- Email: [email protected]
Structure Determinants for Accurate Processing of miR172a in Arabidopsis thaliana
Current Biology 20, 42–48, January 12, 2010 ª2010 Elsevier Ltd All rights reserved
DOI 10.1016/j.cub.2009.10.073
Report Structure Determinants for Accurate Processing of miR172a in Arabidopsis thaliana Schallum Werner,1 Heike Wollmann,1,2 Korbinian Schneeberger,1 and Detlef Weigel1,* 1Department of Molecular Biology, Max Planck Institute for Developmental Biology, 72076 Tu¨bingen, Germany
Summary Plant microRNAs (miRNAs) are processed by the RNase III-like enzyme DICER-LIKE1 acting in concert with the double-stranded RNA-binding protein HYPONASTIC LEAVES1 and the zinc finger protein SERRATE. Together, they excise a miRNA/miRNA* duplex with a 2 nucleotide 30 overhang from the primary miRNA (pri-miRNA) transcript [1–4]. pri-miRNAs include a partially self-complementary foldback or stem loop, which gives rise to the mature miRNA. In animals, pri-miRNAs are very similar, with a stereotypic position of the miRNA within the foldback. Accordingly, rules for miRNA excision from the precursor are quite simple in animals [5]. In contrast, how miRNA sequences are recognized in the structurally much more diverse foldbacks of plants is unknown. We have performed an extensive in vivo structure-function analysis of Arabidopsis thaliana pri-miRNA 172a (pri-miR172a). A junction of single-stranded and double-stranded RNA 15 nucleotides proximal from the miRNA/miRNA* duplex appears to be essential for accurate miR172a processing. This attribute is found in several other but not all plant miRNA foldbacks. In addition, we have identified features of the distal foldback structure important for miR172a processing. Our ability to engineer de novo a functional minimal miRNA precursor highlights that we have discovered several elements both necessary and sufficient for accurate miRNA processing. Results and Discussion Effects of Point Mutations on pri-miR172a Processing Efficiency Processing of animal primary microRNAs (pri-miRNAs) begins in the nucleus with the release of an w70 nucleotide-long stem loop, called precursor miRNA (pre-miRNA), with which the proximal end of the mature miRNA/miRNA* is defined. Excision of the miRNA/miRNA* duplex from the pre-miRNA, the second step, occurs in the cytoplasm (reviewed in [6]). In plants, processing occurs in the nucleus, and there is no evidence for a similar two-step mechanism (reviewed in [7]). We refer to the pre-miRNA equivalent, miRNA containing stem loop as foldback (Figure 1A; see also Figures S1a and S1b available online). Because naturally occurring small point mutations have shed light on miRNA processing in humans [8, 9], we performed mutagenesis of a miRNA precursor to identify sites in
*Correspondence: [email protected] 2Present address: Chromatin and Reproduction Group, Temasek Life Sciences Laboratory, 1 Research Link, National University of Singapore, Singapore 117604, Singapore
the miRNA foldback required for efficient processing. For an assay system we used overexpression of miR172a, which causes very early flowering in most transformants [2, 10–12]. The usefulness of this assay was initially tested by ethyl methanesulfonate mutagenesis of miR172a-overexpressing plants. Three lines, in which the early flowering phenotype was partially suppressed, had mutations in the transgene (Figure S1c). Two of the mutations were in the miRNA itself, but one was 4 base pairs proximal to the miRNA/miRNA* duplex. Because it was outside of the miRNA, it was likely to interfere with the accuracy or efficiency of processing rather than the ability of miR172a to reduce activity of its target genes, indicating the usefulness of mutagenizing the miR172a foldback. We introduced a series of additional point mutations that either disrupted base pairing or closed unpaired bases in the foldback, with the goal of identifying structural features important for miRNA processing. All point mutations were on the 50 arm of the miRNA foldback, opposite to the mature miRNA, to avoid confounding effects caused by reduced activity of miR172a itself. See Figure S3 for the effects of each mutation on foldback structure and the exact phenotype. Most point mutations did not alter pri-miR172a activity, as deduced from the early flowering phenotype of transgenic plants overexpressing the mutant constructs (Figure 1; Table S1; Figures S2–S4). Several mutations were around the first, proximal processing site, indicating a tolerance to minor structural changes at this position (e.g., mut18; see Figure S3w). In humans, a single-nucleotide polymorphism in the miRNA duplex can strongly attenuate processing of primiRNAs to pre-miRNAs [8], suggesting that the Arabidopsis thaliana processing machinery is more tolerant to structural changes inside the miRNA duplex than its animal counterpart. Processing Determinants in the Proximal Region of miR172a Foldback In the following, we divide the miRNA foldback into three parts: the proximal region, which contains three major unpaired regions between the base of the foldback and the miRNA/ miRNA* duplex; the miRNA/miRNA* duplex itself; and the distal region with the terminal loop (Figure 1A). In animals, the proximal cleavage site of the DCL1-containing processing complex is determined by the distance from the miRNA/ miRNA* duplex to the base of the foldback, which constitutes the 50 and 30 ends of the pre-miRNA. In plants, by contrast, the distance from the miRNA/miRNA* duplex to the base of the foldback is highly variable [13]. To determine the importance of features within the foldback base, we introduced several deletions (Figure 2). In the primiR172a foldback, there are three unpaired regions proximal to the miRNA/miRNA* duplex. Deletion of the entire proximal region, including only the third and largest unpaired region or the second and third unpaired regions, did not compromise pri-miR172a activity (mutants stem1 and stem2; Figure 2; Figures S3c and S3d). On the contrary, deletion of the proximal portion including the first unpaired region abolished the ability of pri-miR172a to cause early flowering (mutant stemcore; Figure 2; Figure S3e). The phenotypic effects of the mutant
miRNA Processing in Arabidopsis 43
Figure 1. miR172a Overexpression (A) Secondary structure of the miR172a foldback predicted with mfold 3.2 [29]. Arrows indicate the position of point mutations that do not affect the overexpression phenotype. ‘‘1st,’’ ‘‘2nd,’’ and ‘‘3rd’’ indicate the different proximal unpaired regions referred to in the text. The proximal cleavage site mapped by 50 RACE is indicated, with the fraction of corresponding clones given. (B) Phenotype of untransformed plants (Col-0), negative control transformed only with vector, and plants overexpressing miR172a from the wild-type precursor. (C) Flowering time of point mutants shown in (A), measured as number of leaves produced on the main stem before the first flower. Error bar indicates standard error of the mean. At least 15 T1 plants were analyzed for each construct.
transgenes were closely paralleled by the amount of miR172a detected on small RNA blots (Figure 2B). To investigate the effect of the stemcore mutation on miR172a processing, we performed 50 rapid amplification of cDNA ends (50 RACE) to map the proximal, and therefore 30 -most, cleavage site of the DCL1 processor complex in the pri-miR172a foldback (Figure S2e) [14]. Compared to wildtype pri-miR172a, the cleavage sites were shifted 12 to 13 nucleotides distally (Figure 1A; Figure 2A), which was 16 to 17 base pairs from the base of the stemcore stem loop. An intermediate cleavage site, proximal to the first processing site [15], was not detected. Concentrating on this processing-sensitive region, we introduced more subtle mutations proximally to the first unpaired region. Closing the first unpaired region (mut54) had no effect, consistent with results obtained with point mutations in this region (Figure S3ba). Closing the second unpaired region, however, led to a loss of miR172a product in mut55 due to a proximal shift of the first processing site by 8 nucleotides
(Figures 2A and 2B). Notably, the new processing site was now 15 nucleotides distal from the third unpaired region. A very interesting point mutant was mut29, in which the second unpaired region was enlarged at its distal end and the stem separating the first and second unpaired regions was correspondingly shortened (Figure 2; Figure S3ah). In mut29, not only was the main cleavage site shifted distally, but the accuracy of processing was also compromised. Together with the results from mutant stem1, this observation suggests that it is not the unpaired region per se but the transition from single-stranded RNA to double-stranded RNA that is required for positioning the DCL1 complex 14 to 15 base pairs proximal to the miRNA/miRNA* duplex (see also mut30; Figure S3ai). The intervening double-stranded stem of at least 6 base pairs can tolerate small 1 nucleotide bulges (see mut24, mut25, mut26, and mut27 in Figure 1 and Figures S3ac–S3af). An alternative explanation for the effect of mut29 on processing could be that the distance between the first and second unpaired regions was changed. The results with
Current Biology Vol 20 No 1 44
g a
A
80
u
u
u
g
c
g
a
u
c
g
u
g
c
c
u
a
c
g
a
u
U
g
c
u
a
g
c
c
g
g
c
u
g
u
a
c
g
c
g
g
c
u
a
g
c
g
u
u
a
a
u
a
u
a
u
c
g
c
c
u
a
u
a
a
a
u
u
g
c
u
a
u
c
g u u
c
100
a
u
a a
g u
g a
c c
u
mut54
c
u
g
u
a
u
20
mut55
u
a
A
c
U
g
c
u
g
c
g
g
c
a u u c
2nd
a
c
u
u
g
g
C
c
c
g 20
a
g u
140
G
g 40
g u u
mut29
1
g
C
g
U
u
G
g
C
u
A
* cg
U
u
a
u
a
g
c
40
u
st
a a u
u
140
* mut -4nt
a c
60
u
a
a
u
c
g
a
A
c
G
u
A
u
A
a
U
g a 80
u
a
g
C
a
U
a
U
c
G
u
A
a
U
c
G
u
A
a
U
c
G
u
(04/16) (04/16)
u
a a
u
G
g
C
u
A
c
U
g
c
u
g
c
g
stemcore g u u
a
g u
a
a
u
g
a
A
a
A
a
a
c
G
c
G
c
g
u
A
u
A
u
a
u
A
u
A
u
a
a
U
a
U
a
u
g
C
g
C
g
c
a
U
a
U
a
u
a
U
a
U
a
c
G
c
G
c
u
A
u
A
a
U
a
c
G
u a c
C U
u
a
c
U
a
u
c
G
c
g
A
u
A
u
a
U
a
U
a
u
G
c
G
c
g
g
C
C U G
g
C
u
A
c
U
g
c
g
U
u
G
g
C
u
A
c
U
g
c
u
g
c
g
g
c
40
u
g
c
g
g
c
u
a
u
a
a a u
g u u
1st
140
140
g
c
g
u
u
g
g
c
u
a
c
u
g
c
u
g
c
g
g
c
40
g u u
(13/16)
c
g
u
a
WT
(3/16) (1/16) (7/16)
c
u
a
g
c
g
a
u
a
u
a
u u
a
g
c
u
a
g
c
c
g
g
c
c
g
u
a
c
g
a
g
c
a
u
u
u
a
g
c
g
u
g
c
u
u
a
a
u
a
u
g
c
g
u
a
u
g
u
g
c
g
2nd
c c
stem2 c
c c 20
c
u
a
u
g
c
g
c
c
g
c
3
3
rd
a
160
c
160
c
u
u
c
u
u c
u
c
u
3
u
g
c
a
u
a
u
g
c
g
c
rd
c
u
g
c
a
u
a
u
g
c
u
5'
5'
g
c
G
g
C
g
U
u
G
g
C
u
A
u
a
u
g
c
g
c
WT
g
U
u
a
u
a
g
c
u
a
*
a
g
c
c
g
u
a
g
c
g
u
140
a
u
20
g
u
a
u c
g 160
c
u u c
2nd c
u
c
g
c
3
rd
c
u
u u
c
u
u
a
c
5'
u c u c
a u
c
u c
a
u
g
c
c
160
a
u
g
c
a
u
a
u
a
u
a
u
g
c
g
c
g
c
g
c
c
c
3'
u
c
u
u c
(13/16)
c
c
c
c
3'
g
a
c
a
c
a
3rd u
120
c
a
u
*
c
u
c u
U
a u u c
c u u
c u
c
c
c
160
c u
a
c
a
c
c
A
c
c 20
u
a
u
c
a
rd
a
c
G
a
140
2nd
c
c
U
u
u
g
u
A
c
u
c
c
u
a
u
20
c
u
g
c u u u
g
u u c u
c u u
u
a
2nd
a
u
g
a
c
G
a
u
a a u
1st
U
u
40
WT
g
a
U
a
c
20
A
g
a
u u c u
A
u
120
a
a
G
u a
c
u
A
c
u
u
a
a
g
a
g
g
u
c
u
c
a
a
a
120
u
g
a u
(1/16)
u
stem1
g u
u
g
120
g
140
u
g
c
c
c
u
a a
c
u
u 60
u a
c
a
g 60
c
u c
u
a
a
a
g
100
a g
g a
c u c
c
a
u
a
g u
c
60
u
a a
u
40
a a u
u
a
a
c
100
a
a
c
g
g u
g u
u
g a
u
u
c
g
a
WT
1st
u
a a
c
u
u
g
g
u
a
a
a
60
a
u
g
c
U
g
g u
g
C
g
u
u
u
c
100
a
c
g
a
u
c
100
g
c
a
a
c
g
u
a
c
u u
c
c
u
g c
u
a
a
c
g a
u
u
c
g
120
c
u
g
a c
g
a
u
c
g
c
g
u
a
c
g
c
c
c
u
a
c
g
u
g
a
u
c
80
c
g
a
g
u
u
g
c
u
u
u
u
g
c
g
c
c
u
u g
a
g
u
g
80
c
u
c
g
u u
g
a
u
u g
c
g
40
80
u
g
g a
g a
u
u
g a
3'
5'
3'
c
a u
u c
c
a
u
g
c
a
u
a
u
g
c
g
c
c
B
C
vector stemcore mut -4nt mut55
vector
35S::pri-miR172a
mut -4nt
mut29
mut55
mut54
vector
stemcore
stem1
3’
stem2
5’
35S::pri-miR172a
35S::pri-miR172a
mut29 mut54 stem1 stem2
miR172a
35S::miR172a 0
U6 rRNA
2
4
6
8
10
12 leaves
Figure 2. Analysis of the Proximal miR172a Foldback Region (A) Predicted secondary structures and phenotypes of mutants. Stemcore, stem1, and stem2 are mutants in which the regions below the horizontal lines were deleted. Mapped cleavage sites of mutants are indicated with blue lines, with the fraction of clones corresponding to each position given. The wild-type proximal cleavage site is indicated in gray. Arrowheads indicate mutated bases in mut54, mut55, and mut29; the small black frame indicates a deletion in mut 24nt. Black stars indicate asymmetrical, unpaired bases. (B) Small RNA blots. RNAs extracted from three biological replicates were loaded consecutively in each lane. As a loading control, U6 ribosomal RNA was used. Because of its larger size, the replicates are not clearly resolved for U6. (C) Flowering time of mutants. Error bars indicate standard error of the mean. At least 15 T1 plants were analyzed for each construct.
mut54 argue against such a scenario because mut54 eliminates the first region but does not affect processing (Figure 2). In contrast, consistent with the DCL1 complex using the second unpaired region as a landmark, closing the region in mut55 abolishes processing (Figure 2). All proximal mutants that retained the 15 base pair distance between the second unpaired region and the miRNA/miRNA* were able to induce very early flowering. In further support of the 15 base pair rule, deleting 4 base pairs of the stem just
proximal to the miRNA/miRNA* in mut 24nt caused a distal shift of the processing site (Figure 2). Importantly, the distal shift of the cleavage site in mut 24nt was 5, not 4, nucleotides. The shift by 1 extra base pair was apparently caused by the unpaired bases (indicated by asterisks in Figure 2A) located proximally to the cleavage site. Positioning of the processing site in mut 24nt suggests that asymmetrical, unpaired nucleotides do not contribute to the measured distance. We propose that this is due to the covalent
miRNA Processing in Arabidopsis 45
bond on the opposite side of the molecule, which cannot be stretched. The 15 base pair rule, however, does not apply to all A. thaliana miRNAs. Specifically, the proximal region is dispensable for the closely related plant miRNAs miR159 and miR319 [16, 17], indicating that there are different miRNA biogenesis mechanisms. Processing Determinants in the Distal Region of miR172a Foldback After identifying structural features of the proximal region important for precise miRNA processing, we investigated the contribution of the distal portion of the miR172a foldback. Minor changes of the structure of the distal region, such as pairing the single-stranded first nucleotide of the miRNA in mut01 (Figure 3A), had little effect on the ability of primiR172a to induce early flowering. Even deleting almost the entire distal portion in mut63, leaving only a single-paired base pair beyond the miRNA/miRNA* and a predicted 4 nucleotide terminal loop, did not completely abolish the miR172a overexpression phenotype (Figure 3; Figure S3bg). Somewhat surprisingly, stronger effects were seen when we disrupted the defined stem-like secondary structure at the distal processing site of the miRNA/miRNA* duplex itself in mut14 and mut15. Opening this region strongly reduced the levels of overexpressed miR172a (Figure 3; Figures S3s and S3t). The correct proximal cleavage site was, however, unaffected, suggesting that proximal and distal cleavage can be separated. mut63 supports this hypothesis. In addition to the miRNA, a larger RNA species of about 50 bases is seen, which likely corresponds to the pre-miRNA, i.e., the miRNA/miRNA* duplex including the much-shortened terminal loop (Figure 3B). Design of a Minimal miRNA Based on the insights gleaned from the mutants, we designed a minimal pri-miRNA construct called pri-amiR-CH42 (Figure 4A). The mature amiR-CH42 sequence was adopted from an artificial miRNA (amiRNA), amiR-SUL [18], that targets At4G18480 (CH42), a gene involved in chlorophyll biosynthesis [19]. Knockdown of CH42 with this amiRNA leads to an easily recognized bleaching phenotype [20]. The pattern of paired and unpaired bases of the miRNA/miRNA* duplex for amiRCH42 was copied from miR172a, because the structure of the miRNA duplex contributed very little to correct processing as long as the duplex was in a defined stemlike shape (Figure 1; Figure 3; Figure S3). The non-miRNA/miRNA* portion of the foldback was designed de novo based on the information gleaned from the mutant analyses. The proximal portion of the foldback, which was kept as short as possible, contained a 5 nucleotide unpaired region and a stem of 14 base pairs of random sequence, which ended with G:C base pairs before the unpaired region, to maximize stability. This structural feature was adopted because several miRNA foldbacks, including miR172a, have a partially paired stem of 14 to 17 base pairs proximal to the miRNA/miRNA* duplex (Figure 4B). A systematic survey of all A. thaliana miRNAs conserved in Populus trichocarpa [13], excluding the related miR159 and miR319 miRNA, which are processed by a distinct loop-to-base mechanism [21], confirmed that unpaired regions proximal to the miRNA/miRNA* duplex increase substantially in size beyond a point that is about 15 nucleotides from the duplex (Figure 4C). Distal to the miRNA/miRNA* duplex of pri-amiR-CH42, a 7 base pair stem was placed because the results with mut63 had suggested that too short a distal stem reduced processing
efficiency (Figure 3; Figure S3bg). To disentangle direct effects of amiR-CH42 on its target from those potentially caused by amplification and secondary small RNAs, we expressed priamiR-CH42 under the SUC2 (At1G22710) promoter, which is active only in phloem companion cells [22]. Plants expressing pri-amiR-CH42 showed a similar but somewhat weaker bleaching phenotype around leaf veins (Figure 4D) as controls expressing an amiRNA against CH42 from the standard miR319a backbone [18]. A small RNA blot confirmed that the predicted amiR-CH42 was produced (Figure 4E). A second, larger species was also detected. Several A. thaliana MIRNA genes have recently been shown to give rise to miRNAs of 23 to 25 nucleotide length in a DCL3-dependent fashion, in addition to the canonical 21 nucleotide species produced by DCL1, and this appears to be correlated with the length and extent of mismatches in the precursor [23]. The amiR-CH42 foldback contains two long, perfectly paired stems of 14 and 15 base pairs, which is unusual for evolutionarily old miRNAs. The production of longer miRNAs could likely be reduced by introducing mismatches in the stems. Conclusions Animal miRNA foldbacks are usually w33 base pairs long, with a proximal 11 base pair stem, the 21 base pair miRNA/miRNA* duplex, and a terminal loop at the distal end (reviewed in [6]). In plants, the miRNA-containing foldbacks are much more diverse in length and structure, with the miRNA/miRNA* duplex being at variable positions, including being very close to the proximal base of the foldback, such as in pri-miR159 and pri-miR319 [24–26]. Both miRNAs have recently been shown to be processed in an unusual loop-to-base pathway [16, 17]. Inspection of other A. thaliana pri-miRNAs did not reveal any obvious common feature, be it sequence or structure, that is present in all of them. However, several primiRNAs, such as pri-miR168a, pri-miR164a, pri-miR171a, and pri-miR390a, share with pri-miR172a a transition between a major unpaired region and a paired stem 14 to 15 base pairs proximal to the miRNA/miRNA* duplex (Figures 4D and 4E). This transition is also conserved in the foldback of A. lyrata pri-miR172a (see Figure S4). Both the structure of endogenous pri-miRNAs and some of our pri-miR172a mutants indicate that the 14 to 15 base pair stem proximal to the miRNA/miRNA* duplex tolerates small unpaired bulges or similar variants as long as the linear structure of the foldback is maintained. Similarly, the extent of pairing in the miRNA/miRNA* duplex is flexible. Only the distal processing site appears to be quite sensitive and needs to have a defined structure. In addition, distally to the miRNA/miRNA* duplex, a certain minimum length of the stem is important for accurate and efficient processing. In summary, we have identified structural features required for accurate and efficient processing of miR172a, and these appear to apply to the majority of A. thaliana miRNAs, with the notable exceptions of miR159 and miR319 [16, 17]. It will be interesting to determine whether there are additional alternative pathways for miRNA processing in plants and whether they all evolved at the same time. Experimental Procedures Plants The Columbia-0 (Col-0) accession was used as background, and plants were grown at 23 C in 16 hr light and 8 hr dark. Plants were transformed by floral dip [27]. At least 15 T1 plants were analyzed for each construct.
Current Biology Vol 20 No 1 46
A u u
u
u
g
a
u
c
g
u
g
60
c
c
u
a
c
g
c
g
a
u
c
u
a
u
u u
u
a a
g u
c
g
u
A
c
G
u
A
u
u
A
u
60
mut63
mut01
80
u
g a 80
A
c
G
A
u
A
G
u
A
u
a
U
a
g
C
a
U
u g
c
60
u
A
u
A
g
G
u
A
a
U
120
a
g
100
u
U
c
G
c
G
u
A
u
A
a
U
c
G
A
U
u
a
g
C
u
A
c
G
g
U
a
U C
A
u
G
g
u
a
U
a
U
g
C
a
U
u
A
c
U
g
c
U
u
g
G
c
g
g u u
1
U
g
c
g
c
u
a a
u
g
u g
c
g
g u u
a a u
1st g
c
u
a
u
a
g
u
2
nd
c
g
u
a
u
g
c
g
c
c
g
g
c
a
u g
c
g
u
a c
2
nd
g
u
c 20
c
a
u
g
c
g
c
u
a
g
C
c
G
U
g
C
u
G
g
U
g
C
u
G
u
A
g
C
c
U
g
c
u c
c
160
5’
U
g
g
c
g
u
g
c
c
g
g
c
a a u
a
u
g
c
a
u
c a
g
c
a
u
a
g
c
u
a
c
g
u
a
g
c
g
u
140
g
u
a
u
g
c
g
c
u
g
c
g
c
a c
3
u a
g
u
a
u
g
c
g
c
c
rd
c u
u
g
c
a
u
a
u
g
c
g
c
160
c
u
c
c
u u
u
3rd
u a
c u c
a
c
a
u u c u
u
u
u c
c
c
a
u
g
c
a
u
a
u
g
c
g
c
c 3’
u
c
c
2
160
c
c
5’
a
nd
20
u
3’
140
c
a c
g
c
u
u
c
c
g
c
u
g
g u u c u
a a u
1st
u
2 c
g u u
a
nd
a
a
(12/16)
u
c
c
c
c
u
A
u
40
u
u u
c
U
g
c 120
A
u
1
u u
G
c
st
20
c
a
U
G
c
a
u u c u
a
120
U
u
3rd
u
a
c
A
A
c
c
u u
G
u
G
a
c
U
c
c
c
c
U
a
a
u u c u
c
C
a
u
a
c 20
U
g
c
g
c a
A
a
a
g u u
c
g
140
u g A G
A
u
(13/16)
a
a
a
U
100
a
u
u
120
g
a
u
c
g
G
40
u
c
g
g a
u
a
u g A G
g
c
100
c
g
A
a a u
st
a
a
60
a
u
c
g
C
u
c
C
a
u
c
g
u
u
g a
u
a
u
u
c
g
g
c
u
a
60
a
u
g
a
80
u
a
u
U
g
a
u
a
u
g
c
g
a
c
G
a
a
c
u
c
c
u
a
40
c
g
a
u
c
u
g
c
g
C
c
mut14
c
a
c
g
g
c
C
u
c
u
a
U
A
U
g
c
g
c
c
a
u
A
a
a
g
u
g
U
40
A
u
c
u
c
g
a
g
u g A G
u
g
g
a
a
a
u
u g
a
g
u
u
mut15
a
u
c
c
A
c
g
u
u
A
u
g
a
a
u g A G
u
u
a
u
5’
3’
c
u u
u c
u c
a
u
g
c
a
u
c
c
5’
a
u
g
c
g
c
3’
B
35S::pri-miR172a
u
mut01
c
a
mut63
u
a
vector
3
u
rd
35S::pri-miR172a
60
g a
u g c
c a
c
a
a
a
a
a u
u
g
a
g u
u
c
g
u
g
a
u
a
u
c
100
a
u
c
a
60
a
a
100
u
g
c
g
u
a
c
g
100
u
a
c
c
g
a
u
u
u
u
u
g
c
C
vector mut14
vector
u
a
u
g u
mut14
80
a a
g
mut15
g a
mut15 mut63 mut01
miR172a
35S::miR172a 0
2
4
6
U6 rRNA
8
10 12 leaves
Figure 3. Analysis of the Distal miR172a Foldback Region (A) For comparison, the wild-type miR172a foldback is shown on the left; insets show details of mutants. (B) Small RNA blots. (C) Flowering time of mutants. Error bars indicate standard error of the mean. At least 15 T1 plants were analyzed for each construct. Cloning Mutants were generated from the plasmid HW042 by DpnI-mediated sitedirected mutagenesis and polymerase chain reaction amplification. Mutant precursors were recombined into pGREEN-IIS destination vectors, with the CaMV 35S promoter in front of a modified Gateway recombination cassette. pGREEN-IIS, a derivative of pGREEN-II [28], confers resistance to spectinomycin in bacteria. Sequences of constructs are available on request.
RNA Analysis Total RNA was isolated from inflorescences of at least 12 pooled T1 plants with Trizol reagent (Invitrogen) and resolved by 17% polyacrylamide gel electrophoresis under denaturing conditions (7 M urea). Small RNAs were detected with end-labeled oligonucleotide probes (DNA for U6 ribosomal RNA detection and locked nucleic acid [Exiqon] for miR172a and amiRCH42 detection). Cleavage sites in pri-miR172a were mapped by 50 RACE like described in [14] (see Figure S2e).
miRNA Processing in Arabidopsis 47
Figure 4. Minimal amiR-CH42 and Secondary Structures of MicroRNA Foldbacks (A) Predicted secondary structure of pri-amiR-CH42. (B) Examples of microRNA (miRNA) foldbacks with a similar structure as pri-miR172a, including a 14 to 15 nucleotide stem proximal to the miRNA/miRNA* duplex. The miRNA is indicated in blue, the miRNA* in gray. (C) Position weight matrix analysis for all Arabidopsis thaliana miRNAs conserved in Populus trichocarpa. X axis indicates distance from miRNA/miRNA* duplex (green boxes). Y axes indicate average run of unpaired bases. Dashed lines indicate the overall average of unpaired bases. From 15 nucleotides distal to the miRNA/miRNA* duplex, the size of unpaired regions is above the average (red). (D) Representative plants expressing amiR-SUL, produced from the natural miR319a backbone, and amiR-CH42, processed from an entirely artificial precursor. (E) Small RNA blots. The first three lanes were loaded with 4, 2, and 1 mg of RNA from a SUC2::pri-amiR-SUL line and the last two lanes with 4 mg of RNA each from two different SUC2::pri-amiR-CH42 T1 plants.
Current Biology Vol 20 No 1 48
miRNA Secondary Structure Analysis All miRNA structures were folded with mfold 3.2 [29]. For all A. thaliana miRNAs conserved in Populus trichocarpa [13], with the exception of the miR159 and miR319 families, the annotated miRNA stem sequences were extended and folded with RNAfold (ViennaRNA 1.8.3) [30]. A position weight matrix analysis was performed separately for 50 and 30 arms. A value was assigned based on the size of the unpaired region that a base participated in, with 0 for paired bases, 1 for unpaired regions of size 1, 2 for unpaired regions of size 2, etc. Supplemental Information Supplemental Information includes one table and four figures and can be found with this article online at doi:10.1016/j.cub.2009.10.073. Acknowledgments We thank J. Carrington, J. Cuperus, N. Fedoroff, J. Palatnik, S. Laubinger, L. Song, M. Todesco, and O. Voinnet for helpful discussion and sharing unpublished information, and F.F. de Felippes for the amiR-SUL line and the amiR-SUL sequence. This work was supported by a Boehringer Ingelheim Foundation doctoral fellowship (H.W.), by a Deutsche Forschungsgemeinschaft Sonderforschungsbereich 446 grant, by European Community FP6 Integrated Project SIROCCO (contract LSHG-CT-2006-037900), and by the Max Planck Society. Received: August 20, 2009 Revised: September 29, 2009 Accepted: October 29, 2009 Published online: December 14, 2009 References 1. Reinhart, B.J., Weinstein, E.G., Rhoades, M.W., Bartel, B., and Bartel, D.P. (2002). MicroRNAs in plants. Genes Dev. 16, 1616–1626. 2. Park, W., Li, J., Song, R., Messing, J., and Chen, X. (2002). CARPEL FACTORY, a Dicer homolog, and HEN1, a novel protein, act in microRNA metabolism in Arabidopsis thaliana. Curr. Biol. 12, 1484–1495. 3. Han, M.H., Goud, S., Song, L., and Fedoroff, N. (2004). The Arabidopsis double-stranded RNA-binding protein HYL1 plays a role in microRNAmediated gene regulation. Proc. Natl. Acad. Sci. USA 101, 1093–1098. 4. Vazquez, F., Gasciolli, V., Cre´te´, P., and Vaucheret, H. (2004). The nuclear dsRNA binding protein HYL1 is required for microRNA accumulation and plant development, but not posttranscriptional transgene silencing. Curr. Biol. 14, 346–351. 5. Han, J., Lee, Y., Yeom, K.H., Nam, J.W., Heo, I., Rhee, J.K., Sohn, S.Y., Cho, Y., Zhang, B.T., and Kim, V.N. (2006). Molecular basis for the recognition of primary microRNAs by the Drosha-DGCR8 complex. Cell 125, 887–901. 6. Winter, J., Jung, S., Keller, S., Gregory, R.I., and Diederichs, S. (2009). Many roads to maturity: MicroRNA biogenesis pathways and their regulation. Nat. Cell Biol. 11, 228–234. 7. Voinnet, O. (2009). Origin, biogenesis, and activity of plant microRNAs. Cell 136, 669–687. 8. Duan, R., Pak, C., and Jin, P. (2007). Single nucleotide polymorphism associated with mature miR-125a alters the processing of pri-miRNA. Hum. Mol. Genet. 16, 1124–1131. 9. Sun, G., Yan, J., Noltner, K., Feng, J., Li, H., Sarkis, D.A., Sommer, S.S., and Rossi, J.J. (2009). SNPs in human miRNA genes affect biogenesis and function. RNA 15, 1640–1651. 10. Chen, X. (2004). A microRNA as a translational repressor of APETALA2 in Arabidopsis flower development. Science 303, 2022–2025. 11. Schwab, R., Palatnik, J.F., Riester, M., Schommer, C., Schmid, M., and Weigel, D. (2005). Specific effects of microRNAs on the plant transcriptome. Dev. Cell 8, 517–527. 12. Aukerman, M.J., and Sakai, H. (2003). Regulation of flowering time and floral organ identity by a MicroRNA and its APETALA2-like target genes. Plant Cell 15, 2730–2741. 13. Fahlgren, N., Howell, M.D., Kasschau, K.D., Chapman, E.J., Sullivan, C.M., Cumbie, J.S., Givan, S.A., Law, T.F., Grant, S.R., Dangl, J.L., and Carrington, J.C. (2007). High-throughput sequencing of Arabidopsis microRNAs: Evidence for frequent birth and death of MIRNA genes. PLoS ONE 2, e219.
14. Llave, C., Xie, Z., Kasschau, K.D., and Carrington, J.C. (2002). Cleavage of Scarecrow-like mRNA targets directed by a class of Arabidopsis miRNA. Science 297, 2053–2056. 15. Kurihara, Y., and Watanabe, Y. (2004). Arabidopsis micro-RNA biogenesis through Dicer-like 1 protein functions. Proc. Natl. Acad. Sci. USA 101, 12753–12758. 16. Addo-Quaye, C., Snyder, J.A., Park, Y.B., Li, Y.F., Sunkar, R., and Axtell, M.J. (2009). Sliced microRNA targets and precise loop-first processing of MIR319 hairpins revealed by analysis of the Physcomitrella patens degradome. RNA 15, 2112–2121. 17. Bologna, N.G., Mateos, J.L., Bresso, E.G., and Palatnik, J.F. (2009). A loop-to-base processing mechanism underlies the biogenesis of plant microRNAs miR319 and miR159. EMBO J. 28, 3646–3656. 18. Schwab, R., Ossowski, S., Riester, M., Warthmann, N., and Weigel, D. (2006). Highly specific gene silencing by artificial microRNAs in Arabidopsis. Plant Cell 18, 1121–1133. 19. Koncz, C., Mayerhofer, R., Koncz-Kalman, Z., Nawrath, C., Reiss, B., Redei, G.P., and Schell, J. (1990). Isolation of a gene encoding a novel chloroplast protein by T-DNA tagging in Arabidopsis thaliana. EMBO J. 9, 1337–1346. 20. Felippes, F.F., and Weigel, D. (2009). Triggering the formation of tasiRNAs in Arabidopsis thaliana: The role of microRNA miR173. EMBO Rep. 10, 264–270. 21. Palatnik, J.F., Wollmann, H., Schommer, C., Schwab, R., Boisbouvier, J., Rodriguez, R., Warthmann, N., Allen, E., Dezulian, T., Huson, D., et al. (2007). Sequence and expression differences underlie functional specialization of arabidopsis microRNAs miR159 and miR319. Dev. Cell 13, 115–125. 22. Imlau, A., Truernit, E., and Sauer, N. (1999). Cell-to-cell and longdistance trafficking of the green fluorescent protein in the phloem and symplastic unloading of the protein into sink tissues. Plant Cell 11, 309–322. 23. Vazquez, F., Blevins, T., Ailhas, J., Boller, T., and Meins, F., Jr. (2008). Evolution of Arabidopsis MIR genes generates novel microRNA classes. Nucleic Acids Res. 36, 6429–6438. 24. Axtell, M.J., and Bartel, D.P. (2005). Antiquity of microRNAs and their targets in land plants. Plant Cell 17, 1658–1673. 25. Qi, Y., Denli, A.M., and Hannon, G.J. (2005). Biochemical specialization within Arabidopsis RNA silencing pathways. Mol. Cell 19, 421–428. 26. Warthmann, N., Das, S., Lanz, C., and Weigel, D. (2008). Comparative analysis of the MIR319a microRNA locus in Arabidopsis and related Brassicaceae. Mol. Biol. Evol. 25, 892–902. 27. Clough, S.J., and Bent, A.F. (1998). Floral dip: A simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743. 28. Hellens, R.P., Edwards, E.A., Leyland, N.R., Bean, S., and Mullineaux, P.M. (2000). pGreen: A versatile and flexible binary Ti vector for Agrobacterium-mediated plant transformation. Plant Mol. Biol. 42, 819–832. 29. Zuker, M. (2003). Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res. 31, 3406–3415. 30. Hofacker, I.L., Fontana, W., Stadler, P.F., Bonhoeffer, L.S., Tacker, M., and Schuster, P. (1994). Fast folding and comparison of RNA secondary structures. Monatsh. Chem. 125, 167–188.
DOI 10.1016/j.cub.2009.10.073
Report Structure Determinants for Accurate Processing of miR172a in Arabidopsis thaliana Schallum Werner,1 Heike Wollmann,1,2 Korbinian Schneeberger,1 and Detlef Weigel1,* 1Department of Molecular Biology, Max Planck Institute for Developmental Biology, 72076 Tu¨bingen, Germany
Summary Plant microRNAs (miRNAs) are processed by the RNase III-like enzyme DICER-LIKE1 acting in concert with the double-stranded RNA-binding protein HYPONASTIC LEAVES1 and the zinc finger protein SERRATE. Together, they excise a miRNA/miRNA* duplex with a 2 nucleotide 30 overhang from the primary miRNA (pri-miRNA) transcript [1–4]. pri-miRNAs include a partially self-complementary foldback or stem loop, which gives rise to the mature miRNA. In animals, pri-miRNAs are very similar, with a stereotypic position of the miRNA within the foldback. Accordingly, rules for miRNA excision from the precursor are quite simple in animals [5]. In contrast, how miRNA sequences are recognized in the structurally much more diverse foldbacks of plants is unknown. We have performed an extensive in vivo structure-function analysis of Arabidopsis thaliana pri-miRNA 172a (pri-miR172a). A junction of single-stranded and double-stranded RNA 15 nucleotides proximal from the miRNA/miRNA* duplex appears to be essential for accurate miR172a processing. This attribute is found in several other but not all plant miRNA foldbacks. In addition, we have identified features of the distal foldback structure important for miR172a processing. Our ability to engineer de novo a functional minimal miRNA precursor highlights that we have discovered several elements both necessary and sufficient for accurate miRNA processing. Results and Discussion Effects of Point Mutations on pri-miR172a Processing Efficiency Processing of animal primary microRNAs (pri-miRNAs) begins in the nucleus with the release of an w70 nucleotide-long stem loop, called precursor miRNA (pre-miRNA), with which the proximal end of the mature miRNA/miRNA* is defined. Excision of the miRNA/miRNA* duplex from the pre-miRNA, the second step, occurs in the cytoplasm (reviewed in [6]). In plants, processing occurs in the nucleus, and there is no evidence for a similar two-step mechanism (reviewed in [7]). We refer to the pre-miRNA equivalent, miRNA containing stem loop as foldback (Figure 1A; see also Figures S1a and S1b available online). Because naturally occurring small point mutations have shed light on miRNA processing in humans [8, 9], we performed mutagenesis of a miRNA precursor to identify sites in
*Correspondence: [email protected] 2Present address: Chromatin and Reproduction Group, Temasek Life Sciences Laboratory, 1 Research Link, National University of Singapore, Singapore 117604, Singapore
the miRNA foldback required for efficient processing. For an assay system we used overexpression of miR172a, which causes very early flowering in most transformants [2, 10–12]. The usefulness of this assay was initially tested by ethyl methanesulfonate mutagenesis of miR172a-overexpressing plants. Three lines, in which the early flowering phenotype was partially suppressed, had mutations in the transgene (Figure S1c). Two of the mutations were in the miRNA itself, but one was 4 base pairs proximal to the miRNA/miRNA* duplex. Because it was outside of the miRNA, it was likely to interfere with the accuracy or efficiency of processing rather than the ability of miR172a to reduce activity of its target genes, indicating the usefulness of mutagenizing the miR172a foldback. We introduced a series of additional point mutations that either disrupted base pairing or closed unpaired bases in the foldback, with the goal of identifying structural features important for miRNA processing. All point mutations were on the 50 arm of the miRNA foldback, opposite to the mature miRNA, to avoid confounding effects caused by reduced activity of miR172a itself. See Figure S3 for the effects of each mutation on foldback structure and the exact phenotype. Most point mutations did not alter pri-miR172a activity, as deduced from the early flowering phenotype of transgenic plants overexpressing the mutant constructs (Figure 1; Table S1; Figures S2–S4). Several mutations were around the first, proximal processing site, indicating a tolerance to minor structural changes at this position (e.g., mut18; see Figure S3w). In humans, a single-nucleotide polymorphism in the miRNA duplex can strongly attenuate processing of primiRNAs to pre-miRNAs [8], suggesting that the Arabidopsis thaliana processing machinery is more tolerant to structural changes inside the miRNA duplex than its animal counterpart. Processing Determinants in the Proximal Region of miR172a Foldback In the following, we divide the miRNA foldback into three parts: the proximal region, which contains three major unpaired regions between the base of the foldback and the miRNA/ miRNA* duplex; the miRNA/miRNA* duplex itself; and the distal region with the terminal loop (Figure 1A). In animals, the proximal cleavage site of the DCL1-containing processing complex is determined by the distance from the miRNA/ miRNA* duplex to the base of the foldback, which constitutes the 50 and 30 ends of the pre-miRNA. In plants, by contrast, the distance from the miRNA/miRNA* duplex to the base of the foldback is highly variable [13]. To determine the importance of features within the foldback base, we introduced several deletions (Figure 2). In the primiR172a foldback, there are three unpaired regions proximal to the miRNA/miRNA* duplex. Deletion of the entire proximal region, including only the third and largest unpaired region or the second and third unpaired regions, did not compromise pri-miR172a activity (mutants stem1 and stem2; Figure 2; Figures S3c and S3d). On the contrary, deletion of the proximal portion including the first unpaired region abolished the ability of pri-miR172a to cause early flowering (mutant stemcore; Figure 2; Figure S3e). The phenotypic effects of the mutant
miRNA Processing in Arabidopsis 43
Figure 1. miR172a Overexpression (A) Secondary structure of the miR172a foldback predicted with mfold 3.2 [29]. Arrows indicate the position of point mutations that do not affect the overexpression phenotype. ‘‘1st,’’ ‘‘2nd,’’ and ‘‘3rd’’ indicate the different proximal unpaired regions referred to in the text. The proximal cleavage site mapped by 50 RACE is indicated, with the fraction of corresponding clones given. (B) Phenotype of untransformed plants (Col-0), negative control transformed only with vector, and plants overexpressing miR172a from the wild-type precursor. (C) Flowering time of point mutants shown in (A), measured as number of leaves produced on the main stem before the first flower. Error bar indicates standard error of the mean. At least 15 T1 plants were analyzed for each construct.
transgenes were closely paralleled by the amount of miR172a detected on small RNA blots (Figure 2B). To investigate the effect of the stemcore mutation on miR172a processing, we performed 50 rapid amplification of cDNA ends (50 RACE) to map the proximal, and therefore 30 -most, cleavage site of the DCL1 processor complex in the pri-miR172a foldback (Figure S2e) [14]. Compared to wildtype pri-miR172a, the cleavage sites were shifted 12 to 13 nucleotides distally (Figure 1A; Figure 2A), which was 16 to 17 base pairs from the base of the stemcore stem loop. An intermediate cleavage site, proximal to the first processing site [15], was not detected. Concentrating on this processing-sensitive region, we introduced more subtle mutations proximally to the first unpaired region. Closing the first unpaired region (mut54) had no effect, consistent with results obtained with point mutations in this region (Figure S3ba). Closing the second unpaired region, however, led to a loss of miR172a product in mut55 due to a proximal shift of the first processing site by 8 nucleotides
(Figures 2A and 2B). Notably, the new processing site was now 15 nucleotides distal from the third unpaired region. A very interesting point mutant was mut29, in which the second unpaired region was enlarged at its distal end and the stem separating the first and second unpaired regions was correspondingly shortened (Figure 2; Figure S3ah). In mut29, not only was the main cleavage site shifted distally, but the accuracy of processing was also compromised. Together with the results from mutant stem1, this observation suggests that it is not the unpaired region per se but the transition from single-stranded RNA to double-stranded RNA that is required for positioning the DCL1 complex 14 to 15 base pairs proximal to the miRNA/miRNA* duplex (see also mut30; Figure S3ai). The intervening double-stranded stem of at least 6 base pairs can tolerate small 1 nucleotide bulges (see mut24, mut25, mut26, and mut27 in Figure 1 and Figures S3ac–S3af). An alternative explanation for the effect of mut29 on processing could be that the distance between the first and second unpaired regions was changed. The results with
Current Biology Vol 20 No 1 44
g a
A
80
u
u
u
g
c
g
a
u
c
g
u
g
c
c
u
a
c
g
a
u
U
g
c
u
a
g
c
c
g
g
c
u
g
u
a
c
g
c
g
g
c
u
a
g
c
g
u
u
a
a
u
a
u
a
u
c
g
c
c
u
a
u
a
a
a
u
u
g
c
u
a
u
c
g u u
c
100
a
u
a a
g u
g a
c c
u
mut54
c
u
g
u
a
u
20
mut55
u
a
A
c
U
g
c
u
g
c
g
g
c
a u u c
2nd
a
c
u
u
g
g
C
c
c
g 20
a
g u
140
G
g 40
g u u
mut29
1
g
C
g
U
u
G
g
C
u
A
* cg
U
u
a
u
a
g
c
40
u
st
a a u
u
140
* mut -4nt
a c
60
u
a
a
u
c
g
a
A
c
G
u
A
u
A
a
U
g a 80
u
a
g
C
a
U
a
U
c
G
u
A
a
U
c
G
u
A
a
U
c
G
u
(04/16) (04/16)
u
a a
u
G
g
C
u
A
c
U
g
c
u
g
c
g
stemcore g u u
a
g u
a
a
u
g
a
A
a
A
a
a
c
G
c
G
c
g
u
A
u
A
u
a
u
A
u
A
u
a
a
U
a
U
a
u
g
C
g
C
g
c
a
U
a
U
a
u
a
U
a
U
a
c
G
c
G
c
u
A
u
A
a
U
a
c
G
u a c
C U
u
a
c
U
a
u
c
G
c
g
A
u
A
u
a
U
a
U
a
u
G
c
G
c
g
g
C
C U G
g
C
u
A
c
U
g
c
g
U
u
G
g
C
u
A
c
U
g
c
u
g
c
g
g
c
40
u
g
c
g
g
c
u
a
u
a
a a u
g u u
1st
140
140
g
c
g
u
u
g
g
c
u
a
c
u
g
c
u
g
c
g
g
c
40
g u u
(13/16)
c
g
u
a
WT
(3/16) (1/16) (7/16)
c
u
a
g
c
g
a
u
a
u
a
u u
a
g
c
u
a
g
c
c
g
g
c
c
g
u
a
c
g
a
g
c
a
u
u
u
a
g
c
g
u
g
c
u
u
a
a
u
a
u
g
c
g
u
a
u
g
u
g
c
g
2nd
c c
stem2 c
c c 20
c
u
a
u
g
c
g
c
c
g
c
3
3
rd
a
160
c
160
c
u
u
c
u
u c
u
c
u
3
u
g
c
a
u
a
u
g
c
g
c
rd
c
u
g
c
a
u
a
u
g
c
u
5'
5'
g
c
G
g
C
g
U
u
G
g
C
u
A
u
a
u
g
c
g
c
WT
g
U
u
a
u
a
g
c
u
a
*
a
g
c
c
g
u
a
g
c
g
u
140
a
u
20
g
u
a
u c
g 160
c
u u c
2nd c
u
c
g
c
3
rd
c
u
u u
c
u
u
a
c
5'
u c u c
a u
c
u c
a
u
g
c
c
160
a
u
g
c
a
u
a
u
a
u
a
u
g
c
g
c
g
c
g
c
c
c
3'
u
c
u
u c
(13/16)
c
c
c
c
3'
g
a
c
a
c
a
3rd u
120
c
a
u
*
c
u
c u
U
a u u c
c u u
c u
c
c
c
160
c u
a
c
a
c
c
A
c
c 20
u
a
u
c
a
rd
a
c
G
a
140
2nd
c
c
U
u
u
g
u
A
c
u
c
c
u
a
u
20
c
u
g
c u u u
g
u u c u
c u u
u
a
2nd
a
u
g
a
c
G
a
u
a a u
1st
U
u
40
WT
g
a
U
a
c
20
A
g
a
u u c u
A
u
120
a
a
G
u a
c
u
A
c
u
u
a
a
g
a
g
g
u
c
u
c
a
a
a
120
u
g
a u
(1/16)
u
stem1
g u
u
g
120
g
140
u
g
c
c
c
u
a a
c
u
u 60
u a
c
a
g 60
c
u c
u
a
a
a
g
100
a g
g a
c u c
c
a
u
a
g u
c
60
u
a a
u
40
a a u
u
a
a
c
100
a
a
c
g
g u
g u
u
g a
u
u
c
g
a
WT
1st
u
a a
c
u
u
g
g
u
a
a
a
60
a
u
g
c
U
g
g u
g
C
g
u
u
u
c
100
a
c
g
a
u
c
100
g
c
a
a
c
g
u
a
c
u u
c
c
u
g c
u
a
a
c
g a
u
u
c
g
120
c
u
g
a c
g
a
u
c
g
c
g
u
a
c
g
c
c
c
u
a
c
g
u
g
a
u
c
80
c
g
a
g
u
u
g
c
u
u
u
u
g
c
g
c
c
u
u g
a
g
u
g
80
c
u
c
g
u u
g
a
u
u g
c
g
40
80
u
g
g a
g a
u
u
g a
3'
5'
3'
c
a u
u c
c
a
u
g
c
a
u
a
u
g
c
g
c
c
B
C
vector stemcore mut -4nt mut55
vector
35S::pri-miR172a
mut -4nt
mut29
mut55
mut54
vector
stemcore
stem1
3’
stem2
5’
35S::pri-miR172a
35S::pri-miR172a
mut29 mut54 stem1 stem2
miR172a
35S::miR172a 0
U6 rRNA
2
4
6
8
10
12 leaves
Figure 2. Analysis of the Proximal miR172a Foldback Region (A) Predicted secondary structures and phenotypes of mutants. Stemcore, stem1, and stem2 are mutants in which the regions below the horizontal lines were deleted. Mapped cleavage sites of mutants are indicated with blue lines, with the fraction of clones corresponding to each position given. The wild-type proximal cleavage site is indicated in gray. Arrowheads indicate mutated bases in mut54, mut55, and mut29; the small black frame indicates a deletion in mut 24nt. Black stars indicate asymmetrical, unpaired bases. (B) Small RNA blots. RNAs extracted from three biological replicates were loaded consecutively in each lane. As a loading control, U6 ribosomal RNA was used. Because of its larger size, the replicates are not clearly resolved for U6. (C) Flowering time of mutants. Error bars indicate standard error of the mean. At least 15 T1 plants were analyzed for each construct.
mut54 argue against such a scenario because mut54 eliminates the first region but does not affect processing (Figure 2). In contrast, consistent with the DCL1 complex using the second unpaired region as a landmark, closing the region in mut55 abolishes processing (Figure 2). All proximal mutants that retained the 15 base pair distance between the second unpaired region and the miRNA/miRNA* were able to induce very early flowering. In further support of the 15 base pair rule, deleting 4 base pairs of the stem just
proximal to the miRNA/miRNA* in mut 24nt caused a distal shift of the processing site (Figure 2). Importantly, the distal shift of the cleavage site in mut 24nt was 5, not 4, nucleotides. The shift by 1 extra base pair was apparently caused by the unpaired bases (indicated by asterisks in Figure 2A) located proximally to the cleavage site. Positioning of the processing site in mut 24nt suggests that asymmetrical, unpaired nucleotides do not contribute to the measured distance. We propose that this is due to the covalent
miRNA Processing in Arabidopsis 45
bond on the opposite side of the molecule, which cannot be stretched. The 15 base pair rule, however, does not apply to all A. thaliana miRNAs. Specifically, the proximal region is dispensable for the closely related plant miRNAs miR159 and miR319 [16, 17], indicating that there are different miRNA biogenesis mechanisms. Processing Determinants in the Distal Region of miR172a Foldback After identifying structural features of the proximal region important for precise miRNA processing, we investigated the contribution of the distal portion of the miR172a foldback. Minor changes of the structure of the distal region, such as pairing the single-stranded first nucleotide of the miRNA in mut01 (Figure 3A), had little effect on the ability of primiR172a to induce early flowering. Even deleting almost the entire distal portion in mut63, leaving only a single-paired base pair beyond the miRNA/miRNA* and a predicted 4 nucleotide terminal loop, did not completely abolish the miR172a overexpression phenotype (Figure 3; Figure S3bg). Somewhat surprisingly, stronger effects were seen when we disrupted the defined stem-like secondary structure at the distal processing site of the miRNA/miRNA* duplex itself in mut14 and mut15. Opening this region strongly reduced the levels of overexpressed miR172a (Figure 3; Figures S3s and S3t). The correct proximal cleavage site was, however, unaffected, suggesting that proximal and distal cleavage can be separated. mut63 supports this hypothesis. In addition to the miRNA, a larger RNA species of about 50 bases is seen, which likely corresponds to the pre-miRNA, i.e., the miRNA/miRNA* duplex including the much-shortened terminal loop (Figure 3B). Design of a Minimal miRNA Based on the insights gleaned from the mutants, we designed a minimal pri-miRNA construct called pri-amiR-CH42 (Figure 4A). The mature amiR-CH42 sequence was adopted from an artificial miRNA (amiRNA), amiR-SUL [18], that targets At4G18480 (CH42), a gene involved in chlorophyll biosynthesis [19]. Knockdown of CH42 with this amiRNA leads to an easily recognized bleaching phenotype [20]. The pattern of paired and unpaired bases of the miRNA/miRNA* duplex for amiRCH42 was copied from miR172a, because the structure of the miRNA duplex contributed very little to correct processing as long as the duplex was in a defined stemlike shape (Figure 1; Figure 3; Figure S3). The non-miRNA/miRNA* portion of the foldback was designed de novo based on the information gleaned from the mutant analyses. The proximal portion of the foldback, which was kept as short as possible, contained a 5 nucleotide unpaired region and a stem of 14 base pairs of random sequence, which ended with G:C base pairs before the unpaired region, to maximize stability. This structural feature was adopted because several miRNA foldbacks, including miR172a, have a partially paired stem of 14 to 17 base pairs proximal to the miRNA/miRNA* duplex (Figure 4B). A systematic survey of all A. thaliana miRNAs conserved in Populus trichocarpa [13], excluding the related miR159 and miR319 miRNA, which are processed by a distinct loop-to-base mechanism [21], confirmed that unpaired regions proximal to the miRNA/miRNA* duplex increase substantially in size beyond a point that is about 15 nucleotides from the duplex (Figure 4C). Distal to the miRNA/miRNA* duplex of pri-amiR-CH42, a 7 base pair stem was placed because the results with mut63 had suggested that too short a distal stem reduced processing
efficiency (Figure 3; Figure S3bg). To disentangle direct effects of amiR-CH42 on its target from those potentially caused by amplification and secondary small RNAs, we expressed priamiR-CH42 under the SUC2 (At1G22710) promoter, which is active only in phloem companion cells [22]. Plants expressing pri-amiR-CH42 showed a similar but somewhat weaker bleaching phenotype around leaf veins (Figure 4D) as controls expressing an amiRNA against CH42 from the standard miR319a backbone [18]. A small RNA blot confirmed that the predicted amiR-CH42 was produced (Figure 4E). A second, larger species was also detected. Several A. thaliana MIRNA genes have recently been shown to give rise to miRNAs of 23 to 25 nucleotide length in a DCL3-dependent fashion, in addition to the canonical 21 nucleotide species produced by DCL1, and this appears to be correlated with the length and extent of mismatches in the precursor [23]. The amiR-CH42 foldback contains two long, perfectly paired stems of 14 and 15 base pairs, which is unusual for evolutionarily old miRNAs. The production of longer miRNAs could likely be reduced by introducing mismatches in the stems. Conclusions Animal miRNA foldbacks are usually w33 base pairs long, with a proximal 11 base pair stem, the 21 base pair miRNA/miRNA* duplex, and a terminal loop at the distal end (reviewed in [6]). In plants, the miRNA-containing foldbacks are much more diverse in length and structure, with the miRNA/miRNA* duplex being at variable positions, including being very close to the proximal base of the foldback, such as in pri-miR159 and pri-miR319 [24–26]. Both miRNAs have recently been shown to be processed in an unusual loop-to-base pathway [16, 17]. Inspection of other A. thaliana pri-miRNAs did not reveal any obvious common feature, be it sequence or structure, that is present in all of them. However, several primiRNAs, such as pri-miR168a, pri-miR164a, pri-miR171a, and pri-miR390a, share with pri-miR172a a transition between a major unpaired region and a paired stem 14 to 15 base pairs proximal to the miRNA/miRNA* duplex (Figures 4D and 4E). This transition is also conserved in the foldback of A. lyrata pri-miR172a (see Figure S4). Both the structure of endogenous pri-miRNAs and some of our pri-miR172a mutants indicate that the 14 to 15 base pair stem proximal to the miRNA/miRNA* duplex tolerates small unpaired bulges or similar variants as long as the linear structure of the foldback is maintained. Similarly, the extent of pairing in the miRNA/miRNA* duplex is flexible. Only the distal processing site appears to be quite sensitive and needs to have a defined structure. In addition, distally to the miRNA/miRNA* duplex, a certain minimum length of the stem is important for accurate and efficient processing. In summary, we have identified structural features required for accurate and efficient processing of miR172a, and these appear to apply to the majority of A. thaliana miRNAs, with the notable exceptions of miR159 and miR319 [16, 17]. It will be interesting to determine whether there are additional alternative pathways for miRNA processing in plants and whether they all evolved at the same time. Experimental Procedures Plants The Columbia-0 (Col-0) accession was used as background, and plants were grown at 23 C in 16 hr light and 8 hr dark. Plants were transformed by floral dip [27]. At least 15 T1 plants were analyzed for each construct.
Current Biology Vol 20 No 1 46
A u u
u
u
g
a
u
c
g
u
g
60
c
c
u
a
c
g
c
g
a
u
c
u
a
u
u u
u
a a
g u
c
g
u
A
c
G
u
A
u
u
A
u
60
mut63
mut01
80
u
g a 80
A
c
G
A
u
A
G
u
A
u
a
U
a
g
C
a
U
u g
c
60
u
A
u
A
g
G
u
A
a
U
120
a
g
100
u
U
c
G
c
G
u
A
u
A
a
U
c
G
A
U
u
a
g
C
u
A
c
G
g
U
a
U C
A
u
G
g
u
a
U
a
U
g
C
a
U
u
A
c
U
g
c
U
u
g
G
c
g
g u u
1
U
g
c
g
c
u
a a
u
g
u g
c
g
g u u
a a u
1st g
c
u
a
u
a
g
u
2
nd
c
g
u
a
u
g
c
g
c
c
g
g
c
a
u g
c
g
u
a c
2
nd
g
u
c 20
c
a
u
g
c
g
c
u
a
g
C
c
G
U
g
C
u
G
g
U
g
C
u
G
u
A
g
C
c
U
g
c
u c
c
160
5’
U
g
g
c
g
u
g
c
c
g
g
c
a a u
a
u
g
c
a
u
c a
g
c
a
u
a
g
c
u
a
c
g
u
a
g
c
g
u
140
g
u
a
u
g
c
g
c
u
g
c
g
c
a c
3
u a
g
u
a
u
g
c
g
c
c
rd
c u
u
g
c
a
u
a
u
g
c
g
c
160
c
u
c
c
u u
u
3rd
u a
c u c
a
c
a
u u c u
u
u
u c
c
c
a
u
g
c
a
u
a
u
g
c
g
c
c 3’
u
c
c
2
160
c
c
5’
a
nd
20
u
3’
140
c
a c
g
c
u
u
c
c
g
c
u
g
g u u c u
a a u
1st
u
2 c
g u u
a
nd
a
a
(12/16)
u
c
c
c
c
u
A
u
40
u
u u
c
U
g
c 120
A
u
1
u u
G
c
st
20
c
a
U
G
c
a
u u c u
a
120
U
u
3rd
u
a
c
A
A
c
c
u u
G
u
G
a
c
U
c
c
c
c
U
a
a
u u c u
c
C
a
u
a
c 20
U
g
c
g
c a
A
a
a
g u u
c
g
140
u g A G
A
u
(13/16)
a
a
a
U
100
a
u
u
120
g
a
u
c
g
G
40
u
c
g
g a
u
a
u g A G
g
c
100
c
g
A
a a u
st
a
a
60
a
u
c
g
C
u
c
C
a
u
c
g
u
u
g a
u
a
u
u
c
g
g
c
u
a
60
a
u
g
a
80
u
a
u
U
g
a
u
a
u
g
c
g
a
c
G
a
a
c
u
c
c
u
a
40
c
g
a
u
c
u
g
c
g
C
c
mut14
c
a
c
g
g
c
C
u
c
u
a
U
A
U
g
c
g
c
c
a
u
A
a
a
g
u
g
U
40
A
u
c
u
c
g
a
g
u g A G
u
g
g
a
a
a
u
u g
a
g
u
u
mut15
a
u
c
c
A
c
g
u
u
A
u
g
a
a
u g A G
u
u
a
u
5’
3’
c
u u
u c
u c
a
u
g
c
a
u
c
c
5’
a
u
g
c
g
c
3’
B
35S::pri-miR172a
u
mut01
c
a
mut63
u
a
vector
3
u
rd
35S::pri-miR172a
60
g a
u g c
c a
c
a
a
a
a
a u
u
g
a
g u
u
c
g
u
g
a
u
a
u
c
100
a
u
c
a
60
a
a
100
u
g
c
g
u
a
c
g
100
u
a
c
c
g
a
u
u
u
u
u
g
c
C
vector mut14
vector
u
a
u
g u
mut14
80
a a
g
mut15
g a
mut15 mut63 mut01
miR172a
35S::miR172a 0
2
4
6
U6 rRNA
8
10 12 leaves
Figure 3. Analysis of the Distal miR172a Foldback Region (A) For comparison, the wild-type miR172a foldback is shown on the left; insets show details of mutants. (B) Small RNA blots. (C) Flowering time of mutants. Error bars indicate standard error of the mean. At least 15 T1 plants were analyzed for each construct. Cloning Mutants were generated from the plasmid HW042 by DpnI-mediated sitedirected mutagenesis and polymerase chain reaction amplification. Mutant precursors were recombined into pGREEN-IIS destination vectors, with the CaMV 35S promoter in front of a modified Gateway recombination cassette. pGREEN-IIS, a derivative of pGREEN-II [28], confers resistance to spectinomycin in bacteria. Sequences of constructs are available on request.
RNA Analysis Total RNA was isolated from inflorescences of at least 12 pooled T1 plants with Trizol reagent (Invitrogen) and resolved by 17% polyacrylamide gel electrophoresis under denaturing conditions (7 M urea). Small RNAs were detected with end-labeled oligonucleotide probes (DNA for U6 ribosomal RNA detection and locked nucleic acid [Exiqon] for miR172a and amiRCH42 detection). Cleavage sites in pri-miR172a were mapped by 50 RACE like described in [14] (see Figure S2e).
miRNA Processing in Arabidopsis 47
Figure 4. Minimal amiR-CH42 and Secondary Structures of MicroRNA Foldbacks (A) Predicted secondary structure of pri-amiR-CH42. (B) Examples of microRNA (miRNA) foldbacks with a similar structure as pri-miR172a, including a 14 to 15 nucleotide stem proximal to the miRNA/miRNA* duplex. The miRNA is indicated in blue, the miRNA* in gray. (C) Position weight matrix analysis for all Arabidopsis thaliana miRNAs conserved in Populus trichocarpa. X axis indicates distance from miRNA/miRNA* duplex (green boxes). Y axes indicate average run of unpaired bases. Dashed lines indicate the overall average of unpaired bases. From 15 nucleotides distal to the miRNA/miRNA* duplex, the size of unpaired regions is above the average (red). (D) Representative plants expressing amiR-SUL, produced from the natural miR319a backbone, and amiR-CH42, processed from an entirely artificial precursor. (E) Small RNA blots. The first three lanes were loaded with 4, 2, and 1 mg of RNA from a SUC2::pri-amiR-SUL line and the last two lanes with 4 mg of RNA each from two different SUC2::pri-amiR-CH42 T1 plants.
Current Biology Vol 20 No 1 48
miRNA Secondary Structure Analysis All miRNA structures were folded with mfold 3.2 [29]. For all A. thaliana miRNAs conserved in Populus trichocarpa [13], with the exception of the miR159 and miR319 families, the annotated miRNA stem sequences were extended and folded with RNAfold (ViennaRNA 1.8.3) [30]. A position weight matrix analysis was performed separately for 50 and 30 arms. A value was assigned based on the size of the unpaired region that a base participated in, with 0 for paired bases, 1 for unpaired regions of size 1, 2 for unpaired regions of size 2, etc. Supplemental Information Supplemental Information includes one table and four figures and can be found with this article online at doi:10.1016/j.cub.2009.10.073. Acknowledgments We thank J. Carrington, J. Cuperus, N. Fedoroff, J. Palatnik, S. Laubinger, L. Song, M. Todesco, and O. Voinnet for helpful discussion and sharing unpublished information, and F.F. de Felippes for the amiR-SUL line and the amiR-SUL sequence. This work was supported by a Boehringer Ingelheim Foundation doctoral fellowship (H.W.), by a Deutsche Forschungsgemeinschaft Sonderforschungsbereich 446 grant, by European Community FP6 Integrated Project SIROCCO (contract LSHG-CT-2006-037900), and by the Max Planck Society. Received: August 20, 2009 Revised: September 29, 2009 Accepted: October 29, 2009 Published online: December 14, 2009 References 1. Reinhart, B.J., Weinstein, E.G., Rhoades, M.W., Bartel, B., and Bartel, D.P. (2002). MicroRNAs in plants. Genes Dev. 16, 1616–1626. 2. Park, W., Li, J., Song, R., Messing, J., and Chen, X. (2002). CARPEL FACTORY, a Dicer homolog, and HEN1, a novel protein, act in microRNA metabolism in Arabidopsis thaliana. Curr. Biol. 12, 1484–1495. 3. Han, M.H., Goud, S., Song, L., and Fedoroff, N. (2004). The Arabidopsis double-stranded RNA-binding protein HYL1 plays a role in microRNAmediated gene regulation. Proc. Natl. Acad. Sci. USA 101, 1093–1098. 4. Vazquez, F., Gasciolli, V., Cre´te´, P., and Vaucheret, H. (2004). The nuclear dsRNA binding protein HYL1 is required for microRNA accumulation and plant development, but not posttranscriptional transgene silencing. Curr. Biol. 14, 346–351. 5. Han, J., Lee, Y., Yeom, K.H., Nam, J.W., Heo, I., Rhee, J.K., Sohn, S.Y., Cho, Y., Zhang, B.T., and Kim, V.N. (2006). Molecular basis for the recognition of primary microRNAs by the Drosha-DGCR8 complex. Cell 125, 887–901. 6. Winter, J., Jung, S., Keller, S., Gregory, R.I., and Diederichs, S. (2009). Many roads to maturity: MicroRNA biogenesis pathways and their regulation. Nat. Cell Biol. 11, 228–234. 7. Voinnet, O. (2009). Origin, biogenesis, and activity of plant microRNAs. Cell 136, 669–687. 8. Duan, R., Pak, C., and Jin, P. (2007). Single nucleotide polymorphism associated with mature miR-125a alters the processing of pri-miRNA. Hum. Mol. Genet. 16, 1124–1131. 9. Sun, G., Yan, J., Noltner, K., Feng, J., Li, H., Sarkis, D.A., Sommer, S.S., and Rossi, J.J. (2009). SNPs in human miRNA genes affect biogenesis and function. RNA 15, 1640–1651. 10. Chen, X. (2004). A microRNA as a translational repressor of APETALA2 in Arabidopsis flower development. Science 303, 2022–2025. 11. Schwab, R., Palatnik, J.F., Riester, M., Schommer, C., Schmid, M., and Weigel, D. (2005). Specific effects of microRNAs on the plant transcriptome. Dev. Cell 8, 517–527. 12. Aukerman, M.J., and Sakai, H. (2003). Regulation of flowering time and floral organ identity by a MicroRNA and its APETALA2-like target genes. Plant Cell 15, 2730–2741. 13. Fahlgren, N., Howell, M.D., Kasschau, K.D., Chapman, E.J., Sullivan, C.M., Cumbie, J.S., Givan, S.A., Law, T.F., Grant, S.R., Dangl, J.L., and Carrington, J.C. (2007). High-throughput sequencing of Arabidopsis microRNAs: Evidence for frequent birth and death of MIRNA genes. PLoS ONE 2, e219.
14. Llave, C., Xie, Z., Kasschau, K.D., and Carrington, J.C. (2002). Cleavage of Scarecrow-like mRNA targets directed by a class of Arabidopsis miRNA. Science 297, 2053–2056. 15. Kurihara, Y., and Watanabe, Y. (2004). Arabidopsis micro-RNA biogenesis through Dicer-like 1 protein functions. Proc. Natl. Acad. Sci. USA 101, 12753–12758. 16. Addo-Quaye, C., Snyder, J.A., Park, Y.B., Li, Y.F., Sunkar, R., and Axtell, M.J. (2009). Sliced microRNA targets and precise loop-first processing of MIR319 hairpins revealed by analysis of the Physcomitrella patens degradome. RNA 15, 2112–2121. 17. Bologna, N.G., Mateos, J.L., Bresso, E.G., and Palatnik, J.F. (2009). A loop-to-base processing mechanism underlies the biogenesis of plant microRNAs miR319 and miR159. EMBO J. 28, 3646–3656. 18. Schwab, R., Ossowski, S., Riester, M., Warthmann, N., and Weigel, D. (2006). Highly specific gene silencing by artificial microRNAs in Arabidopsis. Plant Cell 18, 1121–1133. 19. Koncz, C., Mayerhofer, R., Koncz-Kalman, Z., Nawrath, C., Reiss, B., Redei, G.P., and Schell, J. (1990). Isolation of a gene encoding a novel chloroplast protein by T-DNA tagging in Arabidopsis thaliana. EMBO J. 9, 1337–1346. 20. Felippes, F.F., and Weigel, D. (2009). Triggering the formation of tasiRNAs in Arabidopsis thaliana: The role of microRNA miR173. EMBO Rep. 10, 264–270. 21. Palatnik, J.F., Wollmann, H., Schommer, C., Schwab, R., Boisbouvier, J., Rodriguez, R., Warthmann, N., Allen, E., Dezulian, T., Huson, D., et al. (2007). Sequence and expression differences underlie functional specialization of arabidopsis microRNAs miR159 and miR319. Dev. Cell 13, 115–125. 22. Imlau, A., Truernit, E., and Sauer, N. (1999). Cell-to-cell and longdistance trafficking of the green fluorescent protein in the phloem and symplastic unloading of the protein into sink tissues. Plant Cell 11, 309–322. 23. Vazquez, F., Blevins, T., Ailhas, J., Boller, T., and Meins, F., Jr. (2008). Evolution of Arabidopsis MIR genes generates novel microRNA classes. Nucleic Acids Res. 36, 6429–6438. 24. Axtell, M.J., and Bartel, D.P. (2005). Antiquity of microRNAs and their targets in land plants. Plant Cell 17, 1658–1673. 25. Qi, Y., Denli, A.M., and Hannon, G.J. (2005). Biochemical specialization within Arabidopsis RNA silencing pathways. Mol. Cell 19, 421–428. 26. Warthmann, N., Das, S., Lanz, C., and Weigel, D. (2008). Comparative analysis of the MIR319a microRNA locus in Arabidopsis and related Brassicaceae. Mol. Biol. Evol. 25, 892–902. 27. Clough, S.J., and Bent, A.F. (1998). Floral dip: A simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743. 28. Hellens, R.P., Edwards, E.A., Leyland, N.R., Bean, S., and Mullineaux, P.M. (2000). pGreen: A versatile and flexible binary Ti vector for Agrobacterium-mediated plant transformation. Plant Mol. Biol. 42, 819–832. 29. Zuker, M. (2003). Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res. 31, 3406–3415. 30. Hofacker, I.L., Fontana, W., Stadler, P.F., Bonhoeffer, L.S., Tacker, M., and Schuster, P. (1994). Fast folding and comparison of RNA secondary structures. Monatsh. Chem. 125, 167–188.