Abasic pivot substitution harnesses target specificity of RNA interference

Abstract

Gene silencing via RNA interference inadvertently represses hundreds of off-target transcripts. Because small interfering RNAs (siRNAs) can function as microRNAs, avoiding miRNA-like off-target repression is a major challenge. Functional miRNA-target interactions are known to pre-require transitional nucleation, base pairs from position 2 to the pivot (position 6). Here, by substituting nucleotide in pivot with abasic spacers, which prevent base pairing and alleviate steric hindrance, we eliminate miRNA-like off-target repression while preserving on-target activity at ∼ 80-100%. Specifically, miR-124 containing dSpacer pivot substitution (6pi) loses seed-mediated transcriptome-wide target interactions, repression activity and biological function, whereas other conventional modifications are ineffective. Application of 6pi allows PCSK9 siRNA to efficiently lower plasma cholesterol concentration in vivo, and abolish potentially deleterious off-target phenotypes. The smallest spacer, C3, also shows the same improvement in target specificity. Abasic pivot substitution serves as a general means to harness the specificity of siRNA experiments and therapeutic applications.

Figure 1. Widespread nucleation bulge sites in siRNA off-targets.

(a) Structure of the human Ago–miRNA complex; bases in blue; α-helices of Ago in grey (with indication of I365 and A369); the surface of miR-20a in faint red and Ago in faint green; the possible hydrogen bond between N562 and 2′-OH at nucleotide position 2 is indicated with a dashed-line, and may be abrogated by 2′-OMe. (b) miRNA-like off-targeting of siRNA (for example, an siRNA targeting Renilla luciferase; siRL) caused by seed matches (Seed, left panel) or nucleation bulge sites (Nuc, middle panel) through transitional nucleation (base-pairs from position 2 to 6, red shade, right panel); pivot (position 6) in yellow box; siRL in blue; off-target mRNA in black. (c) Meta-analysis of putative siRNA off-target transcripts that contain miRNA-like target sites (Seed or Nuc) in 3′-UTR. On the basis of compiled microarray data from expression of 35 different siRNAs, cumulative fraction of transcripts depending on fold changes (log₂ ratio, relative to control) was analysed (left panel) and compared with control transcripts (No site; transcripts with neither Seed nor Nuc); P values from KS-test. Heatmap analysis of putative off-target transcripts, which were clustered depending on fold changes at different concentrations (Cons), times (Time) and sequences (Others) of siRNAs targeting Mapk14 (ref. ; right panel). (d) The same analysis as in left panel of c except only for transcripts with significant fold changes (P<0.05) depending on the expression of seven different siRNAs; no modification (left panel) versus 2me (2′-OMe in position 2, right panel). Of note, siRNAs with 2′-OMe still showed off-target repression especially where there was marginal fold repression (right panel, grey inset). The highest P value (2me, Nuc, P=7.4 × 10⁻⁵, KS-test) among all the samples (except for the negative control) was still statistically significant even after Bonferroni correction was applied (n=7, P=1.9 × 10⁻⁴).

Figure 2. Effect of abasic spacer substitution for a nucleotide in the nucleation region of siRNA.

(a) Abasic deoxynucleotide (dSpacer, φ), applied to a nucleotide in siRNA as abasic spacer substitution. (b) dSpacer substitution (pi) in the nucleation region causes a single mismatch to seed sites in off-target transcripts (for example, dSpacer pivot substitution, 6pi), leading to unstable transitional nucleation (for example, siRL-6pi). However, siRNA-6pi may induce a stable interaction only for on-target with a perfect match site through compensatory near-perfect matches (right panel). Details are in Supplementary Fig. 2. Of note, the nomenclature ‘pi' is derived from ‘φ' which here stands for abasic spacer substitution with a deoxynucleotide linker, dSpacer. (c) Luciferase reporter assays for miRNA-like off-target repression, mediated by seed sites for siRL (75 nM) with pi. Relative activity (average Renilla luciferase activity normalized to firefly luciferase) was analysed as a percentage relative to the control ('NT', non-targeting control siRNA); error bars, s.d. WT indicates the unmodified siRNA (red bar). Asterisk denotes P<0.01 (t-test, n=6). RL′ indicates a different sequence of Renilla luciferase gene, that could not be targeted by siRL. (d) The same luciferase assay as in c except for measuring on-target repression (inner set). Repression efficiency was measured at different concentrations of siRL with pi (outer set; 2–6pi, indicated by different colours). IC₅₀ and I_max values are represented in Supplementary Fig. 3c. (e) On-target activity (solid line) was examined together with off-target activity (dotted line) for siRL (red) versus siRL-6pi (blue) by luciferase reporter assays. General siRNA concentrations used for cell culture are indicated (grey, 1–100 nM). (f–g) Effects of dSpacer substitution (pi) in the nucleation region were also examined for siPCSK9-A1 (ref. 36) in f as in c and in g as in d. (h) The same analysis as in e except for siPCSK9-A1; the grey colour indicates the therapeutic concentration. (i) In vitro Ago2 cleavage assays for let-7 (upper panel) and siPCSK9-A1 (lower panel). The triangle denotes expected size of cleaved product from the target substrate (indicated with a line).

Figure 3. Comparison of various abasic spacer conformations within the nucleation region.

(a) A positional effect of rSpacer (abasic ribonucleotide) substitution (pi-r), when applied to a nucleotide within the nucleation region, was examined for seed-mediated off-target repression as in Fig. 2c. siRLs harbouring pi-r within positions 3–6 showed significant derepression. (b) Efficiency of on-target activity of siRL with pi-r (within position 3–6) was examined as in Fig. 2d to compare the results with unmodified siRL (WT, IC₅₀=0.04 nM, red dashed line, as measured in parallel with Supplementary Fig. 3b) and siRL-6pi (IC₅₀=0.62 nM, blue dashed line, as measured in parallel with Fig. 2d). Details on the IC₅₀ values are provided in Supplementary Fig. 4c. (c) Insertion of dSpacer (pi-b) into the nucleation region of siRL that results in bulge formation when siRL anneals to an on-target site (upper panel): examination for seed-mediated off-target repression by luciferase reporter assays as in a (lower panel). siRLs harbouring pi-b within positions 4–6 showed significant derepression. Of note, the nucleation bulge can be caused by the Ago–miRNA structure because the bulge occurs in target RNA, where there is no significant contact with Ago. (d) The same assay as in b except siRL contained pi-b within positions 4–6; ND indicates that IC₅₀ could not be determined. (e) The effect of inserting an rSpacer (pi-rb) into the nucleation region of siRL on seed-mediated off-target repression, when measured as in a. (f) The same experiment as in d except for pi-rb. Of note, every spacer conformation that showed derepression of seed-mediated off-targets had limited potency to induce on-target silencing, in comparison with the superior on-target activity of 6pi. Asterisk denotes P<0.01 (t-test, n=6).

Figure 4. Abasic pivot substitution abolishes seed-mediated miRNA-like target repression.

(a) miRNA-like off-target activity of siPCSK9-A1, mediated by nucleation (Nuc) bulge sites, was examined through luciferase reporter assays in the presence and absence (WT) of 6pi, as in Fig. 2h (n=6); error bars, s.d. Asterisk denotes P<0.01 (t-test). (b) The same experiment as in a, except for siRL. (c–d) Efficiency of seed-mediated target repression by miRNAs was measured in the presence of 6pi, as in a. Elimination of seed-mediated target repression is shown for (c) miR-708-6pi and (d) cel-miR-67-6pi. Of note, because cel-miR-67 is expressed only in C.elegans, it has been widely used as a control in miRNA or siRNA experiments. Nonetheless, cel-miR-67 showed miRNA-like off-target repression. IC₅₀ values of all small RNAs containing 6pi could not be determined (IC₅₀=ND) because there was no significant repression in every range of siRNA concentration (0.05–75 nM). Breaks in y axis were indicated in grey. ND, not determined.

Figure 5. Abasic pivot substitution improves target specificity of siRNA compared with other modifications.

(a) In the absence (WT) and presence of dSpacer pivot substitution (6pi) or 2′-OMe (2me, in position 2), luciferase reporter assays with a perfectly matched site for miR-124 were performed as in Fig. 2e (IC₅₀[WT]=0.05, IC₅₀[6pi]=0.10, IC₅₀[2me]=0.53 nM). (b) The same analysis as in a except for siMAPK14 (IC₅₀[6pi]=0.021 versus IC₅₀[WT]=0.017 nM). (c) Immunoblot analysis of MAPK14 at different concentrations of siRNA transfection (0, 0.5, 5 or 50 nM) relative to control (α-tubulin). (d) Efficiency of on-target activity was estimated as in a, except for siPCSK9-A2 (IC₅₀[6pi]=0.013 versus IC₅₀[WT]=0.008 nM, left panel). Of note, IC₅₀ for siMAPK14 (b) or siPCSK-A2 (d) was approximately estimated owing to its strong repressive activity. siPCSK9-A2 has the same nucleotide sequence as siPCSK9-A1 except it contains 2′-OMe in several positions other than the seed regions to avoid innate immune responses. PCSK9 mRNA levels in HepG2 cells were measured by qPCR after transfection of siPCSK9-A2 (50 nM, relative expression to control (NT) transfection, normalized to GAPDH, right panel). (e) Effect of various modifications (6pi, 2me and 7UNA; UNA modification at position 7) on seed-mediated target repression was measured for miR-124 (IC₅₀[WT]=0.07, IC₅₀[2me]=0.65, and IC₅₀[7UNA]=7.2 nM) as performed in Fig. 4. (f) miR-124-6pi derepresses a validated miR-124 target, PTBP1 (ref. 38), confirmed by immunoblot analysis; γ-tubulin served as a control. Transfected miR-124 and miR-124-6pi was confirmed by northern blotting. ‘RNA' indicates staining of total RNA. (g) The same experiments as in b except for measuring off-target repression mediated by seed sites. (h) Seed-mediated off-target activity of siPCSK9-A2 (IC₅₀=0.91 nM) was examined in the presence of 7UNA (IC₅₀=0.97 nM) or 2bulge (bulge-siRNA, which contains a bulge at position 2 of the guide strand in siRNA duplex; IC₅₀=0.96 nM) as in e. All small RNAs containing 6pi showed depression of miRNA-like targets mediated by seed regions (IC₅₀=ND, 0% repression). ND, not determined. Asterisk denotes P<0.01 (t-test, n=6).

Figure 6. Loss of global target repression and function in miR-124 by abasic pivot substitution.

(a) Transcriptome-wide derepression of miR-124 by dSpacer pivot substitution is represented as a heat map where transcripts are sorted by fold change (log₂ ratio) of miR-124-dependent repression (upper panel). Distribution of the transcripts with a significant fold change depending on miR-124 (red bar) or miR-124-6pi expression (blue bar) was plotted together with all transcripts from control (NT, grey line; lower panel). The dotted box (upper panel) indicates a significant derepression range shown by comparing the distribution of fold changes (sorted fold change <0, upper panel; P<0.01, binomial test, lower panel). All RNA-Seq data from HeLa cells (NT versus miR-124 transfection), which do not express miR-124. (b) Cumulative distribution of transcripts with seed sites in de novo Ago–miR-124 clusters (Ago|Seed) were analysed depending on transcriptome profiles under miR-124 (red) and miR-124-6pi expression (blue), together with all transcripts (total) containing de novo Ago–miR-124 clusters. miR-124-6pi showed no significant difference in the distribution showing repression (fold change <0) relative to total (P=0.30, KS-test), indicating that miR-124-6pi cannot bind and repress target transcripts containing seed sites bound by Ago–miR-124. (c) The same analysis as in b, except for nucleation bulge sites (Nuc). Cumulative fraction analysis of transcriptome profiles under miR-124-6pi expression for Ago|Nuc showed no significant difference (P=0.11, KS-test). (d) miR-124-dependent repression of a known endogenous target mRNA (ITGB1 (ref. 29), normalized to GAPDH) was measured by qPCR; 6pi versus 2me, relative to control (NT). Asterisk denotes P<0.01 (t-test, n=3). (e) The same analysis as in the upper panel of a, except 6pi is compared with 2me in siRL. (f) miR-124-induced neurite outgrowth in N2a cells was prevented by 6pi but not by 2me, likely because of remaining seed-mediated repression activity, detected as in d. See details in Supplementary Fig. 11. Scale bar indicates 500 μm. Time lapse images are also available (miR-124, Supplementary Movie 1; miR-124-6pi, Supplementary Movie 2; miR-124-2me, Supplementary Movie 3).

Figure 7. Abasic pivot substitution eliminates miRNA-like off-target effect in vivo.

(a) In vivo delivery of siRNAs (5 mg kg⁻¹) to the mouse liver was confirmed by measuring the amount of siPCSK9-A2 or siPCSK9-A2-6pi in the liver (qPCR, indicated as log₂ ratio relative to NT, normalized to U6, n=5) after tail vein injection; error bars, s.d. The double asterisk denotes P<0.01 (t-test). (b) The same analysis as in a, except PCSK9 mRNA is quantified to estimate on-target activity (relative amount to NT, normalized to GAPDH, n=5). (c) The concentration of plasma cholesterol, measured by a quantitative colorimetric cholesterol determination assay (n=5). (d) The same transcriptome-wide analysis of off-target repression comparing siPCSK9-A2 with siPCSK9-A2-6pi in the mouse liver, as in the upper panel of Fig. 6a. GO analysis elucidated the enrichment of metal binding function (including ‘copper metabolism') among the off-targets (Supplementary Table 4). (e) The amount of intracellular copper in NCTC clone 1469 cells was significantly increased by siPCSK9-A2 (25±2.2 μg dl⁻¹, P<0.01, relative to NT, t-test, n=3), but not by siPCSK9-6pi (16±5.0 μg dl⁻¹). ‘Cu²⁺' indicates treatment with 32 μM CuSO₄. A single asterisk denotes P<0.05 (t-test). (f) Cell death assays of NCTC clone 1469 cells measured by FACS analysis with propidium iodide (PI) and Annexin V staining. The percentage of cells in the phase of early apoptosis (red box, mean±s.d., n=3) was significantly increased by siPCSK9-A2 but not by siPCSK9-A2-6pi (details in Supplementary Fig. 12d). (g) The same transcriptome-wide analysis of off-targets as in d except on HepG2 cells. (h) Comparison of fold changes between siPCSK9-A2 and siPCSK9-A2-6pi, in a set of off-target transcripts functioning in cell cycle regulation, identified by GO analysis in 6pi-dependent derepressed transcripts (dotted box in g, Supplementary Fig. 13c–f). (i) Cell cycle analysis of HepG2 cells, performed by FACS analysis using PI staining. A defect in cell cycle regulation was observed with siPCSK9-A2, but not with siPCSK9-A2-6pi (n=3).

Figure 8. Models of siRNA containing abasic pivot substitution implicated by the usage of c3 spacer.

(a) A surface model of Ago–miRNA structure with dSpacer pivot substitution (6pi), presented as in Fig. 1a (from 4F3T in PDB). Of note, 6pi reduced steric hindrance by generating space (4.9 Å from I365, 5.1 Å from A369) in the kink between position 6 and 7 (3.6 Å from I365, 3.9 Å from A369, Fig. 1a). Details are provided in Supplementary Fig. 16. (b) A surface model of Ago–miRNA–target structure with 6pi (from 4W5O in PDB). Target mRNA is yellow. (c) The C3 spacer (upper panel) substitution for pivot (6c3) was applied to miR-124 and its effect on repressing perfectly matched sites was analysed by estimating IC₅₀ using luciferase reporter assays, as in Fig. 5a. (d) Effect of 6c3 on seed-mediated target repression was measured for miR-124 by luciferase reporter assays (IC₅₀[6c3]=ND, IC₅₀[WT]=0.07 nM) as in Fig. 5e. ND, not determined. Asterisk denotes P<0.01 (t-test, n=6). (e) miRNA-like off-target sites mediated by siRNA seed regions may be unable to induce stable transitional nucleation owing to abasic pivot substitution, avoiding miRNA-like off-target recognition and repression (upper panel). In contrast, the on-target site that is perfectly complementary to siRNA except for the abasic pivot overcomes unstable transitional nucleation (only four consecutive base pairs in positions 2–5) and the structural kink by 3′ compensatory pairing with help of the reduced steric hindrance (as modelled in b). This event leads to formation of the siRNA–target duplex and on-target repression (lower panel). Shaded ovals represent the Ago protein.