Genomics 94 (2009) 263–268

Contents lists available at ScienceDirect

Genomics

j ourna l homepage: www.e lsev ie r.com/ locate /ygeno

Conserved miRNA analysis in Gossypium hirsutum through small RNA sequencing

Meng-Bin Ruan a, Ying-Tao Zhao b,c, Zhao-Hong Meng a, Xiu-Jie Wang b,⁎, Wei-Cai Yang a,⁎a Key Laboratory of Molecular and Developmental Biology, National Center for Plant Gene Research (Beijing), Institute of Genetics and Developmental Biology, Chinese Academy of Sciences,Beijing 100080, Chinab State Key Laboratory of Plant Genomics, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, 100101, Chinac Graduate School of Chinese Academy of Sciences, 19 Yuquan Road, Beijing, China

⁎ Corresponding authors. Fax: +86 10 62551272.E-mail addresses: xjwang@genetics.ac.cn (X.-J. Wang

(W.-C. Yang).

0888-7543/$ – see front matter © 2009 Elsevier Inc. Adoi:10.1016/j.ygeno.2009.07.002

a b s t r a c t

a r t i c l e i n f o

Article history:Received 9 February 2009Accepted 12 July 2009Available online 21 July 2009

Keywords:Gossypium hirsutummiRNAIdentificationSequencing

Several miRNA family and their targets in cotton had been identified by computational methods based on theconserved characterization of miRNAs. So far, there are no experiments to validate the existence of miRNAs incotton. In this study, to analyze the miRNAs in cotton, a small RNA library of sequences from 18 to 26 nt ofGossypium hirsutum seedling has been built by high-throughput sequencing. In this library, 34 conservedmiRNAfamilies were identified by homology search and the miRNA⁎ sequences of themwere also found in the library.Furthermore, potential targets of these conservedmiRNA families were predicted in cotton TC library. However,based on the mature miRNAs and their miR⁎ sequences, only 8 conserved miRNA encoding loci (miR156,miR157a, miR157b, miR162, miR164, miR393, miR399, miR827) were identified from cotton EST sequences.Multiple encoding loci of some miRNAs were identified by comparing the cloned miRNA and miR⁎ sequences.

© 2009 Elsevier Inc. All rights reserved.

Introduction

Gene expression is regulated by an elaborate network of multiplemechanisms. miRNA, as a class of 20 to 24 nt small regulatory RNAs ofgene expression in eukaryotes [1], negatively regulates gene expres-sion at the posttranscriptional levels through RNA-induced silencingcomplex (RISC) by binding target mRNAs for mRNA cleavage orinhibition ofmRNA translation [2]. miRNAs play crucial roles in variousbiological and metabolic processes in plants and animals [3–11].Identified miRNAs are found with important regulatory functions inspecific biological processes through the whole life cycle of plants[6,10,12–19]. These processes include control of tissue (leaf, root, stem,and flower) differentiation and development, phase switch fromvegetative growth to reproductive growth, signal transduction, andresponse to different biotic and abiotic stresses (eg. salinity, drought,and pathogens), etc.

Some known miRNA loci are found in clusters on the genome. ThemiRNAs in a given cluster are often related to each other. Theseclusters of miRNAs maybe are produced by gene duplication [20–22].Many reported miRNAs are phylogenetically conserved. This con-servation of miRNAs has been used as a powerful strategy for iden-tification or prediction of miRNAs by homology search in other species[23]. Bioinformatically predicted miRNAs should be validated for theirexpression by experimental methods. Northern blotting and PCR-based amplification of adaptor-ligated cDNA had been used for vali-dation of predicted miRNAs [24,25]. Northern blotting might not be

), wcyang@genetics.ac.cn

ll rights reserved.

sensitive enough to detect less-abundant miRNAs and it does notreveal the actual miRNA sequences, and PCR-based amplification canbe difficult in practice when the actual mature miRNA region isunknown. The recent, dramatic improvements in the second genera-tion sequencing technology make it possible to acquire even lowabundant small RNA sequences of a sample at significantly reducedtime and cost compared with traditional approaches [26].

Cotton is an important economic crop. Following many studiesinvolving miRNA computational prediction and identification in Ara-bidopsis, recent works using similar approach have identified manymiRNAs in cotton. Qiu et al. [27] and Zhang et al. [28] reported 21 and22miRNA families fromGSS and EST sequences of cotton by homologysearch. Using the identified plant miRNA precursors as orthologues tosearch all sequences of cotton in GenBank, 13 miRNA families werepredicted [29]. Recently, 682 miRNAs were identified in 155 diverseplant species using all publicly available nucleotide databases [30]. Inthis research, 18 conserved miRNA families were identified in cotton.For studying the potential roles of the small RNAs in developing cottonovules, researchers have cloned and sequenced small RNAs derivedfrom eleven DPA periods (0–10 DPA) of cotton fiber development [31].Only 2 miRNAs were identified in this research. However, the recentmiRNA analysis in Arabidopsis and rice by using deep sequencingapproach discovered that the encoding loci of non-conserved miRNAsare more than that of conserved miRNAs [32,33]. These suggest that,besides conserved miRNAs, many non-conserved miRNA encodingloci may present in the cotton genome.

For fully identification of conserved miRNAs in cotton, it is neces-sary to sequence all expressed small RNAs. In this paper, a total of1,358,811 unique small RNAs have been sequenced in Gossypiumhirsutum seedlings. After miRNA homology analysis, 34 conserved

Fig. 1. Sequence length distribution of cotton small RNA library.

Table 2Distribution of small RNA reads in the sequenced cotton small RNA library.

Small RNA Unique reads Total reads

Global 1,358,811 3,129,095Repeat region 614 10,899rRNA 30,152 374,997tRNA 573 22,310Protein coding 407 687snoRNA 27 30miRNA 892 135,298

Table 3Conserved miRNAs in cotton small RNA library.

miRNA family Readsa Precursorb miRNA⁎sequenced

Predicted target family

miR156/miR157 22,560 Y Y Squamosa-promoter bindingprotein

miR159 178 N Y ATP synthasemiR160 67 N Y Auxin response factormiR162 1 Y Y Zinc finger proteinmiR164 297 Y Y NAC domain protein NAC1miR165 49 N Y Class III HD-Zip proteinmiR166 1715 N Y Class III HD-Zip proteinmiR167 6030 N Y LIM domain proteinmiR168 11,627 N Y AGO1-1miR169 362 N Y UnknownmiR170/miR171 77 N Y GRAS transcription factormiR172 843 N Y AP2 related transcription factormiR319 26 N Y ATP synthasemiR390 19 N N UnknownmiR393 601 Y Y Transport inhibitor response 1miR394 11 N Y UnknownmiR395 13 N Y ATP sulfurylasemiR396 268 N Y UnknownmiR397 115 N Y Laccase, diphenol oxidase,

iron-stress related proteinmiR398 1 Y N Cytoplasmic Cu/Zn SODmiR399 1 Y Y UnknownmiR408 4 N N Basic blue copper proteinmiR414 2 N N Alpha chain of nascent

polypeptide associated complexmiR444a.2 130 N Y Heat stress transcription

factor A-1bmiR444b.1 113 N Y Choline monooxygenaseOsa-miR528 122 N Y Pathogen-related proteinmiR827 42 Y Y Expp1 protein

264 M.-B. Ruan et al. / Genomics 94 (2009) 263–268

miRNA families have been identified from the small RNA library. Basedon the identified miRNA and miRNA⁎ sequences, 8 miRNA (miR156,miR157a, miR157b, miR162, miR164, miR393, miR399, miR827)encoding loci have been identified from the cottonTC library. Diversityanalysis of conserved miRNA and miRNA⁎ sequences suggested thatthe conversed miRNA encoding loci in cotton is far more than that inArabidopsis and rice.

Results and discussion

The small RNA profile

In plant, three small RNA pathways were discovered recently,including siRNA, miRNA and ta-siRNA [34]. A single miRNA can beproduced by the processing of one to several longer precursors whichpossess stem–loop shaped secondary structures. Deep sequencingapproach has been proved as an effective method for analysis of smallRNAs including miRNAs in Arabidopsis and rice [32,33]. The small RNAworld of cotton developing ovule has been uncovered by using theregular small RNA clone. Now, we are beginning to lift the curtain onthe small RNAs especially miRNAs in cotton by high-throughputsequencing method.

A cotton small RNA library was built by high-throughput small RNAsequencing, including 3,129,095 sequences with length from 18 to26 nt. As the Fig. 1 distribution chart showed, almost 2/3 sequencesare redundant sequences. 1,189,529 (~38%) small RNAs have beensequenced only once, 169,282 small RNAs have been sequenced atleast two times. Moreover, 322 small RNAs have been sequencedmorethan 500 times, but only 14 (4.3%) of them were respectivelyidentified as members of conserved miRNA families (miR156/157,miR164, miR166, miR167, miR168, miR169, miR172, miR393, miR396,miR444).

For sequence analysis of the small RNA library, 144 BAC and 21,405TC of cotton were downloaded from GenBank and TIGR. The uniquesequences of the small RNA library were mapped to the sequences ofthese two dataset. As Table 1 showed, 1003 unique sequences weremapped to cotton BACs, and 30,521 unique sequences were located incotton TCs. These unique sequences that can be located in BAC and TCsequences only comprise of 2.32% of total unique sequences identifiedin the library. Also, the fragments of repeat regions, tRNAs, rRNAs,protein coding regions, and snoRNAs were analyzed in the library,distribution of these fragments was listed in Table 2.

Table 1Small RNA mapped to the BAC and TC library of cotton.

Library Mapping loci Unique sRNA Percent of unique sRNAs

BAC (144) 2152 1003 0.07TC (21,405) 51,051 30,521 2.25

Identification of conserved miRNAs

For identification of conserved miRNAs in cotton, unique smallRNA sequences had been aligned with conserved plant miRNAs in themiRBase (release 12.0) with 0–3 bases of mismatch. 892 uniquesequences were identified as homologs of 33 known plant miRNAs. Tobe annotated as an miRNA, sequences representing both miRNA andmiRNA⁎ should be identified for the miRNA canidates [35]. MiRNA⁎sequences of most of these conserved miRNA families have beenidentified from our cloned small RNA library by homology search.Target prediction of the conserved miRNA families was performed bymiRU using nearly perfect sequence complementary as criteria [36].The identified conserved miRNA and their predicted targets werelisted in Table 3. Most conserved miRNAs have different sequences,which were shown in Fig. 2. In this result, 13 conserved miRNAfamilies (miR156/miR157, miR159, miR164, miR165, miR166, miR167,

miR860 1 N N Unknownppt-miR894 1 N Y UnknownmiR845 5 N Y Histone H2BOsa-miR1318 3 N Y RACK1-like proteinOsa-miR1436 1 N Y Triacylglycerol lipase

Targets were predicted by miRU at URL: http://bioinfo3.noble.org/psRNATarget/.a Reads per million in the small library.b Precursor located in cotton EST sequences.

Fig. 2. Small RNA number distribution of conserved miRNAs in cotton.

265M.-B. Ruan et al. / Genomics 94 (2009) 263–268

miR168, miR169, miR172, miR319, miR393, miR399) have more than40 different small RNA sequences. The mature and miRNA⁎ sequencesof 8 miRNAs can be mapped to cotten TCs. The predicted secondstructure of these miRNAs are shown in Fig. 3.

Fig. 3. Hairpin secondary structures of conserved miRNA precursors, mature miRNA positionURL: http://www.bioinfo.rpi.edu/applications/mfold/cgi-bin/rna-form1.cgi. (For interpretaversion of this article.)

Since the G. hirsutum full genome sequence is not available yet, itis unfeasible to determine the precise genomic loci of each miRNAfamily. According to the numbers of different sequences homo-logous to known miRNAs identified in our library, it is likely that

s were highlighted in red. The secondary structures were produced by mfold (V3.2) attion of the references to colour in this figure legend, the reader is referred to the web

Fig. 4.Multiple alignments of miRNA and their miRNA⁎ sequences of miR156/miR157. The sequence Logo was produced byWebLogo at URL: http://weblogo.berkeley.edu/logo.cgi.The miRNA⁎ sequences that have been sequenced for at least 3 times were listed.

266 M.-B. Ruan et al. / Genomics 94 (2009) 263–268

the G. hirsutum genome contains more miRNA encoding loci thanArabidopsis and rice. In the Arabidopsis plant, whose genome hasbeen fully sequenced, 191 miRNA encoding loci have been iden-tified in the Arabidopsis MPSS plus database [33]. And there are 372miRNA encoding loci that have been identified in rice [32]. Com-paring these two libraries suggested that the bigger a genome,the more miRNA encoding loci. Therefore, G. hirsutum, a tetraploid(2n=4x=52) plant, whose genome is bigger than that of Arabi-dopsis and rice, should have more miRNA encoding loci than thatof Arabidopsis and rice. As the MPSS plus database showed, themiR156/miR157 family has 12 encoding loci in both Arabidopsisand rice. We analyzed the sequences that have been identified as

miRNA and miRNA⁎ sequences of miR156/miR157. As shown in Fig.4, many miRNA and miRNA⁎ sequences of miR156/157 have beenidentified in our sequencing results. Although some of these se-quencing variants might be generated as a result of RNA editing orSNP, it is also possible that the G. hirsutum genome contains moremiR156/miR157 loci than that of Arabidopsis and rice.

Bioinformatics prediction of the conservation of the conservedmiRNA families in cotton may not be completely informative at thistime because of the lack of complete genome information and thesearch for these miRNA precursor sequences in TC library has beenunsuccessful (only 8 precursors were identified). Recent deep se-quencing of Arabidopsis and rice small RNAs suggested that the

267M.-B. Ruan et al. / Genomics 94 (2009) 263–268

plant genome encodes more non-conserved miRNA families thanconserved miRNA families [32,33]. Although we have predictedseveral potential precursors of non-conserved miRNA from cottonTC library, these non-conserved miRNAs have been sequenced onlyone or two times in the small library, but the miRNA⁎ sequences ofthese non-conserved miRNAs were failed to be identified in thelibrary.

miRNA and miRNA⁎ sequence diversity analysis

To analyze the diversity of miRNA and miRNA⁎ sequences thathave been identified from our small RNA library, multiple alignmentsof several conserved miRNA families (miR156/miR157, miR167,miR168, miR169, miR393) were performed. As Fig. 4 showed (onlythe 195 sequence that has been read at least 3 times is listed), amongthe sequences identified as miR156/miR157, five nucleotide variablepositions were found, and 9 nucleotide variable positions were foundamong the sequences of miR393 (Fig. 5). Sequence variationwere alsofound in the miRNA⁎ sequences, mostly at positions different fromvariations observed in the mature miRNA sequences.

Some known miRNAs are found in clusters, they are transcribed aspolycistronic primary transcripts and miRNAs in a cluster may be func-tionally related [37]. One evolutionary forces to producemiRNA clustersis gene duplication. The genus Gossypium occurs naturally throughouttropical and subtropical regions, and includes about 45 species splitacross two ploidy levels, diploid (2n=2x=26) and tetraploid (2n=4x=52) [38]. G. hirsutum is a tetraploid (2n=4x=52), so the geneduplications in it might be more complex than other plants. Due to thelack of compelete genome information, most of our sequence reads cannot be mapped to the genomes. Therefore, no miRNA gene cluster wasidentified in this study.

In summary, we have built a small RNA library of G. hirsutumseedling by high-throughput sequencing technology. ConservedmiRNA and miRNA⁎ sequences have been identified from the library.Along with more and more information about cotton genomebecoming available, the miRNA world of cotton will be uncoveredlittle by little. With the availability of cotton BAC and genome

Fig. 5. Multiple alignments of miRNA and their miRNA⁎ sequences of miR393. The seque

sequences in the future, we expect more novel cotton miRNAs beingidentified in the future.

Materials and methods

Plant materials and total RNA extraction

Upland cotton (G. hirsutum) C312 seeds were grown on 1/2 MSmedia and 1/2 MS media containing 200 mM NaCl. Total RNA of6-day seedlings of two treatments was isolated by cold-acidicphenol method with a modified extraction buffer (50 mM Tris–HClpH 6.0, 10 mM EDTA pH 8.0, 100 mM LiCl, 2% SDS, 2% PVP). ThenRNA was precipitated by ethanol, dissolved in DEPC water andstored at −80 °C.

Small RNA isolation and sequencing

The extracted total RNAs were resolved on a denatured 15%polyacrylamide gel. Gel fragments spanning the size range of 18 to26 nt were excised, and small RNAs were eluted overnight with 0.5 MNaCl at 4 °C, precipitated by ethanol. The small RNAs were sequencedby SOLEXA sequencing method (BGI, Beijing).

Small RNA library analysis

144 BAC from GenBank and 21,405 TC from TIGR of cotton weredownloaded. Sequences from 18 to 26 nt in the sequence result weremapped to the cotton BAC and TC library.

Conserved miRNA identification and target prediction

For analysis of conserved miRNAs in cotton, unique small RNAswere aligned with the plant mature miRNAs from miRBase (release12.0) [39]. Cotton small RNA sequenceswith 0–3 basesmismatchwereidentified as known plant miRNAs. miRNA⁎ sequences of conservedmiRNAs were identified by using themiRNA⁎ sequences of Arabidopsisand rice as homology. Conserved miRNAs were mapped to cotton TC

nce Logo was produced by WebLogo at URL: http://weblogo.berkeley.edu/logo.cgi.

268 M.-B. Ruan et al. / Genomics 94 (2009) 263–268

sequences from TIGR. Sequences of 120 bp long around the mappingloci were processed by mfold (V3.2) [40] at the URL: http://www.bioinfo.rpi.edu/applications/mfold/cgi-bin/rna-form1.cgi to findwhether they could be folded into a hairpin secondary structures.Targets of conserved miRNA were predicted by the web tool psRNA-Target [36] at the URL: http://bioinfo3.noble.org/psRNATarget/ usingthe Gossypium (cotton) DFCI Gene index (CGI) release 9 as the se-quence library for target search, set 3 as the maximum expectation.

Conserved miRNA sequence multiple alignments

Conserved miRNA multiple alignments were performed using theWebLogo at URL: http://weblogo.berkeley.edu/logo.cgi.

Acknowledgments

The author gratefully acknowledges the support of the ChineseNational 863 Plans Projects Foundation (grant no. 2006AA10A108),the support of K.C. Wong Education Foundation, Hong Kong, and thesupport of the Knowledge Innovation Program of the ChineseAcademy of Sciences (grant No. KYQY-Y011).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.ygeno.2009.07.002.

References

[1] B.C. Meyers, P.J. Green, C. Lu, miRNAs in the Plant Genome: all things great andsmall, Genome Dyn. 4 (2008) 108–118.

[2] W. Filipowicz, L. Jaskiewicz, F.A. Kolb, R.S. Pillai, Post-transcriptional genesilencing by siRNAs and miRNAs, Curr. Opin. Struct. Biol. 15 (2005) 331–341.

[3] U. Lakshmipathy, B. Love, L.A. Goff, R. Jornsten, R. Graichen, R.P. Hart, J.D. Chesnut,MicroRNA expression pattern of undifferentiated and differentiated humanembryonic stem cells, Stem Cells Dev. 16 (2007) 1003–1016.

[4] M. Raftopoulou, microRNA signals cell fate, Nat. Cell Biol. 8 (2006) 112.[5] I. Bentwich, A postulated role for microRNA in cellular differentiation, FASEB J. 19

(2005) 875–879.[6] S. Schwarz, A.V. Grande, N. Bujdoso, H. Saedler, P. Huijser, The microRNA regulated

SBP-box genes SPL9 and SPL15 control shoot maturation in Arabidopsis, Plant Mol.Biol. 67 (2008) 183–195.

[7] G. Chuck, R. Meeley, E. Irish, H. Sakai, S. Hake, The maize tasselseed4 microRNAcontrols sex determination and meristem cell fate by targeting Tasselseed6/indeterminate spikelet1, Nat. Genet. 39 (2007) 1517–1521.

[8] C. Kutter, H. Schob, M. Stadler, F. Meins Jr., A. Si-Ammour, MicroRNA-mediatedregulation of stomatal development in Arabidopsis, Plant Cell 19 (2007)2417–2429.

[9] T.J. Chiou, K. Aung, S.I. Lin, C.C. Wu, S.F. Chiang, C.L. Su, Regulation of phosphatehomeostasis by MicroRNA in Arabidopsis, Plant Cell 18 (2006) 412–421.

[10] X. Chen, MicroRNA biogenesis and function in plants, FEBS Lett. 579 (2005)5923–5931.

[11] I. Alvarez-Garcia, E.A. Miska, MicroRNA functions in animal development andhuman disease, Development 132 (2005) 4653–4662.

[12] A.C. Mallory, B.J. Reinhart, M.W. Jones-Rhoades, G. Tang, P.D. Zamore, M.K. Barton,D.P. Bartel, MicroRNA control of PHABULOSA in leaf development: importance ofpairing to the microRNA 5′ region, EMBO J. 23 (2004) 3356–3364.

[13] M.T. Juarez, J.S. Kui, J. Thomas, B.A. Heller, M.C. Timmermans, microRNA-mediatedrepression of rolled leaf1 specifies maize leaf polarity, Nature 428 (2004) 84–88.

[14] N.A. McHale, R.E. Koning, MicroRNA-directed cleavage of Nicotiana sylvestrisPHAVOLUTA mRNA regulates the vascular cambium and structure of apicalmeristems, Plant Cell 16 (2004) 1730–1740.

[15] M.C. Timmermans, M.T. Juarez, T.L. Phelps-Durr, A conserved microRNA sig-nal specifies leaf polarity, Cold Spring Harbor Symp. Quant. Biol. 69 (2004)409–417.

[16] K. Guo, K. Xia, Z.M. Yang, Regulation of tomato lateral root development by carbonmonoxide and involvement in auxin and nitric oxide, J. Exp. Bot. 59 (2008)3443–3452.

[17] J.W. Wang, L.J. Wang, Y.B. Mao, W.J. Cai, H.W. Xue, X.Y. Chen, Control of root capformation by MicroRNA-targeted auxin response factors in Arabidopsis, Plant Cell17 (2005) 2204–2216.

[18] H.S. Guo, Q. Xie, J.F. Fei, N.H. Chua, MicroRNA directs mRNA cleavage of thetranscription factor NAC1 to downregulate auxin signals for Arabidopsis lateralroot development, Plant Cell 17 (2005) 1376–1386.

[19] H.H. Liu, X. Tian, Y.J. Li, C.A. Wu, C.C. Zheng, Microarray-based analysis of stress-regulated microRNAs in Arabidopsis thaliana, RNA 14 (2008) 836–843.

[20] X. Guo, Y. Gui, Y. Wang, Q.H. Zhu, C. Helliwell, L. Fan, Selection and mutation onmicroRNA target sequences during rice evolution, BMC Genomics 9 (2008) 454.

[21] A. Li, L. Mao, Evolution of plant microRNA gene families, Cell Res. 17 (2007)212–218.

[22] C. Maher, L. Stein, D. Ware, Evolution of Arabidopsis microRNA families throughduplication events, Genome Res. 16 (2006) 510–519.

[23] T. Dezulian, M. Remmert, J.F. Palatnik, D.Weigel, D.H. Huson, Identification of plantmicroRNA homologs, Bioinformatics 22 (2006) 359–360.

[24] A. Valoczi, C. Hornyik, N. Varga, J. Burgyan, S. Kauppinen, Z. Havelda, Sensitive andspecific detection of microRNAs by northern blot analysis using LNA-modifiedoligonucleotide probes, Nucleic Acids Res. 32 (2004) e175.

[25] V. Ambros, R.C. Lee, Identification of microRNAs and other tiny noncoding RNAs bycDNA cloning, Methods Mol. Biol. 265 (2004) 131–158.

[26] E.A. Glazov, P.A. Cottee, W.C. Barris, R.J. Moore, B.P. Dalrymple, M.L. Tizard, AmicroRNA catalog of the developing chicken embryo identified by a deepsequencing approach, Genome Res. 18 (2008) 957–964.

[27] C.X. Qiu, F.L. Xie, Y.Y. Zhu, K. Guo, S.Q. Huang, L. Nie, Z.M. Yang, Computationalidentification of microRNAs and their targets in Gossypium hirsutum expressedsequence tags, Gene 395 (2007) 49–61.

[28] B. Zhang, Q. Wang, K. Wang, X. Pan, F. Liu, T. Guo, G.P. Cobb, T.A. Anderson,Identification of cotton microRNAs and their targets, Gene 397 (2007) 26–37.

[29] M.Y. Khan Barozai, M. Irfan, R. Yousaf, I. Ali, U. Qaisar, A. Maqbool, M. Zahoor, B.Rashid, T. Hussnain, S. Riazuddin, Identification of micro-RNAs in cotton, PlantPhysiol. Biochem. 46 (2008) 739–751.

[30] R. Sunkar, G. Jagadeeswaran, In silico identification of conserved microRNAs inlarge number of diverse plant species, BMC Plant Biol. 8 (2008) 37.

[31] I.Y. Abdurakhmonov, E.J. Devor, Z.T. Buriev, L. Huang, A. Makamov, S.E.Shermatov, T. Bozorov, F.N. Kushanov, G.T. Mavlonov, A. Abdukarimov, SmallRNA regulation of ovule development in the cotton plant, G. hirsutum L, BMCPlant Biol. 8 (2008) 93.

[32] K. Nobuta, R.C. Venu, C. Lu, A. Belo, K. Vemaraju, K. Kulkarni, W.Wang, M. Pillay, P.J.Green, G.L. Wang, B.C. Meyers, An expression atlas of rice mRNAs and small RNAs,Nat. Biotechnol. 25 (2007) 473–477.

[33] M. Nakano, K. Nobuta, K. Vemaraju, S.S. Tej, J.W. Skogen, B.C. Meyers, Plant MPSSdatabases: signature-based transcriptional resources for analyses of mRNA andsmall RNA, Nucleic Acids Res. 34 (2006) D731–D735.

[34] E. Bonnet, Y. Van de Peer, P. Rouze, The small RNAworld of plants, New Phytol. 171(2006) 451–468.

[35] B.C. Meyers, M.J. Axtell, B. Bartel, D.P. Bartel, D. Baulcombe, J.L. Bowman, X. Cao, J.C.Carrington, X. Chen, P.J. Green, S. Griffiths-Jones, S.E. Jacobsen, A.C. Mallory, R.A.Martienssen, R.S. Poethig, Y. Qi, H. Vaucheret, O. Voinnet, Y. Watanabe, D. Weigel,J.K. Zhu, Criteria for annotation of plant MicroRNAs, Plant Cell 20 (2008)3186–3190.

[36] Y. Zhang, miRU: an automated plant miRNA target prediction server, Nucleic AcidsRes. 33 (2005) W701–704.

[37] V.N. Kim, J.W. Nam, Genomics of microRNA, Trends Genet. 22 (2006) 165–173.[38] J.J. Lee, A.W. Woodward, Z.J. Chen, Gene expression changes and early events in

cotton fibre development, Ann. Bot. (Lond.) 100 (2007) 1391–1401.[39] S. Griffiths-Jones, H.K. Saini, S. van Dongen, A.J. Enright, miRBase: tools for

microRNA genomics, Nucleic Acids Res. 36 (2008) D154–D158.[40] M. Zuker, Mfold web server for nucleic acid folding and hybridization prediction,

Nucleic Acids Res. 31 (2003) 3406–3415.