Limits...
Weak base pairing in both seed and 3' regions reduces RNAi off-targets and enhances si/shRNA designs.

Gu S, Zhang Y, Jin L, Huang Y, Zhang F, Bassik MC, Kampmann M, Kay MA - Nucleic Acids Res. (2014)

Bottom Line: The reduced off-targeting was confirmed by RNA-Seq analyses from mouse liver RNAs expressing various anti-HCV shRNAs.Compared with previously established work, the new algorithm was more effective in reducing off-targeting without jeopardizing on-target potency.These studies provide new rules that should significantly improve on siRNA/shRNA design.

View Article: PubMed Central - PubMed

Affiliation: Departments of Pediatrics and Genetics, Stanford University, Stanford, CA 94305, USA.

Show MeSH
AT-enrichment in both seed and 3′ regions is an optimal design to reduce miRNA-like off-target effect. (A) On-target efficacy of shRNAs with no, one, two or three GC pair(s) in the central region was measured by dual-luciferase reporter assay in HEK293 cells. shRNAs tested had all-AU seed and 3′ region sequences. The guide strand and target were perfectly matched in all cases. Sequences used in the central region of guide strand are labeled on the x-axis. RL-luciferase activities were normalized with FF-luciferase, and the percentage of relative enzyme activity (dark bar) compared to the negative control (treated with sh-scramble, gray bar) was plotted. Error bars represent the SD from two independent experiments, each performed in triplicate transfections. *P (t-test, two tailed) < 0.0001 compared with sh-scramble control treatment. (B) Validation of the new design by dual-luciferase assay in HEK293 cells. As illustrated in the figure, a perfectly matched target was used to measure the on-target effect while a central-mismatched target was used to capture the miRNA-like off-target effects. All tested shRNAs have a GC-enriched central region and an AU-enriched seed and 3′ region. Sequences used in the seed and 3′ regions were indicated in the shRNA name. Symbol before the underline represents the seed sequence. Results were plotted as described above. (C) AU-enriched siRNAs are as potent as regular siRNAs (si-W_N1) with respect to on-target knock-down. Each siRNA was transfected with psi-CHECK2 vector containing one perfectly matched target site in the 3′ UTR. Every siRNA was tested at various concentrations in HEK293 cells. RL-luciferase activities were normalized with FF-luciferase, and the percentage of relative enzyme activity compared to the si-scramble negative control was plotted against the final concentration of siRNAs on a log scale. Error bars represent the SD from two independent experiments, each performed in triplicate transfections. (D) The off-target effects of siRNAs tested in (C) were measured by being co-transfected with psi-CHECK2 vector containing one central-mismatched target in the 3′ UTR in HEK293 cells. All siRNAs were tested at a final concentration of 30 nM, which is five times of the highest concentration used in (C). The result was plotted as described above.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4231738&req=5

Figure 2: AT-enrichment in both seed and 3′ regions is an optimal design to reduce miRNA-like off-target effect. (A) On-target efficacy of shRNAs with no, one, two or three GC pair(s) in the central region was measured by dual-luciferase reporter assay in HEK293 cells. shRNAs tested had all-AU seed and 3′ region sequences. The guide strand and target were perfectly matched in all cases. Sequences used in the central region of guide strand are labeled on the x-axis. RL-luciferase activities were normalized with FF-luciferase, and the percentage of relative enzyme activity (dark bar) compared to the negative control (treated with sh-scramble, gray bar) was plotted. Error bars represent the SD from two independent experiments, each performed in triplicate transfections. *P (t-test, two tailed) < 0.0001 compared with sh-scramble control treatment. (B) Validation of the new design by dual-luciferase assay in HEK293 cells. As illustrated in the figure, a perfectly matched target was used to measure the on-target effect while a central-mismatched target was used to capture the miRNA-like off-target effects. All tested shRNAs have a GC-enriched central region and an AU-enriched seed and 3′ region. Sequences used in the seed and 3′ regions were indicated in the shRNA name. Symbol before the underline represents the seed sequence. Results were plotted as described above. (C) AU-enriched siRNAs are as potent as regular siRNAs (si-W_N1) with respect to on-target knock-down. Each siRNA was transfected with psi-CHECK2 vector containing one perfectly matched target site in the 3′ UTR. Every siRNA was tested at various concentrations in HEK293 cells. RL-luciferase activities were normalized with FF-luciferase, and the percentage of relative enzyme activity compared to the si-scramble negative control was plotted against the final concentration of siRNAs on a log scale. Error bars represent the SD from two independent experiments, each performed in triplicate transfections. (D) The off-target effects of siRNAs tested in (C) were measured by being co-transfected with psi-CHECK2 vector containing one central-mismatched target in the 3′ UTR in HEK293 cells. All siRNAs were tested at a final concentration of 30 nM, which is five times of the highest concentration used in (C). The result was plotted as described above.

Mentions: We next sought to reduce binding stability by maximizing AU content beyond the seed region. Indeed, off-target effects were almost undetectable when five shRNAs with extensive AU sequences were tested (Supplementary Figure S2A). However, the on-target knock-down efficiencies were correspondingly reduced (Supplementary Figure S2A). In vitro biochemical studies indicated that base paring in the central region was critical for on-target cleavage but less important for RISC–target association (7). Consistent with this idea, we found that on-target knock-down efficacy was positively correlated with the GC content in the central region when the rest of the guide strand was AU-rich (Figure 2A and Supplementary Figure S2B). Taken together, these results indicate that AU enrichment in both the seed and 3′ regions, but not the central region, could be an OD scheme. Indeed, five shRNAs designed using these criteria were shown to have potent on-target activity while having minimal off-target effects in both HEK293 cells (Figure 2B) and MEF cells (Supplementary Figure S2C).


Weak base pairing in both seed and 3' regions reduces RNAi off-targets and enhances si/shRNA designs.

Gu S, Zhang Y, Jin L, Huang Y, Zhang F, Bassik MC, Kampmann M, Kay MA - Nucleic Acids Res. (2014)

AT-enrichment in both seed and 3′ regions is an optimal design to reduce miRNA-like off-target effect. (A) On-target efficacy of shRNAs with no, one, two or three GC pair(s) in the central region was measured by dual-luciferase reporter assay in HEK293 cells. shRNAs tested had all-AU seed and 3′ region sequences. The guide strand and target were perfectly matched in all cases. Sequences used in the central region of guide strand are labeled on the x-axis. RL-luciferase activities were normalized with FF-luciferase, and the percentage of relative enzyme activity (dark bar) compared to the negative control (treated with sh-scramble, gray bar) was plotted. Error bars represent the SD from two independent experiments, each performed in triplicate transfections. *P (t-test, two tailed) < 0.0001 compared with sh-scramble control treatment. (B) Validation of the new design by dual-luciferase assay in HEK293 cells. As illustrated in the figure, a perfectly matched target was used to measure the on-target effect while a central-mismatched target was used to capture the miRNA-like off-target effects. All tested shRNAs have a GC-enriched central region and an AU-enriched seed and 3′ region. Sequences used in the seed and 3′ regions were indicated in the shRNA name. Symbol before the underline represents the seed sequence. Results were plotted as described above. (C) AU-enriched siRNAs are as potent as regular siRNAs (si-W_N1) with respect to on-target knock-down. Each siRNA was transfected with psi-CHECK2 vector containing one perfectly matched target site in the 3′ UTR. Every siRNA was tested at various concentrations in HEK293 cells. RL-luciferase activities were normalized with FF-luciferase, and the percentage of relative enzyme activity compared to the si-scramble negative control was plotted against the final concentration of siRNAs on a log scale. Error bars represent the SD from two independent experiments, each performed in triplicate transfections. (D) The off-target effects of siRNAs tested in (C) were measured by being co-transfected with psi-CHECK2 vector containing one central-mismatched target in the 3′ UTR in HEK293 cells. All siRNAs were tested at a final concentration of 30 nM, which is five times of the highest concentration used in (C). The result was plotted as described above.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4231738&req=5

Figure 2: AT-enrichment in both seed and 3′ regions is an optimal design to reduce miRNA-like off-target effect. (A) On-target efficacy of shRNAs with no, one, two or three GC pair(s) in the central region was measured by dual-luciferase reporter assay in HEK293 cells. shRNAs tested had all-AU seed and 3′ region sequences. The guide strand and target were perfectly matched in all cases. Sequences used in the central region of guide strand are labeled on the x-axis. RL-luciferase activities were normalized with FF-luciferase, and the percentage of relative enzyme activity (dark bar) compared to the negative control (treated with sh-scramble, gray bar) was plotted. Error bars represent the SD from two independent experiments, each performed in triplicate transfections. *P (t-test, two tailed) < 0.0001 compared with sh-scramble control treatment. (B) Validation of the new design by dual-luciferase assay in HEK293 cells. As illustrated in the figure, a perfectly matched target was used to measure the on-target effect while a central-mismatched target was used to capture the miRNA-like off-target effects. All tested shRNAs have a GC-enriched central region and an AU-enriched seed and 3′ region. Sequences used in the seed and 3′ regions were indicated in the shRNA name. Symbol before the underline represents the seed sequence. Results were plotted as described above. (C) AU-enriched siRNAs are as potent as regular siRNAs (si-W_N1) with respect to on-target knock-down. Each siRNA was transfected with psi-CHECK2 vector containing one perfectly matched target site in the 3′ UTR. Every siRNA was tested at various concentrations in HEK293 cells. RL-luciferase activities were normalized with FF-luciferase, and the percentage of relative enzyme activity compared to the si-scramble negative control was plotted against the final concentration of siRNAs on a log scale. Error bars represent the SD from two independent experiments, each performed in triplicate transfections. (D) The off-target effects of siRNAs tested in (C) were measured by being co-transfected with psi-CHECK2 vector containing one central-mismatched target in the 3′ UTR in HEK293 cells. All siRNAs were tested at a final concentration of 30 nM, which is five times of the highest concentration used in (C). The result was plotted as described above.
Mentions: We next sought to reduce binding stability by maximizing AU content beyond the seed region. Indeed, off-target effects were almost undetectable when five shRNAs with extensive AU sequences were tested (Supplementary Figure S2A). However, the on-target knock-down efficiencies were correspondingly reduced (Supplementary Figure S2A). In vitro biochemical studies indicated that base paring in the central region was critical for on-target cleavage but less important for RISC–target association (7). Consistent with this idea, we found that on-target knock-down efficacy was positively correlated with the GC content in the central region when the rest of the guide strand was AU-rich (Figure 2A and Supplementary Figure S2B). Taken together, these results indicate that AU enrichment in both the seed and 3′ regions, but not the central region, could be an OD scheme. Indeed, five shRNAs designed using these criteria were shown to have potent on-target activity while having minimal off-target effects in both HEK293 cells (Figure 2B) and MEF cells (Supplementary Figure S2C).

Bottom Line: The reduced off-targeting was confirmed by RNA-Seq analyses from mouse liver RNAs expressing various anti-HCV shRNAs.Compared with previously established work, the new algorithm was more effective in reducing off-targeting without jeopardizing on-target potency.These studies provide new rules that should significantly improve on siRNA/shRNA design.

View Article: PubMed Central - PubMed

Affiliation: Departments of Pediatrics and Genetics, Stanford University, Stanford, CA 94305, USA.

Show MeSH