Content uploaded by Won-Yong Song
Author content
All content in this area was uploaded by Won-Yong Song on Mar 05, 2015
Content may be subject to copyright.
Arabidopsis metallothioneins 2a and 3 enhance resistance to cadmium
when expressed in Vicia faba guard cells
Joohyun Lee
1,2,
, Donghwan Shim
1,
, Won-Yong Song
1,3
, Inhwan Hwang
4
and
Youngsook Lee
1,
*
1
National Research Laboratory of Phytoremediation, Division of Molecular Life Sciences, Pohang University
of Science and Technology, Pohang 790-784, Korea (*author for correspondence; e-mail ylee@postech.
ac.kr);
2
Current address: Biological Science Department, Dartmouth College, Hanover NH 03755-3576,
USA;
3
Current address: Division of Applied Biology and Chemistry, College of Agriculture and Life Sciences,
Chungnam National University, Daejeon 305-764, Korea;
4
Center for Plant Intracellular Trafficking, Pohang
University of Science and Technology, Pohang 790-784, Korea;
These authors contributed equally to the work
Received 18 December 2003; accepted in revised form 23 April 2004
Key words: Arabidopsis thaliana, cadmium resistance, localization, metallothionein, reactive oxygen spe-
cies, Vicia faba guard cells
Abstract
The Arabidopsis metallothionein genes AtMT1 and AtMT2 confer Cd(II) resistance to Cd(II)-sensitive
yeast, but it has not been directly shown whether they or other metallothioneins provide the same pro-
tection to plants. We tested whether AtMT2a and AtMT3 can confer Cd(II) resistance to plant cells by
introducing GFP- or RFP-fused forms into guard cells of Vicia faba by biolistic bombardm ent. AtMT2a
and AtMT3 protected guard cell chloroplasts from degradation upon exposure to Cd(II), an effect that was
confirmed using an FDA assay to test the viability of the exposed guard cells. AtMT2a- and AtMT3-GFP
were localized in the cytoplasm both before and after treatment of V. faba guard cells or Arabidopsis
protoplasts with Cd(II), and the levels of reactive oxygen species were lower in transformed guard cells than
in non-transformed cells after Cd(II)-treatment. These results suggest that the Cd(II)-detoxification
mechanism of AtMT2a and AtMT3 may not include sequestration into vacuoles or other organelles, but
does involve reduction of the level of reactive oxygen species in Cd(II)-treated cells. Increased expression of
AtMT2a and AtMT3 was observed in Arabidopsis seedlings exposed to Cd(II). Together, these data
support a role for the metallothioneins AtMT2a and AtMT3 in Cd(II) resistance in intact plant cells.
Introduction
Metallothioneins (MTs) are proteins rich in cys-
teine residues that bind heavy metals and confer
resistance to Cd(II) and Zn(II) in mammalian cells
(Palmiter, 1998). Plants also have many MT genes
and express some of these genes at high levels
(Cobbett and Goldsbrough, 2002). The MT1 and
MT2 genes from Arabidopsis thaliana (AtMT1 and
AtMT2) enhance resistance to Cu(II) and Cd(II)
when expressed in yeast cells (Zhou and Golds-
brough, 1994). Exposure of Arabidopsis to various
heavy metals, including Cu(II) and Cd(II), in-
creases the expression of AtMT1 and AtMT2
messages (Murphy and Taiz, 1995). Although
these data suggest that plant MT genes play an
important role in resi stance to Cu(II) and Cd(II), it
has not been tested whether plant MTs confer
resistance to heavy metals in intact plant cells.
Several lines of indirect evidence suggest that MTs
are mainly involved in Cu(II) resistance and in
copper homeostasis: the increase in AtMT2
Plant Molecular Biology 54: 805–815, 2004.
2004 Kluwer Academic Publishers. Printed in the Netherlands.
805
mRNA is higher in plants treated with Cu(II) than
in those treated with any other heavy metal (Zhou
and Goldsbrough, 1994; Murphy and Taiz, 1995),
AtMT1 and AtMT2 protein levels increase upon
exposure to Cu(II) in tissue-specific manners
(Murphy et al., 1997), and MT gene expression
increases in senescing leaves together with a set of
genes involved in copper homeostasis (Cobbett
and Goldsbrough, 2002).
The availability of knockout plants has greatly
facilitated the identification of the proteins in-
volved in resistance to individual heavy metals.
For example, phytochelatin synthase knockout
plants are highly sensitive to Cd(II); based on this,
Cd(II) resistance in plants is attributed mainly to
phytochelatins (Howden et al., 1995; Cobbett et
al., 1998). However, the many MT genes in plants
most likely have overlapping functions (Cobbett
and Goldsbrough, 2002), which precludes the use
of available knockout plants for the identification
of their functions. Therefore, the roles of plant
MTs in heavy metal resistance remain less clear,
and a method that could identify the contributions
of individual MTs to the resistance to particular
heavy metals would be highly valuable.
Heavy metals in the cytoplasm can be detoxi-
fied by several different pathways (Hall, 2002).
They can be chelated by various ligands which
often have many cysteine residues; they can
be sequestered into the vacuole; or they can be
transported out to the apoplast. Heavy metal
resistance of plant cells may also include effi-
cient repair of heavy metal-induced damage.
Cd(II) has been repo rted to be sequestered into
vacuoles as a step of detoxification (Wink, 1993).
For example, Cd(II) bound to phytochelatin has
been found in vacuoles of tobacco (Vogeli-Lange
and Wagner, 1990), and Cd(II) is transported into
vacuoles by a Cd
2+
/H
+
antiporter (Hirschi et al. ,
2000). However, the sub-cellular localization of
plant MTs has not been reported, partly due to
their instability in the presence of oxygen, which
limits our understanding of the MT mechanism of
action.
Here we confirm that AtMT2a confers strong
resistance to Cd(II) when expressed in yeast, and
in addition, show that AtMT2a and AtMT3 en-
hance Cd(II) resistance when overexpressed in
guard cells of Vicia faba. Further, we show that
these MTs protect plant cells from Cd(II)-toxicity
by reducing the levels of reactive oxygen species
(ROS) without apparent translocation from the
cytoplasm.
Materials and methods
Screening of Cd(II)-resistance genes
A ycf1 mutant of Saccharomyces cerevisiae,
DTY167 (MATa ura3 leu2 his3 trp3 lys2 suc2
ycf::hisG) was used to screen Cd(II)-resistance
genes from an Arabidopsis leaf cDNA library
cloned in pFL61 (Piao et al., 1999). The library
was introduced into DTY167 and transformants
were selected on SG plates lacking uracil (SG-ura)
containing 30–100 lM CdCl
2
. The plasmids from
the surviving yeasts were rescued in E. coli, ex-
tracted, and then sequenced. The sequences of
AtMT2a and AtMT3 genes were identical to those
published (GenBank accession numbers S57861
and AF013959, respectively).
Heavy metal resistance tests of yeast strains
The pYES2/NT C vector with or without the
AtMT2a or AtMT3 genes was introduced into
DTY167 and its isogenic wild type, DTY165. The
transformants were spotted onto CdCl
2
-containing
half strength (1/2) SG-ura plates and grown for 3–
5 days.
Plant material and growth conditions
Arabidopsis wild-type seeds (Arabidopsis thaliana
ecotype Columbia 0) were surface sterilized, placed
in the dark at 4 C for 2 days, and then sown on
plates of 1/2 Murashige and Skoog with 1.5%
sucrose (Murashige and Skoog, 1962) at 18 C.
Vicia faba plants were grown in cycles of 16 h light
and 8 h dark at 22 ± 2 C in a greenhouse.
Young mature leaves from 3- to 4-week-old plants
were used in all experiments.
Northern analysis of Cd(II)-induced MT gene
expression
Arabidopsis plants were grown vertically on 1/2
MS-agar plates supplemented with 1.5% sucrose
for 2 weeks. The plates were then placed hori-
zontally, and the roots were exposed to 10 ml of
10, 30, 50 lM CdCl
2
solution for 24 h. Subsequent
806
RNA preparation and Northern hybridization
were carried out as described (Sambrook and
Russell, 2001).
Fluorescent gene constructs of AtMTs
AtMT2a and AtMT3 were amplified by PCR and
inserted into the BamHI single site located at the 5¢
termini of GFP and RFP genes in the GFP and
RFP vectors. The primers used were 5¢GGATC-
CATGTCTTGCTGTGGAGGAAAC 3¢ and 5¢G
GATCCTCTTGCGGTGCAAGGAT CAC3¢ for
the AtMT2a:GFP construct, and 5 ¢GGATCCAT
GTCAAGCAACTGCGGAAG3¢ and 5¢GGATC
CTGTTGGGGCAGCAAGTGCAG3¢ for the
AtMT3:GFP construct. For the AtMT2a:RFP
construct, 5¢GGATCCATGTCTTGCTGTGGA
GGAAAC3¢ and 5¢GGATCCTCTTGCGGTG C
AATGGATCAC3 ¢ primers were used, and 5¢GG
ATCCATGTCAAGCAACTGCGGAAG3¢, and
5¢GGATCCTGTTGGGGCAGCAAT GTGCAG
3¢ were used for AtMT3:RFP. DNA manipulations
were performed according to standard methods
(Sambrook and Russell, 2001). The fidelity of all
constructs was confirmed by sequencing.
Transient expression of genes in V. faba and in
Arabidopsis cells
AtMTs:GFP and cytosolic RFP were co-expressed
in Arabidopsis protoplasts and V. faba guard cells
using PEG transformation (Jin et al., 2001) and
bombardment (Jung et al., 2002), respectively. For
PEG transformation, Arabidopsis protoplasts
were isolated from seedlings grown in plates of 1/2
Murashige and Skoog with 1.5% sucrose. For bi-
olistic bombardment, young mature leaves from 3-
to 4-week-old V. faba plants were used.
Viability tests in V. faba guard cells
To test the viability of the transformed cells in the
presence of Cd(II), V. faba guard cells were
transformed with AtMTs:RFP by using the bom-
bardment technique. The bombarded leaves were
kept under darkness on wet filter paper for 2–
3 days and then floated on 50 mM KCl, 10 mM
Mes–KOH (pH 6.15) bathing medium supple-
mented with 10 lM CdCl
2
under darkness for 1 h.
The abaxial epidermis was then peeled from the
leaves and the number of chloroplasts in guard
cells was counted using an Axioplan fluorescence
microscope. In addition, fluorescein diacetate
(FDA) was used as a viability probe at a final
concentration of 50 lg/ml. We quantified the
fluorescence level in guard cells using Adobe
Photoshop 5.5 software as previously described
(Park et al., 2003). In short, areas occupied by each
transformed and non-transformed guard cell pair
were delineated, the number of pixels and their
green fluorescence intensity in the regions were
measured, and then the mean value of green
fluorescence of the region was obtained using the
software.
Localization and translocation assay for AtMTs
To test localization, expression of GFP- or RFP-
fused protein was monitored after 20 h using an
Axioplan fluorescence microscope (Zeiss,
Oberkochen, Germany). The filter sets used were
Filter set 44 (excitation, BP 455–495; beamsplitter,
FT500; emission, BP 505–555) for green fluores-
cent proteins and Filter set 20 (excitation, BP 540–
552; beamsplitter, FT560; emission, BP 575–640)
for red fluorescent proteins (Zeiss). To test whe-
ther AtMT proteins translocate in response to
Cd(II) in plants, cells were obs erved at room
temperature before and after a 1 h incubation in
incubation solution (50 mM KCl and 10 mM
MES, pH 6.2) containing 10 lM CdCl
2
. Images
were captured with a cooled charge-coupled device
camera.
ROS assay
V. faba leaves bombarded with AtMTs:RFP wer e
treated with 10 lM CdCl
2
for 1 h as described
previously for the viability assay. Epidermal strips
of V. faba were peeled and floated for 10 min on a
50 lM dichlorodihydrofluorescein diacetate
(H
2
DCF-DA) solution containing 1:100 diluted
10% (w/w) paraphenylene diamine (an anti-fading
agent). Then, they were incubated for 30 min to
remove ROS generated by the stripping stress. The
green fluorescence was detected in transformed
and non-transformed guard cells using a fluores-
cence microscope, and the mean intensity level was
quantified using Adobe Photoshop 5.5 software as
described above.
807
Results
Expression of AtMT2a and AtMT3 enhances
Cd(II) resistance in Saccharomyces cerevisiae
To select genes that confer Cd( II) resistance, an
Arabidopsis cDNA library was introduced into the
DTY167 strain of Saccharomyces cerevisiae which
is hypersensitive to Cd(II) due to a deletion of
YCF1, a vacuolar sequester of Cd(II)-glutathione
complexes. Transformed cells were grown on
Cd(II)-containing 1/2 SG agar medium for 4 days.
Surviving colonies were harvested, the plas-
mids were isolated from the colonies and amplified
in E. coli. The plasmids were introduced again into
the DTY 167 cells and resistance tests were re-
peated. The plasmids isolated from the surviving
colonies contained cDNAs of AtMT2a and
AtMT3, phytochelatin synthase, and some un-
known genes.
To confirm that AtMT2a and AtMT3 confer
Cd(II)-resistance, we spotted four yeast strains on
1/2 SG medium containing CdCl
2
: DTY167
transformed with AtMT2a, AtMT3, and the empty
vector, and DTY 165, the isogenic wild type of
DTY167, transformed with the empty vector.
DTY167 yeast expressing AtMT2a and AtMT3
grew better than either strain transformed with the
empty vector (Figure 1, right).
AtMT2a and AtMT3 are induced by Cd(II)
treatment in Arabidopsis
If AtMT2a and AtMT3 are involved in Cd resis-
tance in Arabidopsis, their expression levels might
increase upon Cd treatment. To test this possibil-
ity, we performed Northern blot analysis using
RNA from 2-week-old whole seedlings. Treat-
ments with 10, 30 and 50 lM CdCl
2
for 24 h
caused upregulation of AtMT2a and ATMT3
expression (Figure 2). The level of expression of
the AtMT2a gene seemed to reach saturation at
10 lM Cd(II) because it did not increase any fur-
ther at higher concentrations of Cd(II), whereas
the level of AtMT3 continued to increase with
increasing Cd(II) concentra tion. In short, both
MT genes were induced by Cd(II), but AtMT3 was
induced more strongly than AtMT2a.
Expression of AtMT2a or AtMT3 enhances
Cd(II)-resistance in Vicia faba guard cells
The resi stance of transgenic plants to heavy metals
is commonly tested by generating stable lines of
transformants and growing them in medium con-
taining the heavy metal. However, this procedure
is time-consuming. For a fast and simple test for
genes that confer Cd(II)-resistance to plant cells,
we used V. faba guard cells transiently trans-
formed by biolistic bombardment.
To test the role of MT genes in resistance to
Cd(II), we bombarded V. faba guard cells with
AtMT2a:GFP and AtMT3:GFP. The expression
of the introduced genes was detected as green
fluorescence in the cells at about 14 h after bom-
bardment (Figur e 3B and D), at which time the
cells were treated with 10 lM CdCl
2
for 1 h. We
Figure 1. Expression of AtMT2a and AtMT3 strongly en-
hances Cd(II)-resistance of YCF1-null mutant and wild-type
yeast. AtMT2a or AtMT3 was inserted into a pFL61 vector and
transformed into DTY167 (YCF1-null) and DTY165 (wild
type) yeast strains. Yeast cells were grown on 1/2 SG media
lacking uracil with or without 20 lM CdCl
2
at 30 C for
3 days. WT; wild type, Dycf1; YCF1-null mutant, EV; empty
vector only.
Figure 2. Northern blot analysis of AtMT2a and AtMT3
expression in whole seedlings of Arabidopsis. Ethidium bro-
mide staining of rRNA was used as a loading control.
808
selected guard cell pairs of which only one was
transformed and the other was not, and comp ared
the responses of the two cells of each stoma. Since
a pair of guard cells is produced from division of a
single guard mother cell, the neighboring cell
provides a good control. While most of the cells
expressing AtMT2a (Figure 3A and B, right) and
AtMT3 (Figure 2C and D, right) remained intact,
many non-transformed neighboring cells lost their
normal cellular organization (Figure 3A and C,
left). Most strikingly, more chloroplasts remained
intact in AtMT2a- or AtMT3-expressing cells than
in non-tr ansformed neighboring guard cells.
Guard cells transformed with free GFP (Fig-
ure 3E, right) were not any more resistant to
Cd(II) than their non-transformed neighbors
(Figure 3E, left). Cells transformed with AtMT3
had an average of 9.4 chloroplasts, whereas non-
transformed neighboring cells had an average of
6.4 chloroplasts (Figure 3G). Thus, overexpressed
AtMT3 conferred Cd(II)-resistance to V. faba
guard cells. Experiments with guard cells express-
ing AtMT2a showed a similar protective effect on
chloroplasts against Cd(II)-toxicity (data not
shown).
To confirm the AtMT3-mediated Cd(II)-resis-
tance, fluorescein diacetate (FDA) was used as a
viability probe. The intensity of the green fluores-
cence of FDA represents the viability of plant cells
(Widholm, 1972); stronger intensity indicates a
healthier cell. To detect cells expressing AtMT
genes under a fluorescent microscope, AtM T genes
were fused to the gene for red fluorescent protein
(RFP) (Jach et al., 2001) and biolistically intro-
duced into guard cells. Before Cd(II) treatment,
normal guard cells showed strong green FDA
fluorescence (Figure 4A and B). However, when
Cd(II) was applied, non-transformed guard cells,
detectable by the absence of red fluorescence, had
weak (Figure 4D, left) or no (Figure 4G, left)
green FDA fluorescence. In contrast, guard cells
transformed with AtMT3-RFP, identified by the
presence of red fluorescence (Figure 4E and H),
showed strong green FDA fluorescence even after
Cd(II)-treatment (Figure 4D and G, right). Gu ard
cells transformed with free RFP had FDA fluo-
rescence levels similar to those of their non-trans-
formed neighbor cells in the presence of Cd(II)
(Figure 4J and K).
We quantified the FDA fluorescence level in
guard cells using Adobe Photoshop 5.5 software.
Figure 3. Expression of AtMT2a:GFP and AtMT3:GFP pro-
tects chloroplasts in Vicia faba guard cells treated with 10 lM
CdCl
2
for 1 h. (A, C, E) Bright field images. (B, D, F) Fluo-
rescent images of AtMT2a-GFP (B), AtMT3-GFP (D), and
free GFP (F), from the cells shown in A, C, and E, respectively.
The guard cells on the right express GFP-fused or free GFP
proteins, as detectable by their green fluorescence. (G) Number
of chloroplasts in the non-treated control, and Cd-treated
AtMT3:GFP-transformed and their non-transformed neighbor
guard cells (n ¼ 100). Error bars represent SE. Scale
bar ¼ 10 lm.
809
Transformed cells had a mean fluorescence inten-
sity of 68.2 (arbitrary unit) and the non-trans-
formed guard cells had a mean fluorescence
intensity of 39.5 (Figure 4L). This result confirmed
that overexpressed AtMT3 enhanced Cd(II)-resis-
tance in V. faba guard cells. AtMT2a demonstrated
a sim ilar protective effect against Cd(II)-toxicity in
guard cells (data not shown).
AtMT2a and AtMT3 are localized in the cytoplasm
of transformed V. faba guard cells
To understand the mechanism by which AtMTs
protect against Cd(II)-toxicity, we investigated the
localization of AtMT2a and AtMT3 in V. faba
guard cells. When AtMT2a and AtMT3 fused with
GFP under the 35S promo ter were introduced into
V. faba guard cells, the green fluorescence was co-
localized with free RFP (Figure 5), which we have
previously observed to localize in the cytoplasm of
plant cells (data not shown). The same experiment
with guard cells expressing AtMT2a showed sim-
ilar cytoplasmic localization of AtM T2a (data not
shown).
To test MT localization in other types of plant
cells, Arabidopsis protopl asts were transformed
with AtMT3:GFP. The green fluorescence was
found to localize in the cytosol, but not in vacuoles
or chloroplasts in mesophyll cells (Figure S1B) or
in all other cell types observed (data not shown).
Figure 4. Viability tests with FDA of Vicia faba guard cells transformed with AtMT3-RFP. (A, B) Bright field (A) and fluorescent (B)
images of a pair of non-transformed guard cells before Cd(II) treatment; (C–K) images of guard cells after 1 h of treatment with 10 lM
CdCl
2
; (C, F, I) bright field images; (D, G, J) FDA fluorescence images of cells shown in C, F, and I, respectively; (E, H, K) RFP
fluorescence images of cells shown in C, F, and I, respectively. The guard cells on the right express AtMT3-RFP (E, H) or free RFP
(K), as detectable by their red fluorescence. (L) The green fluorescence intensity of FDA from guard cells expressing AtMT3-RFP and
their neighbor cells was quantified from microscopic images using Photoshop software (n ¼ 20). Error bars represent SE. Scale
bar ¼ 10 lm.
810
The same experiment with cells transformed with
AtMT2a showed cytoplasmic localization of
AtMT2a similar to that of AtMT3 (data not
shown). Therefore we concluded that AtMT2a and
AtMT3 are localized in the cytoplasm in tran-
siently transformed V. faba guard cells and Ara-
bidopsis cells in nor mal conditions before Cd( II)
treatment.
AtMT2a and AtMT3 do not translocate from the
cytoplasmic region in response to Cd(II) treatment
One previously report ed mechanisms of Cd(II)
detoxification in plant cells is vacuolar sequestra-
tion of Cd(II) bound to phytochelatin (Vogeli-
Lange and Wagner, 1990). Since MTs, like phyto-
chelatins, have many cysteine residues, we tested
whether the MTs of Arabidopsis are also trans lo-
cated to the vacuole after treatment of intact cells
with CdCl
2
. When V. faba guard cells transformed
with AtMT2a:GFP and AtMT3 :GFP were treated
with 10 lM CdCl
2
, the localization of AtMT2a-
GFP and AtMT3-GFP did not change (Figure 6B
and F); they continued to be colocalized with free
RFP (Figure 6C and G), as shown by the overlap
of the green and red fluorescence (Figure 6D and
H). Even after 3 h of treatment with 1–100 lM
CdCl
2
, AtMT2a and AtMT3 did not change their
localization (data not shown). Since their protec-
tive effect can be observed clearly after 1 h of
Figure 5. Localization of AtMT3 overexpressed in Vicia faba guard cells before Cd(II) treatment. (A) Bright field image; (B) green
fluorescent image of AtMT3-GFP in the guard cells shown in A; (C) red fluorescent image of free RFP in the guard cells shown in A.
AtMT3-GFP and free RFP are co-expressed in the guard cell on the right side; (D) overlap of B and C. Scale bar ¼ 10 lm.
Figure 6. Localization of AtMT2a and AtMT3 overexpressed in V. faba guard cells after treatment with 10 lM CdCl
2
(II) for 1 h. (A)
Bright field image; (B) green fluorescent image of AtMT2a-GFP in the guard cells shown in A; (C) red fluorescent image of free RFP in
the guard cells shown in A; (D) overlap of B and C; (E) bright field image; (F) green fluorescent image of AtMT3-GFP in the guard
cells shown in E; (G) red fluorescent image of free RFP in the guard cells shown in E; (H) overlap of F and G. Scale bar ¼ 10 lm.
811
treatment with 10 lM CdCl
2
(Figures 3 and 4), we
propose that AtMT2a and AtMT3 protect V. faba
guard cells against Cd(II)-toxicity while localized
in the cytosolic region.
To test the possible translocat ion of AtMT2a
and AtMT3 in other cell types, Arabidopsis pro-
toplasts from whole seedlings were transiently
transformed with AtMT2a:GFP or AtMT3:GFP
Figure 7. ROS contents of AtMT3-transformed and non-transformed guard cells before and after treatment with Cd(II). (A, B) Bright
field and fluorescence images of a non-transformed pair of guard cells before heavy metal treatment after staining with H
2
DCF. (C, D)
Bright field and fluorescence images of a pair of non-transformed guard cells stained with DCF after treatment with 10 lM CdCl
2
for
1 h. (E) Bright field image of a guard cell transformed with AtMT3:RFP (right) and its neighbor (left) after treatment with 10 lM
CdCl
2
for 1 h. (F) DCF fluorescence image of the same cells shown in E. (G) Red fluorescence image of AtMT3:RFP in the same cells
shown in E. (H) Bright field image of a guard cell transformed with AtMT3:RFP (right) and its non-transformed neighbor (left) after
treatment with 30 lM CdCl
2
for 1 h. (I) DCF fluorescence image of the same cells shown in H. (J) Red fluorescence image of
AtMT3:RFP in the same cells shown in H. (K) Bright field image of a guard cell transformed with RFP (right) and its non-transformed
neighbor (left) after treatment with 10 lM CdCl
2
for 1 h. (L) DCF fluorescence image of the same cells shown in K. (M) Red
fluorescence image of RFP in the same cells shown in K. (N, O) DCF fluorescence in AtMT3:RFP(N)- and AtMT2a:RFP(O)-
transformed and non-transformed guard cells after treatment with CdCl
2
for 1 h. The level of fluorescence was quantified by measuring
the green fluorescence intensity of DCF from microscopic images using Photoshop software (n ¼ 20). Error bars represent SE. Scale
bar ¼ 10 lm.
812
simultaneously with free RFP (Figure S2). They
were then treated with CdCl
2
for 1 h. The green
fluorescence of both AtMT2a:GFP (Figure S2B)
and AtMT3:GFP (Figure S2F) co-localized with
the red fluorescence of free RFP (Figure S2C, D,
G and H, respectively) in Arabidopsis protoplasts
after CdCl
2
treatment. Even after 3 h of treatment
with 1–100 lM CdCl
2
, translocation of AtMT2a
and AtMT3 was not detectable (data not shown).
Therefore we propose that AtMT2a and AtMT3
localize in the cytoplasm in Arabidopsis proto-
plasts before and after Cd(II) treatment.
Transformed AtMTs prevent ROS generation in
V. faba guard cells
Exposure of cells to Cd(II) induces generation of
reactive oxygen species (Watanabe and Suzuki,
2002), and oxidative stress by ROS damages plant
cells (Bethke and Jones, 2001). To test if overex-
pressed AtMT2a or AtMT3 decreased cell damage
by Cd(II) in plant cells by reducing the level of
ROS, we assayed ROS content in guard cells using
dichlorodihydrofluorescein diacetate (H
2
DCF-
DA), which produces fluorescent DCF upon oxi-
dation (Ohba et al., 1994). V. faba guard cells were
bombarded with AtMT3:RFP and then incubated
in a Cd(II)-containing bath solution for 1 h. Green
DCF fluorescence was hardly detectable in intact
guard cells without Cd(II) treatment (Figure 7B).
However, the green fluor escence appeared in guard
cells after CdCl
2
treatment (Figure 7D). When the
guard cells in bombarded leaves were observed for
DCF fluorescence after a 1 h treatment with
10 lM of Cd(II), the non-transformed cells
showed strong green fluorescence (Figur e 7F, left),
whereas the neighboring cells transformed with
AtMT3:RFP (Figure 7G) showed only weak
fluorescence (Figure 7F, right). When Cd(II) con-
centration was increased to 30 lM. The cells
transformed with AtMT3:RFP (Figure 7J) still
had a lower fluorescence compared to neighboring
non-transformed cells (Figure 7I, right). The mean
values of DCF fluorescence observed at 10 and
30 lM Cd(II) were respectively 46 and 63 for
transformed guard cells and 71 and 73 respectively
for non-transformed guard cells (Figure 7N). The
differences between transformed and non-trans-
formed cells in the level s of fluorescence at 10 and
30 lM Cd(II) were statistically significant at
P < 0.01. When Cd(II) concentration was in-
creased to 50 lM, all guard cells showed high
fluorescence and the difference between trans-
formed and non-transformed cells was no longer
significant (data not shown). Guard cells trans-
formed with free RFP (Figure 7M) gave a level of
DCF fluorescence that was similar to that of their
non-transformed neighboring cells and unrelated
to the concentration of Cd(II) employed (Fig-
ure 7L and data not shown). Similar experiments
using AtMT2a:RFP resulted in similar results. The
mean values of DCF fluorescence observed at 10
and 30 lM Cd(II) were respectively 53 and 69 for
transformed guard cells and 75 and 85 respectively
for non-transformed guard cells (Figure 7O). The
differences between transformed and non-trans-
formed cells in the level of fluorescence at 10 and
30 lM Cd(II) were statistically significant at
P < 0.01. Thus, the results confirm that Cd(II)
induces ROS generation and that overexpressed
AtMT3 or AtMT2a decreases the level of ROS in
guard cells treated with Cd(II).
Discussion
Cd(II) resistance in plant cells has been attributed
mainly to phytochelatins since the discovery of
phytochelatin synthase in plants (Clemens et al.,
1999; Ha et al., 1999; Vatamani uk et al., 1999). In
this paper, we show that plant MTs can also
contribute to Cd(II) resistance in plants as the
expression of AtMT2a and AtMT3 was induced
when Arabidopsis plants were exposed to Cd(II)
(Figure 2) and V. faba guard cells overexpressing
AtMT2a and AtMT3 displayed en hanced Cd(II)
resistance (Figures 3 and 4). We further showed
that reduction of Cd(II)-induced ROS production
is a mechanism by which MT protects the cells
from Cd(II) toxicity (Figure 7). This effect of MTs
does not seem to require translocation from the
cytoplasm, since GFP-fused MTs that did not
show apparent translocation after Cd(II) treat-
ment (Figures 5 and 6) were able to confer im-
proved Cd(II) resistance (Figures 3 and 4).
The resistance of transgenic plants to heavy
metals is commonly tested by generating stable
lines of transformants and growing them in a
medium containing the heavy metal. However, this
procedure is time-consuming. We therefore de-
signed a new and faster method to test trans-
formed cells for their resistance to abiotic stre ss
813
(Figure 3), which was to count the number of
chloroplasts in guard cells transformed by biolistic
bombardment. We found this method to be simple
and easy, since V. faba guard cells are easily
transformed by bombardment, express foreign
genes a t a high level, and have chloroplasts that
are sensitive to stress. The validity of the method
of counting chloroplasts for viability assay was
confirmed with an FDA test (Figure 4).
Besides being a good model system, guard cells
could also be a direct target for Cd toxicity,
especially when Cd is dispersed in the atmosphere
from industrial smoke or as a contaminant in rain.
These cells have been shown to respond sensitively
to Cd(II) treatment, leading to the hypothesis that
plants attempt to reduce Cd-induced damage by
closing their stomata, thereby redu cing transpira-
tion and, consequently, accumulation of the metal
(Perfus-Barbeoch et al., 2002).
In order to elucidate the function of a protein, it
is important to know its location. This has been
difficult for MT proteins, since cell fractionation
destroys MT proteins that are unstable in the
presence of oxygen. We showed that AtMT2a and
AtMT3 fused to GFP or RFP are localized in the
cytosol of V. faba and Arabidopsis cells (Figures 3–
5 and S1), and that this localization did not change
after Cd(II) treatment (Figures 6 and S2). While
the protective effect of AtMT2a and AtMT3 was
evident after treating the guard cells with 10 lM
Cd(II) for 1 h, translocation was not detectable
even after 3 h of treatment with higher concentra-
tions of Cd(II), suggesting that these MTs are not
sequestered into a compartment of low metabolic
activity. Therefore, we suggest that Cd(II) bound
to AtMT2a and AtMT3 most likely stays in the
cytoplasm of the cell; this differs from Cd(II)
bound to phytochelatin, which is sequestered into
vacuoles. However, further studies using reliable
techniques to detect MT proteins in vitro will be
necessary to conclusively determine the localiza-
tion of the MT proteins and their possible trans-
location out of the cytoplasm. It would be very
interesting if indeed MTs and phytochelatins,
which both contain many cysteine residues, are
different in the final step of detoxification of Cd(II).
It has already been shown that Cd(II) induces
the generation of ROS (Watanabe and Suzuki,
2002), but it had not previously been shown whe-
ther any Cd(II) detoxifier actually reduces the Cd-
induced increase in ROS in plant cells. We clearly
showed that AtMT3 reduced the level of ROS in a
plant cell exposed to Cd(II) (Figur e 7), and a sim-
ilar effect was found for AtMT2a (Figure 7O). It is
possible that AtMT2a and AtMT3 chelate Cd(II)
in the cytoplasm, thereby blocking Cd(II) from
freely interacting with cytoplasmic components or
entering into organelles. This in turn would reduce
the generation of ROS by the Cd(II)-induced
disturbance of cellular processes, resulting in
protection of the cell. Alternatively, MT protein
itself might act as an antioxidant (Butt et al.,
1998). Consistent to this hypothesis, MT1 gene
expression in Arabidopsis leaves is induced by
H
2
O
2
and intense light (Dunaeva and Adamska,
2001), which most likely elevat es the ROS level in
the leaf cells.
In summary, we have shown that AtMT2a and
AtMT3 can contribute to Cd(II)-resistance when
overexpressed in guard cells of V. faba , and its
mechanism of detoxification apparently does not
include sequestration of Cd(II) into organelles. In
addition, our method of introducing the genes into
guard cells and assaying the resistance of the
transformed cells to abiotic stress provides a sim-
ple and fast way to evaluate candidate genes that
may confer stress resistance.
Acknowledgements
We thank Dr. Dennis Thiele for the YCF1-null
yeast line and its isogenic wild type, and Ms. Ki-
Youb Park and Ms. Eunsook Jung for technical
assistance. We also than k Ms. Jumok Park, Dr.
Ildoo Hwang, and an anonymous reviewer for
helpful suggestions. This work was supported by
grants awarded to Y.L. from the National Re-
search Laboratory program of the Ministry of
Science and Technology of Korea and to I.H. from
the Creative Research Initiative Program of the
Ministry of Science and Technology of Korea
(M10116000005-02F0000-00310).
References
Bethke, P.C. and Jones, R.L. 2001. Cell death of barley
aleurone protoplasts is mediated by reactive oxygen species.
Plant J. 25: 19–29.
Butt, A., Mousley, C., Morris, K., Beynon, J., Can, C., Holub,
E., Greenberg, J.T. and Buchanan-Wollaston, V. 1998.
Differential expression of a senescence-enhanced metallothio-
814
nein gene in Arabidopsis in response to isolates of Peronospora
parasitica and Pseudomonas syringae. Plant J. 16: 209–221.
Clemens, S., Kim, E.J., Neumann, D. and Schroeder, J.I. 1999.
Tolerance to toxic metals by a gene family of phytochelatin
synthases from plants and yeast. EMBO J. 18: 3325–
3333.
Cobbett, C. and Goldsbrough, P. 2002. Phytochelatins and
metallothioneins: roles in heavy metal detoxification and
homeostasis. Annu. Rev. Plant Biol. 53: 159–182.
Cobbett, C.S., May, M.J., Howden, R. and Rolls, B. 1998. The
glutathione-deficient, cadmium-sensitive mutants, cad2-1, of
Arabidopsis thaliana is deficient in (gamma)- glutamylcysteine
synthetase. Plant J. 16: 73–78.
Dunaeva, M. and Adamska, I. 2001. Identification of genes
expressed in response to light stress in leaves of Arabidopsis
thaliana using RNA differential display. Eur. J. Biochem.
268: 5521–5529.
Ha, S.-B., Smith, A.P., Howden, R., Dietrich, W.M., Bugg, S.,
O’Connell, M.J., Goldsbrough, P.B. and Cobbett, C.S. 1999.
Phytochelatin synthase genes from Arabidopsis and the
Yeast Schizosaccharomyces pombe. Plant Cell 11: 1153–116.
Hall, J.L. 2002. Cellular mechanism for heavy metal detoxifi-
cation and tolerance. J. Exp. Bot. 53: 1–11.
Hirschi, K.D., Korenkov, V.D., Wilganowski, N.L. and Wag-
ner, G.J. 2000. Expression of Arabidopsis CAX2 in tobacco.
Altered metal accumulation and increased manganese toler-
ance. Plant Physiol. 124: 125–133.
Howden, R., Goldsbrough, P.B., Anderson, C.R. and Cobbett,
C.S. 1995. Cadmium-sensitive, cad1 mutants of Arabidopsis
thaliana are phytochelatin deficient. Plant Physiol. 107: 1059–
1066.
Jach, G., Binot, E., Frings, S., Luxa, K. and Schell, J. 2001. Use
of red fluorescent protein from Discosoma sp. (dsRED) as a
reporter for plant gene expression. Plant J. 28: 483–491.
Jin, J.B., Kim, Y.A., Kim, S.J., Lee, S.H., Kim, D.H., Cheong,
G.W. and Hwang, I. 2001. A new dynamin-like protein, ADL6,
is involved in trafficking from the trans-golgi network to the
central vacuole in Arabidopsis. Plant Cell 13: 1511–1525.
Jung, J.Y., Kim, Y.W., Kwak, J.M., Hwang, J.U., Young, J.,
Schroeder, J.I., Hwang, I. and Lee, Y. 2002. Phosphatidyl-
inositol 3- and 4-phosphate are required for normal stomatal
movements. Plant Cell 14: 2399–2412.
Murashige, T. and Skoog, F. 1962. A revised medium for rapid
growth and bioassays with tobacco tissue cultures. Physiol.
Plant 15: 473–497.
Murphy, A. and Taiz, L. 1995. Comparison of metallothionein
gene expression and nonprotein thiols in ten Arabidopsis
ecotypes. Plant Physiol. 109: 945–954.
Murphy, A., Zhou, J., Goldsbrough, P.B. and Taiz, L. 1997.
Purification and immunological identification of metallothi-
oneins 1 and 2 from Arabidopsis thaliana. Plant Physiol. 113:
1293–1301.
Ohba, M., Shibanuma, M., Kuroki, T. and Noes, K. 1994.
Production of hydrogen peroxide by transforming growth
factor-beta 1 and its involvement in induction of egr-1
in mouse osteoblastic cells. J. Cell Biol. 126: 1079–1088.
Palmiter, R.D. 1998. The elusive function of metallothioneins.
Proc. Natl. Acad. Sci. USA 95: 428–430.
Park, K. Y., Jung, J. Y., Park, J., Hwang, J. U., Kim, Y. W.,
Hwang, I., Lee, Y. 2003. A role for phosphatidylinositol 3-
phosphate in abscisic acid-induced reactive oxygen species
generation in guard cells. Plant Physiol. 132: 92–98.
Perfus-Barbeoch, L., Leonhardt, N., Vavasseur, A. and Fores-
tier, C. 2002. Heavy metal toxicity: cadmium permeates
through calcium channels and disturbs the plant water status.
Plant J. 32(4): 539–548.
Piao, H.L., Pih, K.T., Lim, J.H., Kang, S.G., Jin, J.B., Kim,
S.H. and Hwang, I. 1999. An Arabidopsis GSK3/shaggy-like
gene that complements yeast salt stress-sensitive mutants is
induced by NaCl and abscisic acid. Plant Physiol. 119(4):
1527–1534.
Sambrook, J. and Russell, D.W. 2001. Molecular Cloning: A
Laboratory Manual, 3rd edn. Cold Spring Harbor Labora-
tory Press, Cold Spring Harbor, NY.
Vatamaniuk, O.K., Mari, S., Lu, Y.P. and Rea, P.A. 1999.
AtPCS1, a phytochelatin synthase from Arabidopsis: Isola-
tion and in vitro reconstitution. Proc. Natl. Acad. Sci. USA
96: 7110–7115.
Vogeli-Langer, R. and Wagner, G.J. 1990. Sub-cellular local-
ization of cadmium and cadmium-binding peptides in
tobacco leaves. Implication of a transport function for
cadmium binding peptides. Plant Physiol. 92: 1086–1093.
Watanabe, M. and Suzuki, T. 2002. Involvement of reactive
oxygen stress in cadmium-induced cellular damage in Euglena
gracilis. Comp. Biochem. Phys. C 131: 491–500.
Widholm, J.M. 1972. The use of fluorescein diacetate and
phenosafranine for determining viability of cultured plant
cells. Stain Technol. 47: 189–194.
Wink, M. 1993. The plant vacuole: a multifunctional compart-
ment. J. Exp. Bot. 44: 231–246.
Zhou, J. and Goldsbrough, P.B. 1994. Functional homologs of
fungal metallothionein genes from Arabidopsis. Plant Cell 6:
875–884.
815
