Sorting of Drosophila small silencing RNAs

Abstract

In Drosophila, small interfering RNAs (siRNAs), which direct RNA interference through the Argonaute protein Ago2, are produced by a biogenesis pathway distinct from microRNAs (miRNAs), which regulate endogenous mRNA expression as guides for Ago1. Here, we report that siRNAs and miRNAs are sorted into Ago1 and Ago2 by pathways independent from the processes that produce these two classes of small RNAs. Such small-RNA sorting reflects the structure of the double-stranded assembly intermediates--the miRNA/miRNA( *) and siRNA duplexes--from which Argonaute proteins are loaded. We find that the Dcr-2/R2D2 heterodimer acts as a gatekeeper for the assembly of Ago2 complexes, promoting the incorporation of siRNAs and disfavoring miRNAs as loading substrates for Drosophila Ago2. A separate mechanism acts in parallel to favor miRNA/miRNA( *) duplexes and exclude siRNAs from assembly into Ago1 complexes. Thus, in flies small-RNA duplexes are actively sorted into Argonaute-containing complexes according to their intrinsic structures.

Figure 1. RNA duplex structure determines the partitioning of a small RNA between Drosophila Ago1 and Ago2

(A) A schematic of the distinct small RNA duplexes produced by Dcr-1 processing of pre-miRNAs and Dcr-2 processing of long dsRNA. (B) 254 nm UV crosslinking of exemplary small RNA duplexes. (C) A central mismatch directs the duplex into Ago1 instead of Ago2. The fraction of each duplex crosslinked to Ago2 relative to the sum of RNA crosslinked to Ago1 and to Ago2 is presented as the average ± standard deviation for three independent trials.

Figure 2. The Dcr-2/R2D2 heterodimer, as a component of the Ago-2 loading machinery, promotes assembly of Ago2 RISC and competes with assembly of Ago1 RISC

(A) Sequence of the small RNA duplex (mm11) used in (B) and (C). (B) Dcr-2/R2D2, but not Dcr-2 alone, directs the association of a small RNA duplex with Ago2. 20 nM mm11 duplex, whose let-7 strand partitions between Ago1 and Ago2, was incubated with wild-type lysate supplemented with increasing amounts of Dcr-2/R2D2 or Dcr-2 alone. Ago1- and Ago2-association were measured by 254 nm UV crosslinking. The data (average ± standard deviation for three trials) were normalized to the crosslinking observed in the absence of supplemental recombinant Dcr-2/R2D2 or Dcr-2 alone. (C) 1 nM mm11 duplex was incubated with wild-type lysate supplemented with increasing concentrations of Dcr-2/R2D2, the Ago1-associated siRNA recovered by immunoprecipitation with anti-Ago1 monoclonal antibody and quantified by scintillation counting.

Figure 3. RISC activity coincides with the formation of Dcr-2/R2D2/siRNA ternary complex C1, and a central mismatch in a small RNA duplex impairs the complex formation

(A) Quantification of concentration dependence of the two complexes formed when purified, recombinant Dcr-2/R2D2 heterodimer was incubated with siRNA. The native gel used for this analysis appears in Figure S2A. (B) Experimental strategy for (C) and (D). (C) Target cleavage activity was measured for RISC assembled in dcr-2 mutant lysate—which lacks both Dcr-2 and R2D2—rescued with increasing amounts of recombinant Dcr-2/R2D2 heterodimer. (D) Quantification of (C). The peak of the target cleavage activity corresponds to the peak of complex C1 formation in (A). The y-axis reports the relative concentration of RISC, calculated from a standard curve relating relative RISC concentration to the fraction of target cleaved (Figure S4D) (E) Each of the ten let-7 small RNA duplexes was 5′ ³²P-radiolabled and incubated with 8 nM Dcr-2/R2D2. Then, the fraction of RNA present as complex C1 was measured. No C2 was formed at this concentration of the heterodimer. Bars report the average ± standard deviation for three trials.

Figure 4. The Ago1 loading pathway selects small RNAs with central mismatches, even in the absence of the competing Ago2 pathway

(A) Three exemplary small RNA duplexes were incubated with wild type, dcr-2 or ago2 embryo lysate and then photocrosslinked with shortwave UV to identify small RNA-bound proteins. (B) Kinetic analysis of small RNA association with Ago1, monitored by UV photocrosslinking. The let-7/let-7* duplex associated with Ago1 more rapidly than the mm9 duplex, which was more rapidly bound by Ago1 than the mm1 siRNA duplex. In the absence of the Ago2-loading machinery or Ago2 itself, association of the small RNA duplexes with Ago1 was accelerated, consistent with the idea that the Ago1 and Ago2 pathways compete for loading with small RNA duplexes. Each data point represents the average ± standard deviation for three trials.

Figure 5. let-7/let-7* duplex, but not the mm1 siRNA duplex nor the mm9 duplex, efficiently assembled mature Ago1-RISC

(A) The three exemplary small RNA duplexes were incubated with wild-type or ago2 embryo lysate for 1 h, UV photocrosslinked, and then mature RISC, which contains single-stranded let-7 RNA, separated from pre-RISC, which contains double-stranded RNA, separated using an immobilized 2′-O-methyl let-7 ASO. T, total; S, supernatant (double-stranded); and B, bound (single-stranded). The Ago1-associated let-7 mm1 siRNA duplex and the mm9 duplex remained largely double-stranded, suggesting that mature Ago1-RISC was not efficiently formed from the Ago1 pre-RISC assembled with these duplexes. Most of the Ago1-associated let-7 loaded from the let-7/let-7* duplex was present as single-stranded let-7 bound to Ago1. That is, the conversion of let-7/let-7* Ago1-pre-RISC to let-7 Ago1-RISC was very efficient. In contrast, the mm1 siRNA duplex and mm9 duplex efficiently loaded single-stranded let-7 into Ago2; these small RNA duplexes were efficiently converted from Ago2 pre-RISC to mature Ago2-RISC. (B) Each small RNA duplex was incubated with wild-type embryo lysate, the Ago1-associated RNA recovered by immunoprecipitation, and then the small RNA isolated and single-stranded RNA separated from dsRNA by native gel electrophoresis. As in (A), the let-7 mm1 siRNA duplex and mm9 duplexes produced mainly Ago1-associated double-stranded RNA, whereas the let-7/let-7* duplex yielded almost entirely Ago1-bound single-stranded let-7.

Figure 6. The double-stranded structure of small RNA duplexes generated by dicing longer precursors determines how they are partitioned between Ago1-and Ago2-RISC

Two short hairpin RNAs and pre-let-7 were incubated in embryo lysate for 1 h to generate let-7 by dicing and program RISC, then RISC activity in cleaving a let-7–complementary target RNA (0.5 nM) measured. At left, Ago1 was immunodepleted before adding the target RNA. The red data points therefore report Ago2-RISC activity. At right, the precursors were incubated in ago2⁴¹⁴ mutant lysate, so the red data points represent only Ago1-RISC activity. For the Ago1 experiments, the precursor concentration was 20 nM; for the ago2⁴¹⁴ experiments, it was 100 nM.

Figure 7. A model for small silencing RNA sorting in Drosophila

Dcr-2/R2D2 bind well to highly paired small RNA duplexes but poorly to duplexes bearing central mismatches; such duplexes are therefore disfavored for loading into Ago2. Ago1 favors small RNAs with central mismatches, but no Ago1-loading proteins have yet been identified. Ago1- and Ago2-loading compete each other, increasing the selectivity of small RNA sorting. The partitioning of a small RNA duplex between the Ago1- and Ago2-pathways reflects its structure. A typical miRNA/miRNA* duplex, such as let-7 or bantam, loads mainly Ago1, whereas a standard siRNA duplex loads mostly Ago2. Some miRNA/miRNA* duplexes containing extensively paired central regions, such as miR-277/miR-277* (see Förstemann et al., accompanying manuscript), partition between Ago1 and Ago2. Sorting of small RNA duplexes into Ago1 and Ago2 produces pre-RISC, in which the duplex is bound to the Argonaute protein. Subsequently, mature RISC, which contains only the siRNA guide or miRNA strand of the original duplex, is formed. The separation of the miRNA and miRNA* or the siRNA guide and passenger strands also reflects the structure of the small RNA duplex. For Ago1, we hypothesize that mismatches between the miRNA and the miRNA* or siRNA guide and passenger strands in the seed sequence are required for the efficient conversion of pre-RISC to mature RISC. For Ago2, such seed sequence mismatches are not needed because Ago2 can efficiently cleave the passenger or miRNA* strand, liberating the guide or miRNA from the duplex.