MicroR159 regulation of most conserved targets in Arabidopsis has negligible phenotypic effects
- Robert S Allen†1, 2, 3,
- Junyan Li†1, 2,
- Maria M Alonso-Peral1,
- Rosemary G White2,
- Frank Gubler2 and
- Anthony A Millar1Email author
© Allen et al; licensee BioMed Central Ltd. 2010
Received: 23 July 2010
Accepted: 28 October 2010
Published: 28 October 2010
A current challenge of microRNA (miRNA) research is the identification of biologically relevant miRNA:target gene relationships. In plants, high miRNA:target gene complementarity has enabled accurate target predictions, and slicing of target mRNAs has facilitated target validation through rapid amplification of 5' cDNA ends (5'-RACE) analysis. Together, these approaches have identified more than 20 targets potentially regulated by the deeply conserved miR159 family in Arabidopsis, including eight MYB genes with highly conserved miR159 target sites. However, genetic analysis has revealed the functional specificity of the major family members, miR159a and miR159b is limited to only two targets, MYB33 and MYB65. Here, we examine the functional role of miR159 regulation for the other potential MYB target genes.
For these target genes, functional analysis failed to identify miR159 regulation that resulted in any major phenotypic impact, either at the morphological or molecular level. This appears to be mainly due to the quiescent nature of the remaining family member, MIR159c. Although its expression overlaps in a temporal and spatial cell-specific manner with a subset of these targets in anthers, the abundance of miR159c is extremely low and concomitantly a mir159c mutant displays no anther defects. Examination of potential miR159c targets with conserved miR159 binding sites found neither their spatial or temporal expression domains appeared miR159 regulated, despite the detection of miR159-guided cleavage products by 5'-RACE. Moreover, expression of a miR159-resistant target (mMYB101) resulted predominantly in plants that are indistinguishable from wild type. Plants that displayed altered morphological phenotypes were found to be ectopically expressing the mMYB101 transgene, and hence were misrepresentative of the in vivo functional role of miR159.
This study presents a novel explanation for a paradox common to plant and animal miRNA systems, where among many potential miRNA-target relationships usually only a few appear physiologically relevant. The identification of a quiescent miR159c:target gene regulatory module in anthers provides a likely rationale for the presence of conserved miR159 binding sites in many targets for which miR159 regulation has no obvious functional role. Remnants from the demise of such modules may lead to an overestimation of miRNA regulatory complexity when investigated using bioinformatic, 5'-RACE or transgenic approaches.
MicroRNAs (miRNAs) regulate gene expression by guiding the RNA induced silencing complex (RISC) to gene targets via base pairing complementarity . For most plant miRNAs, their target mRNAs contain motifs that have perfect/near perfect complementarity resulting in a regulatory mechanism that includes RISC-directed slicing . Due to these high sequence complementarity requirements, it has been relatively easy to bioinformatically predict potential targets for a particular miRNA in plants . For ancient miRNAs, conservation of target motifs over long evolutionary distances have further aided in the identification of targets, and highlighted the importance of the miRNA:target interaction . Moreover, a hallmark of high complementarity miRNA-mediated regulation has been the isolation of miRNA-guided target mRNA cleavage products by rapid amplification of 5' complementary DNA ends (5'-RACE) methods . The detection of these products designates such mRNAs as experimentally validated miRNA targets [5–7]. This has now been extrapolated onto a genome wide scale, where the sequencing of plant degradomes has identified many miRNA targets [8–10]. With the recent proliferation of genomic data from many plant species, bioinformatics and degradome data will be at the forefront of predicting and identifying a diverse array of new miRNAs and their targets with a high degree of confidence [11, 12].
However, perhaps a more important but overlooked question relates to the identification of miRNA:target relationships that are biologically relevant . Although the commonly taken approaches described above have validated many miRNA:target interactions, the in vivo relevance of such interactions is often difficult to assess. Notably, in the few plant and animal miRNA studies where specific miRNAs have been removed or mutated, most phenotypic abnormalities can be attributed to deregulation of a discrete subset of targets [14–18]. This apparent paradox is hard to reconcile with an abundance of bioinformatic and molecular data that argues many validated targets may exist for a particular miRNA.
This also applies for the miR159 family in Arabidopsis . This family has three different members, miR159a, miR159b and miR159c [19, 20], and represents one of the most ancient miRNAs in the plant kingdom , and it is also the most abundant miRNA family in Arabidopsis [22–24]. Extensive bioinformatic and molecular analysis predicts this family has potentially 20 or more target genes (Additiona file1). This includes eight genes encoding R2R3 MYB domain proteins that have miR159 binding sites that are strongly conserved in both monocot and dicot species. The miR159 family is also related to the miR319 family [19, 20], yet sequence differences prevent miR159 regulating targets of miR319 , while the spatial expression domains and low abundance of miR319 family members prevents major regulation of miR159 targets .
To determine the functional role of miR159 regulation, loss-of-function mutations in the two predominantly expressed miR159 members, miR159a and miR159b, were obtained. Functionally redundancy was demonstrated, as a mir159ab double mutant displayed strong pleiotropic developmental defects not apparent in either single mutant . However only two targets, MYB33 and MYB65 were deregulated in mir159ab. The biological importance of this relationship was implicitly demonstrated by suppression of all pleiotropic mir159ab phenotypes in a quadruple mir159ab/myb33/myb65 mutant, implying that miR159a/miR159b are functionally specific for only 2 of the 20 predicted/validated target genes . This specificity was reconciled by the observation that many targets appear to have mutually exclusive transcriptional domains when compared to that of miR159a/b; they were transcribed predominantly in anthers where the near ubiquitous miR159a/b appeared absent. Therefore MYB33 and MYB65 would be considered switch targets, whereas the other targets would be considered neutral targets .
This raises the question of what is the selective pressure driving miR159 target site conservation in these other potential target genes. Confounding this, multiple lines of evidence exist suggesting the other targets are regulated by miR159, including five genes that have been validated by 5'-RACE (Additional file 1; [25, 27, 28]). For instance, MYB101, the closest related gene to MYB33/MYB65, contains a highly conserved miR159 binding site, is downregulated in flowers of 35S:MIR159a Arabidopsis plants  and miR159-guided cleavage products corresponding to MYB101 have been isolated validating it as a genuine miR159 target [25, 27, 28].
There are numerous possibilities for the requirement of miR159-binding sites in these other potential targets. Firstly, they may be regulated by the third member of the family, miR159c. Although lowly abundant due to probable poor processing , deep sequencing indicates miR159c is still expressed , and at similar levels to other miRNAs, such as miR164c, which is required for proper floral development . Secondly, like myb33.myb65 , the mir159ab/myb33.myb65 mutant was male sterile that may mask any potentially important miR159 regulation in the anther. This is important as of the 20 or so potential miR159 targets, 12 of them are expressed in anthers/pollen or associated with their development (Additional file 1). Another possibility is that mir159ab may only be a hypomorphic mutant, able to produce miR159 at levels sufficient to silence most target genes, but not at high enough levels to fully suppress MYB33 and MYB65. Alternatively, environmental conditions may exist where miR159 is induced, repressing these other target genes. This could include the possibility that miR159 acts as a safeguard against these anther transcribed genes, attenuating 'leaky' transcription that may occur in other tissues, a theme that is common in animals .
From exploring these possibilities, our analyses suggest that miR159c has subfunctionalised and corresponds to a distinct anther miR159c:target gene regulatory module. However in Arabidopsis the activity of MIR159c is so weak it has negligible impact on its potential target genes. Thus the remnants of this miR159c regulatory module may provide an explanation for the presence of conserved miRNA target sites in genes that appear principally independent of miR159 regulation. Our study highlights the need for loss-of-function analysis in identifying key miRNA:target relationships and that in fact other methodologies may be overestimating the scope and extent of miRNA regulation.
A mir159abc triple mutant appears indistinguishable from mir159ab
To determine the functional importance of MIR159c, the transfer DNA (T-DNA) insertional mutant SAIL_248_G11 was obtained from the SIGnAL database , hereafter designated mir159c. Isolation and sequencing of both T-DNA junctions found the T-DNA had integrated in an inverse tandem manner within the stem-loop region (Figure 1a). Although endogenous levels of pri-MIR159c are extremely low, elevated pri-mir159c transcript was detected in mir159c (Figure 1b), suggesting that the T-DNA is affecting the transcription and/or processing of this gene. Given the location of the T-DNA between the miR159c and miR159c* sequences, any transcript from this allele would be unlikely to form the secondary structure required to enable processing into mature miR159c and hence mir159c is most likely a null allele. Despite this, the morphological phenotype of mir159c was indistinguishable from wild type and a mir159abc triple mutant was indistinguishable from mir159ab (Figure 1c), the latter suggesting that no additional redundancy between miR159a/b and miR159c exists.
mir159abcrepresents a very strong loss-of-function miR159 mutant
Next, we determined the precise extent to which miR159 has been downregulated in mir159abc. Previously northern blotting of mir159ab using a miR159a probe failed to detect expression of miR159 in mir159ab , however the mir159a allele may still retain some activity as the T-DNA had inserted outside of the stem-loop region, whereas mir159b and mir159c were assumed to be null alleles because of the stem-loop location of the T-DNA. Because the sequences of the three miR159 members are highly similar (Additional file 1), conventional northern blotting would not be expected to easily differentiate between the three family members. Therefore we used ABI TaqMan microRNA quantitative real-time stem-loop PCR assays (qSL-PCR; http://www.appliedbiosystems.com/) to assess the miR159 levels in all three mir159 mutants.
To assess this, assays were performed on the single mir159 mutants (Figure 2a). This revealed the assays generally had high, but not absolute specificity. For example, comparison of miR159a levels in mir159a (0.24) with mir159ab (0.06) indicates the miR159a TaqMan assay crossreacts with miR159b in mir159a. Similarly, the levels of miR159c found in wild type (0.36) and mir159c (0.32) are likely representative of the miR159c assay crossreacting with miR159a and miR159b, as demonstrated by the level of miR159c in mir159ab (0.04). Therefore to accurately measure the reduction of miR159 members in mir159a, mir159b and mir159c, a comparison between wild type, mir159ab, and mir159abc was required. It was found both mutant alleles of mir159a and mir159b were expressing their respective miR159 products to less than 1% of wild-type levels (Figure 2a). For mir159b, this confirms that the mutant is a likely null allele . For mir159a, the low level of miR159a from this locus was somewhat surprising, given that the T-DNA insertion was outside the stem-loop region and only reduced pri-mir159a levels sixfold , suggesting this T-DNA insertion reduces both transcription and processing of pri-MIR159a, leading to what would be considered a strong loss-of-function mir159a allele. Therefore, mir159ab does not represent a miR159 hypomorphic mutant, but instead is a strong loss-of-function mutant. For mir159c, assays were performed on mir159ab and mir159abc and no differences in the level of miR159c between the two genotypes could be found (Figure 2b), demonstrating the level of miR159c measured in inflorescences is so low that it is beyond the detection limits of qSL-PCR analysis.
A 35S:MIR159c transgene is a potent silencer of MYB33 and MYB65
Expression of MIR159cis limited to a narrow range of cell types
Strikingly, the expression pattern of MIR159c:GUS in inflorescences appeared reciprocal to the MIR159a:GUS and MIR159b:GUS transgenes, being expressed only in anthers (Figure 4b). This β-glucuronidase (GUS) expression was temporally controlled, being absent in young anthers, but visible in floral stages 6-12 and then weakening again in mature anthers. Transverse sections of anthers found staining was restricted to the tapetal cell layer in postmeiotic anthers, with staining strongest during the microspore stage of development (Figure 4c). Degeneration of the tapetum coincided with the loss of staining, hence explaining the transient nature of expression in anthers. In vegetative tissues, the MIR159c:GUS transgene was specifically expressed in hydathodes and in the shoot apical region with no other cell type showing any staining (Figure 4d). Restriction of MIR159c expression to these specific cell types provides part of the explanation for the low abundance of miR159c when compared to the broadly transcribed MIR159a and MIR159b genes .
MYB335'-RACE product abundance appears higher than other miR159 targets
As miR159a and miR159b are functionally specific for MYB33/MYB65, and mir159c does not display a mutant phenotype, the question arose as to whether the other MYB target genes with conserved miR159 binding sites were in fact miR159 regulated. As many of these MYB genes are predominantly transcribed in anthers and/or pollen (Additional file 1), 5'-RACE cleavage assays of genes using RNA isolated from inflorescence tissues was carried out. Furthermore to gain insight into what extent miR159c-guided cleavage may be involved in regulating these MYB genes, this analysis was also performed on mir159ab. To ensure that the integrity of the 5'-RACE cDNA was equivalent between wild type and mir159ab, PCR using MYB101 and MYB81 primers annealing downstream of the 5'-RACE adapter showed similar levels of amplification (Additional file 3, Figure S2).
miR159-guided cleavage products for MYB33, MYB101, MYB81, MYB120 and DUO1 were cloned with far lower frequency from mir159ab 5'-RACE cDNA. This suggests that cleavage of these MYB genes is carried out primarily by miR159a/miR159b. Consistent with this, mRNA abundance of these MYB genes was unchanged in the inflorescences of mir159c when compared to wild type (Figure 5h), demonstrating that the levels of miR159c present in wild-type anthers and any slicing it performs is insufficient to have an impact on mRNA abundance of these target genes.
Finally, the 5'-RACE assays are also able to detect any miR319 regulation, as miR319 guides cleavage one nucleotide upstream of the miR159 cleavage position . miR319-guided cleavage only composed a very minor amount of cleavage events and these were only detected in the absence of miR159a and miR159b, suggesting that none of these MYB genes are under strong miR319 regulation. This is also consistent with previous findings .
Pollen development and germination appears independent of miR159 regulation
MYB101 or MYB120 expression in anthers is not delineated by miR159
The majority of MYB101:GUS (6/10) and mMYB101:GUS (7/8) lines showed anther-specific GUS expression in the inflorescence (Figure 7b, c), a pattern indistinguishable from proMYB101:GUS (MYB101 promoter only) transgenic plants  and consistent with online microarray data showing MYB101 is overwhelmingly expressed in anthers and pollen . At the gross organ level, there was no evidence of an expanded spatial or temporal mMYB101:GUS expression domain (Figure 7b, c). Expression of both the MYB101:GUS and mMYB101:GUS transgenes appeared restricted to anthers of intermediate age (approximately floral stages 6-12) but absent in both younger and older anthers. This spatial and temporal expression pattern appears highly similar to that of MIR159c:GUS (Figure 4b).
For MYB120:GUS and mMYB120:GUS, multiple transgenic lines were isolated, and GUS activity was observed exclusively in anthers (Figure 7e, f). No transgene expression was detected in vegetative tissues. MYB120:GUS and mMYB120:GUS showed indistinguishable anther specific expression patterns in inflorescences, both spatially and temporally, demonstrating that loss of the miR159 target site did not influence the expression pattern of MYB120 (Figure 7e, f). By comparison to MYB101, expression of MYB120 appeared to occur at earlier anther stages, but again included those stages in which MIR159c:GUS was expressed (Figure 4b).
MYB101 and MIR159chave highly similar spatial and temporal transcriptional domains
Despite MIR159c (Figure 4b) and MYB101 (Figure 7) being both transcribed in an anther-specific manner in the inflorescence, miR159c is not the major cleavage regulator of MYB101, as demonstrated by the relative paucity of miR159 cleavage products recovered in mir159ab. However it has been found that in Arabidopsis roots, the MIR395 gene is transcribed in phloem companion cells, adjacent to xylem expressed targets, perhaps preventing 'leaky' target expression . Similarly, to precisely determine if MIR159c and MYB101 are transcribed in adjacent cell types, and to investigate if MYB101 is subject to subtle miR159 regulation not detectable in whole anthers, transverse sections of GUS-stained proMYB101:GUS, MYB101:GUS and mMYB101:GUS anthers were examined using dark field microscopy.
In a highly similar pattern to the MIR159c:GUS transgene, MYB101/mMYB101:GUS expression only occurs in postmeiotic anthers (Figure 8a), and although initial MYB101:GUS expression is found in many cell types, as development proceeds, expression becomes concentrated in the tapetum (Figure 8d). Also similar to MIR159c:GUS, degeneration of the tapteum coincides with the loss of GUS staining, explaining the transient MYB101:GUS expression in anthers. This demonstrated that MIR159c and MYB101 have overlapping transcriptional domains, both spatially and temporally, being both predominantly transcribed in postmeiotic tapetal cells, indicating that MYB101 expression should be subjected to miR159c regulation. Yet the fact that both MYB101 (Figure 8) and MYB33  protein can accumulate in postmeiotic tapetal cells, implies that at most, miR159c could only be acting as a tuning miRNA.
The highly similar family members MYB33 and MYB101 are both expressed in anthers and both strongly in the tapetum. However MYB33 is expressed in young anthers, at premeiotic developmental stages, and in myb33.myb65 plants the block in anther development is premeiotic , before MYB101 is expressed. Consistent with this the MYB101 mRNA level is downregulated in myb33.myb65 flowers (Figure 8b), suggesting a developmental hierarchy of MYB expression in Arabidopsis anthers and a possible contributing factor to the lower MYB101 transcript levels observed in 35S:MIR159a plants , and higher MYB101 transcript levels in mir159ab inflorescences (Figure 3d).
Overexpression of mMYB101does not result in anther/pollen defects
Furthermore, ectopic mMYB101/mMYB101:GUS expression could also lead to rosette phenotypes similar again to mir159ab (data not shown), indicating that this transgene has biological activity, yet is unable to produce anther defects even under strong ectopic expression. Thus rendering MYB101 resistant to miRNA regulation and/or overexpression of this gene resulted in no obvious phenotypic consequences for anther or pollen development.
For many potential miR159 target genes, functional analysis has failed to identify any major morphological or molecular impact associated with miR159 regulation, despite more commonly taken approaches insinuating otherwise. The overlapping transcriptional domains of miR159c and many potential targets in anthers has identified a miR159c:target gene module, which we speculate may have once been important, but now appears to be quiescent in nature. We reason that the relatively recent demise of such a module may account for the conservation of miRNA target sites within potential target genes that appear principally independent of miR159 regulation. If correct, this would illustrate the evolutionarily fluid nature of miRNA:target relationships that exist even within ancient miRNA:target modules; while miR159a and miR159b are ingrained and critical for development, miR159c appears to be veering towards obsolescence. With respect to the targets, miR159 regulation of MYB33 is critical throughout the majority of the plant, while MYB101 expression appears principally independent of miR159 regulation. Therefore closely related members of the same gene family with conserved miRNA binding sites are regulated by very different mechanisms; MYB33 and MYB65 expression is largely determined post-transcriptionally, while MYB101 expression is predominantly determined at the transcriptional level. This case highlights the need for functional analysis of predicted miRNA targets, and that evidence from other methodologies may overestimate the extent of miRNA regulation that confers functional importance. These results reflect similar findings in metazoan systems; where among many potential targets, often only a single or few mRNAs may be relevant in vivo targets for a particular miRNA (reviewed in ). Thus a greater functional specificity of miRNAs than predicted by bioinformatics or molecular methodologies is an emergent and unifying theme in both plant and animal studies.
MIR159chas the hallmarks of obsolesce
Firstly, MIR159c appears to be very weakly expressed; a fact that also appears to be the case in A lyrata and Capsella rubella . The presence of a stretch of three transposable elements 214 bp upstream of the MIR159c stem-loop that are absent in the A lyrata MIR159c locus, indicates major sequence changes are tolerated at the MIR159c locus, possibly reflecting reduced selection pressure. In addition to low transcription of MIR159c, additional processing inefficiencies further reduce miR159c accumulation (; Figure 3). Furthermore mir159c displayed no obvious phenotypic defects or further redundancy with mir159ab (Figure 1). Finally, no potential targets of miR159c showed any deregulation in mir159c (Figure 5). Together, these analyses have failed to find any obvious functional role for MIR159c.
The only indication MIR159c may have any activity was the complementation of mir159ab by a 35S:MIR159c transgene (Figure 3). This demonstrates that when transcribed at sufficiently high levels, it is still able to effectively silence MYB33 and MYB65; hence MIR159c could not yet be considered a pseudogene. This raises the possibility that the gene, if induced, may play a role in controlling MYB activity. However this would seem unlikely in rosette or inflorescence tissues. In the rosettes miR159a and miR159b are so high a 35S:MIR159c transgene has no effect. In the inflorescence, although the 35S:MIR159c transgene could reduce MYB transcript levels in the mir159ab background, it could not repress MYB expression to levels that could cause male sterility. Therefore it is likely that not only would MIR159c transcription need to increase, but also its poor processing efficiency  be overcome for it to be highly expressed to exert a biological outcome.
Evidence suggests that MIR159c may have previously carried out a specialised function that has now become obsolete. In contrast to MIR159a and MIR159b that are broadly expressed, MIR159c appears to be transcribed specifically in the inflorescence in the same discrete anther cell layer (the tapetum) as MYB101 (Figure 8) and MYB33 . It is tempting to speculate that MIR159c had previously subfunctionalised and corresponded to a discrete regulatory module, controlling GAMYB expression in anthers compared to the broader MIR159a and MIR159b module controlling MYB33 and MYB65 in vegetative tissues. Precedence for this is that GAMYB expression in rice anthers appears under miR159 control , a role miR159c could have performed in Arabidopsis if its activity had been stronger; as most of the Arabidopsis GAMYB-like genes are strongly transcribed in anthers . However in Arabidopsis this requirement no longer appears important for two reasons. Firstly, from the perspective of MIR159c, loss of this gene has no noticeable molecular or phenotypic consequence (Figures 1, 2, 5 and 6). Secondly, from the perspective of potential miR159c targets, not only does MYB101 and MYB120 expression appear independent of miR159 regulation, overexpression of a mMYB101 transgene results in no obvious detrimental anther phenotype (Figure 9). By contrast, GAMYB overexpression in transgenic barley results in male sterility , demonstrating that modulation of GAMYB levels is required. Therefore the level of anther GAMYB and its regulation by miR159 appears important in cereals, and may reflect that miR159c regulation of MYB101 may represent a former important regulatory module for anther development that is now obsolete in Arabidopsis.
However these conclusions that are in part based on a lack of demonstrated function cannot be definitive, as we are unable to rule out that in a certain cell type or environmental condition, miR159c or the miR159 binding site within MYB101 and the other targets may have a functional role. Nevertheless, even if this is the case for some of these targets, our functional analysis of the entire miR159 family generally supports the notion of a much narrower functional specificity of plant miRNAs, rather than the possibility that they exert their effects through a broad range of targets.
Diverse modes of regulation of closely related genes resulting in similar expression outcomes
Despite MYB101 having a highly conserved miR159 binding site, detailed analysis in inflorescences of MYB101:GUS and mMYB101:GUS transgenic plants failed to find any miR159 regulation that delineates the expression of this gene (Figure 8). Furthermore, MYB101 and mMYB101 transgenes, when expressed in their native domain in anthers, had no apparent impact on development (Figure 9). Therefore in stark contrast to MYB33 , miR159 appears peripheral to regulation of MYB101.
Curiously MYB33 and MYB101 represent two closely related genes that are predominantly expressed in seeds and anthers [32, 44], but via two very different mechanisms. MYB33 is almost constitutively transcribed throughout the plant, only to be silenced by miR159 except in seeds and anthers , hence post-transcriptional regulation delineates the final expression pattern of the MYB33 protein. Conversely, MYB101 is specifically transcribed in seeds and anthers, tissues in which miR159 activity appears weak or absent  implying transcriptional regulation of this gene predominantly determines where the MYB101 protein is expressed. Thereby these two very closely related gene family members have very different regulatory relationships with miR159, as defined by their transcriptional domains.
Limitations of experimental validation of miRNA targets
This study has illustrated the limitations of several current methodologies used in miRNA target gene identification. Firstly, 5'-RACE recovery of miRNA-guided cleaved mRNA targets has been considered the gold standard for determining whether a particular mRNA is an authentic miRNA target . However, although miR159-guided cleavage products for MYB101 in inflorescences were recovered, it appears that there is no obvious functional role of this miRNA-target relationship. It should be noted that while MYB33 cleavage products were observable after the first round of PCR, this was not the case for any other MYB target genes. This may suggest that the abundance of cleavage is orders of magnitude less for these other genes when compared to MYB33. This agrees with degradome data that has found MYB33 and MYB65 were miR159 regulated, but not MYB81, MYB101, MYB120 or DUO1 [8–10]. This could reflect the extremely sensitive nature of the 5'-RACE assays, where nested PCR and gel purification of the expected size band would detect basal background miRNA activity of no functional consequence. Therefore our findings here and previously  concur with the quantitative degradome approaches that will predominantly detect strong regulatory relationships .
Secondly, analysis of phenotypical changes resulting from disruption of miRNA binding sites in putative mRNA targets has been used to identify the functional importance of miRNA-mediated regulation. However the results here show that transgenic expression can potentially misrepresent the extent of miRNA regulation for a particular target, even when transcribed under its native promoter (Figure 10). This supports the notion that inferring miRNA regulation based on transgene expression can be misleading .
Finally overexpression of miRNAs may misrepresent what genes are miRNA regulated in vivo. By example, it has been reported that mRNA levels of MYB33 and MYB65 were not downregulated by a 35S:miR159a transgene, but rather MYB101 was the most strongly downregulated gene . However the fact that 35S:miR159a anthers were sterile, similar to the myb33.myb65 mutant, suggests that the downregulation of MYB101 could be in part a secondary effect of MYB33 and MYB65 silencing. This was substantiated by the observation that MYB101 levels are considerably lower in myb33.myb65 flowers (Figure 8b).
As many evolutionary changes are thought to occur through alterations to gene regulation, this will also include miRNA:target regulatory modules becoming obsolete. The hypothesis of frequent birth and death of evolutionary non-conserved miRNAs  can now be extended to include conserved miRNAs such as miR159. The remnants of these systems, which include both the MIRNA genes and the binding motifs in target mRNAs, may be leading to the overestimation of the complexity of miRNA regulation that confers a functional impact.
Isolation and genotyping of mir159c
The SAIL_248_G11 T-DNA mutant was found on the SIGnAL 'T-DNA Express' Arabidopsis Gene Mapping Tool http://signal.salk.edu/cgi-bin/tdnaexpress and ordered from the Arabidopsis Biological Resource Center. Amplification using gene specific primers (Additional file 5, Table S1) detected the wild-type allele, and amplification using the T-DNA specific primer LB3 and MIR159c gene specific primers (Additional file 5, Table S1) isolated the T-DNA border junctions.
Total RNA was extracted from inflorescences using TRIzol reagent (Invitrogen; http://www.invitrogen.com/) with the following modifications to the manufacturers protocol. (1) The chloroform extraction was repeated. (2) Precipitation of RNA was carried out overnight at -20°C. (3) Samples were heated only to 37°C after dissolving in nuclease free water. RQ1 DNAse (Promega; http://promega.com/) was used to treat RNA except for qSL-PCR (see below), where no DNase treatment was carried out. RNA was cleaned using Plant RNAeasy columns (Qiagen; http://www.qiagen.com/). cDNA synthesis was carried out using Superscript III reverse transcriptase (Invitrogen) according to the manufacturers protocol with an oligo dT primer (Invitrogen). For each RNA sample, three separate cDNA synthesis reactions were carried out. Real-time quantitative (qRT)-PCR was carried out as described in Allen et al. . Primers used for this analysis are described in Additional file 5, Table S2.
Histochemical analysis of GUS activity
In situ GUS activity staining was performed using the method of Jefferson et al. . Tissues were transferred to this reagent in 1.5 ml Eppendorf tubes, vacuum infiltrated for 2 min, and left overnight or as described elsewhere at 37°C. Afterwards, stained tissues were rinsed three times using 70% ethanol. For preparation of GUS-stained anther sections for light microscopy, inflorescences were stained in GUS reagent for 48 h at 37°C and dehydrated in a graded ethanol series (70%, 95%, 100%). Inflorescences were then infiltrated and embedded with LR white resin (London Resin Company, London, U.K). Transverse sections (2 μm) were made with a Leica Ultracut 6 ultramicrotome (Leica UK; http://www.leica.com/). Sections were stained with 1% toludine blue for 1 min.
Digital photographs of rosettes, siliques and whole plants were taken at the CSIRO Phytotron studio, Canberra, Australia. Scanning electron microscopy of stamens, anthers, pollen and seed was performed by gold-coating tissues using a high resolution splutter coater (Bio-Rad; http://www.bio-rad.com/), and examination with a Cambridge S360 scanning electron microscope (SEM) (Cambridge, UK). Images of GUS-stained anther sections were taken with a Leica DMR upright microscope for dark-field microscopy. Images of GUS-stained seedlings, inflorescences, and individual flowers were taken with a Leica MZFLIII dissecting microscope.
Quantitative stem-loop qRT-PCR miRNA analysis
For TaqMan stem-loop qRT-PCR miRNA analysis (qSL-PCR), RNA was prepared using TRIzol as described above for expression analysis. For the assays, Applied Biosystems ABI TaqMan MicroRNA quantitative real-time stem-loop PCR assays were used http://www.appliedbiosystems.com/, and the manufacturers instructions were followed with the following modifications. (1) For each RNA sample, there were three stem-loop cDNAs made, and the reverse transcriptase (RT) step was multiplexed using both sno101 RT primer and miR159a, miR159b or miR159c primer. (2) The cDNA (15 μl) was diluted with 86.4 μl of nuclease free water, so that 9 μl of RT reaction could be pipetted into 20 μl total PCR reaction volume. (3) Each cDNA was assayed in triplicate on a Corbett real-time PCR machine (Corbett, http://www.corbettlifescience.com/). Primer sequences for qSL-PCR are proprietary to Applied Biosystems. Expression of miR159 was normalised to sno101, using the comparative concentration analysis program of rotor gene software (Corbett).
Generation of binary vectors and transgenic plants
To generate the 2 × 35S:MIR159c construct 1226 bp of MIR159c sequence comprising 525 bp upstream and 485 bp downstream of the MIR159c stem loop was amplified from Arabidopsis and cloned into the vector pMDC32 (Invitrogen). To generate the MIR159c:GUS construct 1.3 kb of genomic sequence immediately upstream of the MIR159c stem loop was cloned into the vector pBI 101.1. For generation of the MYB101 genomic construct 6333 bp of MYB101 genomic sequence comprising upstream and downstream genomic regions were amplified from Arabidopsis genomic DNA. miRNA resistant target sites were introduced by PCR to generate the construct mMYB101. The genomic fragments were cloned into pMDC99 [49 For MYB101 and MYB120 GUS constructs, genomic sequences of 5171 and 3128 bp, respectively were amplified, for MYB101 GUS, this sequence was identical to the genomic fragment used for the genomic construct above except there was no genomic sequence beyond the stop codon. miRNA resistant versions were produced by PCR, and both MYB101/120 and mMYB101/120 fragments were cloned into the vector pMDC164 . All vectors were transformed into Agrobacterium and Arabidopsis as previously described .
Modified 5'-RACE of cleaved miRNA targets
mRNA was purified from the same inflorescence RNA samples used for qRT-PCR analysis of miR159 targets in wild type and mir159ab , using 100 μg of total RNA. A Gene-Racer kit (Invitrogen) was used for 5'-RACE, except the decapping protocol was not carried out, and the adapter was ligated directly to mRNA. PCR of MYB33 and MYB101 sequence downstream of the miR159 cleavage site was used as a control to check 5' cDNA amplification was successful. The products from the second (nested) round of 5'-RACE were gel purified using a Wizard preps PCR purification kit (Promega) and ligated into pGEM-T easy (Promega). Plasmids were transformed into Escherichia coli XL-10 gold and purified. Clones were digested with NotI to verify they contained inserts of the correct size, and were sequenced. Primers used for this analysis are shown in Additional file 4, Table S3.
We thank the Salk Institute Genomic Analysis Laboratory for providing the sequence-indexed Arabidopsis T-DNA insertion mutant SAIL_248_G11. Funding for the SIGnAL indexed insertion mutant collection was provided by the National Science Foundation. We thank Carl Davies for photography and Dr Cheng Huang for assistance with Scanning Electron Microscopy. We thank Dr Mark Talbot and Celia Miller for assistance with microscopy. We thank Dr Ming-Bo Wang and three anonymous reviewers for many constructive and insightful comments. RSA was funded by an ANU postgraduate scholarship and by a CSIRO Emerging Science Initiative. JL is the recipient of an ANU international student postgraduate scholarship. This research was supported by an Australian Research Council grant DP0773270.
- Carthew RW, Sontheimer EJ: Origins and mechanisms of miRNAs and siRNAs. Cell. 2009, 136: 642-655. 10.1016/j.cell.2009.01.035.PubMed CentralView ArticlePubMedGoogle Scholar
- Jones-Rhoades MW, Bartel DP, Bartel B: MicroRNAS and their regulatory roles in plants. Annu Rev Plant Biol. 2006, 57: 19-53. 10.1146/annurev.arplant.57.032905.105218.View ArticlePubMedGoogle Scholar
- Rhoades MW, Reinhart BJ, Lim LP, Burge CB, Bartel B, Bartel DP: Prediction of plant microRNA targets. Cell. 2002, 110: 513-520. 10.1016/S0092-8674(02)00863-2.View ArticlePubMedGoogle Scholar
- Llave C, Kasschau KD, Rector MA, Carrington JC: Endogenous and silencing-associated small RNAs in plants. Plant Cell. 2002, 14: 1605-1619. 10.1105/tpc.003210.PubMed CentralView ArticlePubMedGoogle Scholar
- Zhu QH, Spriggs A, Matthew L, Fan L, Kennedy G, Gubler F, Helliwell C: A diverse set of microRNAs and microRNA-like small RNAs in developing rice grains. Genome Res. 2008, 18: 1456-1465. 10.1101/gr.075572.107.PubMed CentralView ArticlePubMedGoogle Scholar
- Chen HM, Li YH, Wu SH: Bioinformatic prediction and experimental validation of a microRNA-directed tandem trans-acting siRNA cascade in Arabidopsis. Proc Natl Acad Sci USA. 2007, 104: 3318-3323. 10.1073/pnas.0611119104.PubMed CentralView ArticlePubMedGoogle Scholar
- Lu S, Sun YH, Shi R, Clark C, Li L, Chiang VL: Novel and mechanical stress-responsive microRNAs in Populus trichocarpa that are absent from Arabidopsis. Plant Cell. 2005, 17: 2186-2203. 10.1105/tpc.105.033456.PubMed CentralView ArticlePubMedGoogle Scholar
- Gregory BD, O'Malley RC, Lister R, Urich MA, Tonti-Filippini J, Chen H, Millar AH, Ecker JR: A link between RNA metabolism and silencing affecting Arabidopsis development. Dev Cell. 2008, 14: 854-866. 10.1016/j.devcel.2008.04.005.View ArticlePubMedGoogle Scholar
- Addo-Quaye C, Eshoo TW, Bartel DP, Axtell MJ: Endogenous siRNA and miRNA targets identified by sequencing of the Arabidopsis degradome. Curr Biol. 2008, 18: 758-762. 10.1016/j.cub.2008.04.042.PubMed CentralView ArticlePubMedGoogle Scholar
- German MA, Pillay M, Jeong DH, Hetawal A, Luo S, Janardhanan P, Kannan V, Rymarquis LA, Nobuta K, German R, De Paoli E, Lu C, Schroth G, Meyers BC, Green PJ: Global identification of microRNA-target RNA pairs by parallel analysis of RNA ends. Nat Biotechnol. 2008, 26: 941-946. 10.1038/nbt1417.View ArticlePubMedGoogle Scholar
- Zhang B, Pan X, Stellwag EJ: Identification of soybean microRNAs and their targets. Planta. 2008, 229: 161-182. 10.1007/s00425-008-0818-x.View ArticlePubMedGoogle Scholar
- Zhang L, Chia JM, Kumari S, Stein JC, Liu Z, Narechania A, Maher CA, Guill K, McMullen MD, Ware D: A genome-wide characterization of microRNA genes in maize. PLoS Genet. 2009, 5: e1000716-10.1371/journal.pgen.1000716.PubMed CentralView ArticlePubMedGoogle Scholar
- Axtell MJ: Evolution of microRNAs and their targets: are all microRNAs biologically relevant?. Biochim Biophys Acta. 2008, 1779: 725-734.View ArticlePubMedGoogle Scholar
- Allen RS, Li J, Stahle MI, Dubroue A, Gubler F, Millar AA: Genetic analysis reveals functional redundancy and the major target genes of the Arabidopsis miR159 family. Proc Natl Acad Sci USA. 2007, 104: 16371-16376. 10.1073/pnas.0707653104.PubMed CentralView ArticlePubMedGoogle Scholar
- Sieber P, Wellmer F, Gheyselinck J, Riechmann JL, Meyerowitz EM: Redundancy and specialization among plant microRNAs: role of the MIR164 family in developmental robustness. Development. 2007, 134: 1051-1060. 10.1242/dev.02817.View ArticlePubMedGoogle Scholar
- Ambros V: A hierarchy of regulatory genes controls a larva-to-adult developmental switch in C. elegans. Cell. 1989, 57: 49-57. 10.1016/0092-8674(89)90171-2.View ArticlePubMedGoogle Scholar
- Cayirlioglu P, Kadow IG, Zhan X, Okamura K, Suh GS, Gunning D, Lai EC, Zipursky SL: Hybrid neurons in a microRNA mutant are putative evolutionary intermediates in insect CO2 sensory systems. Science. 2008, 319: 1256-1260. 10.1126/science.1149483.PubMed CentralView ArticlePubMedGoogle Scholar
- Xiao C, Calado DP, Galler G, Thai TH, Patterson HC, Wang J, Rajewsky N, Bender TP, Rajewsky K: MiR-150 controls B cell differentiation by targeting the transcription factor c-Myb. Cell. 2007, 131: 146-159. 10.1016/j.cell.2007.07.021.View ArticlePubMedGoogle Scholar
- Jones-Rhoades MW, Bartel DP: Computational identification of plant microRNAs and their targets, including a stress-induced miRNA. Mol Cell. 2004, 14: 787-799. 10.1016/j.molcel.2004.05.027.View ArticlePubMedGoogle Scholar
- Palatnik JF, Allen E, Wu X, Schommer C, Schwab R, Carrington JC, Weigel D: Control of leaf morphogenesis by microRNAs. Nature. 2003, 425: 257-263. 10.1038/nature01958.View ArticlePubMedGoogle Scholar
- Axtell MJ, Bowman JL: Evolution of plant microRNAs and their targets. Trends Plant Sci. 2008, 13: 343-349. 10.1016/j.tplants.2008.03.009.View ArticlePubMedGoogle Scholar
- Kasschau KD, Fahlgren N, Chapman EJ, Sullivan CM, Cumbie JS, Givan SA, Carrington JC: Genome-wide profiling and analysis of Arabidopsis siRNAs. PLoS Biol. 2007, 5: e57-10.1371/journal.pbio.0050057.PubMed CentralView ArticlePubMedGoogle Scholar
- Rajagopalan R, Vaucheret H, Trejo J, Bartel DP: A diverse and evolutionarily fluid set of microRNAs in Arabidopsis thaliana. Genes Dev. 2006, 20: 3407-3425. 10.1101/gad.1476406.PubMed CentralView ArticlePubMedGoogle Scholar
- Nakano M, Nobuta K, Vemaraju K, Tej SS, Skogen JW, Meyers BC: Plant MPSS databases: signature-based transcriptional resources for analyses of mRNA and small RNA. Nucleic Acids Res. 2006, 34: D731-735. 10.1093/nar/gkj077.PubMed CentralView ArticlePubMedGoogle Scholar
- Palatnik JF, Wollmann H, Schommer C, Schwab R, Boisbouvier J, Rodriguez R, Warthmann N, Allen E, Dezulian T, Huson D, Carrington JC, Weigel D: Sequence and expression differences underlie functional specialization of Arabidopsis microRNAs miR159 and miR319. Dev Cell. 2007, 13: 115-125. 10.1016/j.devcel.2007.04.012.View ArticlePubMedGoogle Scholar
- Bartel DP: MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004, 116: 281-297. 10.1016/S0092-8674(04)00045-5.View ArticlePubMedGoogle Scholar
- Reyes JL, Chua NH: ABA induction of miR159 controls transcript levels of two MYB factors during Arabidopsis seed germination. Plant J. 2007, 49: 592-606. 10.1111/j.1365-313X.2006.02980.x.View ArticlePubMedGoogle Scholar
- Alves-Junior L, Niemeier S, Hauenschild A, Rehmsmeier M, Merkle T: Comprehensive prediction of novel microRNA targets in Arabidopsis thaliana. Nucleic Acids Res. 2009, 37: 4010-4021. 10.1093/nar/gkp272.PubMed CentralView ArticleGoogle Scholar
- Schwab R, Palatnik JF, Riester M, Schommer C, Schmid M, Weigel D: Specific effects of microRNAs on the plant transcriptome. Dev Cell. 2005, 8: 517-527. 10.1016/j.devcel.2005.01.018.View ArticlePubMedGoogle Scholar
- Backman TW, Sullivan CM, Cumbie JS, Miller ZA, Chapman EJ, Fahlgren N, Givan SA, Carrington JC, Kasschau KD: Update of ASRP: the Arabidopsis Small RNA Project database. Nucleic Acids Res. 2008, 36: D982-985. 10.1093/nar/gkm997.PubMed CentralView ArticlePubMedGoogle Scholar
- Baker CC, Sieber P, Wellmer F, Meyerowitz EM: The early extra petals1 mutant uncovers a role for microRNA miR164c in regulating petal number in Arabidopsis. Curr Biol. 2005, 15: 303-315. 10.1016/j.cub.2005.02.017.View ArticlePubMedGoogle Scholar
- Millar AA, Gubler F: The Arabidopsis GAMYB-Like genes, MYB33 and MYB65, are microRNA-regulated genes that redundantly facilitate anther development. Plant Cell. 2005, 17: 705-721. 10.1105/tpc.104.027920.PubMed CentralView ArticlePubMedGoogle Scholar
- Flynt AS, Lai EC: Biological principles of microRNA-mediated regulation: shared themes amid diversity. Nat Rev Genet. 2008, 9: 831-842. 10.1038/nrg2455.PubMed CentralView ArticlePubMedGoogle Scholar
- Fahlgren N, Jogdeo S, Kasschau KD, Sullivan CM, Chapman EJ, Laubinger S, Smith LM, Dasenko M, Givan SA, Weigel D, Carrington JC: MicroRNA gene evolution in Arabidopsis lyrata and Arabidopsis thaliana. Plant Cell. 22: 1074-1089. 10.1105/tpc.110.073999.Google Scholar
- Ma Z, Coruh C, Axtell MJ: Arabidopsis lyrata small RNAs: transient MIRNA and small interfering RNA loci within the Arabidopsis genus. Plant Cell. 22: 1090-1103. 10.1105/tpc.110.073882.Google Scholar
- Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen H, Shinn P, Stevenson DK, Zimmerman J, Barajas P, Cheuk R, Gadrinab C, Heller C, Jeske A, Koesema E, Meyers CC, Parker H, Prednis L, Ansari Y, Choy N, Deen H, Geralt M, Hazari N, Hom E, Karnes M, Mulholland C, Ndubaku R, Schmidt I, Guzman P, Aguilar-Henonin L, Schmid M, et al: Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science. 2003, 301: 653-657. 10.1126/science.1086391.View ArticlePubMedGoogle Scholar
- Achard P, Herr A, Baulcombe DC, Harberd NP: Modulation of floral development by a gibberellin-regulated microRNA. Development. 2004, 131: 3357-3365. 10.1242/dev.01206.View ArticlePubMedGoogle Scholar
- Slotkin RK, Vaughn M, Borges F, Tanurdzic M, Becker JD, Feijo JA, Martienssen RA: Epigenetic reprogramming and small RNA silencing of transposable elements in pollen. Cell. 2009, 136: 461-472. 10.1016/j.cell.2008.12.038.PubMed CentralView ArticlePubMedGoogle Scholar
- Hruz T, Laule O, Szabo G, Wessendorp F, Bleuler S, Oertle L, Widmayer P, Gruissem W, Zimmermann P: Genevestigator V3: a reference expression database for the meta-analysis of transcriptomes. Adv Bioinformatics. 2008, 2008: 5-View ArticleGoogle Scholar
- Kawashima CG, Yoshimoto N, Maruyama-Nakashita A, Tsuchiya YN, Saito K, Takahashi H, Dalmay T: Sulphur starvation induces the expression of microRNA-395 and one of its target genes but in different cell types. Plant J. 2009, 57: 313-321. 10.1111/j.1365-313X.2008.03690.x.View ArticlePubMedGoogle Scholar
- Seitz H: Redefining microRNA targets. Curr Biol. 2009, 19: 870-873. 10.1016/j.cub.2009.03.059.View ArticlePubMedGoogle Scholar
- Tsuji H, Aya K, Ueguchi-Tanaka M, Shimada Y, Nakazono M, Watanabe R, Nishizawa NK, Gomi K, Shimada A, Kitano H, Ashikari M, Matsuoka M: GAMYB controls different sets of genes and is differentially regulated by microRNA in aleurone cells and anthers. Plant J. 2006, 47: 427-444. 10.1111/j.1365-313X.2006.02795.x.View ArticlePubMedGoogle Scholar
- Murray F, Kalla R, Jacobsen J, Gubler F: A role for HvGAMYB in anther development. Plant J. 2003, 33: 481-491. 10.1046/j.1365-313X.2003.01641.x.View ArticlePubMedGoogle Scholar
- Penfield S, Li Y, Gilday AD, Graham S, Graham IA: Arabidopsis ABA INSENSITIVE4 regulates lipid mobilization in the embryo and reveals repression of seed germination by the endosperm. Plant Cell. 2006, 18: 1887-1899. 10.1105/tpc.106.041277.PubMed CentralView ArticlePubMedGoogle Scholar
- Voinnet O: Origin, biogenesis, and activity of plant microRNAs. Cell. 2009, 136: 669-687. 10.1016/j.cell.2009.01.046.View ArticlePubMedGoogle Scholar
- Fahlgren N, Howell MD, Kasschau KD, Chapman EJ, Sullivan CM, Cumbie JS, Givan SA, Law TF, Grant SR, Dangl JL, Carrington JC: High-throughput sequencing of Arabidopsis microRNAs: evidence for frequent birth and death of MIRNA genes. PLoS ONE. 2007, 2: e219-10.1371/journal.pone.0000219.PubMed CentralView ArticlePubMedGoogle Scholar
- Jefferson RA, Kavanagh TA, Bevan MW: GUS fusions: beta-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. Embo J. 1987, 6: 3901-3907.PubMed CentralPubMedGoogle Scholar
- Curtis MD, Grossniklaus U: A Gateway cloning vector set for high-throughput functional analysis of genes in planta. Plant Physiol. 2003, 133: 462-469. 10.1104/pp.103.027979.PubMed CentralView ArticlePubMedGoogle Scholar
- Zhang Y: miRU: an automated plant miRNA target prediction server. Nucleic Acids Res. 2005, 33: W701-704. 10.1093/nar/gki383.PubMed CentralView ArticlePubMedGoogle Scholar
- The Zhao Bioinformatics Laboratory: psRNATarget: a plant small RNA regulator target analysis server. [http://bioinfo3.noble.org/psRNATarget/]
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.