Introduction

The ability to use double-stranded RNA (dsRNA) molecules to silence the expression of target genes has greatly expanded the repertoire of questions scientists can address in biological systems. RNA interference (RNAi) is now routinely used in laboratories for loss-of-function analyses and, increasingly, for the rescue of phenotypes caused by dominant acting mutant genes. In the field of dominantly inherited neurodegenerative disorders, for example, scientists have succeeded in applying RNAi in vivo to block the toxic effects of disease genes. RNAi-mediated suppression of mutant transgenes has been shown to delay the onset and progression of disease phenotypes in various animal models of neurodegenerative diseases, including spinocerebellar ataxia type 1 (SCA1),¹ Huntington's disease^{2, 3} and amyotrophic lateral sclerosis.^{4, 5}

Many of the disease-causing proteins in these disorders have essential and/or unknown functions in the organism. Therefore, indiscriminate, sustained interruption of their cellular expression could result in untoward effects. In fact, in many of the studies mentioned above, silencing of the mutant human transgene was achieved while expression of the endogenous mouse gene remained normal. This reflects the fact that the sequences being targeted in the human disease transgenes were not always conserved between the human and mouse genes.

While the field of RNAi therapy is moving quickly toward the goal of clinical application in humans, its success in some diseases may depend on the ability to target mutant transcripts specifically without reducing the levels of wild-type alleles, especially when the targeted gene has an essential function. Therefore, strategies designed to silence the expression of a mutant allele in a selective manner should be considered. Here, we review how the RNAi machinery can be experimentally manipulated to suppress the expression of mutant transcripts in an allele-specific manner. In addition, we discuss recent findings on RNAi mechanisms that can be applied to RNAi design in order to improve the efficacy and specificity of RNAi therapy, as these are critically important factors in achieving allele-specific silencing.

RNA interference: siRNAs and miRNAs

RNA-mediated silencing refers to an evolutionarily conserved process in which cells respond to the presence of dsRNA molecules by initiating a cascade of events that repress target genes in a sequence-specific manner. Three related forms of RNA-mediated gene silencing act on different aspects of gene expression: modulating transcription, messenger RNA (mRNA) stability or mRNA translation. The best studied form of RNA-mediated silencing is RNAi.

First discovered in plants and initially characterized in Caenorhabditis elegans, RNAi is initiated when dsRNA molecules are processed to small-interfering RNAs (siRNAs) by the ribonuclease-III (RNase-III) enzyme, Dicer. siRNAs are 21–23 nucleotide (nt) dsRNA intermediates that associate with and confer substrate specificity upon a protein complex called the RNA-induced silencing complex (RISC). Unwinding of the siRNA duplex is coupled to the incorporation into RISC of one strand, known as the guide strand, and to the selective elimination of the second strand, known as the passenger strand. Groups led by Zamore and Khvorova have demonstrated that strand selection is likely a thermodynamically dependent process dictated by the internal stability of the siRNA molecule.^{6, 7} Once assembled into RISC, siRNAs mediate the sequence-specific cleavage of complementary mRNA molecules. In RISC, cleavage of the targeted transcript is accomplished by a member of the Argonaute family of proteins, Argonaute-2 (Ago2), the 'slicer' component of RISC.⁸

The initial RNAi experiments carried out by Fire and Mello⁹ in C. elegans relied on the use of exogenous dsRNA molecules. Soon afterward, others demonstrated the existence of endogenous siRNA-like molecules termed microRNAs (miRNAs). miRNAs are small, noncoding RNAs that act via the RNAi pathway to regulate endogenous gene expression. They coordinate many aspects of cellular function including development, differentiation, proliferation and apoptosis.¹⁰ They are also necessary for proper stem cell division¹¹ and their aberrant expression has been implicated in the onset and progression of cancer.¹⁰

miRNAs are the product of a two-step process that begins in the nucleus with the cleavage of long primary-miRNA (pri-miRNA) transcripts by the microProcessor, a protein complex consisting of the RNase-III enzyme Drosha and its partner DGCR8 ('Pasha' in flies and worm).¹² The microProcessor recognizes a 70-nt long stem-loop hairpin structure contained within the pri-miRNA transcript and releases this hairpin from the larger transcript. The resulting precursor miRNA hairpin (a pre-miRNA) is exported to the cytoplasm where Dicer then cleaves the RNA to the mature 22-nt miRNA molecule. In mammals, miRNAs do not cleave the targeted mRNA but rather bind to the targeted mRNA and suppress its translation into protein.¹³ The difference in activity of siRNAs (which cleave targeted mRNAs) and miRNAs (which inhibit translation of targeted mRNAs) is attributed to the lack of perfect complementarity between miRNAs and their intended targets. Generally, miRNA targets are perfectly complementary to nucleotide positions 2–8 of the mature miRNA but lack complementarity at one or more nucleotides elsewhere in the sequence. In contrast, siRNAs are designed to be perfectly complementary to their targeted sequence. The basis for cleavage of mRNAs by RISC is increasingly understood, but the mechanism and cellular machinery through which miRNAs cause translational arrest without cleavage is just beginning to be elucidated.

Allele-specificity: application to neurological disease

The ability to achieve allele-specific silencing stems from the exquisite sequence specificity of RNAi. Dominantly inherited CNS disorders constitute particularly fertile ground for the application of allele-specific approaches. This is because affected (or at-risk) persons carry one dominant-acting mutant allele and one normal allele of the disease gene. Dominantly inherited neurodegenerative diseases, of which there are many, represent particularly attractive targets. Among these are familial forms of Alzheimer's disease (AD), Parkinson's disease, and amyotrophic lateral sclerosis; frontotemporal dementia with parkinsonism linked to chromosome-17; and the nine known polyglutamine neurodegenerative disorders including Huntington's disease (HD) and several spinocerebellar ataxias (SCA types 1, 2, 3, 6, 7 and 17). In these devastating brain disorders, expression of a mutant allele at the protein level is the key step in a poorly understood molecular cascade leading to neurodegeneration. Regardless of the details of this cascade, using RNAi to silence expression of the dominant toxic gene product, which appears to be the proximal disease trigger, would be expected to mitigate disease.

RNAi-mediated allele-specific silencing is achieved by exploiting differences in nucleotide sequence between pathological and nonpathological alleles. The goal is to suppress the toxic allele without altering activity of the normal allele. Biochemical analyses suggest that RISC-mediated cleavage of the bound RNA target takes place immediately opposite nucleotide positions 10 and 11 of the guide strand.^{14, 15} Thus, synthetic siRNAs or vector-encoded short hairpin RNAs (shRNAs) that are designed to selectively target a mutant allele usually place the nucleotide(s) complementary to the mutation at or very near the 10th base-pair position of the dsRNA. Central placement seems to favor selective cleavage of the mutant transcript by RISC.

The strategy outlined above has been used successfully in vitro to selectively silence mutant forms of the Cu, Zn superoxide dismutase (SOD1) in ALS,¹⁶ tau in FTDP-17,¹⁷ APP Swedish variant in AD¹⁷ and Torsin A in DYT1 dystonia.¹⁸ In polyglutamine diseases, where the mutation is an expansion of a normally occurring CAG repeat, attempts to directly target the expanded CAG repeat also suppress the normal allele.¹⁹ What's more, the many mammalian genes that normally contain CAG repeats would also be subjected to unintended suppression by this approach. Thus, allelic discrimination in polyglutamine diseases can probably only be achieved by targeting disease-linked polymorphisms neighboring the repeat rather than the repeat itself.

The polyglutamine disorder Spinocerebellar Ataxia Type 3 (SCA3, also known as Machado–Joseph disease) offers a compelling example of how a linked polymorphism can be exploited in this manner. Miller et al.¹⁹ took advantage of a single-nucleotide polymorphism (G C) that exists in tight linkage disequilibirum with the CAG repeat expansion to silence expression of the mutant disease protein, ataxin-3, without altering levels of normal ataxin-3. About 70% of MJD1 disease alleles have a 'C' at this polymorphism, which sits immediately 3' to the repeat, while most normal alleles have a 'G'.²⁰ Miller and co-workers were able to generate siRNAs that selectively reduced expression of either the mutant or the normal allele, depending on the specific nucleotide present at this site in the complementary guide strand and the targeted mRNA. For mechanistic reasons outlined below, the most effective siRNAs for allele-specific silencing of ataxin-3 placed the SNP at the midpoint of the targeted sequence.

The most common polyglutamine disease, HD, is one in which allele-specific silencing may prove critically important because the HD gene product, huntingtin, is essential for proper brain development and function.²¹ Being an extremely large mRNA, the HD transcript contains numerous potential polymorphisms. In principle, these represent potential target differences between wild type and mutant alleles. To date, however, only one such polymorphism (a three base-pair GAG deletion in exon 58) is known to be in linkage disequilibrium with the pathogenic CAG expansion.²² Although two groups recently established the therapeutic feasibility of RNAi in animal models of HD,^{2, 3} no one has yet reported allele-discriminating RNAi in HD models. Unfortunately, given the relatively low frequency of the exon 58 polymorphism, selectively targeting this region would only benefit a minority of HD patients. Several groups are seeking to identify potential disease-linked polymorphisms encoded in the HD transcript that might expand the pool of targetable mutant HD alleles, in the event that further studies show that allele-specific silencing will be required for HD.

DYT1 dystonia is another dominantly inherited neurological disorder for which allele-specific silencing might be required and actually work quite well. This progressive and devastating disease is caused by a GAG deletion in the gene encoding torsinA.²³ The GAG deletion in the TOR1A gene eliminates one of a pair of glutamic acid residues near the carboxyl-terminus of mutant torsinA. Genetic and biochemical evidence support the idea that mutant torsinA acts through a dominant-negative or toxic mechanism.^{24, 25} Importantly, wild-type torsinA expression is required during ontogeny, suggesting an important, as yet undefined role in cellular function. Therefore, efforts to silence the expression of mutant torsinA have focused on an allele-specific strategy that exploits the three base-pair difference between wild type and mutant forms of torsinA (see note added in proof).¹⁸ By designing siRNAs with the GAG mutation placed near position 10 in the duplex, Gonzalez-Alegre et al.¹⁸ were able to selectively silence mutant torsinA in cells expressing both wild type and mutant torsinA. To bring this strategy a step closer to the clinic, demonstrating the feasibility of this approach in animal models of DYT1 dystonia will be needed. Moreover, even though allele-specific silencing seems ideally suited to DYT1 dystonia, which area of the brain should be targeted remains unclear.

We should emphasize that the concept of selective silencing is not limited to allelic differences. It can also be applied widely as a research tool to silence one splice variant over others, taking advantage of splice junction or exonic sequences that are unique to the splice variant under investigation. In the nervous system, where differential splicing plays a major role,²⁶ isoform-specific RNAi may prove invaluable in defining the function of differentially spliced genes.

Allele-specificity: mechanistic insights

The fact that miRNAs are imperfectly complementary to their target mRNAs yet still silence their expression raises an intriguing question: Why is allele-specific silencing by siRNAs possible? Why isn't expression of the non-targeted, wild-type allele also suppressed? One possible explanation is that genes regulated by miRNAs often contain multiple miRNA binding sites in their 3' untranslated regions.¹³ Thus, the mechanism leading to translational repression by an miRNA may require a higher order number of RISC-miRNA complexes and effector proteins in close association with the target mRNA. This would not be the case for allele-specific silencing occurring through cleavage, where a single site is targeted. Another possibility is that it simply reflects a difference in kinetics: that is, the turnover properties of RISC-mediated cleavage versus translational repression favor the elimination of the targeted mutant transcript. In fact, it remains uncertain whether miRNA-mediated translational repression acts through a catalytic mechanism like siRNA-mediated cleavage.

siRNA-loaded RISC complexes cleave mRNA by targeting the scissile phosphate linking positions 10 and 11 in the targeted strand. Although locating a mismatched sequence centrally in the siRNA tends to favor cleavage of a mutant over a normal allele, siRNAs in some cases retain activity against the normal allele. This suggests that an important contributing factor to single base-pair discrimination is the type of nucleotide mismatch between target mRNA and siRNA at or near the 10th position. A favored model is that purine:purine mismatches disrupt RISC activity by preventing the formation of a conventional A-form helix between the guide strand and the target mRNA, a structural requirement for RISC-mediated cleavage.^{27, 28} In contrast, pyrimidine:pyrimidine mismatches have relatively little effect on cleavage by RISC. Analysis of the siRNA sequences used in several of the previously mentioned allele-specific RNAi studies supports this model. For example, the robust allelic discrimination described in SCA3/MJD was likely due to a G:G purine mismatch that occurs between the guide strand and the normal MJD1 transcript.¹⁹

When mutations lead to pyrimidine:pyrimidine or pyrimidine:purine mismatches, allelic discrimination may not be so robust. In some cases, this problem can be circumvented by adding a second mismatch to the siRNA sequence so that the normal allele differs from the guide strand by two nucleotides rather than only one. Indeed, this was demonstrated by designing siRNAs that preferentially silenced the V337M mutant variant of Tau.¹⁷ Three different siRNAs were tested: one siRNA with the single-nucleotide mutation mismatch centrally placed and two additional siRNAs containing the centrally placed mismatched mutation and a second mismatch either downstream or upstream of the first mismatch. Adding this second mismatch 5' to the centrally placed mismatch in the guide strand resulted in effective silencing of mutant, but not wild-type Tau, in transfected cells.

Although central placement of the mismatch in the siRNA duplex seems important, the efficiency and specificity of silencing also depend on proper strand selection and incorporation into RISC. The process by which siRNAs are loaded into RISC is well understood in Drosophila, where a second Dicer enzyme, Dcr-2, acts with its partner R2D2 to initiate strand selection and siRNA unwinding upon binding to the RISC-associated Ago-2.²⁹ Strand selection is a thermodynamic process in which the more unstable of the two 5' ends of the siRNA duplex is recognized and properly oriented to enter RISC.^{6, 7, 29} In flies, the Dcr-2/R2D2 complex discerns this thermodynamic asymmetry. In mammals, there is only one Dicer and no reports of an R2D2 homolog. Nevertheless, systematic analysis of naturally occurring mammalian mature miRNAs suggests that strand selection is a conserved process that favors unstable 5' ends.⁶ Therefore, successful silencing of a specific allele depends, in part, on designing targeting guide strands with 5'ends that are thermodynamically favored for selection over the 5'ends of passenger strands. Algorithms that aid in the design of these 'asymmetric' duplexes are readily accessible on the internet (e.g. http://sfold.wadsworth.org). New insights on the role that 5'end nucleotides play, both in strand selection and in proper identification of the target mRNA, may allow for further optimization of allele-specific silencing.

In RISC-mediated mRNA cleavage, bases in the 5'-half of the guide strand play a critical role in target recognition and binding affinity.¹⁵ Recent analyses of the interactions between Ago2-bound siRNA and its target mRNA lend support to this idea.^{30, 31} In agreement with bioinformatics and biochemical analysis of endogenous miRNAs, structural data suggest that bases 2–7 of the guide strand are accessible and critically important for base-pairing to the target RNA and, when bound to it, assume a conventional RNA A-form helix. In contrast, position 1 at the 5' end of the guide strand does not contribute to target recognition and the role of nucleotides in the 3' end of the guide strand remains uncertain. Thus, when introducing a second mismatch into an siRNA to favor its activity against a mutant versus a wild-type allele, placing this second mismatch in the 5' half of the guide strand may be preferable. Keep in mind, however, that introducing more than one mismatch into an siRNA will likely reduce its efficacy against the intended target allele. Thus, it is important to screen such doubly mismatched-siRNAs in vitro for efficacy and selectivity before moving to more difficult in vivo applications.

Structural models of Ago2-siRNA-target complexes reveal that the catalytic site is situated a fixed distance from the 5' end of the guide strand.^{30, 31} Therefore, the nucleotide sequence comprising this 5' end will influence the efficiency of an intended allele-specific siRNA because it determines where an engineered mutation mismatch will be placed in relation to the catalytic site. Although central placement of the mismatch is easily accomplished with synthetic siRNAs, this may not always be so straightforward for shRNAs, which must undergo processing by Dicer. Dicer is responsible for specificity in the enzymatic process that generates siRNAs from either exogenous shRNAs or endogenous pre-microRNA precursors. Although much is known about Dicer function, we still do not know precisely how the sequence and structure of a dsRNA precursor molecule affects the position of Dicer cleavage and, accordingly, the 5' end nucleotide composition of an siRNA guide strand.

In a recent series of elegant in vitro experiments, Khvorova and co-workers demonstrated that the type of termini in precursor dsRNA or shRNA molecules influences the specificity and efficiency of Dicer cleavage.³² With precursors having a defined 3' overhang of 3 or fewer nucleotides, Dicer was found to cleave the substrate a fixed distance of 23 nucleotides from the end of the 3' overhang. In contrast, overhangs greater than three nucleotides resulted in a relatively invariant site of cleavage in the complementary strand. Thus, the length and structure of the 3' overhang is a critical determinant of Dicer specificity and efficiency. This process thus dictates the nucleotide composition at the 5' end of the guide strand and, accordingly, the potential catalytic site between positions 10 and 11. Unfortunately, polymerase-III based expression vectors can generate shRNA precursor molecules with 3' ends of variable lengths;³³ consequently, the products of Dicer cleavage will also vary. For allele-specific approaches in which precise placement of the mismatch maybe critical, this variability might pose serious problems. In the light of the results obtained by Khvorova and co-workers, these expression vectors should be modified in order to minimize the production of unwanted Dicer cleavage products that could hinder allele-specific silencing and overall silencing activity. One modification that may increase the efficiency and predictability of siRNA processing is to incorporate designed duplexes into primary-miRNA structures.

Allele-specificity: future directions

A major challenge in bringing RNAi to the clinic will be delivery. The field of gene therapy has made great strides in viral vector technology, liposome-mediated delivery and the development of chemically modified RNA duplexes that can be delivered locally or systemically (reviewed elsewhere in this issue). However, given the limitations inherent to all of these delivery methods, the activity and specificity of therapeutic siRNAs will need to be optimized in order to ensure effective silencing in transduced areas. Continued efforts to better understand the mechanistic details of the initiatior and effector phases of RNAi should help resolve some of the main problems surrounding efficiency and specificity.

New delivery strategies are emerging from studies focusing on the biogenesis of miRNAs.³⁴ For example, Cullen and co-workers^{35, 36} have pioneered the use of primary miRNA-like transcripts to deliver specific, exogenous siRNA molecules. They have demonstrated that embedding siRNAs in pri-miRNA-like structures with long, flanking sequences results in predictable and efficient processing by Drosha.³⁵ The use of such polymerase-II driven miRNA-like constructs to deliver therapeutic siRNAs would ensure the biological processing of precursors by Drosha in the nucleus and Dicer in the cytosol, yielding siRNA molecules of desired size and sequence. Tissue-specific or cell-specific polymerase-II based promoters may also serve to direct the silencing of toxic mutant proteins to certain cells or tissues. The use of polymerase-II based systems also opens the door to pharmacological regulation of siRNA expression (e.g. tetracycline-regulatable promoters). While promising, these second generation siRNA expression vectors still require testing in animal models of disease to determine whether they offer improvements over first generation vectors in efficiency, specificity and safety.

Although not the focus of this review, chemically stabilized siRNAs have also been shown to effectively silence genes in vivo and could be used to achieve allele-specific silencing.^{37, 38, 39} In fact, naked RNA duplexes are currently being used in a clinical setting to treat macular degeneration.⁴⁰ Soutscheck et al.⁴¹ reported efficient in vivo silencing of the apoB gene by siRNA duplexes modified with phosphorothioates and 2'-O-methyl sugars near the 3' of the guide strand and conjugated to cholesterol at the 3'end of the passenger strand. These modifications reduce degradation by nucleases and facilitate entry of the duplex into cells. Most recently, Song et al.⁴² reported the use of siRNA–protamine–antibody complexes that made use of cell-surface receptors to achieve cell-type specific silencing after systemic delivery. Clearly, further studies addressing the long-term efficacy of these siRNA methods for allele-specific silencing in neurological diseases and other conditions (e.g. cancer) are warranted.

Conclusion

Allele-specific silencing of dominant toxic genes can be achieved through RNAi. Gene therapies based on this technique may aid the treatment of many currently untreatable neurological diseases. For some of these devastating disorders, clinical success will most likely depend on the ability to preferentially suppress the expression of the mutant allele. Recent advances in the understanding of RNAi biology should contribute to the development of allele-discriminating siRNAs that act with greater efficiency and specificity. For optimal therapy, future work should include the development and testing of reagents that mirror features of endogenously expressed miRNAs.

Notes

While this article was in press, Gonzalez-Alegre et al. published further advances on allele-specific silencing of torsinA.
Gonzalez-Alegre P, Bode N, Davidson BL, Paulson HL. Silencing primary dystonia: lentiviral-mediated RNA interference therapy for DYT1 dystonia. J Neurosci 2005; 25: 10502–10509.

References

1. Xia H, Mao Q, Eliason EL, Harper SQ, Martins IH, Orr HT et al. RNAi suppresses polyglutamine-induced neurodegeneration in a model of spinocerebellar ataxia. Nat Med 2004; 10: 816–820. | Article | PubMed | ISI | ChemPort |
2. Rodriguez-Lebron E, Denovan-Wright EM, Nash K, Lewin AS, Mandel RJ. Intrastriatal rAAV-mediated delivery of anti-huntingtin shRNAs induces partial reversal of disease progression in R6/1 Huntington's disease transgenic mice. Mol Ther 2005; 12: 618–633. | Article | PubMed | ISI | ChemPort |
3. Harper SQ, Staber PD, He X, Eliason SL, Martins IH, Mao Q et al. RNA interference improves motor and neuropathological abnormalities in a Huntington's disease mouse model. Proc Natl Acad Sci USA 2005; 102: 5820–5825. | Article | PubMed | ChemPort |
4. Raoul C, Abbas-Terki T, Bensadoun JC, Guillot S, Haase G, Szulc J et al. Lentiviral-mediated silencing of SOD1 through RNA interference retards disease onset and progression in a mouse model of ALS. Nat Med 2005; 11: 423–428. | Article | PubMed | ISI |
5. Ralph GS, Radcliffe PA, Day DM, Carthy JM, Leroux MA, Lee DC et al. Silencing mutant SOD1 using RNAi protects against neurodegeneration and extends survival in an ALS model. Nat Med 2005; 11: 429–433. | Article | PubMed | ISI |
6. Khvorova A, Reynolds A, Jayasena SD. Functional siRNAs and miRNAs exhibit strand bias. Cell 2003; 115: 209–216. | Article | PubMed | ISI | ChemPort |
7. Schwarz DS, Hutvagner G, Du T, Xu Z, Aronin N, Zamore PD. Asymmetry in the assembly of the RNAi enzyme complex. Cell 2003; 115: 199–208. | Article | PubMed | ISI | ChemPort |
8. Liu J, Carmell MA, Rivas FV, Marsden CG, Thomson JM, Song JJ et al. Argonaute2 is the catalytic engine of mammalian RNAi. Science 2004; 305: 1437–1441. | Article | PubMed | ISI | ChemPort |
9. Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 1998; 391: 806–811. | Article | PubMed | ISI | ChemPort |
10. Mendell JT. MicroRNAs: critical regulators of development, cellular physiology and malignancy. Cell Cycle 2005; 4: 1179–1184. | PubMed | ISI |
11. Murchison EP, Partridge JF, Tam OH, Cheloufi S, Hannon GJ. Characterization of Dicer-deficient murine embryonic stem cells. Proc Natl Acad Sci USA 2005; 102: 12135–12140. | Article | PubMed | ChemPort |
12. Denli AM, Tops BB, Plasterk RH, Ketting RF, Hannon GJ. Processing of primary microRNAs by the microprocessor complex. Nature 2004; 432: 231–235. | Article | PubMed | ISI | ChemPort |
13. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004; 116: 281–297. | Article | PubMed | ISI | ChemPort |
14. Elbashir SM, Martinez J, Patkaniowska A, Lendeckel W, Tuschl T. Functional anatomy of siRNAs for mediating efficient RNAi in Drosophila melanogaster embryo lysate. EMBO J 2001; 20: 6877–6888. | Article | PubMed | ISI | ChemPort |
15. Haley B, Zamore PD. Kinetic analysis of the RNAi enzyme complex. Nat Struct Mol Biol 2004; 11: 599–606. | Article | PubMed | ISI | ChemPort |
16. Ding H, Schwarz DS, Keene A, Affar el B, Fenton L, Xia X et al. Selective silencing by RNAi of a dominant allele that causes amyotrophic lateral sclerosis. Aging Cell 2003; 2: 209–217. | Article | PubMed | ISI | ChemPort |
17. Miller VM, Gouvion CM, Davidson BL, Paulson HL. Targeting Alzheimer's disease genes with RNA interference: an efficient strategy for silencing mutant alleles. Nucleic Acids Res 2004; 32: 661–668. | Article | PubMed | ISI | ChemPort |
18. Gonzalez-Alegre P, Miller VM, Davidson BL, Paulson HL. Toward therapy for DYT1 dystonia: allele-specific silencing of mutant TorsinA. Ann Neurol 2003; 53: 781–787. | Article | PubMed | ISI | ChemPort |
19. Miller VM, Xia H, Marrs GL, Gouvion CM, Lee G, Davidson BL et al. Allele-specific silencing of dominant disease genes. Proc Natl Acad Sci USA 2003; 100: 7195–7200. | Article | PubMed | ChemPort |
20. Gaspar C, Lopes-Cendes I, Hayes S, Goto J, Arvidsson K, Dias A et al. Ancestral origins of the Machado–Joseph disease mutation: a worldwide haplotype study. Am J Hum Genet 2001; 68: 523–528. | Article | PubMed | ISI | ChemPort |
21. Reiner A, Dragatsis I, Zeitlin S, Goldowitz D. Wild-type huntingtin plays a role in brain development and neuronal survival. Mol Neurobiol 2003; 28: 259–276. | Article | PubMed | ISI | ChemPort |
22. Almqvist E, Spence N, Nichol K, Andrew SE, Vesa J, Peltonen L et al. Ancestral differences in the distribution of the delta 2642 glutamic acid polymorphism is associated with varying CAG repeat lengths on normal chromosomes: insights into the genetic evolution of Huntington disease. Hum Mol Genet 1995; 4: 207–214. | PubMed | ISI | ChemPort |
23. Ozelius LJ, Hewett JW, Page CE, Bressman SB, Kramer PL, Shalish C et al. The early-onset torsion dystonia gene (DYT1) encodes an ATP-binding protein. Nat Genet 1997; 17: 40–48. | Article | PubMed | ISI | ChemPort |
24. Breakefield XO, Kamm C, Hanson PI. TorsinA: movement at many levels. Neuron 2001; 31: 9–12. | Article | PubMed | ISI | ChemPort |
25. Gonzalez-Alegre P, Paulson HL. Aberrant cellular behavior of mutant torsinA implicates nuclear envelope dysfunction in DYT1 dystonia. J Neurosci 2004; 24: 2593–2601. | Article | PubMed | ISI | ChemPort |
26. Lipscombe D. Neuronal proteins custom designed by alternative splicing. Curr Opin Neurobiol 2005; 15: 358–363. | Article | PubMed | ISI | ChemPort |
27. Parker JS, Roe SM, Barford D. Crystal structure of a PIWI protein suggests mechanisms for siRNA recognition and slicer activity. EMBO J 2004; 23: 4727–4737. | Article | PubMed | ISI | ChemPort |
28. Song JJ, Smith SK, Hannon GJ, Joshua-Tor L. Crystal structure of Argonaute and its implications for RISC slicer activity. Science 2004; 305: 1434–1437. | Article | PubMed | ISI | ChemPort |
29. Tomari Y, Matranga C, Haley B, Martinez N, Zamore PD. A protein sensor for siRNA asymmetry. Science 2004; 306: 1377–1380. | Article | PubMed | ISI | ChemPort |
30. Parker JS, Roe SM, Barford D. Structural insights into mRNA recognition from a PIWI domain-siRNA guide complex. Nature 2005; 434: 663–666. | Article | PubMed | ISI | ChemPort |
31. Ma JB, Yuan YR, Meister G, Pei Y, Tuschl T, Patel DJ. Structural basis for 5'-end-specific recognition of guide RNA by the A. fulgidus Piwi protein. Nature 2005; 434: 666–670. | Article | PubMed | ISI | ChemPort |
32. Vermeulen A, Behlen L, Reynolds A, Wolfson A, Marshall WS, Karpilow J et al. The contributions of dsRNA structure to Dicer specificity and efficiency. Rna 2005; 11: 674–682. | Article | PubMed | ISI | ChemPort |
33. Miyagishi M, Taira K. U6 promoter-driven siRNAs with four uridine 3' overhangs efficiently suppress targeted gene expression in mammalian cells. Nat Biotechnol 2002; 20: 497–500. | Article | PubMed | ISI | ChemPort |
34. Cullen BR. Transcription and processing of human microRNA precursors. Mol Cell 2004; 16: 861–865. | Article | PubMed | ISI | ChemPort |
35. Zeng Y, Cullen BR. Efficient processing of primary microRNA hairpins by Drosha requires flanking nonstructured RNA sequences. J Biol Chem 2005; 280: 27595–27603. | Article | PubMed | ISI | ChemPort |
36. Zeng Y, Cai X, Cullen BR. Use of RNA polymerase II to transcribe artificial microRNAs. Methods Enzymol 2005; 392: 371–380. | Article | PubMed | ISI | ChemPort |
37. Morrissey DV, Lockridge JA, Shaw L, Blanchard K, Jensen K, Breen W et al. Potent and persistent in vivo anti-HBV activity of chemically modified siRNAs. Nat Biotechnol 2005; 23: 1002–1007. | Article | PubMed | ISI | ChemPort |
38. Manoharan M. RNA interference and chemically modified small interfering RNAs. Curr Opin Chem Biol 2004; 8: 570–579. | Article | PubMed | ISI | ChemPort |
39. Eckstein F. Small non-coding RNAs as magic bullets. Trends Biochem Sci 2005; 30: 445–452. | Article | PubMed | ISI | ChemPort |
40. Check E. A crucial test. Nat Med 2005; 11: 243–244. | Article | PubMed | ISI | ChemPort |
41. Soutschek J, Akinc A, Bramlage B, Charisse K, Constien R, Donoghue M et al. Therapeutic silencing of an endogenous gene by systemic administration of modified siRNAs. Nature 2004; 432: 173–178. | Article | PubMed | ISI | ChemPort |
42. Song E, Zhu P, Lee SK, Chowdhury D, Kussman S, Dykxhoorn DM et al. Antibody mediated in vivo delivery of small interfering RNAs via cell-surface receptors. Nat Biotechnol 2005; 23: 709–717. | Article | PubMed | ISI | ChemPort |