...and SNIP1 in humans act in small RNA biogenesis | PNAS

来自 : 发布时间：2024-12-22

Edited by Richard M. Amasino, University of Wisconsin, Madison, WI, and approved May 21, 2008 (received for review May 1, 2008) AbstractProteins containing the forkhead-associated domain (FHA) are known to act in biological processes such as DNA damage repair, protein degradation, and signal transduction. Here we report that DAWDLE (DDL), an FHA domain-containing protein in Arabidopsis, acts in the biogenesis of miRNAs and endogenous siRNAs. Unlike mutants of genes known to participate in the processing of miRNA precursors, such as dcl1, hyponastic leaves1, and serrate, ddl mutants show reduced levels of pri-miRNAs as well as mature miRNAs. Promoter activity of MIR genes, however, is not affected by ddl mutations. DDL is an RNA binding protein and is able to interact with DCL1. In addition, we found that SNIP1, the human homolog of DDL, is involved in miRNA biogenesis and interacts with Drosha. Therefore, we uncovered an evolutionarily conserved factor in miRNA biogenesis. We propose that DDL participates in miRNA biogenesis by facilitating DCL1 to access or recognize pri-miRNAs.DCL1microRNAsiRNADroshaSMADA class of sequence-specific repressors of gene expression in eukaryotes is 20- to 24-nt small RNAs, which include miRNAs and siRNAs. miRNAs are processed from stem-loop precursor RNAs, called pri-miRNAs. In animals, pri-miRNAs are processed in the nucleus by Drosha to form pre-miRNAs, which are exported to the cytoplasm by exportin 5 and further processed by Dicer to produce mature miRNAs (reviewed in ref. 1). In Arabidopsis, mature miRNAs are produced through two processing steps (pri-miRNAs to pre-miRNAs and pre-miRNAs to miRNAs) in the nucleus by DCL1 with the assistance of HYL1 and SERRATE (reviewed in ref. 2). After processing, miRNAs are 2′-O-methylated by HEN1 (3). siRNAs are produced from long, double-stranded RNAs. Plants contain several classes of endogenous siRNAs, such as transacting siRNAs (ta-siRNAs), natural antisense siRNAs (nat-siRNAs), and siRNAs from endogenous repeat sequences and transposons (reviewed in ref. 4).The forkhead-associated (FHA) domain is an 80- to 100-aa module that is thought to recognize phosphothreonine-containing motifs and mediate protein–protein interactions in prokaryotes and eukaryotes (reviewed in ref. 5). DAWDLE (DDL) is a nuclear-localized FHA domain-containing protein in Arabidopsis (6). DDL appears to act in multiple developmental processes such as growth, fertility, and root, shoot, and floral morphogenesis (6).Smad nuclear interacting protein 1 (SNIP1) is a human FHA domain-containing protein that functions as an inhibitor of TGF-β and NF-κB signaling pathways by competing with the TGF-β signaling protein Smad4 and the NF-κB transcription factor p65/RelA for binding to the transcriptional coactivator p300 (7, 8). Recently, Fujii et al. (9) reported that SNIP1 interacts with the transcription factor/oncoprotein c-Myc and enhances its activity by bridging its interaction with p300.Here we report that DDL is required for the accumulation of miRNAs and endogenous siRNAs in Arabidopsis. Its affinity for RNA, its potential association with DCL1, and the reduction in pri-miRNA levels in ddl loss-of-function mutants suggest that DDL is a candidate protein recruiting DCL1 to its substrates. In addition, we show that SNIP1 is a human ortholog of DDL and that it also acts in miRNA biogenesis.ResultsDDL Acts in miRNA Biogenesis in Arabidopsis.ddl-1 and ddl-2 are two recessive, potentially null alleles in the DDL gene in the Wassilewskija genetic background (6). The ddl-1 and ddl-2 mutants show delayed growth and reduced fertility, and have defects in root, shoot, and floral morphology. These pleiotropic developmental defects resemble those of mutants deficient in miRNA biogenesis and prompted us to test whether the ddl mutants are compromised in miRNA accumulation. We examined the abundance of various miRNAs in ddl-1 and ddl-2 by RNA filter hybridization. Indeed, the levels of 9 of 10 tested DCL1-dependent miRNAs were reduced by 2- to 3.3-fold in ddl-1 and ddl-2, relative to WT [Fig. 1 and supporting information (SI) Fig. S1]. The levels of an antisense miRNA, miR172*, were reduced by 3.3 times in ddl mutants (Fig. 1). Introduction of a DDL transgene into ddl-1 rescued the morphological defects of the mutants (6) and fully recovered the levels of miRNAs and miR172* (Fig. 1 and Fig. S1), demonstrating that the defects in miRNA accumulation in the two mutants were due to DDL loss of function. To determine whether DDL plays a role in the methylation of miRNAs, we evaluated the methylation status of miR161 in ddl mutants by treating total RNAs with sodium periodate followed by β-elimination (3) and analyzing miR161 by filter hybridization. Loss of methylation would result in faster migration of the RNA in this assay (3). We found that the ddl mutations had no detectable effects on the methylation of miR161 (Fig. S1).Download figureOpen in new tabDownload powerpointFig. 1. DDL is required for the accumulation of miRNAs. The accumulation of various miRNAs and miR172* as detected by Northern blotting in Ws (WT), ddl-1, ddl-2, and a ddl-1 transgenic line harboring DDL genomic DNA. Note that except for miR822, which is DCL4-dependent, all miRNAs are DCL1-dependent. Total RNAs were extracted from inflorescences. The control 5S rRNA blots were below the corresponding miRNA blots. In cases where a membrane was used for several miRNAs, there would be one 5S rRNA blot for several miRNA blots. The numbers indicate the relative abundance of miRNAs or miR172* among the four genotypes.DDL Is Required for the Biogenesis of ta-siRNAs and Repeated DNA-Associated siRNAs.We next tested whether DDL is involved in the biogenesis of endogenous siRNAs. We found that two DCL4-dependent siRNAs—siRNA1511, a ta-siRNA from the TAS2 locus (10), and siRNA255, a ta-siRNA from the TAS1 locus (10)—were reduced in abundance in ddl mutants (Fig. 2A). The reduced accumulation of these ta-siRNAs was rescued by the DDL transgene (Fig. 2A). Because the ta-siRNAs require a DCL1-dependent miRNA in their biogenesis, the reduction in ta-siRNA accumulation in ddl mutants could not support a direct role of DDL in the biogenesis of DCL4-dependent small RNAs. We tested the effect of ddl mutations on the accumulation of a DCL4-dependent miRNA, miR822 (11). We found that the ddl mutations led to reduced levels of this miRNA, and that DDL genomic DNA rescued the molecular defect in ddl-1 (Fig. 1).Download figureOpen in new tabDownload powerpointFig. 2. DDL is required for the accumulation of endogenous siRNAs. (A) ta-siRNAs were detected in various genotypes by Northern blotting. (B) Repeated DNA-associated siRNAs were detected by Northern blotting in various genotypes. The same membrane was used to probe for siRNA 255, siRNA 02, and siRNA 1003, and the corresponding 5S rRNA control is shown twice to aid visual comparison. The numbers below the hybridization images indicate the relative abundance of siRNAs among the four genotypes. Ws, the WT control for the ddl mutants; ddl-1+DDL, a ddl-1 mutant rescued by DDL genomic DNA.We also examined the levels of DCL3-dependent siRNAs from repeated DNA or transposons, such as siRNA02, siRNA1003, AtSN1 siRNAs, and cluster 2 siRNAs. Three of the four siRNAs were reduced by 2- to 10-fold in the ddl mutants, and this reduction was rescued by the DDL transgene (Fig. 2B).The Amount of pri- and pre-miRNAs Is Reduced in the ddl Mutants.To determine the step at which a defect in miRNA biogenesis occurred in ddl mutants, we examined the levels of pri-miRNAs and pre-miRNAs in WT and ddl mutants. We determined the levels of pri-miRNAs at five MIR loci (MIR159a, MIR166a, MIR167a, MIR171a, and MIR172b) by RT-PCR. The levels of the five tested pri-miRNAs were reduced by 1.7- to 3.0-fold in ddl-1 and 1.9- to 4.8-fold in ddl-2 relative to WT (Fig. 3A). Next we examined the levels of pre-miR166a and pre-miR164b, which were shown to be detectable by RNA filter hybridization (12). We also included dcl1-9, which has reduced levels of pre-miRNAs (12), and its WT control (Ler) in the analysis. The levels of the two pre-miRNAs were reduced in ddl mutants (Fig. 3B). Furthermore, in the ddl mutants, the levels of pri-miRNAs, pre-miRNAs, and miRNAs appeared to be reduced to a similar extent (Figs. 1 and 3).Download figureOpen in new tabDownload powerpointFig. 3. The accumulation of both pri- and pre-miRNAs in inflorescences is reduced in ddl mutants. (A) RT-PCR analysis of the levels of pri-miRNAs in WT (Ws), ddl-1, and ddl-2 inflorescences. The levels of pri-miRNAs in ddl mutants were normalized to those of UBIQUITIN 5 and compared with WT. (B) Northern blot analysis of pre-miRNAs in various genotypes. Ler, the WT control for dcl1-9; ddl-1+DDL, a ddl-1 mutant rescued by DDL genomic DNA. The numbers below the hybridization images indicate the relative abundance of pre-miRNAs among the four genotypes.DDL Does Not Control the Transcription of MIR Genes.The reduction in pri-miRNA levels in ddl mutants raised the possibility that DDL is a general transcription factor for MIR genes or that DDL is a general regulator of transcription for most or all genes. We used two strategies to determine whether DDL controls the transcription of MIR genes. First, we tested the effect of the ddl-1 mutation on the expression of a GUS reporter gene under the control of the promoter of MIR172a or MIR172b. If DDL was a transcriptional regulator of MIR genes, the ddl-1 mutation would be expected to affect the expression of the GUS transgene like it affects the endogenous MIR genes. We first generated transgenic lines containing a single-locus pMIR172a::GUS or pMIR172b::GUS transgene. The promoters of the two genes were defined as the genomic DNA from the upstream neighboring gene to the start of transcription of MIR172a and MIR172b as determined by 5′ RACE (data not shown) and were 6.3 kb and 3.4 kb, respectively, for MIR172a and MIR172b. The transgenic plants were crossed to ddl-1, and DDL+ (DDL/DDL or DDL/ddl-1) or ddl-1 genotypes containing the GUS transgene were obtained in the F2 generation. The ddl-1 mutation did not affect the amount of GUS staining resulting from pMIR172a::GUS or pMIR172b::GUS (Fig. 4A). Because GUS staining might not be a quantitative measure of transgene expression, we also determined the levels of GUS transcripts from the transgenes by RT-PCR. No difference in GUS transcript levels was found between DDL+ and ddl-1 plants (Fig. S2). In addition, the expression of a GUS reporter driven by the promoter of PISTILLATA (13), a protein coding gene, was also similar between DDL+ and ddl genotypes (Fig. 4A), suggesting that DDL is unlikely a general regulator of transcription.Download figureOpen in new tabDownload powerpointFig. 4. DDL does not act through the promoters of MIR genes. (A) The ddl-1 mutation has no obvious effects on GUS expression directed by the promoter of MIR172a or MIR172b. Many promoter-GUS plants were analyzed. A representative image is shown in each case. DDL+: DDL/DDL or DDL/ddl-1. (B) RT-PCR analysis of the levels of pri-miR172a in DDL+ and ddl-1 plants harboring a p35S::MIR172a transgene. (C) Northern blotting to determine the accumulation of pre-miR172a and miR172 in DDL+ and ddl-1 plants with (+) or without (−) the 35S::MIR172a transgene.Second, we monitored the steady-state levels of pri-miR172a, pre-miR172a, and miR172 in DDL+ and ddl-1 plants containing a single-locus MIR172a transgene under the control of the cauliflower mosaic virus 35S promoter. If DDL functions as a transcriptional regulator of MIR172a through its promoter, the ddl-1 mutation would not be expected to affect the expression of MIR172a driven by the 35S promoter. We crossed a transgenic line containing a single locus of 35S::MIR172a (14) to ddl-1 and obtained DDL+ and ddl-1 plants containing the transgene in the F2 generation. As expected, the presence of the 35S::MIR172 transgene led to a large increase in the levels of miR172 relative to control Ws plants (Fig. 4C; compare lanes with the 35S::MIR172a transgene to those without the transgene), which indicated that the miR172 signal detected in the presence of the transgene largely reflected the miRNA pool produced from the transgene. It was obvious that both miR172 and pre-miR172a levels were reduced in the ddl-1 mutant (Fig. 4C). The levels of pri-miR172a, as determined by RT-PCR, were also reduced in ddl-1 (Fig. 4B). These data demonstrated that DDL did not act through the promoters of MIR genes.DDL Is Associated with DCL1.Because our results did not support a role of DDL in the transcriptional control of MIR genes, we investigated other possibilities that may explain the reduced levels of miRNAs in ddl mutants. One possibility was that DDL regulates the expression of genes involved in miRNA biogenesis. We determined the levels of DCL1, HYL1, and SERRATE RNAs by RT-PCR and the levels of HEN1 protein by using Western blot analysis. We found that the ddl mutations had no obvious effects on the expression of these genes (data not shown). Next, we tested for interaction of DDL with proteins involved in miRNA biogenesis with a yeast two-hybrid assay and did not detect any interaction between DDL and AGO1, HEN1, or HYL1 (data not shown). However, DDL was found to interact with an N-terminal portion (amino acids 1–833), but not a C-terminal portion (amino acids 814-1909), of DCL1 in the yeast two-hybrid assay (Fig. 5A).Download figureOpen in new tabDownload powerpointFig. 5. DDL interacts with DCL1. (A) The N-terminal fragment of DCL1 (DCL1N) is necessary and sufficient to interact with DDL in yeast. Interaction between DCL1N with DDL was manifested by the ability of yeast cells (pJ69-4A) to grow in adenine-deficient medium (-Ade–Leu–Trp). The pGBKT7 and pGAD10 vectors contained the DNA binding and activation domains of GAL4, respectively. Yeast colonies containing two plasmids were first selected in -Leu–Trp medium. Then, cells from one colony were resuspended and spotted onto -Leu–Trp–Ade medium. (B) CoIP of DDL and DCL1. IP was performed on extracts containing transiently expressed DCL1-YFP and HA-DDL or extracts containing transiently expressed HA-DDL alone by using polyclonal antibodies against GFP and GFP variants. After IP, DCL1-YFP and HA-DDL were detected by using Western blot analysis with anti-GFP and anti-HA antibodies, respectively. CoIP was also performed similarly for DCL1-YFP and HA-HEN1, which served as a control for HA-DDL.To confirm the interaction between DDL and DCL1 with an independent assay, we tested coimmunoprecipitation (coIP) of the two proteins transiently expressed in Nicotiana benthamiana. The presence of the viral-silencing suppressor p19 of tomato bushy stunt virus was reported to result in at least a 50-fold increase in the expression of a target protein in a transient expression system in N. benthamiana (15). We used this system to separately express the DCL1 protein fused to a yellow fluorescent protein (DCL1-YFP) and the DDL protein fused to an HA epitope (HA-DDL). The HA-DDL extract was then mixed with the DCL1-YFP extract at a 1:1 ratio. We then used anti-GFP (and GFP variants) polyclonal antibodies conjugated to agarose beads to pull down DCL1-YFP from mixed DCL1-YFP/HA-DDL extracts or from HA-DDL extracts alone (control). As shown in Fig. 5B, an anti-HA antibody detected the enrichment of HA-DDL in the DCL1-YFP immunoprecipitate relative to the control immunoprecipitate, suggesting the association between DDL and DCL1. As another negative control, we expressed HA-HEN1 and performed coIP on DCL1-YFP and HA-HEN1 with the same procedure. We did not find any association between DCL1-YFP and HA-HEN1 (Fig. 5B).It should be noted that only a small portion of the input HA-DDL was associated with DCL1-YFP in this assay. In fact, HA-DDL was not detected in the DCL1-YFP immunoprecipitate when the levels of DCL1-YFP were low (i.e., in the absence of p19). These observations suggested that only a portion of DCL1 molecules could interact with DDL or that DDL interacted transiently with DCL1.DDL Is an RNA Binding Protein.BLAST analysis showed that an N-terminal domain of DDL was conserved in several proteins with roles in RNA metabolism (Fig. 6A) (data not shown). The presence of this domain prompted us to test whether DDL was an RNA binding protein. We performed a GST pull-down assay to test whether DDL could bind to pri-miR162b, which was transcribed in vitro in the presence of [α-32P]UTP. We chose to test pri-miR162b because its 5′ and 3′ ends were determined experimentally (16). GST-DDL or the control GST was expressed in E. coli and purified with glutathione beads. Labeled pri-miR162b was added to the beads containing GST-DDL or GST. After washes and elution, the RNA was resolved on a polyacrylamide gel. Pri-miR162b was retained by GST-DDL but not GST alone (Fig. 7 A and B). Under the same conditions, double-stranded DNA corresponding to pri-miR162b was not retained by GST-DDL (Fig. 7B). We found that GST-DDL could also bind to an in vitro transcribed RNA corresponding to part of the APETALA1 mRNA (data not shown). These observations indicate that DDL is an RNA binding protein, but it does not bind specifically to pri-miRNAs in vitro.Download figureOpen in new tabDownload powerpointFig. 6. SNIP1, a human ortholog of DDL, functions in miRNA biogenesis. (A) Diagrams show DDL and its homologs in animals. The gray box represents a domain of unknown function (DUF) present in proteins with roles in RNA metabolism. The hatched box indicates the FHA domain. The overall amino acid similarity between DDL and its homologs is 50–60%. The similarity between DDL and its homologs in the DUF and FHA regions is ≈40–50% and 80–90%, respectively. Bt, Bos taurus; Hs, Homo sapiens; Pt, Pan troglodytes; Mm, Mus musculus; Dm, Drosophila melanogaster; Gg, Gallus gallus; Dd, Danio rerio. (B) The levels of endogenous SNIP1 were reduced in HeLa cells transfected with SNIP1 siRNA1 or siRNA2 as compared with cells transfected with a control nontargeting siRNA. The SNIP protein was detected by using Western blot analysis with anti-SNIP1 antibodies. Part of the stained protein gel is shown below to indicate near equal loading. (C) The accumulation of miRNAs as determined by Northern blotting. The signals were quantified with a phosphoimager and normalized against 5S rRNA. The numbers indicate the relative abundance of the miRNAs between control siRNA- and SNIP1 siRNA-treated cells. (D) coIP of SNIP1 and Drosha. IP was performed on extracts containing Drosha-Flag and HA-SNIP1 or extracts containing Drosha-Flag alone by using an immobilized anti-HA monoclonal antibody. After IP, Drosha-Flag and HA-SNIP1 were detected by using Western blot analysis with anti-Flag and anti-HA antibodies, respectively. The \"− HA-SNIP1” lanes represent extracts from cells transfected with the pcDNA3 vector alone. Additional negative controls (Myc-exportin 5 and Dicer) are shown in Fig. S3.Download figureOpen in new tabDownload powerpointFig. 7. DDL is an RNA binding protein. (A) The two proteins used for the RNA binding assay, GST and GST-DDL, were resolved on an SDS/polyacrylamide gel to show that near equal amounts of the two proteins were used. The lower bands in the GST-DDL lane were truncated GST-DDL proteins because they were all recognized by anti-GST antibodies (data not shown). (B) DDL binds pri-miR162b but not DNA in vitro. In vitro-synthesized and 32P-labeled pri-miR162b or PCR-amplified and labeled DNA corresponding to pri-miR162b was incubated with purified GST or GST-DDL. After washing, the bound nucleic acids were eluted, denatured, and resolved on a polyacrylamide gel. (C) A proposed model for DDL\'s function in miRNA biogenesis. We hypothesize that DDL promotes the access of DCL1 to its substrates. Without DDL, a portion of the pri-miRNA pool cannot be processed by DCL1 and is degraded.The Human Homolog of DDL Is Involved in miRNA Biogenesis.By BLAST searches of the nonredundant protein databases of the National Center for Biotechnology Information, we found that the SNIP1 proteins from human, mouse, and other organisms were obvious homologs of DDL (Fig. 6A). The human SNIP1 protein was reported to function in TGF-β and NF-κB signaling pathways, but a potential role in miRNA metabolism was not evaluated.We tested whether the human SNIP1 protein was involved in miRNA biogenesis. We first used an siRNA (siRNA1) targeting SNIP1 to knock down SNIP1 expression in HeLa cells. Indeed, siRNA1 targeting SNIP1 reduced the levels of SNIP1 protein by ≈50% relative to a nontargeting control siRNA (Fig. 6B). The accumulation of five tested miRNAs (let-7i, miR21, miR22, miR23, and miR25) showed 1.7- to 2-fold reduction after SNIP1 siRNA1 treatment (Fig. 6C). We further tested the effect of a different SNIP1 targeting siRNA (siRNA2). Transfection of HeLa cells with siRNA2 resulted in 50% reduction in SNIP1 protein levels relative to a nontargeting control siRNA (Fig. 6B). A 2-fold reduction in three tested miRNAs was observed in siRNA2-treated cells (Fig. 6C). These data demonstrated that human SNIP1 functions in miRNA metabolism and is therefore likely an ortholog of DDL.Both SNIP1 and Drosha, which processes pri-miRNAs in humans, are localized in the nucleus. We reasoned that, like the association between DDL and DCL1, which processes pri-miRNAs in plants, SNIP1 might be associated with Drosha. We tested the association of SNIP1 and Drosha by coIP. We first transfected 293T cells with an HA-SNIP1 plasmid (8), the vector alone control for HA-SNIP1, or a Drosha-Flag plasmid (17). Western blot analyses showed that HA-SNIP1 and Drosha-Flag were indeed expressed in the transfected cells (Fig. 6D, \"Input” lanes). Next, we mixed the extract from Drosha-Flag cells with the HA-SNIP1 extract or the vector alone extract and performed coIP. After immunoprecipitation of HA-SNIP1 with an anti-HA antibody, we were able to detect the presence of Drosha-Flag in the SNIP1 immunoprecipitate, but not in the control immunoprecipitate, with an anti-Flag antibody (Fig. 6D). To further test the specificity of the SNIP1-Drosha interaction, we performed coIP of SNIP1 and exportin 5 and of SNIP1 and Dicer. We first transfected 293T cells with the HA-SNIP1 plasmid, the vector alone control, a Dicer plasmid (Gregory Hannon, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), or a Myc-exportin 5 plasmid (18). Next, we confirmed that HA-SNIP1, Dicer, and Myc-exportin 5 were expressed in the cells (Fig. S3, \"Input” lanes). We then mixed Myc-exportin 5 or Dicer extracts with the HA-SNIP1 extract or the vector alone extract and performed immunoprecipitation with anti-HA antibodies. Although HA-SNIP1 was successfully immunoprecipitated, neither Dicer nor exportin 5 was detected in the immunoprecipitate (Fig. S3). Therefore, SNIP1 was associated with Drosha in vivo.DiscussionBecause miRNAs are crucial regulators of plant development, several genes (DCL1, HYL1, HASTY, HEN1, and SERRATE) involved in miRNA biogenesis were first identified from forward genetic screens for mutants defective in plant development (19–22). One common feature of these mutants is the pleiotropic developmental defects such as delayed growth, abnormal leaves and flowers, and, in most cases, reduced fertility. In this study, we showed that DDL, whose loss-of-function mutations cause pleiotropic developmental defects (6), is also a player in miRNA biogenesis. However, it should be noted that DDL must have a broader role than one in miRNA biogenesis because ddl mutants have more severe morphological phenotypes than the dcl1-9 mutant whereas the reduction in miRNA levels is less in ddl mutants than in dcl1-9.How does DDL function in miRNA biogenesis? It appears that DDL acts differently from DCL1, SERRATE, and HYL1, which likely form a complex to process pri-miRNAs (23, 24). We found that the amount of pri-miRNAs in ddl mutants was reduced, whereas the levels of pri-miRNAs were shown to be increased in dcl1, hyl1, and serrate mutants (12, 25, 26). This argues against a role of DDL in the processing of pri-miRNAs. The reduced accumulation of pri-miRNAs in ddl mutants is not due to reduced transcription of MIR genes. We hypothesize that DDL promotes the access to, or recognition of, pri-miRNAs by DCL1 (Fig. 7C). Without DDL, a portion of pri-miRNAs cannot be properly channeled to DCL1 and is likely degraded, hence the reduced levels of pri-miRNAs in ddl mutants. The ability of DDL to interact with DCL1 and the fact that DDL binds RNA are consistent with, but are insufficient to prove, this hypothesis.DDL not only functions in miRNA biogenesis but also acts in siRNA biogenesis as evidenced by the reduction of ta-siRNAs and repeat-associated siRNAs in ddl mutants. Whether DDL is directly involved in ta-siRNA biogenesis is questionable because the biogenesis of ta-siRNAs requires miRNAs that are affected in ddl mutants.However, DDL clearly plays a role in the biogenesis of a DCL4-dependent miRNA and several DCL3-dependent siRNAs. How DDL acts in multiple DCL-dependent processes is currently unknown.DDL appears to be an evolutionarily conserved protein across plant and animal kingdoms. Because small RNAs are general regulators of gene expression in eukaryotes (reviewed in ref. 1), it is not surprising that some of the machinery involved in the biogenesis or function of small RNAs is present in both plants and animals. Examples include AGO, DICER, HEN1, and exportin 5. DDL can also be added to this list because SNIP1 and DDL show high degrees of sequence similarity, and siRNA-mediated knockdown of SNIP1 results in reduced accumulation of miRNAs in HeLa cells. The association of DDL with DCL1 and the interaction between SNIP1 and Drosha further support a conserved role of DDL and SNIP1 in miRNA biogenesis.Given that miRNAs in animals regulate large numbers of target genes and play crucial roles in various developmental processes, the fact that SNIP1 acts in miRNA biogenesis would be consistent with the described pleiotropic defects associated with SNIP1 knockdown or overexpression. It was reported that siRNA-mediated knockdown of SNIP1 in human cell lines results in reduced cell proliferation and cell-cycle arrest (27). Knocking down the C. elegans SNIP1 protein in a genome-wide RNAi screen results in embryonic lethality (www.wormbase.org). Injection of SNIP1 RNA into dorsal blastomeres of Xenopus embryos results in the suppression of dorsal mesoderm fate, which is specified by Nodal (a TGF-β ligand) and the TGF-β signaling pathway (8). Intriguingly, a recent study showed that miR-15 and miR-16 inhibit Nodal/TGF-β signaling by repressing the expression of the Nodal receptor ActRIIA in Xenopus (28). One plausible explanation for the inhibitory effect of SNIP1 on dorsal mesoderm fate specification (8) would be that overexpression of SNIP1 leads to increased levels of miR-15 and miR-16, which in turn inhibit Nodal/TGF-β signaling (28).Although we have uncovered a role of SNIP1 in miRNA biogenesis, SNIP has been found to interact with three transcription factors as well as with the transcriptional coactivator p300 (7–9). At present, it is not clear whether SNIP1 is a multifunctional protein involved in several molecular processes or whether the observed molecular functions of SNIP1 (interactions with p300 and promotion of miRNA accumulation) represent aspects of a unified function.Materials and MethodsPlant Materials.The ddl-1 and ddl-2 mutants and the ddl-1 mutant expressing a DDL genomic construct were reported by Morris et al. (6). To generate DDL+ (DDL/DDL or DDL/ddl-1) or ddl-1 plants harboring a single-locus GUS transgene, transgenic lines (in Ler background) containing a single locus of pMIR172a::GUS, pMIR172b::GUS, or pPISTILLATA::GUS (13) were crossed to ddl-1. In the F2 population, plants showing the WT (of the DDL/DDL or DDL/ddl-1 genotypes) and ddl-1 phenotypes were screened for the GUS transgene by GUS staining. To generate DDL+ (DDL/DDL or DDL/ddl-1) or ddl-1 plants expressing 35S::MIR172a, we crossed a transgenic line (in Ler background) containing a single locus of 35S::MIR172a (14) to ddl-1. F2 seeds were selected on Kanamycin medium for the presence of the transgene, and DDL+ (DDL/DDL or DDL/ddl-1) or ddl-1 plants were identified by their phenotypes.Plasmid Construction.The construction of GST-DDL, pMIR172a::GUS, and pMIR172b::GUS plasmids is described in SI Materials and Methods.RT-PCR Analysis of pri-miRNAs.cDNA was synthesized from 5 μg of total RNA by using reverse transcriptase (Invitrogen) and oligo(dT). Quantitative PCR was performed in triplicate on a Bio-Rad IQcycler apparatus with the Quantitech SYBR green kit (Bio-Rad). The primers used are listed in Table S1.RNA Analyses.RNA isolation and hybridization for miRNAs and endogenous siRNAs were performed as described (29). 5′-End-labeled 32P antisense DNA oligonucleotides or LNA oligonucleotides were used to detect miRNAs, ta-siRNAs, AtSN1 siRNA, siRNA02, and siRNA1003. For cluster 2 siRNAs, a DNA fragment was amplified from genomic DNA with forward and reverse primers (Table S1), gel-purified, and labeled by random priming. The detection of pre-miR166a and pre-miR164b was performed as described (12).GST-DDL RNA binding assays were performed as described (30, 31). A DNA fragment corresponding to pri-miR162b was amplified by PCR from genomic DNA with primers miR162bp1, which contained a T7 promoter, and miR162bp2 (Table S1). The resulting PCR product was purified and used as a template for in vitro transcription with T7 polymerase in the presence of [α-32P]UTP to generate pri-miR162b.Transient Expression in N. benthamiana and coIP.Transient expression of DCL1-YFP and HA-DDL was performed as described (32). For coIP between DDL and DCL1, the harvested leaves of N. benthamiana were ground in liquid nitrogen and homogenized in 3 vol of protein lysis buffer [50 mM Tris·HCl (pH 8.0), 150 mM NaCl, 0.2% Nonidet P-40, 2 mM DTT, 5% glycerol, complete proteinase inhibitor mixture (Roche)] and centrifuged for 15 min at 16,110 × g. After preclearing with protein-A agarose, half of the HA-DDL or HA-HEN1 lysate was mixed with the DCL1-YFP lysate. The mixed lysate or the HA-DDL or HA-HEN1 lysate alone was incubated with anti-GFP (and GFP variants) antibodies coupled to protein A agarose beads (Clontech) for 2 h. The immune complexes were then washed four times with 1 ml of lysis buffer. Proteins retained on the beads were resolved on SDS/polyacrylamide gels. Anti-YFP (Covance), anti-HA (Sigma-Aldrich), and anti-HEN1 antibodies were used to detect DCL1-YFP, HA-DDL, and HA-HEN1, respectively, by using Western blot analysis.siRNA-Mediated Knockdown of SNIP1.HeLa cells were transfected with siRNAs targeting SNIP1 or a nontargeting control siRNA (Table S1) (Dharmacon) by using Oligofectamine (Invitrogen). Cells were harvested for protein and RNA analyses 3 days after transfection. Anti-SNIP1 antibodies were purchased from Bethyl Laboratories.AcknowledgmentsWe thank Mahmoud El-Shami and Dominique Pontier for technical assistance; Julien Curaba, Theresa Dinh, and Shengben Li for comments on the manuscript; Drs. Popescu and Dinesh-Kumar for suggestions on transient expression of proteins in N. benthamiana; Dr. Shou-Wei Ding\'s laboratory for help with human cell culture; Drs. Lewis Bowman and Vicki Vance (University of South Carolina, Columbia, SC) for the DCL1 yeast two-hybrid plasmids; Drs. Richard Kim and Anita Roberts (National Cancer Institute, Bethesda, MD) for a HA-SNIP1 plasmid; Dr. Narry Kim (Seoul National University, Seoul, Korea) for a Drosha-Flag plasmid; Dr. Gregory Hannon for a Dicer plasmid; Dr. Brian Cullen (Duke University, Durham, NC) for an exportin 5 plasmid; and Dr. Li Zhao in the Chen laboratory for sharing pMIR172a::GUS and pMIR172b::GUS transgenic lines. This work was supported by National Institutes of Health Grant GM61146 (to X.C.); National Science Foundation Grants MCB-0718029 (to X.C.), MCB-0112278 (to J.C.W.), and MCB-0418946 (to J.C.W.); and Agence Nationale de Recherches Grant NT05-3_45717 (to T.L.). Funding was provided from the University of Missouri, the Food for the 21st Century Program (J.C.W.), and Centre National de la Recherche Scientifique.Footnotes‡‡To whom correspondence should be addressed. E-mail: xuemei.chen{at}ucr.eduAuthor contributions: B.Y. and X.C. designed research; B.Y., L.B., B.Z., L.J., M.A., and V.R. performed research; D.C., W.L., T.L., and J.C.W. contributed reagents/analytic tools; B.Y., L.B., and X.C. analyzed data; and B.Y. and X.C. wrote the paper.↵§Present address: Department of Biological Sciences, Mississippi State University, Mississippi State, MS 39762.↵¶Present address: Teri University, India Habitat Centre, Lodhi Road, New Delhi, India, 110003.The authors declare no conflict of interest.This article is a PNAS Direct Submission.This article contains supporting information online at www.pnas.org/cgi/content/full/0804218105/DCSupplemental.© 2008 by The National Academy of Sciences of the USA Bartel DP (2004) MicroRNAs: Genomics, biogenesis, mechanism, and function. Cell 116:281–297.OpenUrlCrossRefPubMed↵Chen X (2008) MicroRNA metabolism in plants. Curr Top Microbiol Immunol 320:117–136.OpenUrlCrossRefPubMed↵Yu B, et al. (2005) Methylation as a crucial step in plant microRNA biogenesis. Science 307:932–935.OpenUrlAbstract/FREE Full Text↵Vazquez F (2006) Arabidopsis endogenous small RNAs: Highways and byways. Trends Plants Sci 11:460–468.OpenUrlCrossRef↵Li J, Lee GI, Van Doren SR, Walker JC (2000) The FHA domain mediates phosphoprotein interactions. J Cell Sci 113:4143–4149.OpenUrlAbstract/FREE Full Text↵Morris ER, Chevalier D, Walker JC (2006) DAWDLE, a forkhead-associated domain gene, regulates multiple aspects of plant development. Plant Physiol 141:932–941.OpenUrlAbstract/FREE Full Text↵Kim RH, et al. (2001) SNIP1 inhibits NF-κ B signaling by competing for its binding to the C/H1 domain of CBP/p300 transcriptional coactivators. J Biol Chem 276:46297–46304.OpenUrlAbstract/FREE Full Text↵Kim RH, et al. (2000) A novel smad nuclear interacting protein, SNIP1, suppresses p300-dependent TGF-β signal transduction. Genes Dev 14:1605–1616.OpenUrlAbstract/FREE Full Text↵Fujii M, et al. (2006) SNIP1 is a candidate modifier of the transcriptional activity of c-Myc on E box-dependent target genes. Mol Cell 24:771–783.OpenUrlCrossRefPubMed↵Allen E, Xie Z, Gustafson AM, Carrington JC (2005) microRNA-directed phasing during trans-acting siRNA biogenesis in plants. Cell 121:207–221.OpenUrlCrossRefPubMed↵Rajagopalan R, Vaucheret H, Trejo J, Bartel DP (2006) A diverse and evolutionarily fluid set of microRNAs in Arabidopsis thaliana. Genes Dev 20:3407–3425.OpenUrlAbstract/FREE Full Text↵Kurihara Y, Takashi Y, Watanabe Y (2006) The interaction between DCL1 and HYL1 is important for efficient and precise processing of pri-miRNA in plant microRNA biogenesis. RNA 12:206–212.OpenUrlAbstract/FREE Full Text↵Chen X, Riechmann JL, Jia D, Meyerowitz EM (2000) The minimal regulatory region of the floral homeotic gene PISTILLATA does not contain a CArG box. Sexual Plant Reprod 13:85–94.OpenUrlCrossRef↵Chen X (2004) A microRNA as a translational repressor of APETALA2 in Arabidopsis flower development. Science 303:2022–2025.OpenUrlAbstract/FREE Full Text↵Voinnet O, Rivas S, Mestre P, Baulcombe D (2003) An enhanced transient expression system in plants based on suppression of gene silencing by the p19 protein of tomato bushy stunt virus. Plant J 33:949–956.OpenUrlCrossRefPubMed↵Xie Z, et al. (2005) Expression of Arabidopsis miRNA genes. Plant Physiol 138:2145–2154.OpenUrlAbstract/FREE Full Text↵Lee Y, et al. (2003) The nuclear RNase III Drosha initiates microRNA processing. Nature 425:415–419.OpenUrlCrossRefPubMed↵Yi R, Doehle BP, Qin Y, Macara IG, Cullen BR (2005) Overexpression of exportin 5 enhances RNA interference mediated by short hairpin RNAs and microRNAs. RNA 11:220–226.OpenUrlAbstract/FREE Full Text↵Chen X, Liu J, Cheng Y, Jia D (2002) HEN1 functions pleiotropically in Arabidopsis development and acts in C function in the flower. Development 129:1085–1094.OpenUrlAbstract/FREE Full Text↵Clarke JH, Tack D, Findlay K, Van Montagu M, Van Lijsebettens M (1999) The SERRATE locus controls the formation of the early juvenile leaves and phase length in Arabidopsis. Plant J 20:493–501.OpenUrlCrossRefPubMed↵Jacobsen SE, Running MP, Meyerowitz EM (1999) Disruption of an RNA helicase/RNase III gene in Arabidopsis causes unregulated cell division in floral meristems. Development 126:5231–5243.OpenUrlAbstract↵Lu C, Fedoroff N (2000) A mutation in the Arabidopsis HYL1 gene encoding a dsRNA binding protein affects responses to abscisic acid, auxin, and cytokinin. Plant Cell 12:2351–2366.OpenUrlAbstract/FREE Full Text↵Fang Y, Spector DL (2007) Identification of nuclear dicing bodies containing proteins for microRNA biogenesis in living Arabidopsis plants. Curr Biol 17:818–823.OpenUrlCrossRefPubMed↵Song L, Han MH, Lesicka J, Fedoroff N (2007) Arabidopsis primary microRNA processing proteins HYL1 and DCL1 define a nuclear body distinct from the Cajal body. Proc Natl Acad Sci USA 104:5437–5442.OpenUrlAbstract/FREE Full Text↵Lobbes D, Rallapalli G, Schmidt DD, Martin C, Clarke J (2006) SERRATE: A new player on the plant microRNA scene. EMBO Rep 7:1052–1058.OpenUrlCrossRefPubMed↵Yang L, Liu Z, Lu F, Dong A, Huang H (2006) SERRATE is a novel nuclear regulator in primary microRNA processing in Arabidopsis. Plant J 47:841–850.OpenUrlCrossRefPubMed↵Roche KC, Wiechens N, Owen-Hughes T, Perkins ND (2004) The FHA domain protein SNIP1 is a regulator of the cell cycle and cyclin D1 expression. Oncogene 23:8185–8195.OpenUrlCrossRefPubMed↵Martello G, et al. (2007) MicroRNA control of Nodal signalling. Nature 449:183–188.OpenUrlCrossRefPubMed↵Park W, Li J, Song R, Messing J, Chen X (2002) CARPEL FACTORY, a Dicer homolog, and HEN1, a novel protein, act in microRNA metabolism in Arabidopsis thaliana. Curr Biol 12:1484–1495.OpenUrlCrossRefPubMed↵Jiao X, Trifillis P, Kiledjian M (2002) Identification of target messenger RNA substrates for the murine deleted in azoospermia-like RNA-binding protein. Biol Reprod 66:475–485.OpenUrlAbstract/FREE Full Text↵Trifillis P, Day N, Kiledjian M (1999) Finding the right RNA: Identification of cellular mRNA substrates for RNA-binding proteins. RNA 5:1071–1082.OpenUrlAbstract↵Popescu SC, et al. (2007) Differential binding of calmodulin-related proteins to their targets revealed through high-density Arabidopsis protein microarrays. Proc Natl Acad Sci USA 104:4730–4735.OpenUrlAbstract/FREE Full Text Thank you for your interest in spreading the word on PNAS.NOTE: We only request your email address so that the person you are recommending the page to knows that you wanted them to see it, and that it is not junk mail. We do not capture any email address.CAPTCHAThis question is for testing whether or not you are a human visitor and to prevent automated spam submissions. The FHA domain proteins DAWDLE in Arabidopsis and SNIP1 in humans act in small RNA biogenesis Bin Yu, Liu Bi, Binglian Zheng, Lijuan Ji, David Chevalier, Manu Agarwal, Vanitharani Ramachandran, Wanxiang Li, Thierry Lagrange, John C. Walker, Xuemei Chen Proceedings of the National Academy of Sciences Jul 2008, 105 (29) 10073-10078; DOI: 10.1073/pnas.0804218105 The FHA domain proteins DAWDLE in Arabidopsis and SNIP1 in humans act in small RNA biogenesis Bin Yu, Liu Bi, Binglian Zheng, Lijuan Ji, David Chevalier, Manu Agarwal, Vanitharani Ramachandran, Wanxiang Li, Thierry Lagrange, John C. Walker, Xuemei Chen Proceedings of the National Academy of Sciences Jul 2008, 105 (29) 10073-10078; DOI: 10.1073/pnas.0804218105 Sign up for the PNAS Highlights newsletter to get in-depth stories of science sent to your inbox twice a month: Relatively clean snow and ice in the Indus River Basin during the COVID-19 pandemic may have reduced meltwater in 2020, compared with the 20-year average. Atmospheric and climate conditions could have created a cloud greenhouse effect to warm Mars and support liquid surface water. Researchers report a safety guideline to limit airborne transmission of COVID-19 that goes beyond the six-foot social distancing guideline. Interventions include using rice husks, manipulating paddy water and soil, and genetic changes that could stop arsenic from reaching the grain. Going beyond conventional approaches, researchers are using carefully cultured bacterial communities to improve sewage treatment.