The C-terminal dsRNA-binding domain of Drosophila Dicer-2 is crucial for efficient and high-fidelity production of siRNA and loading of siRNA to Argonaute2

Abstract

Drosophila Dicer-2 efficiently and precisely produces 21-nucleotide (nt) siRNAs from long double-stranded RNA (dsRNA) substrates and loads these siRNAs onto the effector protein Argonaute2 for RNA silencing. The functional roles of each domain of the multidomain Dicer-2 enzyme in the production and loading of siRNAs are not fully understood. Here we characterized Dicer-2 mutants lacking either the N-terminal helicase domain or the C-terminal dsRNA-binding domain (CdsRBD) (ΔHelicase and ΔCdsRBD, respectively) in vivo and in vitro. We found that ΔCdsRBD Dicer-2 produces siRNAs with lowered efficiency and length fidelity, producing a smaller ratio of 21-nt siRNAs and higher ratios of 20- and 22-nt siRNAs in vivo and in vitro. We also found that ΔCdsRBD Dicer-2 cannot load siRNA duplexes to Argonaute2 in vitro. Consistent with these findings, we found that ΔCdsRBD Dicer-2 causes partial loss of RNA silencing activity in vivo. Thus, Dicer-2 CdsRBD is crucial for the efficiency and length fidelity in siRNA production and for siRNA loading. Together with our previously published findings, we propose that CdsRBD binds the proximal body region of a long dsRNA substrate whose 5'-monophosphate end is anchored by the phosphate-binding pocket in the PAZ domain. CdsRBD aligns the RNA to the RNA cleavage active site in the RNase III domain for efficient and high-fidelity siRNA production. This study reveals multifunctions of Dicer-2 CdsRBD and sheds light on the molecular mechanism by which Dicer-2 produces 21-nt siRNAs with a high efficiency and fidelity for efficient RNA silencing.

Keywords: Dicer; RNA silencing; dsRNA; fidelity; siRNA.

FIGURE 1.

Domain structures of Drosophila Dicer-2. (A) Domain structures of the wild-type and truncated mutant Drosophila Dicer-2 used in this study. (B) Anti-HA Western blot detecting the transgenic HA-Dicer-2 proteins in whole transgenic flies.

FIGURE 2.

N-terminal helicase domain and C-terminal dsRBD of Dicer-2 are important for efficient RNA silencing in vivo. (A) Images of the eyes of the female wild-type and mutant Dicer-2 rescue transgenic flies in the background of the endogenous dicer-2^null or dicer-2^G31R in the presence of white inverted repeat (wIR). (Top) wIR; dicer-2^null, Act5C-Gal4/dicer-2^null; UAST-HA-Dicer-2 (wild-type or mutant)/+. (Bottom) wIR; dicer-2^null, Act5C-Gal4/dicer-2^G31R; UAST-HA-Dicer-2 (wild-type or mutant)/+. Control flies are wIR; dicer-2^null, Act5C-Gal4/CyO; +. (B) Measurement of eye pigment chemically extracted from hand-dissected fly eyes. Of note, 480-nm absorbance normalized to mean of CantonS samples was shown. Data are mean ± SD (n = 5). (*) P-value <0.05. (C) Levels of white mRNA normalized by rp49 in the flies with dicer-2^null or dicer-2^G31R background relative to the mean of the control flies, determined by qRT-PCR. Data are mean ± SD (n = 3). (*) P-value <0.05. Data for the control, dicer-2^null, and wild-type rescue are from Kandasamy and Fukunaga (2016).

FIGURE 3.

C-terminal dsRBD of Dicer-2 is important for high-fidelity 21-nt siRNA production in vivo. Sequencing results of small RNAs prepared from female fly heads are shown. The reads were normalized by the sequencing depth. (A) Normalized number of reads of 19- to 23-nt wIR-derived siRNAs. (B) Abundance and 5′ position of wIR-derived siRNAs. Antisense siRNAs are shown in red, sense in blue. Graphs on the right have different y-axis scales. (C) Length distribution of wIR-derived siRNAs. Data for the control, dicer-2^null, wild-type rescue, phosphate-binding pocket mutant are from Kandasamy and Fukunaga (2016).

FIGURE 4.

C-terminal dsRBD of Dicer-2 is important for high-fidelity 21-nt siRNA production by purified recombinant proteins in vitro. (A) Silver-staining of the purified recombinant proteins (28 ng) run on a SDS–PAGE gel. (B) In vitro dicing assay using the purified recombinant proteins. As long dsRNA, body-labeled 104-bp dsRNAs with 2-nt 3′ overhang and either 5′ monophosphate or 5′ hydroxyl were tested. As short dsRNA, 5′ labeled 30-bp dsRNAs with 2-nt 3′ overhang and either 5′ monophosphate or 5′ hydroxyl were tested. The other end of the 30-bp dsRNAs was blocked by two deoxynucleotides (Cenik et al. 2011). Representative gel images (left panels) and quantification of the signals (right panels) are shown. Data are mean ± SD for three independent trials. (C) Length distribution of siRNAs produced from 104-bp dsRNA with 2-nt 3′ overhang and 5′ monophosphate by recombinant Dicer-2 proteins in test tube revealed by high-throughput sequencing. Data for the wild-type and phosphate-binding pocket mutant are from Kandasamy and Fukunaga (2016).

FIGURE 5.

N-terminal helicase domain, but not C-terminal dsRBD, of Dicer-2 is crucial for interaction with R2D2. HA-tagged transgenic Dicer-2 proteins in ovary lysate were immunoprecipitated by anti-HA beads and the copurified proteins were examined by Western blotting. αTubulin and Vasa served as loading controls and were not coimmunoprecipitated with HA-Dicer-2 proteins. The results were reproduced with three biologically replicated samples and representative images are shown.

FIGURE 6.

C-terminal dsRBD, but not N-terminal helicase domain, of Dicer-2 is important for siRNA production in ovary lysate in vitro. (A) Western blotting of the ovary lysates used in the in vitro assays in Figures 6 and 7. Anti-HA antibody detects HA-tagged transgenic Dicer-2 proteins, while anti-Dicer-2 antibody detects the transgenic proteins and the endogenous wild-type Dicer-2. αTubulin and Vasa served as loading controls. (B,C) In vitro dicing assay using the fly ovary lysates. (B) Body-labeled 104-bp dsRNA with 2-nt 3′ overhang and 5′ monophosphate and (C) 5′ labeled 30-bp dsRNAs with 2-nt 3′ overhang and 5′ monophosphate were tested. The other end of the 30-bp dsRNA was blocked by two deoxynucleotides (Cenik et al. 2011). Representative gel images (left panels) and quantification of the signals (right panels). Plotted until the 60-min time points (since at 120 min, significant nonspecific degradation of the substrates were observed) are shown. Data are mean ± SD for three independent trials.

FIGURE 7.

Both N-terminal helicase domain and C-terminal dsRBD of Dicer-2 are important for siRNA loading to Argonaute2 in ovary lysate in vitro. (A) Representative gel image of in vitro siRNA loading assay using ovary lysate. (B) Quantification of the signals in the gels normalized to the background signals in the ovary lysate of dicer-2^null without any rescue gene (negative control). Data are mean ± SD for four independent trials. (*) P-value <0.05.

FIGURE 8.

Models for efficient and high-fidelity siRNA production by Dicer-2. The phosphate-binding pocket in the Dicer-2 PAZ domain anchors the 5′ monophosphate of long dsRNA. Then C-terminal dsRBD binds the proximal body region of the long dsRNA and aligns the RNA to the RNase III active sites. The distance between the anchoring phosphate-binding pocket and RNase III active sites correspond to 21-nt length, allowing high-fidelity 21-nt siRNA production. In the absence of C-terminal dsRBD, dsRNA is misaligned, resulting in lower-efficiency and lower-fidelity production of siRNAs. In the absence of the phosphate-binding pocket, the RNA terminal end is not properly anchored, resulting in lower-fidelity production of siRNAs. The models align well with the previous structural model of Dicer-2 (Lau et al. 2012).