ISSN 1672-9145 Acta Biochimica et Biophysica Sinica 2007, 39(11): 829–834 CN 31-1940/Q

Cloning of Novel Repeat-associated Small RNAs Derived from

Hairpin Precursors in Oryza sativa

Chengguo YAO, Botao ZHAO, Wei LI, Yang LI, Wenming QIN, Bing HUANG, and Youxin JIN*
State Key Laboratory of Molecular Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences,
Chinese Academy of Sciences; Graduate School of Chinese Academy of Sciences, Shanghai 200031, China

Downloaded from http://abbs.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on August 27, 2014

Abstract Plant small non-coding RNAs including microRNAs (miRNAs), small interfering RNAs
(siRNAs) and trans-acting siRNAs, play important roles in modulating gene expression in cells. Here we
isolated 21 novel endogenous small RNA molecules, ranging from 18 to 24 nucleotides, in Oryza sativa that
can be mapped to 111 hairpin precursors. Further analysis indicated that most of these hairpin sequences
originated from putative miniature inverted-repeat transposable elements, a major type of DNA transposon.
Considering that miRNA is characteristic of hairpin-like precursor and plant endogenous siRNAs are often
located at transposon regions, we hypothesized that our cloned small RNAs might represent the intermediate
product in the evolutionary process between siRNAs and miRNAs. Northern blot analysis indicated that five
of them were much more abundantly expressed in flower compared to other tissues, implying their potential
function in inflorescence. In conclusion, our results enrich rice small RNA data and provide a meaningful
perspective for small RNA annotation in plants.

Keywords small RNA; microRNA; small interfering RNA; hairpin; Oryza sativa

MicroRNAs (miRNAs), small interfering RNAs (siRNAs),
and trans-acting siRNAs have emerged in recent years as
small but mighty gene attenuators in plants [1]. As small
non-coding RNAs, they are almost indistinguishable in
chemical composition and length, and their classification
mainly depends on their biogenesis and mode of action
[2]. miRNA is processed to a ~22 nucleotide (nt) mature
sequence from an incompletely base-paired hairpin-like RNA
transcript and then acts in trans on target messenger RNA
[3]. siRNAs are primarily processed from long double-
stranded RNAs, but trans-acting siRNAs act in trans just
like miRNAs, whereas other siRNAs mainly target the
transcripts that originate from siRNAs themselves [4].
Given their relatively detailed characterization, a number
of open questions remain concerning the differentiation of
these small RNAs. For example, siRNAs are well known
to be related to transposable elements (TEs), as opposed
Received: April 4, 2007 Accepted: July 1, 2007
This work was supported by the National Natural Science Foundation
of China (No. 3 0430 210 ), the National Key Basic Research a nd
Development Program (No. 2005CB724602) and Chinese Academy of
Science (KSCXI-YW-R-64)
*Corresponding a u thor : Tel, 8 6 -2 1 -5 4 9 2 1 2 2 2 ; Fa x, 8 6 -2 1 -
54921011; E-mail, yxjin@sibs.ac.cn
to them, miRNAs are thought to derive from loci different
from other genes or TEs [5]. However, several examples
of miRNA genes have been identified to be located at TEs
in animals [6], and recent bioinformatics analysis shows
that 55 experimentally characterized human miRNA genes
are derived from TEs, and these TE-derived miRNAs have
the potential to regulate thousands of genes [7]. Miniature
inverted-repeat transposable elements (MITEs) are notable
types of TE associated with miRNA, as they have
palindromic structures with terminal inverted repeats
(TIRs) that flank short internal regions. Their expression
as RNA results in the formation of the kinds of hairpins
seen for pre-miRNAs [8]. Indeed, MITEs have previously
been shown to contribute miRNA genes in human and
Arabidopsis genomes [8,9].
The rice genome contains approximately 90,000 MITEs,
constituting approximately 26% of the genome sequence
and highest-copy-number TEs in rice [10]. Previous
sequence analyses of cloned rice small RNAs ignored the
correlation between hairpin-derived small RNA and TEs
[11−13]. We constructed several rice small RNA libraries
DOI: 10.1111/j.1745-7270.2007.00346.x

©Institute of Biochemistry and Cell Biology, SIBS, CAS

830 Acta Biochim Biophys Sin
from typical organs at different development stages, and
were able to obtain 3003 sequences. Subsequent analysis
allowed us to pick out 21 novel small RNA sequences
mapped to 111 hairpin precursors that are encoded by
putative tra nsposons. These sma ll RNAs sha re
characteristics with both miRNA and siRNA, and might
represent the evolutionary link between both.
Materials and Methods
Cloning of small RNA sequences from rice
Whole rice plants (Oryza sativa L. ssp. japonica) were
grown under natural conditions. Different plant tissues at
various development stages (root, leaf, flower, and stem)
were collected, washed with double distilled H2O, and
frozen in liquid nitrogen. Total RNAs from O. sativa were
prepared using the Trizol method (Invitrogen, Carlsbad,
USA) in which the isopropanol precipitation was replaced
by ethanol precipitation. In brief, small RNAs from 18 to
28 nt were size-fractionated, purified, and ligated sequen-
tially to the 5' DNA adapter (ACCGAATTCACAGTCA-
GACC, EcoRI site) and 3' adapter (GCAGATCGTCAGA-
ATTCCAG, EcoRI site) with T4 RNA ligase from New
England Biolabs (Beverly, USA). The 3' adapter was
blocked by ddA with terminal transferase (NEB) at its 3'
terminus and phosphorylated by T4 polynucleotide kinase
(NEB) at the 5' terminus. The ligated RNA was reverse
transcribed into cDNA by the Access Quick reverse
transcription-polymerase chain reaction (RT-PCR) system
(Promega, Madison, USA) with the 5' adapter sequence
and another primer complementary to the 3' adapter. The
RT-PCR product was amplified by PCR with the same
primers. After purification, the PCR product was digested
with EcoRI (NEB) and concatemerized with T4 DNA ligase
(NEB) followed by purification with the kit (Bioasia,
Shanghai, China). PCR was carried out with the previous
primers and the DNA of approximately 800 bp was cloned
into pGEM-T vector (Promega) for sequencing.
To avoid losing the cDNAs containing the EcoRI site,
the adaptors with a SalI site but not an EcoRI site were
also used to generate some cDNA libraries.
Sequence analysis and prediction of fold-back structures
Using BLASTN on the National Center for Biotechnology
Information (NCBI) website (http:www.ncbi.nlm.nih.gov/
blast), cloned sequences shorter than 16 nt and possible
ribosomal RNA, messenger RNA, transfer RNA, and small
nucleolar RNA fragments were discarded. This step allowed
©Institute of Biochemistry and Cell Biology, SIBS, CAS
Vol. 39, No. 11
the removal of most non-siRNA and non-miRNA species
and identification of cloned known miRNAs that could be
confirmed in the miRNA registry (http://microrna.sanger.
ac.uk/sequences/). After redundancy analysis and exclusion
of previously reported rice small RNAs that were also cloned
through construction of libraries, the remaining sequences
were compared with the latest version of the rice genome
at the Rice Genome Database (RGP; http://riceblast.dna.
affrc.go.jp/). The surrounding sequences of 200 nt in both
orientations were extracted and, together with the perfectly
matched sequences in the RGP, were submitted for RNA
folding using the Mfold program [14]. Folding results were
Downloaded from http://abbs.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on August 27, 2014
inspected, and fold-back structures with small RNA in the
stem were considered hairpin precursors.
Next, we ran BLAST search against the TIGR rice repeat
database (http://tigrblast.tigr.org/euk-blast/index.cgi?
project=osa1) with the hairpin precursors. Sequences with
at least 80% similarity, with one hit including the mature
region, were regarded as candidates for rice repeat genome
data [7], and some of them were perfect matches. Genomic
annotation was examined using the Rice Genome Automated
Annotation System (http://ricegaas.dna.affrc.go.jp/).
Northern blot analysis of small RNAs
Approximately 10 µg total RNA was isolated from
different rice samples (embryo, seedling, leaf, stem, flower,
and fruit), loaded and resolved on the denaturing 15%
polyacrylamide gel, and transferred to a Hybond N+ nylon
membrane (Amersham, Piscataway, USA). The membrane
was ultraviolet cross-linked. DNA probes complementing
miRNA sequences were end-labeled with [γ-32P]ATP (3000
Ci/mM; Amersham) using T4 polynucleotide kinase (MBI,
Vilnius, Lithuania). Unincorporated [γ- 32P]ATP was
removed using a Sephadex G-25 column (Pharmacia,
Uppsala, Sweden). Methylene blue staining of membranes
prior to hybridization was used to detect ribosomal RNA.
The membrane was pre-hybridized and hybridized using
ExpressHyb Hybridization Solution (Clontech, Palo Alto,
USA) then washed according to the user manual. The
membrane was briefly air-dried then exposed to a
PhosphorImager (Amersham).
Results
Novel small RNA gene families identified in rice show
features of both miRNA and siRNA
Construction of a small RNA library is a practical and
effective way to study small RNAs. In fact, many rice
Nov., 2007 Chengguo YAO et al.: Cloning of Novel Repeat-associated Small RNAs Derived from Hairpin Precursors in Oryza sativa 831

miRNAs, endogenous siRNAs and other uncharacterized
tiny non-coding RNAs were identified through this start-
ing point. We constructed six small RNA libraries ranging
from embryo to flower in rice plant (O. sativa L. ssp.
japonica). Small RNAs of 18−26 nt in length were isolated
by size fractionation and ligated to 5' and 3' adapters, then
cloned and sequenced. A total of ~460 clones comprising
3003 small cDNA insertion sequences were collected. After
removal of sequences shorter than 16 nt, we blasted the
remained sequences against the rice genome in the NCBI
database to discard the query sequences bearing no hits,
and those that could be degradation products of ribosomal
eleven loci corresponding to 21 small RNAs were predicted
to hold hairpin-like fold-backs (Table 1), characteristic of
miRNAs. However, sequences BLASTed against the rice
repeat database indicated that they all showed high sequence
similarity with one or more hits. Most of them are putative
MITEs, and others are putative unclassified transposons
or putative retrotransposons (data not shown), which are
obviously features of siRNA-generating sequences.
Genomic annotation of new rice small RNAs
Non-coding small RNAs are usually located in the
genomic segments that are distinct from protein-coding

Downloaded from http://abbs.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on August 27, 2014

RNA, transfer RNA, small nucleolar RNA, or small nuclear
RNA that constituted approximately half of the overall
sequences. The BLAST results in the NCBI database and
miRNA registry (http://www.sanger.ac.uk/Software/Rfam/
mirna/search.shtml) also showed that we had cloned many
known miRNAs and endogenous siRNAs. A total of 1416
small RNA sequences were matched to the rice genome
(japonica genome database at RGP). One hundred and
regions. Examination results of the corresponding loci with
the Rice Genome Automated Annotation System (http://
ricegaas.dna.affrc.go.jp/) conformed to the traditional
knowledge, in that almost all of the loci were in the
intergenic or intron, non-coding expressed sequence tag
regions, except for a few in the exon of a hypothetical
protein and cDNA region, which could be pseudo genes.
The proteins specified by the intron region are nearly all

Table 1 New rice small RNAs derived from hairpin precursors identified by cloning
ID Sequence (5'→3') Annotation Length No. of hairpin Arm a
precursors

I4-136 CAGCUCCACAACUCUACUCCA EST region 21 2 5

I4-165 UGCACCAUCUCAUGCCACUAC Intergenic 21 1 5
I4-304 CUGGGAGUGGAGUUGGGUGGAG EST region 22 2 5, 3
II1-077 UUUGACCGGAUGUCGGAAGGGG Intergenic 22 1 3
II1-130 AAACGUUCACACGGAAACCUAGGG Intergenic 24 24 3
II1-196 AAAAUUCAAAUAAGACGGACGGU Intron of putative peroxidase 23 1 3
II1-200 GCACUGUUUGACCACUCGUCUU EST, intergenic 22 30 5
II2-187 AAUCCAUCAUUAGCACAUGUGGGU EST, intergenic, exon of hypothetical 24 25 5
protein
II2-316 UUCACCCAAUAAUCUGGAAAAA Intergenic 22 1 5
II2-419 AAUCGAGUCUAUUGCUGAGAUGU EST, intergenic 23 3 5, 3
II2-492 ACAAACCGGGACUACAAAGGGU Intron of putative reverse transcriptase 22 1 5
IV4-8 GGAGCGGUCCAUUAGCGCGUGAUU Intergenic 24 2 3
IV4-25 GGCCCUCACGUGUACACACCGUGU Intron of hypothetical protein 24 1 3
IV4-123 UAUGGGAAACGCUAGAAUG EST 19 1 3
IV4-149 ACACUCAAGGCUGUUGUUGG Unknown 20 1 3
IV4-235 ACCGCGAGACAAAUCUUUUGAGCC Intergenic, intron of unknown protein 24 17 5
IV4-298 AAGUUGUGUGUAUGAUAGGUUGAU Intron of unknown protein 24 1 5
IV4-299 AAGGACUGAAGUUUGGAU EST, intergenic 18 7 5, 3
IV4-351 GACGGAUGGUCAAAACGCUGAGCA Intergenic 24 1 3
V3-27 GGUAGCGUGCUGCCGCCUUCCCGC Intergenic, exon of putative protein 24 4 5
VI2-16 AGAUGACGUGACUGUUUCACC Intergenic 21 8 5
a
Location of small RNAs in certain fold-back hairpins. 3, located in the arm close to the 3' terminus; 5, located in the arm close to the 5' terminus; EST, expressed
sequence tag.

http://www.abbs.info; www.blackwellpublishing.com/abbs
832 Acta Biochim Biophys Sin Vol. 39, No. 11

hypothetical or putative, indicating that they might also be
pseudo genes. For those small RNAs encoded by multiple
loci, we could not assign their exact origin unambiguously,
and some of them might not be genuine genes, as referred
to above.
Cloning of putative influorescence-associated small
RNAs
To validate the cloning results, we carried out Northern
blot analysis to examine the existence of these small RNAs
using total RNA samples isolated from diverse organs from
different development stages. Interestingly, five of them
trans-acting siRNAs are being proffered [15−17], the
differences between them seem to be much clearer.
However, some aspects still remain controversial, such as
their conservation, origin, and length [18], which convinced
us of the view that the diversity of plant small RNAs is
more complex than previously expected. Recent cloning
and analysis of rice endogenous small non-coding RNAs
corroborate this point [13]. To date, most reports have
emphasized their distinction and classification, but ignored
the correlation and possible evolutionary intermediates. Here
we present data to provide a link between plant siRNAs
and miRNAs. A full-length DNA-type transposable element

Downloaded from http://abbs.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on August 27, 2014

were exclusively abundant in flower compared to other
tissues (Fig. 1), suggesting that they might specially
function in the process of inflorescence. However, their
targets and the mechanisms through which they play the
role are yet to be discovered. As a control, miR159c was
also examined, and showed ubiquitous expression in the
samples listed in our experimental conditions. Some small
RNAs were not detected by Northern blot analysis with
our available rice tissues. They might be expressed at very
low levels or confined to some specific tissues or cell types,
therefore the amount of these miRNAs was inappreciable
in the limited total RNA samples.
Discussion
As more and more detailed information about plant small
RNAs, including miRNAs, endogenous siRNAs, and
contains an open reading frame flanking two TIRs. When
it becomes a non-autonomous MITE, it is deprived of an
open reading frame, and the hairpin structure would be
formed by base-pair interaction of MITE TIRs. Previously,
an inverted duplication model for miRNA gene evolution
in plants was proposed [19], however, the role of MITEs
was not discussed in that the researchers used known
miRNA fold-back sequences. There is a postulation that
miRNAs could have evolved from TE-encoded siRNAs in
the way that the TIRs that are processed from longer RNAs
to form siRNAs could be similarly processed to form
miRNAs [8]. Thus we assumed that small RNAs that
originated from MITEs were matured by miRNA biogenesis
pathways (Fig. 2). However, additional analysis of these
small RNAs in rice mutants relating to the biogenesis of
miRNAs, such as DCL1, is necessary to validate this
hypothesis [20,21]. Predictably, the evolutionary
relationship and distinction between plant miRNA and

Fig. 1 Expression pattern analysis of some newly cloned rice small RNAs using Northern blot
Total RNAs from Oryza sativa L. ssp. japonica fruit of autumn stage (lane 1), flower of inflorescence (lane 2), leaf of tillering stage (lane 3), stem of jointing stage (lane
4), seedling (lane 5), and embryo (lane 6) were analyzed on the denaturing 15% polyacrilamide gel. Known miR159c was used as the positive control; 5S ribosomal RNA
(Rrna) as the loading control. Positions are indicated with size markers. nt, nucleotide.

©Institute of Biochemistry and Cell Biology, SIBS, CAS

Nov., 2007 Chengguo YAO et al.: Cloning of Novel Repeat-associated Small RNAs Derived from Hairpin Precursors in Oryza sativa 833

Downloaded from http://abbs.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on August 27, 2014

Fig. 2 Schematic representation of the possible biogenesis pathway of rice repeat-associated small RNAs derived from hairpin
precursors
(A) Typical fold-back structure predicted with the Mfold program. Green regions indicate the mature small RNA sequences; red rectangles include the genome repeats.
MITE-adh type B, type D are subtypes defined by MITE (miniature inverted-repeat transposable element) structural features. (B) Assumption model of microRNA
(miRNA) biogenesis pathways for small RNAs that are originated from MITEs. Red sections indicate mature small RNA region. TIR, terminal inverted repeat.

siRNA will become clearer as the biogenesis knowledge
about these MITE-derived small RNAs accumulates.
The average copy number of rice MITEs ranges from
dozens to thousands. Here we only extracted hairpin
precursors perfectly matched to the cloned sequences,
thus many homologs might have been missed. In view of
our limited number of small RNA libraries, predicting the
average copy number of the MITE-derived small RNAs is
not practical. However, taking account of the variability
of plant fold-back precursors and 90,000 MITEs in rice,
it is a really interesting area of study, and our cloned small
RNA sequences from MITEs just show the tip of the
iceberg.
Although plant small RNA target prediction is conve-
nient for the extensive complementarity between small RNA
and its target according to the accepted principles [22,
23], the authentic targets of the vast majority of plant small
RNAs are not characterized, and the phenotypic conse-
quences of disrupted or altered small RNA regulation are
also obscure because projects often take place over long
time periods, and chance events take place, especially in
the study of rice [11,24−26]. Northern blot data revealed
that we have cloned five flower-exclusive small RNAs
among the tissues analyzed, hinting that they might act
like miRNAs to get involved in inflorescence. The molecular
scenarios are waiting to be unraveled.
References
1 Johnson C, Bowman L, Adai AT, Vance V, Sundaresan V. CSRDB: A small
RNA integrated database and browser resource for cereals. Nucleic Acids Res
2007, 35: D829−D833
2 Vazquez F. Arabidopsis endogenous small RNAs: Highways and byways.
Trends Plant Sci 2006, 11: 460−468
3 Chen X. MicroRNA biogenesis and function in plants. FEBS Lett 2005, 579:
5923−5931
4 Yoshikawa M, Peragine A, Park MY, Poethig RS. A pathway for the biogen-
esis of trans-acting siRNAs in Arabidopsis. Genes Dev 2005, 19: 2164−2175
5 Bartel DP. MicroRNAs: Genomics, biogenesis, mechanism, and function.
Cell 2004, 116: 281−297
6 Smalheiser NR, Torvik VI. Mammalian microRNAs derived from genomic
repeats. Trends Genet 2005, 21: 322−326
7 Piriyapongsa J, Marino-Ramirez L, Jordan IK. Origin and evolution of hu-
man microRNAs from transposable elements. Genetics 2007, 176: 1323−
1337
8 Piriyapongsa J, Jordan IK. A family of human microRNA genes from minia-
ture inverted-repeat transposable elements. PLoS ONE 2007, 2: e203
9 Mette MF, van der Winden J, Matzke M, Matzke AJ. Short RNAs can
identify new candidate transposable element families in Arabidopsis. Plant
Physiol 2002, 130: 6−9
10 Jiang N, Feschotte C, Zhang X, Wessler SR. Using rice to understand the
origin and amplification of miniature inverted repeat transposable elements
(MITEs). Curr Opin Plant Biol 2004, 7: 115−119

http://www.abbs.info; www.blackwellpublishing.com/abbs
834 Acta Biochim Biophys Sin
11 Sunkar R, Girke T, Jain PK, Zhu JK. Cloning and characterization of
microRNAs from rice. Plant Cell 2005, 17: 1397−1411
12 Sunkar R, Girke T, Zhu JK. Identification and characterization of endogenous
small interfering RNAs from rice. Nucleic Acids Res 2005, 33: 4443−4454
13 Chen Z, Zhang J, Kong J, Li S, Fu Y, Li S, Zhang H et al. Diversity of
endogenous small non-coding RNAs in Oryza sativa. Genetica 2006, 128:
21−31
14 Zuker M. Mfold web server for nucleic acid folding and hybridization prediction.
Nucleic Acids Res 2003, 31: 3406−3415
15 Allen E, Xie Z, Gustafson AM, Carrington JC. MicroRNA-directed phasing
during trans-acting siRNA biogenesis in plants. Cell 2005, 121: 207−221
16 Nakano M, Nobuta K, Vemaraju K, Tej SS, Skogen JW, Meyers BC. Plant
MPSS databases: Signature-based transcriptional resources for analyses of
mRNA and small RNA. Nucleic Acids Res 2006, 34: D731−D735
17 Johnson C, Sundaresan V. Regulatory small RNAs in plants. EXS 2007, 97:
99−113
18 Carrington JC. Small RNAs and Arabidopsis. A fast forward look. Plant
Physiol 2005, 138: 565−566
19 Allen E, Xie Z, Gustafson AM, Sung GH, Spatafora JW, Carrington JC.
Evolution of microRNA genes by inverted duplication of target gene sequences
©Institute of Biochemistry and Cell Biology, SIBS, CAS
Vol. 39, No. 11
in Arabidopsis thaliana. Nat Genet 2004, 36: 1282−1290
20 Liu B, Li P, Li X, Liu C, Cao S, Chu C, Cao X. Loss of function of OsDCL1
affects microRNA accumulation and causes developmental defects in rice.
Plant Physiol 2005, 139: 296−305
21 Axtell MJ, Jan C, Rajagopalan R, Bartel DP. A two-hit trigger for siRNA
biogenesis in plants. Cell 2006, 127: 565−577
22 Zhang Y. miRU: An automated plant miRNA target prediction server. Nucleic
Acids Res 2005, 33: W701−W704
23 Li Y, LI W, Jin YX. Computational identification of novel family members
of microRNA genes in Arabidopsis thaliana and Oryza sativa. Acta Biochim
Biophys Sin 2005, 37: 75−87
24 Luo YC, Zhou H, Li Y, Chen JY, Yang JH, Chen YQ, Qu LH. Rice
embryogenic calli express a unique set of microRNAs, suggesting regulatory
roles of microRNAs in plant post-embryogenic development. FEBS Lett
Downloaded from http://abbs.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on August 27, 2014
2006, 580: 5111−5116
25 Reinhart BJ, Weinstein EG, Rhoades MW, Bartel B, Bartel DP. MicroRNAs
in plants. Genes Dev 2002, 16: 1616−1626
26 Jones-Rhoades MW, Bartel DP. Computational identification of plant
microRNAs and their targets, including a stress-induced miRNA. Mol Cell
2004, 14: 787−799
Edited by
Blake C. MEYERS