|
|
|
|
|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | ||||
First published online 7 February 2007
doi: 10.1242/dev.02817
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Division of Biology, California Institute of Technology, Pasadena, CA 91125,
USA.
2 Institute of Plant Biology, University of Zurich, Zollikerstrasse 107, 8008
Zurich, Switzerland.
* Author for correspondence (e-mail: meyerow{at}its.caltech.edu)
Accepted 22 January 2007
 |
SUMMARY |
|---|
 TOP SUMMARY INTRODUCTIONMATERIALS AND METHODSRESULTS Analysis ofDISCUSSIONREFERENCES |
|---|
Key words: MicroRNA, miR164, Organogenesis, Developmental robustness, Phyllotaxis, CUP-SHAPED COTYLEDON
|
INTRODUCTION |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTS Analysis ofDISCUSSIONREFERENCES |
|---|
21 nucleotide) non-translated RNAs that
are generated by the enzymatic processing of stem-loop regions of longer
precursor RNAs (Bartel, 2004
;
Valencia-Sanchez et al., 2006;
Vaucheret, 2006). miRNAs are
present in both plants and animals, and they regulate gene expression in a
sequence-specific manner by targeting mRNAs for cleavage or translational
repression. Animal miRNAs typically act by inhibiting translation of their
targets, with which they usually share relatively low sequence complementarity
(Lim et al., 2005). As a
result, a given animal miRNA might target many different genes, and a large
fraction of the animal transcriptome has been proposed to be directly
influenced by miRNA control (Farh et al.,
2005; Stark et al.,
2005). By contrast, plant miRNAs use cleavage as the preferential
mechanism for target gene regulation, they tend to exhibit a high degree of
complementarity to their targets, and appear to have fewer target genes per
miRNA (Llave et al., 2002;
Schwab et al., 2005). The
existence of perfect or nearly perfect complementarity between plant miRNAs
and mRNAs has greatly facilitated the identification of putative targets for
many of the characterized miRNAs (Rhoades
et al., 2002). In Arabidopsis, 165 miRNA loci have been
identified so far, which are grouped into 93 different miRNA families
(Lu et al., 2006;
Rajagopalan et al., 2006;
Reinhart et al., 2002)
(miRBase, release 9.0,
http://microrna.sanger.ac.uk/).
At least 70 of these miRNA loci, comprising 21 gene families, have been
predicted or demonstrated to target genes encoding transcription factors. This
prevalence of regulatory genes points to a central role for miRNAs in the
control of gene regulatory networks in plants. Consistent with this notion, it
has been shown that miRNAs are required in many developmental processes in
plants, including organ polarity determination, meristem function, floral
patterning, vascular development, lateral root development and hormone
response (Baulcombe, 2004;
Chen, 2005;
Mallory and Vaucheret, 2006).
The participation of miRNAs in these processes has been established primarily
through their overexpression, or by generating plants that express
miRNA-resistant versions of their target gene(s). By contrast, mutants have
been isolated for only a few miRNAs
(Aukerman and Sakai, 2003;
Baker et al., 2005;
Guo et al., 2005;
Kim et al., 2005;
Palatnik et al., 2003;
Williams et al., 2005), and
thus far only a single loss-of-function mutant for a plant miRNA has been
identified in a forward genetic screen: the Arabidopsis early extra
petals1 mutant (eep1), in which miR164c function is
disrupted (Baker et al.,
2005).
The scarcity of identified loss-of-function mutants and phenotypes might be
attributed to the fact that many miRNAs belong to multigene families, which
are predicted to target the same (or overlapping) sets of genes, opening the
possibility of substantial functional redundancy among miRNAs in plants.
Although for Caenorhabditis elegans it has been shown that some
members of the let-7 family can have redundant functions
(Abbott et al., 2005), evidence
for redundancy among plant miRNAs has only been circumstantial. For example,
loss of a single miRNA of a multigene family did not result in an aberrant
phenotype in tissues where the miRNA was expressed
(Mallory et al., 2004).
The Arabidopsis MIR164 family comprises three members
(miR164a, miR164b and miR164c) and negatively regulates,
through mRNA cleavage, several genes that encode NAC-like transcription
factors (Baker et al., 2005;
Guo et al., 2005;
Kasschau et al., 2003;
Laufs et al., 2004;
Mallory et al., 2004;
Park et al., 2002). These
genes include CUP-SHAPED COTYLEDON1 (CUC1) and
CUC2, which are expressed in, and are necessary for, the formation of
boundaries between meristems and emerging organ primordia
(Aida et al., 1999;
Heisler et al., 2005;
Takada et al., 2001). Failure
to establish organ boundaries leads to severe developmental consequences, and
in loss-of-function cuc1 cuc2 double-mutant seedlings the two
cotyledons fail to separate and the seedling meristem arrests
(Aida et al., 1997). Expression
of miRNA cleavage-resistant versions of CUC1 and CUC2 in
Arabidopsis has revealed that miR164-mediated repression of
CUC1 and CUC2 is necessary for proper control of organ
number (Baker et al., 2005;
Mallory et al., 2004) and for
organ boundary formation (Laufs et al.,
2004). Analysis of eep1 mutants has shown that
miR164c functions to prevent extra petals in early-arising flowers by
repressing CUC1 and CUC2
(Baker et al., 2005). The role
of the MIR164 family is not limited to flower development, however,
as both miR164a and miR164b have been reported to prevent
lateral root initiation by repressing the miR164 target NAC1
(Guo et al., 2005). In
addition, ectopic expression of miR164-resistant versions of CUC1 and
CUC2, respectively, was shown to lead to abnormal vegetative development
(Laufs et al., 2004;
Mallory et al., 2004;
Nikovics et al., 2006). Taken
together, these results suggest that miR164 miRNAs may act throughout
plant development.
Here, we report on the elimination of the activity of the entire MIR164 family, and its consequences for development, demonstrating that all miR164 miRNAs function redundantly during Arabidopsis shoot development, and uncovering new functions for these genes, including the regulation of phyllotaxis (the arrangement of organs along the stem) and developmental robustness.
|
MATERIALS AND METHODS |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTS Analysis ofDISCUSSIONREFERENCES |
|---|
).
mir164a-4 (this study) (Nikovics
et al., 2006) and mir164b-1
(Baker et al., 2005;
Mallory et al., 2004) were
isolated in the accession Columbia (Col-0). eep1, here referred to as
mir164c-1, was isolated in Landsberg erecta (Ler).
Although mir164c-1 mutants transcribe reduced amounts of
pri-miR164c, they fail to effectively repress CUC1 and
CUC2 function and thus may represent functionally null mutants
(Baker et al., 2005). The
mir164c-2 allele (Col-0 background) represents a dSpm transposon
insertion (dSpm_3_25571) 85 bp 3' of the mature miR164c miRNA.
Disruption of the predicted pre-miR164c stem loop in this allele does
not cause any obvious mutant phenotype. In addition, mir164a-4 b-1
c-2 triply homozygous mutant plants fail to show any increase in severity
of their phenotype as compared with the respective single mutants (data not
shown). The mir164a-4 allele, isolated from the GABI-Kat collection
(Rosso et al., 2003), was
crossed to plants doubly homozygous for mir164b-1 and
mir164c-1 (Baker et al.,
2005). Progeny of three independent crosses
(ntot=246) were genotyped at all three MIR164
loci (see below). Twelve plants were homozygous for all three mir164
mutant alleles. mir164a-4 b-1 c-1 triply homozygous plants always
showed a novel phenotype, referred to as the mir164abc mutant
phenotype. Floral organ counts were performed on 15 mir164abc triple
mutants and 15 mir164aAbBcC `wild-type' control plants, in both cases
of a mixed Ler/Col background. The mir164abc triple mutant
was backcrossed once to Col-0 and to Ler wild-type plants to assess
potential contributions of either background to the phenotype.
Plasmid constructs
The pCUC1::CUC1-GFP and pCUC1::CUC1m-GFP
constructs have been described previously
(Baker et al., 2005).
pCUC2::CUC2-GFP was cloned by recombining DNA fragments essentially
as described for pCUC2::CUC2-VENUS-N7
(Heisler et al., 2005), except
that mGFP5 was used as an alternative green fluorescent protein (GFP) instead
of VENUS-N7. To generate pCUC2::CUC2m-GFP, the
CUC2 coding region was mutated by exchanging nucleotides 772-792
(5'-GAGCACGTGTCCTGTTTCTCC-3') for
5'-GAACATGTATCATGCTTTAGC-3'
(base changes are underlined), thereby introducing eight silent mutations,
which left the amino acid sequence of the CUC2 protein unchanged. Mutations
were introduced by applying a PCR-mediated in vitro mutagenesis strategy as
described for pCUC1::CUC1m-GFP
(Baker et al., 2005). The
transcriptional reporter pCUC1::3XVENUS-N7 was generated as follows:
a 1.4 kb EcoRI-SfiI fragment that contained the endogenous
CUC1 regulatory sequences (Baker
et al., 2005) was blunt-ended using T4 DNA polymerase; this
promoter fragment was subsequently introduced into the SmaI site of
plasmid pPD35 (Heisler et al.,
2005) (kindly provided by Dr P. Das) and the resulting plasmid was
tested for the correct orientation of the insert. pCUC2::3XVENUS-N7
has been previously described (Heisler et
al., 2005). 1882 bp of MIR164a 5' promoter sequence
was amplified from Col-0 genomic DNA using the primer combination PS272/PS273
(Table 1). The
NdeI-XhoI digested PCR fragment was cloned into the
corresponding restriction sites of pPD35 to generate
pMIR164a::3XVENUS-N7. The fragment corresponding to 2548 bp of
MIR164b 5' upstream sequence was cloned analogously with
primers PS240/PS241 and using XhoI and BamHI to generate
pMIR164b::3XVENUS-N7. pMIR164c::3XVENUS-N7 was made by cutting
pMIR164c::GUS (Baker et al.,
2005) with XhoI and BamHI and ligating into the
pPD35 vector. The NotI cassette from each pPD35 subclone was shuttled
into the pMLBART (Eshed et al.,
1999) binary vector as previously described
(Baker et al., 2005).
|
).
Microscopy
Protocols for light microscopy (LM) and SEM were as previously described
(Baker et al., 2005). Confocal
laser scanning microscopy (CLSM) imaging of live plants was performed using a
ZEISS LSM 510 Meta using either a 63x 0.95 W or a 40x 0.05 W
Achroplan water objective as described
(Heisler et al., 2005;
Reddy et al., 2004). FM4-64
dye (Molecular Probes) was used as a plasma membrane marker. Specimens of the
VENUS-N7/FM4-64 combination were excited with an argon laser that was
attenuated to 10% at 514 nm. Single tracking line-scan mode was used in
combination with a NFT 635 VIS main dichroic short-pass filter. Each scan
represents the mean of two scans. The emission was split by a 545 nm secondary
dichroic filter and sent through a 530-600 nm band pass for detection of
VENUS, and a 650 nm long-pass filter for FM4-64 signal, respectively. A single
tracking line-scan was used for GFP/FM4-64 covisualization. GFP/FM4-64
specimens were excited using the 488 nm laser line together with a NFT 635 VIS
main dichroic short-pass filter in combination with a 545 nm secondary
dichroic to split the emission. GFP and FM4-64 were detected using a 505-530
nm band pass and a 650 nm long-pass filter, respectively.
Genotyping
Presence of the mir164a-4 T-DNA insertion was confirmed by PCR
amplification across the junction between the left border of the T-DNA and the
genomic DNA by using the primer pair PS321/PS322
(Table 1). The PCR product was
sequenced with primer PS323 to confirm the presence of the insertion site.
Insertion of the dSpm transposon in the mir164c-2 allele was
confirmed by PCR with primers PS385/PS386/PS387
(Table 1). The resulting
650 bp PCR fragment was sequenced with primers PS385 and PS386.
Genotyping of the eep1 allele and the miR164b-1 allele have
been described previously (Baker et al.,
2005).
Gene expression analysis
RNA blot analysis was performed essentially as described
(Chen, 2004;
Reinhart et al., 2002;
Williams et al., 2005) with
the following modifications. RNA (15 µg) enriched for small RNAs was
isolated from Arabidopsis plants using the mirVana RNA
Isolation Kit (Ambion, cat. #1560) in combination with Plant RNA Isolation Aid
(Ambion, cat. #9690) according to the manufacturer's instructions. Blots were
hybridized using a [
-32P]-ATP end-labeled locked nucleic
acid ath-MIR164a oligonucleotide (Exiqon A/S, Denmark, cat. #30024).
A mixture of 0.1 µM PS388/PS389 oligonucleotides were used as 20- and 30-nt
size standards, respectively. As a loading control, blots were stripped and
re-probed with a [-32P]-ATP end-labeled DNA oligonucleotide
(5'-TTGCGTGTCATCCTTGCGCAGG-3') complementary to U6 RNA
(Mallory et al., 2004). ImageJ
1.34s software
(http://rsb.info.nih.gov/ij)
was used for quantification of miR164. Quantitative reverse
transcriptase-mediated polymerase chain reaction (qRT-PCR) analyses were
performed as described (Baker et al.,
2005). Transcript levels were normalized to the level of the
non-target gene TUB4 (At5g44340). The primer pairs used to
detect transcripts of TUB4, CUC1, CUC2, NAC1, At5g07560 and
At5g61430 were described previously
(Mallory et al., 2004). Primer
pair PS396/PS397 (Table 1) was
used to detect At5g39610.
RNA in situ hybridization analyses for CUC1 and CUC2 were
performed as described (Baker et al.,
2005). For the miRNA in situ hybridizations, we followed the
protocol of Valoczi and co-workers as originally described
(Valoczi et al., 2006). The
locked nucleic acid (LNA) miRNA oligos were end-labelled using the DIG
Oligonucleotide 3'-End Labelling Kit, 2nd Generation, from Roche (cat.
#03 353 575 910) according to the manufacturer's recommendations.
LNA-ath-miR164a antisense oligo
(5'-TGCACGTGCCCTGCTTCTCCA-3') was used to detect miR164
miRNAs. Scramble-miR (Exiqon A/S, Denmark, cat. #99001;
5'-TTCACAATGCGTTATCGGATGT-3') was used as a negative control
(rather than miR164 sense probes, which could hybridize to
miR164 target transcripts). Hybridization was performed at 50°C
overnight for all slides. A total of five washes were performed at 50°C
with 2x SSC-50% formamide. An RNaseA digest was included after the third
wash to remove non-specific background signal. Standard blocking and washing
steps in combination with anti-DIG antibody (Roche) were used for the
immunological detection (Long and Barton,
1998). Western Blue Reagent (Promega) in combination with
levamisole was used for the detection reaction. Slides were mounted in
Glycerol in TE.
|
RESULTS Analysis of |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTS Analysis ofDISCUSSIONREFERENCES |
|---|
). This
allele was named mir164a-4, as three other mir164a alleles
had been described previously (Guo et al.,
2005). mir164a-4 is likely to represent a null allele
because the T-DNA is inserted 29 bp 3' of the last nucleotide of the
processed miR164a sequence and thus disrupts the predicted stem-loop
structure of the miR164a precursor, which is essential for miRNA
biogenesis (see Fig. S1 in the supplementary material)
(Nikovics et al., 2006;
Parizotto et al., 2004).
Plants homozygous for mir164a-4 have been reported to show deepened
serration of the leaf margins (see Fig. S2 in the supplementary material)
(Nikovics et al., 2006).
Shoots of the previously described mir164b-1 mutant also displayed an
essentially normal phenotype (Mallory et
al., 2004), whereas mir164c-1 mutant plants formed extra
petals in early-arising flowers, but otherwise largely resembled wild-type
plants (Baker et al., 2005). We
used these single-mutant lines to generate plants homozygous for
mir164a-4, mir164b-1 and mir164c-1 (henceforth referred to
as mir164abc triple mutants). The results of RNA blot analyses
revealed that in mir164abc triple mutants, miR164-type 21 nt
miRNAs are severely reduced in abundance, if not completely abolished,
compared with wild-type plants (Fig.
1A). mir164abc triple mutants were indistinguishable from
wild-type plants during vegetative development, except for rosette leaves that
appeared slightly more serrated in the triple mutant (see Fig. S2 in the
supplementary material) (Nikovics et al.,
2006). Severe phenotypic alterations were observed, however, after
the switch from vegetative to reproductive development. Arabidopsis
wild-type flowers are composed of an almost invariant number of floral organs,
with four sepals, four petals, six stamens, and two fused carpels arranged in
four concentric circles or whorls (Fig.
1B). By contrast, most flowers of the mir164abc triple
mutant had an increased number of sepals and petals, but slightly fewer
stamens than Ler wild-type and mir164aAbBcC
(Ler/Col) control plants (Fig.
1C-G, and data not shown). Organ numbers in mir164abc
triple mutants were highly variable, which is reflected in relatively large
standard deviations of the organ counts for flowers at different positions
along the stem (Fig. 1F,G). In
addition to the organ-number defects, individual organs varied in size and
carpels typically failed to fuse in the triple mutants, resulting in a severe
reduction in fertility (Fig.
1C-E). Thus, loss of miR164a and miR164b function substantially enhances the floral defects of mir164c plants, as floral organs are affected in all four whorls and in all flowers independent of their time of initiation (Fig. 1E-G). These miRNAs might therefore control flower development in a redundant manner.
Control of phyllotaxis by miR164 miRNAs
In addition to the flower defects described above, the phyllotaxis of
mir164abc triple-mutant plants was severely disrupted
(Fig. 2, and see Fig. S3 in the
supplementary material). In wild-type plants, flower primordia are
successively initiated on the flanks of the inflorescence meristem, so that an
incipient primordium is initiated in a position that is furthest away from the
two preceding primordia (Reinhardt,
2005). This leads to a spiral arrangement, in which developing
flowers are positioned both radially and vertically at regular intervals along
the stem (Fig. 2E). By
contrast, mir164abc triple-mutant plants displayed a highly unequal
and distorted arrangement of flowers, both with respect to the angle formed by
two consecutive flowers and their distance from each other along the axis of
the stem (Fig. 2C,D). The
average internode distance (the distance between two flowers) of the mixed
Ler/Col-0 wild-type control was 8.7±3.6 mm (s.d.,
ntot=149), and 8.6±8.6 mm (s.d.,
ntot=150) for mir164abc triple-mutant plants
(Fig. 2K). Thus, the mean
internode length was almost unaffected in the mir164abc mutant.
However, whereas in the wild-type control 87% (129/149) of all flowers
remained within the standard deviation of the mean value for the internode
distance of the wild type (categories 5 mm to 12 mm), this was true for only
32% (48/150) of mir164abc triplemutant flowers. Notably, in
mir164abc mutants, 35% of all flowers were separated by a distance of
1 mm or less, compared with fewer than 2% (2/149) in the control
(Fig. 2J,K).
The initial positioning of flower primordia, however, was normal in mir164abc triple mutants when compared with the wild type (Fig. 2A,B), indicating that the disruption of the phyllotaxis in miR164abc mutants occurs after flowers have been initiated. Thus, the positioning of flowers in Arabidopsis appears to be dependent not only on their initiation pattern at the shoot apex, but also on mechanisms that actively retain their initial arrangement during flower maturation and growth.
In order to identify the cause of the phyllotaxis defects in
mir164abc mutants, stem segments were examined by scanning electron
microscopy at a position basal to the meristem, where flowers are at an
advanced stage of development. Whereas mature flowers were separated by
uniformly elongated and rectangular epidermal cells in the wild type
(Fig. 2F,H), cells between the
clustered flowers of mir164abc triple-mutant plants appeared smaller
and more variable in shape (Fig.
2G,I). In addition, successive mir164abc flowers were
often radially separated by only five or six cells
(Fig. 2I), which is similar to
the number of cells found between two neighboring floral primordia at the time
of initiation (Heisler et al.,
2005; Reddy et al.,
2004). This suggests that in mir164abc plants, cell
division activities are repressed in the internodes that separate individual
flowers.
|
;
Guo et al., 2005;
Nikovics et al., 2006).
Confocal laser scanning microscopy (CLSM) was used to image reporter gene
expression in vivo (Fig.
3A-D).
During reproductive development, when phenotypic alterations become
apparent in miR164abc plants, all three reporter genes were expressed
in inflorescence tissue and predominantly in epidermal cells
(Fig. 3, and data not shown).
For MIR164c, GFP fluorescence was detected in the inflorescence
meristem, in lateral boundary cells between flower primordia and the
inflorescence meristem, as well as in floral meristems, sepal margins and
carpels (Fig. 3C,D). Expression
of the GFP-based reporter is in agreement with that of the previously
described ß-glucuronidase-based transcriptional reporter for
MIR164c (Baker et al.,
2005), which showed expression in meristems and young floral buds.
MIR164a reporter expression was detected in leaves
(Nikovics et al., 2006), and
it was observed in the boundaries between the inflorescence meristem and
floral primordia, in young floral buds [stages 2-4; stages according to Smyth
et al. (Smyth et al., 1990)],
as well as in the adaxial domains of older flowers
(Fig. 3A). Thus, the expression
patterns of MIR164a and MIR164c are partially overlapping
and are consistent with the regions of the plant affected in
mir164abc triple-mutants. By contrast, GFP expression in the
MIR164b reporter line appeared to be excluded from meristems and was
strongest in abaxial epidermal cells of sepals
(Fig. 3B). The RNA blot
analysis (Fig. 1A) suggests
that the miR164b locus contributes substantially to the overall
population of miR164 RNA molecules in the inflorescence. However, it
is unknown whether the three miR164 miRNAs are equally well
processed, or how the efficiency of processing varies among cells, which may
explain the apparent discrepancy between the results obtained in the RNA blot
analysis and those obtained through the use of the transcriptional reporter.
Alternatively, additional regulatory sequences that could affect the degree of
miR164b accumulation may not have been included in the reporter
construct.
|
Mature miR164 miRNAs are essentially identical in sequence and are
predicted to target the same set of transcripts. However, the availability of
the triple mir164abc mutant, as well as of the various double
mutants, allowed the use of in situ hybridization to infer characteristics of
the expression patterns of the individual MIR164 family members.
DIG-labeled LNA oligo probes were used to detect miR164 miRNA
accumulation patterns on tissue sections of inflorescences. In Ler
wild-type plants, combined signal of all three miR164 miRNAs was
detected in vegetative leaves, in inflorescence meristems, in young flower
primordia as well as in floral organ primordia. Strong signal was also
detected in the locules of the anthers
(Fig. 3E,F). The
miR164 expression pattern in A. thaliana thus resembled the
pattern of miR164 expression in N. benthamiana
(Valoczi et al., 2006). There
was no signal above background in the Scramble-miR control
(Fig. 3G,H), and thus the
detected signal was miR164-specific. Weak, but specific signal was
also detected in the mir164abc triplemutant background
(Fig. 3O,P). This signal might
represent processed miR164 originating from leaky expression of one
or more of the three mir164 mutant loci. Alternatively, the probe
might hybridize to another RNA fragment, for instance to the 24 nt band
that was detected on the RNA blot (Fig.
1A). miR164c accumulation, as detected in mir164a-4
b-1 double-mutant plants (Fig.
3I,J), was found in the expected tissue but did not accumulate to
levels significantly above the level of miR164 signal observed in
mir164abc triple mutants. miR164b on the other hand, as
detected in mir164a-4 c-1 double mutants, reached a level of
expression that was comparable to miR164 accumulation in the wild
type (Fig. 3K,L). Accumulation
of miR164a, when examined in mir164b-1 c-1 double-mutant
plants was comparable to the result obtained for miR164c, with the
difference that the expression in leaves remained strong in mir164b-1
c-1 double mutants (Fig.
3M,N).
Altogether, the in situ hybridization data are consistent with the results obtained from RNA blot experiments, which indicate that miR164b miRNAs contribute substantially to the miR164 miRNA pool in shoots.
|
). Cleavage products of the target transcripts that are
consistent with miR164 miRNA-dependent degradation were detected in
wild-type plants (Guo et al.,
2005; Kasschau et al.,
2003; Mallory et al.,
2004). Analysis of mir164 single mutants showed that
CUC1 and CUC2 transcript levels, but not those of the other
predicted targets, are elevated in shoot apices of mir164c plants as
compared with the wild type (Baker et al.,
2005), and that NAC1 transcripts are enriched in roots of
mir164a and mir164b mutants
(Guo et al., 2005).
Furthermore, expression of miR164 from the constitutive 35S promoter
led to a reduction in transcript levels of the predicted targets
(Guo et al., 2005;
Laufs et al., 2004),
indicating that these transcripts can be under miRNA-dependent regulation when
miR164 miRNAs are ectopically expressed. However, the individual
miR164 miRNAs are not expressed ubiquitously, but rather in specific
patterns (Fig. 3). Thus, it is
possible that certain targets, at least in some tissues, are not regulated by
miRNAs because their expression domains and those of the MIR164 genes
do not overlap. To test whether all of the predicted targets are subjected to miRNA-dependent regulation and whether there are tissue-specific differences in the degree to which individual transcripts are controlled by miR164 miRNAs, their transcript levels were measured in inflorescences, rosette leaves and seedlings of mir164abc triple-mutant plants by qRT-PCR. All of the predicted targets accumulated in mir164abc mutants to higher levels than in wild-type plants (Fig. 4A-C), confirming that they are indeed regulated by the endogenous miR164 miRNAs. Moreover, the extent to which transcripts of the targets accumulated in the different tissue samples varied substantially, indicating tissue-specific effects of miR164 miRNAs on target gene expression. These differences might be due to variable degrees of overlap between regions of target gene and miRNA expression in the tissues tested.
Regulation of CUC gene expression by miR164 miRNAs
It has been proposed that miRNAs control development by selectively
clearing cells of mRNAs that encode cell fate determinants, thereby promoting
rapid cell fate transitions and differentiation of cell lineages
(Rhoades et al., 2002). In
accordance with this idea, the plant miRNAs miR171, miR172 and
miR165/166 and their respective target mRNAs were found in adjacent,
but non-overlapping domains (Chen,
2004; Juarez et al.,
2004; Kidner and Martienssen,
2004; Parizotto et al.,
2004; Williams et al.,
2005). For miR164 miRNAs, however, a different mechanism
for the control of target gene expression has been proposed. In
mir164c single mutants, transcripts of CUC1 and
CUC2 were found to be elevated when compared with the wild type, but
remained restricted to cells in boundary regions
(Baker et al., 2005), implying
that miR164c does not act by clearing CUC1 and CUC2
mRNAs from non-boundary cells, but rather by regulating transcript abundance
in a pre-existing pattern.
In addition to miR164c, miR164a and miR164b are also
likely to be involved in regulating CUC1 and CUC2
expression, as inactivation of miR164a and miR164b leads to
an enhancement of the defects shown by mir164c plants (Figs
1 and
2). The possibility of
additional effects of miR164-dependent regulation on CUC1
and CUC2 expression was tested by examination of the distribution of
wild-type CUC1 and CUC2 and of miRNA cleavage-resistant
versions of these genes (CUC1m and
CUC2m), translationally fused to GFP and expressed from
their own promoters, in wild-type plants. CUC1-GFP and CUC2-GFP fusion
proteins were detected by CSLM in narrow columns of cells that separate flower
primordia and the inflorescence meristem
(Fig. 5A,B), in agreement with
the reported boundary-specific expression of the corresponding genes
(Aida et al., 1999;
Takada et al., 2001). By
contrast, the miRNA-resistant versions CUC1m-GFP and
CUC2m-GFP accumulated in boundaries to much higher levels than the
wild-type proteins, as well as weakly in the center of meristems
(Fig. 5, compare C,D with A,B,
respectively). These differences in protein accumulation strongly suggest that
miR164 miRNAs function by dampening the transcript levels of
CUC1 and CUC2, so that the initially strong expression in
boundary regions is greatly reduced, whereas the weak expression in meristems
is repressed below the detection limit of CLSM. These results also imply that
miRNA-dependent regulation is not required per se for the expression of
CUC1 and CUC2 in boundaries, and thus that the establishment
of the expression patterns for these genes is largely under transcriptional
control. In agreement with this idea, transcriptional reporters for
CUC1 and CUC2 showed strong expression in the expected
pattern (Fig. 5E,F).
|
|
) (this
study), suggest that overexpression of CUC2, and not of
CUC1, is responsible for this aspect of the mutant phenotype. For
CUC1, elevated levels of transcript accumulation were seen in
inflorescences of mir164abc triple-mutant plants within the normal
domain of expression (Fig.
6D,F) as compared with wild type
(Fig. 6C,E). In addition,
expression was observed in apparently random patches of cells within and
especially in between floral primordia
(Fig. 6D,F), where cell
division activities are often reduced in the triple mutant
(Fig. 2). Thus, certain
phenotypic alterations of mir164abc plants, such as the carpel fusion
defect and the reduced growth between flower primordia, are tightly correlated
with ectopic target transcript accumulation.
|
|
). These
plants form flowers with misshapen sepals, petals and stamens, which are
significantly reduced in size compared with the organs of wild-type flowers
(Baker et al., 2005). To
determine whether these growth defects are a consequence of reduced cell
division rates, a decrease in cell elongation, or both, we examined the
abaxial epidermis of sepals of 35S::CUC1m-GFP plants
(Fig. 7C), which were about
fourfold shorter than those of the wild type
(Fig. 7E). Compared with
wild-type sepals (Fig. 7A) and
those of miR164abc triple mutants
(Fig. 7B), the average size of
the epidermal cells was not significantly changed in the
CUC1-overexpressor lines (Fig.
7D,F,G). Thus, the dramatically reduced length of sepals of the
transgenic plants is not caused by an inhibition of cell elongation, and
therefore must be due to a reduction in cell number, suggesting that
CUC genes function by repressing cell division. The elevated levels
of CUC activity in between floral primordia, where growth is severely reduced
in mir164abc triple-mutant plants, might lead to a suppression of
growth between neighboring primordia, thus keeping primordia together while
the stem continues to grow. Similarly, the expansion of the CUC
expression domain in mir164abc triple mutants into meristems (Figs
5 and
6) might interfere with
primordium formation in the flower and hence cause organ-number defects. |
DISCUSSION |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTS Analysis ofDISCUSSIONREFERENCES |
|---|
Functional redundancy is often found in plants, as well as in animals,
among protein-coding genes that originated from gene or genome duplications.
Although gene duplicates are thought to be often lost over time, retaining
duplicated genes can be beneficial for an organism because they might buffer
fundamental developmental processes from the detrimental effects of random
mutations (Chapman et al.,
2006). It is also possible that duplicated genes functionally
diverge over time and undergo functional specialization
(subfunctionalization), or acquire functions other than that of the progenitor
gene (neofunctionalization). These processes are often a result of mutations
in the regulatory regions of the gene duplicates that can lead to distinct
spatial and/or temporal expression patterns. It has been proposed that miRNAs
evolved from their targets by inverted duplication
(Allen et al., 2004), and
recent evidence suggests that large-scale segmental duplications may play a
key role in the establishment of miRNA families in plants, including the
MIR164 family (Maher et al.,
2006). Thus, miRNA genes might evolve similarly to protein-coding
genes. The finding that the miR164 miRNAs, though largely
functionally redundant, contribute differently to certain aspects of
development is therefore in agreement with functional diversification through
subfunctionalization, an idea that is further supported by the distinct, but
partially overlapping expression patterns of the individual MIR164
miRNA genes.
miR164 miRNAs contribute to the robustness of development
The absence of miR164 miRNAs leads to phenotypic alterations that
are correlated with elevated and/or ectopic target transcript accumulation.
This suggests that the role of miR164 miRNAs in development is to
prevent fluctuations in target gene expression and, thus, to increase the
precision of the developmental programs underlying organogenesis and to
protect them from the intrinsic stochasticity of biochemical processes such as
transcription and translation. miR164 miRNAs appear to control
development by dampening transcript accumulation of their targets, where their
expression patterns and those of the targets overlap. Furthermore,
CUC1 and CUC2 expression domains are enlarged in
mir164abc mutant inflorescence meristems
(Fig. 6), indicating that the
miR164 miRNAs can spatially limit target mRNA accumulation in
addition to reducing the levels of target transcripts. These seemingly
different effects are likely to be a consequence of spatial differences in
target transcript accumulation. Where target gene expression is high, the pool
of miR164 miRNAs might not suffice to efficiently clear the target
transcripts from cells. By contrast, the level of miR164 miRNAs may
be high enough to completely eliminate target transcripts where they are
expressed at comparatively low levels. This mode of action would be consistent
with findings that showed an miRNA-dependent reduction, but not an
elimination, of highly expressed transcripts in mammalian tissues
(Farh et al., 2005;
Sood et al., 2006), as well as
with the results of a recent study that reported miR168 and its
target AGO1 as being co-expressed in Arabidopsis
(Vaucheret et al., 2006).
Thus, dampening of gene expression is a mechanism of miRNA-target interaction
that is likely to be found in both plants and animals.
The ability of miRNAs to reduce fluctuations in transcript abundance
suggests that miRNAs may be involved in buffering developmental processes. In
the absence of the miR164 miRNAs, the domain of CUC
expression is less precise and can expand seemingly at random from boundary
regions into peripheral regions of the inflorescence meristem and also into
flower primordia (Fig. 6). This
indicates that transcriptional control per se lacks the accuracy to prevent
fluctuations in the CUC expression domains. The variability in flower
positioning in mir164abc mutants correlates with local alterations in
the CUC expression pattern and can be explained by the lack of
precision in the control of the CUC expression domain. These
observations are in agreement with a role in stabilizing developmental
processes, a function that has also been proposed for animal miRNAs
(Hornstein and Shomron,
2006).
Supplementary material
Supplementary material for this article is available at
http://dev.biologists.org/cgi/content/full/134/6/1051/DC1
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTS Analysis ofDISCUSSIONREFERENCES |
|---|
Abbott, A. L., Alvarez-Saavedra, E., Miska, E. A., Lau, N. C., Bartel, D. P., Horvitz, H. R. and Ambros, V. (2005). The let-7 MicroRNA family members mir-48, mir-84, and mir-241 function together to regulate developmental timing in Caenorhabditis elegans. Dev. Cell 9,403 -414.[CrossRef][Medline]
Aida, M., Ishida, T., Fukaki, H., Fujisawa, H. and Tasaka,
M. (1997). Genes involved in organ separation in Arabidopsis:
an analysis of the cup-shaped cotyledon mutant. Plant
Cell 9,841
-857.
Aida, M., Ishida, T. and Tasaka, M. (1999). Shoot apical meristem and cotyledon formation during Arabidopsis embryogenesis: interaction among the CUP-SHAPED COTYLEDON and SHOOT MERISTEMLESS genes. Development 126,1563 -1570.[Abstract]
Allen, E., Xie, Z., Gustafson, A. M., Sung, G. H., Spatafora, J. W. and Carrington, J. C. (2004). Evolution of microRNA genes by inverted duplication of target gene sequences in Arabidopsis thaliana. Nat. Genet. 36,1282 -1290.[CrossRef][Medline]
Aukerman, M. 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.
Baker, C. C., Sieber, P., Wellmer, F. and Meyerowitz, E. M. (2005). The early extra petals1 mutant uncovers a role for microRNA miR164c in regulating petal number in Arabidopsis. Curr. Biol. 15,303 -315.[CrossRef][Medline]
Bartel, D. P. (2004). MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116,281 -297.[CrossRef][Medline]
Baulcombe, D. (2004). RNA silencing in plants. Nature 431,356 -363.[CrossRef][Medline]
Chapman, B. A., Bowers, J. E., Feltus, F. A. and Paterson, A.
H. (2006). Buffering of crucial functions by paleologous
duplicated genes may contribute cyclicality to angiosperm genome duplication.
Proc. Natl. Acad. Sci. USA
103,2730
-2735.
Chen, X. (2005). micro RNA biogenesis and function in plants. FEBS Lett. 579,5923 -5931.[CrossRef][Medline]
Chen, X. M. (2004). A miRNA as a translational
repressor of APETALA2 in Arabidopsis flower development.
Science 303,2022
-2025.
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.[CrossRef][Medline]
Dunoyer, P., Lecellier, C. H., Parizotto, E. A., Himber, C. and
Voinnet, O. (2004). Probing the microRNA and small
interfering RNA pathways with virus encoded suppressors of RNA silencing.
Plant Cell 16,1235
-1250.
Eshed, Y., Baum, S. F. and Bowman, J. L. (1999). Distinct mechanisms promote polarity establishment in carpels of Arabidopsis. Cell 99,199 -209.[CrossRef][Medline]
Farh, K. K., Grimson, A., Jan, C., Lewis, B. P., Johnston, W.
K., Lim, L. P., Burge, C. B. and Bartel, D. P. (2005). The
widespread impact of mammalian MicroRNAs on mRNA repression and evolution.
Science 310,1817
-1821.
Guo, H. S., Xie, Q., Fei, J. F. and Chua, N. H.
(2005). MicroRNA directs mRNA cleavage of the transcription
factor NAC1 to downregulate auxin signals for arabidopsis lateral root
development. Plant Cell
17,1376
-1386.
Heisler, M., Ohno, C., Das, P., Sieber, P., Reddy, G. V. and Meyerowitz, E. M. (2005). Patterns of auxin transport and gene expression during primordium development revealed by live imaging of the Arabidopsis inflorescence mersistem. Curr. Biol. 15,1899 -1911.[CrossRef][Medline]
Hornstein, E. and Shomron, N. (2006). Canalization of development by microRNAs. Nat. Genet. 38,S20 -S24.
Juarez, M., Kui, J., Thomas, J., Heller, B. and Timmermans, M. (2004). microRNA-mediated repression of rolled leaf1 specifies maize leaf polarity. Nature 428, 84-88.[CrossRef][Medline]
Kasschau, K. D., Xie, Z. X., Allen, E., Llave, C., Chapman, E. J., Krizan, K. A. and Carrington, J. C. (2003). P1/HC-Pro, a viral suppressor of RNA silencing, interferes with Arabidopsis development and miRNA function. Dev. Cell 4, 205-217.[CrossRef][Medline]
Kidner, C. A. and Martienssen, R. (2004). Spatially restricted microRNA directs leaf polarity through ARGONAUTE1. Nature 428,81 -84.[CrossRef][Medline]
Kim, J., Jung, J. H., Reyes, J. L., Kim, Y. S., Kim, S. Y., Chung, K. S., Kim, J. A., Lee, M., Lee, Y., Narry Kim, V. et al. (2005). microRNA-directed cleavage of ATHB15 mRNA regulates vascular development in Arabidopsis inflorescence stems. Plant J. 42,84 -94.[CrossRef][Medline]
Laufs, P., Peaucelle, A., Morin, H. and Traas, J.
(2004). MicroRNA regulation of the CUC genes is required
for boundary size control in Arabidopsis meristems.
Development 131,4311
-4322.
Lim, L. P., Lau, N. C., Garrett-Engele, P., Grimson, A., Schelter, J. M., Castle, J., Bartel, D. P., Linsley, P. S. and Johnson, J. M. (2005). Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature 433,769 -773.[CrossRef][Medline]
Llave, C., Xie, Z. X., 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.
Long, J. A. and Barton, M. K. (1998). The development of apical embryonic pattern in Arabidopsis.Development 125,3027 -3035.[Abstract]
Lu, C., Kulkarni, K., Souret, F. F., MuthuValliappan, R., Tej,
S. S., Poethig, R. S., Henderson, I. R., Jacobsen, S. E., Wang, W., Green, P.
J. et al. (2006). MicroRNAs and other small RNAs enriched in
the Arabidopsis RNA-dependent RNA polymerase-2 mutant. Genome
Res. 16,1276
-1288.
Maher, C., Stein, L. and Ware, D. (2006).
Evolution of Arabidopsis microRNA families through duplication events.
Genome Res. 16,510
-519.
Mallory, A. C. and Vaucheret, H. (2006). Functions of microRNAs and related small RNAs in plants. Nat. Genet. 38,S31 -S36.
Mallory, A. C., Dugas, D. V., Bartel, D. P. and Bartel, B. (2004). MicroRNA regulation of NAC-domain targets is required for proper formation and separation of adjacent embryonic, vegetative, and floral organs. Curr. Biol. 14,1035 -1046.[CrossRef][Medline]
Nikovics, K., Blein, T., Peaucelle, A., Ishida, T., Morin, H.,
Aida, M. and Laufs, P. (2006). The balance between the
MIR164A and CUC2 genes controls leaf margin serration in
Arabidopsis. Plant Cell
18,2929
-2945.
Palatnik, J. F., Allen, E., Wu, X. L., Schommer, C., Schwab, R., Carrington, J. C. and Weigel, D. (2003). Control of leaf morphogenesis by microRNAs. Nature 425,257 -263.[CrossRef][Medline]
Parizotto, E. A., Dunoyer, P., Rahm, N., Himber, C. and Voinnet,
O. (2004). In vivo investigation of the transcription,
processing, endonucleolytic activity, and functional relevance of the spatial
distribution of a plant miRNA. Genes Dev.
18,2237
-2242.
Park, W., Li, J. J., Song, R. T., Messing, J. and Chen, X. M. (2002). CARPEL FACTORY, a Dicer homolog, and HEN1, a novel protein, act in microRNA metabolism in Arabidopsis thaliana. Curr. Biol. 12,1484 -1495.[CrossRef][Medline]
Rajagopalan, R., Vaucheret, H., Trejo, J. and Bartel, D. P.
(2006). A diverse and evolutionarily fluid set of microRNAs in
Arabidopsis thaliana. Genes Dev.
20,3407
-3425.
Reddy, G. V., Heisler, M. G., Ehrhardt, D. W. and Meyerowitz, E.
M. (2004). Real-time lineage analysis reveals oriented cell
divisions associated with morphogenesis at the shoot apex of Arabidopsis
thaliana. Development
131,4225
-4237.
Reinhardt, D. (2005). Phyllotaxis - a new chapter in an old tale about beauty and magic numbers. Curr. Opin. Plant Biol. 8,487 -493.[CrossRef][Medline]
Reinhart, B. J., Weinstein, E. G., Rhoades, M. W., Bartel, B. and Bartel, D. P. (2002). MicroRNAs in plants.
