Download Characterization of two rice DNA methyltransferases

Original Article
Characterization of two rice DNA methyltransferases
and RNAi-mediated restoration of promoter activity in
silenced rice callus
Prapapan Teerawanichpan†*, Mahesh B. Chandrasekharan†, Yiming Jiang†,
Jarunya Narangajavana* and Timothy C. Hall††
† Institute
of Developmental and Molecular Biology, and Department of Biology, Texas A&M
University, College Station, TX 77843-3155, U.S.A.
* Department of Biotechnology, Faculty of Science, Mahidol University, Rama 6 Road, Bangkok,
10400, Thailand
∗Correspondence author:
E-mail: [email protected]
Fax: +979-8624098
The original publication is available at http://www.springerlink.com
Planta An International Journal of Plant Biology © Springer-Verlag 2003
DO I: 10.1007/s00425-0 03-1112 -6
http://www.springerlink.com/app/home/content.asp?wasp=b0cywllvrq5vb6ykpee7&referrer=contribution&f
orm
at=13&page=1&pagecount=0
Fo r pe rsonal use only
Abstract Two genomic clones (OsMET1-1, AF 462029 and OsMET1-2), each encoding a
cytosine-5 DNA methyltransferase (MTase), were isolated from rice (Oryza sativa L.) BAC
libraries. OsMET1-1 has an open reading frame of 4,566 nucleotides with twelve exons and
eleven introns while OsMET1-2 has an open reading frame of 4,452 nucleotides with eleven
exons and ten introns. Although OsMET1-1 and OsMET1-2 have high sequence similarity
overall, they share only 24% identity in exon 1 and intron 3 of OsMET1-1 is absent from
OsMET1-2. As for other eukaryotic DNA MTases of the Dnmt1/MET l class, the derived amino
acid sequences of OsMET1-1 and OsMET1-2 suggest that they are comprised of two-thirds
regulatory domain and one-third of catalytic domain. Most functional domains identified for
other MTases were present in the rice MET1 sequences. Amino acid sequence comparison
indicated high similarity (56-75% identity) of rice MET1s to other plant MET1 sequences but
limited similarity (~24% identity) to animal Dnmt1s. Genomic blot and database analysis
indicated the presence of a single copy of OsMET1-1 (on chromosome 3) and single copy of
OsMET1-2 (on chromosome 7). Ribonuclease protection assays revealed expression of both
OsMET1-1 and OsMET1-2 in highly dividing cells, but the steady state level of OsMET1-2 was
7-12 fold higher than that for OsMET1-1 in callus, root and inflorescence. The functional
involvement of the rice DNA MTases in gene silencing was investigated using an RNAi strategy.
Inverted repeat constructs of either the N- or C-terminal regions of OsMET1-1 were
supertransformed into calli derived from a rice line bearing a silenced 35S-uidA-nos transgene.
Restoration of uidA expression in the bombarded calli was consistent with the inactivation of
maintenance methylation and with previous evidence for the involvement of methylation in
silencing of this line.
Keywords methylation • methyltransferase • reactivation • rice • silencing
Abbreviations 35S: cauliflower mosaic virus 35S promoter • GFP: green fluorescent protein •
GUS: ß -glucuronidase • uidA: ß-glucuronidase gene from Escherichia coli
2
Introduction
Cytosine methylation is an epigenetic process that contributes to tissue- and developmental
stage-specific regulation of gene expression (Chaudhury et al. 2001) , genome defense
mechanisms (Matzke et al. 2002;Vaucheret and Fagard 2001) , gene silencing (Kumpatla et al.
1997) , genomic imprinting (Baroux et al. 2002) , chromosome X inactivation (Rakyan et al.
2001) , and sex determination (Siroky et al. 1998) . Cytosine-5 DNA methyltransferases (DNA
MTases), enzymes capable of transferring a methyl group from S-adenosyl methionine to
cytosine residues of the double helix, are found in both prokaryotes and eukaryotes. DNA
methylation in animals is generally confined to cytosines in CpG dinucleotides but, in plants,
methylation has been observed in both CpG dinucleotides and CpNpG triplets (Pradhan et al.
1995) .
The simplest model of transcriptional repression by methylation is that a methyl group on
the DNA duplex can impede the binding of the basal transcriptional machinery or of
transcription factors that require contact with the major groove of the double helix. Alternatively,
cytosine methylation can cause structural changes of chromatin that limit promoter accessibility
by transcriptional machinery. The discovery of several methyl binding domain (MBD)containing proteins (such as MeCP1, MeCP2, MBD1, MBD2, MBD3 and MBD4), that interact
with DNA MTases and are capable of recruiting repressive complexes and histone deacetylases,
suggests the existence of additional mechanisms for gene suppression that may act through
histone deacetylation (Brackertz et al. 2002;Curradi et al. 2002;Feng et al. 2002) . This, together
with the finding that MBD proteins are associated with distinct histone deacetylase (HDAC)
complexes suggests that various MBD/HDAC interactions have different roles in gene silencing
and may act at different stages of development (Hendrich and Bird 1998;Jiang et al. 2002) .
However, the finding in Arabidopsis that MOM1 can modulate silencing in the absence of
changes in CpG methylation status (Scheid et al. 2002) , suggests that some forms of gene
silencing are either methylation-independent or act downstream of cytosine methylation
(Amedeo et al. 2000) .
3
Eukaryotic DNA MTases can be divided into 4 classes based on their structure and
function. The Dnmt1/ MET1 class has maintenance methylation activity in vivo (Finnegan et al.
1996;Li et al. 1992;Ronemus et al. 1996) . Dnmt2 MTases contain only a methyltransferase
domain and lack significant activity both in vivo and in vitro (Okano et al. 1998) . Whereas the
Dnmt3 class contains putative de novo DNA MTases that are found in animals and plants (Cao
and Jacobsen 2002;Cao et al. 2000;Okano et al. 1999;Robertson et al. 2000) , chromomethylases
(CMTs) have thus far only been found in plants and are implicated in CpNpG methylation
(Bartee et al. 2001;Henikoff and Comai 1998;Papa et al. 2001) and, possibly, in heterochromatin
methylation.
MET1 MTases include a long N-terminal domain (that does not exist in prokaryotic DNA
MTases) and a shorter C-terminal domain that contains some or all of ten motifs (I to X) that are
well conserved among prokaryotes and eukaryotes (Lauster et al. 1989;Posfai et al. 1989;Sankpal
and Rao 2002) . The presence of several functional regions in the N-terminal domain of
eukaryotic DNA MTases implies the involvement of additional proteins and a connection
between DNA methylation and cellular mechanisms such as genome defense and DNA
replication (Bahler and Rhoads 2002;Callebaut et al. 1999) .
MET1 clones have been reported from several plant sources, including three from
Arabidopsis (Finnegan and Dennis 1993;Genger et al. 1999) , two from carrot (Bernacchia et al.
1998) , one from tobacco (Nakano et al. 2000) , one from pea (Pradhan et al. 1998) and one from
maize (Steward et al. 2000) . We are not aware of previous work in which rice DNA MTase has
been characterized, although a report exists of MTase purification from cultured rice cells
(Giordano et al. 1991) . Here, we report the isolation of two rice MET1 genes that differ in
organization and steady state expression levels. We also show reactivation of GUS expression by
bombardment of rice callus containing a silenced gus transgene with constructs designed to
generate dsRNA complementary to conserved MET1 regions.
Materials and Methods
4
Plant materials
Rice (Oryza sativa subspecies japonica cv. Taipei 309) seeds were surface sterilized in sodium
hypochlorite (3% v/v) for 45 min, then thoroughly rinsed with sterile distilled water. Subsequently, the
seeds were germinated on MS medium (Murashige and Skoog 1962) at 26°C under a 18 h/6 h (light/dark)
regime for 15 days, transferred to soil and grown to maturity in greenhouse at 26°C with a 14 h/10 h
photoperiod. Root and leaf samples were collected 10 days after germination. Inflorescence and leaf
samples were collected upon flowering. Callus was induced from mature seeds and maintained on N6
medium (Chu and Bi 1975) for 3-5 weeks..
Nucleic acid extraction and Southern blot analysis
Rice genomic DNA was isolated from leaves as described previously (Buchholz et al. 1998a) . Genomic
DNA (2 µg) was digested with 10 units of appropriate restriction enzymes (NEB) for 3 h. After
electrophoretic separation in a 0.8% agarose gel, the DNA fragments were transferred to Hybond-N+
membrane (Amersham). DNA probes were labeled using a DECAprime II™ kit (Ambion).
Hybridizations were performed using ULTRAhyb
(Ambion) according to the manufacturer’s
recommendations. Copy number estimations were conducted as described by Buchholz et al. (1998a), a
3.4 kb HindIII fragment of OsMET1-1 was gel-purified and diluted to the required concentration to obtain
the desired copy number standards.
PCR amplification of DNA fragments encoding conserved motifs of DNA MTases
Gene-specific probes for screening DNA MTase were amplified from rice genomic DNA with the forward
primer:5'-CGTCTAGCTACTCTTGACATTTTTG
and
reverse
primer:
5'-
AAAGTTCCGAACATTTTCTAACAG. The primers were nucleotide sequences corresponding to motifs
I (FGG) and VI (ENV) present in tobacco MET1 (NtMET1, AB030726) (Nakano et al. 2000) . PCR
amplification was performed using Taq DNA polymerase (Promega). A 10 µl sample was subjected to
5% PAGE for visualization of the PCR products, and the remainder was used for gel purification of the
products. The PCR fragments were treated with 0.5 units of Pfu DNA polymerase (Stratagene) at 72EC
for 30 min and cloned into pPCR-Script Amp (SK+)™ (Stratagene). The clones containing the insert
were identified by digestion with BssHII (NEB). Clone number 4, designated as R4, and clone number 8,
designated as R8, were sequenced. Nucleotide sequence comparison by the ClustalW program (AlignX)
5
from Vector NTI (InforMax) indicated that the nucleotide sequence of R4 (744 bp) and R8 (719 bp)
respectively showed 76% and 72% identity to sequences containing motifs I to VI of maize MET1
(AF063403). The R4 and R8 fragments were used in further screening of OsMET1-1 and OsMET1-2 in
BAC libraries and for genomic blot analyses.
Screening of OsMET1-1 and OsMET1-2 from rice genomic BAC libraries
Filter sets of rice genomic BAC libraries (Nipponbare-BamHI, NPB; Teqing-BamHI, TQB and TeqingHindIII, TQH) were purchased from GENEfinder genomic resources at Texas A&M University. The
filters were initially screened with an R8-specific probe (719 bp) using ULTRAhyb (Ambion) at 42EC
for 16 h. Following two sequential washes with 2X SSC and 0.1% SDS at 50EC for 5 min and 0.1X SSC
and 0.1% SDS at 50EC for 15 min, the filters were subjected to phosphorimager analysis (FUJIX
BAS2000 Phosphorimaging System). Eight positive clones out of 21,504 clones of the NPB library, three
positive clones out of 12,288 clones in the TQH library and one positive clone out of 7,296 clones in the
TQB library were obtained. To further analyze these clones, the BAC plasmid was isolated and digested
with EcoRI, HindIII, NcoI or SacI. The digested fragments were resolved using 0.8% agarose gel
electrophoresis and transferred to Hybond™ N+ membrane. The membranes were then hybridized to
radiolabeled probes and positive clones were confirmed by phosphorimager analysis.
Cloning of the OsMET1-1 full-length genomic sequence
Various fragments of OsMET1-1 present in NPB clone 33-O-20 were assembled to yield the complete
genomic sequence. These included a HindIII-1 fragment (H1), a SacI-1 fragment (S1), a HindIII-2
fragment (H2), a SacI-2 fragment (S2), and a SacI-3 fragment (S3). Initially, the BAC plasmid was
digested with HindIII or SacI. Subsequently, the fragments were cloned into HindIII or SacI-digested and
dephosphorylated vector pBlueScript- SK+. Positive clones bearing various fragments were obtained by
colony-lifts and hybridization to radiolabeled probes using standard procedures. The following probes
were used to obtain the various subclones: R8 probe for the H1 fragment, ES probe for the S1 fragment,
SN probe for the H2 fragment, PS probe for the S2 fragment and BH probe for the S3 fragment (see Fig
2A for probe locations). The subclones were sequenced as described below and the complete sequence
was then assembled using Gene Construction Kit 2.5 software (Textco).
Cloning of the OsMET1-2 full-length genomic sequence
6
The full-length sequence of OsMET1-2 was obtained by comparison of OsMET1-1 with the rice genome
database available at http://www.tigr.org. Nucleotide sequence comparison using AlignX indicated that
OsMET1-1 had 99% identity to a region in clone AC093713, located on chromosome 3 and showed
68.5% identity to a region of accession AP003835, located on chromosome 7, respectively. This
similarity indicated that AP003835 includes a second DNA MTase and a gap in this sequence was
resolved by PCR amplification (35 cycles) using rice (cv. Nipponbare) genomic DNA (100 ng) as a
template
and
Pfu
DNA
polymerase
GATGTCTGTCAATCTCCAATCATTGC
(Stratagene).
and
The
the
forward
reverse
primer
primer
was:
was:
5'5'-
GGTCTAATATACAAATAGTCATTGACATTGTAGG.
DNA sequencing and analysis
All sequencing was performed on Applied Biosystems 373XL and 377 XL DNA sequencers (Gene
Technologies Lab, Institute of Developmental and Molecular Biology, Texas, USA). Nucleotide sequence
and amino acid sequence comparisons were conducted using AlignX from Vector NTI (InforMax). A
determination of conserved domains was obtained using the programs Prosite (ExPaSy database) and
RBS Blast (NCBI database).
RT-PCR and 3’RACE
Total RNA was extracted from 10 g of leaf tissue using the RNeasy plant mini kit under conditions
detailed by the supplier (Qiagen) and treated with DNaseI (Invitrogen, 1 unit/Fl), 20 mM Tris-HCl, pH
8.3, 50 mM KCl, 2 mM MgCl2 at 25EC for 15 min to remove DNA and then heated at 65EC for 10 min in
the presence of 2.5 mM EDTA to inactivate DNase. RT-PCR reactions were performed using a OnestepRT-PCR kit (Qiagen) according to the manufacturer’s recommendations. Reverse transcription was
performed at 50EC for 30 min using a mixture of Omniscript and Sensiscript Reverse Transcriptases
(Qiagen) employing the manufacturer’s recommendations Introns 2, 3 and 5 to 8 of OsMET1-1 were
confirmed
using
the
following
primers.
For
intron
2,
forward
primer
5'-
TTGGAGCTCATCCCTATGAAAGC and reverse primer 5'-AAGGGTGATGATACTAACAGCTAGC.
For
intron
3,
forward
primer
AAGATCATGCAAGACAAGCTCG.
5'-ATGGTCCTGCTGGATCAAGG
For
intron
5
to
and
8,
reverse
forward
primer
5'-
primer
TTGAGCTAGGTGGTTCAGACAAACC and reverse primer 5'-AAGAGTCTCTCCAGGTGCAGCG.
7
For visualization of the RT-PCR products, a 10 Fl aliquot was resolved in a 1 or 2% agarose gel,
depending on the expected size of the amplified fragment. The remainder of the reaction was used for gel
purification. Purified fragments were blunted using Pfu DNA polymerase and cloned using a PCRScript™ cloning kit. Subsequently, the clones bearing the desired insert were screened by digestion with
BssHII and sequenced.
3’RACE was performed using a First Choice
TM
RLM-RACE kit (Ambion) under the conditions
recommended by the manufacturer. First strand cDNA was synthesized at 42EC for 1 h using the adapter
supplied with the kit (5'-(A)12CCTATAGTGAGTCGTATTAATTCTGTGCTCGC) and RNA (2 Fg)
from mature leaves as the template. A 1 Fl of reverse transcription reaction was subsequently used in 35
cycles of PCR amplification using the 3' RACE outer primer (5'-GCGAGCACAGAATTAATACGACT)
and the 3' RACE gene-specific outer primer (5'-AAGAGATGAATGAGCTGAACCTC) under the
following cycling conditions: 94EC for 30 sec, 65EC for 30 sec, and 72EC for 30 sec. A 10 Fl aliquot was
resolved in a 2% agarose gel for visualization. The remainder of the reaction was used for gel purification
of the PCR product. The purified 3’RACE product was blunted using Pfu DNA polymerase, cloned into
vector pPCR-Script™ (Stratagene) and plasmids bearing the expected inserts were sequenced.
Ribonuclease Protection Assay
A 227 bp PCR fragment corresponding to the 5' region of OsMET1-1, designated as sM1, was amplified
and cloned into pCR-Script Amp (SK+)™ (Stratagene) to generate an antisense construct, psM1/pPCR.
An antisense riboprobe of 312 nucleotides (227 nucleotides of coding region plus 85 nucleotides of
polylinker) was synthesized by in vitro transcription (MAXIscript™ T7/T3 Kit) (Ambion) using T3 RNA
polymerase and HindIII-linearized psM1/pPCR as the template. Similarly, a 125 bp PCR fragment
corresponding to exon 4 of OsMET1-2, named sM2, was amplified and subcloned into pPCR-Script
(Stratagene) to generate an antisense construct, psM2/pPCR. An antisense riboprobe of 180 nucleotides
(125 nucleotides of coding region plus 55 nucleotides of polylinker) was synthesized by in vitro
transcription using T3 RNA polymerase and HindIII-linearized psM2/pPCR as the template. Total RNA
was extracted from callus tissue, roots, inflorescence and 10 day old leaves using TRIzol Reagent
(Invitrogen). RNase protection assays were performed on the extracted RNA using an RPAIII
TM
kit
(Ambion) according to the manufacturer’s instructions. The protected fragments were analyzed by
electrophoresis on 5% polyacrylamide/8 M urea gels.
The amount of OsMET1-1, OsMET1-2 and 18S rRNA transcripts was determined by measuring
the intensity of signal obtained from phosphorimager using MacBas image analysis software. Expression
8
levels for OsMET1-1 and OsMET1-2 were calculated following normalization to 18S rRNA transcript
levels.
RNAi constructs corresponding to the N- and C-terminal regions of OsMET1-1
A modified version of pHANNIBAL (Wesley et al. 2001) was used as a vector for cloning and expressing
inverted repeat sequences corresponding to selected 5' and 3' regions of OsMET1-1. To modify this
vector, the 35S CaMV promoter in pHANNIBAL was replaced with the rice Actin1 promoter (McElroy et
al. 1990) to obtain vector pHA. Sense and anti-sense orientations of a 5' region were designated as sNt
and asNt, respectively. Sense and anti-sense orientations of a conserved 3' region were designated as sCt
and asCt, respectively. The primers used for amplification of the sense orientation of Nt (forward:
GGTACCTTCCAAGAAAGAGAGCAATGG; reverse: GGTACCATGGAACGGACAACACCAG) and
Ct
(forward:
GGTACCTTGAGCTAGGTGGTTCAGACAAACC;
GGTACCAAGAGTCTCTCCAGGTGCAGCG)
(underlined).
Primers
for
the
regions
antisense
of OsMET1-1
orientation
contained a KpnI site
of
TCTAGATTGAGCTAGGTGGTTCAGACAAAC;
AAGCTTAAGAGTCTCTCCAGGTGCAGCG)
reverse:
Nt
(forward:
reverse:
and
TCTAGATTGAGCTAGGTGGTTCAGACAAACC;
Ct
(forward:
reverse:
AAGCTTAAGAGTCTCTCCAGGTGCAGCG) included either an XbaI or a HindIII site (underlined).
Following PCR amplification, the purified sense and anti-sense orientation products were digested with
KpnI or HindIII and XbaI. The purified PCR fragments of sNt and sCt were separately cloned into KpnIdigested, dephosphorylated pHA to obtain constructs psNt/pHA and psCt/pHA. The PCR products of
asNt and asCt were digested with HindIII-XbaI and ligated to HindIII and XbaI-digested psNt/pHA or
psCt/pHA, respectively. The plasmids containing the inverted repeats of the 5' and 3' regions of OsMET11 were named pir-Nt and pir-Ct, respectively. In addition, 35S-gfp-nos was inserted as a NotI-SphI
fragment downstream of the ir-Nt and ir-Ct expression cassettes.
Reactivation of GUS in silenced transgenic callus
JDV92-8 is rice line transformed with a casette containing three genes (35S-hpt-35S::35S-uidAnos::mUbi1-bar-nos), of which uidA is silenced but capable of reactivation in the presence of 5azacytidine (Hall et al. 2001) , suggesting the involvement of methylation. Callus was induced from
mature seeds of this line by culture on N6 (Chu and Bi 1975) medium at 28EC in the dark for 3-5 weeks.
9
Prior to bombardment, calli were transferred to N60 medium (N6 medium containing 3% w/v sorbitol) for
4 h and then bombarded (PDS 1000/ He biolistics system, BioRad) using gold particles (3 mg; 1 µm o.d.)
coated with plasmid DNA (5 µg). For each construct, two Petri dishes (10 cm dia.), each bearing 25 calli,
were bombarded.
Following bombardment, the calli were cultured on N60 medium in the dark; after three days they
were examined for GFP expression using a Stemi SV11 APO Microscope (Zeiss) fitted with an AxioCam
HRc camera. Calli that expressed GFP were imaged using a 500 nm filter and an exposure time of 3 sec,
separated from non-expressing calli and subjected to histochemical staining (Jefferson et al. 1987) to
visualize GUS expression.
Results
Isolation and characterization of genomic DNA encoding DNA MTases in rice
We noted that the rice putative DNA MTase present in pRM1 (N. Wiriyawutikorn and J.
Narangajavana; submitted as GenBank accession AF155874) displayed 99% nucleotide
sequence identity to mouse Dnmt1 (NM 010066), casting doubt on its authenticity. To clarify
this situation, a genomic blot of rice (cv. Taipei 309) was probed using a DNA fragment of
pRM1 that corresponds to the conserved C-terminal region of DNA MTase. No hybridization
was detected, even after extended (32 h) exposure (data not shown). This revealed that pRM1 did
correspond to a rice DNA MTase. In contrast, PCR amplification of rice (Taipei 309) genomic
DNA using primers corresponding to highly conserved motifs I and VI from tobacco NtMET1
(GenBank AB030726: Nakano et al. 2000) , yielded two distinct products: R4, 744 bp and R8,
719 bp (Fig. 1A). This observation suggested the presence of two closely related genes encoding
DNA MTase in rice.
Cloning and sequencing of the R4 and R8 fragments revealed that each includes a
predicted coding region of four exons and three introns. The two fragments share 73% nucleotide
and 91% amino acid identity, with several small, dispersed, nucleotide insertions accounting for
the larger size of R4 (Fig. 1B). Also, the R4 and R8 fragments showed 76% and 72% identity,
respectively, to sequences corresponding to motifs I to IV of maize MET1 (AF063403). To
10
obtain full-length genomic clones encoding these putative rice MTases, three rice genomic BAC
libraries: Nipponbare-BamHI (NPB), Teqing-BamHI (TQB) and Teqing-HindIII (TQH), were
screened, initially using R8 as a probe. Eight candidate clones were obtained from the NPB
library; three from the TQH library and one from the TQB library.
The subcloning and sequencing strategies used to characterize the BAC clones are shown
in Fig. 2A. Southern blot analysis using R8 as the probe revealed that all of the clones from the
NPB library (representative clones are shown in Fig. 2B, panel 1) and two clones from the TQH
library (Fig. 2B, panel 2) displayed similar restriction patterns, and were designated as
corresponding to OsMET1-1. A different pattern was obtained for two clones (TQH3-F-21 and
TQB10-B-12) that hybridized weakly to the R8 probe (Fig. 2B, panel 3) but hybridized strongly
to the R4 probe (Fig. 2B, panel 4) that corresponds to OsMET1-2. The positions of the strongly
hybridizing bands in Fig. 2B, panel 4 differ from those in panels 1 and 2, reflecting the
differences in OsMET1-1 and OsMET1-2. Similarly, the weak signals obtained for the PS probe
in panel 7 differ in location from the OsMET1-1 fragments shown in panels 5 and 6. These
findings further confirm that clones TQH 3-F-21 and TQB 10-B-12 contained OsMET1-2. The
differences in restriction patterns for Fig. 2, panels 5 and 6 reflect the fact that the TQH clones
lack the upstream region present in the NPB clones.
The complete DNA sequence of OsMET1-1 (GenBank Accession No. AF462029) was
obtained by sub-cloning various fragments derived from the R8-hybridizing BAC clone NPB33O-20 (Fig 2A). The sequence of OsMET1-2 was derived from the TIGR (http://www.tigr.org)
rice genome database using OsMET1-1 as the query. A region missing from the OsMET1-2
sequence in the database was obtained by PCR amplification using rice (cv. Nipponbare)
genomic DNA as a template and GenBank accession number AY230205). OsMET1-1 contains
an open reading frame of 4,566 nucleotides, corresponding to a protein of 1,522 amino acids
with a predicted Mol. Wt. of 170 kDa. The assembled sequence of OsMET1-2 has an open
reading frame of 4,452 nucleotides (1,484 amino acids: 167 kDa).
As summarized in Fig. 3A, comparison of the OsMET1-1 and OsMET1-2 sequences (see
electronic supplementary material) revealed that, with the exception of exon 1, their coding
regions share 69-88% identity. Nevertheless, striking differences exist between these two genes:
they share only 24% identity in exon 1 and intron 3 of OsMET1-1 is absent from OsMET1-2.
11
While most of the introns are of similar size for the two genes, introns 2 and 11 are substantially
longer in OsMET1-1 than in OsMET1-2. The presence of most of the predicted introns and exons
of OsMET1-1 was confirmed by RT-PCR and the 3' polyadenylation site was determined using
RACE analyses. The amplified RT-PCR (Fig. 3B) and 3' RACE (Fig. 3C) products were cloned.
Sequencing of the cloned RT-PCR products validated the presence of predicted introns 2 (RE-3
region), 3 (RE-2 region) and 5 to 8 (RE-1 region). Additionally, sequencing of the 3' RACE
products established the presence of intron 11, the 3' UTR and the polyadenylation site
Comparison of OsMET1-1 and OsMET1-2
A comparison of the entire OsMET1-1 and OsMET1-2 sequences is provided in electronic
supplementary material and is summarized in Fig. 4A. A lysine-glycine repeat that is highly
conserved in DNA MTases (Finnegan and Kovac 2000) was present near the center of the
predicted coding region in both OsMET1-1 and OsMET1-2. The amino acid sequence
downstream of this motif in OsMET1-1 (the catalytic, C-terminal, domain: residues 1066 - 1522)
showed a higher conservation (86.5%) with that of OsMET1-2 than did that (67.7%) for the Nterminal domain (residues 1 - 1059). As shown in Fig. 4, the inferred amino acid sequences of
OsMET1-1 and OsMET1-2 contain several functional regions. These include a glutamic acid
rich region (residues 660-684 for OsMET1-1 and residues 637-662 for OsMET1-2), an IQ
calmodulin binding region (residues 809-838, for OsMET1-1 and 781-810 for OsMET1-2), two
bromo adjacent homology domains (BAH, residues 737-869; 905-1,044 for OsMET1-1 and 709841; 878-1,017 for OsMET1-2), a bipartite nuclear localization signal (NLS, residues 955-972
for OsMET1-1 and 928-944 for OsMET1-2) and a methyltransferase catalytic domain (residues
1,087-1,515 for OsMET1-1 and 1,061-1,489 for OsMET1-2). Eight conserved motifs (motif I;
FxGxG, motif II; EW; motif IV; PCQ, motif VI; ENV, motif VII; DY, motif VIII; QxRxR, motif
IX; RE, and motif X; GN) (Kumar et al. 1994) , arranged in order, were present in the
methyltransferase domain of both rice MET1 clones (see electronic supplementary material).
Additionally, as shown in Fig 4B, the rice DNA MTases are related to DNA MTases of
Dnmt1/MET1 class in diverse organisms from several kingdom.
12
Genomic organization of OsMET1-1 and OsMET1-2
To determine the copy number of OsMET1-1, rice (cv. Taipei 309) genomic DNA was digested
with EcoRI, HindIII, NcoI or SacI and separated by electrophoresis (Fig. 5). The 3.4 kb HindIII
fragment hybridized with an intensity approximately equivalent to the 1 copy reconstruction
upon hybridization with the R8 probe (Fig. 5, panel 1). Several other, weakly hybridizing,
fragments were also observed, indicative of the presence of another DNA MTase, possibly
OsMET1-2. This possibility was confirmed by stripping and rehybridizing the blot with the R4
probe, yielding strong bands with positions corresponding to those of the weak signal from probe
R8 (Fig. 5, compare panels 1 and 2). Furthermore, a single band was obtained for each digest
when the blot was stripped and hybridized with the PS (Fig. 5, panel 3) or the ES (Fig. 5, panel
4) probes, that are located in the OsMET1-1 clone (Fig. 2A). These data support the conclusion
that rice contains a single copy each of OsMET1-1 and OsMET1-2 and are in agreement with the
TIGR database which predicts one DNA MTase on chromosome 3 (OsMET1-1) and another
(OsMET1-2)
on
chromosome 7.
13
Spatial and developmental profiles of OsMET1-1 and OsMET1-2 expression
The spatial expression patterns for OsMET1-1 and OsMET1-2 were evaluated by ribonuclease
protection assays. To permit discrimination between the expression of these two highly similar
genes, the sM1 fragment (Fig. 3A) in exon 2 of OsMET1-1 and the sM2 fragment in exon 4 of
OsMET1-2 were used as riboprobes. Since probe sM1 has only 46% identity to OsMET1-2 and
sM2 has only 68% identity with OsMET1-1, their differences are sufficient to permit the
desired discrimination. As shown in Fig. 6A, expression of both OsMET1-1 and OsMET1-2
was detected in callus, root and inflorescence. However, the steady state level of OsMET1-2
was 7-12 fold higher than that for OsMET1-1 in these tissues (Fig. 6B). No transcript for
OsMET1-2 was detectable in differentiated tissue (10 day-old leaf), and no expression for
either gene was found in mature leaves.
Reactivation of a GUS-silenced line by inverted repeat constructs of OsMET1-1
DNA methylation is frequently associated with gene silencing and mutation of Arabidopsis
METI has been shown to interfere with both methylation and silencing (Finnegan et al.
1996;Jones et al. 2001;Morel et al. 2000;Ronemus et al. 1996) . Consequently, we sought to
explore the functionality of OsMET1-1 by using it to debilitate MTase in calli of a transgenic
rice line (JDV92-8) that bears a silenced copy of 35S-uidA-nos that is capable of reactivation
upon exposure to the hypomethylating agent 5-azacytidine (Hall et al. 2001) . RNAi has been
shown to be a potent approach to induce gene silencing, both in stably transformed plants
(Chuang and Meyerowitz 2000;Stoutjesdijk et al. 2002;Wesley et al. 2001) and in transient
experiments (Schweizer et al. 2000;Zentella et al. 2002) . To debilitate rice MTase, 5’ (Nt) and
3’ (Ct) regions of OsMET1-1 were cloned as inverted repeat (ir) constructs pir-Nt and pir-Ct
(Fig. 7A) suitable for transient expression in bombardment experiments. These constructs were
designed to produce hairpin RNA structures that contain an intron. Excision of the intron by
splicing yields dsRNA that is an effective gene-specific mediator in triggering RNAi (Smith et
al. 2000;Wesley et al. 2001) . The pir-Nt and pir-Ct constructs also include 35S-gfp-nos to
express GFP as an easily detectable reporter of successful bombardment. Transient expression
of RNAi corresponding to OsMET1-1 by particle bombardment (Buchholz et al. 1998b) was
predicted to inactivate maintenance methylation and thereby restore GUS activity.
14
As expected, calli bombarded with pJD7 did not express GFP (Fig. 7B, panel 1),
but did express GUS (Fig. 7B, panel 5), showing successful bombardment and histochemical
staining. Expression of GFP but not GUS (Fig. 7B, panels 2 and 6, respectively) showed that
bombardment with pPT1 was successful but did not result in reactivation of the endogenous,
silenced, 35S-uidA gene. However, calli bombarded with the pir-Nt (Fig. 7B, panels 3 and 7) or
pir-Ct (Fig. 7B, panels 4 and 8) constructs exhibited green fluorescence, confirming successful
bombardment, that was coincident with GUS expression from the endogenous 35S-uidA gene.
These results reveal that inactivation of DNA MTase, and hence of maintenance methylation,
led to reactivation.
Discussion
Features and comparison of the DNA and amino acid sequences of OsMET1-1 and OsMET1-2
The sequences and corresponding translated proteins of both OsMET1-1 and OsMET1-2 were
identified as DNA MTases, based on their homology with other eukaryotic DNA MTases
(Fig. 4). OsMET1-1 and OsMET1-2 are similar at both the nucleotide and amino acid levels,
particularly in the C-terminal domain. Differences at the nucleotide level are primarily within
introns. The predicted rice DNA MTase proteins were similar, with amino acid substitutions
scattered along the ORFs. A glutamic acid-rich region previously shown to be conserved
among plant MET1 sequences (Finnegan and Kovac 2000) was present, but no homology was
apparent to the cysteine-rich or Zn ion-binding regions thought to be unique to animal Dnmt1s.
Two long BAH domains were identified downstream of the glutamic acid-rich region. The
BAH domain has been proposed to be a protein-protein interaction module specialized for gene
silencing and is present in proteins involved in replication and transcriptional processes
(Callebaut et al. 1999) . This suggests a possible involvement of the BAH domains of rice
MTases in mediating DNA methylation. The presence in the first BAH domain of an IQ motif,
an extremely basic element known to bind calmodulin and calmodulin-like proteins (Bahler
and Rhoads 2002) , suggests a possible connection between DNA methylation and Ca2+
signaling. The second BAH domain in the rice MET1 sequence bears a putative nuclear
localization sequence, making it likely that it, like other enzymes in the Dnmt1/MET1 class, is
targeted to the nucleus.
15
Basic amino acid-rich regions, known to be preferential targets for proteases are present
towards the N-terminus and in the K-G repeat motifs of both rice DNA MTases, suggesting
that post-translational proteolytic processing may occur. Indeed proteins of 35 to 55 kDa
exhibiting DNA MTase activity have been reported from wheat (Theiss et al. 1987) and rice
(Giordano et al. 1991) that may result from proteolytic cleavage at the K-G repeat. The Cterminal domains of the rice DNA MTases bear six motifs (I, IV, VI, VIII, IX, and X)
characteristic of conserved elements found in prokaryotic and eukaryotic MTases (Finnegan
and Kovac 2000) , and it is likely that all, or some, of these are active in catalytic function.
Comparison between rice MET1s and enzymes of the Dnmt1/MET1 class in other animals and
plants.
The phylogenetic relationships shown in Fig. 4B revealed that maize MET1 is more closely
related to OsMET1-2 than to OsMET1-1. Amino acid sequence comparison of rice MET1s to
other plant MET1 sequences revealed high similarity (56-75% identity) but limited similarity
(~24% identity) to animal Dnmt1. The N- and C-terminal domains of the rice DNA MTases
have 42-51% and 71-78% similarity, respectively, to those of dicots. The C-terminal (catalytic)
domain of the rice MTases share 44-50% identity with those of animal Dnmt1s but the Nterminal regions bear little (13-15%) similarity. Although the methyltransferase domains in
plant DNA MTases are very similar to those in animals, ascomycetes and prokaryotes, some
differences do exist (see supplementary material, Figure 2). Notably, the highly conserved
leucine residue of motif V in animals, ascomycetes and prokaryotes is replaced by methionine
in MET1 of rice and other plants. Plant MTases appear to lack a substantial stretch of the
variable region between motifs VIII and IX (Kumar et al. 1994) . However, a target recognition
domain having the weak consensus T(V/I/L)XXXXXXG(V/L), discerned by Lauster et al.
(1989), may be represented within this region by TSVTDPQPMGKV at positions 491-503 of
the rice MET1 sequences. A GenBank database search retrieved yet another putative DNA
MTase, named MET2a (AC069324). This protein bears a chromomethylase domain and, as
shown in Fig 4B, MET2a from rice and maize form a distinct class and are not closely related
to MET1 class DNA MTases.
Copy number and spatial expression of OsMET1-1 and OsMET1-2
16
In agreement with recent information from the TIGR rice genome project, both OsMET1-1 and
OsMET1-2 were found to be present as one copy per haploid genome (Fig. 5). An appraisal of
the spatial expression profiles for the rice MTases (Fig. 6) shows a high abundance of
transcripts in young plant tissues, suggesting that these MTases may be involved in
maintenance of DNA methylation. Indeed, the expression profiles of the rice MTases are
similar to that observed for tobacco MET1 (Nakano et al. 2000) . Additionally, Arabidopsis
MET1 has been shown to play a pivotal role in maintaining methylation patterns at CpG
residues. It is expressed to high levels in both flower and vegetative tissue as compared to other
DNA MTases (Genger et al. 1999) , with the highest expression in meristematic cells
(Ronemus et al. 1996) . Interestingly, the two carrot DNA Mtases are expressed to high levels
in proliferating tissues but differ from each other in their relative abundance (Bernacchia et al.
1998) . As shown in Fig. 6, the relatively high levels of OsMET1-2 transcript in actively
proliferating tissues compared to OsMET1-1 transcripts provokes the speculation that
OsMET1-2 may be the predominant MTase in rice and may be orthologous in its function to
known tobacco, carrot and Arabidopsis DNA MTases.
Function of OsMET1
The biochemistry of RNA-mediated post-transcriptional gene regulation has rapidly advanced
through studies on nematodes (Fire et al. 1998;Tabara et al. 1998) , flies (Elbashir et al.
2001;Yang et al. 2000) , and plants (Hamilton et al. 2002;Hamilton and Baulcombe 1999) .
Given the high sequence similarity of OsMET1-1 and OsMET1-2, the RNAi approach used in
the experiments shown in Fig. 8 can be expected to lead to degradation of mRNA for both
genes, thereby debilitating maintenance methylation in the silenced rice callus. The time course
for reversal of silencing was three days in the present experiments and approximately one week
for exposure of germinating seeds to 5-azacytidine (Hall et al. 2001 and unpublished
observations). This is in accord with activation of the silenced promoter through incorporation
of the non-methylatable cytidine analog into newly replicated DNA in the case of 5-azacytidine
treatment and debilitation of maintenance methylation in the tissues bombarded with RNAi
constructs. An alternative mechanism could involve DNA demethylation, but a report of an
active DNA demethylase (Ramchandani et al. 1999) was challenged on the basis of the high
energetic requirements of this reaction (Wolffe et al. 1999) . It would appear that the wellcharacterized, silenced rice line and RNAi constructs described here will be valuable for
gaining further insight to maintenance methylation in rice. Indeed, the transient assay system
17
described here provides a versatile and rapid approach to evaluate the role of any candidate
protein(s) in gene silencing.
Acknowledgements This work was funded by the Texas Advanced Technology Program (0103660092-1999) and NSF (MCB 0110477). MBC was supported by NSF grant MCB 9974706. PT was
funded in part by The Royal Golden Jubilee Ph.D. Program (PHD/00075/2541), The Thailand Research
Fund. We thank Guojun Yang, Lakshminarayan Iyer and Amaret Bhumiratana for valuable discussions.
References
Amedeo P, Habu Y, Afsar K, Scheid OM, Paszkowski J (2000) Disruption of the plant gene MOM releases transcriptional
silencing of methylated genes. Nature 405: 203-206
Bahler M, Rhoads A (2002) Calmodulin signaling via the IQ motif. FEBS Lett. 513: 107-113
Baroux C, Spillane C, Grossniklaus U (2002) Genomic imprinting during seed development. Adv. Genet. 46: 165-214
Bartee L, Malagnac F, Bender J (2001) Arabidopsis cmt3 chromomethylase mutations block non-CG methylation and silencing
of an endogenous gene. Genes Dev. 15: 1753-1758
Bernacchia G, Primo A, Giorgetti L, Pitto L, Cella R (1998) Carrot DNA-methyltransferase is encoded by two classes of genes
with differing patterns of expression. Plant J. 13: 317-329
Brackertz M, Boeke J, Zhang R, Renkawitz R (2002) Two highly related p66 proteins comprise a new family of potent
transcriptional repressors interacting with MBD2 and MBD3. J. Biol. Chem. 277: 40958-40966
Buchholz WG, Connell JP, Kumpatla SP, Hall TC (1998a) Molecular analysis of transgenic rice. Methods Mol. Biol. 81: 397415
Buchholz WG, Teng W, Wallace D, Ambler JR, Hall TC (1998b) Production of transgenic rice (Oryza sativa subspecies
japonica cv. Taipei 309). Methods Mol. Biol. 81: 383-396
Callebaut I, Courvalin JC, Mornon JP (1999) The BAH (bromo-adjacent homology) domain: a link between DNA methylation,
replication and transcriptional regulation. FEBS Lett. 446: 189-193
Cao X, Jacobsen SE (2002) Locus-specific control of asymmetric and CpNpG methylation by the DRM and CMT3
methyltransferase genes. Proc. Natl. Acad. Sci. U.S.A. 99: 16491-16498
Cao X, Springer NM, Muszynski MG, Phillips RL, Kaeppler S, Jacobsen SE (2000) Conserved plant genes with similarity to
mammalian de novo DNA methyltransferases. Proc. Natl. Acad. Sci. U.S.A. 97: 4979-4984
Chaudhury AM, Koltunow A, Payne T, Luo M, Tucker MR, Dennis ES, Peacock WJ (2001) Control of early seed
development. Ann. Rev. Cell Dev. Biol. 17: 677-699
Chu C, Bi Y (1975) Establishment of an efficient medium for anther culture of rice thorough comparative experiments on the
nitrogen sources. Sci. Sin. 18: 659–668
Chuang CF, Meyerowitz EM (2000) Specific and heritable genetic interference by double-stranded RNA in Arabidopsis
thaliana. Proc. Natl. Acad. Sci. U.S.A. 97: 4985-4990
Curradi M, Izzo A, Badaracco G, Landsberger N (2002) Molecular mechanisms of gene silencing mediated by DNA
methylation. Mol. Cell. Biol. 22: 3157-3173
Elbashir SM, Lendeckel W, Tuschl T (2001) RNA interference is mediated by 21- and 22-nucleotide RNAs. Genes Dev. 15:
188-200.
Feng Q, Cao R, Xia L, Erdjument-Bromage H, Tempst P, Zhang Y (2002) Identification and functional characterization of the
p66/p68 components of the MeCP1 complex. Mol. Cell. Biol. 22: 536-546
Finnegan EJ, Dennis ES (1993) Isolation and identification by sequence homology of a putative cytosine methyltransferase
from Arabidopsis thaliana. Nucleic Acids Res. 21: 2383-2388
18
Finnegan EJ, Kovac KA (2000) Plant DNA methyltransferases. Plant Mol. Biol. 43: 189-201
Finnegan EJ, Peacock WJ, Dennis ES (1996) Reduced DNA methylation in Arabidopsis thaliana results in abnormal plant
development. Proc. Natl. Acad. Sci. U.S.A. 93: 8449-8454
Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC (1998) Potent and specific genetic interference by doublestranded RNA in Caenorhabditis elegans. Nature 391: 806-811
Genger RK, Kovac KA, Dennis ES, Peacock WJ, Finnegan EJ (1999) Multiple DNA methyltransferase genes in Arabidopsis
thaliana. Plant Mol. Biol. 41: 269-278
Giordano M, Mattachini ME, Cella R, Pedrali-Noy G (1991) Purification and properties of a novel DNA methyltransferase
from cultured rice cells. Biochem. Biophys. Res. Commun. 177: 711-719
Hall TC, Kumpatla SP, Kharb P, Iyer L, Cervera M, Jiang Y, Wang T, Yang G, Teerawanichpan P, Narangajavana J, Dong J
(2001) Gene silencing and its reactivation in transgenic rice. In: Khush GS, Brar DS, Hardy B (eds) Rice Genetics.
Science Publishers, Inc., New Delhi, pp 465-481
Hamilton A, Voinnet O, Chappell L, Baulcombe D (2002) Two classes of short interfering RNA in RNA silencing. EMBO J.
21: 4671-4679
Hamilton AJ, Baulcombe DC (1999) A species of small antisense RNA in posttranscriptional gene silencing in plants. Science
286: 950-952
Hendrich B, Bird A (1998) Identification and characterization of a family of mammalian methyl-CpG binding proteins. Mol.
Cell Biol. 18: 6538-6547
Henikoff S, Comai L (1998) A DNA methyltransferase homolog with a chromodomain exists in multiple polymorphic forms in
Arabidopsis. Genetics 149: 307-318
Jefferson RA, Kavanagh TA, Bevan MW (1987) GUS fusions: ß-glucuronidase as a sensitive and versatile gene fusion marker
in higher plants. EMBO J. 6: 3901-3907
Jiang CL, Jin SG, Lee DH, Lan ZJ, Xu X, O'Connor TR, Szabo PE, Mann JR, Cooney AJ, Pfeifer GP (2002) MBD3L1 and
MBD3L2, two new proteins homologous to the methyl-CpG-binding proteins MBD2 and MBD3: Characterization of
MBD3L1 as a testis-specific transcriptional repressor. Genomics 80: 621-629
Jones L, Ratcliff F, Baulcombe DC (2001) RNA-directed transcriptional gene silencing in plants can be inherited
independently of the RNA trigger and requires Met1 for maintenance. Curr. Biol. 11: 747-757
Kumar S, Cheng X, Klimasauskas S, Mi S, Posfai J, Roberts RJ, Wilson GG (1994) The DNA (cytosine-5) methyltransferases.
Nucleic Acids Res. 22: 1-10
Kumpatla SP, Teng W, Buchholz WG, Hall TC (1997) Epigenetic transcriptional silencing and 5-azacytidine-mediated
reactivation of a complex transgene in rice. Plant Physiol. 115: 361-373
Lauster R, Trautner TA, Noyer-Weidner M (1989) Cytosine-specific type II DNA methyltransferases. A conserved enzyme
core with variable target-recognizing domains. J. Mol. Biol. 206: 305-312
Li E, Bestor TH, Jaenisch R (1992) Targeted mutation of the DNA methyltransferase gene results in embryonic lethality. Cell
69: 915-926
Matzke MA, Aufsatz W, Kanno T, Mette MF, Matzke AJ (2002) Homology-dependent gene silencing and host defense in
plants. Adv. Genet. 46: 235-275
McElroy D, Zhang W, Cao J, Wu R (1990) Isolation of an efficient actin promoter for use in rice transformation. Plant Cell 2:
163-171
Morel JB, Mourrain P, Beclin C, Vaucheret H (2000) DNA methylation and chromatin structure affect transcriptional and posttranscriptional transgene silencing in Arabidopsis. Curr. Biol. 10: 1591-1594
Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant.
15: 431-497
Nakano Y, Steward N, Sekine M, Kusano T, Sano H (2000) A tobacco NtMET1 cDNA encoding a DNA methyltransferase:
molecular characterization and abnormal phenotypes of transgenic tobacco plants. Plant Cell Physiol. 41: 448-457
19
Okano M, Bell DW, Haber DA, Li E (1999) DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo
methylation and mammalian development. Cell 99: 247-257
Okano M, Xie S, Li E (1998) Dnmt2 is not required for de novo and maintenance methylation of viral DNA in embryonic stem
cells. Nucleic Acids Res. 26: 2536-2540
Papa CM, Springer NM, Muszynski MG, Meeley R, Kaeppler SM (2001) Maize chromomethylase Zea methyltransferase2 is
required for CpNpG methylation. Plant Cell 13: 1919-1928
Posfai J, Bhagwat AS, Posfai G, Roberts RJ (1989) Predictive motifs derived from cytosine methyltransferases. Nucleic Acids
Res. 17: 2421-2435
Pradhan S, Cummings M, Roberts RJ, Adams RL (1998) Isolation, characterization and baculovirus-mediated expression of
the cDNA encoding cytosine DNA methyltransferase from Pisum sativum. Nucleic Acids Res. 26: 1214-1222
Pradhan S, Houlston C, Cummings M, Adams RL (1995) CG and CNG methyltransferases in plants. Gene 157: 289-291
Rakyan VK, Preis J, Morgan HD, Whitelaw E (2001) The marks, mechanisms and memory of epigenetic states in mammals.
Biochem. J. 356: 1-10
Ramchandani S, Bhattacharya SK, Cervoni N, Szyf M (1999) DNA methylation is a reversible biological signal. Proc. Natl.
Acad. Sci. U.S.A. 96: 6107-6112
Robertson KD, Keyomarsi K, Gonzales FA, Velicescu M, Jones PA (2000) Differential mRNA expression of the human DNA
methyltransferases (DNMTs) 1, 3a and 3b during the G(0)/G(1) to S phase transition in normal and tumor cells. Nucleic
Acids Res. 28: 2108-2113
Ronemus MJ, Galbiati M, Ticknor C, Chen J, Dellaporta SL (1996) Demethylation-induced developmental pleiotropy in
Arabidopsis. Science 273: 654-657
Sankpal UT, Rao DN (2002) Structure, function, and mechanism of HhaI DNA methyltransferases. Crit. Rev. Biochem. Mol.
Biol. 37: 167-197
Scheid OM, Probst AV, Afsar K, Paszkowski J (2002) Two regulatory levels of transcriptional gene silencing in Arabidopsis.
Proc. Natl. Acad. Sci. U.S.A. 99: 13659-13662
Schweizer P, Pokorny J, Schulze-Lefert P, Dudler R (2000) Double-stranded RNA interferes with gene function at the singlecell level in cereals. Plant J. 24: 895-903
Siroky J, Castiglione MR, Vyskot B (1998) DNA methylation patterns of Melandrium album chromosomes. Chromosome Res.
6: 441-446
Smith NA, Singh SP, Wang MB, Stoutjesdijk PA, Green AG, Waterhouse PM (2000) Total silencing by intron-spliced hairpin
RNAs. Nature 407: 319-320
Steward N, Kusano T, Sano H (2000) Expression of ZmMET1, a gene encoding a DNA methyltransferase from maize, is
associated not only with DNA replication in actively proliferating cells, but also with altered DNA methylation status in
cold-stressed quiescent cells. Nucleic Acids Res. 28: 3250-3259
Stoutjesdijk PA, Singh SP, Liu Q, Hurlstone CJ, Waterhouse PA, Green AG (2002) hpRNA-mediated targeting of the
Arabidopsis FAD2 gene gives highly efficient and stable silencing. Plant Physiol. 129: 1723-1731
Tabara H, Grishok A, Mello CC (1998) RNAi in C. elegans: soaking in the genome sequence. Science 282: 430-431
Theiss G, Schleicher R, Schimpff-Weiland G, Follmann H (1987) DNA methylation in wheat. Purification and properties of
DNA methyltransferase. Eur. J. Biochem. 167: 89-96
Vaucheret H, Fagard M (2001) Transcriptional gene silencing in plants: targets, inducers and regulators. Trends Genet. 17: 2935
Wesley SV, Helliwell CA, Smith NA, Wang MB, Rouse DT, Liu Q, Gooding PS, Singh SP, Abbott D, Stoutjesdijk PA,
Robinson SP, Gleave AP, Green AG, Waterhouse PM (2001) Construct design for efficient, effective and highthroughput gene silencing in plants. Plant J. 27: 581-590
Wolffe AP, Jones PL, Wade PA (1999) DNA demethylation. Proc. Natl. Acad. Sci. U.S.A. 96: 5894-5896
Yang D, Lu H, Erickson JW (2000) Evidence that processed small dsRNAs may mediate sequence-specific mRNA
degradation during RNAi in Drosophila embryos. Curr. Biol. 10: 1191-1200
20
Zentella R, Yamauchi D, Ho TH (2002) Molecular dissection of the gibberellin/abscisic acid signaling pathways by transiently
expressed RNA interference in barley aleurone cells. Plant Cell 14: 2289-22301
Figure legends
Fig. 1. PCR amplification and sequencing of putative MTase-encoding fragments from rice
genomic DNA. A. PCR products R4 and R8 amplified in the presence (+) and absence (-) of rice
(cv. Taipei 309) genomic DNA using primers corresponding to a conserved region of tobacco
NtMET1 (see text). B. Sequence comparison of R4 and R8, later shown to be fragments of
OsMET1-2 and OsMET1-1, respectively. Regions of identity are underlined and intron-exon
boundaries are denoted by arrows.
Fig. 2. Sequencing strategy for OsMET1-1 and identification of OsMET1-1 and OsMET1-2. A.
Proportional diagram of OsMET1-1 showing the coding (black line) and promoter (black arrow)
regions. NPB BAC clone 33-O-20 was digested with BamHI (B), EcoRI (E), HindIII (H), NcoI
(N), PvuII (P) and SacI (S). HindIII and SacI fragments were subcloned as H1 and H2 and S1, S2
and S3 and sequenced. Fragment sizes are indicated in kb. The locations of DNA fragments used
as probes for screening during subcloning are shown as white letters on a black background (PS,
SN, ES, R4, R8, and BH). B. Southern blot analysis of candidate BAC clones. Hybridization
patterns for OsMET1-1 candidates NPB 33-O-20 and NPB 40-B-18 from the NPB library and
TQH 9-A-23 and TQH 4-G-4 from the TQH library were digested with EcoRI (E), HindIII (H),
NcoI (N), and SacI (S), and hybridized with R8 (panels 1, 2) or PS probes (panels 5, 6).
Hybridization patterns for OsMET1-2 candidates TQH 3-F-21 and TQB 10-B-12 were digested
with the same set of enzymes and hybridized with R8 (panel 3), R4 (panel 4) or PS (panel 7)
probes. Asterisks denote weakly hybridizing bands. The blot shown in panel 3 was stripped and
hybridized with R4 probe to yield the profile shown in panel 4. A 1 kb DNA ladder (NEB) was
used as a size marker.
Fig. 3. Identification of intron and exon junctions of OsMET1-1. A. Schematic diagram of
predicted exons (filled boxes; bold numerals) and introns (thin lines) are shown to scale for
OsMET1-1 (top) and OsMET1-2 (bottom); the intron regions confirmed by RT-PCR are indicated
(RE-1, RE-2, RE-3). B. RT-PCR products from rice (T309) total RNA for RE1, RE2 and RE3
(panels 1 to 3, respectively). Lanes N1 and N2 are negative controls in which no template was
added for PCR and RT-PCR, respectively; lanes R are negative controls for PCR in which total
RNA was used as the template. Positive controls for PCR reactions are included in which plasmid
DNA (lanes P) and genomic DNA (lanes G) were used as templates. White arrows indicate
expected bands for PCR products. A 1 kb DNA ladder (NEB) was used as standard marker (lanes
M). Expected sizes (kb) of RT-PCR products are indicated by black arrows. C. 3'RACE analysis
to locate the positions of intron 11, the 3’ UTR and the polyadenylation site of OsMET1-1. Lanes
met1-1 denote products obtained using nested outer (pane 1) and inner (panel 2) primer sets.
21
Negative controls for PCR reactions in which no template was added are shown in lanes N. The
3'RACE product for ß-actin amplified from mouse thymus RNA served as a positive control (lanes
A). White arrows indicate the sizes of 3’RACE products. A 100 bp ladder (NEB) used as standard
marker (lanes M).
Fig. 4. Positions of conserved domains in OsMET1-1 and OsMET1-2 and phylogenetic
relationship with other MTases. A. Proportionate diagram of OsMET1-1 showing conserved
domains: S-stretch, serine-rich region; E-rich region, glutamic acid-rich region; BAH (bromo
adjacent homology domain), black boxes with white dots; IQ motif (calmodulin binding)
isoleucine (I) and glutamine(Q) motif, hatched box; NLS (putative nuclear localization signal),
oval; KG linker (the lysine-glycine repeat, KNKGKG), gray box; I to X, the catalytic domain with
eight highly conserved DNA MTase motifs, black bars. B. Phylogenetic relationships between
eukaryotic and prokaryotic MTases were derived using the AlignX program of the VectorNTI
suite with the following GenBank accessions: maize MET1 (AF063403), rice MET1-1 (OsMET11, AF462029), rice MET1-2 (OsMET1-2, AY230205), arabidopsis MET1(AtMET1,AT5G49160)
and MET2 (MET2-type C-5 DNA MTase,AT4G08990), pea MET1 (putative C-5 DNA MTase,
AF034419), carrot MET1 (carrot C-5 DNA MTase, AF007807) and MET2 (MET2-type C-5 DNA
MTase, AF007808), tobacco MET1 (NtMET1, AB030726), tomato (putative C-5 DNA MTase
AJ002140), rat DNMT1 (AB012214), human DNMT1 (AF180682), Xenopus DNMT1(D78638),
zebra fish DNMT1 (AF483203), chicken DNMT1 (2112268A), Neurospora DIM2 (AF348971),
Ascobolus DNMTase (Z96933), rice MET2a (AC069324), maize MET2a (AF243043) and
Haemophilus haemolyticus M.HhaI (XYH1H1).
Fig. 5. Genomic blot analysis of DNA MTases in rice. Rice (T309) genomic DNA was digested
with EcoRI (E), HindIII (H), NcoI (N), or SacI (S), and hybridized with the R8 (panel 1), R4
(panel 2), PS (panel 3) or ES (panel 4) probes. In panel 1, 1x, 2x and 5x copy reconstructions of a
3.4 kb HindIII fragment (see Fig. 2A) are shown; size markers (kb) corresponding to a 1 kb ladder
(NEB) are shown to the left of each panel.
Fig. 6. Analysis of spatial expression levels of OsMET1-1 and OsMET1-2 in rice (T309). A.
Ribonuclease protection assays of OsMET1-1 and OsMET1-2 expression levels in calli (C), 10 dold leaf (L), 10 d.-old root (R), leaf after flowering (ML), and young inflorescence (I) using either
the sM1 (for OsMET1-1) or sM2 (for OsMET1-2) probes (see text). B. Relative quantitative
analysis of OsMET1-1 and OsMET1-2 expression levels (see Materials and Methods). Transcript
abundance for 18S rRNA was used as a loading control and normalization.
Fig. 7. Reactivation of GUS expression in silenced rice calli. A. Gene constructs used to
debilitate endogenous MTase: (1) control plasmid (pJD7) containing only a GUS expression
cassette that contains an intron from castor bean (intcb); (2) control plasmid (pPT1) containing only
a GFP expression cassette. Panels (3) and (4) show pir-Nt and pir-Ct constructs, containing sense
22
(s) and antisense (as) sequences separated by an intron from pyruvate orthophosphate dikinase
(intpdk) (Wesley et al. 2001) , designed to express double-stranded RNA complementary to
conserved motifs within MTase, fused with the reporter GFP expression cassette. Arrows within
boxes indicate the orientation of coding regions. A schematic structure of the predicted dsRNA
stem with a single-stranded loop generated by the inverted repeat (ir) constructs (see Materials and
Methods) is shown in (5). B. Effect of supertransformation of silenced JDV92-8 calli. Calli in
panels (1) and (5) were bombarded with pJD7; those in (2) and (6) with pPT1; those in (3) and (7)
with pir-Nt, and those in (4) and (8) with pir-Ct. Calli were examined 3 d post-bombardment for
GFP expression (panels 1-4) and were then histochemically stained to show GUS expression
(panels 5-8).
Legends to supplementary material
Supplemental Fig 1. Alignment of OsMET1-1 and OsMET1-2 coding sequences. Identical bases
are denoted by red letters and regions of identity are highlighted in yellow.
Supplemental Fig 2. Alignment of OsMET1-1 and OsMET1-2 showing conserved motifs.
Identical amino acids are highlighted in yellow; similar amino acids are shown in maroon letters
on a gray background. Highly conserved domains are indicated by colored lines and sequences
conserved in rice and other plant MTases are labeled.
Supplemental Fig 3. A comparison of highly conserved catalytic domains in representative DNA
MTases from various organisms. The phylogenetic relationships between diverse MTases were
derived using the AlignX program of the VectorNTI suite using the following GenBank
accessions: maize MET1 (AF063403), rice MET1-1 (OsMET1-1, AF462029), rice MET1-2
(OsMET1-2, AY230205), arabidopsis MET1(AtMET1,AT5G49160) and MET2 (MET2-type C-5
DNA MTase,AT4G08990), pea MET1(putative C-5 DNA MTase, AF034419), carrot
MET1(carrot C-5 DNA MTase, AF007807) and MET2 (MET2-type C-5 DNA MTase,
AF007808), tobacco MET1 (NtMET1, AB030726), tomato (putative C-5 DNA MTase
AJ002140), rat DNMT1 (AB012214), human DNMT1 (AF180682), Xenopus DNMT1(D78638),
zebra fish DNMT1 (AF483203), chicken DNMT1 (2112268A), Neurospora DIM2 (AF348971),
Ascobolus DNMTase (Z96933), rice MET2a (AC069324), maize MET2a (AF243043) and
Haemophilus haemolyticus M.HhaI (XYH1H1).Color coding for amino acid relationships among
sequences (letter color/background color): identical among all sequences, red/yellow; consensus
residue derived from block of similar residues, dark blue/light blue; consensus residue derived
from >50% of residues at a given position, black/green; residue weakly similar to consensus
residue at a given position, black/gray; dissimilar residues, black/white. Motifs conserved in all
MTases are labeled; motif V (grey letter) is methionine in plants but leucine in all other organisms.
23
(A)
kb M
Template
_ +
1
R4 (744 bp)
0.5
R8 (719 bp)
(B)
R8
R4
Fig 1
R4
(A)
S
HE
H
E
E
PS SN
NH H
HP S
E
R8
ES
ES
E HE
S
H2
S1
7.3
1.3
11.3
10.0
TQH 9-A-23
kb
H E N S
H E N
4.0
3.0
2.0
1.5
1.0
3.0
2.0
1.5
1.0
*
*
*
*
R4 probe
TQB 10-B-12
kb H E N S
S
10.0
4.0
3.0
2.0
1.5
10.0
4.0
TQH 3-F-21
TQH 4-G-4
H E N S
kb
*
10.0
4.0
3.0
*
SacI
2.6
R8 probe
R8 probe
H E N S
NcoI
0.8
2.2
R8 probe
H E N S
EcoRI
5.7
7.9
kb
HindIII
3.4
3.0
NPB 40-B-18
S
S3
1.1
NPB 33-O-20
H
H1
S2
(B)
BH
B
*
* *
*
TQH 3-F-21
TQB 10-B-12
H E N S
H E N S
2.0
1.5
1.0
1.0
0.5
0.5
0.5
0.5
(1)
(2)
(3)
PS probe
NPB 33-O-20
kb
H
E N S
PS probe
NPB 40-B-18
H
10.0
E
N
S
TQH 9-A-23
kb
H
E N S
4.0
2.0
1.5
PS probe
TQH 4-G-4
H
E
N
S
TQH 3-F-21
kb
10.0
10.0
4.0
3.0
2.0
3.0
(4)
4.0
3.0
2.0
1.5
1.5
H
*
TQB 10-B-12
E N S
*
H
*
*
*
1.0
0.5
0.5
0.5
(5)
(6)
Fig. 2
N S
*
*
*
1.0
1.0
E
(7)
RE2
(A)
1
kb M
A met1-1
N
RE1
RE3
247
51
(C)
669 bp
110 109 77 88 101 85
2
3
4
5 6 7
kb M N
8
104
99
9
10 11
520 bp
469 bp
513 bp
390
12
0.6
0.5
0.4
A met1-1
0.6
0.5
0.4
sM1
24%
72%
69%
76% 79% 88% 87% 83% 82%
83%
80%
(1)
1
2
3
9
4 5 6 7 8
10 11
sM2
85
(B)
85 90 87 115 101 95
120
kb M N1 P G R N2 RE1
kb M N1 P G R N2 RE2
1.3
1
95 98
kb M N1 P G R N2 RE3
0.5
0.6
1
1
0.5
0.5
0.5
0.7
(1)
0.4
(2)
(3)
Fig 3
0.33
(2)
S-stretch
E-rich
region
IQ
motif
NLS
KG
linker
100
I II IV VI VII VIII
(A)
BAH domain
(B)
IX X
Catalytic
domain
Rice MET1-2
Maize MET1
Rice MET1-1
Carrot MET1
Carrot MET2
Tobacco MET1
Tomato MET1
Pea MET1
Arabidopsis MET1
Arabidopsis MET2
Rat DNMT1
Human DNMT1
Xenopus DNMT1
Chicken DNMT1
Zebra fish DNMT1
Neurospora DIM1
M.HhaI
Ascobolus DNMTase
Rice MET2a
Maize MET2a
Fig 4
R8 probe
1x 2x 5x
H
R4 probe
E
N
S
2x 5x
H
E
PS probe
N
H
S
E
N
ES probe
S
1x 2x 5x
10.0
10.0
10.0
10.0
6.0
5.0
6.0
5.0
6.0
5.0
6.0
5.0
3.0
3.0
3.0
3.0
2.0
2.0
2.0
2.0
1.5
1.5
1.5
1.5
1.0
1.0
1.0
1.0
0.5
0.5
0.5
0.5
(1)
(2)
Fig. 5
(3)
H
(4)
E
N
S
(A)
C
L
R
ML
I
sM1
18S
sM2
18S
(B)
Inflorescence
Mature
leaf
Root
Leaf
Osmet1-1
Osmet1-2
Callus
0
5
10
15
Fig. 6
20
25
(A)
(1)
intcb
35S
gus
nos
(2)
35S
gfp
sNt
nos
asNt
(3)
intpdk
actin
sCt
oct 35S
gfp
nos
oct 35S
gfp
nos
asCt
(4)
actin
intpdk
Inverted repeat
(5)
dsRNA
(B)
(1)
(2)
(3)
(5)
(6)
(7)
Fig. 7
(4)
(8)
LEGENDS TO SUPPLEMENTAL FIGURES
Supplemental Figure 1. Alignment of OsMET1-1 and OsMET1-2 coding sequences. Identical
bases are denoted by red letters and regions of identity are highlighted in yellow.
Supplemental Figure 2. Alignment of OsMET1-1 and OsMET1-2 showing conserved motifs.
Identical amino acids are highlighted in yellow; similar amino acids are shown in maroon letters
on a gray background. Highly conserved domains are indicated by colored lines and sequences
conserved in rice and other plant MTases are labeled.
Supplemental Figure 3. A comparison of highly conserved catalytic domains in representative
DNA MTases from various organisms. The phylogenetic relationships between diverse MTases
were derived using the AlignX program of the VectorNTI suite using the following GenBank
accessions: maize MET1 (AF063403), rice MET1-1 (OsMET1-1, AF462029), rice MET1-2
(OsMET1-2, AY230205), arabidopsis MET1(AtMET1,AT5G49160) and MET2 (MET2-type C5 DNA MTase,AT4G08990), pea MET1(putative C-5 DNA MTase, AF034419), carrot
MET1(carrot C-5 DNA MTase, AF007807) and MET2 (MET2-type C-5 DNA MTase,
AF007808), tobacco MET1 (NtMET1, AB030726), tomato (putative C-5 DNA MTase
AJ002140), rat DNMT1 (AB012214), human DNMT1 (AF180682), Xenopus DNMT1(D78638),
chicken DNMT1 (2112268A), Neurospora DIM2 (AF348971), Ascobolus DNMTase (Z96933),
rice MET2a (AC069324), maize MET2a (AF243043) and Haemophilus haemolyticus M.HhaI
(XYH1H1).Color coding for amino acid relationships among sequences (letter color/background
color): identical among all sequences, red/yellow; consensus residue derived from block of
similar residues, dark blue/light blue; consensus residue derived from >50% of residues at a
given position, black/green; residue weakly similar to consensus residue at a given position,
black/gray; dissimilar residues, black/white. Motifs conserved in all MTases are labeled; motif V
(grey letter) is methionine in plants but leucine in all other organisms.