3 Recent Papers on the Significance of siRNA Design
Recently, three articles were published in the journal RNA detailing cutting-edge siRNA research conducted by Dharmacon scientists concerning the importance of siRNA design on target specificity and cell viability. The first article, Reynolds, et al., demonstrates that the delivery of double-stranded RNA (dsRNA) greater than 23 basepairs (bp) can affect not only cell viability but also induce a strong interferon response in a cell type-specific manner. The second article, Fedorov, et al., shows that the knockdown of unintended targets (off-targets) can induce toxic phenotypes. Finally, Jackson, et al., details a position-specific, sequence-independent chemical modification that reduces the silencing of partially complementary transcripts (off-targets) but does not affect the silencing of perfectly matched targets. These three articles highlight the impact Dharmacon research has on advancing the understanding of the RNAi mechanism and ultimately its utility as an indispensable tool for functional gene analysis.
1) A. Reynolds, E. M. Anderson, A. Vermeulen, Y. Fedorov, K. Robinson, D. Leake, J. Karpilow, W. S. Marshall, and A. Khvorova. Induction of the interferon response by siRNA is cell type and duplex length dependent. RNA, 2006 12(6):988-993.
Reynolds, et al. recently delved into the debate over the ability of longer siRNAs to cause an immune response as recent studies suggest that they provide more efficient gene silencing than shorter siRNAs. To investigate the general belief that molecules less than 30 bp in length avoid inducing the interferon pathways, Reynolds, et al. systematically examined the effects of dsRNA lengths on cell viability and immune response in a range of cell lines. First, the effect of 24 different dsRNAs with lengths varying from 17 to 31 bp (plus 2 bp 3’ overhangs on sense strand) on cell viability was studied in HeLa S3 cells. They found that all duplexes greater than 23 bp induced 20-60% cell death, where the acceptable threshold for RNAi experiments is 15% cell death. To determine if this effect was cell-line dependent, four additional cell lines (DU145, HEK293, HeLa, and MCF7) were investigated. The 27 bp dsRNA induced 30-60% cell death in DU145, HeLa S3, and MCF7 cells (defined as sensitive cells) while the same dsRNA induced minimal cell death in HeLa and HEK293 cells (defined as insensitive cells). Conversely, the 19 bp dsRNA did not affect cell viability in any of the cell lines tested. Further microarray analysis on the “sensitive” HeLa S3 cells and “insensitive” HEK293 cells showed a 10-fold difference in the number of genes induced by long dsRNA (27 bp), ~34 IFN-related genes in the insensitive cells compared to greater than 300 in the sensitive cells. A prominent gene associated with the immune response, TLR3, was up-regulated almost 10-fold, strongly correlating with the observed phenotypes in both sensitive and insensitive cell lines transfected with 27 bp dsRNA. Thus, these findings suggest that for RNAi experimentation, great care should be taken when considering dsRNAs of lengths greater than 19 bp for a specific cell line. This work has significant implications with respect to how RNAi will be implemented especially in the development of therapeutic applications.
2) Y. Fedorov, E. M. Anderson, A. Birmingham, A. Reynolds, J. Karpilow, K. Robinson, D. Leake, W. S. Marshall, A. Khvorova. Off-Targeting By siRNA Can Induce Toxic Phenotype. RNA, 2006 12(7) 1188-1196.
Fedorov, et al. investigated whether off-target gene modulation can induce observable phenotypes. A population of 176 randomly selected siRNAs targeting two genes (DBI and firefly luciferase) were evaluated to determine if a relationship existed between cell survival and lipid-mediated siRNA delivery. Using a cell viability cutoff of 75% for toxicity, 51 siRNAs (29%) were identified as affecting cell viability under the conditions tested. Additional studies demonstrated that the toxic phenotype was dependent on siRNA concentration but independent of siRNA activity and observed in all cell lines tested. The apparent toxicity was eliminated when a critical component of the RNAi pathway, eIF2C2 (hAgo2), was disabled, suggesting that the RNAi pathway is necessary for the toxic phenotype. Furthermore, the analysis of the siRNA sequences identified several sequence motifs in common, suggesting that the siRNA toxicity was associated with one or more of these motifs. To test this theory, 289 siRNAs having a predicted RISC-entry strand bias were studied for effects on cell viability. Ninety-two siRNAs caused obvious toxic cell viability while 98 siRNAs clearly exhibited non-toxic cell viability (190 of the original 289 siRNAs). Subsequent analysis of these 190 siRNAs identified a single UGGC motif, suggesting a strong correlation between the siRNA-induced toxic phenotype and this motif. Consequently, the incorporation of this specific siRNA attribute into the designing of highly functional and specific siRNA may possibly reduce off-target related phenotypes.
3) A. L. Jackson, J. Burchard, D. Leake, A. Reynolds, J. Schelter, J. Guo, J. M. Johnson, L. Lim, J. Karpilow, K. Nichols, W. S. Marshall, A. Khvorova, and P. S. Linsley. Position-specific Chemical Modification Increases Specificity of siRNA-mediated Gene Silencing. RNA, 2006 12(7): 1197-1205.
Jackson, et al. examined the effect of siRNA chemical modifications on unintended off-target silencing. Specifically, a walk of single or paired 2’-O-methyl modifications on the first twelve nucleotides of the antisense strand was performed to assess if backbone modifications disrupt the stability of partial complementarity between programmed RISC and unintended mRNA targets. Jackson et al. focused on the impact of the modification within the seed region (positions 2-7) of the antisense, or guide strand, and found that modification of a single position or paired bases across positions 1 through 5 reduced the total number of off-targeted genes, while retaining potent on-target silencing. The greatest reduction of off-target events was observed when positions 1 and 2, or position 2 alone, were modified. The effect of the 2’-O-methyl modifications on positions 1 and 2 of the sense strand and on position 2 of the antisense strand was further examined on seven additional siRNAs. All siRNAs tested showed full silencing of the intended target, while reducing the silencing of 80% of off-target genes. Although the modification pattern significantly improved specificity, it did not eliminate all off-target events. Utilizing multiple independent siRNAs to target a gene of interest was proposed as a strategy to discern true knockdown from off-target events. Consequently, this study demonstrates that the position-specific modification of both the siRNA sense and antisense strand confers increased siRNA-specificity in RNAi experiments. This research has significant implications on the ability to produce both potent and specific siRNAs using a design algorithm and the chemical modification.
^ back to top
