The Interactive Fly
Zygotically transcribed genes
RNAi and PTGS - functions and processes
Principles of microRNA-target recognition
An mRNA m7G cap binding-like motif within human Ago2 represses translation
Although hundreds of distinct animal microRNAs (miRNAs) are known, the specific biological functions of only a handful are understood at present. Three different families of Drosophila miRNAs directly regulate two large families of Notch target genes, including basic helix-loop-helix (bHLH) repressor and Bearded family genes. These miRNAs regulate Notch target gene activity via GY-box (GUCUUCC), Brd-box (AGCUUUA), and K-box (cUGUGAUa) motifs. These are conserved sites in target 3'-untranslated regions (3'-UTRs) that are complementary to the 5'-ends of miRNAs, or 'seed' regions. Collectively, these motifs represent >40 miRNA-binding sites in Notch target genes, and all three classes of motif are shown to be necessary and sufficient for miRNA-mediated regulation in vivo. Importantly, many of the validated miRNA-binding sites have limited pairing to miRNAs outside of the "box:seed" region. Consistent with this, it was found that seed-related miRNAs that are otherwise quite divergent can regulate the same target sequences. Finally, it is demonstrated that ectopic expression of several Notch-regulating miRNAs induces mutant phenotypes that are characteristic of Notch pathway loss of function, including loss of wing margin, thickened wing veins, increased bristle density, and tufted bristles. Collectively, these data establish insights into miRNA target recognition and demonstrate that the Notch signaling pathway is a major target of miRNA-mediated regulation in Drosophila (Lai, 2005).
The E(spl)-C and Brd-C of Drosophila melanogaster (Dm) contain two large families of direct Notch target genes, including seven bHLH repressor-encoding genes and 10 Bearded family genes. With the exception of E(spl)mbeta and Ocho, all of these genes contain GY-box (GUCUUCC), Brd-box (AGCUUUA), and/or K-box (UGUGAU) motifs in their 3'-UTRs, which are propose to be miRNA-binding sites. Nine of these genes contain three or more box sites, a density that is especially remarkable when one considers how short their 3'-UTRs are (often <350 nt in length). The conservation of these sites were systematically assessed in their orthologs from Drosophila pseudoobscura (Dp) and Drosophila virilis (Dv), species that are ~30 million and 60 million years diverged from Dm, respectively. 33/51 Brd-boxes, GY-boxes, and K-boxes have been perfectly conserved and reside in syntenic locations among all three species; 11 additional sites are identical in two of the three species. This indicates that all three motifs are under strong selective constraint (Lai, 2005).
Closer examination of nucleotide divergence surrounding these boxes has revealed some unexpected features that are germane to the proposition that these boxes represent miRNA-binding sites. These features are best illustrated by comparing rapidly evolving genes. Notably, Bearded is an unusually rapidly evolving protein, with only 15 residues preserved between Dm and Dv orthologs (out of 81 and 66 amino acids, respectively), and Dv Bearded has a significantly different arrangement of these 3'-UTR motifs. The 3'-UTR of Dv E(spl)m5 is also quite different from its counterparts in Dm/Dp. Alignment of Dm/Dp orthologs of Bearded and E(spl)m5 reveals that sequences upstream of most GY-boxes are well conserved; these regions include most sequences presumed to pair with miR-7. Similar patterns are seen for many other GY-boxes in other Notch target genes. However, the sequence upstream of many Brd- and K-boxes is strongly diverged, so that only 'box'-pairing is often preserved. In fact, many Brd- and K-boxes generally lack extensive pairing to miRNAs outside of the 'box' sequence. These factors likely preclude their identification by various published computational algorithms for miRNA-binding sites. Indeed, Brd- and K-boxes in Notch target genes have been deemed unlikely to represent miRNA-binding sites. In contrast, rapid divergence of the upstream portion of miRNA-binding sites is consistent with the idea that pairing between the miRNA "seed" (positions ~2-8) and the 3'-UTR 'box' (approximately the last one-third of the miRNA-binding site) is most critical for miRNA-mediated regulation (Lai, 2005).
It is also noted that precise spacing of several motif occurrences that are closely paired is also conserved, even though orthologous 3'-UTRs otherwise display significant insertions and deletions. In these cases, one would presume that simultaneous binding of miRNAs to their respective sites would not be possible unless the 3'-end of the downstream miRNA was unpaired, a configuration that unexpectedly proved functional in vitro. Finally, there are a few nonconserved boxes in these 3'-UTRs (7/51 total sites). In several cases, the nonconserved site is highly related to a neighboring conserved site [i.e., the first and second GY-boxes of Dp E(spl)m4 are equally similar to the first GY-box in Dm E(spl)m4; the third and fourth Brd-boxes in Dp E(spl)m5 are highly related to the third Brd-box in Dm E(spl)m5], implying that these nonconserved sites may be functional, newly evolved miRNA-binding sites (Lai, 2005).
GY-box-, Brd-box-, and K-box-class miRNAs are highly conserved among diverse insects, and many are, indeed, identical. Therefore Brd-boxes, GY-boxes, and K-boxes were sought in the predicted 3'-UTRs of E(spl)bHLH and Brd genes from mosquitoes, bees, and moths; these species cover ~350 million years of divergence from Drosophila. Impressively, homologs of both E(spl)bHLH and Brd genes in these highly diverged species all contain multiple copies and multiple classes of 'box' motifs in their 3'-UTRs. This strongly suggests that regulation by all three families of miRNAs is an ancient feature of Notch target gene regulation in insects (Lai, 2005).
To directly test the capacity of miRNAs to regulate the 3'-UTRs of these Notch target genes, an in vivo assay was used. The target in this assay is a ubiquitously expressed reporter (tub>GFP or arm>lacZ) fused to an endogenous 3'-UTR (a 3'-UTR sensor). The reporter transgene is introduced into a genetic background in which a UAS-DsRed-miRNA transgene is activated with dpp-Gal4 or ptc-Gal4. This results in ectopic miRNA production in a stripe of red-fluorescing cells at the anterior–posterior boundary of imaginal discs. Inhibition of the green reporter within the red miRNA-misexpressing domain reflects direct miRNA-mediated negative regulation. Focus was placed on the central wing pouch region of the wing imaginal disc (Lai, 2005).
The ability of sensor transgenes for most Bearded family genes [Bob, Bearded, Tom, Ocho, E(spl)malpha, and E(spl)m4] and most E(spl)bHLH repressor genes [E(spl)mgamma, E(spl)mdelta, E(spl)m3, E(spl)m5, and E(spl)m8] to be regulated by ectopic GY-box-, Brd-box-, and K-box-class miRNAs was extensively analyzed. Sensor expression is influenced by the level to which it is negatively regulated by endogenous factors, including miRNAs. In this assay, the disc sensor must be expressed at sufficient levels before one can observe its knock-down by ectopic miRNAs. 3'-UTR sensor constructs for different Notch target genes accumulate to different levels in vivo, consistent with variable amounts of endogenous miRNA-mediated regulation. Nevertheless, it was possible to reliably detect expression of all sensors excepting E(spl)m8. As detailed in the following three sections, these sensors were used to unequivocally demonstrate GY-boxes, Brd-boxes, and K-boxes to be sites of miRNA-mediated negative regulation by corresponding families of complementary miRNAs in vivo (Lai, 2005).
miR-7 is the only known Drosophila miRNA whose 5'-end is complementary to the GY-box (GUCUUCC). miR-7 has been shown to regulate three GY-box targets, including two members of the E(spl)-C, E(spl)m3 and E(spl)m4. While these two genes scored well in a genome-wide prediction of miR-7 targets, many other members of the Brd-C and E(spl)-C also contain between one and three GY-boxes in their 3'-UTRs [Bob, Bearded, Tom, E(spl)mgamma, E(spl)m5]. Of these, only Tom was computationally identified as a compelling candidate for miR-7 (Lai, 2005).
The specificity of the disc sensor assay was assayed by showing that neither an empty tub-GFP sensor nor an Ocho sensor were affected by miR-7. The previous experiments done with E(spl)m3 and E(spl)m4 were repeated and it was observed that both were, indeed, inhibited by ectopic miR-7. This assay was used to demonstrate that miR-7 negatively regulates all seven GY-box-containing members of the Brd-C and E(spl)-C, including those with single sites [E(spl)m3, E(spl)mgamma, and Bearded], those with two sites [E(spl)m4, Tom, Bob], and those with three sites [E(spl)m5]. These data convincingly support the hypothesis that GY-boxes are general signatures of miR-7-binding sites in Notch target genes, irrespective of the overall amount of pairing between miR-7 and sequences outside of the GY-box. In order to more definitively demonstrate that miR-7-mediated regulation occurs through identified GY-boxes, mutant sensors bearing point mutations in the GY-boxes were tested. A Bearded sensor carrying five point mutations in its single GY-box no longer responded to miR-7. In a more stringent test, an E(spl)m5 sensor carrying 2-nt mutations in each of its three GY-boxes was generated. These targeted changes also abolished the ability of miR-7 to negatively regulate E(spl)m5. Therefore, ~7 continuous base pairs between the 'box' motif and its cognate miRNA seed are critical for in vivo target regulation. It is also noted that when mutant 3'-UTRs are tested, a mild increase in reporter activity in miRNA-misexpressing cells was sometimes observed, the reason for which has not been determined (Lai, 2005).
Previous work has suggested synergism between miRNA-binding sites on the same transcript. Multiple GY-box 3'-UTRs were generally subject to greater regulation than single-site 3'-UTRs, even though the amount of miR-7 pairing to individual GY-boxes in multiple-site 3'-UTRs is often less than its pairing with single GY-box 3'-UTRs. Indeed, negative regulation of E(spl)m4, Tom, Bob, and E(spl)m5 by miR-7 was qualitatively indistinguishable from an artificial sensor containing two perfectly miR-7-complementary sites, even though many sites in these genes display relaxed pairing with miR-7 outside of GY-boxes. This suggests that as little as 7–8 nt of complementarity may suffice for miRNA target recognition, especially where multiple sites are present. However, since all three single GY-box-containing 3'-UTRs were also regulated by miR-7, synergism is not required for biologically significant regulation by miRNAs (Lai, 2005).
There are two Drosophila miRNAs, miR-4 and miR-79, whose 5'-ends are complementary to the Brd-box (AGCUUUA). Both miRNAs are resident in miRNA clusters, and miR-4 resides in particularly dense clusters containing several unrelated miRNAs. Use was made of a UAS-DsRed-miR-286, miR-4, miR-5 transgene that is referred to as "UAS-miR-4" and a UAS-DsRed-miR-79 transgene. miR-4 and miR-79 have only limited similarity outside of their Brd-box seed, and there is little indication from pairwise alignments that these miRNAs are specifically "tuned" to different Brd-box sites in Notch target genes. In fact, all of these Brd-boxes lack the extended complementarity to miRNAs that is typical of miR-7:GY-box pairs, and no Notch target genes were previously predicted computationally as targets of miR-4 or miR-79 (Lai, 2005).
Seven Brd-box-containing Notch target genes were validated as being regulated by Brd-box-family miRNAs, including those with single sites [Tom, E(spl)mdelta, E(spl)mgamma] and those with multiple sites [Bearded, E(spl)malpha, E(spl)m4, and E(spl)m5]. Curiously, the negative regulatory effects of miR-4 on E(spl)mgamma, E(spl)malpha, E(spl)m4, and E(spl)m5 were greater than those of miR-79 on these same 3'-UTRs, even though miR-4 is no more complementary to these sites than is miR-79. Nevertheless, the common ability of miR-4 and miR-79 to down-regulate individual sensors indicates that cross-regulation of individual sites by multiple members of a given miRNA family may occur. Notably, both miRNAs are expressed at high levels during embryonic development (Lai, 2005).
The specificity of miR-4 and miR-79 was tested using two mutant Bearded sensors, one bearing several point mutations in each of its three Brd-boxes and another containing mutations in the Brd-boxes and the GY-box. In both cases, the mutant transgenes accumulate to higher levels, consistent with relief from negative regulation by endogenous Brd-box-class miRNAs in the wing disc. In addition, they are no longer responsive to ectopic Brd-box-class miRNAs, indicating that the observed regulation occurs directly via Brd-boxes. As well, this experiment demonstrates that regulation by the miR-4 transgene is not attributable to miR-286 and miR-5 carried on this construct. Nevertheless, this miRNA construct efficiently down-regulates a miR-5 sensor containing two miR-5 sites, indicating that the other miRNAs carried on this construct are functional. As a final test of the specificity of this assay, it was observed that this three-miRNA construct fails to inhibit the expression of an empty tub-GFP sensor (Lai, 2005).
Having demonstrated that Brd-boxes are bona fide miRNA-binding sites, it was asked whether regulation of the Bearded 3'-UTR by miR-7 requires the presence of Brd-boxes. This might be the case, for example, if negative regulation of a given 3'-UTR required synergism between different types of miRNA-binding sites. A Bearded 3'-UTR carrying mutations in each of the three Brd-boxes was observed to be still strongly inhibited by miR-7, indicating that individual types of miRNA-binding sites suffice for regulation in this assay (Lai, 2005).
The largest family of Drosophila miRNAs includes those whose 5'-ends are complementary to the K-box (cUGUGAUa, where the lowercase nucleotides represent positions of strong bias). The K-box is also the most pervasive motif within these Notch target genes; it is present in almost every member of the Brd-C and E(spl)-C [excepting E(spl)mbeta and Ocho, which lack any box motifs]. The maximum overall site complementarity of any given K-box site to any K-box family miRNAs is generally modest, and less than that seen with other demonstrated targets of the K-box family miRNA miR-2, namely, the proapoptotic genes grim, reaper, and sickle. In fact, the sole Notch target gene that was predicted informatically as a target of a K-box family miRNA in any study was E(spl)m8: miR-11, and this pair ranked only 46th (Lai, 2005).
The ability was tested of two quite distinct K-box family miRNAs, those of the miR-2 cluster (miR-2a-1, miR-2a-2, and miR-2b-2) and miR-11, to regulate K-box-containing 3'-UTRs. Given the abundance of K-box complementary miRNAs (as a class, they are among the more frequently cloned fly miRNAs), the occupancy of K-box sites by endogenous K-box-class miRNAs may be near-saturating in some cases. In fact, negative regulation of E(spl)m8, whose K-boxes mediate 10-fold negative regulation and nearly eliminate expression of this sensor, could not be convincingly demonstrated. In spite of this, positive evidence was obtained that four other K-box-containing 3'-UTRs, E(spl)m4, Bob, E(spl)malpha, and E(spl)mdelta, are directly regulated by K-box-family miRNAs, although the amount of regulation observed was weaker than that seen with GY-box- or Brd-box-class miRNAs. As was the case with the two Brd-box-class miRNAs, both miR-2 and miR-11 are capable of regulating some of the same K-box-containing targets. This constitutes further evidence for the possibility of cross-regulation of miRNA-binding sites, even where the miRNAs in question display very little similarity outside of their seeds (Lai, 2005).
In performing pairwise tests of these miRNAs with Notch target gene sensors, two instances were observed of miRNA-mediated regulation of sensors lacking canonical boxes. (1) It was observed that the E(spl)mdelta sensor was inhibited by miR-7. Although E(spl)mdelta lacks a canonical GY-box, it does contain a GY-box-like site that would have a single G:U base pair with the miR-7 seed. The nucleotides that are 5' and 3' to the box are also paired with miR-7, and there is a significant region of pairing to the 3'-end of the miRNA. These factors may allow this site to be recognized by miR-7. The 9-mer AGUUUUCCA is found in both Dp and Dv orthologs of E(spl)mdelta, indicating that this site is under selection and therefore is likely important for regulation of E(spl)mdelta. (2) It was observed that the Bob sensor was negatively regulated by both Brd-box-class miRNAs, miR-4 and miR-79. Although Bob lacks a canonical Brd-box, it does contain two matches to positions 2-7 of the Brd-box, which would pair to positions 2-7 of the miR-4/79. In this regard, this type of site is reminiscent of the 6-mer K-box, which pairs to positions 2-7 of K-box miRNAs. One of these Brd-box-like sites is conserved in Dp, and the syntenic site in Dv is, in fact, a canonical Brd-box, further indicating a functional relationship between Bob and miRNAs of the Brd-box family (Lai, 2005).
The apparent functionality of these noncanonical sites led to a search for other such sites in Notch target 3'-UTRs. Although one might expect to find many-fold more copies of degenerate sites relative to canonical sites, instead only a few additional examples of relaxed GY-box-like or Brd-box-like sites were found. For comparison, there are 28 canonical sites of these classes in Notch target 3'-UTRs (16 Brd-boxes and 12 GY-boxes), but only three additional examples of a 7-mer box-like site with a G:U base-pair to a miRNA seed [all are GY-box-like sites in E(spl)mdelta, E(spl)m3, and E(spl)m7]. In addition, there are only five additional examples of sites that match only positions 2-7 of the GY-box or the Brd-box [all of which are Brd-box-like sites: the two in Bob, one in E(spl)m7, one in E(spl)malpha, and one in E(spl)mdelta]. These considerations strongly suggest that the much more restricted, canonical sites are actively selected for function in these Notch target 3'-UTRs, a conclusion that is bolstered by the patterns of evolutionary conservation of these sites (Lai, 2005).
These experiments presented thus far demonstrate that target gene 3'-UTRs harboring sequence elements with Watson-Crick complementarity to the 5'-ends of miRNAs are, indeed, regulated by these miRNAs in vivo, and that such sites are necessary for miRNA-mediated regulation. Are these sites sufficient for regulation by complementary miRNAs? Although a variety of studies of model sites in tissue culture assays indicate site sufficiency, tests in animals suggest that miRNA site context can be less forgiving in vivo. For example, certain reporters containing multimers of six lin-4 or three let-7 sites are not appropriately regulated by lin-4 or let-7 in nematodes. In addition, mutation of sequences outside of the let-7-binding sites in lin-41 abolishes regulation by let-7 in vivo. Therefore, it was of interest to test the functionality of GY-boxes, Brd-boxes, and K-boxes when abstracted from endogenous 3'-UTR context (Lai, 2005).
To do so, a tandem of isolated GY-box, Brd-box, and K-box elements were cloned from Bob, Bearded, and E(spl)m8, respectively, into tub-GFP transgenes. Also mutant versions were cloned containing single changes in the Brd-box sites or dual changes in the GY-boxes. The ability of these 'box' sensors to respond to exogenously expressed miRNAs was tested. It was found that wild-type GY-box, Brd-box, and K-box sensors are all negatively regulated by corresponding miRNAs. These data directly demonstrate that all three types of box sites are sufficient for miRNA-mediated negative regulation. In contrast, mutant box sensors are nonfunctional in this assay. Since the mutant box sensors contain only one or two changes in each site, these data provide strong in vivo support for the idea that Watson-Crick pairing to the 5'-end of the miRNA (the "seed") is the key essential feature of miRNA target recognition. As a further test of this idea, the ability of the three different K-box miRNAs, miR-6, miR-2, and miR-11, to down-regulate a miR-6 sensor was tested. All three inhibited miR-6 sensor expression, consistent with the ability of seed-pairing to mediate regulation by miRNAs (Lai, 2005).
With these UAS-miRNA transgenic lines in hand, the consequences of ectopically expressing miRNAs on Drosophila development were tested. It should be noted that Notch target-regulating miRNAs were fully expected to regulate other functionally unrelated targets in vivo. For example, it has been established that K-box-family miRNAs also negatively regulate the proapoptotic genes reaper, sickle, and grim via K-boxes in their 3'-UTRs, while Brd-box-family miRNAs target the mesodermal determinant bagpipe via a Brd-box in its 3'-UTR. Therefore, even if ectopic miRNAs are able to affect normal development, they would not necessarily be expected to affect Notch signaling exclusively. Nevertheless, it has been previously reported that ectopic miR-7 induces loss of molecular markers of wing margin development, resulting in wing notching. This indicates that phenotypic characterization of miRNA misexpression can be informative (Lai, 2005).
Using an independently derived UAS-miR-7 construct lacking DsRed, it was verified that dpp-Gal4>miR-7 wings display notching and loss of Cut expression at the developing wing margin of wing imaginal discs; the size of the L3-L4 intervein domain was also reduced. It was next observed that ectopic K-box miRNAs of the miR-2a-1, miR-2a-2, miR-2b-2 cluster or miR-6-1, miR-6-2, miR-6-3 cluster had similar effects on wing margin development, although two UAS-transgenes were necessary to produce this effect. Also loss of anterior crossvein and occasional L3 vein breaks was observed, although these are not indicative of loss of N signaling. More generalized expression of miR-7 using bx-Gal4 induced strong thickening of wing veins, which is indicative of compromised Notch signaling during lateral inhibition of wing veins. Expression of K-box miRNAs using bx-Gal4 had severe effects on wing development, resulting in tiny, crumpled wings. It is suspected that this results from misregulation of non-Notch-pathway-related targets. The Brd-box miRNAs miR-4 and miR-79 and the K-box miRNA miR-11 did not affect wing margin development, even when these transgenes were present in two copies, indicating that this phenotype is not generally due to misexpression of miRNAs. However, miR-79 induced strong wing curling at high levels, potentially due to misregulation of non-Notch-pathway-related targets (Lai, 2005).
Next, focus was placed on development of the adult peripheral nervous system, as exemplified by the bristle sensory organs that decorate the body surface. A classic role for Notch signaling is to restrict the number of sensory organ precursors. It was found that misexpression of miR-6 using bx-Gal4 results in a strong increase in microchaete bristle density and clustered dorsocentral macrochaetes, phenotypes that are consistent with loss of Notch signaling during lateral inhibition of sensory organ precursors. Ectopic miR-2 had a similar, but milder, effect and mostly induced clustered dorsocentral and scutellar macrochaetes. Therefore, divergent members of the K-box miRNA family have similar effects on sensory organ development, consistent with data indicating that seed-related miRNAs can regulate overlapping sets of target genes. Ectopic miR-7 also induces macrochaete tufting, which correlates with the differentiation of supernumerary sensory organ precursors in wing imaginal discs. Finally, occasional duplication of bristles was observed upon misexpression of the Brd-box miRNA mir-79, although this construct also induced occasional bristle loss. Ectopic expression of miRNAs does not in itself induce bristle defects per se, since misexpression of miR-4 or miR-11 does not interfere with bristle development (Lai, 2005).
Overall, the ability of different classes of Notch-regulating miRNAs to specifically induce phenotypes that are characteristic of Notch pathway loss of function in multiple developmental settings is a strong indication that Notch pathway targets validated in this study are key endogenous targets of these miRNAs (Lai, 2005).
It appears, therefore, that cells go through a significant amount of trouble to actively inhibit Notch signaling. Core components of the Notch pathway are subject to significant negative regulation at every step in their life cycle, including at the transcriptional, post-transcriptional, and post-translational levels. For example, in the absence of activated nuclear Notch, CSL proteins are transcriptional repressors that actively repress Notch target gene activity. In addition, multiple dedicated ubiquitin ligases promote degradation of Notch pathway components, including the receptor Notch itself. To this list, may be added transcripts of most direct Notch target genes in Drosophila that are negatively regulated by multiple families of miRNAs (Lai, 2005).
The evidence provided in this study to support this conclusion is that (1) three different classes of miRNA-binding sites (GY-boxes, Brd-boxes, and K-boxes) are pervasive among two major classes of Notch target genes; (2) all three classes of motif are selectively constrained in 3'-UTRs during evolution; (3) transcripts bearing these box sites are negatively regulated by complementary miRNAs in vivo; (4) all three classes of sites are both necessary and sufficient for miRNA-mediated regulation in vivo; and (5) ectopic expression of Notch target-regulating miRNAs phenocopies Notch pathway loss of function during multiple developmental settings. Perhaps most importantly, it has been shown that genomic transgenes specifically mutated for miRNA-binding sites are sufficiently hyperactive so as to perturb normal development of the peripheral nervous system. This places these Drosophila Notch target genes in a relatively select group of miRNA targets for which miRNA-mediated regulation is phenotypically essential for normal development (Lai, 2005).
While most of the previously characterized in vivo targets of miRNAs are of the 'extensive pairing' variety, many of the validated targets in this study display much more limited 'box:seed'-pairing to miRNAs. In fact, within the context of the set of Notch target gene 3'-UTRs, the presence of conserved GY-boxes, Brd-boxes, and K-boxes allowed for highly effective prediction of miRNA:target relationships. This is the case even without first taking into account the extent of miRNA-pairing outside of box motifs. Rapid divergence of sequences upstream of box motifs, particularly those of the Brd-box and K-box classes, further indicates that extensive pairing is not selected for in these bona fide target sites. Consistent with this, multiple lines of evidence are presented that show that divergent seed-related miRNAs can regulate overlapping sets of target in vivo. Conversely, the importance of pairing between 3'-UTR boxes to miRNA seeds was demonstrated by endogenous 3'-UTR and box sufficiency tests, where even single-nucleotide disruption of seed-pairing abolishes regulation by miRNAs in vivo (Lai, 2005).
Identification and characterization of miRNA-binding sites in these Notch target 3'-UTRs mesh well with other recent bioinformatics and experimental studies that together help to define the 'look' of miRNA-binding sites. The concept of using conserved 'boxes' with Watson-Crick complementarity to miRNA seeds to identify miRNA targets is at the heart of the TargetScanS approach. A recent study has identify statistically significant signal not only for conserved 3'-UTR sites that match positions 2-8 of the miRNA (as is characteristic of the Brd-box and GY-box), but also for matches to positions 2-7 of the miRNA (as is characteristic of the K-box). In addition, a significant bias was identified for the nucleotide corresponding to position one of the miRNA to be an adenosine in predicted target sites. Interestingly, 27/42 (64%) of GY-boxes, Brd-boxes, and K-boxes in Dm Notch target genes also have an adenosine in this position, consistent with the notion that this feature can help to identify genuine target sites. These results are also consistent with directed tests of model sites using an imaginal disc sensor assay. Together with the recent observation that miRNAs can down-regulate large numbers of transcripts that contain box:seed matches in their 3'-UTRs, a current view emerges that conserved 3'-UTR boxes that are 6-7 nt in length and complementary to the 5'-ends of miRNAs need to be considered seriously as functional regulatory sites. While seed-pairings with G:U base pairs are evidently not generally selected for, evidence is shown that rare sites of this class are functional. This is consistent with other studies that demonstrate that G:U seed-pairing impairs, but does not necessarily abolish target site function (Lai, 2005).
Finally, the presence of multiple classes of miRNA-binding sites in most Notch target gene 3'-UTRs raises the possibility of combinatorial regulation. Although this has been widely suggested as a formal possibility, extensive evidence has been provided that 3'-UTRs can bear multiple classes of functional sites. Phylogenetic considerations indicate that 10 different Notch target genes are likely regulated by multiple classes of miRNAs, and direct experimental support of this was provided for six Notch target genes. Multiple Brd-box-, K-box-, and GY-box-class miRNAs are present at high levels in the Drosophila embryo, and the Brd-box miRNA miR-4 is co-transcribed with the K-box miRNAs miR-6-1, miR-2, miR-3, suggesting that combinatorial control of Notch target genes actually occurs during normal development. Future studies are aimed at examining how different miRNA-binding sites collectively contribute to overall regulation of an individual gene (Lai, 2005).
Of the few animal miRNAs whose in vivo functions and targets are well understood, most act as genetic switches that determine binary, on/off states of target gene activity. For example, lin-4 and let-7 are temporal switches that control progression through nematode larval stages by inhibiting their targets at designated times in development. lsy-6 and miR-273 are spatial switches whose extremely restricted cell-type-specific expression patterns control neuronal identity. In these cases, temporally or spatially restricted miRNA expression is central to their control of specific processes, and each of these miRNAs appears to have a small number of key targets (Lai, 2005).
A different rationale is proposed for Brd-box and K-box miRNAs during Drosophila development. Although endogenous territories of GY-box-mediated regulation are not known, negative regulation by Brd-boxes and K-boxes appears spatially and temporally ubiquitous. Thus, Notch target transcripts of the Brd family and E(spl)bHLH families are subject to modes of miRNA-mediated regulation that operate in all cells, even though the genes themselves display highly restricted patterns of spatial expression. This suggests that these miRNAs are not dedicated to regulating Notch signal transduction, but may 'tune' the expression of many target genes. Indeed, the K-box-family miRNAs miR-2, miR-6, and miR-11 also directly regulate K-box-containing proapoptotic genes, and the Brd-box-family miRNAs miR-4 and miR-79 regulate the mesodermal determinant bagpipe. One prediction is that even though mutation of Brd-boxes and K-boxes in individual Notch target genes results in specific defects in Notch-mediated cell fate decisions, mutation of Brd-box and K-box miRNAs would have more general developmental consequences. This is supported by the observation that many, but not all, of the phenotypes induced by ectopic expression of Notch-regulating miRNAs appear to be obviously related to repression of Notch pathway activity (Lai, 2005).
An important advance of this study is the in vivo validation of a large number of biologically relevant miRNA targets that are minimally paired to miRNAs outside of the 'box:seed' region. Since modestly complementary sites are both necessary and sufficient for miRNA-mediated regulation, it might be relatively easy for novel miRNA-binding sites to arise in 'tuning' targets. Indeed, a subset of box sites has apparently newly evolved during Drosophilid radiation. In the greater context of insect Notch target genes, it appears to have been important that they be negatively regulated by miRNAs, although the precise numbers and arrangement of different sites is variable. These features of tuning targets seem to allow for highly customized regulation of individual genes (Lai, 2005).
The experimental validation of many tuning targets may be challenging or impossible to obtain where quantitative regulation is subtle. Nevertheless, minor changes in gene activity, even of a fraction of a percent, could become highly significant when selecting the fitness of individuals at the population level. Deep evolutionary profiling of related species will therefore be key to revealing the full complement of biologically important miRNA-binding sites. The data suggest that multiple classes of miRNA-binding sites can be recognized with confidence as highly conserved 3'-UTR 'boxes' complementary to miRNA seeds, and this approach has been applied to the analysis of mammalian genomes. By mid-2005, 12 Drosophila genomes will be completed, which should enable high-confidence identification of miRNA-binding sites on the genome-wide scale -- even in cases in which only 7 nt of the target are paired to a miRNA (Lai, 2005).
Recent computational work pointed to regulation of vertebrate Notch and Delta by miR-34; however, no Notch target genes were similarly singled out in various bioinformatics efforts. miR-34 is conserved in flies; however, inspection of fly Notch or its ligands Delta and Serrate failed to reveal 'boxes' that might indicate similar regulation by miR-34. Brd-box-, GY-box-, and K-box-complementary miRNAs are likewise conserved between flies and vertebrates. Are any vertebrate Notch target genes predicted to be targeted by these miRNAs by virtue of 'boxes'? Although Brd proteins have thus far been found only in insects, E(spl)bHLH proteins are conserved in and are primary effectors of Notch signaling in all vertebrates. No enrichment for Brd-boxes, GY-boxes, and K-boxes is observed across the set of vertebrate E(spl)bHLH 3'-UTRs as a whole. However, members of a specific subset of E(spl)-related repressors, named the Hey genes, contain a preponderance of these boxes in their 3'-UTRs. This appears to be the case in a variety of mammals (human, mouse, and rat) and fish (fugu and zebrafish). Therefore, miRNA-mediated regulation may be a conserved feature of Notch target genes, a scenario that is under current experimental investigation (Lai, 2005).
MicroRNAs (miRNAs) are short non-coding RNAs that regulate gene expression in plants and animals. Although their biological importance has become clear, how they recognize and regulate target genes remains less well understood. This study systematically evaluates the minimal requirements for functional miRNA-target duplexes in vivo and classes of target sites with different functional properties are distinguished. Target sites can be grouped into two broad categories. 5' dominant sites have sufficient complementarity to the miRNA 5' end to function with little or no support from pairing to the miRNA 3' end. Indeed, sites with 3' pairing below the random noise level are functional given a strong 5' end. In contrast, 3' compensatory sites have insufficient 5' pairing and require strong 3' pairing for function. Examples and genome-wide statistical support is presented to show that both classes of sites are used in biologically relevant genes. Evidence is provided that an average miRNA has approximately 100 target sites, indicating that miRNAs regulate a large fraction of protein-coding genes and that miRNA 3' ends are key determinants of target specificity within miRNA families (Brennecke, 2005).
To improve understanding of the minimal requirements for a functional miRNA target site, use was made of a simple in vivo assay in the Drosophila wing imaginal disc. A miRNA was expressed in a stripe of cells in the central region of the disc and its ability to repress the expression of a ubiquitously transcribed enhanced green fluorescent protein (EGFP) transgene containing a single target site in its 3′ UTR was assessed. The degree of repression was evaluated by comparing EGFP levels in miRNA-expressing and adjacent non-expressing cells. Expression of the miRNA strongly reduced EGFP expression from transgenes containing a single functional target site (Brennecke, 2005).
In a first series of experiments it was asked which part of the RNA duplex is most important for target regulation. A set of transgenic flies was prepared, each of which contained a different target site for miR-7 in the 3′ UTR of the EGFP reporter construct. The starting site resembled the strongest bantam miRNA site in its biological target hid and conferred strong regulation when present in a single copy in the 3′ UTR of the reporter gene. The effects were tested of introducing single nucleotide changes in the target site to produce mismatches at different positions in the duplex with the miRNA (note that the target site mismatches were the only variable in these experiments). The efficient repression mediated by the starting site was not affected by a mismatch at positions 1, 9, or 10, but any mismatch in positions 2 to 8 strongly reduced the magnitude of target regulation. Two simultaneous mismatches introduced into the 3′ region had only a small effect on target repression, increasing reporter activity from 10% to 30%. To exclude the possibility that these findings were specific for the tested miRNA sequence or duplex structure, the experiment was repeated with miR-278 and a different duplex structure. The results were similar, except that pairing of position 8 was not important for regulation in this case. Moreover, some of the mismatches in positions 2-7 still allowed repression of EGFP expression up to 50%. Taken together, these observations support previous suggestions that extensive base-pairing to the 5′ end of the miRNA is important for target site function (Brennecke, 2005).
Next the minimal 5′ sequence complementarity necessary to confer target regulation was determined. The core of 5′ sequence complementarity essential for target site recognition is referred to as the 'seed'. All possible 6mer, 5mer, and 4mer seeds complementary to the first eight nucleotides of the miRNA were tested in the context of a site that allowed strong base-pairing to the 3′ end of the miRNA. The seed was separated from a region of complete 3′ end pairing by a constant central bulge. 5mer and 6mer seeds beginning at positions 1 or 2 are functional. Surprisingly, as few as four base-pairs in positions 2-5 confers efficient target regulation under these conditions, whereas bases 1-4 are completely ineffective. 4mer, 5mer, or 6mer seeds beginning at position 3 are less effective. These results suggest that a functional seed requires a continuous helix of at least 4 or 5 nucleotides and that there is some position dependence to the pairing, since sites that produce comparable pairing energies differ in their ability to function. These experiments also indicate that extensive 3′ pairing of up to 17 nucleotides in the absence of the minimal 5′ element is not sufficient to confer regulation. Consequently, target searches based primarily on optimizing the extent of base-pairing or the total, and ranking miRNA target sites according to overall complementarity or free energy of duplex formation might not reflect their biological activity (Brennecke, 2005).
To determine the minimal lengths of 5′ seed matches that are sufficient to confer regulation alone, single sites were tested that pair with eight, seven, or six consecutive bases to the miRNA's 5′ end, but that do not pair to its 3′ end. Surprisingly, a single 8mer seed (miRNA positions 1-8) is sufficient to confer strong regulation by the miRNA. A single 7mer seed (positions 2-8) is also functional, although less effective. The magnitude of regulation for 8mer and 7mer seeds is strongly increased when two copies of the site are introduced in the UTR. In contrast, 6mer seeds show no regulation, even when present in two copies. Comparable results have been reported for two copies of an 8mer site with limited 3′ pairing capacity in a cell-based assay. These results do not support a requirement for a central bulge (Brennecke, 2005).
From these experiments it is concluded that (1) complementarity of seven or more bases to the 5′ end miRNA is sufficient to confer regulation, even if the target 3′ UTR contains only a single site; (2) sites with weaker 5′ complementarity require compensatory pairing to the 3′ end of the miRNA in order to confer regulation, and (3) extensive pairing to the 3′ end of the miRNA is not sufficient to confer regulation on its own without a minimal element of 5′ complementarity (Brennecke, 2005).
While recognizing that there is a continuum of base-pairing quality between miRNAs and target sites, the experiments presented here suggest that sites that depend critically on pairing to the miRNA 5′ end (5′ dominant sites) can be distinguished from those that cannot function without strong pairing to the miRNA 3′ end (3′ compensatory sites). The 3′ compensatory group includes seed matches of four to six base-pairs and seeds of seven or eight bases that contain G:U base-pairs, single nucleotide bulges, or mismatches (Brennecke, 2005).
It is useful to distinguish two subgroups of 5′ dominant sites: those with good pairing to both 5′ and 3′ ends of the miRNA (canonical sites) and those with good 5′ pairing but with little or no 3′ pairing (seed sites). Seed sites are considered to be those where there is no evidence for pairing of the miRNA 3′ end to nearby sequences that is better than would be expected at random. The possibility cannot be excluded that some sites identified as seed sites might be supported by additional long-range 3′ pairing. Computationally, this is always possible if long enough loops in the UTR sequence are allowed. Whether long loops are functional in vivo remains to be determined (Brennecke, 2005).
Canonical sites have strong seed matches supported by strong base-pairing to the 3′ end of the miRNA. Canonical sites can thus be seen as an extension of the seed type (with enhanced 3′ pairing in addition to a sufficient 5′ seed) or as an extension of the 3′ compensatory type (with improved 5′ seed quality in addition to sufficient 3′ pairing). Individually, canonical sites are likely to be more effective than other site types because of their higher pairing energy, and may function in one copy. Due to their lower pairing energies, seed sites are expected to be more effective when present in more than one copy (Brennecke, 2005).
Most currently identified miRNA target sites are canonical. For example, the hairy 3′ UTR contains a single site for miR-7, with a 9mer seed and a stretch of 3′ complementarity. This site has been shown to be functional in vivo , and it is strikingly conserved in the seed match and in the extent of complementarity to the 3′ end of miR-7 in all six orthologous 3′ UTRs (Brennecke, 2005).
Although seed sites have not been previously identified as functional miRNA target sites, there is some evidence that they exist in vivo. For example, the Bearded (Brd) 3′ UTR contains three sequence elements, known as Brd boxes, that are complementary to the 5′ region of miR-4 and miR-79. Brd boxes have been shown to repress expression of a reporter gene in vivo, presumably via miRNAs; expression of a Brd 3′ UTR reporter is elevated in dicer-1 mutant cells, which are unable to produce any miRNAs. All three Brd box target sites consist of 7mer seeds with little or no base-pairing to the 3′ end of either miR-4 or miR-79. The alignment of Brd 3′ UTRs shows that there is little conservation in the miR-4 or miR-79 target sites outside the seed sequence, nor is there conservation of pairing to either miRNA 3′ end. This suggests that the sequences that could pair to the 3′ end of the miRNAs are not important for regulation as they do not appear to be under selective pressure. This makes it unlikely that a yet unidentified Brd box miRNA could form a canonical site complex (Brennecke, 2005).
The 3′ UTR of the HOX gene Sex combs reduced (Scr) provides a good example of a 3′ compensatory site. Scr contains a single site for miR-10 with a 5mer seed and a continuous 11-base-pair complementarity to the miRNA 3′ end. The miR-10 transcript is encoded within the same HOX cluster downstream of Scr, a situation that resembles the relationship between miR-iab-5p and Ultrabithorax in flies and miR-196/HoxB8 in mice. The predicted pairing between miR-10 and Scr is perfectly conserved in all six drosophilid genomes, with the only sequence differences occurring in the unpaired loop region. The site is also conserved in the 3′ UTR of the Scr genes in the mosquito, Anopheles gambiae, the flour beetle, Tribolium castaneum, and the silk moth, Bombyx mori. Conservation of such a high degree of 3′ complementarity over hundreds of millions of years of evolution suggests that this is likely to be a functional miR-10 target site. Extensive 5′ and 3′ sequence conservation is also seen for other 3′ compensatory sites, e.g., the two let-7 sites in lin-41 or the miR-2 sites in grim and sickle (Brennecke, 2005).
Several families of miRNAs have been identified whose members have common 5′ sequences but differ in their 3′ ends. In view of the evidence that 5′ ends of miRNA are functionally important, and in some cases sufficient, it can be expected that members of miRNA families may have redundant or partially redundant functions. According to this model, 5′ dominant canonical and seed sites should respond to all members of a given miRNA family, whereas 3′ compensatory sites should differ in their sensitivity to different miRNA family members depending on the degree of 3′ complementarity. This is being tested using the wing disc assay with 3′ UTR reporter transgenes and overexpression constructs for various miRNA family members (Brennecke, 2005).
miR-4 and miR-79 share a common 5′ sequence that is complementary to a single 8mer seed site in the bagpipe 3′ UTR. The 3′ ends of the miRNAs differ. miR-4 is predicted to have 3′ pairing at approximately 50% of the maximally possible level (~10.8 kcal/mol), whereas the level of 3′ pairing for miR-79 is approximately 25% maximum (~6.1 kcal/mol), which is below the average level expected for random matches. Both miRNAs repressed expression of the bagpipe 3′ UTR reporter, regardless of the 3′ complementarity. This indicates that both types of site are functional in vivo and suggests that bagpipe is a target for both miRNAs in this family (Brennecke, 2005).
To test whether miRNA family members can also have non-overlapping targets, 3′ UTR reporters were used of the pro-apoptotic genes grim and sickle, two recently identified miRNA targets. Both genes contain K boxes in their 3′ UTRs that are complementary to the 5′ ends of the miR-2, miR-6, and miR-11 miRNA family. These miRNAs share residues 2-8 but differ considerably in their 3′ regions. The site in the grim 3′ UTR is predicted to form a 6mer seed match with all three miRNAs, but only miR-2 shows the extensive 3′ complementarity that would be needed for a 3′ compensatory site with a 6mer seed to function (~19.1 kcal/mol, 63% maximum 3′ pairing, versus ~10.9 kcal/mol, 46% maximum, for miR-11 and ~8.7 kcal/mol, 37% maximum, for miR-6). Indeed, only miR-2 is able to regulate the grim 3′ UTR reporter, whereas miR-6 and miR-11 are non-functional (Brennecke, 2005).
The sickle 3′ UTR contains two K boxes and provides an opportunity to test whether weak sites can function synergistically. The first site is similar to the grim 3′ UTR in that it contains a 6mer seed for all three miRNAs but extensive 3′ complementarity only to miR-2. The second site contains a 7mer seed for miR-2 and miR-6 but only a 6mer seed for miR-11. miR-2 strongly downregulates the sickle reporter, miR-6 has moderate activity (presumably via the 7mer seed site), and miR-11 has nearly no activity, even though the miRNAs were overexpressed. The fact that a site is targeted by at least one miRNA argues that it is accessible (e.g., miR-2 is able to regulate both UTR reporters), and that the absence of regulation for other family members is due to the duplex structure. These results are in line with what would be expected based on the predicted functionality of the individual sites, and indicate that the model of target site functionality can be extended to UTRs with multiple sites. Weak sites that do not function alone also do not function when they are combined (Brennecke, 2005).
To show that endogenous miRNA levels regulate all three 3′ UTR reporters, EGFP expression was compared in wild-type cells and dicer-1 mutant cells, which are unable to produce miRNAs. dicer-1 clones did not affect a control reporter lacking miRNA binding sites, but showed elevated expression of a reporter containing the 3′ UTR of the previously identified bantam miRNA target hid. Similarly, all 3′ UTR reporters above were upregulated in dicer-1 mutant cells, indicating that bagpipe, sickle, and grim are subject to repression by miRNAs expressed in the wing disc. Taken together, these experiments indicate that transcripts with 5′ dominant canonical and seed sites are likely to be regulated by all members of a miRNA family. However, transcripts with 3′ compensatory sites can discriminate between miRNA family members (Brennecke, 2005).
Experimental tests such as those presented in this study and the observed evolutionary conservation suggest that all three types of target sites are likely to be used in vivo. To gain additional evidence the occurrence of each site type was examined in all Drosophila 3′ UTRs. Use was made of the D. pseudoobscura genome, the second assembled drosophilid genome, to determine the degree of site conservation for the three different site classes in an alignment of orthologous 3′ UTRs. From the 78 known Drosophila miRNAs, a set of 49 miRNAs with non-redundant 5′ sequences was chosen. Whether sequences complementary to the miRNA 5′ ends are better conserved than would be expected for random sequences was tested. For each miRNA, a cohort of ten randomly shuffled variants was constructed. To avoid a bias for the number of possible target matches, the shuffled variants were required to produce a number of sequence matches comparable (±15%) to the original miRNAs for D. melanogaster 3′ UTRs. 7mer and 8mer seeds complementary to real miRNA 5′ ends were significantly better conserved than those complementary to the shuffled variants. Conserved 8mer seeds for real miRNAs occur on average 2.8 times as often as seeds complementary to the shuffled miRNAs. For 7mer seeds this signal was 2:1, whereas 6mer, 5mer, and 4mer seeds did not show better conservation than expected for random sequences. To assess the validity of these signals and to control for the random shuffling of miRNAs, this procedure was repeated with 'mutant' miRNAs in which two residues in the 5′ region were changed. There was no difference between the mutant test miRNAs and their shuffled variants. This indicates that a substantial fraction of the conserved 7mer and 8mer seeds complementary to real miRNAs identify biologically relevant target sites (Brennecke, 2005).
MicroRNAs are small noncoding RNAs that control gene function posttranscriptionally through mRNA degradation or translational inhibition. Much has been learned about the processing and mechanism of action of microRNAs, but little is known about their biological function. Injection of 2′O-methyl antisense oligoribonucleotides (2'OM-ORNs) into early Drosophila embryos leads to specific and efficient depletion of microRNAs and thus permits systematic loss-of-function analysis in vivo. Twenty-five of the forty-six embryonically expressed microRNAs show readily discernible defects; pleiotropy is moderate and family members display similar yet distinct phenotypes. Processes under microRNA regulation include cellularization and patterning in the blastoderm, morphogenesis, and cell survival. The largest microRNA family in Drosophila (miR-2/6/11/13/308) is required for suppressing embryonic apoptosis; this is achieved by differential posttranscriptional repression of the proapoptotic factors hid, grim, reaper, and sickle. These findings demonstrate that microRNAs act as specific and essential regulators in a wide range of developmental processes (Leaman, 2005).
miR-9 affects cellularization: Embryos injected with miR-9 antisense 2′OM-ORNs rarely form any cuticle and show virtually no internal differentiation. Examination of early embryogenesis, using phalloidin and DNA staining as well as DIC, reveal severe defects in nuclear division and migration, pole cell formation, cellularization, and in the basal movement of yolk droplets. To establish that these defects are in fact due to depletion of miR-9, whether they can be rescued by genomic overexpression of mir-9 was tested. Expression of mir-9 with a strong maternal driver (nos-Gal4VP16;UAS-mir-9a) has no effect on its own, but significantly ameliorates the phenotype of miR-9 antisense injection, confirming that a reduction in miR-9 activity is responsible for the defect. Most of the processes affected by miR-9 depletion are complex, but all share an involvement of the microtubule cytoskeleton. Therefore, miR-9 may have a single or a small number of phenocritical targets involved in microtubule function, but a more pleiotropic role cannot be excluded (Leaman, 2005).
miR-31 affects segmentation: In contrast to miR-9, miR-31 depleted embryos complete development but show severe segmentation defects. Embryos show abnormal cuticle patterns, ranging from partial fusions of denticle belts to a complete loss of alternating segments, suggesting that pattern formation is disrupted at the level of the pair rule genes. Further examination of pair rule gene expression in the blastoderm shows severe pattern abnormalities for even skipped (eve) and fushi tarazu (ftz), as well as hairy, indicating that misregulation must occur above the pair rule gene level in the segmentation gene hierarchy. Since pattern formation is affected throughout the segmented portion of the embryo, the regional gap factors are less likely to be responsible than ubiquitous or widely expressed factors such as components of the JAK/STAT pathway, Dichaete, grainy head, or Grunge (Leaman, 2005).
The miR-310 family affects dorsal closure: Embryos injected with antisense 2′OM-ORNs for the miR-310/311/312/313/92 family show morphogenetic defects in later development. In cuticle preparations, all family members show head-involution defects; in addition, miR-311 and miR-312 show mild dorsal-closure defects, and miR-313 occasional germ band-retraction defects; miR-310 and miR-313 also show occasional segmentation defects. Germ band retraction, dorsal closure, and head involution are interconnected morphogenetic processes that share the involvement of several cellular structures and pathways, including the cytoskeleton and cell junctions, and JNK and Dpp signaling. Note that despite sequence identity at positions 2–8, the members of the miR-310 family show some differences in their depletion phenotypes, suggesting that the 3′ end of the miRNA contributes to the specificity of the miRNA:mRNA pairing (Leaman, 2005).
miR-2/13 and miR-6 depletion results in catastrophic apoptosis: Embryos injected with miR-2/13 and miR-6 antisense 2′OM-ORNs fail to differentiate normal internal and external structures. At the end of embryogenesis, the embryos fall apart on touch, and no cuticle is recovered. To determine the onset of these problems, blastoderm embryos were examined, and it was found that cellularization and early pattern formation along the anteroposterior axis occur normally for both miRNAs, indicating that early fating and morphogenesis are intact. Interestingly, in miR-6, but not miR-2/13 depleted embryos, pole cell formation at the posterior end is disrupted (Leaman, 2005).
One possible cause of the catastrophic defects observed in miR-2/13 and miR-6 depleted embryos is excessive and widespread apoptosis. In both miR-2/13 and miR-6 antisense injected embryos, the number of apoptotic cells is greatly increased compared to wild-type by stage 13. Notably, the overall morphology of miR-6 depleted embryos is much more affected than that of miR-2/13 depleted embryos. miR-6 depleted embryos are generally smaller in size and have fewer and abnormally large (para-) segments, suggesting greater excess or earlier onset of apoptosis (Leaman, 2005).
To determine the specificity of the effects of miR-6 and miR-2/13 antisense injections, genomic rescue experiments were carried out. Embryos ubiquitously overexpressing mir-6 or mir-2 (Actin-Gal4;UAS-mir6-3/2b-2) show normal cell-death patterns. When injected with miR-6 or miR-2/13 antisense, they show significant rescue of miR-6 antisense by mir-6, with respect to both cell death and morphology, and of miR-2/13 antisense by mir-2. Interestingly, crossrescue of miR-6 antisense by mir-2 overexpression and of miR-2/13 antisense by mir-6 is weak (Leaman, 2005).
The miRNA sequence family miR-6 and miR-2/13 belong to has two additional members, miR-11 and miR-308. Depletion of miR-11 results in a moderate and of miR-308 in a mild increase in apoptosis in midembryogenesis. Thus, for all members of the miR-2 family, antisense-induced depletion results in excess embryonic cell death, but with marked differences in phenotypic strength. This differential could be due to differences in expression level or to sequence divergence and thus differential interaction with target mRNAs (Leaman, 2005).
The miR-2 family regulates cell survival by translational repression of proapoptotic factors: In Drosophila, three pathways are known to control caspase activity. The main control is thought to come from the proapoptotic factors Hid, Grim, and Reaper (Rpr), which are transcriptionally activated in response to a range of natural and toxic conditions; they promote caspase activation through inhibition of the caspase inhibitor Diap1. The three factors appear to act independently, with each being sufficient to drive apoptosis. When miR-2/13 and miR-6 antisense 2′OM-ORNs are injected into embryos deficient for the hid, grim, and rpr genes (H99 deficiency), they are unable to trigger apoptosis, indicating that these miRNAs act through hid, grim, and/or rpr (Leaman, 2005).
To determine whether the regulation of the three proapoptotic factors occurs at the transcriptional or at the posttranscriptional level, their RNA expression was examined in miR-2/13 and miR-6 depleted embryos using in situ hybridization and quantitative PCR. No significant increase was found in the expression level or broadening of the pattern compared to control embryos for any of three transcripts, either at embryonic stage 13 or 1 hr earlier at embryonic stage 12. By contrast, the protein expression of Hid is dramatically increased in miR-6 depleted embryos and modestly in miR-2/13 depleted embryos. These results strongly argue against a transcriptional and in favor of a posttranscriptional regulation of the proapoptotic factors by miR-2/13 and miR-6 (Leaman, 2005).
To test this directly, two existing translation control assays were adapted to the embryonic paradigm. In the first assay, full-length 3′UTRs are fused to a ubiquitously transcribed sensor (tub-GFP); transgenic embryos are injected with sense or antisense 2′OM-ORNs, and GFP fluorescence is measured. The 3′UTRs of hid, grim, rpr, and sickle (skl, a structurally related but less potent proapoptotic factor display marked differences in sensor expression, with rpr showing no expression, hid and skl low uniform expression, and grim strong and spatially modulated expression, indicating that these proapoptotic factors experience quite different levels of translation control. To gauge the efficacy of the assay, hid GFP sensor embryos were injected with bantam antisense 2′OM-ORNs, and mild but statistically significant derepression of GFP expression was found as compared to control, consistent with the weak cell-death phenotype of bantam depleted embryos. Antisense injection of miR-2 family members reveals strong derepression of the hid GFP sensor by miR-6 antisense, but not by miR-2/13, 11, or 308 antisense. Conversely, the grim GFP sensor shows significant derepression as a result of miR-2/13, 11, and 308, but not miR-6 depletion. Finally, the skl GFP sensor shows significant derepression for all four family members (Leaman, 2005).
To assess effects on rpr, a second, more sensitive assay was developed that employs transient expression of a dual-luciferase vector in injected embryos. For initial comparison with the GFP assay, a hid luciferase sensor was tested against the entire miR-2 family and the same profile was found. The rpr luciferase sensor shows strong derepression in miR-6 and 2/13, moderate derepression in miR-11, and no significant effect in miR-308 depleted embryos. Thus, the 3′UTRs of all four proapoptotic factors are subject to translational repression by the miR-2 family, but each miRNA displays a distinct interaction profile. The interaction preferences correlate well with the observed differences in phenotype: miR-6 has the most severe death phenotype and is the only family member to regulate hid, the factor with the broadest expression and the strongest proapoptotic effect. mir-2/13 and miR-11 have the same overall profile, but they differ in the strength of their interaction with rpr and show a corresponding differential in phenotypic strength. Finally, miR-308, which has the mildest death phenotype, interacts only with the weakly proapoptotic skl and with grim (Leaman, 2005).
The differences in target interaction profile between the miR-2 family members are pronounced and do not merely reproduce differences in the strength or onset of miRNA expression. This suggests that differential pairing outside the 5′ core sequence shared by all members has an important role in target selection. Computational predictions indicate that miR-2 family binding sites are present in the 3′UTRs of all four proapoptotic factors: rpr and grim have one, hid and skl two predicted sites. All six miRNA target sites lie in sequence blocks that are conserved between the six sequenced Drosophilid species, spanning an evolutionary distance of 40 Myr. Interestingly, for all sites, absolute conservation extends well beyond the bases complementary to the 5′ core of the miRNA and includes adjacent stretches suitable for pairing with the 3′ end. All but one of the sites show Watson-Crick pairing with miRNA positions 2-7 and variable pairing at the 3′ end. One of the hid sites (hid468) has a mismatch in the core but shows strong pairing with miR-6 at the 3′ end. The rules for 3′ pairing between miRNAs and their targets are not yet well understood, but it is clear that the miR-2 family members differ considerably in their ability to form 3′ matches with the six target sites. Further experimentation will be required to better understand how the observed differences in regulatory effect relate to differences in sequence pairing (Leaman, 2005).
RNA interference (RNAi) is a phylogenetically widespread gene-silencing process triggered by double-stranded RNA. In plants and Caenorhabditis elegans, two distinct populations of small RNAs have been proposed to participate in RNAi: 'Primary siRNAs' (derived from Dicer nuclease-mediated cleavage of the original trigger) and 'secondary siRNAs' [additional small RNAs whose synthesis requires an RNA-directed RNA polymerase (RdRP)]. Analyzing small RNAs associated with ongoing RNAi in C. elegans, it was found that secondary siRNAs constitute the vast majority. The bulk of secondary siRNAs exhibit structure and sequence indicative of a biosynthetic mode whereby each molecule derives from an independent de novo initiation by RdRP. Analysis of endogenous small RNAs indicated that a fraction derive from a biosynthetic mechanism that is similar to that of secondary siRNAs formed during RNAi, suggesting that small antisense transcripts derived from cellular messenger RNAs by RdRP activity may have key roles in cellular regulation (Pak, 2007).
Double-stranded RNA (dsRNA)–triggered gene silencing in eukaryotes appears universally to involve 21- to 25-nucleotide (nt) siRNA effectors. In Drosophila and mammals, siRNAs derive primarily from processing of longer duplexes by Dicer nuclease, forming 21- to 25-nt duplexes possessing 5'-monophosphates, 3'-hydroxyl groups, and 2-nt 3' overhangs. Along with this 'primary' siRNA response, amplification of the RNA trigger population has been proposed to contribute to potency and persistence of gene silencing in several systems. Amplification mechanisms are accompanied in some cases by 'transitive RNAi' phenomena in which dsRNA matching one mRNA region can silence targets bearing homology to other parts of the mRNA. Unlike the situation in plants where 'spreading' of the effector population occurs bidirectionally relative to the target mRNA, transitive RNAi in C. elegans exhibits a strong bias toward sequences upstream of trigger homology. Transitive RNAi requires function of a putative RdRP (RRF-1 in C. elegans soma, SDE1/SGS2 in Arabidopsis thaliana), suggesting several conceivable means for secondary siRNA production. One possibility is that antisense primary siRNAs could act as primers in the RdRP-mediated synthesis of new dsRNAs on an mRNA template. Alternatively, primary siRNAs may merely guide the RdRP to a target, allowing unprimed synthesis either at the cleaved end of the targeted transcript, at a location close to the trigger-target complex, or at a structure such as a free end that might be revealed as aberrant through consequences of the initial RNA-induced silencing complex (RISC)::target interaction (Pak, 2007).
To better understand signal amplification in C. elegans, small RNAs were characterized from animals undergoing RNAi against an abundantly expressed endogenous gene, sel-1. After reverse transcription, 245,420 18- to 25-nt RNAs were sequenced by means of single-molecule pyrosequencing. Among these sequences, 534 exhibited either a perfect match (428 instances) or single mismatches (106 instances) to sel-1 mRNA. A similar analysis of ~850,000 clones from animals not exposed to dsRNA yielded just one sel-1 small RNA. Most sel-1 small RNAs induced during interference (483) had an antisense orientation, consistent with previous hybridization-based analyses. Of the 51 sense strand clones, 22 showed complementarity to at least one antisense clone (Pak, 2007).
An incomplete bias was observed in siRNA positions relative to the trigger; of 138 antisense siRNAs outside the original trigger, 110 (80%) occurred on the 5' side. This bias could certainly account for preferential detection of upstream secondary responses in functional and biochemical assays. Twenty-eight observed instances of small antisense RNAs completely downstream of the trigger homology were of particular interest, since these would not have been expected if the sole mode of amplification involved extension by RdRP of existing siRNA triggers that hybridize to the target transcript (Pak, 2007).
Exon-exon junctions offer a unique opportunity to unequivocally distinguish de novo synthesis of antisense nucleic acids from an mRNA template. It was found 50 sel-1 small antisense RNA sequences that span exon/exon junctions. Of these, 43 fall within the trigger [458 base pairs (bp) of sel-1 cDNA sequence] and thus could have derived directly from triggering dsRNA. Six antisense exon-exon junction sequences upstream of the trigger were recovered (four matching perfectly and two with single mismatches). These imply de novo copying of the mature mRNA template (Pak, 2007).
The apparent scarcity of sel-1 siRNAs suggested that the procedure for cloning small RNAs (including ligation of linkers to 3' and 5' ends) might underrepresent the siRNA population. To analyze small RNA termini in detail, a number of structure-specific treatments were used. Treatment of RNA with periodate followed by ß elimination results in a shift on a denaturing acrylamide gel, indicating at least one unmodified (cis-diol) 3' terminus. Ribonuclease T (RNaseT) requires a 3'-hydroxyl to degrade single-stranded RNA. Finally, Terminator exonuclease preferentially degrades substrates with a single 5'-phosphate. Although sel-1 siRNAs are susceptible to both ß elimination and RNaseT reactions, they are resistant to Terminator. Control synthetic 25-nt sel-1 RNA oligonucleotides s with 5'-monophosphate and 3'-OH were sensitive to all three treatments. It is surmised that sel-1 siRNAs are blocked at their 5' ends (Pak, 2007).
It was next asked if a cloning protocol could be designed that would not be biased by the structure at the 5' end on an siRNA. The resulting protocol avoids both (1) the requirement for ligation of the 5' end of the RNA and (2) the possibility that modified 5' ends on small RNAs could affect enzymatic treatments of the paired cDNA strand. 127 sel-1 antisense sequences were observed and zero sense sequences from 1612 total clones using this protocol. For sel-1 antisense sequences, this represents a 40-fold enrichment compared to the 5'-ligation-dependent cloning method, providing further evidence for a prominent population of 5'-blocked siRNAs (Pak, 2007).
Secondary siRNAs are still recovered in 5'-ligation-dependent cloning, albeit inefficiently, as indicated by the representation of sequences outside the trigger (presumably most siRNAs within the trigger are also secondary). Notably, it was found that small antisense segments cloned with a 5'-ligation-independent procedure were on average 1 nt longer than those cloned with a 5'-ligation-dependent procedure. The substantial increase in incidence of sel-1 clones that followed 5'-ligation-independent cloning indicates that the vast majority of small sel-1 RNAs are modified on their 5' ends, while at most 2 to 3% have simple 5'-phosphate termini that are exposed in vivo or produced by 5' cleavage during the cloning procedure. An assumption that sense and antisense are roughly equal in the primary siRNA pool leads to primary siRNA estimates of <0.6% of the total sel-1 siRNA population and <0.05% of the total 21- to 25-nt RNAs in the animal (Pak, 2007).
The two methods of cloning were selective for different classes of endogenous small RNAs. microRNAs (miRNAs) appeared much less frequently with the 5'-ligation-independent cloning method, seemingly replaced by endogenous small RNAs corresponding to antisense sequence from coding regions. This analysis suggests that miRNAs and small antisense RNAs could be comparably abundant in C. elegans, with 5' modification of the small antisense RNAs accounting for the predominance of miRNA clones in libraries derived using ligation-dependent schemes. 612 out of 245,420 clones from the 5'-ligation-dependent method and 9 out of 1612 clones from the 5'-ligation-independent method were observed that were perfect antisense copies of exon/exon junctions, suggesting synthesis by RdRP acting on an mRNA template (Pak, 2007).
To further characterize the modification of siRNAs in C. elegans, a ligation assay and Terminator 5'-exonuclease treatment (both requiring a 5'-phosphate) were used. Sensitivity of the predominant fraction of the siRNAs could be restored by sequential treatment with alkaline phosphatase (which removes any number of 5'-phosphates) and T4 polynucleotide kinase (which adds a single 5'-phosphate), suggesting that the 5' modification was likely to involve additional 5'-phosphate groups on the siRNA. How many phosphates do these molecules have on their 5' ends? Examining relative gel mobilities of the native and dephosphorylated siRNAs, using a variety of gel porosities (and using a series of synthetic RNA markers with different numbers of phosphates), indicated that the predominant fraction of the untreated siRNAs have triphosphate 5' termini (Pak, 2007).
The results presented here define an RNA population produced de novo during RNAi in C. elegans as a pool of 5'-triphosphate–terminated small antisense molecules templated by the mature mRNA target and covering sequences both upstream and downstream of the original dsRNA trigger. The current working model for amplified gene silencing in C. elegans is that rare primary siRNAs, formed from a long dsRNA trigger, act as guides (presumably in an Argonaute-dependent manner) to recruit RdRP to targeted transcripts. This recruitment leads to de novo synthesis of short antisense RNAs that must be stripped off the template mRNA and incorporated into complexes that are capable of finding additional silencing targets. This model differs from other examples of RdRP action, such as in the generation of ta-siRNAs in A. thaliana where repeated Dicer activity on long RdRP-generated dsRNAs produces a phased distribution of small RNAs (Pak, 2007).
Although ongoing RNAi is certainly important for de novo synthesis of antisense siRNAs, this process appears to contribute by providing guidance to the RdRP rather than priming activity. Primary silencing targets may or may not be degraded; whatever their fate, however, they remain intact for a sufficient period to be substrates for RdRP activity upstream and (somewhat less efficiently) downstream of the targeting site. This model is consistent with the biochemical properties of characterized cellular RdRPs in that these enzymes are capable of unprimed (as well as primed) synthesis. Initiation at 3' ends of potential templates has been reported for fungal and plant RdRPs; this might allow initial Argonaute-mediated cleavage of mRNA targets to yield ready RdRP substrates (Pak, 2007).
Previous investigations of siRNA structure revealing double-stranded character, 3' overhangs, and 5'-monophosphate termini were performed in organisms whose genomes do not encode canonical RdRPs. In addition, crystal structures of Argonaute proteins [key executioners in the RNAi pathway indicate considerable specificity in recognizing specific 5' structures in RNA. It is possible that 5'-triphosphate antisense RNAs are themselves inactive in gene silencing, requiring either removal of two terminal phosphates or of the entire first base for activity. Alternatively, triphosphate-terminated small RNAs may be active directly as silencing triggers, potentially through distinct members of the Argonaute family that might recognize guide RNAs with a 5'-triphosphate (Pak, 2007).
Brennecke, J., Stark, A., Russell, R. B. and Cohen, S. M. (2005). Principles of microRNA-target recognition. PLoS Biol. 3(3):e85. 15723116
Kiriakidou, M., et al. (2007). An mRNA m7G cap binding-like motif within human Ago2 represses translation. Cell 129: 1141-1151. Medline abstract: 17524464
Lai, E. C., Tam, B. and Rubin, G. M. (2005). Pervasive regulation of Drosophila Notch target genes by GY-box-, Brd-box-, and K-box-class microRNAs. Genes Dev. 19: 1067-1080. 15833912
Leaman, D., et al. (2005). Antisense-mediated depletion reveals essential and specific functions of microRNAs in Drosophila development. Cell 121: 1097-1108. 15989958
Pak, J. and Fire, A. (2007). Distinct populations of primary and secondary effectors during RNAi in C. elegans. Science 315(5809): 241-4. Medline abstract: 17124291
date revised: 10 June 2007
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.