Designing siRNA that distinguish between genes that differ by a single nucleotide

Small interfering RNAs (siRNAs), the guides that direct RNA interference (RNAi), provide a powerful tool to reduce the expression of a single gene in human cells. Ideally, dominant, gain-of-function human diseases could be treated using siRNAs that specifically silence the mutant disease allele, while leaving expression of the wild-type allele unperturbed. Previous reports suggest that siRNAs can be designed with single nucleotide specificity, but no rational basis for the design of siRNAs with single nucleotide discrimination has been proposed. We systematically identified siRNAs that discriminate between the wild-type and mutant alleles of two disease genes: the human Cu, Zn superoxide dismutase (SOD1) gene, which contributes to the progression of hereditary amyotrophic lateral sclerosis through the gain of a toxic property, and the huntingtin (HTT) gene, which causes Huntington disease when its CAG-repeat region expands beyond approximately 35 repeats. Using cell-free RNAi reactions in Drosophila embryo lysate and reporter assays and microarray analysis of off-target effects in cultured human cells, we identified positions within an siRNA that are most sensitive to mismatches. We also show that purine:purine mismatches imbue an siRNA with greater discriminatory power than other types of base mismatches. siRNAs in which either a G:U wobble or a mismatch is located in the “seed” sequence, the specialized siRNA guide region responsible for target binding, displayed lower levels of selectivity than those in which the mismatch was located 3′ to the seed; this region of an siRNA is critical for target cleavage but not siRNA binding. Our data suggest that siRNAs can be designed to discriminate between the wild-type and mutant alleles of many genes that differ by just a single nucleotide.     

Combinatorial delivery of small interfering RNAs reduces RNAi efficacy by selective incorporation into RISC

Despite the great potential of RNAi, ectopic expression of shRNA or siRNAs holds the inherent risk of competition for critical RNAi components, thus altering the regulatory functions of some cellular microRNAs. In addition, specific siRNA sequences can potentially hinder incorporation of other siRNAs when used in a combinatorial approach. We show that both synthetic siRNAs and expressed shRNAs compete against each other and with the endogenous microRNAs for transport and for incorporation into the RNA induced silencing complex (RISC). The same siRNA sequences do not display competition when expressed from a microRNA backbone. We also show that TAR RNA binding protein (TRBP) is one of the sensors for selection and incorporation of the guide sequence of interfering RNAs. These findings reveal that combinatorial siRNA approaches can be problematic and have important implications for the methodology of expression and use of therapeutic interfering RNAs.     

Functional and intracellular localization properties of U6 promoter-expressed siRNAs, shRNAs, and chimeric VA1 shRNAs in mammalian cells

RNA polymerase III (Pol III) expression systems for short hairpin RNAs (U6 shRNAs or chimeric VA1 shRNAs) or individually expressed sense/antisense small interfering RNA (siRNA) strands have been used to trigger RNA interference (RNAi) in mammalian cells. Here we show that individually expressed siRNA expression constructs produce 21-nucleotide siRNAs that strongly accumulate as duplex siRNAs in the nucleus of human cells, exerting sequence-specific silencing activity similar to cytoplasmic siRNAs derived from U6 or VA1-expressed hairpin precursors. In contrast, 29-mer siRNAs separately expressed as sense/antisense strands fail to elicit RNAi activity, despite accumulation of these RNAs in the nucleus. Our findings delineate different intracellular accumulation patterns for the three expression strategies and suggest the possibility of a nuclear RNAi pathway that requires 21-mer duplexes.     

siRNA function in RNAi: a chemical modification analysis

Various chemical modifications were created in short-interfering RNAs (siRNAs) to determine the biochemical properties required for RNA interference (RNAi). Remarkably, modifications at the 2'-position of pentose sugars in siRNAs showed the 2'-OHs were not required for RNAi, indicating that RNAi machinery does not require the 2'-OH for recognition of siRNAs and catalytic ribonuclease activity of RNA-induced silencing complexes (RISCs) does not involve the 2'-OH of guide antisense RNA. In addition, 2' modifications predicted to stabilize siRNA increased the persistence of RNAi as compared with wild-type siRNAs. RNAi was also induced with chemical modifications that stabilized interactions between A-U base pairs, demonstrating that these types of modifications may enhance mRNA targeting efficiency in allele-specific RNAi. Modifications altering the structure of the A-form major groove of antisense siRNA-mRNA duplexes abolished RNAi, suggesting that the major groove of these duplexes was required for recognition by activated RISC*. Comparative analysis of the stability and RNAi activities of chemically modified single-stranded antisense RNA and duplex siRNA suggested that some catalytic mechanism(s) other than siRNA stability were linked to RNAi efficiency. Modified or mismatched ribonucleotides incorporated at internal positions in the 5' or 3' half of the siRNA duplex, as defined by the antisense strand, indicated that the integrity of the 5' and not the 3' half of the siRNA structure was important for RNAi, highlighting the asymmetric nature of siRNA recognition for initiation of unwinding. Collectively, this study defines the mechanisms of RNAi in human cells and provides new rules for designing effective and stable siRNAs for RNAi-mediated gene-silencing applications.     

Effect of target secondary structure on RNAi efficiency

RNA interference (RNAi) mediated by small interfering RNAs (siRNAs) or short hairpin RNAs (shRNAs) has become a powerful tool for gene knockdown studies. However, the levels of knockdown vary greatly. Here, we examine the effect of target disruption energy, a novel measure of target accessibility, along with other parameters that may affect RNAi efficiency. Based on target secondary structures predicted by the Sfold program, the target disruption energy represents the free energy cost for local alteration of the target structure to allow target binding by the siRNA guide strand. In analyses of 100 siRNAs and 101 shRNAs targeted to 103 endogenous human genes, we find that the disruption energy is an important determinant of RNAi activity and the asymmetry of siRNA duplex asymmetry is important for facilitating the assembly of the RNA-induced silencing complex (RISC). We estimate that target accessibility and duplex asymmetry can improve the target knockdown level significantly by nearly 40% and 26%, respectively. In the RNAi pathway, RISC assembly precedes target binding by the siRNA guide strand. Thus, our findings suggest that duplex asymmetry has significant upstream effect on RISC assembly and target accessibility has strong downstream effect on target recognition. The results of the analyses suggest criteria for improving the design of siRNAs and shRNAs.     

Increased siRNA duplex stability correlates with reduced off-target and elevated on-target effects

Argonaute (Ago) proteins form the core of RNA-induced silencing complexes (RISCs) and mediate small RNA-guided gene silencing. In RNAi, short interfering RNAs (siRNAs) guide RISCs to complementary target RNAs, leading to cleavage by the endonuclease Ago2. Noncatalytic Ago proteins, however, contribute to RNAi as well but cannot cleave target RNA and often generate off-target effects. Here we show that synthetic siRNA duplexes interact with all Ago proteins, but a functional RISC rapidly assembles only around Ago2. By stabilizing the siRNA duplex, we show that the noncatalytic Ago proteins Ago1, -3, and -4 can be selectively blocked and do not form functional RISCs. In addition, stabilized siRNAs form an Ago2-RISC more efficiently, leading to increased silencing activity. Our data suggest novel parameters for the design of siRNAs with selective activation of the endonuclease Ago2.     

Cleavage of the star strand facilitates assembly of some microRNAs into Ago2-containing silencing complexes in mammals

In animals, microRNAs (miRNAs) and small interfering RNAs (siRNAs) repress expression of protein coding genes by assembling distinct RNA-induced silencing complexes (RISCs). It has previously been shown that passenger-strand cleavage is the predominant mechanism when siRNA duplexes are loaded into Argonaute2 (Ago2)-containing RISC, while an unwinding bypass mechanism is favored for miRNA duplexes with mismatches. Here I present experimental data indicating that some mammalian miRNAs are assembled into Ago2-containing RISC by cleaving their corresponding miRNA star strands. This phenomenon may depend on the secondary structure near the scissile phosphate of the miRNA duplex. In addition, I show that ATP is not required for star-strand cleavage in this process. Taken together, the data here provide insight into the miRNA-loading mechanisms in mammals.     

Drosophila microRNAs are sorted into functionally distinct argonaute complexes after production by dicer-1

Small interfering RNAs (siRNAs) and microRNAs (miRNAs) guide distinct classes of RNA-induced silencing complexes (RISCs) to repress mRNA expression in biological processes ranging from development to antiviral defense. In Drosophila, separate but conceptually similar endonucleolytic pathways produce siRNAs and miRNAs. Here, we show that despite their distinct biogenesis, double-stranded miRNAs and siRNAs participate in a common sorting step that partitions them into Ago1- or Ago2-containing effector complexes. These distinct complexes silence their target RNAs by different mechanisms. miRNA-loaded Ago2-RISC mediates RNAi, but only Ago1 is able to repress an mRNA with central mismatches in its miRNA-binding sites. Conversely, Ago1 cannot mediate RNAi, because it is an inefficient nuclease whose catalytic rate is limited by the dissociation of its reaction products. Thus, the two members of the Drosophila Ago subclade of Argonaute proteins are functionally specialized, but specific small RNA classes are not restricted to associate with Ago1 or Ago2.     

Conversion of pre-RISC to holo-RISC by Ago2 during assembly of RNAi complexes

In the Drosophila RNA interference (RNAi) pathway, small interfering RNAs (siRNAs) direct Argonaute2 (Ago2), an endonuclease, within the RNA-induced silencing complex (RISC) to cleave complementary mRNA targets. In vitro studies have shown that, for each siRNA duplex, RISC retains only one strand, the guide, and releases the other, the passenger, to form a holo-RISC complex. Here, we have isolated a new Ago2 mutant allele and provide, for the first time, in vivo evidence that endogenous Ago2 slicer activity is important to mount an RNAi response in Drosophila. We demonstrate in vivo that efficient removal of the passenger strand from RISC requires the cleavage activity of Ago2. We have also identified a new intermediate complex in the RISC assembly pathway, pre-RISC, in which Ago2 is stably bound to double-stranded siRNA.     

Potent RNAi by short RNA triggers

RNA interference (RNAi) is a gene-silencing mechanism by which a ribonucleoprotein complex, the RNA-induced silencing complex (RISC) and a double-stranded (ds) short-interfering RNA (siRNA), targets a complementary mRNA for site-specific cleavage and subsequent degradation. While longer dsRNA are endogenously processed into 21- to 24-nucleotide (nt) siRNAs or miRNAs to induce gene silencing, RNAi studies in human cells typically use synthetic 19- to 20-nt siRNA duplexes with 2-nt overhangs at the 3′-end of both strands. Here, we report that systematic synthesis and analysis of siRNAs with deletions at the passenger and/or guide strand revealed a short RNAi trigger, 16-nt siRNA, which induces potent RNAi in human cells. Our results indicate that the minimal requirement for dsRNA to trigger RNAi is an ~42 Å A-form helix with ~1.5 helical turns. The 16-nt siRNA more effectively knocked down mRNA and protein levels than 19-nt siRNA when targeting the endogenous CDK9 gene, suggesting that 16-nt siRNA is a more potent RNAi trigger. In vitro kinetic analysis of RNA-induced silencing complex (RISC) programmed in HeLa cells indicates that 16-nt siRNA has a higher RISC-loading capacity than 19-nt siRNA. These results suggest that RISC assembly and activation during RNAi does not necessarily require a 19-nt duplex siRNA and that 16-nt duplexes can be designed as more potent triggers to induce RNAi.     

Competition potency of siRNA is specified by the 5'-half sequence of the guide strand

Small-interfering RNAs (siRNAs) execute specific cellular gene silencing by exploiting the endogenous RNA interference (RNAi) pathway. Therefore, excess amounts of siRNAs can saturate cellular RNAi machineries. Indeed, some siRNAs saturate the RNA-induced silencing complex (RISC) and competitively inhibit silencing by other siRNAs. However, the molecular feature of siRNAs that specifies competition potency has been undetermined. While previous reports suggested a correlation between the competition potency and silencing efficiency of siRNAs, we found that the silencing efficiency was insufficient to explain the competition potency. Instead, we show that the nucleotide sequence of the 5′-half of the guide strand determines the competition potency of an siRNA. Our finding provides important information for understanding the mechanistic basis of competition in combinatorial RNAi treatment.     

Sequence,chemical, and structural variation of small interfering RNAs and short hairpin RNAs and the effect on mammalian gene silencing

Small interfering RNAs (siRNAs) induce sequence-specific gene silencing in mammalian cells and guide mRNA degradation in the process of RNA interference (RNAi). By targeting endogenous lamin A/C mRNA in human HeLa or mouse SW3T3 cells, we investigated the positional variation of siRNA-mediated gene silencing. We find cell-type-dependent global effects and cell-type-independent positional effects. HeLa cells were about 2-fold more responsive to siRNAs than SW3T3 cells but displayed a very similar pattern of positional variation of lamin A/C silencing. In HeLa cells, 26 of 44 tested standard 21-nucleotide (nt) siRNA duplexes reduced the protein expression by at least 90%, and only 2 duplexes reduced the lamin A/C proteins to <50%. Fluorescent chromophores did not perturb gene silencing when conjugated to the 5'-end or 3'-end of the sense siRNA strand and the 5'-end of the antisense siRNA strand, but conjugation to the 3'-end of the antisense siRNA abolished gene silencing. RNase-protecting phosphorothioate and 2'-fluoropyrimidine RNA backbone modifications of siRNAs did not significantly affect silencing efficiency, although cytotoxic effects were observed when every second phosphate of an siRNA duplex was replaced by phosphorothioate. Synthetic RNA hairpin loops were subsequently evaluated for lamin A/C silencing as a function of stem length and loop composition. As long as the 5'-end of the guide strand coincided with the 5'-end of the hairpin RNA, 19-29 base pair (bp) hairpins effectively silenced lamin A/C, but when the hairpin started with the 5'-end of the sense strand, only 21-29 bp hairpins were highly active.     

Small interfering RNAs containing full 2'-O-methylribonucleotide-modified sense strands display Argonaute2/eIF2C2-dependent activity

RNA interference (RNAi) is a process by which short interfering RNAs (siRNAs) direct the degradation of complementary single-strand RNAs. In this study, we investigated the effects of full-strand phosphorothioate (PS) backbone and 2'-O-methyl (2'-OMe) sugar modifications on RNAi-mediated silencing. In contrast to previous reports, we have identified active siRNA duplexes containing full 2'-OMe-modified sense strands that display comparable activity to the unmodified analog of similar sequence. The structure of these modified siRNAs is the predominant determinant of their activity, with sequence and backbone composition being secondary. We further show, by using biotin-tagged siRNAs and affinity-tagged hAgo2/eIF2C2, that activity of siRNA duplexes containing full 2'-OMe substitutions in the sense strand is mediated by the RNA-induced silencing complex (RISC) and that strand-specific loading (or binding) to hAgo2 may be modulated through selective incorporation of these modifications.     

Efficient allele-specific targeting of LRRK2 R1441 mutations mediated by RNAi

Since RNA interference (RNAi) has the potential to discriminate between single nucleotide changes, there is growing interest in the use of RNAi as a promising therapeutical approach to target dominant disease-associated alleles. Mutations in the leucine-rich repeat kinase 2 (LRRK2) gene have been linked to dominantly inherited Parkinson's disease (PD). We focused on three LRRK2 mutations (R1441G/C and the more prevalent G2109S) hoping to identify shRNAs that would both recognize and efficiently silence the mutated alleles preferentially over the wild-type alleles. Using a luciferase-based reporter system, we identified shRNAs that were able to specifically target the R1441G and R1441C alleles with 80% silencing efficiency. The same shRNAs were able to silence specifically mRNAs encoding either partial or full-length mutant LRRK2 fusion proteins, while having a minimal effect on endogenous wild-type LRRK2 expression when transfected in 293FT cells. Shifting of the mutant recognition site (MRS) from position 11 to other sites (4 and 16, within the 19-mer window of our shRNA design) reduced specificity and overall silencing efficiency. Developing an allele-specific RNAi of G2019S was problematic. Placement of the MRS at position 10 resulted in efficient silencing of reporters (75-80%), but failed to discriminate between mutant and wild-type alleles. Shifting of the MRS to positions 4, 5, 15, 16 increased the specificity of the shRNAs, but reduced the overall silencing efficiency. Consistent with previous reports, these data confirm that MRS placement influences both allele-specificity and silencing strength of shRNAs, while further modification to hairpin design or MRS position may lead to the development of effective G2019S shRNAs. In summary, the effective shRNA against LRRK2 R1441 alleles described herein suggests that RNAi-based therapy of inherited Parkinson's disease is a viable approach towards developing effective therapeutic interventions for this serious neurodegenerative disease.     

Single-Stranded RNAs Use RNAi to Potently and Allele-Selectively Inhibit Mutant Huntingtin Expression

Mutant huntingtin (HTT) protein causes Huntington disease (HD), an incurable neurological disorder. Silencing mutant HTT using nucleic acids would eliminate the root cause of HD. Developing nucleic acid drugs is challenging, and an ideal clinical approach to gene silencing would combine the simplicity of single-stranded antisense oligonucleotides with the efficiency of RNAi. Here, we describe RNAi by single-stranded siRNAs (ss-siRNAs). ss-siRNAs are potent (>100-fold more than unmodified RNA) and allele-selective (>30-fold) inhibitors of mutant HTT expression in cells derived from HD patients. Strategic placement of mismatched bases mimics micro-RNA recognition and optimizes discrimination between mutant and wild-type alleles. ss-siRNAs require Argonaute protein and function through the RNAi pathway. Intraventricular infusion of ss-siRNA produced selective silencing of the mutant HTT allele throughout the brain in a mouse HD model. These data demonstrate that chemically modified ss-siRNAs function through the RNAi pathway and provide allele-selective compounds for clinical development.