or 25-bp siRNAs.
A recent paper evaluated the levels of Dicer and Drosha in cell lines
and in tumor samples from patients with ovarian cancer [126]. The
distribution of Dicer mRNA levels was bimodal and 60% of specimens
had decreased Dicer mRNA. Furthermore, low Dicer levels were found
to be a predictor of reduced disease-specific survival in multivariate
analysis. Of particular interest was the finding that, compared to siRNA
mediated silencing of the galectin-3 gene, poor silencing was achieved
with shRNA in ovarian cancer cell lines with low versus high Dicer
expression. These data will need to be confirmed and evaluated
further. Although there is a global downregulation of miRNA
expression in cancer [127], whether this is, in large part, due to low
Dicer expression or to shortened 3′ UTRs with fewer miRNA-binding
sites [128] in highly proliferating tumors with modulated feedback
mechanisms is not known. miRNA functionality has been confirmed in
the three tumor types (ovary, breast, and lung) evaluated in this study
raising questions regarding both qualitative and quantitative issues.
As the authors note, there are data correlating high Dicer expression
with poor prognostic features in other tumor types [129,130]. In their
retrospective evaluation of lung cancer, the role of let-7 as both a
regulator of Dicer and ras and their feedback networks could not be
assessed [131,132]. The Dicer levels in the tumors used for functional
assay of gene silencing are not given nor is there confirmation of the
equivalency of strand biasing between the siRNA and shRNA
constructs.
RNAi therapeutics have been shown to be well tolerated in
numerous animal models allowing for transition into the clinic. At
least 10 RNAi-based drugs are currently in early phase clinical trials
[133], two of which are cancer related; one targeted against the M2
subunit of ribonucleotide reductase (RRM2) [134] and the other
targeted against tenascin-C [135]. Animal studies with siRNA
inhibitors for RRM2 show efficacy [134] and safety in non-human
primates [136]. shRNA for the treatment of hepatitis B was also
approved for clinical trial by FDA.
Both siRNA and shRNA have their respective advantages and
disadvantages from the mechanistic point of view. However, safety of
this new therapeutic paradigm is of the utmost importance. Although
no significant adverse event involving initial RNAi-based clinical trials
has been reported, there are concerns over the potential off-target
effect of RNAi-based agents which will be discussed separately below.
4. Delivery
Efficacy of an RNAi cancer therapeutic is limited by the quantity of
the oligomer that effectively enters the tumor cells. In the clinical
setting this is primarily dependent on the method of delivery. An ideal
delivery vehicle must be able to selectively and differentially target
tumors versus normal tissue, homogeneously distribute through the
tumor mass and penetrate the tumor cells following systemic
administration. If cell entry is mediated by endocytosis the delivery
vehicle must negotiate endosomal/lysosomal escape and, in the case
of shRNA, the payload must penetrate the nuclear membrane as well.
Viral vectors are popular for laboratory delivery of shRNA because of
their high transfection efficiency and effective integration of exogenous DNA, but they have been losing support in recent years because of
concerns over safety and immunogenicity [137,138]. Non-viral polymeric delivery systems, in particular those with biodegradable
components, have much better safety profiles than their viral
753
counterparts though their transfection efficiency is generally lower
[139–141]. Non-viral vehicles for delivery of siRNA and shRNA are
typically cationic preparations. Their positive charge facilitates
complexation with negatively charged nucleic acids and also binding
to the negatively charged glycocalyx on external cell membranes
promoting endocytosis. Both tumor targeting and cell entry can be
enhanced by decoration or complexation of the vehicle with targeting
moieties, such as monoclonal antibodies, peptides, small molecule
ligands, and aptamers to recognize cell surface markers [124,142].
Once endocytosed, the vehicle's positive charge facilitates early escape
from the endosome [143,144]. Though the positive charge of these
vehicles improves their transfection efficiency, it is also associated
with increased toxicity [140,145].
A wide variety of potential vehicles are being developed to address
the different issues associated with the delivery of shRNA and siRNA.
There are three major classes of non-viral delivery vehicle systems:
synthetic polymers, natural/biodegradable polymers, and lipids;
many of the vehicles that are showing promise are actually hybrids
of these classes. For instance, there is a cyclodextrin-based cationic
polymer which has been used successfully to deliver siRNA targeted to
RRM2 in various in vivo cancer models [134,146]. This preparation is
currently in Phase I clinical trials. Lipid based nanoparticles are
showing potential for the delivery of shRNA and siRNA [147]. Protiva
Biotherapeutics and Alnaylam have developed nanoparticles composed of a lipid–PEG conjugate that is capable of encapsulating and
protecting nucleic acids for the purpose of systemic delivery. These
stable nucleic acid lipid particles (SNALPs) were used in the first
successful administration of siRNAs to a non-human primate
[148,149]. Silence Therapeutics has developed a lipid-based delivery
vehicle specifically designed for siRNA delivery to endothelial cells.
This vehicle, called AtuPLEX, contains a mix of cationic and fusogenic
lipids [150,151]. This vehicle has been used effectively to knockdown
protein kinase N3 in murine prostate and pancreatic cancer models,
inhibiting cancer progression [152,153]. More detailed discussions of
delivery vehicles for shRNA [124,154] and siRNA [141,155–157] as well
as general discussions of organ and tissue specific RNAi delivery
[138,158,159] may be found elsewhere.
The magnitude of cytokine induction associated with in vivo
delivery of siRNA has been noted to vary widely based on the delivery
vehicle used [160]. The well recognized conundrum in cationic nonviral nucleic delivery is that transfection efficiency usually correlates
with toxicity [161,162]. Effective strategies being pursued to break this
correlation include molecular modifications to shield positive charge
and the use of biodegradable polymers [154,163].
5. Off-target effects
Despite initial results showing excellent specificity in RNAi
mediated gene silencing, over the past 5 years many studies have
shown that there are multiple specific and nonspecific mechanisms
through which siRNA and shRNA can cause effects other than the
intended mRNA suppression. Unintended effects on gene expression
mediated by RNAi are termed “off-target effects.” Specific off-target
effects are mediated by partial sequence complementarity of the RNAi
construct to mRNAs other than the intended target. Nonspecific offtarget effects include a wide variety of immune and toxicity related
effects that are intrinsic to the RNAi construct itself or its delivery
vehicle. The following provides a brief review to the mechanisms
surrounding off-target effects and addresses strategies that are being
developed to minimize those effects.
5.1. Specific off-target effects
Expression profiling experiments have shown that partial sequence
complementarity in the passenger or guide strands of the RNAi
construct can produce off-target gene suppression. The first group to
754
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
definitively demonstrate this used an unbiased genome-wide microarray profile to search for downregulated mRNA immediately after
transfecting a variety of different siRNA constructs directed to different
genes into HeLa cells. This study demonstrated that few off-target genes
were regulated in common. Furthermore, the off-target expression
patterns observed for each individual siRNA were consistent, with
repeated runs revealing the same off-target expression profile. It was
also noted that the number and identity of the off-target transcripts was
unrelated to the ability of the siRNA to silence the target gene [164]. It
has subsequently been shown that both siRNA and shRNA constructs
with complementarity in the “seed region” can produce the same offtarget expression profiles, even across cell lines, independent of delivery
method [165].
Even limited sequence siRNA:mRNA complementarity with the
intended target, which is usually less than optimal, can produce offtarget suppression. Off-target silencing effects have been demonstrated in transcripts with complementarity as low as 7 nucleotides
with the guide siRNA strand [166]. Due to a variety of known and
unknown mechanisms, not all transcripts with this level of similarity
are silenced. The location of the region of complementarity within the
RNAi construct and the mRNA transcript are important predictors of
potential for suppression. Complementarity within nucleotides 2–7 at
the 5′ end of either the siRNA passenger or guide strands has been
shown to be a key determinant in directing off-target effects [167].
This region of the construct is reminiscent of the “seed” region within
miRNA, i.e., a heptameric sequence beginning at the first or second
position from the 5′ end of the miRNA that is complementary to sites
in the mRNA 3′-UTRs, which guide silencing in endogenous RNAi.
Indeed, RNAi off-targeting seems to be mechanistically related to an
miRNA-like effect in that complementarity between the miRNA seed
hexamer and the 3′UTR of the off-targeted mRNA produce effective
suppression of gene expression [103,168,169].
Various in vitro and in silico methods are either available or under
development for analysis of the off-target effects of a given RNAi
construct. Screening can be accomplished effectively using mRNA
expression data from transfected cells. Expression profiling for offtarget screening must be temporally controlled to ensure observation
of primary changes in mRNA levels and not secondary changes as a
result of downregulation of target protein expression. Results can then
be correlated to qRT-PCR data and to protein expression data through
Western blots. Methods for microarray-based off-target screening are
reviewed in detail elsewhere [169,170].
Computational methods are also being developed so that RNAi
constructs may be more effectively screened prior to in vitro testing.
Global complementarity search algorithms such as BLASTn and
Smith–Waterman have been shown to be poor measures of off-target
potential in an RNAi construct [168]. Methods for in silico screening of
RNAi constructs must be optimized to consider seed complementarity
in the constructs as well as 3′ UTR complementarity in the
transcriptome. Several groups are working to develop algorithms
and some are freely available online [171–174].
The specific off-target effects of a given construct can be mitigated by
several methods. siRNAs with an asymmetric unilateral 2-nt-overhang
on the antisense strand have greater potency than conventional siRNAs
as well as reduced off-target effects due to preferential strand selection
[113]. Sequence based modifications designed to reduce specific offtarget effects are likely to benefit both siRNA and shRNA approaches.
Single and double base mismatches between an RNAi construct and its
target transcript are often tolerated without reducing the potency of
suppression [168,175–177]. This allows for the optimization of the
construct sequence to minimize complementarity with 3′UTRs of
unintended targets.
Although the vector-driven shRNA approach to RNAi does not
permit specific chemical modification of the silencing construct, siRNA
oligomers can be chemically modified in order to reduce direct offtarget effects. These modifications must be carefully applied as
reductions in off-target effects are sometimes accompanied by
deceases in the overall potency of suppression. [178–180].
Chemical modifications can also be asymmetrically applied to the
passenger strand of the siRNA construct in order to specifically inhibit its
participation in silencing. Annealing of the guide strand to a passenger
strand composed of two segments has been shown to reduce off-target
effects mediated by passenger strand complementarity [181]. The
addition of other chemical moieties to one or both strands can also
limit specific off-target effects. The 5′-phosphorylation status of the
siRNA strands has been shown to be a determinant of which strand is
involved in silencing, and thus replacement of the 5′ phosphate by of an
siRNA strand with a methyl group has been shown to reduce its
participation in silencing [182].
Chemical modifications to nucleotides within the seed region of
the passenger or guide strand can reduce unintended entry of siRNA
constructs into the endogenous miRNA gene silencing pathway by
inhibiting the interaction of the RISC complex with mRNA. It has been
shown that 2′-O-methyl ribosyl substitution at position 2 in the siRNA
guide strand can reduce off-target silencing of transcripts with
complementarity to the seed region of the siRNA guide strand [165].
Asymmetric replacement of seed region nucleotides with DNA bases
has also been shown to reduce off-targeting as a result of seed region
complementarity within the passenger strand [183].
Recent in vitro studies have shown that shRNA produces fewer offtarget effects than siRNA. In one study shRNA and siRNA of the same
core sequence directed towards TP53 were applied to HCT-116 colon
carcinoma cells in concentrations necessary to achieve comparable
levels of target knockdown. Microarray profiling demonstrated a
much higher degree of up- and downregulation of off-target
transcripts in the siRNA transfected cells (M. Mehaffey, T. Ward, and
M. Cleary, in prep.). It has been suggested that these differences arise
from the fact that shRNA is transcribed in the nucleus and is therefore
subject to endogenous processing and regulatory mechanisms.
Additionally, siRNA is more susceptible to degradation in the
cytoplasm, which may also lead to off-target silencing [184].
5.2. Nonspecific off-target effects
Nonspecific off-target effects are those unintended perturbations
in gene expression not resulting from the direct interaction of an RNAi
construct with an mRNA transcript. Included in this category are
interferon (IFN) and other immune mediated responses to exogenous
RNAi, cellular toxicities due to the nucleotide construct, and effects
related to the delivery vehicle. There is strong reason to believe that
shRNA and siRNA would have very different profiles of nonspecific offtarget effects because of the mechanistic differences between the two
approaches.
Introduction of dsRNA longer than 29–30 bp into mammalian cells
results in a potent induction of the innate immune system via PKR,
similar to the mammalian cell defense mechanism against viral
infection [185]. Activation of the innate immune response by receptors
sensitive to exogenous nucleic acids leads to global degradation of
mRNA and thus broad inhibition of translation as well as global
upregulation of IFN-stimulated gene expression. Although shorter
siRNA constructs have been shown to avoid receptor activation and
were initially considered to be non-immunogenic [18], subsequent in
vitro data have shown that the introduction of both synthetic siRNA
oligomers and shRNA can induce a partial interferon response [186–
188] some of which are sequence dependent (e.g. GU-dependent, 5′UGUGU-3′ and GU-independent, 5′-GUCCUUCAA-3′) and, therefore,
avoidable. Activation of the innate immune system in the case of
exogenous RNAi is likely mediated through several cytoplasmic and
endosomal mechanisms attuned to recognize exogenous nucleic acids
from infectious agents. The relevant mediators of nucleic acidstimulated immunoactivation at the level of the endosome are Tolllike receptors (TLRs) 7 and 8 (typically activated by ssRNA), TLR 9 (via
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
unmethylated CpG activation) and TLR 3 (via dsRNA) [189–193].
Immune activation by nucleic acids at the cytoplasmic level is
mediated through RNA sensing receptors such as RIG-I and MDA-5
[114,194]. Mechanisms surrounding immune activation by RNAi are
reviewed more thoroughly elsewhere [160,195].
Recently, naked siRNA has been shown to activate TLR-3 on the
surface of vascular endothelial cells and trigger the release of IFN-γ
and IL-12 that mediate nonspecific anti-angiogenic effects in-vivo
[196]. Immune stimulation often is largely responsible for the
observed therapeutic effects of siRNA rather than the direct targeted
effect [197]. Misinterpreting the therapeutic effect of siRNA needs to
be carefully monitored.
Immune activation at the endosomal level is more readily avoidable
by shRNA constructs because the construct is presented on a DNA
plasmid obviating dsRNA activation of TLR 3. However, TLR 9, as noted, is
activated by unmethylated CpG motifs, which are typically found in
bacterial DNA [198,199]. Careful shRNA-encoding plasmid design,
avoiding unmethylated CpG motifs, can effectively attenuate if not
eliminate TLR 9 mediated endosomal immunoactivation [200].
There is also reason to believe that shRNA is less likely to induce an
inflammatory response through cytoplasmic dsRNA receptors in vivo
because shRNA is spliced by endogenous mechanisms. It has been
suggested that the 5′ ends of the endogenous-dicer spliced RNA
oligomers are less immunogenic than the 5′ ends of exogenous siRNA
oligomers [114,194]. This has been supported in vitro in an experiment
that compared liposome delivered siRNAs versus Pol III promoter—
expressed shRNAs of the same sequence in primary CD34+ progenitor—
derived hematopoietic cells. In this study it was found that siRNA
induced IFN-alpha and type I IFN genes, while the shRNA of the same
sequence did not induce an immune response [201].
Sequence modifications can be made to shRNA or siRNA in order to
reduce immunogenicity. It has been shown that endosomal immunoactivation by siRNA through TLR7 and TLR8 can be sequence
dependant [189,190,202]. Some simple sequence modifications, such
as the introduction of G:U mismatches into the sequence also seem to
lower the IFN response in vitro [203,204]. One recent study showed
that a marked reduction in the expression of the interferon-stimulated
gene oligoadenylate synthetase 1 (Oas1) could be achieved by
modifying the shRNA to contain features of the naturally occurring
microRNA-30 (miR-30) precursor [205].
As in the case of specific off-target effects, chemical modification to
siRNA oligomers can make them less immunostimulatory, however
these modifications must be fine-tuned so as not to negatively affect
the potency of intended target silencing. Suppression of the TLR
mediated immune response has been achieved by substituting the 2′hydroxyl uridines in the construct with 2′-fluoro, 2′-deoxy, or 2′-Omethyl uridines [206–208], the products of which do retain targetsilencing potency while reducing immunogenicity.
Usage of endogenous processing systems gives shRNA an advantage over siRNA in terms of its propensity for induction of IFN but its
over-saturation of these systems has been shown to have other
consequences that are more easily avoided by siRNA. In a key in vivo
study of the safety effects of long term expression of shRNA in the
livers of adult mice, a type 8 adeno-associated virus with a Pol III
promoter was used to drive expression of 49 different shRNAs of
different lengths and sequences directed against six targets. 36 of the
constructs tested resulted in a dose dependant liver injury that was
determined to be associated with the downregulation of critical
endogenous miRNAs. The degree of miRNA downregulation was
related neither to the shRNA sequence nor to the degree of
downregulation of the target mRNA [209]. Subsequent transfection
studies suggested the degree of miRNA downregulation to result from
a competitive bottleneck in shared miRNA/shRNA processing, most
likely at the level of exportin-5 (the nuclear membrane export protein
used to transfer pre-miRNAs into the cytoplasm) and the RISC
component, Argonaute-2 [210]. A similar in vivo study of hepatic
755
toxicity of siRNAs in mice and hamsters showed that systemic
introduction of synthetic siRNAs does not result in the suppression
of endogenous miRNA levels. In this study siRNAs targeting two
hepatocyte-specific genes (apolipoprotein B and factor VII) and a
scramble control were administered to mice and one siRNA targeting
the hepatocyte-expressed gene Scap was administered to hamsters.
Robust suppression of target genes was achieved in all cases. No
changes in levels of the hepatocyte endogenously expressed miR-122
were observed in the mice or the hamsters and no changes in the
broadly expressed miRNAs, miR-16 and let-7a, were observed in the
mice [211].
Though siRNA likely does not compete with endogenous miRNA
for processing proteins, care must be taken when using shRNA as an
effector of RNAi in order to minimize the potential for damage
mediated by over-saturation of exportin-5. In one in vitro study, overexpression of the exportin-5 protein has been shown to eliminate the
nuclear export bottleneck and allow cells to tolerate higher dosages of
shRNA without toxicity [212]. Another study showed that an adenoassociated virus construct using a Pol-II promoter was able to achieve
stable target gene suppression at high shRNA doses for over 1 year
after the initial dosing [213]. In order to minimize the risk of toxicity
from over-saturation of miRNA professing systems, data to date
suggest both selective promoter integration (e.g. pol II versus pol III)
and limiting the dosage of shRNA so as to stay below the threshold of
competitive inhibition of the endogenous miRNA biogenesis
machinery.
6. The future outlook
The ability to precisely and differentially target functionally biorelevant molecular signals in patient's cancers will establish a new
paradigm in cancer management; one which focuses on defining the
uniqueness of each patient's tumor and tumor-host processes and
interactions following rational target prioritization using computational systems biology algorithms. This, then, would allow for
exploitation of the “attack vulnerability” of the rewired cancer
network by deconstructing essential hubs and linkages, multiply
targeting and eliminating them. Both siRNA and shRNA effectors are
attractive opportunities. The capability of potentiating activity using a
bi-functional design may further enhance safety and efficacy. The
simplicity of siRNA manufacturing and the transient nature of the
effect per dose may be optimal for certain medical disorders in which
high vector doses are required, e.g. some of the viral infections,
however, by using the endogenous processing machinery, optimized
shRNA constructs allow for high potency sustainable effects using
low-copy numbers resulting in less off-target effects (particularly if
embedded in an miRNA scaffold) thereby ensuring greater safety.
Though shRNA seems ideal for cancer-related therapeutic development, new technology such as bi-functional RNA interference may
provide an even greater opportunity for enhancement in potency as
well as heightening safety thereby increasing the opportunities for
multiple target therapy. This, of course, is contingent on optimization
of delivery and minimization of off-target effect which will need to be
established through early clinical testing.
7. Conclusions
Our understanding and application of RNAi has dramatically
advanced over the last 5 years. Despite limitations in developing
effective delivery vehicles and concerns regarding potential off-target
activity, clinical development has been initiated. As the science of this
fledgling technology advances, it is evident that issues such as target
selection, effector potency, delivery vehicle design, and off-target
effects will continue to be addressed and resolved. Bi-functional RNAi
products are evolving components in this transition to clinically
effective and safer therapeutics.
756
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
Acknowledgements
The authors would like to acknowledge Brenda Marr and Susan
Mill for their competent and knowledgeable assistance in the
preparation of this manuscript, and M. Cleary for providing the data
before publication.
References
[1] V.G. Kaklamani, W.J. Gradishar, Gene expression in breast cancer, Curr. Treatm.
Opt. Oncol. 7 (2006) 123–128.
[2] L. Marchionni, R.F. Wilson, S.S. Marinopoulos, A.C. Wolff, G. Parmigiani, E.B. Bass,
S.N. Goodman, Impact of gene expression profiling tests on breast cancer
outcomes, Evid. Rep./Technol. Assess. (Full Rep.) (2007) 1–105.
[3] J.K. Habermann, J. Doering, S. Hautaniemi, U.J. Roblick, N.K. Bundgen, D. Nicorici, U.
Kronenwett, S. Rathnagiriswaran, R.K. Mettu, Y. Ma, S. Kruger, H.P. Bruch, G. Auer, N.L.
Guo, T. Ried, The gene expression signature of genomic instability in breast cancer is
an independent predictor of clinical outcome, Int. J. Cancer. 124 (7) (2009 Apr 1)
1552–1564.
[4] P. Wirapati, C. Sotiriou, S. Kunkel, P. Farmer, S. Pradervand, B. Haibe-Kains, C.
Desmedt, M. Ignatiadis, T. Sengstag, F. Schutz, D.R. Goldstein, M. Piccart, M.
Delorenzi, Meta-analysis of gene expression profiles in breast cancer: toward a
unified understanding of breast cancer subtyping and prognosis signatures,
Breast Cancer Res. 10 (2008) R65.
[5] T. Sorlie, R. Tibshirani, J. Parker, T. Hastie, J.S. Marron, A. Nobel, S. Deng, H. Johnsen,
R. Pesich, S. Geisler, J. Demeter, C.M. Perou, P.E. Lonning, P.O. Brown, A.L. BorresenDale, D. Botstein, Repeated observation of breast tumor subtypes in independent
gene expression data sets, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 8418–8423.
[6] M. Davies, B. Hennessy, G.B. Mills, Point mutations of protein kinases and
individualised cancer therapy, Expert Opin. Pharmacother. 7 (2006) 2243–2261.
[7] A.G. Letai, Diagnosing and exploiting cancer's addiction to blocks in apoptosis,
Nat. Rev., Cancer 8 (2008) 121–132.
[8] O.I. Olopade, T.A. Grushko, R. Nanda, D. Huo, Advances in breast cancer: pathways
to personalized medicine, Clin. Cancer Res. 14 (2008) 7988–7999.
[9] J. Nemunaitis, N. Senzer, I. Khalil, Y. Shen, P. Kumar, A. Tong, J. Kuhn, J. Lamont, M.
Nemunaitis, D. Rao, Y.A. Zhang, Y. Zhou, J. Vorhies, P. Maples, C. Hill, D. Shanahan,
Proof concept for clinical justification of network mapping for personalized
cancer therapeutics, Cancer Gene Ther. 14 (2007) 686–695.
[10] C.A. Stein, J.S. Cohen, Oligodeoxynucleotides as inhibitors of gene expression: a
review, Cancer Res. 48 (1988) 2659–2668.
[11] E. Zamaratski, P.I. Pradeepkumar, J. Chattopadhyaya, A critical survey of the
structure–function of the antisense oligo/RNA heteroduplex as substrate for
RNase H, J. Biochem. Biophys. Methods 48 (2001) 189–208.
[12] S.T. Crooke, Molecular mechanisms of action of antisense drugs, Biochim.
Biophys. Acta 1489 (1999) 31–44.
[13] D. Castanotto, M. Scherr, J.J. Rossi, Intracellular expression and function of
antisense catalytic RNAs, Methods Enzymol. 313 (2000) 401–420.
[14] J.J. Rossi, Ribozymes in the nucleolus, Science 285 (1999) 1685.
[15] S.W. Santoro, G.F. Joyce, A general purpose RNA-cleaving DNA enzyme, Proc. Natl.
Acad. Sci. U. S. A. 94 (1997) 4262–4266.
[16] Y. Wu, L. Yu, R. McMahon, J.J. Rossi, S.J. Forman, D.S. Snyder, Inhibition of bcr-abl
oncogene expression by novel deoxyribozymes (DNAzymes), Hum. Gene Ther. 10
(1999) 2847–2857.
[17] T.E. Ichim, M. Li, H. Qian, I.A. Popov, K. Rycerz, X. Zheng, D. White, R. Zhong, W.P. Min,
RNA interference: a potent tool for gene-specific therapeutics, Am. J. Transplant.
4 (2004) 1227–1236.
[18] S.M. Elbashir, J. Harborth, W. Lendeckel, A. Yalcin, K. Weber, T. Tuschl, Duplexes of
21-nucleotide RNAs mediate RNA interference in cultured mammalian cells,
Nature 411 (2001) 494–498.
[19] J. Liu, M.A. Carmell, F.V. Rivas, C.G. Marsden, J.M. Thomson, J.J. Song, S.M.
Hammond, L. Joshua-Tor, G.J. Hannon, Argonaute2 is the catalytic engine of
mammalian RNAi, Science 305 (2004) 1437–1441.
[20] Z. Paroo, D.R. Corey, Challenges for RNAi in vivo, Trends Biotechnol. 22 (2004)
390–394.
[21] Y. Aoki, D.P. Cioca, H. Oidaira, J. Kamiya, K. Kiyosawa, RNA interference may be
more potent than antisense RNA in human cancer cell lines, Clin. Exp. Pharmacol.
Physiol. 30 (2003) 96–102.
[22] B.L. Bass, RNA interference. The short answer, Nature 411 (2001) 428–429.
[23] S. Coma, V. Noe, C. Lavarino, J. Adan, M. Rivas, M. Lopez-Matas, R. Pagan, F.
Mitjans, S. Vilaro, J. Piulats, C.J. Ciudad, Use of siRNAs and antisense oligonucleotides against survivin RNA to inhibit steps leading to tumor angiogenesis,
Oligonucleotides 14 (2004) 100–113.
[24] M. Miyagishi, M. Hayashi, K. Taira, Comparison of the suppressive effects of
antisense oligonucleotides and siRNAs directed against the same targets in
mammalian cells, Antisense Nucleic Acid Drug Dev. 13 (2003) 1–7.
[25] M. Scherr, K. Battmer, T. Winkler, O. Heidenreich, A. Ganser, M. Eder, Specific inhibition
of bcr-abl gene expression by small interfering RNA, Blood 101 (2003) 1566–1569.
[26] T.R. Brummelkamp, R. Bernards, R. Agami, Stable suppression of tumorigenicity
by virus-mediated RNA interference, Cancer Cells 2 (2002) 243–247.
[27] L.A. Martinez, I. Naguibneva, H. Lehrmann, A. Vervisch, T. Tchenio, G. Lozano, A.
Harel-Bellan, Synthetic small inhibiting RNAs: efficient tools to inactivate oncogenic
mutations and restore p53 pathways, Proc. Natl. Acad. Sci. U. S. A. 99 (2002)
14,849–14,854.
[28] M. Yoshinouchi, T. Yamada, M. Kizaki, J. Fen, T. Koseki, Y. Ikeda, T. Nishihara, K.
Yamato, In vitro and in vivo growth suppression of human papillomavirus 16positive cervical cancer cells by E6 siRNA, Mol. Ther. 8 (2003) 762–768.
[29] A. Choudhury, J. Charo, S.K. Parapuram, R.C. Hunt, D.M. Hunt, B. Seliger, R.
Kiessling, Small interfering RNA (siRNA) inhibits the expression of the Her2/neu
gene, upregulates HLA class I and induces apoptosis of Her2/neu positive tumor
cell lines, Int. J. Cancer 108 (2004) 71–77.
[30] G. Yang, K.Q. Cai, J.A. Thompson-Lanza, R.C. Bast Jr., J. Liu, Inhibition of breast and
ovarian tumor growth through multiple signaling pathways by using retrovirusmediated small interfering RNA against Her-2/neu gene expression, J. Biol. Chem.
279 (2004) 4339–4345.
[31] B. Farrow, P. Rychahou, C. Murillo, K.L. O'Connor, T. Iwamura, B.M. Evers,
Inhibition of pancreatic cancer cell growth and induction of apoptosis with novel
therapies directed against protein kinase A, Surgery 134 (2003) 197–205.
[32] E. Yague, C.F. Higgins, S. Raguz, Complete reversal of multidrug resistance by stable
expression of small interfering RNAs targeting MDR1, Gene Ther.11 (2004) 1170–1174.
[33] B.A. Kosciolek, K. Kalantidis, M. Tabler, P.T. Rowley, Inhibition of telomerase activity
in human cancer cells by RNA interference, Mol. Cancer Ther. 2 (2003) 209–216.
[34] D.P. Cioca, Y. Aoki, K. Kiyosawa, RNA interference is a functional pathway with
therapeutic potential in human myeloid leukemia cell lines, Cancer Gene Ther.
10 (2003) 125–133.
[35] H. Kawasaki, K. Taira, Short hairpin type of dsRNAs that are controlled by tRNA(Val)
promoter significantly induce RNAi-mediated gene silencing in the cytoplasm of
human cells, Nucleic Acids Res. 31 (2003) 700–707.
[36] K. Li, S.Y. Lin, F.C. Brunicardi, P. Seu, Use of RNA interference to target cyclin
E-overexpressing hepatocellular carcinoma, Cancer Res. 63 (2003) 3593–3597.
[37] U.N. Verma, R.M. Surabhi, A. Schmaltieg, C. Becerra, R.B. Gaynor, Small interfering
RNAs directed against beta-catenin inhibit the in vitro and in vivo growth of colon
cancer cells, Clin. Cancer Res. 9 (2003) 1291–1300.
[38] S. Aharinejad, P. Paulus, M. Sioud, M. Hofmann, K. Zins, R. Schafer, E.R. Stanley, D.
Abraham, Colony-stimulating factor-1 blockade by antisense oligonucleotides and
small interfering RNAs suppresses growth of human mammary tumor xenografts in
mice, Cancer Res. 64 (2004) 5378–5384.
[39] H. Uchida, T. Tanaka, K. Sasaki, K. Kato, H. Dehari, Y. Ito, M. Kobune, M. Miyagishi,
K. Taira, H. Tahara, H. Hamada, Adenovirus-mediated transfer of siRNA against
survivin induced apoptosis and attenuated tumor cell growth in vitro and in vivo,
Mol. Ther. 10 (2004) 162–171.
[40] A.J. Salisbury, V.M. Macaulay, Development of molecular agents for IGF receptor
targeting, Horm. Metab. Res. 35 (2003) 843–849.
[41] M.S. Duxbury, H. Ito, M.J. Zinner, S.W. Ashley, E.E. Whang, Focal adhesion kinase
gene silencing promotes anoikis and suppresses metastasis of human pancreatic
adenocarcinoma cells, Surgery 135 (2004) 555–562.
[42] M.S. Duxbury, E. Matros, H. Ito, M.J. Zinner, S.W. Ashley, E.E. Whang, Systemic
siRNA-mediated gene silencing: a new approach to targeted therapy of cancer,
Ann. Surg. 240 (2004) 667–674 [discussion 675–666].
[43] S. Filleur, A. Courtin, S. Ait-Si-Ali, J. Guglielmi, C. Merle, A. Harel-Bellan, P.
Clezardin, F. Cabon, SiRNA-mediated inhibition of vascular endothelial growth
factor severely limits tumor resistance to antiangiogenic thrombospondin-1 and
slows tumor vascularization and growth, Cancer Res. 63 (2003) 3919–3922.
[44] Y. Takei, K. Kadomatsu, Y. Yuzawa, S. Matsuo, T. Muramatsu, A small interfering
RNA targeting vascular endothelial growth factor as cancer therapeutics, Cancer
Res. 64 (2004) 3365–3370.
[45] S.S. Lakka, C.S. Gondi, N. Yanamandra, W.C. Olivero, D.H. Dinh, M. Gujrati, J.S. Rao,
Inhibition of cathepsin B and MMP-9 gene expression in glioblastoma cell line via
RNA interference reduces tumor cell invasion, tumor growth and angiogenesis,
Oncogene 23 (2004) 4681–4689.
[46] A. Singh, S. Boldin-Adamsky, R.K. Thimmulappa, S.K. Rath, H. Ashush, J. Coulter, A.
Blackford, S.N. Goodman, F. Bunz, W.H. Watson, E. Gabrielson, E. Feinstein, S.
Biswal, RNAi-mediated silencing of nuclear factor erythroid-2-related factor 2
gene expression in non-small cell lung cancer inhibits tumor growth and
increases efficacy of chemotherapy, Cancer Res. 68 (2008) 7975–7984.
[47] S. Nakahira, S. Nakamori, M. Tsujie, Y. Takahashi, J. Okami, S. Yoshioka, M.
Yamasaki, S. Marubashi, I. Takemasa, A. Miyamoto, Y. Takeda, H. Nagano, K. Dono,
K. Umeshita, M. Sakon, M. Monden, Involvement of ribonucleotide reductase M1
subunit overexpression in gemcitabine resistance of human pancreatic cancer,
Int. J. Cancer 120 (2007) 1355–1363.
[48] A. Fire, S. Xu, M.K. Montgomery, S.A. Kostas, S.E. Driver, C.C. Mello, Potent and
specific genetic interference by double-stranded RNA in Caenorhabditis elegans,
Nature 391 (1998) 806–811.
[49] J.Y. Yu, S.L. DeRuiter, D.L. Turner, RNA interference by expression of short-interfering
RNAs and hairpin RNAs in mammalian cells, Proc. Natl. Acad. Sci. U. S. A. 99 (2002)
6047–6052.
[50] M. Miyagishi, K. Taira, U6 promoter-driven siRNAs with four uridine 3′ overhangs
efficiently suppress targeted gene expression in mammalian cells, Nat.
Biotechnol. 20 (2002) 497–500.
[51] A. Jarve, J. Muller, I.H. Kim, K. Rohr, C. MacLean, G. Fricker, U. Massing, F. Eberle, A.
Dalpke, R. Fischer, M.F. Trendelenburg, M. Helm, Surveillance of siRNA integrity
by FRET imaging, Nucleic Acids Res. 35 (2007) e124.
[52] J.P. Leonetti, N. Mechti, G. Degols, C. Gagnor, B. Lebleu, Intracellular distribution of
microinjected antisense oligonucleotides, Proc. Natl. Acad. Sci. U. S. A. 88 (1991)
2702–2706.
[53] G.B. Robb, K.M. Brown, J. Khurana, T.M. Rana, Specific and potent RNAi in the
nucleus of human cells, Nat. Struct. Mol. Biol. 12 (2005) 133–137.
[54] S.Y. Berezhna, L. Supekova, F. Supek, P.G. Schultz, A.A. Deniz, siRNA in human cells
selectively localizes to target RNA sites, Proc. Natl. Acad. Sci. U. S. A. 103 (2006)
7682–7687.
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
[55] B.A. Janowski, K.E. Huffman, J.C. Schwartz, R. Ram, R. Nordsell, D.S. Shames, J.D.
Minna, D.R. Corey, Involvement of AGO1 and AGO2 in mammalian transcriptional
silencing, Nat. Struct. Mol. Biol. 13 (2006) 787–792.
[56] D.H. Kim, L.M. Villeneuve, K.V. Morris, J.J. Rossi, Argonaute-1 directs siRNA-mediated
transcriptional gene silencing in human cells, Nat. Struct. Mol. Biol. 13 (2006)
793–797.
[57] S. Rudel, A. Flatley, L. Weinmann, E. Kremmer, G. Meister, A multifunctional
human Argonaute2-specific monoclonal antibody, RNA 14 (2008) 1244–1253.
[58] T. Ohrt, J. Mutze, W. Staroske, L. Weinmann, J. Hock, K. Crell, G. Meister, P. Schwille,
Fluorescence correlation spectroscopy and fluorescence cross-correlation spectroscopy reveal the cytoplasmic origination of loaded nuclear RISC in vivo in human
cells, Nucleic Acids Res. 36 (2008) 6439–6449.
[59] L. Weinmann, J. Hock, T. Ivacevic, T. Ohrt, J. Mutze, P. Schwille, E. Kremmer, V.
Benes, H. Urlaub, G. Meister, Importin 8 is a gene silencing factor that targets
argonaute proteins to distinct mRNAs, Cell 136 (2009) 496–507.
[60] S. Guang, A.F. Bochner, D.M. Pavelec, K.B. Burkhart, S. Harding, J. Lachowiec, S.
Kennedy, An Argonaute transports siRNAs from the cytoplasm to the nucleus,
Science 321 (2008) 537–541.
[61] A. Grunweller, C. Gillen, V.A. Erdmann, J. Kurreck, Cellular uptake and localization
of a Cy3-labeled siRNA specific for the serine/threonine kinase Pim-1,
Oligonucleotides 13 (2003) 345–352.
[62] T. Ohrt, D. Merkle, K. Birkenfeld, C.J. Echeverri, P. Schwille, In situ fluorescence
analysis demonstrates active siRNA exclusion from the nucleus by exportin 5,
Nucleic Acids Res. 34 (2006) 1369–1380.
[63] Y.L. Chiu, A. Ali, C.Y. Chu, H. Cao, T.M. Rana, Visualizing a correlation between
siRNA localization, cellular uptake, and RNAi in living cells, Chem. Biol. 11 (2004)
1165–1175.
[64] K. Forstemann, Y. Tomari, T. Du, V.V. Vagin, A.M. Denli, D.P. Bratu, C. Klattenhoff,
W.E. Theurkauf, P.D. Zamore, Normal microRNA maturation and germ-line stem
cell maintenance requires Loquacious, a double-stranded RNA-binding domain
protein, PLoS. Biol. 3 (2005) e236.
[65] F. Jiang, X. Ye, X. Liu, L. Fincher, D. McKearin, Q. Liu, Dicer-1 and R3D1-L catalyze
microRNA maturation in Drosophila, Genes Dev. 19 (2005) 1674–1679.
[66] K. Saito, A. Ishizuka, H. Siomi, M.C. Siomi, Processing of pre-microRNAs by the
Dicer-1-Loquacious complex in Drosophila cells, PLoS. Biol. 3 (2005) e235.
[67] X. Liu, J.K. Park, F. Jiang, Y. Liu, D. McKearin, Q. Liu, Dicer-1, but not Loquacious, is
critical for assembly of miRNA-induced silencing complexes, RNA 13 (2007)
2324–2329.
[68] E.P. Murchison, J.F. Partridge, O.H. Tam, S. Cheloufi, G.J. Hannon, Characterization of
Dicer-deficient murine embryonic stem cells, Proc. Natl. Acad. Sci. U. S. A. 102 (2005)
12,135–12,140.
[69] E. Bernstein, A.A. Caudy, S.M. Hammond, G.J. Hannon, Role for a bidentate
ribonuclease in the initiation step of RNA interference, Nature 409 (2001) 363–366.
[70] P. Provost, D. Dishart, J. Doucet, D. Frendewey, B. Samuelsson, O. Radmark,
Ribonuclease activity and RNA binding of recombinant human Dicer, EMBO J.
21 (2002) 5864–5874.
[71] S.M. Hammond, A.A. Caudy, G.J. Hannon, Post-transcriptional gene silencing by
double-stranded RNA, Nat. Rev., Genet. 2 (2001) 110–119.
[72] M.A. Carmell, G.J. Hannon, RNase III enzymes and the initiation of gene silencing,
Nat. Struct. Mol. Biol. 11 (2004) 214–218.
[73] I.J. Macrae, K. Zhou, F. Li, A. Repic, A.N. Brooks, W.Z. Cande, P.D. Adams, J.A.
Doudna, Structural basis for double-stranded RNA processing by Dicer, Science
311 (2006) 195–198.
[74] K.H. Kok, M.H. Ng, Y.P. Ching, D.Y. Jin, Human TRBP and PACT directly interact with
each other and associate with dicer to facilitate the production of small
interfering RNA, J. Biol. Chem. 282 (2007) 17,649–17,657.
[75] G.B. Robb, T.M. Rana, RNA helicase A interacts with RISC in human cells and
functions in RISC loading, Mol. Cell 26 (2007) 523–537.
[76] C. Matranga, Y. Tomari, C. Shin, D.P. Bartel, P.D. Zamore, Passenger–strand
cleavage facilitates assembly of siRNA into Ago2-containing RNAi enzyme
complexes, Cell 123 (2005) 607–620.
[77] T.A. Rand, S. Petersen, F. Du, X. Wang, Argonaute2 cleaves the anti-guide strand of
siRNA during RISC activation, Cell 123 (2005) 621–629.
[78] P.J. Leuschner, S.L. Ameres, S. Kueng, J. Martinez, Cleavage of the siRNA passenger
strand during RISC assembly in human cells, EMBO Rep. 7 (2006) 314–320.
[79] J.B. Preall, E.J. Sontheimer, RNAi: RISC gets loaded, Cell 123 (2005) 543–545.
[80] B. Haley, P.D. Zamore, Kinetic analysis of the RNAi enzyme complex, Nat. Struct.
Mol. Biol. 11 (2004) 599–606.
[81] R.I. Gregory, T.P. Chendrimada, N. Cooch, R. Shiekhattar, Human RISC couples
microRNA biogenesis and posttranscriptional gene silencing, Cell 123 (2005)
631–640.
[82] E. Maniataki, Z. Mourelatos, A human, ATP-independent, RISC assembly machine
fueled by pre-miRNA, Genes Dev. 19 (2005) 2979–2990.
[83] R.E. Collins, X. Cheng, Structural and biochemical advances in mammalian RNAi,
J. Cell. Biochem. 99 (2006) 1251–1266.
[84] R.L. Boudreau, A.M. Monteys, B.L. Davidson, Minimizing variables among hairpinbased RNAi vectors reveals the potency of shRNAs, RNA 14 (2008) 1834–1844.
[85] B.R. Cullen, RNAi the natural way, Nat. Genet. 37 (2005) 1163–1165.
[86] Y. Zeng, R. Yi, B.R. Cullen, Recognition and cleavage of primary microRNA
precursors by the nuclear processing enzyme Drosha, EMBO J. 24 (2005)
138–148.
[87] J.M. Silva, M.Z. Li, K. Chang, W. Ge, M.C. Golding, R.J. Rickles, D. Siolas, G. Hu, P.J.
Paddison, M.R. Schlabach, N. Sheth, J. Bradshaw, J. Burchard, A. Kulkarni, G. Cavet,
R. Sachidanandam, W.R. McCombie, M.A. Cleary, S.J. Elledge, G.J. Hannon, Secondgeneration shRNA libraries covering the mouse and human genomes, Nat. Genet.
37 (2005) 1281–1288.
757
[88] Y. Lee, C. Ahn, J. Han, H. Choi, J. Kim, J. Yim, J. Lee, P. Provost, O. Radmark, S. Kim, V.N.
Kim, The nuclear RNase III Drosha initiates microRNA processing, Nature 425 (2003)
415–419.
[89] H. Zhang, F.A. Kolb, V. Brondani, E. Billy, W. Filipowicz, Human Dicer
preferentially cleaves dsRNAs at their termini without a requirement for ATP,
EMBO J. 21 (2002) 5875–5885.
[90] Y. Lee, K. Jeon, J.T. Lee, S. Kim, V.N. Kim, MicroRNA maturation: stepwise
processing and subcellular localization, EMBO J. 21 (2002) 4663–4670.
[91] B.R. Cullen, Transcription and processing of human microRNA precursors, Mol.
Cell 16 (2004) 861–865.
[92] R. Yi, Y. Qin, I.G. Macara, B.R. Cullen, Exportin-5 mediates the nuclear export of
pre-microRNAs and short hairpin RNAs, Genes Dev. 17 (2003) 3011–3016.
[93] E. Lund, S. Guttinger, A. Calado, J.E. Dahlberg, U. Kutay, Nuclear export of
microRNA precursors, Science 303 (2004) 95–98.
[94] Y.S. Lee, K. Nakahara, J.W. Pham, K. Kim, Z. He, E.J. Sontheimer, R.W. Carthew,
Distinct roles for Drosophila Dicer-1 and Dicer-2 in the siRNA/miRNA silencing
pathways, Cell 117 (2004) 69–81.
[95] M. Landthaler, D. Gaidatzis, A. Rothballer, P.Y. Chen, S.J. Soll, L. Dinic, T. Ojo, M. Hafner,
M. Zavolan, T. Tuschl, Molecular characterization of human Argonaute-containing
ribonucleoprotein complexes and their bound target mRNAs, RNA 14 (2008)
2580–2596.
[96] S.M. Hammond, S. Boettcher, A.A. Caudy, R. Kobayashi, G.J. Hannon, Argonaute2, a
link between genetic and biochemical analyses of RNAi, Science 293 (2001)
1146–1150.
[97] E.J. Sontheimer, R.W. Carthew, Molecular biology. Argonaute journeys into the
heart of RISC, Science 305 (2004) 1409–1410.
[98] G. Hutvagner, P.D. Zamore, A microRNA in a multiple-turnover RNAi enzyme
complex, Science 297 (2002) 2056–2060.
[99] S. Yekta, I.H. Shih, D.P. Bartel, MicroRNA-directed cleavage of HOXB8 mRNA,
Science 304 (2004) 594–596.
[100] D.T. Humphreys, B.J. Westman, D.I. Martin, T. Preiss, MicroRNAs control translation
initiation by inhibiting eukaryotic initiation factor 4E/cap and poly(A) tail function,
Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 16,961–16,966.
[101] R.S. Pillai, S.N. Bhattacharyya, C.G. Artus, T. Zoller, N. Cougot, E. Basyuk, E.
Bertrand, W. Filipowicz, Inhibition of translational initiation by Let-7 MicroRNA
in human cells, Science 309 (2005) 1573–1576.
[102] R. Thermann, M.W. Hentze, Drosophila miR2 induces pseudo-polysomes and
inhibits translation initiation, Nature 447 (2007) 875–878.
[103] M.A. Valencia-Sanchez, J. Liu, G.J. Hannon, R. Parker, Control of translation and
mRNA degradation by miRNAs and siRNAs, Genes Dev. 20 (2006) 515–524.
[104] R. Parker, U. Sheth, P bodies and the control of mRNA translation and degradation,
Mol. Cell 25 (2007) 635–646.
[105] K. Okamura, A. Ishizuka, H. Siomi, M.C. Siomi, Distinct roles for Argonaute
proteins in small RNA-directed RNA cleavage pathways, Genes Dev. 18 (2004)
1655–1666.
[106] K. Miyoshi, H. Tsukumo, T. Nagami, H. Siomi, M.C. Siomi, Slicer function of
Drosophila Argonautes and its involvement in RISC formation, Genes Dev. 19 (2005)
2837–2848.
[107] Y. Tomari, T. Du, P.D. Zamore, Sorting of Drosophila small silencing RNAs, Cell
130 (2007) 299–308.
[108] K. Forstemann, M.D. Horwich, L. Wee, Y. Tomari, P.D. Zamore, Drosophila
microRNAs are sorted into functionally distinct argonaute complexes after
production by dicer-1, Cell 130 (2007) 287–297.
[109] F.A. Steiner, S.W. Hoogstrate, K.L. Okihara, K.L. Thijssen, R.F. Ketting, R.H. Plasterk,
T. Sijen, Structural features of small RNA precursors determine Argonaute loading
in Caenorhabditis elegans, Nat. Struct. Mol. Biol. 14 (2007) 927–933.
[110] A. Azuma-Mukai, H. Oguri, T. Mituyama, Z.R. Qian, K. Asai, H. Siomi, M.C. Siomi,
Characterization of endogenous human Argonautes and their miRNA partners in
RNA silencing, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 7964–7969.
[111] D.G. Hendrickson, D.J. Hogan, D. Herschlag, J.E. Ferrell, P.O. Brown, Systematic
identification of mRNAs recruited to argonaute 2 by specific microRNAs and
corresponding changes in transcript abundance, PLoS ONE 3 (2008) e2126.
[112] Y. Tay, J. Zhang, A.M. Thomson, B. Lim, I. Rigoutsos, MicroRNAs to Nanog, Oct4 and Sox2
coding regions modulate embryonic stem cell differentiation, Nature 455 (2008)
1124–1128.
[113] M. Sano, M. Sierant, M. Miyagishi, M. Nakanishi, Y. Takagi, S. Sutou, Effect of
asymmetric terminal structures of short RNA duplexes on the RNA interference
activity and strand selection, Nucleic Acids Res. 36 (2008) 5812–5821.
[114] D.H. Kim, M.A. Behlke, S.D. Rose, M.S. Chang, S. Choi, J.J. Rossi, Synthetic dsRNA Dicer
substrates enhance RNAi potency and efficacy, Nat. Biotechnol. 23 (2005) 222–226.
[115] M.A. McAnuff, G.R. Rettig, K.G. Rice, Potency of siRNA versus shRNA mediated
knockdown in vivo, J. Pharm. Sci. 96 (2007) 2922–2930.
[116] D. Siolas, C. Lerner, J. Burchard, W. Ge, P.S. Linsley, P.J. Paddison, G.J. Hannon, M.A.
Cleary, Synthetic shRNAs as potent RNAi triggers, Nat. Biotechnol. 23 (2005) 227–231.
[117] A.V. Vlassov, B. Korba, K. Farrar, S. Mukerjee, A.A. Seyhan, H. Ilves, R.L. Kaspar, D.
Leake, S.A. Kazakov, B.H. Johnston, shRNAs targeting hepatitis C: effects of
sequence and structural features, and comparision with siRNA, Oligonucleotides
17 (2007) 223–236.
[118] M. Gossen, H. Bujard, Tight control of gene expression in mammalian cells by
tetracycline-responsive promoters, Proc. Natl. Acad. Sci. U. S. A. 89 (1992) 5547–5551.
[119] S. Gupta, R.A. Schoer, J.E. Egan, G.J. Hannon, V. Mittal, Inducible, reversible, and stable
RNA interference in mammalian cells, Proc. Natl. Acad. Sci. U. S. A. 101 (2004)
1927–1932.
[120] R.A. Dickins, M.T. Hemann, J.T. Zilfou, D.R. Simpson, I. Ibarra, G.J. Hannon, S.W.
Lowe, Probing tumor phenotypes using stable and regulated synthetic microRNA
precursors, Nat. Genet. 37 (2005) 1289–1295.
758
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
[121] A.W. Tong, Y.A. Zhang, J. Nemunaitis, Small interfering RNA for experimental
cancer therapy, Curr. Opin. Mol. Ther. 7 (2005) 114–124.
[122] A. Aigner, Cellular delivery in vivo of siRNA-based therapeutics, Curr. Pharm. Des.
14 (2008) 3603–3619.
[123] A.W. Tong, C.M. Jay, N. Senzer, P.B. Maples, J. Nemunaitis, Systemic therapeutic gene
delivery for cancer: crafting Paris' arrow, Curr. Gene Ther. 9 (1) (2009 Feb) 45–60.
[124] J.S. Vorhies, J. Nemunaitis, Nonviral delivery vehicles for use in short hairpin RNAbased cancer therapies, Expert Rev. Anticancer Ther. 7 (2007) 373–382.
[125] A.P. McCaffrey, L. Meuse, T.T. Pham, D.S. Conklin, G.J. Hannon, M.A. Kay, RNA
interference in adult mice, Nature 418 (2002) 38–39.
[126] W.M. Merritt, Y.G. Lin, L.Y. Han, A.A. Kamat, W.A. Spannuth, R. Schmandt, D. Urbauer,
L.A. Pennacchio, J.F. Cheng, A.M. Nick, M.T. Deavers, A. Mourad-Zeidan, H. Wang, P.
Mueller, M.E. Lenburg, J.W. Gray, S. Mok, M.J. Birrer, G. Lopez-Berestein, R.L. Coleman,
M. Bar-Eli, A.K. Sood, Dicer, Drosha, and outcomes in patients with ovarian cancer, N.
Engl. J. Med. 359 (2008) 2641–2650.
[127] J. Lu, G. Getz, E.A. Miska, E. Alvarez-Saavedra, J. Lamb, D. Peck, A. Sweet-Cordero, B.L.
Ebert, R.H. Mak, A.A. Ferrando, J.R. Downing, T. Jacks, H.R. Horvitz, T.R. Golub,
MicroRNA expression profiles classify human cancers, Nature 435 (2005) 834–838.
[128] R. Sandberg, J.R. Neilson, A. Sarma, P.A. Sharp, C.B. Burge, Proliferating cells
express mRNAs with shortened 3′ untranslated regions and fewer microRNA
target sites, Science 320 (2008) 1643–1647.
[129] S. Chiosea, E. Jelezcova, U. Chandran, M. Acquafondata, T. McHale, R.W. Sobol, R.
Dhir, Up-regulation of dicer, a component of the MicroRNA machinery, in prostate
adenocarcinoma, Am. J. Pathol. 169 (2006) 1812–1820.
[130] N. Sugito, H. Ishiguro, Y. Kuwabara, M. Kimura, A. Mitsui, H. Kurehara, T. Ando, R.
Mori, N. Takashima, R. Ogawa, Y. Fujii, RNASEN regulates cell proliferation and affects
survival in esophageal cancer patients, Clin. Cancer Res. 12 (2006) 7322–7328.
[131] S. Tokumaru, M. Suzuki, H. Yamada, M. Nagino, T. Takahashi, Let-7 regulates Dicer
expression and constitutes a negative feedback loop, Carcinogenesis 29 (2008)
2073–2077.
[132] J.J. Forman, A. Legesse-Miller, H.A. Coller, A search for conserved sequences in
coding regions reveals that the let-7 microRNA targets Dicer within its coding
sequence, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 14,879–14,884.
[133] Breakdown of RNAi-Based Drugs in the Clinic, RNAi News, vol. 6, 2008.
[134] J.D. Heidel, J.Y. Liu, Y. Yen, B. Zhou, B.S. Heale, J.J. Rossi, D.W. Bartlett, M.E. Davis,
Potent siRNA inhibitors of ribonucleotide reductase subunit RRM2 reduce cell
proliferation in vitro and in vivo, Clin. Cancer Res. 13 (2007) 2207–2215.
[135] R. Zukiel, S. Nowak, E. Wyszko, K. Rolle, I. Gawronska, M.Z. Barciszewska, J.
Barciszewski, Suppression of human brain tumor with interference RNA specific
for tenascin-C, Cancer Biol. Ther. 5 (2006) 1002–1007.
[136] J.D. Heidel, Z. Yu, J.Y. Liu, S.M. Rele, Y. Liang, R.K. Zeidan, D.J. Kornbrust, M.E. Davis,
Administration in non-human primates of escalating intravenous doses of
targeted nanoparticles containing ribonucleotide reductase subunit M2 siRNA,
Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 5715–5721.
[137] E. Check, A tragic setback, Nature 420 (2002) 116–118.
[138] T. Nguyen, E.M. Menocal, J. Harborth, J.H. Fruehauf, RNAi therapeutics: an update
on delivery, Curr. Opin. Mol. Ther. 10 (2008) 158–167.
[139] J. Luten, C.F. van Nostrum, S.C. De Smedt, W.E. Hennink, Biodegradable polymers as
non-viral carriers for plasmid DNA delivery, J. Control. Release 126 (2008) 97–110.
[140] S. Zhang, B. Zhao, H. Jiang, B. Wang, B. Ma, Cationic lipids and polymers mediated
vectors for delivery of siRNA, J. Control. Release 123 (2007) 1–10.
[141] A. Aigner, Nonviral in vivo delivery of therapeutic small interfering RNAs, Curr.
Opin. Mol. Ther. 9 (2007) 345–352.
[142] J.A. Hughes, G.A. Rao, Targeted polymers for gene delivery, Expert Opin. Drug
Deliv. 2 (2005) 145–157.
[143] W.T. Godbey, M.A. Barry, P. Saggau, K.K. Wu, A.G. Mikos, Poly(ethylenimine)mediated transfection: a new paradigm for gene delivery, J. Biomed. Mater. Res.
51 (2000) 321–328.
[144] M. Thomas, A.M. Klibanov, Enhancing polyethylenimine's delivery of plasmid
DNA into mammalian cells, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 14,640–14,645.
[145] Y. Kim, M. Tewari, J.D. Pajerowski, S. Cai, S. Sen, J. Williams, S. Sirsi, G. Lutz, D.E.
Discher, Polymersome delivery of siRNA and antisense oligonucleotides, J. Control.
Release 134 (2) (2009 Mar 4) 132–140.
[146] J.D. Heidel, Linear cyclodextrin-containing polymers and their use as delivery
agents, Expert Opin. Drug Deliv. 3 (2006) 641–646.
[147] J.N. Moreira, A. Santos, V. Moura, M.C. Pedroso de Lima, S. Simoes, Non-viral lipidbased nanoparticles for targeted cancer systemic gene silencing, J. Nanosci.
Nanotechnol. 8 (2008) 2187–2204.
[148] N. Blow, Small RNAs: delivering the future, Nature 450 (2007) 1117–1120.
[149] T.S. Zimmermann, A.C. Lee, A. Akinc, B. Bramlage, D. Bumcrot, M.N. Fedoruk, J.
Harborth, J.A. Heyes, L.B. Jeffs, M. John, A.D. Judge, K. Lam, K. McClintock, L.V.
Nechev, L.R. Palmer, T. Racie, I. Rohl, S. Seiffert, S. Shanmugam, V. Sood, J.
Soutschek, I. Toudjarska, A.J. Wheat, E. Yaworski, W. Zedalis, V. Koteliansky, M.
Manoharan, H.P. Vornlocher, I. MacLachlan, RNAi-mediated gene silencing in
non-human primates, Nature 441 (2006) 111–114.
[150] A. Santel, M. Aleku, O. Keil, J. Endruschat, V. Esche, B. Durieux, K. Loffler, M.
Fechtner, T. Rohl, G. Fisch, S. Dames, W. Arnold, K. Giese, A. Klippel, J. Kaufmann,
RNA interference in the mouse vascular endothelium by systemic administration
of siRNA-lipoplexes for cancer therapy, Gene Ther. 13 (2006) 1360–1370.
[151] A. Santel, M. Aleku, O. Keil, J. Endruschat, V. Esche, G. Fisch, S. Dames, K. Loffler, M.
Fechtner, W. Arnold, K. Giese, A. Klippel, J. Kaufmann, A novel siRNA-lipoplex
technology for RNA interference in the mouse vascular endothelium, Gene Ther.
13 (2006) 1222–1234.
[152] M. Aleku, G. Fisch, K. Mopert, O. Keil, W. Arnold, J. Kaufmann, A. Santel,
Intracellular localization of lipoplexed siRNA in vascular endothelial cells of
different mouse tissues, Microvasc. Res. 76 (2008) 31–41.
[153] M. Aleku, P. Schulz, O. Keil, A. Santel, U. Schaeper, B. Dieckhoff, O. Janke, J. Endruschat,
B. Durieux, N. Roder, K. Loffler, C. Lange, M. Fechtner, K. Mopert, G. Fisch, S. Dames, W.
Arnold, K. Jochims, K. Giese, B. Wiedenmann, A. Scholz, J. Kaufmann, Atu027, a
liposomal small interfering RNA formulation targeting protein kinase N3, inhibits
cancer progression, Cancer Res. 68 (2008) 9788–9798.
[154] J.S. Vorhies, J.J. Nemunaitis, Synthetic vs. natural/biodegradable polymers for
delivery of shRNA-based cancer therapies, Methods Mol. Biol. 480 (2009) 11–29.
[155] P. Kumar, H.S. Ban, S.S. Kim, H. Wu, T. Pearson, D.L. Greiner, A. Laouar, J. Yao, V.
Haridas, K. Habiro, Y.G. Yang, J.H. Jeong, K.Y. Lee, Y.H. Kim, S.W. Kim, M. Peipp, G.H.
Fey, N. Manjunath, L.D. Shultz, S.K. Lee, P. Shankar, T cell-specific siRNA delivery
suppresses HIV-1 infection in humanized mice, Cell 134 (2008 Dec 30) 577–586.
[156] K. Gao, L. Huang, Nonviral methods for siRNA Delivery, Mol. Pharmacol. 2008 Dec 30.
[157] 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. Narayanannair Jayaprakash, M. Jayaraman, M.
John, V. Koteliansky, M. Manoharan, L. Nechev, J. Qin, T. Racie, D. Raitcheva, K.G.
Rajeev, D.W. 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.
[158] S. Ghatak, V.C. Hascall, F.G. Berger, M.M. Penas, C. Davis, E. Jabari, X. He, J.S. Norris,
Y. Dang, R.R. Markwald, S. Misra, Tissue-specific shRNA delivery: a novel
approach for gene therapy in cancer, Connect. Tissue Res. 49 (2008) 265–269.
[159] P. Kumar, H. Wu, J.L. McBride, K.E. Jung, M.H. Kim, B.L. Davidson, S.K. Lee, P.
Shankar, N. Manjunath, Transvascular delivery of small interfering RNA to the
central nervous system, Nature 448 (2007) 39–43.
[160] A. Judge, I. MacLachlan, Overcoming the innate immune response to small
interfering RNA, Hum. Gene Ther. 19 (2008) 111–124.
[161] H. Lv, S. Zhang, B. Wang, S. Cui, J. Yan, Toxicity of cationic lipids and cationic
polymers in gene delivery, J. Control. Release 114 (2006) 100–109.
[162] T.G. Park, J.H. Jeong, S.W. Kim, Current status of polymeric gene delivery systems,
Adv. Drug Deliv. Rev. 58 (2006) 467–486.
[163] M. Breunig, U. Lungwitz, R. Liebl, A. Goepferich, Breaking up the correlation between
efficacy and toxicity for nonviral gene delivery, Proc. Natl. Acad. Sci. U. S. A.104 (2007)
14,454–14,459.
[164] A.L. Jackson, S.R. Bartz, J. Schelter, S.V. Kobayashi, J. Burchard, M. Mao, B. Li, G.
Cavet, P.S. Linsley, Expression profiling reveals off-target gene regulation by RNAi,
Nat. Biotechnol. 21 (2003) 635–637.
[165] A.L. Jackson, J. Burchard, J. Schelter, B.N. Chau, M. Cleary, L. Lim, P.S. Linsley,
Widespread siRNA “off-target” transcript silencing mediated by seed region
sequence complementarity, RNA 12 (2006) 1179–1187.
[166] X. Lin, X. Ruan, M.G. Anderson, J.A. McDowell, P.E. Kroeger, S.W. Fesik, Y. Shen,
siRNA-mediated off-target gene silencing triggered by a 7 nt complementation,
Nucleic Acids Res. 33 (2005) 4527–4535.
[167] A.L. Jackson, J. Burchard, D. Leake, A. Reynolds, J. Schelter, J. Guo, J.M. Johnson, L.
Lim, J. Karpilow, K. Nichols, W. Marshall, A. Khvorova, P.S. Linsley, Positionspecific chemical modification of siRNAs reduces “off-target” transcript silencing,
RNA 12 (2006) 1197–1205.
[168] A. Birmingham, E.M. Anderson, A. Reynolds, D. Ilsley-Tyree, D. Leake, Y. Fedorov, S.
Baskerville, E. Maksimova, K. Robinson, J. Karpilow, W.S. Marshall, A. Khvorova, 3′
UTR seed matches, but not overall identity, are associated with RNAi off-targets, Nat.
Methods 3 (2006) 199–204.
[169] E. Anderson, Q. Boese, A. Khvorova, J. Karpilow, Identifying siRNA-induced offtargets by microarray analysis, Methods Mol. Biol. 442 (2008) 45–63.
[170] S. van Dongen, C. Abreu-Goodger, A.J. Enright, Detecting microRNA binding and
siRNA off-target effects from expression data, Nat. Methods 5 (2008) 1023–1025.
[171] S. Qiu, C.M. Adema, T. Lane, A computational study of off-target effects of RNA
interference, Nucleic Acids Res. 33 (2005) 1834–1847.
[172] Y.K. Park, S.M. Park, Y.C. Choi, D. Lee, M. Won, Y.J. Kim, AsiDesigner: exon-based
siRNA design server considering alternative splicing, Nucleic Acids Res. 36 (2008)
W97–103.
[173] T. Yamada, S. Morishita, Accelerated off-target search algorithm for siRNA,
Bioinformatics 21 (2005) 1316–1324.
[174] Y. Naito, T. Yamada, T. Matsumiya, K. Ui-Tei, K. Saigo, S. Morishita, dsCheck: highly
sensitive off-target search software for double-stranded RNA-mediated RNA
interference, Nucleic Acids Res. 33 (2005) W589–591.
[175] Q. Du, H. Thonberg, J. Wang, C. Wahlestedt, Z. Liang, A systematic analysis of the
silencing effects of an active siRNA at all single-nucleotide mismatched target
sites, Nucleic Acids Res. 33 (2005) 1671–1677.
[176] D.S. Schwarz, H. Ding, L. Kennington, J.T. Moore, J. Schelter, J. Burchard, P.S. Linsley,
N. Aronin, Z. Xu, P.D. Zamore, Designing siRNA that distinguish between genes
that differ by a single nucleotide, PLoS Genet. 2 (2006) e140.
[177] C. Dahlgren, H.Y. Zhang, Q. Du, M. Grahn, G. Norstedt, C. Wahlestedt, Z. Liang,
Analysis of siRNA specificity on targets with double-nucleotide mismatches,
Nucleic Acids Res. 36 (2008) e53.
[178] J.K. Watts, G.F. Deleavey, M.J. Damha, Chemically modified siRNA: tools and
applications, Drug Discov. Today 13 (2008) 842–855.
[179] O. Snove Jr., J.J. Rossi, Chemical modifications rescue off-target effects of RNAi,
ACS Chem. Biol. 1 (2006) 274–276.
[180] M.A. Behlke, Chemical modification of siRNAs for in vivo use, Oligonucleotides
18 (2008) 305–319.
[181] J.B. Bramsen, M.B. Laursen, C.K. Damgaard, S.W. Lena, B.R. Babu, J. Wengel, J. Kjems,
Improved silencing properties using small internally segmented interfering RNAs,
Nucleic Acids Res. 35 (2007) 5886–5897.
[182] P.Y. Chen, L. Weinmann, D. Gaidatzis, Y. Pei, M. Zavolan, T. Tuschl, G. Meister,
Strand-specific 5′-O-methylation of siRNA duplexes controls guide strand
selection and targeting specificity, RNA 14 (2008) 263–274.
D.D. Rao et al. / Advanced Drug Delivery Reviews 61 (2009) 746–759
[183] K. Ui-Tei, Y. Naito, S. Zenno, K. Nishi, K. Yamato, F. Takahashi, A. Juni, K. Saigo,
Functional dissection of siRNA sequence by systematic DNA substitution: modified
siRNA with a DNA seed arm is a powerful tool for mammalian gene silencing with
significantly reduced off-target effect, Nucleic Acids Res. 36 (2008) 2136–2151.
[184] D. Rao, N. Senzer, M.A. Cleary, J. Nemunaitis, Comparative assessment of siRNA
and shRNA off target effects: what is slowing clinical development. Cancer Gene
Therapy (2009).
[185] D. Bumcrot, M. Manoharan, V. Koteliansky, D.W. Sah, RNAi therapeutics: a
potential new class of pharmaceutical drugs, Nat. Chem. Biol. 2 (2006) 711–719.
[186] A.J. Bridge, S. Pebernard, A. Ducraux, A.L. Nicoulaz, R. Iggo, Induction of an
interferon response by RNAi vectors in mammalian cells, Nat. Genet. 34 (2003)
263–264.
[187] C.A. Sledz, M. Holko, M.J. de Veer, R.H. Silverman, B.R. Williams, Activation of the
interferon system by short-interfering RNAs, Nat. Cell Biol. 5 (2003) 834–839.
[188] K. Kariko, P. Bhuyan, J. Capodici, D. Weissman, Small interfering RNAs mediate
sequence-independent gene suppression and induce immune activation by
signaling through toll-like receptor 3, J. Immunol. 172 (2004) 6545–6549.
[189] V. Hornung, M. Guenthner-Biller, C. Bourquin, A. Ablasser, M. Schlee, S. Uematsu,
A. Noronha, M. Manoharan, S. Akira, A. de Fougerolles, S. Endres, G. Hartmann,
Sequence-specific potent induction of IFN-alpha by short interfering RNA in
plasmacytoid dendritic cells through TLR7, Nat. Med. 11 (2005) 263–270.
[190] A.D. Judge, V. Sood, J.R. Shaw, D. Fang, K. McClintock, I. MacLachlan, Sequencedependent stimulation of the mammalian innate immune response by synthetic
siRNA, Nat. Biotechnol. 23 (2005) 457–462.
[191] S.S. Diebold, T. Kaisho, H. Hemmi, S. Akira, C. Reis e Sousa, Innate antiviral
responses by means of TLR7-mediated recognition of single-stranded RNA,
Science 303 (2004) 1529–1531.
[192] F. Heil, H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira, G. Lipford,
H. Wagner, S. Bauer, Species-specific recognition of single-stranded RNA via tolllike receptor 7 and 8, Science 303 (2004) 1526–1529.
[193] M. Karin, Y. Yamamoto, Q.M. Wang, The IKK NF-kappa B system: a treasure trove
for drug development, Nat. Rev., Drug Discov. 3 (2004) 17–26.
[194] J.T. Marques, T. Devosse, D. Wang, M. Zamanian-Daryoush, P. Serbinowski, R.
Hartmann, T. Fujita, M.A. Behlke, B.R. Williams, A structural basis for
discriminating between self and nonself double-stranded RNAs in mammalian
cells, Nat. Biotechnol. 24 (2006) 559–565.
[195] M. Sioud, RNA interference and innate immunity, Adv. Drug Deliv. Rev. 59 (2007)
153–163.
[196] M.E. Kleinman, K. Yamada, A. Takeda, V. Chandrasekaran, M. Nozaki, J.Z. Baffi, R.J.
Albuquerque, S. Yamasaki, M. Itaya, Y. Pan, B. Appukuttan, D. Gibbs, Z. Yang, K.
Kariko, B.K. Ambati, T.A. Wilgus, L.A. DiPietro, E. Sakurai, K. Zhang, J.R. Smith, E.W.
Taylor, J. Ambati, Sequence- and target-independent angiogenesis suppression by
siRNA via TLR3, Nature 452 (2008) 591–597.
[197] M. Robbins, A. Judge, E. Ambegia, C. Choi, E. Yaworski, L. Palmer, K. McClintock, I.
Maclachlan, Misinterpreting the therapeutic effects of siRNA caused by immune
stimulation, Hum. Gene Ther. 19 (10) (2008 Oct) 991–999.
[198] S. Bauer, C.J. Kirschning, H. Hacker, V. Redecke, S. Hausmann, S. Akira, H. Wagner,
G.B. Lipford, Human TLR9 confers responsiveness to bacterial DNA via speciesspecific CpG motif recognition, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 9237–9242.
759
[199] H. Hemmi, O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, K.
Hoshino, H. Wagner, K. Takeda, S. Akira, A toll-like receptor recognizes bacterial
DNA, Nature 408 (2000) 740–745.
[200] H.K. de Wolf, N. Johansson, A.T. Thong, C.J. Snel, E. Mastrobattista, W.E. Hennink,
G. Storm, Plasmid CpG depletion improves degree and duration of tumor gene
expression after intravenous administration of polyplexes, Pharm. Res. 25 (2008)
1654–1662.
[201] M.A. Robbins, M. Li, I. Leung, H. Li, D.V. Boyer, Y. Song, M.A. Behlke, J.J. Rossi, Stable
expression of shRNAs in human CD34+ progenitor cells can avoid induction of
interferon responses to siRNAs in vitro, Nat. Biotechnol. 24 (2006) 566–571.
[202] M. Sioud, Induction of inflammatory cytokines and interferon responses by
double-stranded and single-stranded siRNAs is sequence-dependent and
requires endosomal localization, J. Mol. Biol. 348 (2005) 1079–1090.
[203] H. Akashi, M. Miyagishi, T. Yokota, T. Watanabe, T. Hino, K. Nishina, M. Kohara, K.
Taira, Escape from the interferon response associated with RNA interference using
vectors that encode long modified hairpin-RNA, Mol. Biosyst. 1 (2005) 382–390.
[204] T. Watanabe, M. Sudoh, M. Miyagishi, H. Akashi, M. Arai, K. Inoue, K. Taira, M.
Yoshiba, M. Kohara, Intracellular-diced dsRNA has enhanced efficacy for silencing
HCV RNA and overcomes variation in the viral genotype, Gene Ther. 13 (2006)
883–892.
[205] M. Bauer, N. Kinkl, A. Meixner, E. Kremmer, M. Riemenschneider, H. Forstl, T.
Gasser, M. Ueffing, Prevention of interferon-stimulated gene expression using
microRNA-designed hairpins, Gene Ther. 16 (2009) 142–147.
[206] M. Sioud, Single-stranded small interfering RNA are more immunostimulatory
than their double-stranded counterparts: a central role for 2′-hydroxyl uridines
in immune responses, Eur. J. Immunol. 36 (2006) 1222–1230.
[207] J.M. Layzer, A.P. McCaffrey, A.K. Tanner, Z. Huang, M.A. Kay, B.A. Sullenger, In vivo
activity of nuclease-resistant siRNAs, RNA 10 (2004) 766–771.
[208] K. Kariko, M. Buckstein, H. Ni, D. Weissman, Suppression of RNA recognition by
Toll-like receptors: the impact of nucleoside modification and the evolutionary
origin of RNA, Immunity 23 (2005) 165–175.
[209] D. Grimm, K.L. Streetz, C.L. Jopling, T.A. Storm, K. Pandey, C.R. Davis, P. Marion, F.
Salazar, M.A. Kay, Fatality in mice due to oversaturation of cellular microRNA/short
hairpin RNA pathways, Nature 441 (2006) 537–541.
[210] D. Grimm, M.A. Kay, Therapeutic application of RNAi: is mRNA targeting finally
ready for prime time? J. Clin. Invest. 117 (2007) 3633–3641.
[211] M. John, R. Constien, A. Akinc, M. Goldberg, Y.A. Moon, M. Spranger, P. Hadwiger, J.
Soutschek, H.P. Vornlocher, M. Manoharan, M. Stoffel, R. Langer, D.G. Anderson, J.D.
Horton, V. Koteliansky, D. Bumcrot, Effective RNAi-mediated gene silencing without
interruption of the endogenous microRNA pathway, Nature 449 (2007) 745–747.
[212] R. Yi, B.P. Doehle, Y. Qin, I.G. Macara, B.R. Cullen, Overexpression of exportin 5
enhances RNA interference mediated by short hairpin RNAs and microRNAs, RNA
11 (2005) 220–226.
[213] J.C. Giering, D. Grimm, T.A. Storm, M.A. Kay, Expression of shRNA from a tissuespecific pol II promoter is an effective and safe RNAi therapeutic, Mol. Ther.16 (2008)
1630–1636.
`