DRUG DEVELOPMENTRNAi AND miRNA-ORIENTED IN CONTEMPORARY THERAPY ONCOLOGY RNA interference and micro- RNA –oriented therapy in cancer: rationales, promises, and challenges † T.F. Duchaine PhD,* and F.J. Slack PhD ABSTRACT the late 1990s, the tenets of the “central dogma” have undergone revision as scientists began to appreciate The discovery that RNA interference (RNA i) and its that RNA s, by themselves, play key regulatory roles. functional derivatives, small interfering RNA s (siR- Leading the charge in this revolution are the NA s) and micro-RNA s (miRNA s) could mediate potent small (18–25 nucleotide) non-coding RNA s called and specific gene silencing has raised high hopes for “small interfering” RNA s (siRNA s) and micro-RNA s cancer therapeutics. The prevalence of these small (miRNA s), which direct RNA interference (RNA i) (18–25 nucleotide) non-coding RNA s in human gene phenomena. The siRNA s and miRNA s are newly networks, coupled with their unique specificity, discovered regulatory molecules that control a va- has paved the way for the development of new and riety of biologic processes, including development, promising therapeutic strategies in re-directing or metabolism, and aging 1–3. The miRNA s often bind inhibiting small RNA phenomena. to complementary sequences in the 3′ untranslated Three strategies are currently being developed: regions of target messenger RNA s (m RNA s) to repress target gene expression; the siRNA s can bind to any • De novo RNA i programming using synthetic siRNA s complementary sequence in a target mRNA . Impor- to target the expression of genes tantly, functional roles have been demonstrated for • Strengthening or recapitulation of the physiologic miRNA s in cancer 4,5, and the potential exists to use targeting of messenger RNA s by specific miRNA s similar molecules as cancer therapeutics. • Sequence-specific inhibition of miRNA functions The present review outlines the underlying by nucleic acid analogs rationales for considering small RNA therapeutics. Although most of the attention has been given to the Each strategy, currently being developed both in use of siRNA s to inhibit gene function, we provide a academia and in industry, holds promise in cancer broader overview and include novel and emerging therapeutics. strategies that are instead based on miRNA s. We highlight the potential of each strategy for cancer KEY WORDS treatment and summarize the challenges that need to be overcome to fully harness RNA i- and miRNA - Cancer therapy, siRNA , miRNA , RNA i therapy, oncomirs, based therapies. nucleic acid analogs, tumour-suppressor miRNA s 2. THE MOLECULAR BASIS FOR siRNA- AND 1. INTRODUCTION miRNA-MEDIATED SILENCING These are very interesting times in the field of gene The silencing mediated by RNA i and miRNA is initi- regulation as it affects development and cancer. Since ated when the double-stranded RNA (dsRNA )–specific type III endonuclease, Dicer, “recognizes” a dsRNA substrate. Processing by Dicer and its accessory proteins yields 18- to 25-nucleotide heteroduplex small RNA s (either siRNA s or miRNA s, depending on their origin). The miRNA s are usually endogenously This scientific paper is the work of the authors and was made transcribed; the siRNA s often derive from exogenous possible through the support of Merck Frosst Canada Ltd. The sources such as viruses 6,7. One of the strands of the opinions and information contained herein are those of the small RNA duplex is selected, loaded into Argonaute author and do not necessarily reflect the views or opinions proteins, and assembled into an effector complex of Merck Frosst Canada Ltd. such as the RNA -induced silencing complex (RISC ). 61 Copyright © 2009 Multimed Inc. CURRENT ON co L O GY —VO LUME 16, NUMBER 4 DUCHAINE and SLACK The other strand is often cleaved or unwound and miRNA s in transformation and tumour growth 21–23. degraded 8,9. A key feature for the use of RNA i in In an increasing number of reports, miRNA s are now therapy is that the RISC can be directly programmed being shown to be capable of acting as bona fide by exogenous mature siRNA s or miRNA s. In effect, this tumour suppressors or oncogenes, bringing about a programming bypasses cleavage by Dicer and the major shift in the landscape of cancer research 24. interferon response triggered by dsRNA s larger than 30 bp in mammalian cells 10. Generally, when a siRNA 4. siRNA- AND miRNA-BASED THERAPEUTIC fully pairs with a target mRNA , the result is rapid and STRATEGIES specific cleavage, an activity named “slicer” 11. In contrast to siRNA s, miRNA s normally originate The chemical properties of the various miRNA s vary from gene-encoded precursor transcripts that are little, and yet, through the sequences that they encode, processed into precursor miRNA s. Those precursors they direct completely distinct activities. At least fold as hairpins and serve as Dicer substrates 12. three distinct overall strategies were adopted early Embedded within the RISC , mature miRNA s pair only on for siRNA - and miRNA -based therapies. Each one partially with their mRNA targets, thereby directing aims (reasonably) to exploit the sequence-directed translation inhibition or reducing the stability of properties of the small RNA s: their target 13. The gene-targeting information is non-symmet- • First, de novo reprogramming of the RNA i ma- rically encoded within mature siRNA s or miRNA s. The chinery can be utilized against a gene that has 5′ portion of the miRNA at positions 2–7 is thought deleterious effects (Figure 1, panel 1). to encode the most important determinant to direct • Second, a physiologic miRNA –mRNA pair could base-pairing with mRNA targets. This portion (called be reproduced by programming the RISC using a the “core,” “nucleus,” or “seed” sequence) is most known, beneficial miRNA —for example, a tumour commonly conserved; it defines paralogous and suppressor miRNA (Figure 1, panel 2). orthologous families of miRNA s 14. Structural stud- • Third, a nonhydrolyzable mimic of an RNA target ies show a tight coordination of this 5′ determinant could be generated to effectively sequester delete- within the Argonaute proteins 15. The 3′ part of the rious miRNA s from their pro-oncogenic functions miRNA or siRNA is more amenable to chemical modi- (Figure 1, panel 3). fications, some of which occur naturally. 4.1 Harnessing RNAi: De Novo Programming with 3. THE BIOLOGIC IMPLICATIONS OF miRNAS siRNAs FOR CANCER Perhaps the simplest and the most direct therapeutic To evaluate the potential of small RNA s in cancer strategy involving RNA i is to harness and redirect its therapy, an appreciation of how extensively they are machinery against a detrimental mRNA target. It there- tied to the normal cell’s gene regulation networks is fore comes as no surprise that this strategy is by far the essential. In human cells, miRNA s are estimated to most advanced in clinical trials. The siRNA -directed account for more than 1000 genes (approximately targeting of genes is currently being developed as a 3% of the total estimated number of genes) and to potential therapy in a great number of diseases 25,26. exert direct negative regulative pressure on more than The keys to this strategy are identification of a func- 30% of the entire coding gene set 14,16. Accordingly, tional and effective RNA i target and optimization of miRNA s are known to regulate genes implicated in the targeting process. The strategy can include, but is a wide range of cellular functions in development not limited to, knowing the uniqueness of the siRNA ’s and homeostasis 17. In Caenorhabditis elegans, for target sequence in the genome, optimizing its loading example, the first two known miRNA s—lin-4 and into the RISC , obtaining a measure of the accessibility let-7—control timing of cell differentiation during of the siRNA target site (structure, protein–RNA interac- development. In let-7 mutants, stem cells in the skin tions), and possibly even adding tolerated covalent frequently fail to terminally differentiate, electing modifications to the siRNA . In addition, and common instead to divide again. Such cellular events are the to each RNA i- or miRNA -based therapy, is the need to defining hallmarks of cancer 18. deliver the RNA molecule or analog to the correct tissue Because misregulation of genes that control and mRNA target (see the “Challenges” subsection). cell proliferation and cell fate often contribute to Many siRNA -based therapies are under develop- cancer development, miRNA s are unsurprisingly ment, but surprisingly few are directed at oncogenic also implicated in cancer biology. Early evidence targets. Among the cancer-directed approaches is the in humans indicates that miRNA loci are associated targeting of pleiotrophin, which effectively reduced with chromosome sites that are often deleted or tumour growth in mouse xenograft models of glioblas- amplified 19, and whose expression is dramatically toma multiforme 27. Another initiative of interest is the altered during the onset of cancer 20. Intense current successful targeting of the Ews–Fli1 fusion oncoprotein research aims to precisely delineate the functions of in mouse models of metastatic Ewing sarcoma 28. 62 CURRENT ON co L O GY —VO LUME 16, NUMBER 4 RNAi AND miRNA-ORIENTED THERAPY cancer involving an activated K-ras allele suppresses activating oncogenic ras mutations just as such overexpression does in C. elegans 33,34. Signalling by ras plays an important role in radioresistance, and radiotherapy is one of the three primary modalities used in combatting cancer. However, many cancers, including lung cancer, are radioresistant. Because let-7 represses ras, let-7 overexpression has been proposed to be a possible cause of radiosensitivity, a hypothesis that indeed proved to be the case in both lung cancer cells and in an in vivo C.