The 3'-to-5' exoribonuclease Nibbler shapes the 3' ends of microRNAs bound to Drosophila Argonaute1

Abstract

Background: MicroRNAs (miRNAs) are ~22 nucleotide (nt) small RNAs that control development, physiology, and pathology in animals and plants. Production of miRNAs involves the sequential processing of primary hairpin-containing RNA polymerase II transcripts by the RNase III enzymes Drosha in the nucleus and Dicer in the cytoplasm. miRNA duplexes then assemble into Argonaute proteins to form the RNA-induced silencing complex (RISC). In mature RISC, a single-stranded miRNA directs the Argonaute protein to bind partially complementary sequences, typically in the 3' untranslated regions of messenger RNAs, repressing their expression.

Results: Here, we show that after loading into Argonaute1 (Ago1), more than a quarter of all Drosophila miRNAs undergo 3' end trimming by the 3'-to-5' exoribonuclease Nibbler (CG9247). Depletion of Nibbler by RNA interference (RNAi) reveals that miRNAs are frequently produced by Dicer-1 as intermediates that are longer than ~22 nt. Trimming of miRNA 3' ends occurs after removal of the miRNA* strand from pre-RISC and may be the final step in RISC assembly, ultimately enhancing target messenger RNA repression. In vivo, depletion of Nibbler by RNAi causes developmental defects.

Conclusions: We provide a molecular explanation for the previously reported heterogeneity of miRNA 3' ends and propose a model in which Nibbler converts miRNAs into isoforms that are compatible with the preferred length of Ago1-bound small RNAs.

Figure 1. miR-34 is trimmed after its production by Dicer-1

(A) Structure of pre-miR-34. Red, miR-34 (24 nt); blue, miR-34* (23 nt). (B) miR-34 isoforms detected in total RNA from S2 cells by Northern hybridization. (C) 52 ³²P-radiolabeled pre-miR-34 was incubated with purified, recombinant Dicer-1/ Loquacious-PB heterodimer (Dcr-1/ Loqs-PB), S2 cell lysate or 0–2 hour embryo lysate. Products were resolved by denaturing polyacrylamide gel electrophoresis. (D) Abundance of miR-34 isoforms detected in fly heads and S2 cells by high throughput sequencing. Only reads with the annotated miR-34 52 ends were analyzed. Black, genome-matching; gray, prefix-matching reads. (E) 52 ³²P-radiolabeled pre-miR-34, 24 nt miR-34 or 21 nt let-7 RNA were incubated in 0–2 h embryo lysate. Products were resolved by denaturing polyacrylamide gel electrophoresis. See also Figure S1.

Figure 2. Trimming of miR-34 requires Ago1 and is limited by the removal of the miRNA* strand

(A) 52 ³²P-radiolabeled pre-miR-34 was incubated in 0–2 h embryo lysate or lysate immunodepleted of Ago1. Products were resolved by denaturing polyacrylamide gel electrophoresis. (B) DsRNA-triggered RNAi targeting Ago1, but not Ago2, decreased trimming of miR-34, compared to treatment with a control dsRNA targeting GFP. “Trimmed” indicates the fraction of all miR-34 corresponding to 21 and 22 nt isoforms. The bantam miRNA and 2S rRNA served as controls. (C) The fraction of long miR-34 isoforms, measured by high throughput sequencing, increased when S2 cells were depleted of Ago1 by RNAi. Only isoforms with the annotated miR-34 52 end were analyzed. The abundance of miR-34 in the two libraries was 3499 ppm (control) and 4506 ppm (ago1 RNAi). (D, E) Synthetic duplexes of 52 ³²P-radiolabeled miR-34 (red) paired to variants of miR-34* (D) were incubated in 0–2 h embryo lysate (E), and the products analyzed by denaturing polyacrylamide gel electrophoresis. (F) Mean ± standard deviation for three independent replicates of the experiment in (E). See also Figure S2.

Figure 3. The 32-to-52 exoribonuclease Knabber (CG9247) trims miR-34, enhancing miR-34 function in S2 cells

(A) S2 cells were transfected with dsRNA against a panel of predicted exonucleases and the effect on miR-34 length analyzed by high resolution Northern hybridization. bantam and 2S rRNA served as controls. The fraction of miR-34 trimmed to 21–22 nt is indicated below each lane. (B) The predicted structure of the knabber (CG9247) gene, mRNA, and protein. (C) S2 cells were transfected with three dsRNAs targeting the second exon or the 32 UTR of knabber as indicated in (B). All four dsRNAs decreased miR-34 trimming, relative to a control dsRNA targeting firefly luciferase. bantam and 2S rRNA served as controls. (D) S2 cells stably expressing wild -type or D435A,E437A mutant Knabber were transfected with dsRNA targeting the 32 UTR of endogenous knabber, and the effect on miR-34 trimming measured. bantam and 2S rRNA served as controls. (E) Reporter construct used in (F). The three miR-34 binding sites pair with miR-34 nucleotides 2–8 and 13–15, mimicking typical animal miRNA binding sites [54]. Rr luc, Renilla reniformis luciferase. (F) Knabber trimming of miR-34 enhances miRNA function. Repression by miR-34 in S2 cells expressing wild-type or D435A,E437A mutant Knabber was measured by blocking miR-34 using a 22-O-methyl-modified anti-miRNA oligonucleotide and measuring the increase in Rr luciferase expression compared to a control oligonucleotide targeting let-7, a miRNA not normally expressed in S2 cells. See also Figure S3.

Figure 4. Knabber trims a quarter of all miRNAs in S2 cells

(A) Analysis of mean miRNA and miRNA* length in S2 cells transfected with dsRNA targeting knabber or a control dsRNA targeting firefly luciferase. miRNA, red; miRNA*, blue; filled circles indicate miRNAs with a significant increase in mean length. In S2 cells, miR-7 (green) does not match our conservative criteria for Knabber substrates, but in flies, miR-7 is trimmed by Knabber (Figure S5C). (B) Knabber trimming explains miRNA 32 heterogeneity. 32 heterogeneity was determined for all S2 cell miRNAs that were more abundant than 200 ppm in high throughput sequencing data. Red, the 11 Knabber substrates identified in this study. Boxplots illustrate 32 heterogeneity of Knabber substrate miRNAs (red) versus all other miRNAs (black). P-value was determined using the Mann-Whitney U test. (C) The mean length of Knabber substrate miRNAs is longer than non-Knabber substrate miRNAs in S2 cells treated with knabber dsRNA. P-value was determined using the Mann-Whitney U test. (D, E) Synthetic miRNA/ miRNA* duplexes comprising a 24 or 22 nt 52 ³²P-radiolabeled miR-305 RNA and the corresponding miRNA* strand (D) were incubated in embryo lysate, and the products analyzed by denaturing polyacrylamide gel electrophoresis (E). See also Figure S4 and Tables S1 and 2.

Figure 5. miRNA trimming requires Knabber in vivo

(A, C, and E) High resolution Northern hybridization of miR-34 from 3–5 day-old male flies. 2S rRNA served as a loading control. (B, D and F) Abundance of miR-34 isoforms in flies carrying a knabber mutant allele (B) or in which Knabber was depleted by RNAi (D and F). miR-34 isoform abundance was measured relative to the indicated controls. (G) A model for miRNA trimming by Knabber. See also Figure S5.