|Home | About | Journals | Submit | Contact Us | Français|
We have previously reported that the RNA-binding protein Dull slicer 16 (DUS16) plays a key role in the processing of primary miRNAs (pri-miRNAs) in the unicellular green alga Chlamydomonas reinhardtii. In the present report, we elaborate on the interaction of DUS16 with Dicer-like 3 (DCL3) during pri-miRNA processing. Comprehensive analyses of small RNA libraries derived from mutant and wild-type algal strains allowed the de novo prediction of 35 pri-miRNA genes, including 9 previously unknown ones. The pri-miRNAs dependent on DUS16 for processing largely overlapped with those dependent on DCL3. Our findings suggest that DUS16 and DCL3 work cooperatively, presumably as components of a microprocessor complex, in the processing of the majority of pri-miRNAs in C. reinhardtii.
MicroRNAs (miRNAs) are loaded into Argonaute (AGO) proteins during the formation of the RNA-induced silencing complex (RISC).1 The main function of miRNAs in RNA silencing is guiding RISC to target transcripts for inducing endonucleolytic RNA cleavage and/or translational repression. In general, miRNAs are embedded in long primary miRNA (pri-miRNA) transcripts containing stem-loop structures and have to be processed to mature miRNAs with the assistance of RNase III Dicer and associated RNA-binding proteins.2,3 We have recently reported that in the unicellular green alga Chlamydomonas reinhardtii, an RNA-binding protein, Dull slicer 16 (DUS16), is required for pri-miRNA processing and associates with Dicer-like 3 (DCL3), which in turn is involved in the biogenesis of the majority of miRNAs (Fig. 1).4,5 We also reported that AGO3, which is one of the 3 AGOs encoded in the C. reinhardtii genome, predominantly binds to mature miRNAs and determines miRNA-mediated post-transcriptional gene silencing (Fig. 1).6 The present report contains a comprehensive analysis of our previously published small RNA-seq (sRNA-seq) data [from the AGO3 mutant (ago3–1); the DUS16 mutant (dus16–1); the parental strain of these mutants Gluc(1×), which expresses a reporter luciferase transgene in the wild-type background; and the wild-type strain CC-124] to predict de novo pri-miRNAs and gain insight into the functional coupling between DUS16 and DCL3.
From the sRNA-seq raw data of CC-124, Gluc(1×), ago3–1, and dus16–1, adaptor sequences were removed and reads ranging from 17 to 25 nucleotides in length were selected for further analyses. The alignment of sorted sRNA reads from the Gluc(1×) sRNA library to the C. reinhardtii genome (Ch_genome_v5.0) using miRA,7 an miRNA discovery tool for plants and algae, led to the identification of 1,062 inverted repeat loci encoding stem-loop RNAs. To stringently screen for genuine pri-miRNA genes, sRNA sequences with <10 read counts were excluded from the libraries, and the remaining redundant sRNA reads were aligned with C. reinhardtii gene models encompassing the inverted repeats using CLC genomic workbench (QIAGEN, https://www.qiagenbioinformatics.com/products/clc-genomics-workbench/). Gene models with <90 mapped-sRNA read counts in the sRNA libraries of CC-124 and Gluc(1×) and/or those without a predominant sRNA species on an arm of the predicted stem-loop structure were discarded. Based on the above workflow, 35 gene models were annotated as pri-miRNA genes, including 9 previously unknown ones (Table 1).
A comparison of total sRNA read counts, mapped on the predicted pri-miRNA genes, from dus16–1 and Gluc(1×) revealed that the production of mature sRNAs from 33 of the 35 pri-miRNAs is significantly lower in dus16–1, suggesting that these pri-miRNAs are mainly processed in a DUS16-dependent manner (Table 1, Fig. S1). Twenty-four of the 35 identified miRNA genes were previously annotated as pri-miRNAs by Valli et al. and are predominantly processed by DCL3 (annotated as “high confidence,” “medium confidence” and/or “upregulated” in Table 1, Fig. S1).5 Furthermore, 22 of these 24 pri-miRNAs (91%) appear to require DUS16 for processing (Table 1; Fig. S1). This result suggests that, in addition to our previous finding of DUS16 physically interacting with DCL3,4 DUS16 is functionally coupled to DCL3, presumably as part of a microprocessor complex involved in the processing of the majority of C. reinhardtii pri-miRNAs.
On the other hand, 2 pri-miRNA transcripts corresponding to Cre04.g217925 and Cre06.g274550, which give rise to mature miR-1144 and miR-1162, respectively, are processed in a DCL3-dependent and DUS16-independent manner (Table 1). In the ago3–1 mutant, the number of mature sRNAs generated from these pri-miRNAs is very low, indicating that most likely, they are authentic pri-miRNAs (Table 1, Fig. 2, Fig. S1). Some sRNAs are also produced from the transcripts of inverted repeats in a DCL3-independent manner.5 These results imply the presence of minor DUS16- and/or DCL3-independent pri-miRNA-processing pathways in C. reinhardtii.
C. reinhardtii appears to possess canonical miRNA biogenesis pathways and miRNA-mediated post-transcriptional gene regulation with certain similarities to those in animals and plants8,9,10,11,12 Mutant analyses revealed that the initial processing of the majority of pri-miRNAs relies on a putative microprocessor complex comprising both DUS16 and DCL3.4,5 In addition, our analyses also uncovered a minor set of pri-miRNAs that are likely processed in a DUS16 and/or DCL3-independent manner.
Small RNA-seq raw data has been deposited in the DDBJ sequence read archive (DRA) under accession numbers DRA003930 and DRA004107 (CC-124 replicate #1, DRX040414; CC-124 replicate #2, CCDRX040415; Gluc1(×) replicate #1, DRX040416; Gluc1(×) replicate #2, DRX040417; ago3–1 replicate#1, DRR045098; ago3–1 replicate#2 DRR045099; dus16–1 replicate #1, DRX043778; and dus16–1 replicate #2, DRX043779).
No potential conflicts of interest were disclosed.
We would like to thank the Functional Genomics Facility, NIBB Core Research Facilities for technical support.
This work was supported by NIBB Collaborative Research Program 15–103 (to T.Y.), JSPS Grant-in-Aid for Young Scientists (B) 16K18480 (to T.Y.), and a grant from the National Science Foundation (to H.C.).