Dicer-like Enzymes with Sequence Cleavage Preferences

Abstract

Dicer proteins are known to produce small RNAs (sRNAs) from long double-stranded RNA (dsRNA) templates. These sRNAs are bound by Argonaute proteins, which select the guide strand, often with a 5' end sequence bias. However, Dicer proteins have never been shown to have sequence cleavage preferences. In Paramecium development, two classes of sRNAs that are required for DNA elimination are produced by three Dicer-like enzymes: Dcl2, Dcl3, and Dcl5. Through in vitro cleavage assays, we demonstrate that Dcl2 has a strict size preference for 25 nt and a sequence preference for 5' U and 5' AGA, while Dcl3 has a sequence preference for 5' UNG. Dcl5, however, has cleavage preferences for 5' UAG and 3' CUAC/UN, which leads to the production of RNAs precisely matching short excised DNA elements with corresponding end base preferences. Thus, we characterize three Dicer-like enzymes that are involved in Paramecium development and propose a biological role for their sequence-biased cleavage products.

Keywords: Dicer-like enzymes; genome rearrangements; precise DNA elimination; sequence specific cleavage; small RNAs.

Graphical abstract

Figure 1

In Vitro Cleavage Assays with Random Hairpin RNA (A) Four Dicer-like proteins from Paramecium tetraurelia are shown schematically. (B) SDS-PAGE analysis of immunopurified Dcl3 and Dcl4 proteins and purified recombinant Dcl2 and Dcl5 from insect cells. 500 ng of purified proteins and 20 μL beads were loaded on 8% SDS-PAGE gels and stained with either Coomassie or silver. Dcl3-3xFlagHA and Dcl4-3xFlagHA have a size of 88 kDa, Dcl2 has a size of 74 kDa, and Dcl5 has a size of 97 kDa. (C) A schematic of DNA oligo used to generate a hairpin RNA with random nucleotides for in vitro cleavage assays. (D) 15% polyacrylamide-urea gel with small RNAs 5′-labeled with ³²P. M is the small RNA ladder. No enzyme controls are marked as (−). “Dcl3 beads only” is a no template RNA control for Dcl3. (E) Size distribution graphs of Dcl2, Dcl3, and Dcl5 cleavage products based on sRNA sequencing. See also Figure S1.

Figure S1

Analysis of the Bead Contamination and Dcl2 and Dcl3 Results with ND7, Related to Figures 1 and 5 (A) Graphs of the mapping for Dcl3 beads control and Dcl4 beads control. y axis shows number of reads. x axis shows different nucleotides sizes. Different sRNAs mapped against the Paramecium MAC plus IES genome of strain 51. Number of reads mapped to the genome are shown in dark color. Non-mapping reads are shown in bright color. (B) The table shows an overview of 25 nt and 27 nt reads of both bead controls mapping to the genome and how many of those reads are found in the NNN cleavage product reads to assess the background. (C) Graphical representation of the table in (B). (D) Sequence logos for both bead controls (Dcl3 and Dcl4) for either 25 nt reads or 27 nt reads. (E) Sequence logos for 25 nt or 27 nt sRNA cleavage products with the NNN hairpin after extraction of the background reads. (F) Results of the processing assay of Dcl2 and Dcl3 with the ND7 fragment RNA. For Dcl3 processing assay, wild-type lysates (wt) were used as a control. (-) indicates no enzyme controls. Size distribution graphs are shown for Dcl2 and Dcl3 deep sequencing after processing assay with ND7.

Figure 2

Sequence Bias Analysis of Dcl Cleavage Products (A) 5′ end nucleotide enrichment graph for Dcl2 cleavage products showing relative abundance of the first 3 nt. Enrichment factor of 1 (dashed line) represents expected frequency based on the nucleotide composition of the input dsRNA. (B) 3′ end nucleotide enrichment graph for Dcl2 cleavage products. (C) Sequence logos for 25 nt long Dcl2 cleavage products. A typical Dcl2 double-stranded cleavage product is shown schematically with end base signatures and the 2nt-3′ overhangs. (D) 5′ end nucleotide enrichment graph for Dcl3 cleavage products. (E) 3′ end nucleotide enrichment graph for Dcl3 cleavage products. (F) Sequence logos for 27 nt long Dcl3 cleavage products. A typical Dcl3 double-stranded cleavage product is shown schematically. (G–I) Enrichment graphs for Dcl5. Shown for the first three positions (G) and the last five positions (H). Sequence logo shown for 32 nt long sRNAs (I).

Figure 3

In Vitro dsRNA Processing by Dcl5 Using Templates Containing Paramecium MAC and IES Sequences (A) dsRNA templates used for in vitro assays with Dcl5 are shown schematically: a 500 nt long fragment of ND7 coding sequence (black line) and two templates made of three concatenated IESs (blue, purple, and orange). (B) 15% polyacrylamide-urea gel with small RNAs 5′-labeled with ³²P. M is the small RNA ladder. (+) and (−) indicate reactions performed with or without the enzyme. Dashed boxes indicate the bands corresponding to the most abundant products after sRNA sequencing. (C) Size distribution graphs of the cleavage products for Dcl5 with Concatemer and Concatemer Longer templates. (D and E) Enrichment graphs for the 5′ (D) and 3′ (E) ends of Dcl5 cleavage products.

Figure 4

Dcl5 Cleavage of dsRNA Templates with and without Nucleotide Substitutions in IES-IES Junction Sequence (A) dsRNA templates used for processing assays are shown schematically. Blue, purple, and orange represent different IES sequences. Red line indicates IES-IES junction. IES-IES junction sequence is shown in black font. Nucleotide substitutions are shown in red. (B and C) 15% polyacrylamide-urea gel with small RNAs 5′-labeled with ³²P for the cleavage with MAC.IES templates (B) and single mutated MAC.IES templates (C). M is the small RNA ladder. (+) and (−) indicate reactions performed with or without the enzyme. Dashed boxes indicate the bands corresponding to the most abundant products after sRNA sequencing (see Figure S2). (D and E) Mapping of all 27 nt long sRNA reads to the MAC.IES templates (D) and to the single mutated MAC.IES templates (E) used for processing assays. Arrow indicates the sRNA peaks most affected upon changes in the template sequence.

Figure S2

Size Distributions of IES-IES Junction Templates and Mutated Templates, Related to Figure 4 (A–C) Size distribution graphs for all the different templates used for Dcl5 processing assays.

Figure 5

In Vitro dsRNA Processing by Dcl4 (A) Expression profiles of four Dicer-like enzymes according to microarray data. y axis shows the expression level. x axis shows the developmental stages. (B) 15% polyacrylamide-urea gel with small RNAs 5′-labeled with ³²P. M is the small RNA ladder. No enzyme controls are marked as (−). “Beads only” is a no template RNA control for Dcl4. (C) Size distribution graphs of Dcl4 cleavage products based on sRNA sequencing. (D–F) 5′ nucleotide enrichment (D) and 3′ end nucleotide enrichment graph (E) for Dcl4 cleavage products showing relative abundance of the nucleotides. The sequence logos are shown for 25 nt sRNAs (F). A typical Dcl4 double-stranded cleavage product is shown schematically (F). See also Figures S1 and S3.

Figure S3

Alignment of Dcl3 and Dcl4 and Results for Dcl4 Processing Assay with ND7, Related to Figure 5 (A) Protein alignment of all Dcls using the EMBOSS Needle Software. (B) Gel of processing assay of Dcl4 with ND7. M is the marker. (−) is the no enzyme control. (C) Size distribution graph for deep sequencing of Dcl4 processing assay with ND7.

Figure 6

Paramecium Dicer-like Cleavage Preferences and the Role of sRNA End Sequence Bias during DNA Elimination Process (A) A Neighbor-Joining phylogenetic tree of Paramecium Dicer-like enzymes and a summary of their sequence and length preferences. (B) A schematic of iesRNA-guided IES excision. Sequence cleavage preference of Dcl5 leads to the production of iesRNAs, which match preferentially to IES ends, thus enabling their precise excision.