Characterization of a rice class II metallothionein gene: tissue expression patterns and induction in response to abiotic factors.

Abstract

Data mining the complete rice genome sequences revealed a genomic fragment encoding a characteristic metallothionein (MT) protein, and its full-length cDNA was isolated from rice developing seeds by RT-PCR. This cDNA, designated OsMT-II-1a, contains an open reading frame of 264 bp encoding a protein of 87 amino acid residues. The predicted amino acid sequence was shown to have structural features characteristic of plant class II MT proteins. By sequence analysis of its 5'-flanking region, one putative TATA box, four putative CAAT boxes, and several short sequences homologous to previously reported regulatory cis-elements were identified. Northern blot analysis showed that accumulation of OsMT-II-1a mRNA is specifically abundant in developing seeds and 2-day glumes after pollination, and OsMT-II-1a transcription can markedly be induced by H2O2, paraquat, SNP, ethephon, ABA and SA, but barely by metal ions or other exogenous abiotic factors such as low temperature and PEG. These results coincide with the prediction of existing regulatory cis-elements in its 5'-flanking region. Taken together, the above results suggest that the processes of pollination and seed development might be mediated, at least in part, by expression of the OsMT-II-1a gene that is regulated by several abiotic factors.

Full text

Page 1

Journal of Plant Physiology 162 (2005) 686—696
Characterization of a rice class II metallothionein
gene: Tissue expression patterns and induction in
response to abiotic factors
Gong-Ke Zhou, Yu-Feng Xu, Jin-Yuan Liu?
Laboratory of Molecular Biology and MOE Laboratory of Protein Science, Department of Biological Sciences and
Biotechnology, Tsinghua University, Beijing 100084, PR China
Received 1 September 2004; accepted 9 November 2004
Summary
Data mining the complete rice genome sequences revealed a genomic fragment
encoding a characteristic metallothionein (MT) protein, and its full-length cDNA was
isolated from rice developing seeds by RT-PCR. This cDNA, designated OsMT-II-1a,
contains an open reading frame of 264bp encoding a protein of 87 amino acid
residues. The predicted amino acid sequence was shown to have structural features
characteristic of plant class II MT proteins. By sequence analysis of its 50-flanking
region, one putative TATA box, four putative CAAT boxes, and several short sequences
homologous to previously reported regulatory cis-elements were identified. Northern
blot analysis showed that accumulation of OsMT-II-1a mRNA is specifically abundant in
developing seeds and 2-day glumes after pollination, and OsMT-II-1a transcription can
markedly be induced by H2O2, paraquat, SNP, ethephon, ABA and SA, but barely by
metal ions or other exogenous abiotic factors such as low temperature and PEG. These
results coincide with the prediction of existing regulatory cis-elements in its 50-
flanking region. Taken together, the above results suggest that the processes of
pollination and seed development might be mediated, at least in part, by expression
of the OsMT-II-1a gene that is regulated by several abiotic factors.
& 2005 Elsevier GmbH. All rights reserved.
ARTICLE IN PRESS
www.elsevier.de/jplph
KEYWORDS
Exogenous factor;
Expression pattern;
Metal ion;
Regulatory cis-ele-
ment;
Rice metallothionein
gene
0176-1617/$-see front matter & 2005 Elsevier GmbH. All rights reserved.
doi:10.1016/j.jplph.2004.11.006
Abbreviations: ABA, abscisic acid; MT, metallothionein; PEG, polyethylene glycol; RT-PCR, reverse transcriptase polymerase chain
reaction; SA, salicylic acid; SNP, sodium nitroprussidew
?Corresponding author. Tel.: +861062783845; fax: +861062772243.
E-mail address: liujy@mail.tsinghua.edu.cn (J.-Y. Liu).

Page 2

Introduction
Metallothioneins (MTs) are defined as a class of
evolutionally conserved, low molecular-weight,
cysteine-rich, metal-binding proteins. Since MTs
were firstly purified from horse kidney as Cd-
binding proteins in 1956, they have been widely
found in diverse organisms including fungi, plants,
mammals and cyanobacteria (Yu et al., 1998;
Cobbett and Goldsbrough, 2002). Unlike in animals,
only EcMT from wheat germ and MT1, MT2 and MT3
from Arabidopsis (Lane et al., 1987; Murphy et al.,
1997) have been purified in plants. In most cases,
however, the presence of proteins has been largely
inferred from DNA sequence, so the term MT-like
proteins is preferred for plant MTs. Plant MT-like
proteins are divided into three major classes (class
I, II and III), which are distinguishable on the basis
of the distribution of cysteine residues in their
sequence (Robinson et al., 1993). Class I MTs are
monomers with two cysteine-rich clusters sepa-
rated by a central cysteine free spacer. Most of the
deduced products of MT genes identified in both
monocots and dicots belong to class I. Based on the
arrangement of cysteine residues, plant class I MT
proteins have been subdivided into several types.
Class II MTs are translational monomers in which
cysteine residues are scattered throughout the
entire sequence, lacking the internal spacer. This
class of protein is exemplified by the wheat EcMT
(formerly cysteine-labeled metallothionein) pro-
tein, the first characterized plant MT protein
(Kawashima et al., 1992). Class III MTs are non-
translationally synthesized polypeptides based on
repeating units of g-glutamlcysteine (g-Glu-Cys),
and are broadly found in the plant kingdom
including algae and certain fungi (Rauser, 1999,
Cobbett, 2000). On the other hand, Cobbett and
Goldsbrough (2002) proposed a classification sys-
tem which was able to place the known plant MT
proteins into four types based on amino acid
sequence. Based on their nomenclature, the class
II MT-like protein (OsMT-II) described in this
paper may be classified as type 4 MT. Recently,
the number of plant MT genes discovered has
increased dramatically, and many have not been
included in these four types. Because the former
nomenclature of three classes can accommodate
for an increasing number of plant MTs to be
subdivided into different types, we tend to regard
the OsMT-like protein described in this paper as
class II MT.
Plant MTs have been suggested to play important
roles in maintaining the homoeostasis of essential
transition metals, detoxification of toxic metals,
and protecting against intercellular oxidative stress
(Palmiter, 1998; Adams et al., 2002; Coyle et al.,
2002; Wong et al., 2004; Akashi et al., 2004).
Moreover, the diverse expression patterns of MT
genes suggest that MT-like protein isoforms may
differ in sequence, as well as in the functions that
these perform in specific tissues (Cobbett and
Goldsbrough, 2002). A number of investigations
demonstrated these to have distinct but over-
lapping functions in the multigene family in single
species (Zhou and Goldsbrough, 1995; Giritch
et al., 1998; Garcia-Hernandez et al., 1998; Guo
et al., 2003). Recently, several studies indicated
that plant MT-like proteins, especially the char-
acterized plant class II MT-like proteins, are also
involved in important developmental processes
(Kawashima et al., 1992; Reynolds and Crawford,
1996; Chatthai et al., 1997; Guo et al., 2003).
Kawashima et al. (1992) showed that the 50-flanking
genomic DNA for the wheat EcMT gene contains a
core sequence in the 50-flanking region known
to be an ABA-responsive element (ABARE), but
doesnotcontain metal-responsive
seen in animal MT genes. Consistent with this,
accumulation of the EcMT mRNA is strongly induced
during development of immature embryos but not
during germination of mature wheat embryos
unless ABA is added to the germination medium.
Also, supplementation of the medium with zinc
does not induce accumulation of EcMT mRNA in
germinationembryos.Reynolds
(1996) further showed that ABA biosynthesis is
accompanied by increased expression of the EcMT
gene transcript concommitant with the differentia-
tion of pollen embryoids in wheat anther cultures,
and suggested that the EcMT gene plays an
important role in pollen embryogenesis. To date,
however, it is not known if different class II MT
genes have specific functions in different organs, or
at different developmental stages besides embry-
ogenesis. Also their response to the different
exogenous factors except zinc remain unclear.
Therefore,wewere very
relationship between expression of plant class II
MT genes and various developmental or environ-
mental signals.
In this report, we described and characterized
the gene of a class II MT-like protein from rice, and
investigated its expression in response to various
exogenous factors. Moreover, an analysis of the
spatial expression pattern revealed the unexpected
finding that this novel gene was highly expressed
in developing seeds and 2-day glumes after
pollination.Thepossible
regulated expression of the gene in response to
complex developmental and environmental cues is
discussed.
elements
andCrawford
interestedinthe
significanceofthe
ARTICLE IN PRESS
Characterization of a rice class II metallothionein gene687

Page 3

Materials and methods
Database searching and sequence analysis
The sequence data used in this study were
collected from a query
database using the EcMT gene sequence (P30570).
DNA sequences homologous to EcMT were then
used to screenthe RGP
rgp.dna.affrc.go.jp) again to confirm the target
gene and its location. The deduced amino acid
sequences of ORFs were initially aligned using the
program ClustalX version 1.8 with default gap
penalties (Thompson et al., 1997). A phylogram
tree was constructed from amino acid sequences of
class II MT-like proteins by neighbor-joining algo-
rithms and then displayed by TreeView (V 1.6.6)
(Page, 1996).
Intron delimitation of target genes within the
genomic DNA sequence was made by comparing the
cDNA sequence and its genomic DNA sequences
derived from the available RGP database. About
1500bp the upstream of the translation initiation
codon ATG were chosen as the upstream regulatory
region for cis-regulatory element analysis. Se-
quences of the known regulatory elements were
used to search the PLACE database (http://
www.dna.affrc.go.jp/htdocs/PLACE/) (Higo et al.,
1999) for regulatory cis-element analysis of the
target gene.
searchin the NCBI
database(http://
Plant materials and treatments with
exogenous factors
Rice (Oryza sativa L. cv. Zhonghua 10) were
grown in a greenhouse at 261C. Samples of roots,
stems,leaves, sheaths
before pollination, 2-day glumes after pollination
and 15-day-old seeds were harvested from mature
plants. For various stress treatments, roots of 10-
day-old seedlings were submerged separately in
aqueous solutions containing
glycol-8000, H2O2, paraquat (generators of super-
oxide anions), ethephon (an ethylene-releasing
compound), ABA, salicylic acid, sodium nitroprus-
side (generators of NO) and in different concentra-
tion aqueous solutions of FeCl3, CuSO4, CdSO4,
ZnSO4, AlCl3 and Pb(Ac)2 for 24h, respectively.
Moreover, 10-day-old seedlings were also treated
separately with heat shock (421C) for 2h, low
temperature (41C) for 24h and wounding for
45min. All these prepared materials were immedi-
ately frozen in liquid nitrogen and stored at ?701C
until use.
andrachises, glumes
polyethylene
Isolation and sequencing of the cDNA
Total RNA was extracted from various rice tissues
using RNeasy Plant Mini Kit (Qiagen), in accordance
with the instructions provided by the manufacturer.
In order to isolate the cDNA, total RNA from the
developing seeds of rice was reverse transcribed
with MMLV reverse transcriptase (BioLabs) and
oligo(dT) following the enzyme manufacturer’s
instructions. The resulting cDNA was used as a
template for PCR amplification using two specific
primers: 50-TCGACGACCATGGGGTGCGA-30(sense)
and 50-GTCTCGTTGGGCATTTTTGGG-30(antisense).
Thermocycling was performed at 951C for 40s,
601C for 30s, and 721C for 50s, for 30 cycles, using
Taq-DNA polymerase (Promega). The amplified
fragments were isolated, gel-purified using the
GeneClean III kit (Bio 101), cloned into a pGEM-T
easy plasmid vector (Promega), and then the
inserts were sequenced entirely on both strands
using BigDyeTM Terminator Cycle Sequencing Ready
Reaction Kit(PerkinElmer,
PRISMTM 377 DNA Sequencer.
German)on ABI
Northern blotting
Total RNA extracted from various tissues of rice
was subjected to electrophoresis in 1.2% formalde-
hyde (2.2M) agarose gels in 1?MOPS buffer
(Sambrook et al., 1989), and blotted to a Nylon
membrane (Hybond N+, Amersham) using the
manufacturer’s recommended techniques. Hybridi-
zations were performed by
probes with random labeled probes (Prime-a-Gene
random labeling system, Promega), overnight at
651C, in a hybridization buffer containing 5?SSC,
5?Denhardt’s solution, 0.5% SDS and 100mg/mL
heat-denatured salmon sperm. After washing the
membrane once in 2?SSC and 0.1% SDS at room
temperature for 10min, and twice in 1?SSC and
0.1% SDS at 651C for 30min, the filter was exposed
to X-ray film.
32P-labeled cDNA
Results and discussion
Molecular cloning and sequence analysis of
OsMT-II-1a
The cysteine-rich signature sequence of EcMT
from wheat (Kawashima et al., 1992) was used to
screen the RGP database. Finally, a genomic DNA
sequence encoding an open reading frame of 87
amino acid residues was identified. The cDNA,
designated as OsMT-II-1a, was isolated from the
ARTICLE IN PRESS
G.-K. Zhou et al.688

Page 4

developing seeds of rice by RT-PCR as described in
materials and methods, with an open reading frame
of 264bp (GenBank accession AAS78805). However,
unlike animal MT multigene families (Sato and
Kondoh, 2002), the OsMT-II-1a gene is single-copy
and located in chromosome 10.
The coding region of OsMT-II-1a, when trans-
lated, is demonstrated to contain 17 cysteine
residues arranged into three groups of 6, 6 and 5
cysteines, which are separated by two interdomain
regions of 13 and 15 cysteine-free amino acid
residues, respectively (Fig. 1). The abundance
(about 15%) and the distribution of cysteines in
OsMT-II-1a were shown to have structural features
characteristic of the class II MT-like proteins
(Kawashima et al., 1992; Reynolds, 2000; Guo et
al., 2003). An alignment of the deduced OsMT-II-1a
protein with all known class II MT-like proteins is
shown in Fig. 1. It has high homology with the plant
class II MT-like proteins, with the overall sequence
similarity varying substantially from 82% to 49%. As
shown in Fig. 1A and B, we discovered that, unlike
ARTICLE IN PRESS
Figure 1. An alignment (A) and phylogenetic tree (B) were constructed from the deduced amino acid sequence of
OsMT-II-1a together with the sequences of other class II MT-like proteins. (A) Sequences were aligned with the ClustalX
program. Amino acid residues identical to OsMT-II-1a are indicated by shaded boxes. Dashes indicate gaps in the
sequence to allow for maximal alignment. The conserved cysteine patterns are indicated under each sequence
alignment and cysteine residues are in bold letters. ‘x’ corresponds to any amino acid besides cysteine. (B) Phylogenetic
tree of all known class II MT-like proteins from rice and other plant species. EMBL bank accession numbers are indicated
in parentheses.
Characterization of a rice class II metallothionein gene 689

Page 5

class I MT-like genes with several members, the
members of class II MT-like genes are no more than
2 in a single species, and the arrangement patterns
of cysteines are the same in each member. These
results imply that this class of proteins might be
conserved in structure and may play a special role
in plants. Compared to those proteins from dicots,
class II proteins from monocots lack 8–13 amino
acids in the N-terminal domain before the first
cysteine residue, but have a few additional amino
acids in the C-terminal domain (Fig. 1), thus
maintaining the similar size of class II proteins
between monocots and dicots. A phylogenetic tree
of all known plant class II MT-like proteins was
constructed, which showed two distinct groups
corresponding to monocots and dicots (Fig. 2).
Thus, the differences in the structures of class II
MT-like proteins may suggest specific functions for
monocots and dicots.
Genomic organization of OsMT-II-1a in rice
Cis-acting DNA elements (in the 50-flanking
region) upstream the sequence encoding MT RNA
respond to multiple signals that regulate both low
level, basal MT mRNA and protein production, and
inducible MT expression in response to a variety of
developmental and environmental signals (Kagi,
1991; Ghoshal and Jacob, 2001; Haq et al., 2003).
The 5’-flanking region of OsMT-II-1a, including a
region of about 1500bp upstream of the translation
initiation codon ATG was derived from the RGP
database. Judging from the consensus sequences
YYYAYYA (Y is pyrimidine) for a number of plant
genes (Joshi, 1987), the predicted transcription
initiation site of OsMT-II-1a could be located at
base 59 upstream of ATG, and was designated as +1
(Fig. 2). The sequence around the transcription
initiation site was TCTACCA (?3 to +4). One putative
TATA box was identified between bases ?29 and ?25.
Four putative CAAT boxes were located between
bases ?300 to ?296, ?305 to ?301, ?320 to ?316
and ?1130 to ?1126, respectively.
As shown in Table 1 and Fig. 2, several DNA motifs
were identified in the promoter of OsMT-II-1a that
are homologous to various previously reported
regulated elements, which might be important in
the transcriptional regulation of this gene. Two
putative ABA responsive elements (ABARE) were
found (one in a forward orientation and the other in
reverse), which were also found in the wheat Em
gene (Marcotte et al., 1989) and rice rab21 gene
(Busk and Pages, 1998). One putative ethylene-
responsive element identified was the GCC box,
GCCGCC, which was found in many ethylene
inducible pathogenesis-related genes (Ohme and
Shinshi, 1995). An antioxidant response element
(ARE), four low-temperature responsive elements
(LTRE),fourABARE-like
CCAAT boxes were also found in the promoter of
OsMT-II-1a. In a word, the presence of these
homologous sequences may be related with the
effects of various stress treatments on the expres-
sion of OsMT-II-1a. Also, unlike animal and most
plant class I MT genes containing the metal-
responsive element TGCRCNC (in which N is
not A, and R is A or G) (Stuart et al., 1985) and/
or copper-responsive element CTGCCA (Quinn and
Merchant, 1995), the promoter of OsMT-II-1a
did not contain any known metal-responsive ele-
ments or metal regulatory motifs. These findings
suggestedthatthe expression
gene might not be modulated by metal ions.
Additionally, we also found many cis-elements
related with embryo-, pollen- and endosperm-
specificgenetranscription
box, ACGT motif, AGAAA motif and (CA)n element.
sequencesandfour
of
OsMT-II-1a
such aslegumin
ARTICLE IN PRESS
Figure. 2. A schematic diagram of the OsMT-II-1a gene structure. The relative positions of homologous sequences are
labeled with ellipses in the figure. The ellipses shown are not proportional to the sizes of the respective elements.
Filled ellipses indicate homologous sequences to stress-responsive elements and empty ellipses represent a TATA box.
Striped ellipses indicate putative regulatory sequences which are related with pollen-, embryo- or endosperm-specific
gene expression. The OsMT-II-1a gene structure is represented by filled box exons and the lines between the filled boxes
correspond to intron. The abbreviations of cis-elements are shown in Table 1.
G.-K. Zhou et al. 690