Transcriptional Regulation of Arabidopsis MIR168a ... · PDF fileARGONAUTE1 Homeostasis in...

Transcriptional Regulation of Arabidopsis MIR168a andARGONAUTE1 Homeostasis in Abscisic Acid and AbioticStress Responses1[W]

Wei Li, Xiao Cui, Zhaolu Meng, Xiahe Huang, Qi Xie, Heng Wu, Hailing Jin,Dabing Zhang, and Wanqi Liang*

School of Life Sciences and Biotechnology, Shanghai Jiao Tong University, Shanghai 200240, China (W.L., X.C.,Z.M., H.W., D.Z., W.L.); State Key Laboratory of Plant Genomics, National Center for Plant Gene Research,Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China (X.H.,Q.X.); and Department of Plant Pathology and Microbiology, University of California, Riverside, Riverside,California 92521 (H.J.)

The accumulation of a number of small RNAs in plants is affected by abscisic acid (ABA) and abiotic stresses, but theunderlying mechanisms are poorly understood. The miR168-mediated feedback regulatory loop regulates ARGONAUTE1(AGO1) homeostasis, which is crucial for gene expression modulation and plant development. Here, we reveal a transcrip-tional regulatory mechanism by which MIR168 controls AGO1 homeostasis during ABA treatment and abiotic stress responsesin Arabidopsis (Arabidopsis thaliana). Plants overexpressing MIR168a and the AGO1 loss-of-function mutant ago1-27 displayABA hypersensitivity and drought tolerance, while the mir168a-2 mutant shows ABA hyposensitivity and drought hyper-sensitivity. Both the precursor and mature miR168 were induced under ABA and several abiotic stress treatments, but noobvious decrease for the target of miR168, AGO1, was shown under the same conditions. However, promoter activity analysisindicated that AGO1 transcription activity was increased under ABA and drought treatments, suggesting that transcriptionalelevation of MIR168a is required for maintaining a stable AGO1 transcript level during the stress response. Furthermore, weshowed both in vitro and in vivo that the transcription of MIR168a is directly regulated by four abscisic acid-responsiveelement (ABRE) binding factors, which bind to the ABRE cis-element within the MIR168a promoter. This ABRE motif is alsofound in the promoter of MIR168a homologs in diverse plant species. Our findings suggest that transcriptional regulation ofmiR168 and posttranscriptional control of AGO1 homeostasis may play an important and conserved role in stress response andsignal transduction in plants.

As sessile organisms, plants have evolved sophisti-cated mechanisms to circumvent the adverse environ-ment. The phytohormone abscisic acid (ABA) is akey inducer of plant responses to abiotic stress. Mucheffort has been made to decipher the mechanism un-derlying ABA signal transduction, in which posttrans-criptional regulation has been shown to be one of themost important modes of regulation (Chinnusamyet al., 2008). Mutants of several components of themicroRNA (miRNA) biogenesis machinery, such ashyponastic leaves1, serrate, dicer-like1, hua enhancer1(hen1), and cap-binding protein (cbp20 and cbp80), arehypersensitive to ABA (Lu and Fedoroff, 2000;Hugouvieux et al., 2001; Kim et al., 2008; Zhang et al.,2008). These mutants, except hen1, were shown to have

lower miRNA levels but higher primary miRNA pre-cursor (pri-miRNA) levels compared with wild-typeplants (Laubinger et al., 2008). In cbp20 and cbp80mutants, ABA induction of miR159 was delayed andthe miR159 target transcripts, which encode two pos-itive regulators of ABA response, MYB DOMAINPROTEIN33 (MYB33) and MYB101, accumulated to ahigher level (Reyes and Chua, 2007; Kim et al., 2008).

ARGONAUTE (AGO) proteins, which contain thewell-characterized Piwi-Argonuaute-Zwille (PAZ)and P element-Induced WImpy testis (PIWI) domains,are considered to be integral players in all knownsmall RNA-targeted regulatory pathways (Vaucheret,2008). Among the AGO proteins in Arabidopsis (Arab-idopsis thaliana), AGO1 is a core component of theRNA-induced silencing complex, which associateswith miRNAs and inhibits target genes by mRNAcleavage and/or translational repression (Vaucheretet al., 2004; Vaucheret, 2008; Voinnet, 2009). Mutationsin AGO1 cause increased accumulation of miRNAtargets (Vaucheret et al., 2004; Ronemus et al., 2006;Kurihara et al., 2009). A hypomorphic allele, ago1-27exhibits hypersusceptibility to Cucumber mosaic virus(Morel et al., 2002), and AGO1 was proved to be amajor determinant for virus-induced gene silencing

1 This work was supported by the National Natural ScienceFoundation of China (grant no. 30600347).

* Corresponding author; e-mail [email protected] author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policydescribed in the Instructions for Authors (www.plantphysiol.org) is:Wanqi Liang ([email protected]).

[W] The online version of this article contains Web-only data.www.plantphysiol.org/cgi/doi/10.1104/pp.111.188789

Plant Physiology�, March 2012, Vol. 158, pp. 1279–1292, www.plantphysiol.org � 2012 American Society of Plant Biologists. All Rights Reserved. 1279 www.plantphysiol.orgon May 7, 2018 - Published by Downloaded from

Copyright © 2012 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

(Zhang et al., 2006). ago1-27 displays resistance to theinfection of a fungal pathogen, Verticillium dahliae, sup-porting the view that AGO1 is involved in the regulationof pathogen defense responses (Ellendorff et al., 2009).Although several AGO proteins have been shown toregulate biotic stresses (Katiyar-Agarwal et al., 2007; Liet al., 2010; Zhang et al., 2011), the role of AGO proteinsin regulating abiotic stress responses is still not wellunderstood. A recent report showed that a decrease inAGO1 confers hypersensitivity to ABA, whereas an in-crease in the level of AGO1 leads to ABA hyposensitiv-ity (Earley et al., 2010). However, the role of AGO1 inabiotic stress response pathways is still unclear.

It is well known that miRNAs play crucial roles incontrolling a variety of developmental processes, suchas organ identity establishment, organ separation, hor-mone signaling, and flowering time control (Chen,2009). Recent evidence also uncovered the role of smallRNA molecules in regulating plant stress responses(Sunkar, 2010; Khraiwesh et al., 2012). Microarrayanalysis and deep-sequencing data demonstrated thatthe accumulation of a number of miRNAs is affected byabiotic stresses (Sunkar and Zhu, 2004; Reyes and Chua,2007; Zhou et al., 2007, 2008; Liu et al., 2008; Jia et al.,2009), implying a potential role of miRNAs in abioticstress responses. In Arabidopsis, the expression ofMIR395, MIR398, and MIR399 is induced by sulfate,copper, and phosphate deprivation, respectively, andtheir subsequent down-regulation of target genes isessential for nutrient reallocation (Fujii et al., 2005;Jagadeeswaran et al., 2009; Kawashima et al., 2009). Up-regulation of miR159 and miR160 by ABA and the sub-sequent degradation of their target transcripts havebeen reported to be important for Arabidopsis seedgermination and seedling development (Liu et al., 2007;Reyes and Chua, 2007). Down-regulation of miR169 byABA and drought leads to the suppression of its cleavageof NUCLEAR FACTOR Y A5 (NFYA5), the positive reg-ulator of drought resistance (Li et al., 2008). In addition,natural cis-antisense small interfering RNAs (siRNAs)and transacting siRNAs also have been reported as im-portant players in adaptation to a variety of abiotic stres-ses (Borsani et al., 2005; Moldovan et al., 2010). However,the roles and regulatory mechanisms of most of the plantsmall RNAs in abiotic stresses remain largely unknown.

MIR168 is one of the most commonly detectedstress-inducibleMIR genes. Homologs ofMIR168 existin various plant species, including dicots such aspoplar (Populus trichocarpa), tobacco (Nicotiana taba-cum), and Arabidopsis and monocots such as maize(Zea mays) and rice (Oryza sativa). These homologshave been found to also respond to salt, drought, andcold stresses or ABA treatment (Liu et al., 2008; Dinget al., 2009; Jia et al., 2009, 2010; Zhou et al., 2010). Anewly identified MIR168a mutant, mir168a-2, exhibitsdevelopmental defects when exposed to temperature,light, and water fluctuations (Vaucheret, 2009). Thesefindings led to the speculation that miR168 and itsexclusive target AGO1 may play a critical role in thestress regulatory network.

In this paper, we reveal the important role ofMIR168aand its target AGO1 in plant responses to ABA andseveral abiotic stresses. Plants either overexpressingMIR168a or carrying a mutation in AGO1 (ago1-27) dis-play ABA hypersensitivity and drought tolerance, whilethe mir168a-2 mutant shows ABA hyposensitivity anddrought hypersensitivity. Consistent with this,MIR168ais transcriptionally up-regulated by ABA and variousstresses, resulting in posttranscriptional control of AGO1homeostasis. Furthermore, we demonstrate thatMIR168ais activated by abscisic acid-responsive element (ABRE)-binding transcription factors ABF1, ABF2, ABF3, andABF4. A typical ABRE motif within the MIR168a pro-moter, which can be bound by the four ABRE-bindingtranscription factors, is conserved in the MIR168 ho-mologs in other plant species. These results imply acommon and conserved mechanism of MIR168 tran-scriptional control in plant stress responses.

RESULTS

Transcriptional Regulation of MIR168a and AGO1under ABA and Abiotic Stresses

Recent investigations using microarray followed byreverse transcription (RT)-PCR analyses revealed in-creased accumulation of miR168 transcripts underhigh salinity, drought, and cold conditions in Arabi-dopsis (Liu et al., 2008). To investigate how the level ofmiR168 is induced, we analyzed the abundance ofprecursor and mature miR168 under stress conditions.Given that MIR168a is ubiquitously expressed andcoexpressed with AGO1 while MIR168b expression isrestricted to the shoot apical meristem (Jiang et al.,2006; Gazzani et al., 2009; Vaucheret, 2009), we focusedon MIR168a in this study.

Northern-blot analysis was performed to measurethe abundance of pre-miR168a transcripts and maturemiR168 in 2-week-old wild-type Arabidopsis seed-lings treated by drought, 300 mM NaCl, and cold for0, 6, and 12 h. The levels of mature miR168 and pre-miR168a were increased under drought, salt, and coldtreatment (Fig. 1A). ABA is an important signal inabiotic stress responses. Thus, we investigated theeffect of 100 mM ABA on the accumulation of miR168and pre-miR168a in 2-week-old wild-type Arabidopsisseedlings at 0, 2, 4, 6, and 12 h. The response to ABAand stress treatments was confirmed by the increasedexpression of RESPONSIVE TO DESICCATION29A(RD29A; Supplemental Fig. S1), a gene known to beinduced under such conditions (Narusaka et al., 2003).Northern-blot analysis showed the elevation of miR168and pre-miR168a levels 2 h after treatment with 100 mM

ABA (Fig. 1A). Furthermore, drought-induced accumu-lation of miR168 was substantially reduced in the ABA-deficient mutant aba1-5 (Leon-Kloosterziel et al., 1996;Fig. 1C), suggesting that the enhanced accumulationof miR168 by abiotic stress is at least partially ABAdependent.

Li et al.

1280 Plant Physiol. Vol. 158, 2012 www.plantphysiol.orgon May 7, 2018 - Published by Downloaded from



miR168 is critical for modulating the level of AGO1by miR168-targeted cleavage of the AGO1 mRNA andtranslational repression of AGO1 (Vaucheret et al.,2006; Vaucheret, 2009; Varallyay et al., 2010). To deter-mine whether ABA and stress treatments affect AGO1accumulation, we performed quantitative (q) RT-PCR

analyses. Unexpectedly, AGO1’s mRNA level wasweakly increased by ABA or other stresses and showedabout 2-fold increase at 12 h of drought treatment (Fig.1B). In order to clarify the role of miR168 in ABA andstress responses, we compared the expression levels ofendogenousAGO1 and exogenousGUS in pAGO1:GUS

Figure 1. Expression analysis of mature miR168, pre-miR168a, AGO1, and GUS in wild-type, aba1-5, and pAGO1:GUStransgenic plants under ABA and abiotic stress treatments. A and B, Northern-blot analysis of mature miR168 and pre-miR168a(A) and qRT-PCR analysis of AGO1mRNA (B) in 2-week-old wild-type Arabidopsis seedlings (Col) with drought, 300 mM NaCl,cold, and 100 mM ABA treatments. The significant induction of AGO1 mRNA by a 12-h drought treatment is indicated with theasterisk (n = 3, P , 0.05, Student’s t test). nt, Nucleotides. C, Northern-blot analysis of miR168 and qRT-PCR analysis of AGO1mRNA in 2-week-old wild-type and aba1-5 mutant plants (n = 10) under standard and drought conditions. Numbers below theblots in A and C show the relative abundance compared with the control U6; the level of each target in the nontreated wild type(Col) was set to 1. These experiments were repeated three times, and similar results were observed. D, qRT-PCR analysis of GUSand AGO1 mRNA in three independent 2-week-old pAGO1:GUS transgenic lines with drought treatment for 0, 6, and 12 h. E,qRT-PCR analysis ofGUS and AGO1mRNA in three independent 2-week-old pAGO1:GUS transgenic lines with the applicationof 100 mM ABA over time. The significant difference between GUS and AGO1 is indicated with asterisks (n = 3, P , 0.05,Student’s t test). Error bars denote SD. Quantifications were normalized to the expression of TUBULIN4. The level of AGO1 andGUS in the wild type (Col) or nontreated sample was set to 1.

Transcriptional Regulation of MIR168a

Plant Physiol. Vol. 158, 2012 1281 www.plantphysiol.orgon May 7, 2018 - Published by Downloaded from



transgenic plants with ABA or drought treatment. Thedata showed that the increase of GUS transcript freefrom miR168 regulation was significantly higher ascompared with the increase of endogenous AGO1 tran-script after ABA and drought treatments (Fig. 1, D andE). These results suggest that AGO1 is induced by ABAanddrought at the transcriptional level and thatmiR168-mediated repression occurs at the posttranscriptionallevel. Both regulatory mechanisms modulate the expres-sion of AGO1 in response to ABA and drought stress.

The ago1-27 Mutant and MIR168a Overexpression PlantsDisplay ABA and Salt Hypersensitivity

Although the accumulation of miR168 is affected byABA treatment and several abiotic stresses, the func-tion of miR168 and its exclusive target AGO1 in plantstress responses is still unclear. To assess the role ofmiR168 and AGO1 in stress responses, we examinedMIR168a overexpression lines (Jiang et al., 2006) andthe AGO1 hypomorphic mutant allele ago1-27 (Morelet al., 2002) under ABA and abiotic stress treatments.The levels of mature miR168 in two 35S:MIR168a trans-genic lines (lines 4 and 6) were more than 10-foldhigher than that in the wild type, while the AGO1transcript levels decreased to 35% and 28%, respec-

tively, of the wild-type levels (Supplemental Fig. S2).The overexpression lines 35S:MIR168a#4 and 35S:MIR168a#6 displayed some developmental defects,including shorter roots, smaller plant stature, anddelayed flowering.

We first examined the effects of ABA treatment onseed germination and seedling growth in theMIR168aoverexpression lines and the ago1-27 mutant. Understandard growth conditions, the germination of 35S:MIR168a lines was delayed by about 2 h comparedwith wild-type seeds, and a 4-h delay in germinationwas observed for ago1-27 seeds (Fig. 2A). We nextdetermined their ABA response during and after seedgermination. Treatment with 0.3 mM ABA caused aslight inhibition of seed germination and seedlinggrowth of the wild type, while the growth of 35S:MIR168a and ago1-27 seedlings was severely arrestedafter radicles emerged (Fig. 2B). Consistent with this,the root growth of 35S:MIR168a and ago1-27 seedlingswas hypersensitive to ABA treatment (Fig. 2C). Aftergrowing on Murashige and Skoog (MS) mediumcontaining 30 mM ABA for 5 d, the root length ofago1-27 and the two 35S:MIR168a lines decreased 85%,70%, and 75%, respectively, compared with a 40%decrease in wild-type plants. These results indicatethat disturbance of AGO1 function leads to enhanced

Figure 2. ago1-27 and MIR168a overexpression plants display ABA and salt hypersensitivity. A, 35S:MIR168a and ago1-27seeds showed delayed germination compared with wild-type seeds (Col) on ABA-free medium. Each data point represents themean of triplicate experiments (n = 100 each), and error bars indicate SE. B, Seven-day-old 35S:MIR168a and ago1-27 seedlingson MS medium containing 0.3 mM ABA. Representative images are shown. C, ABA inhibition of primary root elongation of 35S:MIR168a (lines 4 and 6) and ago1-27 seedlings. Seeds were germinated onMSmedium for 4 d, transferred to medium containing30 mM ABA, and root length was measured after a 5-d treatment. Representative images are shown. D, Seed germination onmedium containing 0, 50, 100, or 150 mMNaCl. Experiments were performed in triplicate (n = 100 each), and error bars indicate SE.

Li et al.




sensitivity to exogenous ABA treatment and are con-sistent with the results of Earley et al. (2010).In addition, we investigated the response of 35S:

MIR168a and ago1-27 lines to salt treatment. Thegermination efficiencies of 35S:MIR168a and ago1-27seeds were much lower on medium containing 50 or100 mM NaCl compared with wild-type seeds (Fig.2D). Under the treatment with 100 mM NaCl, the ger-mination efficiencies were 70% for the wild type, 42%and 37% for 35S:MIR168a#4 and -#6, respectively, andonly 18% for ago1-27. These results suggest that 35S:MIR168a and ago1-27 plants are hypersensitive to saltstress. After germination, the difference in growthbetween these lines and the wild type became lessapparent.

MIR168a Overexpression Plants and the ago1-27 MutantDisplay an Enhanced Resistance to Drought

Guard cells form the stoma on the leaf epidermis andrespond to environmental factors by changing stomatalaperture. Drought stress is able to promote the synthe-sis of ABA in plants, leading to a quick closure of stomato prevent water loss by transpiration (Blatt, 2000).Previous transcriptomic and proteomic analyses detectedAGO1 transcripts and proteins in Arabidopsis guardcells (Leonhardt et al., 2004; Zhao et al., 2008). Consis-tent with this, we observed MIR168a expression inguard cells in transgenic lines carrying pMIR168a:GUS(Supplemental Fig. S3) in this study.Given the drought-inducible expression of MIR168a

as well as the expression of MIR168a and AGO1 inguard cells, we hypothesized that miR168a and AGO1may play a role in plant drought tolerance. To addressthis point, we performed a drought tolerance experi-ment and found that both 35S:MIR168a and ago1-27plants displayed higher survival rates under waterdeficit conditions. Three-week-old plants were treated

with drought for 15 d before being rewatered. Twodays after the rewatering, only 46 out of 90 wild-typeplants survived, whereas all of the 35S:MIR168a andago1-27 plants recovered (Fig. 3A). We hypothesizedthat the enhanced drought tolerance of 35S:MIR168aand ago1-27 plants might have resulted, at least par-tially, from their lower transpiration rate. To test thisprediction, we examined the transpiration rate andstomatal aperture of 35S:MIR168a and ago1-27 plants.The water loss rates of 35S:MIR168a and ago1-27 lineswere 71% to 77% of that of wild-type plants (Fig. 3B).We also measured the stomatal opening of 35S:MIR168a and ago1-27 plants under drought conditions.The stomatal apertures of 35S:MIR168a and ago1-27plants were 78% to 84% of the wild-type value. Thestomatal closure was significantly enhanced in 35S:MIR168a and ago1-27 lines under drought conditions(Fig. 3C). These data suggest that overexpression ofMIR168a or mutation in AGO1 promotes stomatalclosure, resulting in drought tolerance.

The mir168a-2 Mutant Displays ABA Hyposensitivity

and Drought Hypersensitivity

To determine whether a loss-of-function mutant ofMIR168a exhibits the opposite phenotype to theMIR168aoverexpression plant in ABA and stress responses, weanalyzed the mir168a-2 mutant, which had been shownto have reduced levels of miR168 and increased AGO1transcripts (Vaucheret, 2009). Under standard condi-tions, mir168a-2 plants had more lateral roots than wild-type plants, but otherwise they grew normally. UnderABA treatment, wild-type plants showed inhibitions ofgermination and root growth, but these inhibitions werenot as serious in the mir168a-2mutants (Fig. 4, A and B).Under 0.3 mM ABA treatment, the germination efficiencyof the wild type decreased more than 90%, while that ofmir168a-2 decreased only about 60%. Consistently, when

Figure 3. ago1-27 and MIR168a overexpressionplants display enhanced drought tolerance. A,Three-week-old plants treated with drought for 15d before being rewatered. The photographs weretaken 2 d after rewatering. Number codes indi-cate the number of surviving plants out of the totalnumber (three repeats, with 90 plants in total). B,Transpiration rate of 4-week-old plants. Rosetteleaves (third and fourth leaves) at the same de-velopmental stages were detached and weighedat various times. Each data point represents themean of triplicate experiments (n = 10 each), anderror bars show SE. C, Stomatal aperture of leavesfrom 3-week-old plants with drought treatmentfor 6 d. Asterisks indicate significant differencecompared with wild type (n = 80, P , 0.05,Student’s t test). Error bars denote SE.





the seedlings were treated with 10 mM ABA, the rootgrowth rate of the wild type was reduced nearly 50%,while that of mir168a-2 displayed much less change.These results suggest that the mir168a-2 mutant hasreduced ABA sensitivity. When measured by the freshweight loss of detached rosette leaves, the water loss rateof mir168a-2 was 30% higher than that of the wild-typeplants (Fig. 4C). At the 6th d of water withholding, thestomatal aperture index of mir168a-2 leaves was 0.24,which was 18% greater than that of the wild type (Fig.4D). In contrast to that of 35S:MIR168a and ago1-27plants, the stomatal closure of the mir168a-2mutant wassignificantly reduced under water limitation conditions.Additionally, when wild-type andmir168a-2 plants weresubjected to 11 d of water-withholding conditions, asshown in Figure 4E, 2 d after the rewatering all of thewild-type plants recovered, whereas only 45 among 87mir168a-2 mutant lines survived, indicating that themir168a-2 mutation led to a reduction in plant toleranceto drought stress.

ABRE Motifs within the MIR168a Promoter AreResponsible for ABA-Responsive MIR168a Expression

To further elucidate the regulatory mechanism ofMIR168a in the ABA response, we screened the pro-

moter region of MIR168a using the PLACE predictionsoftware (http://www.dna.affrc.go.jp/PLACE/signalscan.html) and identified five putative ABRE motifs (M1–M5), all of which contain a core sequence of ACGT.M1 to M5 are located in the MIR168a promoter at2981 to 2978, 2875 to 2871, 2301 to 2297, 2275 to2271, and2126 to2122, respectively. The ABREmotifis a major class of ABA-responsive cis-elements and isfound in most of the genes regulated by ABA (Kim,2006). To test which of the ABREmotifs in theMIR168apromoter are mainly responsible for the ABA response,a series of truncated fragments of theMIR168apromoterwere fused to GUS (Fig. 5A), and the resulting con-structs were transformed into wild-type plants. Trans-genic plants containing an 800-bp upstream region fromthe transcription start site of the MIR168a promoter(pMIR168aD1:GUS) had similar GUS activities to plantswith the 1,000-bp MIR168a promoter (pMIR168aD0:GUS) with or without ABA treatment, suggesting thatthe M1 and M2 motifs are not essential for MIR168aexpression in response to ABA (Fig. 5B). When M3 andM4 were deleted, the pMIR168aD2:GUS plants showedreduced GUS activity, but they still could be signifi-cantly induced by the ABA treatment. When M5 wasremoved (pMIR168aD3:GUS), a decrease of theMIR168apromoter activity was observed with or without ABA

Figure 4. mir168a-2 plants display ABA hyposen-sitivity and drought hypersensitivity. A, Germina-tion of 4-d-old wild-type (WT) and mir168a-2seedlings on MS medium containing differentconcentrations (0, 0.1, 0.3, 0.5 mM) of ABA. Eachdata point represents the mean of triplicate ex-periments (n = 100 each), and error bars show SE.B, ABA inhibition of the primary root elongationof mir168a-2 seedlings. Seeds were germinatedon MS medium for 3 d, transferred to mediumcontaining 10 mM ABA, and root length wasmeasured after a 5-d treatment. Representativeimages are shown. C, Transpiration rate of4-week-old mir168a-2 leaves. Rosette leaves(third and fourth leaves) at the same developmen-tal stages were detached and weighed at varioustimes. Each data point represents the mean oftriplicate experiments (n = 10 each), and errorbars show SE. D, Stomatal aperture of 3-week-oldleaves with drought treatment for 6 d (n = 80, P,0.05, Student’s t test). Error bars denote SE. E,Three-week-old plants treated with drought for11 d before being rewatered. The photographswere taken 2 d after rewatering. Number codesindicate the number of surviving plants out ofthe total number (three repeats, with 87 plants intotal).

Li et al.




treatment, suggesting that M5 is crucial for the basalexpression and response of MIR168a to ABA. Further-more, when theM3,M4, andM5motifs were eliminatedsimultaneously (pMIR168aD4:GUS), the MIR168a pro-moter activity was reduced and almost no ABA re-sponse was detected. These results indicate that thethree ABREs near the transcription start site ofMIR168amainly contribute to MIR168a expression and the ABAresponse.

ABF1, ABF2, ABF3, and ABF4 Directly Interact with the

ABRE Motif in the MIR168a Promoter

The ABF family members, also referred to as AREB(for ABA-responsive element-binding protein), havebeen shown to bind ABRE in yeast one-hybrid screenanalysis (Choi et al., 2000). ABF/AREB proteins be-long to the bZIP class of transcription factors and arerequired for the ABA-mediated stress response by

Figure 5. ABF1, ABF2, ABF3, and ABF4 bind theMIR168a promoter in vitro and in vivo. A, Different constructs of theMIR168apromoter. The blue boxes indicate ABRE; the white boxes indicate mutated ABRE. The five ABREs were named M1 to M5,respectively. pMIR168aD0:GUS contains a 1,000-bp promoter region beginning from the transcription start site. pMIR168aD1:GUS contains an 800-bp promoter beginning from the transcription start site with M1 and M2 deleted. pMIR168aD2:GUScontains a 1,000-bp promoter with mutated M3 and M4. pMIR168aD3:GUS contains mutated M5. pMIR168aD4:GUS containsmutated M3, M4, and M5. B, GUS activity in 10 independent transgenic Arabidopsis plants (n = 15) harboring various constructsshown in Awith or without 6 h of 100 mM ABA treatment. Data show means6 SD for four independent experiments (* P, 0.05,Student’s t test). C, The bait DNA sequence used in yeast one-hybrid analysis contained three tandem repeats of a 25-bp fragment(5#-CAAGAGAACACGTGTCGGAAAATGA-3#) that includes the ABRE motif M5 within the MIR168a promoter. Six plasmidscontaining insert DNA fragments of ABF1, ABF2, ABF3, ABF4, ABI5, and TDR were retransformed into yeast strain YM4271carrying the reporter genes HIS and lacZ under the control of the 75-bp fragment containing three M5 motifs. The transformantswere examined for growth in the presence of 3-AT and b-galactosidase (b-gal) activity. D, Schematic diagram of the MIR168apromoter. The bent arrow indicates the translational start site. The blue boxes indicate ABRE motifs. F1, F2, and F3 representDNA fragments amplified in the quantitative ChIP-PCR assay. E, ChIP enrichment test showing the binding of ABF1-Myc, ABF2-Myc, ABF3-HA, and ABF4-HA to theMIR168a promoter after application of 100 mM ABA for 6 h. F, ChIP enrichment test showingthe binding of ABF1-Myc, ABF2-Myc, ABF3-HA, and ABF3-HA to theMIR168a promoter under drought conditions for 6 h. Errorbars denote SD. Asterisks in E and F indicate significant difference between enrichment fold of F3 and that of F1 (n = 5, P, 0.05,Student’s t test).





binding to the G-ABREmotif. In Arabidopsis, the ABF/AREB family consists of 13 members, eight of whichhave been reported to be involved in ABA signaltransduction (Bensmihen et al., 2002; Jakoby et al.,2002). These eight members can be further dividedinto two classes. One class includes ABF1, ABF2/AREB1, ABF3/AtDPBF5, and ABF4/AREB2, which areubiquitously expressed and induced by ABA and var-ious abiotic stresses (Jakoby et al., 2002; Kang et al.,2002; Kim et al., 2004). The other class includes ABAINSENSITIVE5 (ABI5), AREB3, DC3 PROMOTER-BINDING FACTOR2, and ENHANCED EM LEVEL,which are mainly expressed in embryos and involvedin seed maturation, germination, and early seedlingdevelopment (Bensmihen et al., 2002; Jakoby et al.,2002). Given that members of the ABF class share asimilar expression pattern with MIR168a (Kang et al.,2002; Kim et al., 2004; Gazzani et al., 2009; Vaucheret,2009), we speculated that ABFs may be associated withMIR168a expression.

To test our prediction, we first used the yeast one-hybrid assay to test the binding activity of ABFs to theM5motif within theMIR168a promoter (Fig. 5C). Yeastcells carrying plasmids containing the cDNAs ofABF1,ABF2,ABF3, andABF4were able to grow on amediumlacking His in the presence of 60 mM 3-aminotriazole(3-AT), but cells carrying plasmids containing thecDNAs of ABI5 (Bensmihen et al., 2002) and thenegative control, Tapetum Degeneration Retardation(TDR; Li et al., 2006), did not. All of the 3-AT-resistantyeast strains had induced lacZ activity and formedblue colonies on filter papers containing 5-bromo-4-chloro-3-indolyl-b-D-galactoside. These data indicatethat cDNAs of ABF1, ABF2, ABF3, and ABF4 encodeproteins that specifically bind to the M5 motif of theMIR168a promoter and activate the transcription of thereporter genes in yeast.

To further investigatewhether ABF1, ABF2, ABF3, andABF4 are able to physically interact with the MIR168apromoter in planta, chromatin immunoprecipitation(ChIP)-PCR was performed (Zhang et al., 2010). Trans-genic T2 lines of 35S:ABF1-Myc, 35S:ABF2-Myc, 35S:ABF3-HA, and 35S:ABF4-HA treated with 100 mM ABAor drought for 6 h were used for ChIP assays. Threeprimer pairs were designed to scan the MIR168a pro-moter (F1–F3; Fig. 5D). ChIP-PCR enrichment revealedthat ABF1, ABF2, ABF3, and ABF4 were mainly asso-ciated with the F3 region, which contains M3, M4, andM5 (Fig. 5, E and F), suggesting that ABF1, ABF2, ABF3,and ABF4 directly bind to the MIR168a promoter invivo in response to ABA or drought treatment.

The Expression of MIR168a Is Positively Regulatedby ABF1, ABF2, ABF3, and ABF4

To further determine whether ABF1, ABF2, ABF3,and ABF4 can regulate the expression of MIR168a invivo, the levels of mature miR168 and pre-miR168awere analyzed in 2-week-old lines overexpressingABF1, ABF2, ABF3, and ABF4 (Supplemental Fig. S4).

Without ABA treatment, the mature miR168 levelchanged less in the lines overexpressing ABF1 andABF2 and slightly increased in ABF3- and ABF4-over-expressing lines (Fig. 6A). Likewise, the levels of pre-miR168a were also slightly elevated in ABF2-, ABF3-,and ABF4-overexpressing lines. Surprisingly, AGO1transcript was increased also in the lines overexpress-ing ABF1, ABF2, ABF3, and ABF4 (Fig. 6B). Previousstudies show that ABA treatment enhances the activityof overexpressed ABF proteins to activate the down-stream genes (Finkelstein et al., 2005; Fujita et al., 2005).In accordance, under 6- and 12-h treatment with 100 mM

ABA, a much stronger induction of pre-miR168a andmature miR168 was detected in all of the ABF over-expression lines (Fig. 6A). Furthermore, the AGO1mRNA levels were significantly lower after a 12-htreatment, although they were still higher than thosein 12-h treated wild-type plants (Fig. 6B). Collectively,ABF1, ABF2, ABF3, and ABF4 may directly activate theexpression of MIR168a, but how they activate theexpression of AGO1 remains to be elucidated. In re-sponse to ABA, these four ABFs induce miR168, whichpartially contributes to the reduction of the AGO1transcript. We did not detect significant changes in theexpression of MIR168a in the individual abf mutant,which may be explained by the functional redundancyof ABF family members (Yoshida et al., 2010).

The ABRE Motif Is Conserved in PlantMIR168 Promoters

The increase of miR168 accumulation during ABAtreatment was also observed in poplar and tobacco (Jiaet al., 2009, 2010). Homologs of miR168 were found in15 plant species (Supplemental Table S1; miRBase,version16.0 [http://www.mirbase.org/index.shtml]).We also detected the ABA-induced expression ofmiR168 in rice and maize after a 6-h treatment with100 mM ABA (Supplemental Fig. S5). To understandwhether these diverse plant species share a similarregulatory mechanism for miR168 accumulation toArabidopsis, we scannedMIR168 upstream sequencesin 10 dicot and monocot species (Arabidopsis, Brassicanapus, Glycine max, Medicago truncatula, rice, poplar,Ricinus communis, Sorghum bicolor, Vitis vinifera, andmaize; Supplemental Table S2) using “profit” fromEMBOSS (Rice et al., 2000). The promoter regions ofthese MIR168 genes were predicted by definingthe transcription start site using the transcriptionstart site identification program for plants (http://linux1.softberry.com/berry.phtml?topic=tssp&group=programs&subgroup=promoter). We found that most of the tran-scription start sites were located within the 150-bpupstream region of the predicted hairpins of MIR168genes (Supplemental Table S3). For each upstreamsequence, a list of candidate ABRE sites was generatedby profit with percentage scores representing theirsimilarities to the known ABRE motifs (Gomez-Porraset al., 2007). Using an arbitrary 90% threshold ofsimilarity score, we found 24 putative ABRE motifs in

Li et al.




10MIR168 upstream sequences from eight species (Fig.7; Supplemental Table S3). Some of the ABRE motifsshare high similarity to M3, M4, andM5 in ArabidopsisMIR168. The putative ABRE motifs displayed a highoccurrence in the 500-bp upstream region starting fromthe beginning of the miR168 hairpin precursors(Fig. 7A). WebLogo analysis (http://weblogo.berkeley.edu) and a frequency matrix of the 24 putative ABREmotifs of the eight species showed that the coresequence ACGTG is conserved within the promoterregion of MIR168a homologs (Fig. 7, B and C). As thenegative control, we checked ABRE occurrences withinevery known promoter in the genome of Arabidopsison the POBO Web site (http://ekhidna.biocenter.helsinki.fi/poxo/pobo/pobo). POBO analysis showedthat the distribution of ABRE in MIR168 upstreamsequences was significantly higher than the back-ground set (Supplemental Fig. S6). These observationsindicate that ABA-regulated MIR168 expression is con-served in diverse plant species.

DISCUSSION

Interference with AGO1 Function Alters ABA andSeveral Abiotic Stress Responses

A number of reports on the identification and func-tional analysis of stress-responsiblemiRNAs and siRNAs

suggest that small RNAs may play an essential role instress responses (Zhou et al., 2007; Sunkar, 2010). In ad-dition, mutations in some genes involved in small RNAbiogenesis cause altered ABA and stress responses,confirming the regulatory role of the small RNA path-way in plant stress responses (Hugouvieux et al., 2001;Vazquez et al., 2004; Kim et al., 2008; Zhang et al., 2008).

Our data showed that not only proteins responsiblefor miRNA biogenesis but also the slicer required formiRNA action is employed in the posttranscriptionalregulation of plant abiotic stress. The reduced AGO1level in both the hypomorphic allele ago1-27 and theMIR168a overexpression lines leads to significantlyenhanced sensitivity toABA andmultiple abiotic stres-ses, such as increased seed dormancy, inhibition ofseed germination and root elongation, and whole-plant drought tolerance. In contrast, the mir168a-2mutant, which contains an increased level of AGO1(Vaucheret, 2009), displayed ABA hyposensitivity anddrought hypersensitivity. These results together indi-cate that AGO1 is a negative modulator of ABA sig-naling and stress responses in Arabidopsis.

Hypomorphic alleles of the AGO1 mutation do notsignificantly affect the stability of most miRNAs butreduce the cleavage efficiency of AGO1 (Vaucheretet al., 2004; Kurihara et al., 2009). Previous microarraydata revealed an overaccumulation of a large portionof miRNA target transcripts in ago1-9, ago1-11, and

Figure 6. Expression analysis ofmiR168 and AGO1 in the wild type and ABF1, ABF2, ABF3, and ABF4 overexpression lines withor without ABA treatment. A, Mature miR168 and pre-miR168a accumulated in the individual ABF1, ABF2, ABF3, and ABF4overexpression lines (ABF1OE, ABF2OE, ABF3OE, and ABF4OE) with 0, 6, and 12 h of ABA treatment by northern-blot analysis.This experiment was repeated three times independently, and similar results were observed. Numbers below the blots show therelative abundance comparedwith the control U6; the level of each target in the nontreatedwild type (Col) was set to 1. B, AGO1mRNA was also induced in the individual ABF1, ABF2, ABF3, and ABF4 overexpression lines with 0, 6, and 12 h of ABAtreatment by qRT-PCR analysis. However, the scale of the inductions was decreased by ABA. The expression level of AGO1mRNA in the wild-type plants under nonstressed conditions was defined as 1.0. Error bars denote SD (n = 3).





ago1-25 (Ronemus et al., 2006; Kurihara et al., 2009).Among the list are a number of targets of stress-responsive miRNAs, including the well-characterizedCOPPER/ZINC SUPEROXIDE DISMUTASE1/2(CSD1/2), NFYA5, ATP SULFURYLASE4 (APS4),AUXINRESPONSEFACTOR8 (ARF8), andUBIQUITIN-CONJUGATING ENZYME24 (UBC24). Targets ofmiR398, CSD1 and CSD2 encode reactive oxygen spe-cies scavengers that prevent the deleterious effects ofoxidants. Arabidopsis plants overexpressing a miR398-resistant version of CSD2 exhibit enhanced resistance tooxidative stresses (Sunkar et al., 2006). NFYA5, a targetof miR169, which is the positive regulator of droughtresistance, is up-regulated by ABA and drought (Liet al., 2008). APS4 is involved in miR395-mediatedsulfate assimilation (Hatzfeld et al., 2000). The circuitbetween miR167a and ARF8mediates nitrogen-inducedlateral root initiation and emergence (Gifford et al.,

2008). Down-regulation ofUBC24 bymiR399 is requiredfor primary root inhibition and phosphate transporterinduction in response to inorganic phosphate starvation(Fujii et al., 2005). Up-regulation of these targets affectsthe response of the plant to adverse environments.Indeed, we observed that the transcripts of CSD1/2,NFYA5, and APS4 were up-regulated in ago1-27 and35S:MIR168a lines (Supplemental Fig. S7).

miR168 Modulates AGO1 Homeostasis during ABAand Several Abiotic Stress Treatments

In wild-type Arabidopsis, AGO1 homeostasis ismaintained by an autoregulatory feedback loop thatinvolves the coexpression of MIR168 and AGO1 aswell as their interaction at the posttranscriptional level(Vaucheret et al., 2006; Vaucheret, 2009; Varallyayet al., 2010). AGO1 mRNA abundance is also con-trolled by siRNAs derived from its own transcripts(Mallory and Vaucheret, 2009). In addition, AGO1activity is further regulated negatively by AGO10 andpositively by a cyclophilin protein, SQUINT (Malloryet al., 2009; Smith et al., 2009).

Here, we show that a coordinate transcriptionalregulation of MIR168 and AGO1 also plays an impor-tant role in modulating plant stress responses. Previ-ous studies have demonstrated that miR168 is inducedby various stresses and ABA treatment (Liu et al., 2008;Jia et al., 2009, 2010). Our results further demonstratethat the increase of miR168 level is at least partiallycaused by the transcriptional activation of MIR168a, aubiquitously expressed miR168 gene. On the contrary,no significant change of AGO1 mRNA level wasdetected after a 12-h treatment with ABA, salt, andcold. Simultaneously, the AGO1 protein level displaysonly subtle changes in response to ABA and cold andis slightly induced by drought (Supplemental Fig. S8).Less repression of AGO1 abundance might result fromdistinct regulatory mechanisms at different regulatorylevels. Consistent with this hypothesis, we observedthat the promoter activity of the transgenic plantsharboring the pAGO1:GUS construct, which is freefrom miR168 inhibition, is induced by ABA treatment.The opposite effects of ABA on AGO1 at the transcrip-tional and posttranscriptional levels might reflect anarms race between the forces transmitting and atten-uating ABA signals.

AGO1 is one of the core factors crucial for miRNA-and siRNA-mediated gene silencing. The fluctuationin AGO1 level affects normal development and re-sponses to environmental stimuli in plants (Morelet al., 2002; Ellendorff et al., 2009; Varallyay et al.,2010). Thus, it is not surprising that plants evolvecomplicated mechanisms to fine-tune AGO1 levelsunder stress conditions. Recently, miR168 was repor-ted to be hijacked by plant viruses to alleviate plantantiviral function through inhibiting the translationalcapacity of AGO1 mRNA (Varallyay et al., 2010),indicating that miR168 acts as a rheostat exploited bydifferent regulatory loops to control AGO1 abun-

Figure 7. Plants have conserved ABRE motifs within the MIR168promoters. A, The localizations of ABRE motifs in 2-kb upstreamsequences of 10 plant MIR168 hairpin precursors in eight species (ath,Arabidopsis thaliana; bna, Brassica napus; gma, Glycine max; osa,Oryza sativa; ptc, Populus trichocarpa; rco, Ricinus communis; vvi,Vitis vinifera; zma, Zea mays). Yellow boxes denote the ABRE motif inthe sense strand (ABRE+); red boxes denote the ABRE motif in theantisense strand (ABRE2); blue arrows denote the transcription start site(TSS); two-way arrows denote the TATA box; blue lines denote thehairpin precursor of the MIR168 gene. B, Base composition of ABREmotifs that were found in the promoters of MIR168 from the eightcommon species. C, Frequency matrix of the ABRE motifs in thepromoters of MIR168 from the eight species.

Li et al.




dance. A novel negative regulator of AGO1, F-BOXWITH WD-40 2 (FBW2), was also reported to beresponsible for fine-tuning ABA signaling by control-ling AGO1 protein level (Earley et al., 2010). Similar tomir168a-2, mutations in FBW2 have nearly no effect onplant morphology but affect the sensitivity of plants toABA (Earley et al., 2010). These observations revealthat the AGO1 level is modulated by multiple mech-anisms in response to external stimuli. The observa-tion that AGO1 is transcriptionally up-regulated byABA suggests that there might be other yet unknownmechanisms involved in the transcriptional regulationof AGO1 in response to ABA.

miR168 Is a Component of the ABF TranscriptionalCassette and Mediates ABA-Dependent Stress Signaling

A wealth of evidence reveals that miRNAs play acrucial role in controlling gene expression at the post-transcriptional level by targeting mRNA cleavage ortranslational repression. However, little is knownabout the transcriptional regulation of miRNA genes.Previously, myeloid transcription factors, PU.1 andCAAT enhancer-binding protein a, were shown to di-rectly regulate the transcription of pri-miR223, and thisregulatory mechanism seems to be conserved in mouseand human (Fukao et al., 2007). In Arabidopsis, theMIR398 expression regulator SQUAMOSAPROMOTER-BINDING PROTEIN-LIKE7 (SPL7) is essential for thecopper deficiency response and is required for thedegradation of mRNAs that encode copper/zinc su-peroxide dismutases (Yamasaki et al., 2009). A nega-tive feedback loop, which involves the direct inductionof MIR172b transcription by the miR156 targets SPL9and SPL10, contributes to the stabilization of thetransition from juvenile to adult phases (Wu et al.,2009). Recently, the expression ofMIR164was shown tobe directly induced by TEOSINTE BRANCHED1 ANDCYCLOIDEA AND PCF TRANSCRIPTION FACTOR3to repress the expression of CUP-SHAPED COTYLE-DON genes (Koyama et al., 2010). The above observa-tions indicate that many of these posttranscriptionalregulators are transcriptionally controlled by develop-mental or environmental cues.Rapid accumulation of the MIR168a precursor after

ABA and stress treatments indicates that MIR168a isregulated by stress signaling at the transcriptionallevel. ABA is involved in the activation of MIR168,since the up-regulation of miR168 by drought treat-ment was attenuated in the ABA-deficient mutantaba1-5. A previous survey for cis-acting elements instress-responsive miRNAs revealed the presence ofmultiple ABREs in the MIR168a promoter (Liu et al.,2008), suggesting a possible link between MIR168atranscription and ABRE-binding proteins. A small cladeof basic Leu zipper transcription factors, called AREB/ABF, has been shown to convey ABA-dependent stresssignaling by binding to ABRE motifs within promotersof downstream genes and inducing gene expression(Kim, 2006). The AREB/ABF regulon is the most impor-

tant transcriptional cassette regulating ABA-dependentgene expression (Jakoby et al., 2002; Yoshida et al., 2010).AREB/ABF proteins function at an early step of ABAsignal transduction and regulate a large spectrum ofstress-responsive genes (Kim, 2006).

In this study, we present biochemical and geneticevidence for the direct activation of MIR168a expres-sion by four members of the AREB/ABF family. Sur-prisingly, an elevated AGO1 expression level was alsoobserved in these ABF overexpression lines. Thismight have resulted from an unknown feedback reg-ulatory loop of AGO1 by ABA or the ABF family. Thefluctuation of MIR168 and AGO1 expression impactsglobal miRNA action (Vaucheret et al., 2004; Kuriharaet al., 2009). Therefore, it is efficient to employ an initialregulator of one pathway to modulate the centralmodifier of another pathway, thus coordinating thecross talk of two pathways. In addition, the findingsthat miR168 is up-regulated by ABA and abioticstresses as well as that the conserved ABRE motifswere found in the MIR168 upstream sequences inmultiple species lead us to the conclusion that higherplants may share a common regulatory mechanism tocontrol miR168 expression in response to ABA andabiotic stimuli. However, we cannot exclude the pos-sibility that there may be some other transcriptionalfactors besides ABFs binding to the same or other cis-elements within the promoter of MIR168. Promoterdeletion analysis will help to uncover other playersinvolved in stress-regulated miR168 accumulation.

MATERIALS AND METHODS

Plant Materials and Stress Treatments

Arabidopsis (Arabidopsis thaliana) ecotypes Columbia (Col-0) and Lands-

berg erectawere used as the wild type. ago1-27 and aba1-5mutants in the Col-0

background were kindly provided by Dr. Herve Vaucheret and Hao Yu,

respectively. CS25625 (mir168a-2) in the Landsberg erecta background was

obtained from the Arabidopsis Biological Resource Center. Plants were grown

in a controlled chamber at 22�C, 70% relative humidity, under continuous

illumination. Two-week-old seedlings were subjected to various treatments.

For ABA treatment, plants were sprayed with 100 mM ABA and harvested at

various time points. For salt treatment, seedlings were transferred to solutions

containing 300 mM NaCl. For drought treatment, seedlings were dehydrated

on filter paper. For cold stress treatment, seedlings were kept at 4�C.

Plasmid Construction

The upstream fragment of MIR168a, spanning nucleotides 22,065 to

21 with respect to the MIR168a transcription start site (Jiang et al., 2006),

was amplified from Arabidopsis genomic DNA and then fused in frame with

the GUS coding region in pCAMBIA1301. To generate internal deletion

constructs, the ABREs located at nucleotides 2302 to 2272 and nucleotides

2126 to2119 were replaced with restriction endonuclease-binding sites (NcoI

and BglII) by PCR. The fusion between the promoter of AGO1 and GUS was

constructed according to Vaucheret et al. (2006).

The cauliflower mosaic virus 35S promoter-MIR168a construct was de-

scribed previously (Jiang et al., 2006). The coding regions of ABF1 and ABF2

were retrieved from SalI/BamHI-digested pEASY-Blunt vector and ligated to

SalI/BamHI-digested pCAMBIA1300-221-Myc vector to generate 35S:ABF1-

Myc and 35S:ABF2-Myc. 35S:ABF3-HA and 35S:ABF4-HA were generated by

inserting the fusion of the ABF coding region and the hemagglutinin (HA) tag

to the pCAMBIA1300-221 vector. These gene fusions were controlled by the





cauliflower mosaic virus 35S promoter and used for the transformation of

Arabidopsis.

Histochemical GUS Staining and Quantification of

GUS Activity

Rosette leaves of 3-week-old homozygous transgenic plants were har-

vested. Histochemical localization of GUS activity was determined as described

previously (Kang et al., 2002). For GUS quantification, at least 10 independent

transgenic lines for each construct were used, and 15 2-week-old T2 seedlings

from each line were combined to measure GUS activity. At least four repeats

were carried out for each line. SD between lines was calculated. GUS activity

was quantified using the fluorometric 4-methylumbelliferyl-b-D-glucuronide

assay method (Jefferson, 1987).

RNA Analysis

Total RNAwas extracted from control or treated plants with Trizol reagent

(Invitrogen). miR168 and pre-miR168a were detected by the probe comple-

mented with miR168 and pre-miR168a loop sequences, which were modified

with 3# and 5# biotin. Forty micrograms of total RNAwas loaded in each lane,

and U6 was used as a loading control (Li et al., 2007). After stringent washes,

signals on the membranes were detected according to the manufacturer’s

instructions for the Chemiluminescent Nucleic Acid Detection Module

(Thermo). qRT-PCR analyses were carried out with total RNA using Bio-

Rad’s real-time PCR amplification systems. Each gene was assayed on three

biological replicates and normalized using the TUBULIN4 cDNA level

(Magnan et al., 2008).

Germination and Seedling Growth Assay

Seeds of wild-type, ago1-27, 35S:MIR168a, and mir168a-2 plants were

collected at the same time and stored in the same conditions. They were

plated on ABA-free medium for 4 d at 4�C before switching to 22�C to

germinate, then the germination (fully emerged radicle) efficiency was scored

at various times. For germination under various NaCl doses, seeds were

plated on MS medium containing 0, 50, 100, and 150 mM NaCl for 4 d before

counting. For germination under various ABA doses, seeds were plated onMS

medium containing 0, 0.1, 0.3, and 0.5 mM ABA for 1 week. For the root growth

assay, plants were germinated on MS plates and transferred after 4 d to fresh

plates containing ABA at 10 mM for the assay of mir168a-2 and 30 mM for the

assay of ago1-27 and 35S:MIR168a transgenic lines. Photographs of the

elongated roots were taken 5 d later.

Drought Treatment and Measurement of TranspirationRate and Stomatal Aperture

For drought treatment, 3-week-old soil-grown plants were not watered for

11 to 15 d. Photographs were taken 2 d after rewatering. The transpiration rate

of detached leaves was measured by weighing freshly harvested leaves placed

on open dishes on the laboratory bench, with the abaxial side up. Leaves of

similar developmental stages from 4-week-old soil-grown plants were used

(Kang et al., 2002; Kim et al., 2004).

Stomatal aperture analysis was conducted according to a previous de-

scription (Li et al., 2008). Stomatal apertures were measured 6 d after

dewatering. For each treatment, 80 measurements were collected, and each

experiment was repeated at least three times.

Yeast One-Hybrid Analysis

The coding regions of the transcription factors ABF1, ABF2, ABF3, ABF4,

and ABI5 were amplified from Arabidopsis Col-0 cDNA using gene-specific

primers. The PCR products were then cloned into the pGADT7 prey vector

(Clontech), creating a translational fusion between the GAL4 activation

domain and the transcription factor. Three tandem repeats of the 2135 to

2110 MIR168a promoter region were synthesized and cloned into pHISi-1

vector (Clontech) carrying the HIS3 reporter gene and pLacZi vector

(Clontech) containing the LacZ reporter gene. Induction of HIS3 and LacZ

indicates positive interaction. The yeast one-hybrid assay was performed

according to the Clontech Yeast Protocols Handbook.

Quantitative ChIP-PCR Analysis

The procedure for ChIP of ABF-DNA complexes in the 35S:ABF1-Myc, 35S:

ABF2-Myc, 35S:ABF3-HA, and 35S:ABF4-HA transgenic Arabidopsis leaves

was modified from previous descriptions (Bowler et al., 2004). Chromatin was

isolated from 1.5 g of formaldehyde cross-linked tissue from leaves of 4-week-

old transgenic plants sprayed with 100 mM ABA for 6 h and seedlings of

2-week-old transgenic plants treated with drought for 6 h. Preabsorption

with protein A-Sepharose beads and immunoprecipitation with c-Myc or

HA monoclonal antibody (Sigma-Aldrich) were performed as described by

Bowler et al. (2004). A small aliquot of untreated sonicated chromatin was

reverse cross-linked and used as the total input DNA control. Quantitative

ChIP-PCR was performed with primers specific to different regions upstream

ofMIR168a. ChIP populations from immune or control immunoprecipitations

were used as templates and normalized by the amount of the DNA fragments

in corresponding input DNA populations. The difference between the nor-

malized mean cycle threshold of the immunized population and the pre-

immunized population was calculated to obtain the relative enrichment of

each upstream fragment (Zhang et al., 2010). The primers used in this article

are listed in Supplemental Table S5.

Sequence Analysis of MIR168 Promoters in Plants

Sequences of miR168 from Arabidopsis, Brassica napus, Glycine max,

Medicago truncatula, rice (Oryza sativa), poplar (Populus trichocarpa), Ricinus

communis, Sorghum bicolor, Vitis vinifera, and maize (Zea mays) and their

corresponding genome coordinates were obtained from miRBase (version

16.0; http://www.mirbase.org/index.shtml). Upstream sequences 2 kb rela-

tive to the beginning of the predicted hairpins were extracted from available

genome sequences according to their coordinates. These 2-kb upstream

sequences were then scanned to identify ABRE sites on both plus and minus

strands using profit from EMBOSS (Rice et al., 2000). A previously discovered

matrix of ABRE was adopted as the input matrix for profit (Gomez-Porras

et al., 2007). In addition, the plant promoter identification program (http://

linux1.softberry.com/berry.phtml?topic=tssp&group=programs&subgroup=

promoter) was used to check the transcription start site of these sequences.

Western-Blot Analysis

The AGO1 antibody was purchased from Agrisera and was used at 1:3,000

dilution in the immunoblot analysis. Anti-b-tubulin was purchased from

Santa Cruz Biotechnology and was used at 1:1,000 dilution in the experiments.

Accession numbers for the genes in this article are as follows: MIR168a

(AT4G19395), AGO1 (AT1G48410), TUBULIN4 (AT5G44340), RD29A (AT5G5

2310), ABA1 (AT5G67030), ABF1 (AT1G49720), ABF2 (AT1G45249), ABF3 (AT

4G34000), ABF4 (AT3G19290), ABI5 (AT2G36270), NFYA5 (AT1G54160), APS4

(AT5G43780), CSD1 (AT1G08830), and CSD2 (AT2G28190).

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Figure S1. qRT-PCR analysis of RD29A mRNA under ABA

and abiotic stress treatments.

Supplemental Figure S2. Expression analysis of mature miR168, pre-

miR168a, and AGO1 mRNA in the wild-type and MIR168a overexpres-

sion lines.

Supplemental Figure S3.Histochemical GUS staining in the guard cells of

pMIR168a:GUS transgenic plants.

Supplemental Figure S4. qRT-PCR analysis of ABF1, ABF2, ABF3, and

ABF4 transcript levels in overexpression lines.

Supplemental Figure S5. Northern-blot analysis of mature miR168 in

2-week-old seedlings of rice and maize with or without ABA treatment.

Supplemental Figure S6. POBO output interface.

Supplemental Figure S7. qRT-PCR analysis of miRNA targets in ago1-27

and 35S:MIR168a lines.

Supplemental Figure S8. Western-blot analysis of AGO1 in response to

ABA, cold, and drought.

Li et al.




Supplemental Table S1. Putative miR168 in 15 species.

Supplemental Table S2. Upstream sequences (2 kb) of MIR168 in 10

species.

Supplemental Table S3. The putative transcription start sites of MIR168

genes.

Supplemental Table S4.Candidate ABRE sites in the promoters ofMIR168

genes.

Supplemental Table S5. Primers and probers used in this article.

ACKNOWLEDGMENTS

We thank Dr. Herve Vaucheret for kindly providing the mutant ago1-27,

Dr. Hao Yu for providing the mutant aba1-5, Dr. Jianping Hu for editing the

manuscript, and The Arabidopsis Information Resource for supplying

mir168a-2 seeds. We also thank Dr. Yuke He for providing the protocol for

small RNA blot.

Received October 12, 2011; accepted January 12, 2012; published January 13,

2012.

LITERATURE CITED

Bensmihen S, Rippa S, Lambert G, Jublot D, Pautot V, Granier F,

Giraudat J, Parcy F (2002) The homologous ABI5 and EEL transcription

factors function antagonistically to fine-tune gene expression during

late embryogenesis. Plant Cell 14: 1391–1403

Blatt MR (2000) Cellular signaling and volume control in stomatal move-

ments in plants. Annu Rev Cell Dev Biol 16: 221–241

Borsani O, Zhu JH, Verslues PE, Sunkar R, Zhu JK (2005) Endogenous

siRNAs derived from a pair of natural cis-antisense transcripts regulate

salt tolerance in Arabidopsis. Cell 123: 1279–1291

Bowler C, Benvenuto G, Laflamme P, Molino D, Probst AV, Tariq M,

Paszkowski J (2004) Chromatin techniques for plant cells. Plant J 39:

776–789

Chen XM (2009) Small RNAs and their roles in plant development. Annu

Rev Cell Dev Biol 25: 21–44

Chinnusamy V, Gong ZZ, Zhu JK (2008) Abscisic acid-mediated epige-

netic processes in plant development and stress responses. J Integr Plant

Biol 50: 1187–1195

Choi HI, Hong JH, Ha JO, Kang JY, Kim SY (2000) ABFs, a family of ABA-

responsive element binding factors. J Biol Chem 275: 1723–1730

Ding D, Zhang LF, Wang H, Liu ZJ, Zhang ZX, Zheng YL (2009) Differ-

ential expression of miRNAs in response to salt stress in maize roots.

Ann Bot (Lond) 103: 29–38

Earley K, Smith M, Weber R, Gregory B, Poethig RS (2010) An endoge-

nous F-box protein regulates ARGONAUTE1 in Arabidopsis thaliana.

Silence 1: 15

Ellendorff U, Fradin EF, de Jonge R, Thomma BPHJ (2009) RNA silencing

is required for Arabidopsis defence against Verticillium wilt disease.

J Exp Bot 60: 591–602

Finkelstein R, Gampala SSL, Lynch TJ, Thomas TL, Rock CD (2005)

Redundant and distinct functions of the ABA response loci ABA-

INSENSITIVE(ABI)5 and ABRE-BINDING FACTOR (ABF)3. Plant Mol

Biol 59: 253–267

Fujii H, Chiou TJ, Lin SI, Aung K, Zhu JK (2005) A miRNA involved in

phosphate-starvation response in Arabidopsis. Curr Biol 15: 2038–2043

Fujita Y, Fujita M, Satoh R, Maruyama K, Parvez MM, Seki M, Hiratsu K,

Ohme-Takagi M, Shinozaki K, Yamaguchi-Shinozaki K (2005) AREB1

is a transcription activator of novel ABRE-dependent ABA signaling

that enhances drought stress tolerance in Arabidopsis. Plant Cell 17:

3470–3488

Fukao T, Fukuda Y, Kiga K, Sharif J, Hino K, Enomoto Y, Kawamura A,

Nakamura K, Takeuchi T, Tanabe M (2007) An evolutionarily con-

served mechanism for microRNA-223 expression revealed by micro-

RNA gene profiling. Cell 129: 617–631

Gazzani S, Li MG, Maistri S, Scarponi E, Graziola M, Barbaro E, Wunder

J, Furini A, Saedler H, Varotto C (2009) Evolution ofMIR168 paralogs in

Brassicaceae. BMC Evol Biol 9: 62

Gifford ML, Dean A, Gutierrez RA, Coruzzi GM, Birnbaum KD (2008)

Cell-specific nitrogen responses mediate developmental plasticity. Proc

Natl Acad Sci USA 105: 803–808

Gomez-Porras JL, Riano-Pachon DM, Dreyer I, Mayer JE, Mueller-

Roeber B (2007) Genome-wide analysis of ABA-responsive elements

ABRE and CE3 reveals divergent patterns in Arabidopsis and rice. BMC

Genomics 8: 260

Hatzfeld Y, Lee S, Lee M, Leustek T, Saito K (2000) Functional character-

ization of a gene encoding a fourth ATP sulfurylase isoform from

Arabidopsis thaliana. Gene 248: 51–58

Hugouvieux V, Kwak JM, Schroeder JI (2001) An mRNA cap binding

protein, ABH1, modulates early abscisic acid signal transduction in

Arabidopsis. Cell 106: 477–487

Jagadeeswaran G, Saini A, Sunkar R (2009) Biotic and abiotic stress down-

regulate miR398 expression in Arabidopsis. Planta 229: 1009–1014

Jakoby M, Weisshaar B, Droge-Laser W, Vicente-Carbajosa J, Tiedemann

J, Kroj T, Parcy F (2002) bZIP transcription factors in Arabidopsis. Trends

Plant Sci 7: 106–111

Jefferson RA (1987) Assaying chimeric genes in plants: the GUS gene

fusion system. Plant Mol Biol 5: 387–405

Jia XY, Mendu V, Tang GL (2010) An array platform for identification of

stress-responsive microRNAs in plants. Methods Mol Biol 639: 253–269

Jia XY, WangWX, Ren LG, Chen QJ, Mendu V, Willcut B, Dinkins R, Tang

XQ, Tang GL (2009) Differential and dynamic regulation of miR398 in

response to ABA and salt stress in Populus tremula and Arabidopsis

thaliana. Plant Mol Biol 71: 51–59

Jiang DH, Yin CS, Yu AP, Zhou XF, Liang WQ, Yuan Z, Xu Y, Yu QB, Wen

TQ, Zhang DB (2006) Duplication and expression analysis of multi-

copy miRNA gene family members in Arabidopsis and rice. Cell Res 16:

507–518

Kang JY, Choi HI, Im MY, Kim SY (2002) Arabidopsis basic leucine zipper

proteins that mediate stress-responsive abscisic acid signaling. Plant

Cell 14: 343–357

Katiyar-Agarwal S, Gao S, Vivian-Smith A, Jin H (2007) A novel class of

bacteria-induced small RNAs in Arabidopsis. Genes Dev 21: 3123–3134

Kawashima CG, Yoshimoto N, Maruyama-Nakashita A, Tsuchiya YN,

Saito K, Takahashi H, Dalmay T (2009) Sulphur starvation induces the

expression of microRNA-395 and one of its target genes but in different

cell types. Plant J 57: 313–321

Khraiwesh B, Zhu JK, Zhu JH (2012) Role of miRNAs and siRNAs in biotic

and abiotic stress responses of plants. Biochim Biophys Acta (in press)

Kim S, Kang JY, Cho DI, Park JH, Kim SY (2004) ABF2, an ABRE-binding

bZIP factor, is an essential component of glucose signaling and its

overexpression affects multiple stress tolerance. Plant J 40: 75–87

Kim S, Yang JY, Xu J, Jang IC, PriggeMJ, Chua NH (2008) Two cap-binding

proteins CBP20 and CBP80 are involved in processing primary micro-

RNAs. Plant Cell Physiol 49: 1634–1644

Kim SY (2006) The role of ABF family bZIP class transcription factors in

stress response. Physiol Plant 126: 519–527

Koyama T, Mitsuda N, Seki M, Shinozaki K, Ohme-Takagi M (2010) TCP

transcription factors regulate the activities of ASYMMETRIC LEAVES1

and miR164, as well as the auxin response, during differentiation of

leaves in Arabidopsis. Plant Cell 22: 3574–3588

Kurihara Y, Kaminuma E, Matsui A, KawashimaM, TanakaM,Morosawa

T, Ishida J, Mochizuki Y, Shinozaki K, Toyoda T, et al (2009) Tran-

scriptome analyses revealed diverse expression changes in ago1 and hyl1

Arabidopsis mutants. Plant Cell Physiol 50: 1715–1720

Laubinger S, Sachsenberg T, Zeller G, Busch W, Lohmann JU, Ratsch G,

Weigel D (2008) Dual roles of the nuclear cap-binding complex and

SERRATE in pre-mRNA splicing and microRNA processing in Arabi-

dopsis thaliana. Proc Natl Acad Sci USA 105: 8795–8800

Leonhardt N, Kwak JM, Robert N, Waner D, Leonhardt G, Schroeder JI

(2004) Microarray expression analyses of Arabidopsis guard cells and

isolation of a recessive abscisic acid hypersensitive protein phosphatase

2C mutant. Plant Cell 16: 596–615

Leon-Kloosterziel KMGM, Gil MA, Ruijs GJ, Jacobsen SE, Olszewski

NE, Schwartz SH, Zeevaart JA, Koornneef M (1996) Isolation and

characterization of abscisic acid-deficient Arabidopsis mutants at two

new loci. Plant J 10: 655–661

Li N, Zhang DS, Liu HS, Yin CS, Li XX, LiangWQ, Yuan Z, Xu B, Chu HW,

Wang J, et al (2006) The rice tapetum degeneration retardation gene is

required for tapetum degradation and anther development. Plant Cell

18: 2999–3014





Li WX, Oono Y, Zhu JH, He XJ, Wu JM, Iida K, Lu XY, Cui XP, Jin HL, Zhu

JK (2008) The Arabidopsis NFYA5 transcription factor is regulated

transcriptionally and posttranscriptionally to promote drought resis-

tance. Plant Cell 20: 2238–2251

Li X, Jiang DH, Yong KL, Zhang DB (2007) Varied transcriptional effi-

ciencies of multiple Arabidopsis U6 small nuclear RNA genes. J Integr

Plant Biol 49: 222–229

Li Y, Zhang Q, Zhang J, Wu L, Qi Y, Zhou JM (2010) Identification of

microRNAs involved in pathogen-associated molecular pattern-triggered

plant innate immunity. Plant Physiol 152: 2222–2231

Liu HH, Tian X, Li YJ, Wu CA, Zheng CC (2008) Microarray-based analysis

of stress-regulated microRNAs in Arabidopsis thaliana. RNA 14: 836–843

Liu PP, Montgomery TA, Fahlgren N, Kasschau KD, Nonogaki H,

Carrington JC (2007) Repression of AUXIN RESPONSE FACTOR10 by

microRNA160 is critical for seed germination and post-germination

stages. Plant J 52: 133–146

Lu C, Fedoroff N (2000) A mutation in the Arabidopsis HYL1 gene encoding

a dsRNA binding protein affects responses to abscisic acid, auxin, and

cytokinin. Plant Cell 12: 2351–2366

Magnan F, Ranty B, Charpenteau M, Sotta B, Galaud JP, Aldon D (2008)

Mutations in AtCML9, a calmodulin-like protein from Arabidopsis

thaliana, alter plant responses to abiotic stress and abscisic acid. Plant

J 56: 575–589

Mallory AC, Hinze A, Tucker MR, Bouche N, Gasciolli V, Elmayan T,

Lauressergues D, Jauvion V, Vaucheret H, Laux T (2009) Redundant

and specific roles of the ARGONAUTE proteins AGO1 and ZLL in

development and small RNA-directed gene silencing. PLoS Genet 5:

e1000646

Mallory AC, Vaucheret H (2009) ARGONAUTE 1 homeostasis invokes the

coordinate action of the microRNA and siRNA pathways. EMBO Rep

10: 521–526

Moldovan D, Spriggs A, Yang J, Pogson BJ, Dennis ES, Wilson IW (2010)

Hypoxia-responsive microRNAs and trans-acting small interfering

RNAs in Arabidopsis. J Exp Bot 61: 165–177

Morel JB, Godon C, Mourrain P, Beclin C, Boutet S, Feuerbach F, Proux F,

Vaucheret H (2002) Fertile hypomorphic ARGONAUTE (ago1) mutants

impaired in post-transcriptional gene silencing and virus resistance.

Plant Cell 14: 629–639

Narusaka Y, Nakashima K, Shinwari ZK, Sakuma Y, Furihata T, Abe H,

Narusaka M, Shinozaki K, Yamaguchi-Shinozaki K (2003) Interaction

between two cis-acting elements, ABRE and DRE, in ABA-dependent

expression of Arabidopsis rd29A gene in response to dehydration and

high-salinity stresses. Plant J 34: 137–148

Reyes JL, Chua NH (2007) ABA induction of miR159 controls transcript

levels of two MYB factors during Arabidopsis seed germination. Plant J

49: 592–606

Rice P, Longden I, Bleasby A (2000) EMBOSS: the European Molecular

Biology Open Software Suite. Trends Genet 16: 276–277

Ronemus M, Vaughn MW, Martienssen RA (2006) MicroRNA-targeted

and small interfering RNA-mediated mRNA degradation is regulated

by argonaute, dicer, and RNA-dependent RNA polymerase in Arabi-

dopsis. Plant Cell 18: 1559–1574

Smith MR, Willmann MR, Wu G, Berardini TZ, Moller B, Weijers D,

Poethig RS (2009) Cyclophilin 40 is required for microRNA activity in

Arabidopsis. Proc Natl Acad Sci USA 106: 5424–5429

Sunkar R (2010) MicroRNAs with macro-effects on plant stress responses.

Semin Cell Dev Biol 21: 805–811

Sunkar R, Kapoor A, Zhu JK (2006) Posttranscriptional induction of two

Cu/Zn superoxide dismutase genes in Arabidopsis is mediated by

downregulation of miR398 and important for oxidative stress tolerance.

Plant Cell 18: 2051–2065

Sunkar R, Zhu JK (2004) Novel and stress-regulated microRNAs and other

small RNAs from Arabidopsis. Plant Cell 16: 2001–2019

Varallyay E, Valoczi A, Agyi A, Burgyan J, Havelda Z (2010) Plant virus-

mediated induction of miR168 is associated with repression of ARGO-

NAUTE1 accumulation. EMBO J 29: 3507–3519

Vaucheret H (2008) Plant ARGONAUTES. Trends Plant Sci 13: 350–358

Vaucheret H (2009) AGO1 homeostasis involves differential production of

21-nt and 22-nt miR168 species by MIR168a and MIR168b. PLoS ONE 4:

e6442

Vaucheret H, Mallory AC, Bartel DP (2006) AGO1 homeostasis entails

coexpression of MIR168 and AGO1 and preferential stabilization of

miR168 by AGO1. Mol Cell 22: 129–136

Vaucheret H, Vazquez F, Crete P, Bartel DP (2004) The action of ARGO-

NAUTE1 in the miRNA pathway and its regulation by the miRNA

pathway are crucial for plant development. Genes Dev 18: 1187–1197

Vazquez F, Gasciolli V, Crete P, Vaucheret H (2004) The nuclear dsRNA

binding protein HYL1 is required for microRNA accumulation and

plant development, but not posttranscriptional transgene silencing.

Curr Biol 14: 346–351

Voinnet O (2009) Origin, biogenesis, and activity of plant microRNAs. Cell

136: 669–687

Wu G, Park MY, Conway SR, Wang JW, Weigel D, Poethig RS (2009) The

sequential action of miR156 and miR172 regulates developmental tim-

ing in Arabidopsis. Cell 138: 750–759

Yamasaki H, Hayashi M, Fukazawa M, Kobayashi Y, Shikanai T (2009)

SQUAMOSA Promoter Binding Protein-Like7 is a central regulator for

copper homeostasis in Arabidopsis. Plant Cell 21: 347–361

Yoshida T, Fujita Y, Sayama H, Kidokoro S, Maruyama K, Mizoi J,

Shinozaki K, Yamaguchi-Shinozaki K (2010) AREB1, AREB2, and

ABF3 are master transcription factors that cooperatively regulate

ABRE-dependent ABA signaling involved in drought stress tolerance

and require ABA for full activation. Plant J 61: 672–685

Zhang H, Liang WQ, Yang XJ, Luo X, Jiang N, Ma H, Zhang DB (2010)

Carbon starved anther encodes a MYB domain protein that regulates

sugar partitioning required for rice pollen development. Plant Cell 22:

672–689

Zhang JF, Yuan LJ, Shao Y, DuW, Yan DW, Lu YT (2008) The disturbance of

small RNA pathways enhanced abscisic acid response and multiple

stress responses in Arabidopsis. Plant Cell Environ 31: 562–574

Zhang X, Zhao H, Gao S, Wang WC, Katiyar-Agarwal S, Huang HD,

Raikhel N, Jin H (2011) Arabidopsis Argonaute 2 regulates innate

immunity via miRNA393()-mediated silencing of a Golgi-localized

SNARE gene, MEMB12. Mol Cell 42: 356–366

Zhang XR, Yuan YR, Pei Y, Lin SS, Tuschl T, Patel DJ, Chua NH (2006)

Cucumber mosaic virus-encoded 2b suppressor inhibits Arabidopsis

Argonaute1 cleavage activity to counter plant defense. Genes Dev 20:

3255–3268

Zhao ZX, Zhang W, Stanley BA, Assmann SM (2008) Functional proteo-

mics of Arabidopsis thaliana guard cells uncovers new stomatal signaling

pathways. Plant Cell 20: 3210–3226

Zhou LG, Liu YH, Liu ZC, Kong DY, DuanM, Luo LJ (2010) Genome-wide

identification and analysis of drought-responsive microRNAs in Oryza

sativa. J Exp Bot 61: 4157–4168

Zhou XF, Wang GD, Sutoh K, Zhu JK, Zhang WX (2008) Identification of

cold-inducible microRNAs in plants by transcriptome analysis. Biochim

Biophys Acta 1779: 780–788

Zhou XF, Wang GD, Zhang WX (2007) UV-B responsive microRNA genes

in Arabidopsis thaliana. Mol Syst Biol 3: 103–113

Li et al.




Transcriptional Regulation of Arabidopsis MIR168a ... · PDF fileARGONAUTE1 Homeostasis in...

Documents

Transcript of Transcriptional Regulation of Arabidopsis MIR168a ... · PDF fileARGONAUTE1 Homeostasis in...