| May
1, 2005 |
| By:
Chandu
Ammini, David
Suhy, Sara
Cunningham |
| Pharmaceutical
Discovery |
|
Hailed the 'Breakthrough of the Year" in 2002 by Science
magazine (Washington DC, USA) and representing a staple research tool in
industry and academia, RNAi technology has gone from proof of principle
in animal models to human clinical trials in less than three years.
As a technology, several key events shaped the field of RNA
interference as we know it. The phenomenon of RNA interference first was
reported in plants (1) by Richard Jorgensen in 1990, although the exact
mechanism was only understood a few years later. Jorgensen's group
reported the curious phenomenon that petunia pigment genes were shut
down when they inserted extra copies of the genes in an attempt to
deepen the purple color. In 1998, Andrew Fire and Craig Mello (2)
reported that double-stranded RNAs (dsRNAs) injected into the nematode Caenorhabditis
elegans silenced the corresponding genes containing complementary
sequences. Benitec's (Mountain View, California, USA) Graham et al.,
demonstrated the universality of RNAi and invented DNA constructs to
trigger the RNAi process in human and mammalian cells (3). Tuschl and
colleagues (4) provided a biochemical understanding of the RNAi pathway
and showed that the functional units of RNAi likely are represented by
dsRNAs shorter than 30 base pairs. McCaffrey et al., published the first
in vivo evidence of RNA interference in adult mice (5), verifying
the work of Graham et al., and allowing therapeutic RNAi companies to
capitalize on insights derived from the past decade of gene therapy
research.
Figure 1. General mechanism of RNA
interference and ddRNAi action within a cell.
|
The Mechanism of RNA Interference
The RNA interference pathway is present in every cell of virtually every
multi-cellular organism (6). It is likely that RNAi evolved as an innate
mechanism for cellular defense against dsRNA viruses. As shown in Figure
1, long dsRNA molecules (either RNA virus genomes or short hairpin RNAs
[shRNAs] produced from ddRNAi expression vectors) are processed into
small interfering RNA (siRNA) by the cytoplasmic enzyme called Dicer.
The "diced" siRNAs rapidly unwind and are assembled into the
RNA-induced silencing complex (RISC). RISC mediates sequence-specific
binding of these guide RNAs to a corresponding messenger RNA (mRNA) and
catalyzes the cleavage and destruction of the mRNA, enabling
gene-specific silencing.
Double-stranded RNA can be introduced into cells either by the
delivery of pre-synthesized siRNA duplexes ("delivered" RNAi)
or by the intracellular expression of shRNAs from a DNA template
followed by hairpin processing to create functional siRNAs
("expressed" RNAi). Both delivered and expressed RNAi pathways
converge at the RISC complex and induce gene silencing, as outlined in
Figure 1. The choice of delivered vs. expressed depends on issues of
convenience, desired level of persistence of expression, clinical
indication and the required method of introduction of the RNAi molecules
into target tissues. Delivered RNAi will not be discussed further in
this application note, and the reader is referred to other excellent
reviews on the topic (7-10).
Figure 2. Strategies for
expression of multiple shRNA by ddRNAi.
|
Expressed RNAi (ddRNAi) Induction
of RNAi via a DNA intermediate can be achieved using either viral or
non-viral delivery systems and is covered under the issued and pending
patent estate of Benitec Ltd., worldwide. The ddRNAi molecules can be
generated from many different combinations of promoters, terminators and
hairpin structures (Figure 2). Figure 2 outlines three such strategies
that can be utilized to generate multiple shRNA from a single vector.
Although the pre-processed forms of the hairpins for the three shRNAs
are very different in the three strategies, the cellular RNAi machinery
generates comparable mature, processed shRNA molecules that converge at
the RISC complex and cause target-specific silencing of the three target
genes in this example.
DNA-derived RNA interference
technology may be used in applications as diverse as therapeutic
development, creation of animal models for human diseases, research
reagents and drug target identification and validation.
Table I. Comparison of ddRNAi and
leading drug development platforms
|
Therapeutic development. RNAi
drugs exploit the naturally occurring RNA interference pathway to
silence disease agents and trigger either disease resolution or viral
clearance. The advantages of ddRNAi over small molecules and antibodies
are outlined in Table I. Small-molecule drug development typically is a
12-15 year dedicated process. Antibody drugs have a shorter development
cycle, due to partial overlap of the first two stages. RNAi technology,
on the other hand, is expected to have significantly shorter development
times because stages one and two are completed simultaneously. RNAi also
allows the targeting of almost any disease, including many hitherto
"undruggable" targets, due to the adaptability of the method
to virtually any known nucleic acid sequence. Furthermore, the ddRNAi
approach allows for the delivery and long-term expression of multiple
shRNAs in the same formulation, reducing the need for repeat dosing.
The key challenge to ddRNAi-based
therapies is to ensure safe and efficient drug delivery. Viral systems
based on retrovirus, herpes simplex virus, adenovirus and adeno-associated
virus (AAV) have been employed to exploit the targeting properties of
the viruses for tissue tropism, as demonstrated in multiple gene therapy
clinical trials (11-15). However, due to the real and perceived
drawbacks of viral delivery, the search for non-viral delivery systems
has led to the development of multiple liposomal formulations for
nucleic acid delivery, in addition to the mechanical administration of
naked or modified DNA (16). These methods offer the benefit of ddRNAi
without the associated complications of viral vectors but remain
unproven in the clinic (17).
Table II. RNAi drug development
pipeline
|
The current pipeline of RNAi-derived
drugs — representing delivered as well as expressed approaches — is
shown in Table II. A number of ddRNAi-derived investigational new drug (IND)
applications are expected in 2006 from Benitec, Nucleonics (Horsham,
Pennsylvania, USA) and Sirna Therapeutics (Boulder, Colorado, USA). For
example, Benitec Ltd., currently is manufacturing clinical-grade
materials for an AIDS Lymphoma clinical trial estimated to begin in the
first half of 2006. This RNA combination drug has the potential to
provide a first-in-class therapy for AIDS lymphoma patients who
currently have no viable treatment option.
Table III. Firms using or
considering ddRNAi for the creation of human disease models
|
Creation of animal models for human
diseases. Creation of gene knockdown animals using ddRNAi technology
is a clever application of gene silencing that transgenic companies such
as ARTEMIS Pharmaceuticals (Cologne, Germany), genOway (Lyon, France)
and RxGen (New Haven, Connecticut, USA) use for the creation of new
animal models for human diseases in mice, rats and primates (Table III).
In mice, ddRNAi-mediated knockouts can be developed within four to five
months, representing a significant shortening of the typical year-long
process by conventional means. In general, this method involves the
transfection of embryonic stem (ES) cells with a vector bearing the
gene-specific short hairpin RNA (shRNA). The shRNA gene is targeted to a
specific locus in the mouse genome. Transduced ES cells then are
injected into blastocysts, chimera are generated and transgenic mice are
created by conventional breeding techniques. Multiple variations of this
method, including the use of cre-lox recombination methods, are employed
by different companies.
ARTEMIS creates gene-specific
knockout mice using its proprietary ArteMice™ RNAi technology (18)
(Table III). Using reporter systems, ARTEMIS has demonstrated over 80%
knockdown in almost all tissues of the body and has demonstrated that
such knockdowns are stable for over 25 weeks and are heritable into the
next generation. genOway (19) uses targeted insertion of the shRNA into
the HPRT locus of mice to enable ubiquitous, tissue-specific and
inducible expression of the shRNA construct. The company's in vivo
RNAi™ technology enables inducible, tissue-specific regulation of
shRNA expression, permitting target validation studies under physio-pathological
conditions.
Primate models of human disease
have several advantages over rodents for the prediction of both drug
action and off-target drug effects, due to the close homology to humans.
RxGen (20) specializes in creating human disease models using the
African green monkey, which is genetically closer to humans than other
Old World monkeys and provides the advantages of a smaller body size
than rhesus and cynomologus monkeys, thus requiring less drug for
testing; they also are ideally suited for modeling a wide range of human
diseases. The existence of well-characterized African green monkey cell
lines (COS-7 and Vero) enables the translation of in vitro
optimization studies to the primate models. RxGen is a firm to watch for
the creation of multiple primate models for human diseases using both
ddRNAi and proprietary technologies.
Although transgenic RNAi
technologies have immense potential, they are not yet as well
characterized as traditional gene knockout techniques. It will be up to
these and other pioneering companies to create novel animal models of
human disease that are not addressable by traditional gene knockout
techniques. Transgenic RNAi has the potential to become a mainstream
technology in the near future.
Research reagents. Research
reagent companies were quick to capitalize on the creation of multiple
product lines of ddRNAi expression vectors, kits and reagents for
enabling RNA interference in cells and tissues in vitro and in
vivo. Leading research reagent companies such as Ambion (Austin,
Texas, USA), Promega (Madison, Wisconsin, USA), BD Biosciences Clontech
(Palo Alto, California, USA) and Invitrogen (Carlsbad, California, USA)
now offer a family of vectors, both viral and non-viral, with numerous
transcriptional promoter/terminator combinations to enable the rapid
adoption of ddRNAi for a variety of uses.
Drug target identification and
validation. Following the Human Genome Project and development of
microarrays and other functional genomics tools, a steady stream of
"druggable" genomic targets rapidly saturated the pipelines of
drug development companies. Yet, the lack of efficient tools to
independently validate the roles of these candidate targets in disease
etiology was a major bottleneck for therapeutic development. Antisense
oligonucleotides (ASOs) were used initially to modulate the activity of
disease targets in tissue culture and small animal model systems.
However, ASO technology is not catalytic, adequate intracellular
concentrations are not easily achieved to enable sufficient knockdown of
multiple targets and the ASO must passively diffuse to its target in the
cell. RNA interference, however, is a catalytic event and the RISC
actively and precisely guides the RNAi molecule to its target mRNA.
Thus, it became a staple tool for rapid target validation. In
particular, a vector-based approach for gene silencing (ddRNAi) is
extremely attractive for long-term knockdown of single or multiple
disease targets.
In addition to providing the
research tools for target validation, RNAi technology rapidly is
evolving as a tool for genome-wide screens for gene functionality. The
robustness and universality of RNAi has led several independent groups
to create RNAi libraries for screening new druggable targets. Although
these libraries typically contain RNAi species against several dozen
targets in focused biological pathways, efforts are underway to generate
libraries that would saturate genomic targets. Initially, broad
libraries of synthesized siRNAs were used to rapidly probe a set of
desired targets. Recent technological advances, however, have created
shRNA-based approaches for comparable large-scale screens. System
Biosciences (Mountain View, California, USA) combines lentiviral vectors
and shRNA expression libraries in an attempt to identify genes
functionally responsible for any selectable phenotype in a single
genome-wide screening. Galapagos Genomics (Mechelen, Belgium), through
its Dutch subsidiary Galadeno (Leiden, The Netherlands), offers siRNA
knockdown services targeting a large number of genes or gene families
using shRNA-expressing adenoviral libraries.
Greg Hannon's group (21) at Cold
Spring Harbor Laboratory (Cold Spring Harbor, New York, USA) has
developed an innovative shRNA-based plasmid library comprised of over
28,000 shRNA expression cassettes that can theoretically target greater
than 9600 human genes. Their plasmid system is coupled to a
matrix-assisted genetically integrated cloning technology, which enables
efficient recovery of plasmid DNA. This technique should provide the
ability to screen on the single cell level, as well as generate pools of
highly functional shRNA species following several rounds of enrichment.
Additional features include an shRNA specific 60-nucleotide bar code
sequence, permitting rapid profiling of enriched species using
microarrays.
RNAi in the Future
RNA interference via ddRNAi has emerged as a powerful platform for
diverse biotechnology applications. In less than three years, this
technology has spawned multiple new business entities for drug
development, research reagent manufacture, creation of animal models,
drug target validation and target identification. Investigational drugs
based on ddRNAi technology should be entering the clinic in the next few
years. Development of shRNA-based large-scale screens should provide
additional tools for deciphering biological networks and the etiology of
complex diseases.
Acknowledgment
We would like to thank Sally Brashears for critical reading of the
manuscript.
References
1. C. Napoli, C. Lemiex and R.A. Jorgenson, Plant Cell 2,
279-289 (1990).
2. A. Fire, S. Xu, M.K. Montgomery
et al., Nature 391, 806-811 (1998).
3. M.W. Graham, Genetic constructs
for delaying or repressing the expression of a target gene, US patent
6,573,099.
4. S.M. Elbashir, J. Harborth, W.
Lendeckel et al., Nature 411, 494-498 (2001).
5. A.P. McCaffrey, L. Meuse, T.T.
Pham et al., Nature 418, 38-39 (2002).
6. G.J. Hannon, Nature
418, 244-251 (2002).
7. I. Bantounas, L.A. Phylactou and
J.B. Uney, J Mol. Endocrinol. 33(3), 545-557 (2004).
8. M. Nesterova and Y.S. Cho-Chung,
Curr. Drug Targets 5, 683-689 (2004).
9. T.C. Karagiannis and A. El-Osta,
Cancer Biol. Ther. 3, 1069-1074 (2004).
10. R.C. Ryther, A.S. Flynt, J.A.
Phillips and J.G. Patton, Gene Ther. 12, 5-11 (2004).
11. R.G. Amado, R.T. Mitsuyasu et
al., Hum. Gene Ther. 15, 251-262 (2004).
12. A. Michienzi, D. Castanotto, N.
Lee et al., Ann. NY Acad. Sci. 1002, 63-71 (2003).
13. T.R. Flotte, M.L. Brantly, L.T.
Spencer et al., Hum. Gene Ther. 15, 93-128 (2004).
14. E.A. Chiocca, K.M. Abbed et
al., Mol. Ther. 10, 958-966 (2004).
15. K. Lundstrom, Technol.
Cancer Res. Treat. 3, 467-477 (2004).
16. G. Romano, P. Micheli, C.
Pacilio and A. Giordano Stem Cells 18, 19-39 (2000).
17. G.J. Hannon and J.J. Rossi, Nature
431, 371-378 (2004).
18. ARTEMIS website: http://www.artemispharma.de.
19. genOway website: http://www.genoway.com/safe_
in_vivo_rnai.htm.
20. RxGen
website
21. P.J. Paddison, J.M. Silva, D.S.
Conklin et al., Nature 428(6981), 427-431 (2004).
Chandu Ammini is business
development manager, David Suhy is senior scientist and Sara
Cunningham is chief executive officer at Benitec Ltd. Chandu Ammini
can be reached at 2375 Garcia Avenue, Mountain View, California 94043
USA. Tel. 650-564-9850; e-mail cammini@benitec.com
.
|
