Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Jun 25.
Published in final edited form as: Mol Cell. 2010 Apr 22;38(6):803–814. doi: 10.1016/j.molcel.2010.04.005

Diverse pathways generate microRNA-like RNAs and Dicer-independent small interfering RNAs in fungi

Heng-Chi Lee 1,*, Liande Li 1,*, Weifeng Gu 2,*, Zhihong Xue 1, Susan K Crosthwaite 3, Alexander Pertsemlidis 4, Zachary A Lewis 5, Michael Freitag 6, Eric U Selker 5, Craig C Mello 2, Yi Liu 1,†
PMCID: PMC2902691  NIHMSID: NIHMS195757  PMID: 20417140

Abstract

A variety of small RNAs, including the Dicer-dependent miRNAs and the Dicer-independent Piwi-interacting RNAs, associate with Argonaute family proteins to regulate gene expression in diverse cellular processes. These two species of small RNA have not been found in fungi. Here, by analyzing small RNA associated with the Neurospora Argonaute protein QDE-2, we show that diverse pathways generate miRNA-like small RNAs (milRNAs) and Dicer-independent small interfering RNAs (disiRNAs) in this filamentous fungus. Surprisingly, milRNAs are produced by at least four different mechanisms that use a distinct combination of factors, including Dicers, QDE-2, the exonuclease QIP and an RNAse III domain-containing protein MRPL3. In contrast, disiRNAs originate from loci producing overlapping sense and antisense transcripts, and do not require the known RNAi components for their production. Taken together, these results uncover several pathways for small RNA production in filamentous fungi, shedding light on the diversity and evolutionary origins of eukaryotic small RNAs.

RNA interference (RNAi) is a conserved eukaryotic mechanism in which small RNAs mediate both post-transcriptional and transcriptional gene regulation (Ambros, 2004; Buhler and Moazed, 2007; Ghildiyal and Zamore, 2009; Hannon, 2002). Various types of small RNAs, with sizes ranging from 20–30 nt, guide Argonaute family proteins to RNA targets to regulate expression of diverse genes. Based on whether or not their biogenesis is dependent on the double-stranded RNA (dsRNA) specific RNAse III ribonuclease Dicer, the known eukaryotic small RNAs can be classified into Dicer-dependent and Dicer-independent groups. The Dicer-dependent group of small silencing RNAs includes microRNAs (miRNAs) and various small interfering RNAs (siRNAs), such as exo-siRNAs, endo-siRNAs, and natsiRNAs (Ghildiyal and Zamore, 2009). miRNAs arise from discrete loci and are processed by Dicer-like enzymes from stem-loop RNA precursors (Ambros et al., 2003; Bartel, 2004). Since the discovery of the first miRNA in C. elegans, miRNAs have been found in animals, plants, and algae (Grimson et al., 2008; Lagos-Quintana et al., 2001; Lee and Ambros, 2001; Lee et al., 1993; Llave et al., 2002; Molnar et al., 2007; Zhao et al., 2007), but not in fungi. The absence of fungal miRNAs and the differences between animal and plant miRNAs led to the conclusion that miRNA pathways evolved independently in plants and animals (Jones-Rhoades et al., 2006).

Piwi-interacting RNAs (piRNAs) physically associate with the Piwi clade of Argonaute proteins and are highly enriched in germ line cells (Aravin et al., 2006; Batista et al., 2008; Das et al., 2008; Girard et al., 2006; Grivna et al., 2006; Lau et al., 2006; Vagin et al., 2006). The biogenesis of piRNAs is Dicer-independent. piRNAs originate mostly from repetitive DNA and frequently match only one strand of DNA, suggesting that they arise from single-stranded RNA (ssRNA). Production of piRNA requires Piwi-related proteins, and a “piRNA ping-pong” amplification mechanism was proposed to account for the production of secondary piRNA from primary piRNA via the slicer activity of Piwi-related proteins (Brennecke et al., 2007; Gunawardane et al., 2007). However, such an amplification mechanism does not appear to be involved in piRNA production in C. elegans (Batista et al., 2008; Das et al., 2008). To date, no Dicer-independent small RNAs have been found to associate with Argonaute proteins outside of the animal kingdom.

Previous studies of small RNAs in fungi have been restricted to siRNAs, which are absent from the budding yeast Saccharomyces cerevisiae, but have been identified in the fission yeast Schizosaccharomyces pombe, budding yeast Saccharomyces castellii and Candida albicans, and in filamentous fungi (Buhler and Moazed, 2007; Catalanotto et al., 2006; Drinnenberg et al., 2009). In the filamentous fungus Neurospora crassa, transgene-induced post-transcriptional gene silencing (quelling) requires QDE-1 (QUELLING DEFICIENT-1, an RNA-dependent RNA polymerase), QDE-2 (an Argonaute protein), QDE-3 (a RecQ DNA helicase homologous to the Werner/Bloom Syndrome proteins), and DCL-1/DCL-2 (two partially redundant Dicer proteins) (Catalanotto et al., 2000; Catalanotto et al., 2004; Choudhary et al., 2007; Cogoni and Macino, 1999a, b). The Neurospora Dicers are essential for generating siRNA from dsRNA in vitro and in vivo (Catalanotto et al., 2004).

Neurospora Dicer proteins cleave dsRNA into approximately 25 nucleotide (nt) siRNAs, which are then loaded onto the RNA-induced silencing complex (RISC) containing the core components QDE-2 and QIP (Maiti et al., 2007). We previously showed that QIP, an exonuclease, interacts with QDE-2 and removes the nicked passenger strand from the siRNA duplex, resulting in RISC activation (Maiti et al., 2007). Recently, we identified a class of Neurospora QDE-2-interacting siRNAs (qiRNAs) that are induced after DNA damage (Lee et al., 2009). qiRNAs originate from the highly repetitive rDNA locus and their biogenesis requires QDE-1, QDE-3 and Dicers, but not QDE-2. Here we report that analyses of the Neurospora QDE-2-associated small RNAs uncovered several additional types of small RNAs, including miRNA-like small RNAs (milRs) and dicer-independent small interfering RNAs (disiRNAs), and five different small RNA biogenesis pathways in this filamentous fungus.

Results

Analyses of Neurospora QDE-2 associated small RNAs by deep sequencing

To comprehensively examine small RNA species in N. crassa, we immunoprecipitated c-Myc tagged QDE-2 and isolated its associated RNAs. Myc-QDE-2 specifically associated with small RNAs of around 21 nt in length (data not shown). cDNA libraries of QDE-2-associated small RNAs (17–30 nt) were sequenced using an Illumina/Solexa Genome Analyzer. We obtained 2,180,272 small RNA sequences that match the current Neurospora genome assembly. Most of the small RNAs identified were 19 to 25 nt long, with the peak at 21–22 nt, and had a strong preference (82.2%) for 5’ U (Fig. 1A, 1B and Supplementary Table 1). The relative abundance and average sizes of different classes of small RNAs are shown in Fig. 1C. Although low levels of small RNAs were found throughout the genome (~6/1 kb DNA), sequences corresponding to some genomic regions were highly enriched in our small RNA library, as exemplified by the small RNA distribution across linkage group VII (LG VII) (Fig. 1D). Many of those regions, ranging from a few kb to more than 20 kb, and including sub-telomeric regions, centromeric region and many other disiRNA regions (see below), were represented by nearly equal numbers of sense and antisense small RNAs. A second set of loci was characterized by small RNAs that match only one DNA strand. These loci were considerably smaller (~100–300 nt) and were almost exclusively located in intergenic regions. The different characteristics and distribution of small RNAs suggest the existence of distinct classes of small RNAs in Neurospora.

Figure 1.

Figure 1

Open in a new tab

Characterization of QDE-2-associated small RNAs. (A) Size distribution of QDE-2-associated small RNAs. (B) Nucleotide frequency of the 5’ end of small RNAs. (C) A pie-graph showing the relative abundance of different classes of small RNA. (D) Small RNA distribution on both strands of Linkage Group VII (LGVII). The numbers of small RNAs/10 kb in 1 kb sliding windows are plotted. The locations of subtelomeric and centromeric sequences are indicated.

Potential miRNA-like genes in Neurospora

The strand-specific small RNAs suggest the existence of miRNA-like genes. To identify potential miRNA-like genes, the following criteria were used: First, we identified loci that were highly enriched for small RNA (>240 hits / 1 kb, top 0.3% small RNA-producing loci of the genome). Secondly, we focused on loci that produce small RNA predominantly from one strand (>10 fold difference). Third, we analyzed the predicted ability of these RNAs to form hairpin structures with the flanking sequences, a conserved feature of animal and plant miRNAs. Finally, we checked if the most abundant small RNA species reside in the stem region of the hairpin structure. In this manner, we identified 25 potential miRNA-like loci (Supplementary Table 2). Several individual loci produced multiple, mostly overlapping small RNA species with the same 5’ end. These small RNA producing loci are named milRs for miRNA-like RNA genes. Even though this group of small RNAs and miRNAs from animals and plants share many similarities, they are not named miRNAs and are instead called miRNA-like RNAs (milRNAs) because some of them are produced by similar but not identical mechanisms as conventional miRNAs (see below).

A total of 138,022 small RNAs from milR loci were sequenced. The size distribution of milRNAs was bimodal, peaking at 25 and 19 nt, with very strong preference for U at their 5’ termini (91.3%) (Fig. 2A & 2B). The distinct size distribution of milRNAs compared to that of the total small RNA population suggests the existence of different biogenesis mechanisms in Neurospora.

Figure 2.

Figure 2

Open in a new tab

Characterization of QDE-2-associated milRNAs. (A) Size distribution of the total population of the Neurospora milRNAs. (B) Nucleotide frequency of the 5’ end of milRNAs. (C) Chromosomal and size distributions of small RNAs and predicted structures of pre-milRNAs of the milR-1, milR-2, milR-3 and milR-4 loci. In the chromosomal histograms, the numbers of small RNAs (reads in 10 nt non-overlapping windows), the linkage group and locations of the milRNA* sequences (arrows with *) are indicated. The major peaks in each histogram is the milRNA peak. An RNA more than 20 nts spanned three separate windows, resulting in 3 peaks in the histogram. For each milR, the milRNA sequence and its paired milRNA* sequence (if available) are labeled in red and green, respectively. The position of the 5’ end shared by multiple milRNA species is indicated on the predicted secondary structure by red arrows.

The chromosomal and size distribution of milRNAs, as well as the predicted milRNA precursor (pre-milRNA) structure of the four loci that produce most milRNAs (milR-1, milR-2, milR-3, and milR-4) are illustrated in Figure 2C. All four loci are in intergenic regions and each was represented by at least 9,000 reads. In every case, the vast majority of small RNA sequences correspond to one arm of the hairpin (the milRNA arm) and share a common U at the 5’ends. In addition, small RNAs (milRNA*) matched to complementary arm of the hairpin were also sequenced, but at much lower frequencies. Several milRNA/milRNA* pairs have a 2nt 3’ overhang, suggesting they may be products of a Dicer-like enzyme (Fig. 2C). Although almost all milRNAs from the milRNA arm share the same 5’ U position, they have two or more 3’ends. Similar heterogeneity at 3’ termini has been reported for miRNAs in other eukaryotes (Ruby et al., 2006). Notably, the sizes of the milRNA stem loop precursors are flexible in Neurospora: The milRNA and milRNA* sequences of milR-1, milR-3 and milR-4 are separated by more than 120 nt, whereas they are very close for milR-2. In addition, the size distributions of milRNAs from these four loci showed different profiles: bimodal for milR-1 and -4 and single peaks for milR-2 (25 nt) and milR-3 (19nt). These results suggest that milRNAs from the different milR loci are generated by different mechanisms.

Dicer, QDE-2 and QIP-dependent mechanism for milR-1 milRNAs biogenesis

Under our normal growth conditions, milR-1 is the most prolific small RNA-producing locus in the Neurospora genome. Of >76,000 small RNAs sequenced from this locus, 27,000 were 25 nt long, 23,000 were 24 nt and 7,000 were 19 nt, all starting with a common 5’U. Northern blot analyses of small RNAs from different strains were performed under denaturing or native conditions. A probe specific for the most abundant milRNA species detected five bands in the wild-type strain, with approximate sizes of 19, 24/25, 33, 43, and ~170 nt in the denaturing gels (Fig. 3A, top panel). The 19- and 24/25-nt bands match the size distribution of milRNAs sequenced from milR-1, indicating that they are the mature milRNA products. Interestingly, all four lower bands were lost in the dcl-1/dcl-2 double mutant (dclDKO), while the level of the ~170-nt band increased dramatically. This result indicates that the four smaller RNA species are processed from the long primary-milRNA (pri-milRNA) by Dicer, and that the two intermediate RNAs (33 and 43 nt) are likely the precursor of milRNAs (pre-milRNAs). Consistently, northern blot analysis of total RNA showed that the pri-milRNA was barely detectable in the wild-type strain but was detected at high levels in the dclDKO mutant (Fig. S1A). In addition, the levels of mature milR-1 in the qde-1 and qde-3 mutants were similar to that of wild type, indicating that the RdRP and RecQ helicases are not required for milR-1 milRNA production (Fig. 3A, top). In contrast, the two mature milRNAs species were completely eliminated in the qde-2 mutant, suggesting that this Argonaute protein is required for milR-1 milRNA biogenesis.

Figure 3.

Figure 3

Open in a new tab

Dicers, QDE-2 and QIP are required for the biogenesis of milR-1 milRNAs. (A) Northern blot analyses of small RNA samples showing the levels of milR-1 pri-milRNAs, pre-milRNAs and milRNAs in the indicated strains. The top and middle panel shows Northern Blots probed with the milRNA or milRNA* specific StarFire probes, respectively. The ethidium bromide-stained denaturing gel in the bottom panel shows equal loading of RNA. The sizes of a RNA ladder are shown at left. (B) Northern blot analyses of small RNA samples showing the expression of milR-1 pri-milRNAs, pre-milRNAs and milRNAs. “D” indicates that RNA samples were denatured by boiling. “N” indicates the use of native and undenatured RNA samples. (C) Northern blot analysis of small RNA samples in the qde-2 mutants. (D) Northern blot analysis of small RNA samples showing that QIP and its catalytic activity are required for milR-1 milRNA biogenesis. qip (WT QIP) and qip(QIPH504A) are qip strains that expresses the wild-type QIP or QIP with its catalytic residue mutated. (E) Northern blot analysis showing the association of Myc-QDE-2 with milR-1 mature milRNAs and pre-milRNA. Myc-QDE-2 was immunoprecipitated with c-Myc antibody from the Myc-QDE-2 strain in qde-2 or qip mutant backgrounds.

Because of the small sizes of the pre-milRNAs (33nt and 43 nt), we hypothesize that they are dsRNA with the loop region of the pri-milRNA cleaved. The cloning and sequencing of the pri-milRNA and pre-milRNA species confirmed our prediction (Fig. S1B): First, the pri-milRNA (170nt) is indeed a transcript containing the predicted hairpin structure. Second, two pre-milRNAs are derived from part of the stem region with the loop region removed. Consistent with this conclusion, a probe specific for the milRNA* strand detected three bands in denaturing gel, corresponding to pri-milRNA and the two pre-milRNA* species (33nt and 41nt), but the mature milRNA* was not detected, most likely due to its low level (Fig. 3A, bottom). Furthermore, results of northern blot analyses using a native gel further demonstrated the duplex nature of the pre-milRNA (Fig. 3B). RNA samples denatured by boiling exhibited a milR-1 small RNA profile similar to that shown in Figure 3A, but with native RNA samples, the two pre-milRNA species in the wild-type and qde-2 strains showed reduced gel mobilities, consistent with them being dsRNA. In contrast, mobilities of the mature milRNA were identical under both conditions, indicating that they are single-stranded.

The requirement of QDE-2 for milR-1 milRNAs production prompted us to examine whether the slicer activity of QDE-2 is necessary for milRNA biogenesis. We previously showed that a mutation of QDE-2 in its DDH motif (D664A) completely abolished its slicer activity (Maiti et al., 2007). The production of milRNAs was reduced but, surprisingly, not abolished in the qde-2 (D664A) mutant (Fig. 3C), indicating that the slicer activity of QDE-2 is not required for milRNA generation. Since QDE-2(D664A) retains its ability to bind duplex sRNA (Maiti et al., 2007), this suggests that QDE-2 may bind to pre-milRNA and recruit another factor to processes the pre-milRNAs into mature milRNAs.

We previously showed that the exonuclease QIP interacts with QDE-2 and removes the nicked passenger strand of the siRNA duplex (Maiti et al., 2007). Thus, we examined the role of QIP in milR-1 milRNA biogenesis. The mature milRNA was completely abolished in the qip mutant as in the qde-2 mutant (Fig. 3D). The function of QIP in milRNA processing requires its exonuclease activity since expression of QIP protein with a point-mutation of its catalytic residue (H504A) (Maiti et al., 2007) in the qip mutant did not rescue milR-1 milRNA production (Fig. 3D). In contrast, expression of the wild-type QIP fully restored the mature milRNA profile in the qip mutant. Since QDE-2 still can bind small RNA duplex in the qip mutant (Maiti et al., 2007), this result also suggest that the lack of mature milR-1 milRNAs in the qde-2 mutant is not due to the role of QDE-2 in stabilizing milRNA.

Immunoprecipiation assays showed that MycQDE-2 associated with the short pre-milRNA in the qip mutant, but also bound two mature milRNAs in the wild-type strain (Fig. 3E). These results indicate that the 33nt pre-milRNA is the precursor of the mature milRNAs, whereas the long pre-milRNA may be an intermediate or by-product of pri-milRNA processing.

Based on our results, we propose the following model for milR-1 processing: (1) milR-1 pri-milRNA is first processed by Dicer to generate double-stranded pre-milRNAs without the loop, (2) QDE-2 binds to pre-milRNA and (3) recruits the exonuclease QIP to process the pre-milRNAs into mature milRNAs (Fig. S1C).

Complete or partial Dicer-dependence for milR-3 and milR-4 milRNA biogenesis

We found that Dicer is also important for the generation of milR-3 and milR-4 milRNAs (Fig. 4A). In the dclDKO mutant, both the pre-milRNA and mature milRNA of milR-3 were abolished. Although the level of the mature milRNA of milR-4 was significantly reduced in the dclDKO mutant, surprisingly, it was not eliminated. Unlike milR-1, QDE-2 is not involved in the generation of mature milRNAs for milR-3 and -4. Furthermore, only the mature milRNAs of milR-3 and -4, not their pre-milRNAs, were associated with QDE-2 (Fig. 4B), consistent with their QDE-2-independent biogenesis. Thus, the biogenesis of milR-3 milRNA resembles miRNA maturation in plants. In contrast, the production of milR-4 milRNA is only partially dependent on Dicer, indicating involvement of another nuclease.

Figure 4.

Figure 4

Open in a new tab

Distinct milRNA biogenesis mechanism for milR-2, milR-3 and milR-4. (A) Northern blot analyses of small RNA samples showing that the production of pre-milRNAs and milRNA of milR-3 and milR-4 in different strains. (B) Northern blot analyses of Myc-QDE-2-associated small RNAs. Myc-QDE-2 was immunoprecipitated with c-Myc antibody from the qde-2, Myc-QDE-2 strain. (C) Northern blot analyses of small RNA samples showing that the production of pre-milRNAs and milRNA of milR-2 requires QDE-2 but is independent of Dicers. The top and middle panels show the milR-2 pre-milRNAs and milRNA, respectively. A longer exposure of the milRNA panels is shown. Gels for panels at right were subjected to a longer electrophoresis time to show size differences in milRNAs. (D) Analyses of Myc-QDE-2-associated small RNAs in the wild-type, qde-1 and dclDKO strains. A construct that expresses Myc-QDE-2 was introduced into the indicated strains. In the top panel, the Myc-QDE-2-associated total small RNAs were 3’ end labeled with 32P. In the middle and bottom panels, Northern blot results of the Myc-QDE-2-associated small RNAs show that milR-2 and milR-4 milRNAs are associated with QDE-2 in the dclDKO strain.

A Dicer-independent, QDE-2-dependent mechanism for biogenesis of milR-2 milRNA

The milR-2-specific probe revealed pre-milRNA species of approximately 33, 39, and 52 nt and a mature milRNA of ~25 nt (Fig. 4C). The signals for pre-milRNAs were much stronger than that of the milRNA. Surprisingly, the production of the pre-milRNAs and mature milRNA occurred in the dclDKO mutant, indicating that the biogenesis of milR-2 milRNA is Dicer-independent. In contrast, QDE-2 and its catalytic activity were required for the production of both pre-milRNAs and milRNA, as shown by the disappearance of the two smaller species of pre-milRNAs, and the accumulation of the 52-nt form in the qde-2 and qde-2(D664A) mutants. An RNA band ~2 nt larger than the wild-type mature milR-2 milRNA was observed in the qde-2 mutants. Unlike milR-1, the generation of milR-2 pre-milRNAs and mature milRNAs did not require QIP (data not shown). Furthermore, both its mature milRNA and the two shorter pre-milRNAs were associated with QDE-2 (Fig. 4B). These results suggest that the milR-2 pri-milRNA is processed into the pre-milRNAs by an unidentified nuclease(s). Afterwards, QDE-2 associates with the pre-milRNAs and its slicer activity is involved in generation of mature milRNA.

Because Dicers are known to participate in the process of siRNA loading onto Argonaute proteins in other organisms (Liu et al., 2003), we examined whether the Dicer-independent milRNAs of milR-2 and milR-4 are loaded onto QDE-2 in a dicer-independent manner. We expressed Myc-QDE-2 in wild-type, qde-1, and dclDKO strains and immunoprecipitated Myc-QDE-2. The total levels of Myc-QDE-2 associated small RNA (3’ end-labeled with 32P) were significantly reduced in the qde-1 and dclDKO mutants (Fig. 4D, top panel), showing that most of the small RNAs produced in Neurospora are QDE-1- and Dicer-dependent. Interestingly, both the mature milRNAs of milR-2 and milR-4 were found to be associated with Myc-QDE-2 in the qde-1 and dclDKO mutants (Fig. 4D, lower panels), indicating that loading of these milRNAs onto QDE-2 does not require Dicer.

These results further establish the existence of Dicer-independent processing of small RNAs in Neurospora. Since QIP is known to facilitate the formation of single-stranded siRNA containing-active RISC complex by removing the passenger strand, we tested whether QIP is also responsible for miRNA RISC maturation (Maiti et al., 2007). QDE-2 associated-mature milR-4 remained single-stranded in the qip mutant (Figure S2A), suggesting a qip-independent mechanism for milRNA RISC activation. This result indicates that the activation of siRNA and milRNA RISC are mechanistically distinct processes in Neurospora. Recent reports also suggested that the activation mechanisms of siRNA and miRNA RISC are different (Kawamata et al., 2009; Liu et al., 2009; Maiti et al., 2007).

Because the QDE-2 associated small RNAs were cloned after CIP and PNK treatments, we examined the 5’ phosphate polarity of milRNAs. As shown in Figure S2B, most of the milRNAs could be ligated to a 22nt 5’ adaptor, indicating that they contain 5’ monophosphate.

Taken together, our results reveal the existence of miRNA-like genes, or milRs, in Neurospora. Importantly, these four milRNAs and their precursors are produced by four distinct mechanisms, uncovering surprisingly diverse pathways for small RNA generation in this filamentous fungus.

milRs mediate gene silencing in Neurospora

For most candidate milR loci, the corresponding milRNAs and their predicted secondary structures are conserved in two other Neurospora species: N. discreta and N. tetrasperma (data not shown). For milR-1, the sequences of both the milRNA and milRNA* strands are highly conserved, but nucleotides in the loop region between them are not (Fig. S3A), suggesting functional importance of milR-1 in the genus Neurospora.

To determine whether Neurospora miRNAs can mediate gene silencing, we created a Myc-tagged reporter construct carrying in-frame five milR-1 milRNA complementary sequences in the middle of the open reading frame. This construct was introduced into the wild-type and qde-2 mutant strains. Protein levels of the reporter were ~20 fold higher in the qde-2 mutants than in the wild-type strains (Fig. 5A), indicating that milR-1 expression results in robust silencing of its complementary target. Quantitative RT-PCR analysis showed that the levels of the reporter mRNA in the qde-2 mutant was only modestly higher than that in the wild-type strain (Fig. 5B), suggesting that milR-1 milRNA may mostly mediate gene silencing by repression of translation.

Figure 5.

Figure 5

Open in a new tab

milRs mediate gene silencing in Neurospora. (A) milR-1 milRNA mediates gene silencing of a Myc-tagged reporter gene with milR-1 target sites. The top diagram shows the construct that carries the milR-1 target sequences in the ORF. The middle and bottom panels show Western blot results with c-Myc antibody. Three independent transformants of the wild-type and qde-2 strains were used. Error bars indicate standard deviations in this Figure. (B) qRT-PCR results showing the mRNA levels of the reporter construct in wild-type and qde-2 strains. The error bars indicate standard deviations from independent experiments. (C) Base pairing between milRs and their predicted targets. (D) qRT-PCR results showing the mRNA levels of predicted milRNA targets in the wild-type and dclDKO mutant strains. (E) qRT-PCR analysis of MycQDE-2 immunoprecipitated RNAs showing the association of Myc-QDE-2 with predicted milR targets. Myc-QDE-2 was immunoprecipitated with c-Myc antibody from the Myc-QDE-2 expressing strain. A wild type strain was used as a control.

No perfect matches for any of the major milR miRNA sequences were found elsewhere in the N. crassa genome, which suggests that, like animal miRNAs, milRNAs may target imperfectly complementary sequences. We therefore predicted putative milRNA targets using miRmate, a mammalian miRNA target prediction program (Du et al., 2009; Younger et al., 2009). The mRNA levels of several predicted targets (Fig. 5C) are up-regulated in the dclDKO mutant (Fig. 5D), suggesting they are regulated by milRNAs. In addition, the mRNA levels of these predicted targets were also elevated in the qde-2 mutant (Figure S3).

To demonstrate the specificity of miRNA action on these predicted target genes, we generated a milR-1 knockout strain (milR-1KO), in which the production of milR-1 milRNAs was eliminated (Figure S3B). As expected, the mRNA levels of NCU07311 and NCU09469, two predicted milR-1 target genes, were elevated in the milR-1KO and qde-2 strains (Figure S3C). In contrast, although the mRNA level of NCU05770, a predicted milR-4 target, was up-regulated in the qde-2 strain, it did not change in the milR-1KO strain. These results suggest that the milR-1 milRNA down-regulates the expression of its predicted target genes.

To further test if those mRNA are targets of milRNAs, Myc-QDE-2 was immunoprecipitated and levels of the associated RNAs were measured. Specific association between QDE-2 and these predicted miRNA targets was observed, whereas QDE-2 did not bind to several control RNAs specifically (Fig. 5E). Among these genes, NCU05770 encodes a peroxidase, NCU09469 contains an EF hand and an ubiquitin binding domain, and both NCU07311 and NCU03313 are fungal-specific genes. Except for NCU05770, which has the milRNA target site in its 3’UTR, the other target genes have the milRNA target sites located within their ORFs. Taken together, our results show that milRs regulates gene expression in Neurospora.

MRPL3, an RNAse III domain-containing protein, is involved in Dicer-independent processing of some milRNAs

The evidence of Dicer-independent mechanisms for milRNA generation prompted us to look for Dicer-independent nuclease activity. Because of the similarities between milRNAs and conventional miRNAs, we carried out pre-miRNA processing assays using Neurospora extracts and the Drosophila let-7 pre-miRNA (Jiang et al., 2005). Neurospora extracts from both the wild-type and dclDKO strains exhibited similar miRNA processing activity, indicating Dicer-independent activity (Fig. 6A). Interestingly, the size of the processed let-7 miRNA in the Neurospora extracts was ~1–2 nt larger than that produced by the recombinant Drosophila Dicer-1/Loqs complex (Jiang et al., 2005). In contrast, it was previously shown that the Neurospora Dicers are essential for generating siRNA from dsRNA in vitro and in vivo (Catalanotto et al., 2004).

Figure 6.

Figure 6

Open in a new tab

MRPL3 is important for the biogenesis of milR-1 and milR-4 milRNAs. (A) miRNA processing assay using Drosophila pre-let-7 shows the in vitro miRNA processing activity in the wild-type and dclDKO extracts. (B) A diagram showing the domain structure of MRPL3. (C) Northern blot analysis showing the reduced level of mrpl3 mRNA in the mrpl3KO(het) strain. (D) & (E) Northern blot analyses showing the levels of milRNAs in the indicated strains. dsmrpl3 is a strain in which the expression of mrpl3-specific dsRNA is induced by QA. (F) miRNA processing assay showing the reduced level of in vitro pre-let7 processing activity in the dsmrpl3 strain. (G) In vitro processing assay for milR-4 pre-milRNA. Universally-labeled milR-4 pre-milRNA was incubated with extracts of the indicated strains. The position of the pre-milRNA and two processed products are labeled with arrows.

In an attempt to identify the protein responsible for the observed activity, we searched for putative RNAse III domain-containing proteins. In addition to the two Dicers, one protein (NCU08299) homologous to the yeast mitochondrial ribosomal protein MRPL3 possesses a putative RNAse III domain and a dsRNA recognition motif (Fig. 6B). Although MRPL3 homologs were identified as part of the mitochondrial ribosomal protein large subunit (Smits et al., 2007), their function is unknown. MRPL3 homologs are found in most eukaryotic organisms; MRPL44 is the human homolog. The RNAse III domains of MRPL3 proteins show little sequence similarity to those of Dicers and Drosha, suggesting that they belong to different families of RNAse III domain proteins. mrpl3 is apparently an essential gene in Neurospora, since no homokaryotic deletion strain could be obtained, but a heterokaryotic strain that carries a mixture of wild-type and mrpl3KO nuclei showed reduced mrpl3 mRNA level (Fig. 6C).

Examination of the levels of milRNA by northern blot analyses revealed that the levels of the major mature milR-1 milRNA and the 33nt pre-milRNAs were markedly reduced in the mrpl3KO(het) strain (Fig. 6D), suggesting that MRPL3 may collaborate with Dicer to generate milR-1 pre-milRNAs and milRNAs. For milR-4, whose milRNA production is partially dependent on Dicer, the mature milRNA in the mrpl3 strain was also reduced to a level similar to that in the dclDKO strain. However, the levels of milR-2 and milR-3 milRNAs and pre-milRNAs were not altered in the mrpl3 mutant compared to wild type (data not shown). In addition, milRNA levels for milR-1 and milR-4 were also reduced in a strain (dsmrpl3) in which mrpl3 is silenced by dsRNA (Fig 6E). Furthermore, pre-let-7 processing activity was reduced in the dsmrpl3 strain in vitro, as indicated by the decreased level of let-7 and increased level of pre-let-7 (Fig. 6F). Finally, the pre-milR-4 processing activity was also reduced in the dsmrpl3 strain in vitro (Fig. 6G). Together, these results suggest that MRPL3 may be important for the Dicer-independent biogenesis of some Neurospora milRNAs.

Dicer-independent small interfering RNAs (disiRNAs)

A major group of Neurospora small RNAs are comprised of small RNAs matched symmetrically to both strands of DNA. We have previously shown that the DNA damage-induced qiRNAs originating from both strands of rDNA are dependent on QDE-1, QDE-3, and Dicers. In addition to the rDNA repeats, we identified 50 small RNA-enriched loci in which nearly equal amounts of small RNAs matched both DNA strands (Supplementary Table 3). Unlike rDNA, these loci are not repetitive; rather, they include genes and intergenic regions and have no apparent sequence motifs. Small RNAs resulting from these loci averaged 22 nt in length, with a strong 5’ U preference (94%) (Fig. 7A & B). The small RNAs from these loci are not induced by DNA damage (data not shown), indicating they are not produced by the same mechanism as qiRNAs. We named these regions disiRNA loci because their characteristics are distinct from those of all known types of small RNAs (see below).

Figure 7.

Figure 7

Open in a new tab

Characterization of disiRNAs. (A) Size distribution of disiRNAs. (B) Nucleotide frequency of 5’end of disiRNAs. (C) Distribution of small RNAs at the disiRNA-6 locus. The middle panel shows the small RNA distribution matched to either Watson (red) or Crick (green) strand of DNA. Numbers of small RNAs (logarithmic scale) in 10 nt sliding windows are shown. The bottom panel shows the small RNA hits of both directions in a small representative region of the locus. The red arrows indicate the overlapping sense and antisense ESTs identified. (D) Chromosomal distribution of small RNAs at the disiRNA-36 and disiRNA-37 loci. The red arrows indicate the identified sense and antisense ESTs of qde-2 and overlapping convergent ESTs in the disiRNA-37 locus. (E) Northern blot analyses of small RNA samples showing the levels of different disiRNAs in the indicated strains.

Notably, disiRNAs appear to originate from loci that produce overlapping sense and antisense transcripts. Figures 7C–D show the chromosomal distribution of small RNAs originating from three disiRNA loci. disiRNA-6 is one of the most abundant disiRNA loci with close to 7,000 overlapping sense and antisense disiRNAs covering a 15-kb region. This region contains two predicted Neurospora genes, NCU00911 and NCU00910. The majority of the small RNAs are located between NCU00911 and NCU00912. The overlapping sense and antisense disiRNAs arising from this locus suggest that disiRNAs is derived from dsRNA. Consistent with this notion, overlapping sense and antisense ESTs (red arrows) were found downstream of NCU00911.

The qde-2 gene is one of the disiRNA loci (disiRNA-36); small RNAs cover the entire gene although a much lower levels of small RNAs corresponding to the neighboring gene were also found (Figure 7D). Several ESTs antisense to qde-2 have been found in the Neurospora genome database.

The distribution of small RNAs in the disiRNA-37 locus peak between two convergently transcribed genes in close proximity. As several overlapping sense and antisense ESTs were identified, there may be a role for convergent transcription in disiRNA production. A survey of the available EST sequences complementary to the 50 disiRNA loci showed that out of the 44 loci for which ESTs were found, 32 have overlapping sense and antisense ESTs and another three are likely to have overlapping sense and antisense ESTs. Thus, based on the limited EST data available, nearly 80% of the disiRNA loci produce overlapping sense and antisense transcripts. These results suggest that disiRNAs are likely derived from dsRNA made from naturally occurring complementary sense and antisense transcripts.

We examined expression of disiRNAs from seven loci in different RNAi mutant strains (Fig. 7E). Unlike qiRNAs, disiRNA biogenesis is independent of QDE-1, QDE-2, and QDE-3. Surprisingly, despite evidence that disiRNAs are derived from dsRNA, their levels were not affected in the dclDKO mutant, indicating that Dicers are not involved in generating disiRNAs. In addition, we found that disiRNA levels were not altered in a strain that has both Neurospora Argonaute genes deleted (qde-2 sms-2) or in other known RNAi mutants or mrpl3 mutants (Fig. 7E). These results indicate that, unlike animal piRNAs, an Argonaute-dependent maturation mechanism is not involved in disiRNA production. It was recently shown that in some budding yeast species, a second RNT1 (contains RNase III domain similar to Dicer)-like protein act as the noncanonical Dicer protein to generate siRNA from long hairpin RNAs (Drinnenberg et al., 2009). However, Neurospora has only one gene encoding for an RNT1 homolog and it is essential for cell survival. In addition, its knock-down by dsRNA did not affect disiRNA production (data not shown). Together, these results indicate that the biogenesis of disiRNAs does not require known RNAi components and suggest another Dicer-independent small RNA biogenesis pathway.

Discussion

In this study, we conducted a genome-wide survey of Argonaute-bound small RNA population from Neurospora and identified two groups of small RNAs, milRNAs and disiRNAs. Our discovery of milRs in Neurospora demonstrates that miRNA and miRNA-like small RNAs exist in all major branches of eukaryotic organisms, and our functional studies of milRs suggest that miRNA-like gene regulatory mechanism evolved early in the eukaryotic lineage. In addition, our demonstration of Dicer-independent biogenesis of some Neurospora milRs and disiRNAs suggests that other dsRNA specific nucleases, e.g. other putative RNase III enzymes, participate in the biogenesis of miRNA-like small RNAs. This raises the possibility that miRNA-like RNAs may be produced in eubacteria and archea, which lack homologs of Dicer, but encode Argonaute-like proteins (Song et al., 2004).

Neurospora lacks sequence homologs of Drosha and Pasha, which generate pre-miRNAs from pri-miRNAs in animals. Despite some differences between the Neurospora milRNAs and miRNAs in animal and plants in their biogenesis mechanisms, they share remarkable similarities. First, as miRNAs, the Neurospora milRNAs arise from highly specific stem-loop RNA precursors. Second, most of the milRNAs examined in this study require Dicer for their production. Third, like the animal miRNAs, milRNAs appear to silence endogenous targets with mismatches.

A major finding of this study is the discovery of diverse pathways for small RNA biogenesis. Our examination of four milRs uncovered surprisingly diverse mechanisms for milRNA production: Only the biogenesis of milR-3 milRNAs resembles that in plants, which require only Dicers for pre-milRNA and milRNA processing (Jones-Rhoades et al., 2006). For milR-1, the production milRNAs requires Dicer, QDE-2, QIP and MRPL3. The production of milR-4 milRNAs is partially Dicer-dependent and requires MRPL3. On the other hand, the biogenesis of milR-2 milRNA requires QDE-2 and its catalytic activity but is completely independent of Dicer. The mechanistic diversity observed for milRNA biogenesis in Neurospora stands in stark contrast with the well-defined miRNA biogenesis pathways in animals and plants and suggests that eukaryotic small RNAs generation mechanisms are more diverse than previously thought.

BLAST searches did not identify highly homologous milR genes outside the Neurospora genus, suggesting individual milRs have evolved independently from plant and animal miRNAs. However, mechanisms similar to those used in Neurospora milRNA production may be involved in the biogenesis of some types of small RNAs in animals and plants. For example, Argonaute-dependent maturation of milR-1 and -2 milRNAs is similar to that of animal piRNAs. In addition, we identified a new role of Argonaute in small RNA biogenesis, where QDE-2 functions as an adaptor through binding to pre-milRNA and recruiting its associated QIP exonuclease to produce mature milR-1. Maturation of animal piRNAs may also involve recruitment of a nuclease to trim the 3’ ends of the Piwi-associated piRNA precursors (Ghildiyal and Zamore, 2009).

We identified the putative ribonuclease MRPL3, which has an RNAse III domain, as an important component in milRNA biogenesis. Although the exact biochemical function of MRPL3 in milRNAs biogenesis is not known, its predicted RNAse III domain suggest that it may function as a hairpin-specific RNase that process pri- or pre-milRNAs. The eukaryotic homologs of Neurospora MRPL3 have unknown functions but are present in the large subunit of the mitochondrial ribosomes. Thus, our results suggest a potential link between mitochondrial ribosomal function and milRNA biogenesis.

Our discovery of QDE-2-associated disiRNAs in Neurospora demonstrates that Dicer-independent small RNAs exist in the fungal kingdom. The different characteristics of Neurospora disiRNAs and animal piRNAs suggest that they are two distinct classes of small RNAs. First, disiRNAs originate from non-repetitive parts of the genome and are matched to overlapping sense and antisense strands of DNA. Second, most of the disiRNA-producing loci yield overlapping sense and antisense transcripts, suggesting that disiRNAs arise from dsRNA templates. In this aspect, disiRNAs are similar to endo-siRNAs and natsiRNAs in animals and plants (Ghildiyal and Zamore, 2009). Since Dicers are essential for the generation of siRNA from dsRNA made from long hairpin RNA transcripts in Neurospora (Catalanotto et al., 2004; Choudhary et al., 2007), it is likely that dsRNA formed by sense and antisense transcripts can be recognized by an unidentified ribonuclease in Neurospora. Third, unlike the animal piRNAs, the biogenesis of disiRNAs does not require the Argonaute proteins. Furthermore, their independence of QDE-1 or other RDRPs suggests that disiRNAs are not secondary small RNAs amplified by RdRPs. Taken together, these results indicate that although disiRNAs share similarities to piRNAs and endo-siRNAs, they are produced by a different small RNA biogenesis pathway. Since the submission of this manuscript, a class of Dicer-independent primal RNAs was reported to contribute heterochromatin formation in S. pombe (Halic and Moazed, 2010).

Materials and Methods

Strains and growth conditions

The wild-type strain used in this study was FGSC 4200(a). Mutant strains including qde-1, qde-2, qde-2 (D664A), qde-3, qip, and a dcl-1 dcl-2 double mutant (dclDKO) were generated previously (Choudhary et al., 2007; Maiti et al., 2007). Myc-tagged protein expressing strains, including Myc-QDE-2, Myc-QIP, and Myc-QIP (H504A) strains, were also created previously (Maiti et al., 2007). The homokaryotic milR-1KO strain was generated by replacing the entire milR-1 pri-milRNA with a hph resistance cassette by homologous recombination. mrpl-3KO (heterokaryon, FGSC#18364) was obtained from the Fungal Genetic Stock Center. The dsmrpl3 strain was created by a transforming a wild-type strain with a plasmid that can inducibly express an inverted repeat specific for mrpl3 ORF (Cheng et al., 2005). Liquid cultures were grown in minimal medium (1x Vogel’s, 2% glucose) at room temperature. For liquid cultures containing QA, 0.01 M QA (pH 5.8) was added to the liquid culture medium containing 1xVogel’s, 0.1% glucose, and 0.17% arginine.

Purification and cloning of QDE-2 associated sRNAs

Immunopurification of Myc-QDE-2 ribonucleoprotein complex was performed as previously described (Lee et al., 2009). Small RNAs from the QDE-2 IP were purified on 15% PAGE/7 M urea gels using 18–26 nt synthetic RNA markers, followed by overnight elution with elution buffer containing 10 mM Tris-Cl (pH7.5), 1 mM EDTA, and 0.3 M NaCl, and precipitation with 20 µg glycogen and 1 volume of isopropanol. RNAs were incubated with 1 U/µl Calf Intestinal Phosphatase (NEB) and 1 U/µl SUPERase•In, in Buffer 3 (NEB) at 37°C for 1hr, extracted twice with phenol, and precipitated with ethanol. RNA was incubated with 10 µM of pre-adenylated DNA oligo (AppCTGTAGGCACCATCAAT/ddC/), 1 unit/µl of SUPERase•In, 10% DMSO and 2 unit/µl T4 RNA ligase (Takara Bio Inc.) in 10 µl ligation buffer (50mM Tris-Cl pH7.5, 10mM MgCl2, 0.06µg/ul BSA, 10mM DTT). The 3'-ligated-RNAs were phosphorylated with 1 unit/µl Polynucleotide Kinase in buffer 3 containing 1 unit/µl SUPERase•In and 2 mM ATP at 37°C for 1 hr, extracted with phenol, precipitated with ethanol, and 5'-ligated with 2 unit/µl T4 RNA ligase and 30 µM 5' linker GTT CTA CAG TCC GAC GAT C in 10 µl 1X buffer containing 1 unit/µl SUPERase•In, 0.1 µg/µl BSA and 10% DMSO. cDNA was synthesized using SuperScript III with RT oligo ATT GAT GGT GCC TAC AG. The cDNA was PCR-amplified with oligos containing Illumina/Solexa linkers.

Small RNA analyses

All small RNA analyses were performed with custom Perl (5.8.6) scripts on a chromosome genome assembly based on N. crassa assembly 7 (Lewis et al., 2009) (available upon request from M. F.), rDNA sequence (Lee et al., 2009), and Version 3 gene predictions for the N. crassa assembly 7 release including mitochondria sequence (Broad Institute). Annotations for genes, structural RNAs and repeats in Version 3 gene predictions were mapped and converted to sequence coordinates in the chromosome genome assembly. Solexa reads containing a perfect match to the first 6 nt of the 3' linker were used to extract the inserted sequences. Inserts 17–30 nt long were used to identify perfect matches to the genome. To calculate small RNA reads derived from a single genomic locus, the small RNA reads were normalized using its total number of matched genomic loci (repeat-normalized reads). The density of small RNA was presented in two ways: using a non-overlapping 10 nt sliding window along the Watson or Crick strand of each chromosome (Fig. 2C); and using a 10,000 nt window moving 1000 nt each step along the Watson or Crick strand of each chromosome (Fig. 1D).

Identification of milRs and disiRNA loci

To identify milRs and disiRNAs, structural RNAs such as tRNA or rRNA were first excluded. The following criteria were used to identify milR loci: 1. Small RNAs were highly enriched at the loci for a density over 240 reads/kbp, which resides in the top 0.3% most enriched loci of the genome. The median loci in the genome contain ~6 hits/kbp. 2. Small RNAs were predominantly derived from one strand (at least a 10 fold difference between two DNA strands). 3. Small RNAs form a stem loop structure (hairpin) with flanking sequences, and the most abundant species of the small RNA resides in the stem region. Secondary structures of small RNAs and their flanking sequences (up to 200 nt) were examined by mfold.

disiRNA loci were identified as regions where small RNAs are derived from both strands with similar abundance and the small RNA density is significantly than the background (>70 hits/kb, which consists of the top 1 % loci in the genome).

RACE and sequencing of pre-milR-1 and pri-milR-1

Small RNAs with sizes of 30–60nt and 100–200nt were fractioned from wild type and dicerDKO mutant samples respectively, using the MirVana kit followed by gel purification. The RNAs were cloned and the cDNA libraries were made without CIP and PNK treatments. To identify the 5' and 3' ends of pre-milR-1 and pri-milR-1, the intermediates were amplified using a milR-1 specific primer paired with one of the primers corresponding to the 5' or 3' linker sequence. Since PsiI and BspEI sites partially overlapped with the 5’ and 3’ ends of mature milRNA, respectively, the specific milR-1 amplified products are identified by their sensitivity to PsiI or BsPE1 digestions and were subjected to DNA sequencing. At least three amplified products of each species were sequenced.

Northern blot analyses and qPCR assay

12.5 to 25 µg of small RNA was separated on a 16% denaturing polyacrylamide gel Urea, and transferred onto a Hybond-NX membrane (GE Healthcare). An RNA ladder (Ambion) was used as a size marker. Cross-linking of RNA to Hybond-NX was performed using a carbodiimide-mediated cross-linking method at 60°C for 2 hours as described (Pall et al., 2007), followed by baking at 80°C for 2 hours. Hybridization was performed according to the manufacturer's instructions in ULTRAhyb™-Oligo hybridization buffer (Ambion) for StareFire® (IDT) labeled DNA probes, or in ULTRAhyb™ hybridization buffer (Ambion) for T7-transcribed riboprobes. The Starfire probe sequences and primer sequences for PCR fragments used in T7 transcribed riboprobes was described in supplementary materials and methods. Northern blot and qRT-PCR analyses using total RNA were performed as previously described (Cheng et al., 2001)(Choudhary et al., 2007).

Strains that express the milR-1 targeting reporter protein

To obtain the fragment containing five milR-1 milRNA sequences, we designed two complementary primers with Ban I sites at both 5’ and 3’ ends. (5’ggtgccATAAGCCGCGAGTACGCCTCCGGACTGTAAggtgcc3’; 5’ggcaccTTACAGTCCGGAGGCGTACTCGCGG CTTATggcacc3’). The two primers were annealed, digested with Ban I enzyme, and self-ligated to create concatamers. The concatamers were gel purified and cloned into TOPO-pCRII (Invitrogen). Sequencing validated TOPO-pCRII_milr-1 with five milR-1 miRNA complementary sequences, which was cloned into the EcoRI site of qaMyc-QDE-2, resulting in the insertion of five milR-1 miRNA targeting sequences within the Myc-QDE-2 ORF. The insertion localizes in the Piwi domain and should result in a non-functional protein. The reporter construct was introduced to the wild type strain and the qde-2rip strain at the his-3 locus. The expression of the reporter is induced by the presence of quinic acid.

Target prediction for milRs

Neurospora genes and the regions 1kb of the 3’ ends were used for the search of milR targets by miRmate, a mammalian miRNA target prediction algorithm (Du et al., 2009; Younger et al., 2009). miRmate considers complementarity (allowing the G-U base pairing) based on variable weighting of positions across the miRNA, so as to encourage (but not require) complete complementarity at the seed region (positions 2–7 of the miRNA) and mismatches or insertions in the central bulge at positions 9–11 of the miRNA (Lewis et al., 2005).

Pre-miRNA cleavage assay

A 61 nt synthetic Drosophila pre-let-7 RNA, 5’-UGAGGUAGUAGGUUGUAUAGU AGUAAUUACACAUCAUACUAUACAAUGUGCUAGCUUUCUU-3’ was kindly provided by Dr. Qinghua Liu and labeled at the 5’ end by PNK. For pre-milRNA of milR-4, the predicted hairpin sequence was PCR amplified and T7 transcribed in the presence of 32P-UTP. Labeled pre-RNA was gel purified before applied in the in vitro pre-milRNA processing assays as previously described (Jiang et al., 2005). Cell extracts (30 µg total protein) from different strains were used in the assays.

Supplementary Material

01

Acknowledgements

We thank Hsing-I Ho, Haiyan Yuan and Dr. Qinghua Liu for technical assistance. We thank Drs. Victor Ambros and Jim Carrington for insightful discussions on miRNA. This work was supported by grants from the National Institutes of Health to Yi Liu, Craig Mello, Eric Selker, and Alexander Pertsemlidis, from the Welch Foundation to Yi Liu, and from the American Cancer Society (RSG-08-030-01-CCG) to Michael Freitag. Z.A.L. was supported by fellowship from the Leukemia and Lymphoma Society (3295-09). Craig Mello is a Howard Hughes Medical Institute Investigator.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Ambros V. The functions of animal microRNAs. Nature. 2004;431:350–355. doi: 10.1038/nature02871. [DOI] [PubMed] [Google Scholar]
  2. Ambros V, Bartel B, Bartel DP, Burge CB, Carrington JC, Chen X, Dreyfuss G, Eddy SR, Griffiths-Jones S, Marshall M, et al. A uniform system for microRNA annotation. RNA. 2003;9:277–279. doi: 10.1261/rna.2183803. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Aravin A, Gaidatzis D, Pfeffer S, Lagos-Quintana M, Landgraf P, Iovino N, Morris P, Brownstein MJ, Kuramochi-Miyagawa S, Nakano T, et al. A novel class of small RNAs bind to MILI protein in mouse testes. Nature. 2006;442:203–207. doi: 10.1038/nature04916. [DOI] [PubMed] [Google Scholar]
  4. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004;116:281–297. doi: 10.1016/s0092-8674(04)00045-5. [DOI] [PubMed] [Google Scholar]
  5. Batista PJ, Ruby JG, Claycomb JM, Chiang R, Fahlgren N, Kasschau KD, Chaves DA, Gu W, Vasale JJ, Duan S, et al. PRG-1 and 21U–RNAs interact to form the piRNA complex required for fertility in C. elegans. Mol Cell. 2008;31:67–78. doi: 10.1016/j.molcel.2008.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Brennecke J, Aravin AA, Stark A, Dus M, Kellis M, Sachidanandam R, Hannon GJ. Discrete small RNA-generating loci as master regulators of transposon activity in Drosophila. Cell. 2007;128:1089–1103. doi: 10.1016/j.cell.2007.01.043. [DOI] [PubMed] [Google Scholar]
  7. Buhler M, Moazed D. Transcription and RNAi in heterochromatic gene silencing. Nat Struct Mol Biol. 2007;14:1041–1048. doi: 10.1038/nsmb1315. [DOI] [PubMed] [Google Scholar]
  8. Catalanotto C, Azzalin G, Macino G, Cogoni C. Gene silencing in worms and fungi. Nature. 2000;404:245. doi: 10.1038/35005169. [DOI] [PubMed] [Google Scholar]
  9. Catalanotto C, Nolan T, Cogoni C. Homology effects in Neurospora crassa. FEMS Microbiol Lett. 2006;254:182–189. doi: 10.1111/j.1574-6968.2005.00037.x. [DOI] [PubMed] [Google Scholar]
  10. Catalanotto C, Pallotta M, ReFalo P, Sachs MS, Vayssie L, Macino G, Cogoni C. Redundancy of the two dicer genes in transgene-induced posttranscriptional gene silencing in Neurospora crassa. Mol Cell Biol. 2004;24:2536–2545. doi: 10.1128/MCB.24.6.2536-2545.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cheng P, He Q, He Q, Wang L, Liu Y. Regulation of the Neurospora circadian clock by an RNA helicase. Genes Dev. 2005;19:234–241. doi: 10.1101/gad.1266805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cheng P, Yang Y, Heintzen C, Liu Y. Coiled-coil domain mediated FRQ-FRQ interaction is essential for its circadian clock function in Neurospora. EMBO J. 2001;20:101–108. doi: 10.1093/emboj/20.1.101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Choudhary S, Lee HC, Maiti M, He Q, Cheng P, Liu Q, Liu Y. A doublestranded-RNA response program important for RNA interference efficiency. Mol Cell Biol. 2007;27:3995–4005. doi: 10.1128/MCB.00186-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Cogoni C, Macino G. Gene silencing in Neurospora crassa requires a protein homologous to RNA-dependent RNA polymerase. Nature. 1999a;399:166–169. doi: 10.1038/20215. [DOI] [PubMed] [Google Scholar]
  15. Cogoni C, Macino G. Posttranscriptional gene silencing in Neurospora by a RecQ DNA helicase. Science. 1999b;286:2342–2344. doi: 10.1126/science.286.5448.2342. [DOI] [PubMed] [Google Scholar]
  16. Das PP, Bagijn MP, Goldstein LD, Woolford JR, Lehrbach NJ, Sapetschnig A, Buhecha HR, Gilchrist MJ, Howe KL, Stark R, et al. Piwi and piRNAs act upstream of an endogenous siRNA pathway to suppress Tc3 transposon mobility in the Caenorhabditis elegans germline. Mol Cell. 2008;31:79–90. doi: 10.1016/j.molcel.2008.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Drinnenberg IA, Weinberg DE, Xie KT, Mower JP, Wolfe KH, Fink GR, Bartel DP. RNAi in budding yeast. Science. 2009;326:544–550. doi: 10.1126/science.1176945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Du L, Schageman JJ, Subauste MC, Saber B, Hammond SM, Prudkin L, Wistuba II, Ji L, Roth JA, Minna JD, et al. miR-93, miR-98, and miR-197 regulate expression of tumor suppressor gene FUS1. Mol Cancer Res. 2009;7:1234–1243. doi: 10.1158/1541-7786.MCR-08-0507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Ghildiyal M, Zamore PD. Small silencing RNAs: an expanding universe. Nat Rev Genet. 2009;10:94–108. doi: 10.1038/nrg2504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Girard A, Sachidanandam R, Hannon GJ, Carmell MA. A germline-specific class of small RNAs binds mammalian Piwi proteins. Nature. 2006;442:199–202. doi: 10.1038/nature04917. [DOI] [PubMed] [Google Scholar]
  21. Grimson A, Srivastava M, Fahey B, Woodcroft BJ, Chiang HR, King N, Degnan BM, Rokhsar DS, Bartel DP. Early origins and evolution of microRNAs and Piwi-interacting RNAs in animals. Nature. 2008;455:1193–1197. doi: 10.1038/nature07415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Grivna ST, Beyret E, Wang Z, Lin H. A novel class of small RNAs in mouse spermatogenic cells. Genes Dev. 2006;20:1709–1714. doi: 10.1101/gad.1434406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gunawardane LS, Saito K, Nishida KM, Miyoshi K, Kawamura Y, Nagami T, Siomi H, Siomi MC. A slicer-mediated mechanism for repeat-associated siRNA 5' end formation in Drosophila. Science. 2007;315:1587–1590. doi: 10.1126/science.1140494. [DOI] [PubMed] [Google Scholar]
  24. Halic M, Moazed D. Dicer-independent primal RNAs trigger RNAi and heterochromatin formation. Cell. 2010;140:504–516. doi: 10.1016/j.cell.2010.01.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hannon GJ. RNA interference. Nature. 2002;418:244–251. doi: 10.1038/418244a. [DOI] [PubMed] [Google Scholar]
  26. Jiang F, Ye X, Liu X, Fincher L, McKearin D, Liu Q. Dicer-1 and R3D1-L catalyze microRNA maturation in Drosophila. Genes Dev. 2005;19:1674–1679. doi: 10.1101/gad.1334005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Jones-Rhoades MW, Bartel DP, Bartel B. MicroRNAS and their regulatory roles in plants. Annu Rev Plant Biol. 2006;57:19–53. doi: 10.1146/annurev.arplant.57.032905.105218. [DOI] [PubMed] [Google Scholar]
  28. Kawamata T, Seitz H, Tomari Y. Structural determinants of miRNAs for RISC loading and slicer-independent unwinding. Nat Struct Mol Biol. 2009;16:953–960. doi: 10.1038/nsmb.1630. [DOI] [PubMed] [Google Scholar]
  29. Lagos-Quintana M, Rauhut R, Lendeckel W, Tuschl T. Identification of novel genes coding for small expressed RNAs. Science. 2001;294:853–858. doi: 10.1126/science.1064921. [DOI] [PubMed] [Google Scholar]
  30. Lau NC, Seto AG, Kim J, Kuramochi-Miyagawa S, Nakano T, Bartel DP, Kingston RE. Characterization of the piRNA complex from rat testes. Science. 2006;313:363–367. doi: 10.1126/science.1130164. [DOI] [PubMed] [Google Scholar]
  31. Lee HC, Chang SS, Choudhary S, Aalto AP, Maiti M, Bamford DH, Liu Y. qiRNA is a new type of small interfering RNA induced by DNA damage. Nature. 2009;459:274–277. doi: 10.1038/nature08041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lee RC, Ambros V. An extensive class of small RNAs in Caenorhabditis elegans. Science. 2001;294:862–864. doi: 10.1126/science.1065329. [DOI] [PubMed] [Google Scholar]
  33. Lee RC, Feinbaum RL, Ambros V. The C. elegans heterochromatic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell. 1993;75:843–854. doi: 10.1016/0092-8674(93)90529-y. [DOI] [PubMed] [Google Scholar]
  34. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell. 2005;120:15–20. doi: 10.1016/j.cell.2004.12.035. [DOI] [PubMed] [Google Scholar]
  35. Lewis ZA, Honda S, Khlafallah TK, Jeffress JK, Freitag M, Mohn F, Schubeler D, Selker EU. Relics of repeat-induced point mutation direct heterochromatin formation in Neurospora crassa. Genome Res. 2009;19:427–437. doi: 10.1101/gr.086231.108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Liu Q, Rand TA, Kalidas S, Du F, Kim HE, Smith DP, Wang X. R2D2, a bridge between the initiation and effector steps of the Drosophila RNAi pathway. Science. 2003;301:1921–1925. doi: 10.1126/science.1088710. [DOI] [PubMed] [Google Scholar]
  37. Liu Y, Ye X, Jiang F, Liang C, Chen D, Peng J, Kinch LN, Grishin NV, Liu Q. C3PO, an endoribonuclease that promotes RNAi by facilitating RISC activation. Science. 2009;325:750–753. doi: 10.1126/science.1176325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Llave C, Kasschau KD, Rector MA, Carrington JC. Endogenous and silencing-associated small RNAs in plants. Plant Cell. 2002;14:1605–1619. doi: 10.1105/tpc.003210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Maiti M, Lee HC, Liu Y. QIP, a putative exonuclease, interacts with the Neurospora Argonaute protein and facilitates conversion of duplex siRNA into single strands. Genes Dev. 2007;21:590–600. doi: 10.1101/gad.1497607. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Molnar A, Schwach F, Studholme DJ, Thuenemann EC, Baulcombe DC. miRNAs control gene expression in the single-cell alga Chlamydomonas reinhardtii. Nature. 2007;447:1126–1129. doi: 10.1038/nature05903. [DOI] [PubMed] [Google Scholar]
  41. Pall GS, Codony-Servat C, Byrne J, Ritchie L, Hamilton A. Carbodiimide-mediated cross-linking of RNA to nylon membranes improves the detection of siRNA, miRNA and piRNA by northern blot. Nucleic Acids Res. 2007;35:e60. doi: 10.1093/nar/gkm112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Ruby JG, Jan C, Player C, Axtell MJ, Lee W, Nusbaum C, Ge H, Bartel DP. Large-scale sequencing reveals 21U–RNAs and additional microRNAs and endogenous siRNAs in C. elegans. Cell. 2006;127:1193–1207. doi: 10.1016/j.cell.2006.10.040. [DOI] [PubMed] [Google Scholar]
  43. Smits P, Smeitink JA, van den Heuvel LP, Huynen MA, Ettema TJ. Reconstructing the evolution of the mitochondrial ribosomal proteome. Nucleic Acids Res. 2007;35:4686–4703. doi: 10.1093/nar/gkm441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Song JJ, Smith SK, Hannon GJ, Joshua-Tor L. Crystal structure of Argonaute and its implications for RISC slicer activity. Science. 2004;305:1434–1437. doi: 10.1126/science.1102514. [DOI] [PubMed] [Google Scholar]
  45. Vagin VV, Sigova A, Li C, Seitz H, Gvozdev V, Zamore PD. A distinct small RNA pathway silences selfish genetic elements in the germline. Science. 2006;313:320–324. doi: 10.1126/science.1129333. [DOI] [PubMed] [Google Scholar]
  46. Younger ST, Pertsemlidis A, Corey DR. Predicting potential miRNA target sites within gene promoters. Bioorg Med Chem Lett. 2009;19:3791–3794. doi: 10.1016/j.bmcl.2009.04.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Zhao T, Li G, Mi S, Li S, Hannon GJ, Wang XJ, Qi Y. A complex system of small RNAs in the unicellular green alga Chlamydomonas reinhardtii. Genes Dev. 2007;21:1190–1203. doi: 10.1101/gad.1543507. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS195757-supplement-01.doc (1.1MB, doc)

RESOURCES