- Email: [email protected]
miRNAs in the biogenesis of trans-acting siRNAs in higher plants
Seminars in Cell & Developmental Biology 21 (2010) 798–804
Contents lists available at ScienceDirect
Seminars in Cell & Developmental Biology journal homepage: www.elsevier.com/locate/semcdb
Review
miRNAs in the biogenesis of trans-acting siRNAs in higher plants Edwards Allen ∗ , Miya D. Howell Monsanto Company, 700 Chesterfield Pkwy W, Chesterfield, MO 63017, USA
a r t i c l e
i n f o
Article history: Available online 30 March 2010 Keywords: miRNA Trans-Acting siRNA Plants Gene regulation
a b s t r a c t Multicellular eukaryotes utilize many complex small RNA mechanisms to regulate gene expression from DNA modifications to RNA stability. RNA interference also regulates exogenous gene expression by degrading invading pathogen RNAs or preventing expression of foreign DNA incorporated into the host genome. Here we review the mechanisms for trans-acting (ta)-siRNA biogenesis and function, including pathways that utilize components of the miRNA and transitive RNAi defense. There are several distinguishing features of ta-siRNA pathways including the requirement for a miRNA-guided cleavage event that sets a processing register, RDR6 dependent dsRNA production, and DCL4 dependent processing to create unique, phased 21 nucleotide small RNAs. These phased small RNAs function to suppress target genes that only show similarity at the ta-siRNA recognition site, and act in trans to repress expression non-cell autonomously of specific target genes. Since the advent of high throughput sequencing technologies, phased siRNAs have been identified in a number of organisms [Heisel SE, Zhang Y, Allen E, Guo L, Reynolds TL, Yang X, et al. Characterization of unique small RNA populations from rice grain. PLoS One 2008;3:e2871. Zhao T, Li G, Mi S, Li S, Hannon GJ, Wang XJ, et al. A complex system of small RNAs in the unicellular green alga Chlamydomonas reinhardtii. Genes Dev 2007;21:1190–203. Johnson C, et al. Clusters and superclusters of phased small RNAs in the developing inflorescence of rice. Genome Res 2009;19:1429–40. Zhu QH, Spriggs A, Matthew L, Fan L, Kennedy G, Gubler F, et al. A diverse set of microRNAs and microRNA-like small RNAs in developing rice grains. Genome Res 2008;18:1456–65. Howell MD, Fahlgren N, Chapman EJ, Cumbie JS, Sullivan CM, Givan SA, et al. Genome-wide analysis of the RNA-DEPENDENT RNA POLYMERASE6/DICER-LIKE4 pathway in Arabidopsis reveals dependency on miRNA- and ta-siRNA-directed targeting. Plant Cell 2007;19:926–42.]. These include transcripts generated either from non-protein-coding or protein-coding transcripts, long imperfect dsRNA or through an unknown mechanism; therefore some of these may not necessarily be classified as canonical ta-siRNAs. © 2010 Elsevier Ltd. All rights reserved.
Contents 1.
2.
Trans-Acting siRNA discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1. Classification and function of miRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2. Biogenesis factors distinguish phased trans-acting siRNAs from miRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3. Revised miRNA and ta-siRNA classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mechanisms for biogenesis of trans-acting siRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Basic requirements for Arabidopsis trans-acting siRNA biogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. TAS1/2/4 single miRNA target model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1. miRNA cleavage sets the 5 position of ta-siRNA phasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2. Phased processing of ta-siRNAs by DICER-LIKE4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.3. Nomenclature for phased ta-siRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.4. Incorporation of processed ta-siRNAs into ARGONAUTE effector complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.5. Function of TAS1/2/4 ta-siRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. TAS3 dual miRNA target model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1. TAS3 loci utilize two conserved miR390 binding sites for function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2. ARGONAUTE7 is required for TAS3 ta-siRNA processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
∗ Corresponding author. Tel.: +1 636 737 6724. E-mail address: [email protected] (E. Allen). 1084-9521/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.semcdb.2010.03.008
799 799 799 799 800 800 800 800 800 800 800 800 801 801 801
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
3.
4.
Function of TAS3 trans-acting siRNAs in development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. tasiR-ARFs are mobile signals and establish a suppression gradient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. tasiR-ARFs are required for proper formation of leaf polarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Looking forward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Identification of new TAS loci . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Questions yet to be elucidated for ta-siRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Trans-Acting siRNA discovery 1.1. Classification and function of miRNAs Endogenous small RNAs in plants are generally classified into groups primarily based on factors required for their biogenesis and their function, including miRNAs, heterochromatin associated (hc)siRNAs, trans-acting (ta)-siRNAs, repeat associated (ra)-siRNAs and naturally occurring antisense (nat)-siRNAs. The workhorses of the group, miRNAs and hc-siRNAs, function to post-transcriptionally or transcriptionally regulate gene expression, respectively. Early studies associated plant miRNA function with regulation of many critical regulatory genes, primarily transcription factors required for proper patterning or timing of developmental processes [6–8]. miRNA genes are transcribed by RNA Pol II to generate imperfect self-complementary foldback structures that are subsequently processed by DICER-Like1 (DCL1) to generate double-stranded fragments, abrogating the need for an RNA-dependent RNA polymerase (RdRP) [9]. The resulting miRNA/miRNA* duplex fragments are approximately 21-nt in length and contain 2 nucleotide 3 overhangs [8,10]. The majority of miRNAs are recognized by the AGO1 effector complex due to their 5 terminal uracil specificity [11–13]. miRNAs target transcripts with little to no similarity outside the miRNA recognition site, and generally function to cleave transcripts between bases 10 and 11 relative to the 5 end of the miRNA, resulting in two specific RNA fragments that may subsequently undergo degradation by exoribonucleases [14–16]. Efficient cleavage requires association with AGO1 and near perfect complementarity of the miRNA to the target, although mispairs are generally tolerated towards the end of the miRNA:target duplex. Rules have been established that effectively predict miRNA targets [17–19]. miRNAs act cell autonomously, resulting in specific and localized gene regulation [6]. More recently, miRNAs have been shown to repress translation in plants [20,21]. 1.2. Biogenesis factors distinguish phased trans-acting siRNAs from miRNAs Previous guidelines for miRNA annotation required classification by DCL1 dependence, origination from an imperfect foldback structure, conservation, and specific targeting of an unrelated mRNA [22]. Early studies to identify miRNAs were limited by the availability of characterized silencing mutants, sequenced genomes and target prediction tools. As the small RNA sequence databases grew, a few sequences (miR175 and miR389) appeared to fit the accepted criteria for miRNAs, but were associated with abundant nearby siRNAs uncharacteristic of canonical miRNAs [8,23]. Similar to miRNAs, they targeted unrelated endogenous genes for suppression. Surprisingly, these siRNAs required additional factors, which should not be necessary for a sequence originating from an RNA foldback structure [18,24,25]. The precursor transcripts from which the siRNAs derived were elevated in sgs3 and rdr6 mutants, genes originally associated with viral defense (VIGS, virus induced gene silencing) and transgene silencing (PTGS, posttranscriptional gene silencing) [26]. Unlike small RNAs typical of PTGS, the siRNAs from these transcripts were in register with each
799
802 802 802 802 802 803 803
other, indicative of sequential processing. The key to this mystery came when miRNA target sites were identified [18,24,25,27], explaining the dependence on DCL1 for generation of an initiator miRNA. The components of PTGS/VIGS were required for secondary production of dsRNA after the miRNA cleavage event. Previous attempts to identify miRNA target sites from these transcripts had failed because these were non-coding non-annotated transcripts, or analysis did not allow for mispairing at critical sites (see mechanism section). In recent years, considerable advances have been made in understanding the biogenesis and function of ta-siRNAs. 1.3. Revised miRNA and ta-siRNA classification The current criteria for plant miRNA annotation now require precise excision from a stem–loop hairpin precursor. The criteria ensures the presence of a miRNA and miR* sequence on opposite stem-arms, their base-pairing interactions, and a low number of asymmetric bulges [28]. Aided by the availability of additional mutants in Arabidopsis thaliana, miRNAs are further distinguished by the lack of dependence on DCL4, RDR6, SGS3, or other components of the small RNA processing machinery (Box 1 ) [28]. As small RNA discovery expands to new organisms lacking annotation, it is critical that sequence information is not discarded when exploring small RNA function. Since the discovery of ta-siRNAs, several labs have developed models and algorithms to predict phased siR-
Box 1: Factors required for ta-siRNA function in plants RDR6 (RNA-dependent RNA polymerase6): First identified as a factor required for RNA-mediated virus induced silencing. Also known as SGS2 (suppressor of gene silencing2) and SDE1 (silencing defective1). Required for production of doublestranded RNA from a single-stranded precursor. rdr6 mutants have downward-curled and elongated leaves, with abaxial trichomes appearing early. DCL4 (DICER-like4): Processes phased 21 nucleotide small RNAs from dsRNA precursor. dcl4 mutants also display similar phenotypes to rdr6 mutants, leading to accelerated juvenile-toadult phase change in Arabidopsis. SGS3 (suppressor of gene silencing3)phosphate deficiency: SGS3 was identified in screens for mutations deficient for post-transcriptional gene silencing. Interacts and colocalizes with RDR6, and recently shown to bind 5 overhang containing dsRNA. Also shows potential role in de novo DNA methylation in Arabidopsis. Point mutation sgs33 mutant reduces ta-siRNA accumulation. AGO7 (argonaute7, zippy): Selectively binds miR390 for initiation of ta-siRNA biogenesis in TAS3 transcripts. Data suggests AGO7 is not required for initial cleavage but rather recruitment of RDR6. AGO7 mutants (zip-1) show similar rosette phenotypes to rdr6 and dcl4 mutants. DRB4 (DICER RNA binding factor4): Specifically interacts with DCL4. drb4 mutants phenocopy dcl4 mutants and display elongated and downwardly curled rosette leaves with increased anthocyanin. Mutants display reduced accumulation of both TAS1 and TAS3 ta-siRNAs.
800
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
NAs for screening small RNA libraries to differentiate miRNAs from phased siRNAs when classifying populations [1,3,5,29].
fate of the 5 fragment is likely degradation by an exonuclease [14], as no small RNAs have been identified from these sequences.
2. Mechanisms for biogenesis of trans-acting siRNAs
2.2.2. Phased processing of ta-siRNAs by DICER-LIKE4 Upon miRNA-guided cleavage of TAS transcripts, the synthesized dsRNA is processed by DCL4 in a phased, 21-nt register starting at the miRNA cleavage site. Processing may be limited either by the length of the dsRNA region, or the processivity of DCL4. Phase plots by Howell et al. indicate dominant phasing patterns from the miR173 cleavage site for all loci, but also reveal phase drift patterns in which small RNAs accumulate in the phase position one or two nucleotides from the original phase as early as cycle five [5]. The phase forward drift of ta-siRNA processing may be due to non-21-nt siRNA produced by a DCL other than DCL4, such as DCL2 [5]. In support of this hypothesis, in the absence of DCL4 activity, 22 and 24nt siRNAs are easily detected from TAS loci [38]. The limitation of phasing may be an evolutionary adaptation to prohibit extended signal amplification resulting in suppression of indirect target transcripts.
2.1. Basic requirements for Arabidopsis trans-acting siRNA biogenesis There are currently four TAS families classified in Arabidopsis computationally identified and validated [1–5,30]. These families all require specific miRNAs to guide cleavage as well as the following basic steps for biogenesis of ta-siRNAs: (1) ARGONAUTE directed miRNA-guided cleavage of a single-stranded RNA precursor transcript, (2) generation of double-stranded RNA by an RdRP, and (3) processing of the dsRNA from the miRNA-guided cleavage site to create unique phased 21-nt siRNAs. These four families can be separated by two classes: those that require one miRNA binding site (TAS1, TAS2, TAS4), and those that require two (TAS3) (Fig. 1). The TAS1/2/4 class and TAS3 class each require unique components and unique RNA structural features; therefore they will be discussed separately. 2.2. TAS1/2/4 single miRNA target model 2.2.1. miRNA cleavage sets the 5 position of ta-siRNA phasing The first characterized ta-siRNA families were TAS1 and TAS2 from Arabidopsis [18,24,25,27,31]. The TAS1 family consists of three loci, TAS1a (At2g27400), TAS1b (At1g50055) and TAS1c (At2g39675); TAS2 (At2g39681) is in close proximity to TAS1c on the second chromosome. Both TAS1 and TAS2 families require miR173 to guide cleavage of the transcript for ta-siRNA biogenesis [27,32,33]. The TAS1/2 primary transcripts are typical RNA Pol II polyadenylated and capped transcripts [25]. The longest potential open reading frames are short at under 100 amino acids raising the possibility that they may not be translated. Plant cells recognize non-protein-coding Pol II transcripts as ‘aberrant’, thus TAS precursors could be primed by the lack of a long ORF to enter an RNA silencing pathway. Cleavage by miR173 occurs at the canonical position between bases 10 and 11 in the target transcript [18,27,34]. miR173, like the majority of plant miRNAs, contains a 5 terminal uridine and associates primarily with AGO1 [11,13,35]. Unlike canonical miRNA targets, TAS1/2 transcripts when bound to miR173 create a mismatch at position 9 (TAS1a, -c, TAS2) or 9 and 10 (TAS1b). Mispairing at bases 9, 10, or 11 results in reduced or abolished cleavage activity [13,18,29]. Potentially the mispairing at or near the cleavage site in a single target transcript pauses the RISC and may recruit additional factors such as SGS3, required for ta-siRNA initiation. In the current model, SGS3 is necessary to stabilize the cleaved transcript followed by recruitment of RDR6, generating double-stranded RNA from the 3 fragment (Fig. 1). SGS3 has been shown to interact and colocalize with RDR6 in cytoplasmic granules [36]. The complementary strand is synthesized by RDR6 proceeding from the 3 poly(A) tail towards the cleavage site, thus it is unknown how potential recruitment of SGS3 and RDR6 at the cleavage site could direct synthesis initiated distally. Pairing ability of the miRNA to the transcript may affect how these transcripts are shunted through the ta-siRNA pathway; by altering the stability of the miRNA:target site, ribosomes may be cleared away allowing for entry of RDR6/SGS3. When the miR173 target site is substituted for a perfectly paired miRNA:target sequence (miR159, miR169, miR171, miR167), ta-siRNA production [32,33] is abolished. Likewise by creating mismatches between the miRNA:target site, the miRNA loses ability to guide cleavage and instead acts as a translational suppressor [37]. Factors that recruit RDR6 or RNA elements to direct initiation of second strand synthesis are not known. The
2.2.3. Nomenclature for phased ta-siRNAs ta-siRNAs have been generally named either by a unique sequence ID by the author, or using a standardized nomenclature in which the register is given a number D1, D2, . . . starting at the miRNA target site [18]. The processing direction is noted by a 5 or 3 prefix (3 in the case of TAS1/2); the orientation is indicated by adding the suffix [+] for the positive (original transcript) strand, or [–] for the negative (RDR generated) strand. 2.2.4. Incorporation of processed ta-siRNAs into ARGONAUTE effector complexes Incorporation of ta-siRNAs into AGO effector complexes follows the 5 base specificity rule, where ta-siRNAs possessing a 5 U are incorporated into AGO1 [11]. TAS1 loci code for multiple ta-siRNAs with very similar sequences predicted to target the same genes. The amplification of the silencing signal through production of multiple siRNAs may increase the efficacy of suppression, or establish strong gradients of target gene regulation (see Section 3). 2.2.5. Function of TAS1/2/4 ta-siRNAs It is noteworthy that at least one ta-siRNA produced from the TAS2 transcript targets two clusters of pentatricopeptide repeat gene transcripts (PPRs) on chromosome 1. Interestingly, these same transcripts also undergo phased siRNA biogenesis from not only TAS2 3 D6[–], but also by miR161.1 and miR161.2 and miR1427 in rice [1] (and possibly miR400, and other TAS1 and TAS2 ta-siRNAs that have yet to be experimentally validated). Arabidopsis contains at least 448 PPR-related genes, which code for putative RNA binding proteins [39,40]. The proliferation of this gene family is highly regulated through miRNA and ta-siRNA networks. It is likely the PPRs are targeted for suppression to quench rapid expansion of this gene family. This family, like other non-conserved and/or evolving miRNA targets may undergo a “birth and death” process as these genes show high sequence and location variability in orthologous, closely related regions of Brassica rapa and Arabidopsis [41]. Currently no other targets have been identified for TAS1 and TAS2 ta-siRNAs. A model has emerged that begins to explain the intricacies of ta-siRNA generation in Arabidopsis. Initiation of the each pathway starts by guided cleavage with a specific miRNA (Fig. 1). miR173 specifically targets TAS1 and TAS2 transcripts and does not appear to be conserved in other eukaryotes. In contrast, miR828 is conserved among plants and is specifically involved in regulation of MYB transcription factors, targeting both MYB113 and TAS4 transcripts. The resulting TAS4 ta-siRNAs in turn regulate MYB113 and
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
801
Fig. 1. TAS pathways in plants. Two pathways are known to generate ta-siRNAs in plants, pathway A in which one miRNA guides cleavage at the 5 end of the mRNA transcript, and pathway B in which two miRNA binding sites activate ta-siRNA production. An initial miRNA precursor is processed by DCL1 and the resulting miRNA strand guides cleavage of a non-protein-coding TAS transcript. In pathway A, miR173 or miR828 binds to the transcript and guides cleavage mediated by AGO1. RDR6 synthesizes a double-stranded RNA fragment that is subsequently processed by DCL4 into a phased, 21-nt register starting at the miRNA cleavage site. One strand of the resulting duplex then targets a complementary mRNA in trans. In pathway B, miR390 binds to the transcript at two sites. In Arabidopsis, the 3 miR390 site is cleaved by AGO7, while the 5 miR390 is required but not cleaved. RDR6 synthesizes a double-stranded RNA fragment that is subsequently processed by DCL4 into a phased, 21-nt register from the 3 miR390 cleavage site. One strand of the resulting duplex then targets a complementary mRNA in trans.
related family members. The features that define a trans-acting trigger miRNA have not been characterized, but they appear to be coded by the miRNA sequence, as substitution of an alternative miRNA sequence into the trigger precursor does not permit the alternative miRNA to generate ta-siRNAs [32]. Likewise, ta-siRNAs are produced from heterologous transcripts containing a miR173 target site indicating the miRNA sequence as the only cis-sequence requirement to trigger biogenesis of ta-siRNAs [33]. 2.3. TAS3 dual miRNA target model 2.3.1. TAS3 loci utilize two conserved miR390 binding sites for function The dual target model is currently unique to TAS3 loci. TAS3 transcripts and the trigger miR390 have been identified in mosses and many higher plants indicating a highly conserved mechanism for ta-siRNA generation [5,29,30]. Unlike the TAS1/2 and TAS4 families, TAS3 transcripts utilize two miR390 binding sites flanking the functional ta-siRNAs (Fig. 1). Three TAS3 loci have been identified to date in Arabidopsis: TAS3a (At3g17185), TAS3b (At5g49615), and TAS3c (At5g57735) [5]. In these transcripts, miR390 guides cleavage at the site 3 of the functional ta-siRNAs to set the processing register [18]. The functional ta-siRNAs (tasiARFs) are in phase and set by the 3 site, indicating a critical role for miR390
directed cleavage. In Arabidopsis, tasiARFs are located in positions 5 D7[+] and 5 D8[+], although the phase position varies among monocot plant species [18,29,42]. Therefore, the phase position of the tasiARFs appears not to be critical, rather only the miR390 binding sites and register relative to the 3 cleavage position. The 5 miR390 binding is mispaired with miR390 at critical positions for cleavage, resulting in a non-cleavable binding interaction between miR390 and TAS3. Studies from both Arabidopsis and moss indicate cleavage at the 5 miR390 binding site in TAS3 transcripts is undetectable [5,29]. Recently however, a novel miR390-dependent set of TAS transcripts have been identified in dicots in which the 5 miR390 binding site may undergo cleavage for initiation of processing [43]. 2.3.2. ARGONAUTE7 is required for TAS3 ta-siRNA processing TAS3 transcripts are unique from other non-coding TAS loci, not only in their dual miR390 specificity and direction of processing from the 3 site, but also in their requirement for AGO7. Unlike miR173 and most other miRNAs, miR390 is excluded from AGO1 because of its 5 terminal adenosine nucleotide [13]. This sorting mechanism allows miR390 to avoid the general miRNA 5 base criteria for AGO1 specificity. Based on 5 nucleotide specificity criteria, miR390 would be predicted to associate with primarily with AGO2 or AGO4 [11,12]; however sequencing of RNA-IP experiments
802
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
with HA-tagged AGO2 and AGO7 suggest a dominant AGO7:miR390 interaction [13]. It is not yet known what additional features determine why miR390 preferentially associates with AGO7 and not AGO2 or AGO4. What is the role of AGO7, and can other miRNAs substitute? The answer to this question comes from studies in which alternative miRNA binding sites were substituted for miR390 in cleavable or non-cleavable form, or where AGO specificity for miR390 was altered by substituting the 5 terminal adenine with uridine [13]. A modified miR390 containing a 5 terminal uridine associates with AGO1 and guides cleavage of a synthetic TAS transcript, yet processing of the TAS transcript and ta-siRNA biogenesis was abolished. In contrast, the 3 site is not dependent on AGO7 or miR390 specifically, only that a cleavable miRNA target is present. The 3 site can be substituted with an alternative miRNA such as miR171, resulting in efficient cleavage and ta-siRNA production. Any alterations to the 5 site by either removing miR390, AGO7, or making the site cleavable will disrupt ta-siRNA production [13,29]. Binding of miR390 and AGO7 in a complex to a non-cleavable 5 site are essential requirements for the dual target ta-siRNA model. 3. Function of TAS3 trans-acting siRNAs in development 3.1. tasiR-ARFs are mobile signals and establish a suppression gradient All TAS3 transcripts identified to date contain at least one conserved sequence when processed to mature ta-siRNAs (tasiR-ARFs) that can target Auxin Response Factor family members ARF1, 2, 3, or 4. In general, processing of angiosperm specific TAS3 loci results in two mature abundant tasiR-ARFs that are nearly identical in sequence and target specificity. ARF1 and 2 contain one tasiR-ARF target site, whereas ARF3 and 4 each contain two conserved target sites and are likely the primary targets of tasiR-ARFs [18,29,44–47]. Production of two or more nearly identical ta-siRNAs from a locus is a common feature for TAS1 and TAS3, and may be a mechanism to further amplify a suppression signal while maintaining target specificity. In addition, production of multiple ta-siRNAs could strengthen the suppression gradient developed by ta-siRNAs. Studies have shown that ta-siRNAs are mobile, and can travel across multiple cells to create a gradient of suppression activity [48,49]. When the ta-siRNA source cell is strong, a steeper gradient may be created [50]. 3.2. tasiR-ARFs are required for proper formation of leaf polarity TAS3 ta-siRNAs, or tasiR-ARFs, suppress the juvenile-to-adult phase transition in Arabidopsis through negative regulation of ARF3/ETTIN [45], and are required for proper formation of leaf development by establishing adaxial/abaxial (dorsoventral) polarity. Similarly to rdr6, dcl4, and ago7 mutants, ARF3mut plants (non-targeted ARF3) also display phase change phenotypes. Further evidence for a role in phase transition is shown in both ARF3 overexpression and ARF3mut plants in an rdr6 background; plants exhibit an increased blade/petiole leaf length, an earlier appearance of abaxial trichomes and strong downward leaf curvature indicating severe morphological and patterning defects [45]. These data indicate an ARF3 dosage dependent accelerated phase change phenotype in ta-siRNA deficient mutants. Furthermore, tasiR-ARFs move intercellularly from the adaxial (upper) to abaxial (lower) side of leaf priomordia to create a gradient that patterns ARF3 expression [48]. These leaf patterning phenotypes are dependent on the polarization of miR390 by restricted AGO7 expression. Antagonistic activities of miR166 on class III homeodomain leucine zipper (HD-ZIPIII) transcription factors also contribute to speci-
fication of adaxial cell fate as tasiR-ARFs restrict the expression domain of miR166 in corn and Arabidopsis [50–52]. Nogueira and colleagues show leaf bladeless1 (lbl1), whose function is necessary for TAS3 ta-siRNAs, leads to ectopic accumulation of miR166 resulting in specification of leaf polarity by the ta-siRNA pathway [53,54]. These and other antagonistic interactions between polarity determinants help to establish dorsoventral patterning within developing organs.
4. Looking forward 4.1. Identification of new TAS loci Additional validated ta-siRNA loci have been scarce, therefore it remains to be seen how broadly this mechanism is utilized by plants. Algorithms are available to assist in prediction of phased ta-siRNA loci, and can be used in combination with existing miRNA target prediction rules [1,5,19,55]. Each prediction algorithm is unique, yet they share common features including a requirement for multiple phases to contain siRNAs, and a predominant 21-nt phasing pattern. Additional evidence is supplied by siRNA sequence depletion in RDR6, SGS3, or DCL4 mutants where available. Since the first publications in 2004 on the ta-siRNA pathway in Arabidopsis, only two new TAS loci, AtTAS4 and PpTAS4, have been validated [30,56,57]. In 2006, an additional transcript was identified in Arabidopsis by Rajagopalan et al. and named AtTAS4 [56]. Unlike the TAS1/2 families which require miR173 as the initiator, TAS4 requires miR828 to guide cleavage in an AGO1 dependent manner. This family was not identified until the advent of high throughput sequencing technologies, since the ta-siRNAs identified from the TAS4 transcript show low expression activity. Recently, phosphate starvation studies uncovered a role for TAS4 in regulating the biosynthesis of anthocyanin [57]. Both TAS4 ta-siRNAs and miR828 were upregulated in shoots by Pi and nitrogen deficiency. TAS4-siR81(−) targets several MYB transcription factors including PAP1/MYB75, PAP2/MYB90, and MYB113 involved in biosynthesis of anthocyanin [56]. In addition, both MIR828 and TAS4 genes possess multiple MYB binding sites and a PAP1 cis-regulatory element, resulting in a feedback mechanism for regulation of anthocyanin accumulation during stress. An autoregulatory mechanism was uncovered in which phosphate deficiency resulted in both upregulation of the MYB transcription factors activating anthocyanin biosynthesis, as well as increasing production of ta-siRNAs via the activation of miR828 and/or TAS4. A more extensive search for RDR6/DCL4 dependent loci revealed the existence of several protein-coding loci which behave similarly to TAS loci. Among these are several protein-coding loci that generate phased siRNAs including a CC-NBS-LRR domain targeted by miR472, and transcripts targeted by miR393 [5]. miR472 is conserved in poplar (miRBASE) and several crop plants, as are phased siRNAs generated from orthologous transcripts (authors, unpublished data). These miRNA:target transcripts likely behave similarly to TAS transcripts in that the siRNAs generated from these transcripts target other mRNAs in trans, however they are proteincoding, a characteristic atypical for ta-siRNA loci. In Physcomitrella patens, a phased siRNA generating locus named PpTAS4 has been validated, indicating that ta-siRNA mechanisms are present in primitive species [30,57]. In addition, several loci that form internal foldback structures have been found that produce phased siRNAs in the absence of a miRNA initiator [1,3,4]. These phased siRNAs are likely early intermediates in the evolution of miRNA [9]. Because they do not require a miRNA, they are not considered canonical tasiRNAs although they can processed by DCL4 and target distantly related transcripts.
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
4.2. Questions yet to be elucidated for ta-siRNAs Several questions regarding this specialized small RNA pathway were posed by Vaucheret [31] that we can now answer, although a few details remain unanswered and more questions have arisen. Generation of ta-siRNAs is a highly effective method of suppression for plants using a steep gradient of gene expression (ARFs), or for targeting expanded gene families for repression (PPRs). The tasiRNA pathway could be a mechanism to control and amplify small RNA production while converting a cell autonomous miRNA signal to a non-cell autonomous ta-siRNA signal (in concert with DCL4) for creating strong gradients across cells [58]. It appears generation of all ta-siRNAs requires the activity of RDR6 and DCL4, but it remains a question on whether all phased siRNAs (for example, proteincoding transcript targets and rice siRNA clusters) also require these or orthologous protein components within their silencing machinery. The two variations of single or dual targeting also require different AGO family members, which may aid in determining the directionality of RDR6 dependent dsRNA production. It is still unclear how ta-siRNA transcripts utilize an RdRP for dsRNA amplification, and whether all siRNAs generated from these transcripts are functional. How do plants differentiate a miRNA targeted for degradation from one directed for dsRNA synthesis? Could this be related to pausing of the AGO-RISC complex as a result of a second non-cleavable target, or from a mismatched target site that reduces miRNA activity? In fact, a subset of miRNA targets generate secondary siRNAs, although at a reduced level compared to ta-siRNAs [59]. Finally, how widely do plants utilize the ta-siRNA mechanism? We anticipate answers to these questions as more plant genomes are sequences and small RNA analyses are applied.
References [1] Heisel SE, Zhang Y, Allen E, Guo L, Reynolds TL, Yang X, et al. Characterization of unique small RNA populations from rice grain. PLoS One 2008;3:e2871. [2] Zhao T, Li G, Mi S, Li S, Hannon GJ, Wang XJ, et al. A complex system of small RNAs in the unicellular green alga Chlamydomonas reinhardtii. Genes Dev 2007;21:1190–203. [3] Johnson C, Kasprzewska A, Tennessen K, Fernandes J, Nan GL, Walbot V, et al. Clusters and superclusters of phased small RNAs in the developing inflorescence of rice. Genome Res 2009;19:1429–40. [4] Zhu QH, Spriggs A, Matthew L, Fan L, Kennedy G, Gubler F, et al. A diverse set of microRNAs and microRNA-like small RNAs in developing rice grains. Genome Res 2008;18:1456–65. [5] Howell MD, Fahlgren N, Chapman EJ, Cumbie JS, Sullivan CM, Givan SA, et al. Genome-wide analysis of the RNA-DEPENDENT RNA POLYMERASE6/DICERLIKE4 pathway in Arabidopsis reveals dependency on miRNA- and tasiRNAdirected targeting. Plant Cell 2007;19:926–42. [6] Palatnik JF, Allen E, Wu X, Schommer C, Schwab R, Carrington JC, et al. Control of leaf morphogenesis by microRNAs. Nature 2003;425:257–63. [7] Rhoades MW, Reinhart BJ, Lim LP, Burge CB, Bartel B, Bartel DP. Prediction of plant microRNA targets. Cell 2002;110:513–20. [8] Park W, Li J, Song R, Messing J, Chen X. CARPEL FACTORY, a Dicer homolog, and HEN1, a novel protein, act in microRNA metabolism in Arabidopsis thaliana. Curr Biol 2002;12:1484–95. [9] Voinnet O. Origin, biogenesis, and activity of plant microRNAs. Cell 2009;136:669–87. [10] Reinhart BJ, Weinstein EG, Rhoades MW, Bartel B, Bartel DP. MicroRNAs in plants. Genes Dev 2002;16:1616–26. [11] Mi S, Cai T, Hu Y, Chen Y, Hodges E, Ni F, et al. Sorting of small RNAs into Arabidopsis argonaute complexes is directed by the 5 terminal nucleotide. Cell 2008;133:116–27. [12] Takeda A, Iwasaki S, Watanabe T, Utsumi M, Watanabe Y. The mechanism selecting the guide strand from small RNA duplexes is different among argonaute proteins. Plant Cell Physiol 2008;49:493–500. [13] Montgomery TA, Howell MD, Cuperus JT, Li D, Hansen JE, Alexander AL, et al. Specificity of ARGONAUTE7-miR390 interaction and dual functionality in TAS3 trans-acting siRNA formation. Cell 2008;133:128–41. [14] Souret FF, Kastenmayer JP, Green PJ. AtXRN4 degrades mRNA in Arabidopsis and its substrates include selected miRNA targets. Mol Cell 2004;15:173–83. [15] Ramachandran V, Chen X. Degradation of microRNAs by a family of exoribonucleases in Arabidopsis. Science 2008;321:1490–2. [16] Gy I, Gasciolli V, Lauressergues D, Morel JB, Gombert J, Proux F, et al. Arabidopsis FIERY1, XRN2, and XRN3 are endogenous RNA silencing suppressors. Plant Cell 2007;19:3451–61.
803
[17] Fahlgren N, Carrington JC. miRNA target prediction in plants. Methods Mol Biol 2010;592:51–7. [18] Allen E, Xie Z, Gustafson AM, Carrington JC. microRNA-directed phasing during trans-acting siRNA biogenesis in plants. Cell 2005;121:207–21. [19] Jones-Rhoades MW, Bartel DP. Computational identification of plant microRNAs and their targets, including a stress-induced miRNA. Mol Cell 2004;14:787–99. [20] Lanet E, Delannoy E, Sormani R, Floris M, Brodersen P, Crete P, et al. Biochemical evidence for translational repression by Arabidopsis microRNAs. Plant Cell 2009;21:1762–8. [21] Brodersen P, Sakvarelidze-Achard L, Bruun-Rasmussen M, Dunoyer P, Yamamoto YY, Sieburth L, et al. Widespread translational inhibition by plant miRNAs and siRNAs. Science 2008;320:1185–90. [22] Ambros V, Bartel B, Bartel DP, Burge CB, Carrington JC, Chen X, et al. A uniform system for microRNA annotation. RNA 2003;9:277–9. [23] Sunkar R, Zhu JK. Novel and stress-regulated microRNAs and other small RNAs from Arabidopsis. Plant Cell 2004;16:2001–19. [24] Peragine A, Yoshikawa M, Wu G, Albrecht HL, Poethig RS. SGS3 and SGS2/SDE1/RDR6 are required for juvenile development and the production of trans-acting siRNAs in Arabidopsis. Genes Dev 2004;18:2368–79. [25] Vazquez F, Vaucheret H, Rajagopalan R, Lepers C, Gasciolli V, Mallory AC, et al. Endogenous trans-acting siRNAs regulate the accumulation of Arabidopsis mRNAs. Mol Cell 2004;16:69–79. [26] Mourrain P, Beclin C, Elmayan T, Feuerbach F, Godon C, Morel JB, et al. Arabidopsis SGS2 and SGS3 genes are required for posttranscriptional gene silencing and natural virus resistance. Cell 2000;101:533–42. [27] Yoshikawa M, Peragine A, Park MY, Poethig RS. A pathway for the biogenesis of trans-acting siRNAs in Arabidopsis. Genes Dev 2005;19:2164–75. [28] Meyers BC, Axtell MJ, Bartel B, Bartel DP, Baulcombe D, Bowman JL, et al. Criteria for annotation of plant MicroRNAs. Plant Cell 2008;20:3186–90. [29] Axtell MJ, Jan C, Rajagopalan R, Bartel DP. A two-hit trigger for siRNA biogenesis in plants. Cell 2006;127:565–77. [30] Talmor-Neiman M, Stav R, Klipcan L, Buxdorf K, Baulcombe DC, Arazi T. Identification of trans-acting siRNAs in moss and an RNA-dependent RNA polymerase required for their biogenesis. Plant J 2006;48:511–21. [31] Vaucheret H. MicroRNA-dependent trans-acting siRNA production. Sci STKE 2005;2005:pe43. [32] Montgomery TA, Yoo SJ, Fahlgren N, Gilbert SD, Howell MD, Sullivan CM, et al. AGO1-miR173 complex initiates phased siRNA formation in plants. Proc Natl Acad Sci USA 2008;105:20055–62. [33] Felippes FF, Weigel D. Triggering the formation of tasiRNAs in Arabidopsis thaliana: the role of microRNA miR173. EMBO Rep 2009;10:264–70. [34] Gasciolli V, Mallory AC, Bartel DP, Vaucheret H. Partially redundant functions of Arabidopsis DICER-like enzymes and a role for DCL4 in producing trans-acting siRNAs. Curr Biol 2005;15:1494–500. [35] Baumberger N, Baulcombe DC. Arabidopsis ARGONAUTE1 is an RNA Slicer that selectively recruits microRNAs and short interfering RNAs. Proc Natl Acad Sci USA 2005;102:11928–33. [36] Kumakura N, Takeda A, Fujioka Y, Motose H, Takano R, Watanabe Y. SGS3 and RDR6 interact and colocalize in cytoplasmic SGS3/RDR6-bodies. FEBS Lett 2009;583:1261–6. [37] Dugas DV, Bartel B. Sucrose induction of Arabidopsis miR398 represses two Cu/Zn superoxide dismutases. Plant Mol Biol 2008;67:403–17. [38] Xie Z, Allen E, Wilken A, Carrington JC. DICER-LIKE 4 functions in trans-acting small interfering RNA biogenesis and vegetative phase change in Arabidopsis thaliana. Proc Natl Acad Sci USA 2005;102:12984–9. [39] Saha D, Prasad AM, Srinivasan R. Pentatricopeptide repeat proteins and their emerging roles in plants. Plant Physiol Biochem 2007;45:521–34. [40] Lurin C, Andres C, Aubourg S, Bellaoui M, Bitton F, Bruyere C, et al. Genomewide analysis of Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis. Plant Cell 2004;16:2089–103. [41] Geddy R, Brown GG. Genes encoding pentatricopeptide repeat (PPR) proteins are not conserved in location in plant genomes and may be subject to diversifying selection. BMC Genom 2007;8:130. [42] Axtell MJ, Snyder JA, Bartel DP. Common functions for diverse small RNAs of land plants. Plant Cell 2007;19:1750–69. [43] Krasnikova MS, Milyutina IA, Bobrova VK, Ozerova LV, Troitsky AV, Solovyev AG, et al. Novel miR390-Dependent Transacting siRNA precursors in plants revealed by a PCR-based experimental approach and database analysis. J Biomed Biotechnol 2009;2009:952304. [44] Garcia D, Collier SA, Byrne ME, Martienssen RA. Specification of leaf polarity in Arabidopsis via the trans-acting siRNA pathway. Curr Biol 2006;16: 933–8. [45] Fahlgren N, Montgomery TA, Howell MD, Allen E, Dvorak SK, Alexander AL, et al. Regulation of AUXIN RESPONSE FACTOR3 by TAS3 ta-siRNA affects developmental timing and patterning in Arabidopsis. Curr Biol 2006;16:939–44. [46] Adenot X, Elmayan T, Lauressergues D, Boutet S, Bouche N, Gasciolli V, et al. DRB4-dependent TAS3 trans-acting siRNAs control leaf morphology through AGO7. Curr Biol 2006;16:927–32. [47] Williams L, Carles CC, Osmont KS, Fletcher JC. A database analysis method identifies an endogenous trans-acting short-interfering RNA that targets the Arabidopsis ARF2, ARF3, and ARF4 genes. Proc Natl Acad Sci USA 2005;102:9703–8. [48] Chitwood DH, Nogueira FT, Howell MD, Montgomery TA, Carrington JC, Timmermans MC. Pattern formation via small RNA mobility. Genes Dev 2009;23:549–54.
804
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
[49] Schwab R, Maizel A, Ruiz-Ferrer V, Garcia D, Bayer M, Crespi M, et al. Endogenous TasiRNAs mediate non-cell autonomous effects on gene regulation in Arabidopsis thaliana. PLoS One 2009;4:e5980. [50] Husbands AY, Chitwood DH, Plavskin Y, Timmermans MC. Signals and prepatterns: new insights into organ polarity in plants. Genes Dev 2009;23:1986–97. [51] Nogueira FT, Chitwood DH, Madi S, Ohtsu K, Schnable PS, Scanlon MJ, et al. Regulation of small RNA accumulation in the maize shoot apex. PLoS Genet 2009;5:e1000320. [52] Nogueira FT, Madi S, Chitwood DH, Juarez MT, Timmermans MC. Two small regulatory RNAs establish opposing fates of a developmental axis. Genes Dev 2007;21:750–5. [53] Nogueira F, Timmermans MC. An interplay between small regulatory RNAs patterns leaves. Plant Signal Behav 2007;2:519–21. [54] Nogueira FT, Sarkar AK, Chitwood DH, Timmermans MC. Organ polarity in plants is specified through the opposing activity of two distinct small regulatory RNAs. Cold Spring Harb Symp Quant Biol 2006;71:157–64.
[55] Chen HM, Li YH, Wu SH. Bioinformatic prediction and experimental validation of a microRNA-directed tandem trans-acting siRNA cascade in Arabidopsis. Proc Natl Acad Sci USA 2007;104:3318–23. [56] Rajagopalan R, Vaucheret H, Trejo J, Bartel DP. A diverse and evolutionarily fluid set of microRNAs in Arabidopsis thaliana. Genes Dev 2006;20: 3407–25. [57] Hsieh LC, Lin SI, Shih AC, Chen JW, Lin WY, Tseng CY, et al. Uncovering small RNA-mediated responses to phosphate-deficiency in Arabidopsis by deep sequencing. Plant Physiol 2009. [58] Voinnet O. Use, tolerance and avoidance of amplified RNA silencing by plants. Trends Plant Sci 2008;13:317–28. [59] Ronemus M, Vaughn MW, Martienssen RA. MicroRNA-targeted and small interfering RNA-mediated mRNA degradation is regulated by argonaute, dicer, and RNA-dependent RNA polymerase in Arabidopsis. Plant Cell 2006;18: 1559–74.
Contents lists available at ScienceDirect
Seminars in Cell & Developmental Biology journal homepage: www.elsevier.com/locate/semcdb
Review
miRNAs in the biogenesis of trans-acting siRNAs in higher plants Edwards Allen ∗ , Miya D. Howell Monsanto Company, 700 Chesterfield Pkwy W, Chesterfield, MO 63017, USA
a r t i c l e
i n f o
Article history: Available online 30 March 2010 Keywords: miRNA Trans-Acting siRNA Plants Gene regulation
a b s t r a c t Multicellular eukaryotes utilize many complex small RNA mechanisms to regulate gene expression from DNA modifications to RNA stability. RNA interference also regulates exogenous gene expression by degrading invading pathogen RNAs or preventing expression of foreign DNA incorporated into the host genome. Here we review the mechanisms for trans-acting (ta)-siRNA biogenesis and function, including pathways that utilize components of the miRNA and transitive RNAi defense. There are several distinguishing features of ta-siRNA pathways including the requirement for a miRNA-guided cleavage event that sets a processing register, RDR6 dependent dsRNA production, and DCL4 dependent processing to create unique, phased 21 nucleotide small RNAs. These phased small RNAs function to suppress target genes that only show similarity at the ta-siRNA recognition site, and act in trans to repress expression non-cell autonomously of specific target genes. Since the advent of high throughput sequencing technologies, phased siRNAs have been identified in a number of organisms [Heisel SE, Zhang Y, Allen E, Guo L, Reynolds TL, Yang X, et al. Characterization of unique small RNA populations from rice grain. PLoS One 2008;3:e2871. Zhao T, Li G, Mi S, Li S, Hannon GJ, Wang XJ, et al. A complex system of small RNAs in the unicellular green alga Chlamydomonas reinhardtii. Genes Dev 2007;21:1190–203. Johnson C, et al. Clusters and superclusters of phased small RNAs in the developing inflorescence of rice. Genome Res 2009;19:1429–40. Zhu QH, Spriggs A, Matthew L, Fan L, Kennedy G, Gubler F, et al. A diverse set of microRNAs and microRNA-like small RNAs in developing rice grains. Genome Res 2008;18:1456–65. Howell MD, Fahlgren N, Chapman EJ, Cumbie JS, Sullivan CM, Givan SA, et al. Genome-wide analysis of the RNA-DEPENDENT RNA POLYMERASE6/DICER-LIKE4 pathway in Arabidopsis reveals dependency on miRNA- and ta-siRNA-directed targeting. Plant Cell 2007;19:926–42.]. These include transcripts generated either from non-protein-coding or protein-coding transcripts, long imperfect dsRNA or through an unknown mechanism; therefore some of these may not necessarily be classified as canonical ta-siRNAs. © 2010 Elsevier Ltd. All rights reserved.
Contents 1.
2.
Trans-Acting siRNA discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1. Classification and function of miRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2. Biogenesis factors distinguish phased trans-acting siRNAs from miRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3. Revised miRNA and ta-siRNA classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mechanisms for biogenesis of trans-acting siRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Basic requirements for Arabidopsis trans-acting siRNA biogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. TAS1/2/4 single miRNA target model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1. miRNA cleavage sets the 5 position of ta-siRNA phasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2. Phased processing of ta-siRNAs by DICER-LIKE4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.3. Nomenclature for phased ta-siRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.4. Incorporation of processed ta-siRNAs into ARGONAUTE effector complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.5. Function of TAS1/2/4 ta-siRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. TAS3 dual miRNA target model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1. TAS3 loci utilize two conserved miR390 binding sites for function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2. ARGONAUTE7 is required for TAS3 ta-siRNA processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
∗ Corresponding author. Tel.: +1 636 737 6724. E-mail address: [email protected] (E. Allen). 1084-9521/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.semcdb.2010.03.008
799 799 799 799 800 800 800 800 800 800 800 800 801 801 801
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
3.
4.
Function of TAS3 trans-acting siRNAs in development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. tasiR-ARFs are mobile signals and establish a suppression gradient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. tasiR-ARFs are required for proper formation of leaf polarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Looking forward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Identification of new TAS loci . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Questions yet to be elucidated for ta-siRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Trans-Acting siRNA discovery 1.1. Classification and function of miRNAs Endogenous small RNAs in plants are generally classified into groups primarily based on factors required for their biogenesis and their function, including miRNAs, heterochromatin associated (hc)siRNAs, trans-acting (ta)-siRNAs, repeat associated (ra)-siRNAs and naturally occurring antisense (nat)-siRNAs. The workhorses of the group, miRNAs and hc-siRNAs, function to post-transcriptionally or transcriptionally regulate gene expression, respectively. Early studies associated plant miRNA function with regulation of many critical regulatory genes, primarily transcription factors required for proper patterning or timing of developmental processes [6–8]. miRNA genes are transcribed by RNA Pol II to generate imperfect self-complementary foldback structures that are subsequently processed by DICER-Like1 (DCL1) to generate double-stranded fragments, abrogating the need for an RNA-dependent RNA polymerase (RdRP) [9]. The resulting miRNA/miRNA* duplex fragments are approximately 21-nt in length and contain 2 nucleotide 3 overhangs [8,10]. The majority of miRNAs are recognized by the AGO1 effector complex due to their 5 terminal uracil specificity [11–13]. miRNAs target transcripts with little to no similarity outside the miRNA recognition site, and generally function to cleave transcripts between bases 10 and 11 relative to the 5 end of the miRNA, resulting in two specific RNA fragments that may subsequently undergo degradation by exoribonucleases [14–16]. Efficient cleavage requires association with AGO1 and near perfect complementarity of the miRNA to the target, although mispairs are generally tolerated towards the end of the miRNA:target duplex. Rules have been established that effectively predict miRNA targets [17–19]. miRNAs act cell autonomously, resulting in specific and localized gene regulation [6]. More recently, miRNAs have been shown to repress translation in plants [20,21]. 1.2. Biogenesis factors distinguish phased trans-acting siRNAs from miRNAs Previous guidelines for miRNA annotation required classification by DCL1 dependence, origination from an imperfect foldback structure, conservation, and specific targeting of an unrelated mRNA [22]. Early studies to identify miRNAs were limited by the availability of characterized silencing mutants, sequenced genomes and target prediction tools. As the small RNA sequence databases grew, a few sequences (miR175 and miR389) appeared to fit the accepted criteria for miRNAs, but were associated with abundant nearby siRNAs uncharacteristic of canonical miRNAs [8,23]. Similar to miRNAs, they targeted unrelated endogenous genes for suppression. Surprisingly, these siRNAs required additional factors, which should not be necessary for a sequence originating from an RNA foldback structure [18,24,25]. The precursor transcripts from which the siRNAs derived were elevated in sgs3 and rdr6 mutants, genes originally associated with viral defense (VIGS, virus induced gene silencing) and transgene silencing (PTGS, posttranscriptional gene silencing) [26]. Unlike small RNAs typical of PTGS, the siRNAs from these transcripts were in register with each
799
802 802 802 802 802 803 803
other, indicative of sequential processing. The key to this mystery came when miRNA target sites were identified [18,24,25,27], explaining the dependence on DCL1 for generation of an initiator miRNA. The components of PTGS/VIGS were required for secondary production of dsRNA after the miRNA cleavage event. Previous attempts to identify miRNA target sites from these transcripts had failed because these were non-coding non-annotated transcripts, or analysis did not allow for mispairing at critical sites (see mechanism section). In recent years, considerable advances have been made in understanding the biogenesis and function of ta-siRNAs. 1.3. Revised miRNA and ta-siRNA classification The current criteria for plant miRNA annotation now require precise excision from a stem–loop hairpin precursor. The criteria ensures the presence of a miRNA and miR* sequence on opposite stem-arms, their base-pairing interactions, and a low number of asymmetric bulges [28]. Aided by the availability of additional mutants in Arabidopsis thaliana, miRNAs are further distinguished by the lack of dependence on DCL4, RDR6, SGS3, or other components of the small RNA processing machinery (Box 1 ) [28]. As small RNA discovery expands to new organisms lacking annotation, it is critical that sequence information is not discarded when exploring small RNA function. Since the discovery of ta-siRNAs, several labs have developed models and algorithms to predict phased siR-
Box 1: Factors required for ta-siRNA function in plants RDR6 (RNA-dependent RNA polymerase6): First identified as a factor required for RNA-mediated virus induced silencing. Also known as SGS2 (suppressor of gene silencing2) and SDE1 (silencing defective1). Required for production of doublestranded RNA from a single-stranded precursor. rdr6 mutants have downward-curled and elongated leaves, with abaxial trichomes appearing early. DCL4 (DICER-like4): Processes phased 21 nucleotide small RNAs from dsRNA precursor. dcl4 mutants also display similar phenotypes to rdr6 mutants, leading to accelerated juvenile-toadult phase change in Arabidopsis. SGS3 (suppressor of gene silencing3)phosphate deficiency: SGS3 was identified in screens for mutations deficient for post-transcriptional gene silencing. Interacts and colocalizes with RDR6, and recently shown to bind 5 overhang containing dsRNA. Also shows potential role in de novo DNA methylation in Arabidopsis. Point mutation sgs33 mutant reduces ta-siRNA accumulation. AGO7 (argonaute7, zippy): Selectively binds miR390 for initiation of ta-siRNA biogenesis in TAS3 transcripts. Data suggests AGO7 is not required for initial cleavage but rather recruitment of RDR6. AGO7 mutants (zip-1) show similar rosette phenotypes to rdr6 and dcl4 mutants. DRB4 (DICER RNA binding factor4): Specifically interacts with DCL4. drb4 mutants phenocopy dcl4 mutants and display elongated and downwardly curled rosette leaves with increased anthocyanin. Mutants display reduced accumulation of both TAS1 and TAS3 ta-siRNAs.
800
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
NAs for screening small RNA libraries to differentiate miRNAs from phased siRNAs when classifying populations [1,3,5,29].
fate of the 5 fragment is likely degradation by an exonuclease [14], as no small RNAs have been identified from these sequences.
2. Mechanisms for biogenesis of trans-acting siRNAs
2.2.2. Phased processing of ta-siRNAs by DICER-LIKE4 Upon miRNA-guided cleavage of TAS transcripts, the synthesized dsRNA is processed by DCL4 in a phased, 21-nt register starting at the miRNA cleavage site. Processing may be limited either by the length of the dsRNA region, or the processivity of DCL4. Phase plots by Howell et al. indicate dominant phasing patterns from the miR173 cleavage site for all loci, but also reveal phase drift patterns in which small RNAs accumulate in the phase position one or two nucleotides from the original phase as early as cycle five [5]. The phase forward drift of ta-siRNA processing may be due to non-21-nt siRNA produced by a DCL other than DCL4, such as DCL2 [5]. In support of this hypothesis, in the absence of DCL4 activity, 22 and 24nt siRNAs are easily detected from TAS loci [38]. The limitation of phasing may be an evolutionary adaptation to prohibit extended signal amplification resulting in suppression of indirect target transcripts.
2.1. Basic requirements for Arabidopsis trans-acting siRNA biogenesis There are currently four TAS families classified in Arabidopsis computationally identified and validated [1–5,30]. These families all require specific miRNAs to guide cleavage as well as the following basic steps for biogenesis of ta-siRNAs: (1) ARGONAUTE directed miRNA-guided cleavage of a single-stranded RNA precursor transcript, (2) generation of double-stranded RNA by an RdRP, and (3) processing of the dsRNA from the miRNA-guided cleavage site to create unique phased 21-nt siRNAs. These four families can be separated by two classes: those that require one miRNA binding site (TAS1, TAS2, TAS4), and those that require two (TAS3) (Fig. 1). The TAS1/2/4 class and TAS3 class each require unique components and unique RNA structural features; therefore they will be discussed separately. 2.2. TAS1/2/4 single miRNA target model 2.2.1. miRNA cleavage sets the 5 position of ta-siRNA phasing The first characterized ta-siRNA families were TAS1 and TAS2 from Arabidopsis [18,24,25,27,31]. The TAS1 family consists of three loci, TAS1a (At2g27400), TAS1b (At1g50055) and TAS1c (At2g39675); TAS2 (At2g39681) is in close proximity to TAS1c on the second chromosome. Both TAS1 and TAS2 families require miR173 to guide cleavage of the transcript for ta-siRNA biogenesis [27,32,33]. The TAS1/2 primary transcripts are typical RNA Pol II polyadenylated and capped transcripts [25]. The longest potential open reading frames are short at under 100 amino acids raising the possibility that they may not be translated. Plant cells recognize non-protein-coding Pol II transcripts as ‘aberrant’, thus TAS precursors could be primed by the lack of a long ORF to enter an RNA silencing pathway. Cleavage by miR173 occurs at the canonical position between bases 10 and 11 in the target transcript [18,27,34]. miR173, like the majority of plant miRNAs, contains a 5 terminal uridine and associates primarily with AGO1 [11,13,35]. Unlike canonical miRNA targets, TAS1/2 transcripts when bound to miR173 create a mismatch at position 9 (TAS1a, -c, TAS2) or 9 and 10 (TAS1b). Mispairing at bases 9, 10, or 11 results in reduced or abolished cleavage activity [13,18,29]. Potentially the mispairing at or near the cleavage site in a single target transcript pauses the RISC and may recruit additional factors such as SGS3, required for ta-siRNA initiation. In the current model, SGS3 is necessary to stabilize the cleaved transcript followed by recruitment of RDR6, generating double-stranded RNA from the 3 fragment (Fig. 1). SGS3 has been shown to interact and colocalize with RDR6 in cytoplasmic granules [36]. The complementary strand is synthesized by RDR6 proceeding from the 3 poly(A) tail towards the cleavage site, thus it is unknown how potential recruitment of SGS3 and RDR6 at the cleavage site could direct synthesis initiated distally. Pairing ability of the miRNA to the transcript may affect how these transcripts are shunted through the ta-siRNA pathway; by altering the stability of the miRNA:target site, ribosomes may be cleared away allowing for entry of RDR6/SGS3. When the miR173 target site is substituted for a perfectly paired miRNA:target sequence (miR159, miR169, miR171, miR167), ta-siRNA production [32,33] is abolished. Likewise by creating mismatches between the miRNA:target site, the miRNA loses ability to guide cleavage and instead acts as a translational suppressor [37]. Factors that recruit RDR6 or RNA elements to direct initiation of second strand synthesis are not known. The
2.2.3. Nomenclature for phased ta-siRNAs ta-siRNAs have been generally named either by a unique sequence ID by the author, or using a standardized nomenclature in which the register is given a number D1, D2, . . . starting at the miRNA target site [18]. The processing direction is noted by a 5 or 3 prefix (3 in the case of TAS1/2); the orientation is indicated by adding the suffix [+] for the positive (original transcript) strand, or [–] for the negative (RDR generated) strand. 2.2.4. Incorporation of processed ta-siRNAs into ARGONAUTE effector complexes Incorporation of ta-siRNAs into AGO effector complexes follows the 5 base specificity rule, where ta-siRNAs possessing a 5 U are incorporated into AGO1 [11]. TAS1 loci code for multiple ta-siRNAs with very similar sequences predicted to target the same genes. The amplification of the silencing signal through production of multiple siRNAs may increase the efficacy of suppression, or establish strong gradients of target gene regulation (see Section 3). 2.2.5. Function of TAS1/2/4 ta-siRNAs It is noteworthy that at least one ta-siRNA produced from the TAS2 transcript targets two clusters of pentatricopeptide repeat gene transcripts (PPRs) on chromosome 1. Interestingly, these same transcripts also undergo phased siRNA biogenesis from not only TAS2 3 D6[–], but also by miR161.1 and miR161.2 and miR1427 in rice [1] (and possibly miR400, and other TAS1 and TAS2 ta-siRNAs that have yet to be experimentally validated). Arabidopsis contains at least 448 PPR-related genes, which code for putative RNA binding proteins [39,40]. The proliferation of this gene family is highly regulated through miRNA and ta-siRNA networks. It is likely the PPRs are targeted for suppression to quench rapid expansion of this gene family. This family, like other non-conserved and/or evolving miRNA targets may undergo a “birth and death” process as these genes show high sequence and location variability in orthologous, closely related regions of Brassica rapa and Arabidopsis [41]. Currently no other targets have been identified for TAS1 and TAS2 ta-siRNAs. A model has emerged that begins to explain the intricacies of ta-siRNA generation in Arabidopsis. Initiation of the each pathway starts by guided cleavage with a specific miRNA (Fig. 1). miR173 specifically targets TAS1 and TAS2 transcripts and does not appear to be conserved in other eukaryotes. In contrast, miR828 is conserved among plants and is specifically involved in regulation of MYB transcription factors, targeting both MYB113 and TAS4 transcripts. The resulting TAS4 ta-siRNAs in turn regulate MYB113 and
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
801
Fig. 1. TAS pathways in plants. Two pathways are known to generate ta-siRNAs in plants, pathway A in which one miRNA guides cleavage at the 5 end of the mRNA transcript, and pathway B in which two miRNA binding sites activate ta-siRNA production. An initial miRNA precursor is processed by DCL1 and the resulting miRNA strand guides cleavage of a non-protein-coding TAS transcript. In pathway A, miR173 or miR828 binds to the transcript and guides cleavage mediated by AGO1. RDR6 synthesizes a double-stranded RNA fragment that is subsequently processed by DCL4 into a phased, 21-nt register starting at the miRNA cleavage site. One strand of the resulting duplex then targets a complementary mRNA in trans. In pathway B, miR390 binds to the transcript at two sites. In Arabidopsis, the 3 miR390 site is cleaved by AGO7, while the 5 miR390 is required but not cleaved. RDR6 synthesizes a double-stranded RNA fragment that is subsequently processed by DCL4 into a phased, 21-nt register from the 3 miR390 cleavage site. One strand of the resulting duplex then targets a complementary mRNA in trans.
related family members. The features that define a trans-acting trigger miRNA have not been characterized, but they appear to be coded by the miRNA sequence, as substitution of an alternative miRNA sequence into the trigger precursor does not permit the alternative miRNA to generate ta-siRNAs [32]. Likewise, ta-siRNAs are produced from heterologous transcripts containing a miR173 target site indicating the miRNA sequence as the only cis-sequence requirement to trigger biogenesis of ta-siRNAs [33]. 2.3. TAS3 dual miRNA target model 2.3.1. TAS3 loci utilize two conserved miR390 binding sites for function The dual target model is currently unique to TAS3 loci. TAS3 transcripts and the trigger miR390 have been identified in mosses and many higher plants indicating a highly conserved mechanism for ta-siRNA generation [5,29,30]. Unlike the TAS1/2 and TAS4 families, TAS3 transcripts utilize two miR390 binding sites flanking the functional ta-siRNAs (Fig. 1). Three TAS3 loci have been identified to date in Arabidopsis: TAS3a (At3g17185), TAS3b (At5g49615), and TAS3c (At5g57735) [5]. In these transcripts, miR390 guides cleavage at the site 3 of the functional ta-siRNAs to set the processing register [18]. The functional ta-siRNAs (tasiARFs) are in phase and set by the 3 site, indicating a critical role for miR390
directed cleavage. In Arabidopsis, tasiARFs are located in positions 5 D7[+] and 5 D8[+], although the phase position varies among monocot plant species [18,29,42]. Therefore, the phase position of the tasiARFs appears not to be critical, rather only the miR390 binding sites and register relative to the 3 cleavage position. The 5 miR390 binding is mispaired with miR390 at critical positions for cleavage, resulting in a non-cleavable binding interaction between miR390 and TAS3. Studies from both Arabidopsis and moss indicate cleavage at the 5 miR390 binding site in TAS3 transcripts is undetectable [5,29]. Recently however, a novel miR390-dependent set of TAS transcripts have been identified in dicots in which the 5 miR390 binding site may undergo cleavage for initiation of processing [43]. 2.3.2. ARGONAUTE7 is required for TAS3 ta-siRNA processing TAS3 transcripts are unique from other non-coding TAS loci, not only in their dual miR390 specificity and direction of processing from the 3 site, but also in their requirement for AGO7. Unlike miR173 and most other miRNAs, miR390 is excluded from AGO1 because of its 5 terminal adenosine nucleotide [13]. This sorting mechanism allows miR390 to avoid the general miRNA 5 base criteria for AGO1 specificity. Based on 5 nucleotide specificity criteria, miR390 would be predicted to associate with primarily with AGO2 or AGO4 [11,12]; however sequencing of RNA-IP experiments
802
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
with HA-tagged AGO2 and AGO7 suggest a dominant AGO7:miR390 interaction [13]. It is not yet known what additional features determine why miR390 preferentially associates with AGO7 and not AGO2 or AGO4. What is the role of AGO7, and can other miRNAs substitute? The answer to this question comes from studies in which alternative miRNA binding sites were substituted for miR390 in cleavable or non-cleavable form, or where AGO specificity for miR390 was altered by substituting the 5 terminal adenine with uridine [13]. A modified miR390 containing a 5 terminal uridine associates with AGO1 and guides cleavage of a synthetic TAS transcript, yet processing of the TAS transcript and ta-siRNA biogenesis was abolished. In contrast, the 3 site is not dependent on AGO7 or miR390 specifically, only that a cleavable miRNA target is present. The 3 site can be substituted with an alternative miRNA such as miR171, resulting in efficient cleavage and ta-siRNA production. Any alterations to the 5 site by either removing miR390, AGO7, or making the site cleavable will disrupt ta-siRNA production [13,29]. Binding of miR390 and AGO7 in a complex to a non-cleavable 5 site are essential requirements for the dual target ta-siRNA model. 3. Function of TAS3 trans-acting siRNAs in development 3.1. tasiR-ARFs are mobile signals and establish a suppression gradient All TAS3 transcripts identified to date contain at least one conserved sequence when processed to mature ta-siRNAs (tasiR-ARFs) that can target Auxin Response Factor family members ARF1, 2, 3, or 4. In general, processing of angiosperm specific TAS3 loci results in two mature abundant tasiR-ARFs that are nearly identical in sequence and target specificity. ARF1 and 2 contain one tasiR-ARF target site, whereas ARF3 and 4 each contain two conserved target sites and are likely the primary targets of tasiR-ARFs [18,29,44–47]. Production of two or more nearly identical ta-siRNAs from a locus is a common feature for TAS1 and TAS3, and may be a mechanism to further amplify a suppression signal while maintaining target specificity. In addition, production of multiple ta-siRNAs could strengthen the suppression gradient developed by ta-siRNAs. Studies have shown that ta-siRNAs are mobile, and can travel across multiple cells to create a gradient of suppression activity [48,49]. When the ta-siRNA source cell is strong, a steeper gradient may be created [50]. 3.2. tasiR-ARFs are required for proper formation of leaf polarity TAS3 ta-siRNAs, or tasiR-ARFs, suppress the juvenile-to-adult phase transition in Arabidopsis through negative regulation of ARF3/ETTIN [45], and are required for proper formation of leaf development by establishing adaxial/abaxial (dorsoventral) polarity. Similarly to rdr6, dcl4, and ago7 mutants, ARF3mut plants (non-targeted ARF3) also display phase change phenotypes. Further evidence for a role in phase transition is shown in both ARF3 overexpression and ARF3mut plants in an rdr6 background; plants exhibit an increased blade/petiole leaf length, an earlier appearance of abaxial trichomes and strong downward leaf curvature indicating severe morphological and patterning defects [45]. These data indicate an ARF3 dosage dependent accelerated phase change phenotype in ta-siRNA deficient mutants. Furthermore, tasiR-ARFs move intercellularly from the adaxial (upper) to abaxial (lower) side of leaf priomordia to create a gradient that patterns ARF3 expression [48]. These leaf patterning phenotypes are dependent on the polarization of miR390 by restricted AGO7 expression. Antagonistic activities of miR166 on class III homeodomain leucine zipper (HD-ZIPIII) transcription factors also contribute to speci-
fication of adaxial cell fate as tasiR-ARFs restrict the expression domain of miR166 in corn and Arabidopsis [50–52]. Nogueira and colleagues show leaf bladeless1 (lbl1), whose function is necessary for TAS3 ta-siRNAs, leads to ectopic accumulation of miR166 resulting in specification of leaf polarity by the ta-siRNA pathway [53,54]. These and other antagonistic interactions between polarity determinants help to establish dorsoventral patterning within developing organs.
4. Looking forward 4.1. Identification of new TAS loci Additional validated ta-siRNA loci have been scarce, therefore it remains to be seen how broadly this mechanism is utilized by plants. Algorithms are available to assist in prediction of phased ta-siRNA loci, and can be used in combination with existing miRNA target prediction rules [1,5,19,55]. Each prediction algorithm is unique, yet they share common features including a requirement for multiple phases to contain siRNAs, and a predominant 21-nt phasing pattern. Additional evidence is supplied by siRNA sequence depletion in RDR6, SGS3, or DCL4 mutants where available. Since the first publications in 2004 on the ta-siRNA pathway in Arabidopsis, only two new TAS loci, AtTAS4 and PpTAS4, have been validated [30,56,57]. In 2006, an additional transcript was identified in Arabidopsis by Rajagopalan et al. and named AtTAS4 [56]. Unlike the TAS1/2 families which require miR173 as the initiator, TAS4 requires miR828 to guide cleavage in an AGO1 dependent manner. This family was not identified until the advent of high throughput sequencing technologies, since the ta-siRNAs identified from the TAS4 transcript show low expression activity. Recently, phosphate starvation studies uncovered a role for TAS4 in regulating the biosynthesis of anthocyanin [57]. Both TAS4 ta-siRNAs and miR828 were upregulated in shoots by Pi and nitrogen deficiency. TAS4-siR81(−) targets several MYB transcription factors including PAP1/MYB75, PAP2/MYB90, and MYB113 involved in biosynthesis of anthocyanin [56]. In addition, both MIR828 and TAS4 genes possess multiple MYB binding sites and a PAP1 cis-regulatory element, resulting in a feedback mechanism for regulation of anthocyanin accumulation during stress. An autoregulatory mechanism was uncovered in which phosphate deficiency resulted in both upregulation of the MYB transcription factors activating anthocyanin biosynthesis, as well as increasing production of ta-siRNAs via the activation of miR828 and/or TAS4. A more extensive search for RDR6/DCL4 dependent loci revealed the existence of several protein-coding loci which behave similarly to TAS loci. Among these are several protein-coding loci that generate phased siRNAs including a CC-NBS-LRR domain targeted by miR472, and transcripts targeted by miR393 [5]. miR472 is conserved in poplar (miRBASE) and several crop plants, as are phased siRNAs generated from orthologous transcripts (authors, unpublished data). These miRNA:target transcripts likely behave similarly to TAS transcripts in that the siRNAs generated from these transcripts target other mRNAs in trans, however they are proteincoding, a characteristic atypical for ta-siRNA loci. In Physcomitrella patens, a phased siRNA generating locus named PpTAS4 has been validated, indicating that ta-siRNA mechanisms are present in primitive species [30,57]. In addition, several loci that form internal foldback structures have been found that produce phased siRNAs in the absence of a miRNA initiator [1,3,4]. These phased siRNAs are likely early intermediates in the evolution of miRNA [9]. Because they do not require a miRNA, they are not considered canonical tasiRNAs although they can processed by DCL4 and target distantly related transcripts.
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
4.2. Questions yet to be elucidated for ta-siRNAs Several questions regarding this specialized small RNA pathway were posed by Vaucheret [31] that we can now answer, although a few details remain unanswered and more questions have arisen. Generation of ta-siRNAs is a highly effective method of suppression for plants using a steep gradient of gene expression (ARFs), or for targeting expanded gene families for repression (PPRs). The tasiRNA pathway could be a mechanism to control and amplify small RNA production while converting a cell autonomous miRNA signal to a non-cell autonomous ta-siRNA signal (in concert with DCL4) for creating strong gradients across cells [58]. It appears generation of all ta-siRNAs requires the activity of RDR6 and DCL4, but it remains a question on whether all phased siRNAs (for example, proteincoding transcript targets and rice siRNA clusters) also require these or orthologous protein components within their silencing machinery. The two variations of single or dual targeting also require different AGO family members, which may aid in determining the directionality of RDR6 dependent dsRNA production. It is still unclear how ta-siRNA transcripts utilize an RdRP for dsRNA amplification, and whether all siRNAs generated from these transcripts are functional. How do plants differentiate a miRNA targeted for degradation from one directed for dsRNA synthesis? Could this be related to pausing of the AGO-RISC complex as a result of a second non-cleavable target, or from a mismatched target site that reduces miRNA activity? In fact, a subset of miRNA targets generate secondary siRNAs, although at a reduced level compared to ta-siRNAs [59]. Finally, how widely do plants utilize the ta-siRNA mechanism? We anticipate answers to these questions as more plant genomes are sequences and small RNA analyses are applied.
References [1] Heisel SE, Zhang Y, Allen E, Guo L, Reynolds TL, Yang X, et al. Characterization of unique small RNA populations from rice grain. PLoS One 2008;3:e2871. [2] Zhao T, Li G, Mi S, Li S, Hannon GJ, Wang XJ, et al. A complex system of small RNAs in the unicellular green alga Chlamydomonas reinhardtii. Genes Dev 2007;21:1190–203. [3] Johnson C, Kasprzewska A, Tennessen K, Fernandes J, Nan GL, Walbot V, et al. Clusters and superclusters of phased small RNAs in the developing inflorescence of rice. Genome Res 2009;19:1429–40. [4] Zhu QH, Spriggs A, Matthew L, Fan L, Kennedy G, Gubler F, et al. A diverse set of microRNAs and microRNA-like small RNAs in developing rice grains. Genome Res 2008;18:1456–65. [5] Howell MD, Fahlgren N, Chapman EJ, Cumbie JS, Sullivan CM, Givan SA, et al. Genome-wide analysis of the RNA-DEPENDENT RNA POLYMERASE6/DICERLIKE4 pathway in Arabidopsis reveals dependency on miRNA- and tasiRNAdirected targeting. Plant Cell 2007;19:926–42. [6] Palatnik JF, Allen E, Wu X, Schommer C, Schwab R, Carrington JC, et al. Control of leaf morphogenesis by microRNAs. Nature 2003;425:257–63. [7] Rhoades MW, Reinhart BJ, Lim LP, Burge CB, Bartel B, Bartel DP. Prediction of plant microRNA targets. Cell 2002;110:513–20. [8] Park W, Li J, Song R, Messing J, Chen X. CARPEL FACTORY, a Dicer homolog, and HEN1, a novel protein, act in microRNA metabolism in Arabidopsis thaliana. Curr Biol 2002;12:1484–95. [9] Voinnet O. Origin, biogenesis, and activity of plant microRNAs. Cell 2009;136:669–87. [10] Reinhart BJ, Weinstein EG, Rhoades MW, Bartel B, Bartel DP. MicroRNAs in plants. Genes Dev 2002;16:1616–26. [11] Mi S, Cai T, Hu Y, Chen Y, Hodges E, Ni F, et al. Sorting of small RNAs into Arabidopsis argonaute complexes is directed by the 5 terminal nucleotide. Cell 2008;133:116–27. [12] Takeda A, Iwasaki S, Watanabe T, Utsumi M, Watanabe Y. The mechanism selecting the guide strand from small RNA duplexes is different among argonaute proteins. Plant Cell Physiol 2008;49:493–500. [13] Montgomery TA, Howell MD, Cuperus JT, Li D, Hansen JE, Alexander AL, et al. Specificity of ARGONAUTE7-miR390 interaction and dual functionality in TAS3 trans-acting siRNA formation. Cell 2008;133:128–41. [14] Souret FF, Kastenmayer JP, Green PJ. AtXRN4 degrades mRNA in Arabidopsis and its substrates include selected miRNA targets. Mol Cell 2004;15:173–83. [15] Ramachandran V, Chen X. Degradation of microRNAs by a family of exoribonucleases in Arabidopsis. Science 2008;321:1490–2. [16] Gy I, Gasciolli V, Lauressergues D, Morel JB, Gombert J, Proux F, et al. Arabidopsis FIERY1, XRN2, and XRN3 are endogenous RNA silencing suppressors. Plant Cell 2007;19:3451–61.
803
[17] Fahlgren N, Carrington JC. miRNA target prediction in plants. Methods Mol Biol 2010;592:51–7. [18] Allen E, Xie Z, Gustafson AM, Carrington JC. microRNA-directed phasing during trans-acting siRNA biogenesis in plants. Cell 2005;121:207–21. [19] Jones-Rhoades MW, Bartel DP. Computational identification of plant microRNAs and their targets, including a stress-induced miRNA. Mol Cell 2004;14:787–99. [20] Lanet E, Delannoy E, Sormani R, Floris M, Brodersen P, Crete P, et al. Biochemical evidence for translational repression by Arabidopsis microRNAs. Plant Cell 2009;21:1762–8. [21] Brodersen P, Sakvarelidze-Achard L, Bruun-Rasmussen M, Dunoyer P, Yamamoto YY, Sieburth L, et al. Widespread translational inhibition by plant miRNAs and siRNAs. Science 2008;320:1185–90. [22] Ambros V, Bartel B, Bartel DP, Burge CB, Carrington JC, Chen X, et al. A uniform system for microRNA annotation. RNA 2003;9:277–9. [23] Sunkar R, Zhu JK. Novel and stress-regulated microRNAs and other small RNAs from Arabidopsis. Plant Cell 2004;16:2001–19. [24] Peragine A, Yoshikawa M, Wu G, Albrecht HL, Poethig RS. SGS3 and SGS2/SDE1/RDR6 are required for juvenile development and the production of trans-acting siRNAs in Arabidopsis. Genes Dev 2004;18:2368–79. [25] Vazquez F, Vaucheret H, Rajagopalan R, Lepers C, Gasciolli V, Mallory AC, et al. Endogenous trans-acting siRNAs regulate the accumulation of Arabidopsis mRNAs. Mol Cell 2004;16:69–79. [26] Mourrain P, Beclin C, Elmayan T, Feuerbach F, Godon C, Morel JB, et al. Arabidopsis SGS2 and SGS3 genes are required for posttranscriptional gene silencing and natural virus resistance. Cell 2000;101:533–42. [27] Yoshikawa M, Peragine A, Park MY, Poethig RS. A pathway for the biogenesis of trans-acting siRNAs in Arabidopsis. Genes Dev 2005;19:2164–75. [28] Meyers BC, Axtell MJ, Bartel B, Bartel DP, Baulcombe D, Bowman JL, et al. Criteria for annotation of plant MicroRNAs. Plant Cell 2008;20:3186–90. [29] Axtell MJ, Jan C, Rajagopalan R, Bartel DP. A two-hit trigger for siRNA biogenesis in plants. Cell 2006;127:565–77. [30] Talmor-Neiman M, Stav R, Klipcan L, Buxdorf K, Baulcombe DC, Arazi T. Identification of trans-acting siRNAs in moss and an RNA-dependent RNA polymerase required for their biogenesis. Plant J 2006;48:511–21. [31] Vaucheret H. MicroRNA-dependent trans-acting siRNA production. Sci STKE 2005;2005:pe43. [32] Montgomery TA, Yoo SJ, Fahlgren N, Gilbert SD, Howell MD, Sullivan CM, et al. AGO1-miR173 complex initiates phased siRNA formation in plants. Proc Natl Acad Sci USA 2008;105:20055–62. [33] Felippes FF, Weigel D. Triggering the formation of tasiRNAs in Arabidopsis thaliana: the role of microRNA miR173. EMBO Rep 2009;10:264–70. [34] Gasciolli V, Mallory AC, Bartel DP, Vaucheret H. Partially redundant functions of Arabidopsis DICER-like enzymes and a role for DCL4 in producing trans-acting siRNAs. Curr Biol 2005;15:1494–500. [35] Baumberger N, Baulcombe DC. Arabidopsis ARGONAUTE1 is an RNA Slicer that selectively recruits microRNAs and short interfering RNAs. Proc Natl Acad Sci USA 2005;102:11928–33. [36] Kumakura N, Takeda A, Fujioka Y, Motose H, Takano R, Watanabe Y. SGS3 and RDR6 interact and colocalize in cytoplasmic SGS3/RDR6-bodies. FEBS Lett 2009;583:1261–6. [37] Dugas DV, Bartel B. Sucrose induction of Arabidopsis miR398 represses two Cu/Zn superoxide dismutases. Plant Mol Biol 2008;67:403–17. [38] Xie Z, Allen E, Wilken A, Carrington JC. DICER-LIKE 4 functions in trans-acting small interfering RNA biogenesis and vegetative phase change in Arabidopsis thaliana. Proc Natl Acad Sci USA 2005;102:12984–9. [39] Saha D, Prasad AM, Srinivasan R. Pentatricopeptide repeat proteins and their emerging roles in plants. Plant Physiol Biochem 2007;45:521–34. [40] Lurin C, Andres C, Aubourg S, Bellaoui M, Bitton F, Bruyere C, et al. Genomewide analysis of Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis. Plant Cell 2004;16:2089–103. [41] Geddy R, Brown GG. Genes encoding pentatricopeptide repeat (PPR) proteins are not conserved in location in plant genomes and may be subject to diversifying selection. BMC Genom 2007;8:130. [42] Axtell MJ, Snyder JA, Bartel DP. Common functions for diverse small RNAs of land plants. Plant Cell 2007;19:1750–69. [43] Krasnikova MS, Milyutina IA, Bobrova VK, Ozerova LV, Troitsky AV, Solovyev AG, et al. Novel miR390-Dependent Transacting siRNA precursors in plants revealed by a PCR-based experimental approach and database analysis. J Biomed Biotechnol 2009;2009:952304. [44] Garcia D, Collier SA, Byrne ME, Martienssen RA. Specification of leaf polarity in Arabidopsis via the trans-acting siRNA pathway. Curr Biol 2006;16: 933–8. [45] Fahlgren N, Montgomery TA, Howell MD, Allen E, Dvorak SK, Alexander AL, et al. Regulation of AUXIN RESPONSE FACTOR3 by TAS3 ta-siRNA affects developmental timing and patterning in Arabidopsis. Curr Biol 2006;16:939–44. [46] Adenot X, Elmayan T, Lauressergues D, Boutet S, Bouche N, Gasciolli V, et al. DRB4-dependent TAS3 trans-acting siRNAs control leaf morphology through AGO7. Curr Biol 2006;16:927–32. [47] Williams L, Carles CC, Osmont KS, Fletcher JC. A database analysis method identifies an endogenous trans-acting short-interfering RNA that targets the Arabidopsis ARF2, ARF3, and ARF4 genes. Proc Natl Acad Sci USA 2005;102:9703–8. [48] Chitwood DH, Nogueira FT, Howell MD, Montgomery TA, Carrington JC, Timmermans MC. Pattern formation via small RNA mobility. Genes Dev 2009;23:549–54.
804
E. Allen, M.D. Howell / Seminars in Cell & Developmental Biology 21 (2010) 798–804
[49] Schwab R, Maizel A, Ruiz-Ferrer V, Garcia D, Bayer M, Crespi M, et al. Endogenous TasiRNAs mediate non-cell autonomous effects on gene regulation in Arabidopsis thaliana. PLoS One 2009;4:e5980. [50] Husbands AY, Chitwood DH, Plavskin Y, Timmermans MC. Signals and prepatterns: new insights into organ polarity in plants. Genes Dev 2009;23:1986–97. [51] Nogueira FT, Chitwood DH, Madi S, Ohtsu K, Schnable PS, Scanlon MJ, et al. Regulation of small RNA accumulation in the maize shoot apex. PLoS Genet 2009;5:e1000320. [52] Nogueira FT, Madi S, Chitwood DH, Juarez MT, Timmermans MC. Two small regulatory RNAs establish opposing fates of a developmental axis. Genes Dev 2007;21:750–5. [53] Nogueira F, Timmermans MC. An interplay between small regulatory RNAs patterns leaves. Plant Signal Behav 2007;2:519–21. [54] Nogueira FT, Sarkar AK, Chitwood DH, Timmermans MC. Organ polarity in plants is specified through the opposing activity of two distinct small regulatory RNAs. Cold Spring Harb Symp Quant Biol 2006;71:157–64.
[55] Chen HM, Li YH, Wu SH. Bioinformatic prediction and experimental validation of a microRNA-directed tandem trans-acting siRNA cascade in Arabidopsis. Proc Natl Acad Sci USA 2007;104:3318–23. [56] Rajagopalan R, Vaucheret H, Trejo J, Bartel DP. A diverse and evolutionarily fluid set of microRNAs in Arabidopsis thaliana. Genes Dev 2006;20: 3407–25. [57] Hsieh LC, Lin SI, Shih AC, Chen JW, Lin WY, Tseng CY, et al. Uncovering small RNA-mediated responses to phosphate-deficiency in Arabidopsis by deep sequencing. Plant Physiol 2009. [58] Voinnet O. Use, tolerance and avoidance of amplified RNA silencing by plants. Trends Plant Sci 2008;13:317–28. [59] Ronemus M, Vaughn MW, Martienssen RA. MicroRNA-targeted and small interfering RNA-mediated mRNA degradation is regulated by argonaute, dicer, and RNA-dependent RNA polymerase in Arabidopsis. Plant Cell 2006;18: 1559–74.