Cadmium localization and quantification

Cadmium localization and quantification in the plant
Arabidopsis thaliana using micro-PIXE
F.J. Ager
a,
*
, M.D. Ynsa
a
, J.R. Dom

ıınguez-Sol

ııs
b
, C. Gotor
b
,
M.A. Respaldiza
a
, L.C. Romero
b
a
Centro Nacional de Aceleradores, Av. Thomas A. Edison s/n, E-41092 Sevilla, Spain
b
Instituto de Bioqu

ıımica Vegetal y Fotos

ııntesis, Av. Ameerico Vespucio s/n, E-41092 Sevilla, Spain
Abstract
Remediation of metal-contaminated soils and waters poses a challenging problem due to its implications in the
environment and the human health. The use of metal-accumulating plants to remove toxic metals, including Cd, from

soil and aqueous streams has been proposed as a possible solution to this problem. The process of using plants for
environmental restoration is termed phytoremediation. Cd is a particularly favourable target metal for this technology
because it is readily transported and accumulated in the shoots of several plant species. This paper investigates the sites
of metal localization within Arabidopsis thaliana leaves, when plants are grown in a cadmium-rich environment, by
making use of nuclear microscopy techniques. Micro-PIXE, RBS and SEM analyses were performed on the scanning
proton microprobe at the CNA in Seville (Spain), showing that cadmium is sequestered within the trichomes on the leaf
surface. Additionally, regular PIXE analyses were performed on samples prepared by an acid digestion method in order
to assess the metal accumulation of such plants. Ó 2002 Published by Elsevier Science B.V.
PACS: 89.60; 78.70.E; 82.80.Yc
Keywords: Arabidopsis; Cadmium accumulator; Phytoremediation; Nuclear microprobe; PIXE
1. Introduction
The ability of certain terrestrial plants to absorb
and accumulate metals such as cadmium, nickel,
zinc, manganese, copper or cobalt makes them very
attractive when the decontamination of soils is
sought. These plants are called metal hyperaccu-
mulators [1] if they accumulate for instance more
than 0.01% of Cd, 0.1% of Ni or 1% of Zn per dry
weight in their shoots in a natural environment.
However, the mechanisms for metal uptake, trans-
location and compartmentation are not yet well
understood and in the past recent years an im-
portant effort has been made in this direction.
From the practical point of view, the comprehen-
sion of those mechanisms will probably lead to
the enhancement of metal absorption by means of
plant hybridation or by producing new transgenic
plants forenvironmental remediation purposes. For
example, genetic engineering of Arabidopsis tha-
liana plant has been demonstrated to be a useful

technique to improve heavy metal tolerance by
phytoremediation purposes [2].
Nuclear Instruments and Methods in Physics Research B 189 (2002) 494–498
www.elsevier.com/locate/nimb
*
Corresponding author. Tel.: +34-954460553; fax: +34-
954460145.
E-mail address: (F.J. Ager).
0168-583X/02/$ - see front matter Ó 2002 Published by Elsevier Science B.V.
PII: S 0168- 583 X ( 01) 0 1 130 - 2
There is evidence that trichomes on the leaf
surface may play a role in the detoxification of
heavy metals [3]. In Alyssum lesbiacum [4] micro-
PIXE analysis proved that epidermal trichomes
represent a site of preferential nickel accumula-
tion. In A. thaliana, trichomes are specialized uni-
cellular structures with uncertain functions, but
recent works suggest their possible role as a sink
during detoxification processes [5].
The nuclear microprobe allows investigators to
obtain quantitative or semi-quantitative elemental
distribution maps of major and trace elements
with high resolution and sensitivity, and is be-
coming a useful tool for localizing the sites of el-
ement accumulation in a wide variety of studies.
In the present study we used the scanning nu-
clear microprobe to determine the elemental con-
centrations and the sites of preferential Cd
accumulation within A. thaliana leaves in plants
grown in Cd-enriched soils. Regular PIXE analysis

was also performed to evaluate the Cd accumula-
tion in the plant leaves.
2. Materials and methods
2.1. Plant material
Wild type A. thaliana (ecotype Columbia)
plants were grown, in a controlled environment
room, on moist vermiculite supplemented with
Hoagland medium at 20 °C in the light and 18 °C
in the dark, under a 16-h white light/8-h dark
photoperiod with a photon flux density of 130 lE/
m
2
s and 70% humidity.
Cadmium chloride treatments were performed
by addition to the Hoagland medium of CdCl
2
to
250 and 2500 lM final concentration. The plants
were daily watered with this medium during 14
days.
2.2. Sample preparation for analysis
In order to minimise the possibility of ion mo-
bilisation, plant leaves were cut off from living
plants taken directly from the growth chamber,
rinsed briefly in deionised water, dried, immedi-
ately frozen at À80 °C in an ultralow temperature
freezer (mod. Nuaire NU-6511) and then freeze-
dried for 72 h at À50 °C at a pressure of 10
À3
mbar. After this preparation, two different treat-

ments were used depending on the subsequent
analysis procedure.
Leaves for macrobeam analysis were prepared
by microwave acid digestion. Plant material (100
mg) was digested in a Teflon bomb in a solution of
HNO
3
and Y as internal standard, following
standard procedures [6]. For elemental analysis, a
10 ll volume of the resulting solution was pipetted
on a polycarbonate film and dried in vacuum.
Leaves for microbeam analysis were mounted
on carbon tape on a standard aluminium frame
and just placed on the sample holder inside the
microprobe target chamber (pres. 10
À7
mbar).
2.3. Instrumentation and analytical methods
The microbeam analyses were performed with
3.0 MeV protons focused to a 3 Â 3 lm beam
normal to the sample with a proton current of 100–
300 pA, using the CNA scanning nuclear microp-
robe [7]. PIXE, backscattering spectrometry and
electron imaging were carried out simultaneously.
PIXE spectra were collected using a Si(Li) X-ray
detector manufactured by Canberra at 45° to the
beam, with a 12.5 mm
2
active area, 8 lm Be re-
movable window and a 50 lm Mylar

â
filter to at-
tenuate X-rays from light elements. The distance
from the detector to the sample was chosen to be 40
mm in order to keep good counting statistics with
Table 1
PIXE analysis of A. thaliana leaves treated with CdCl
2
(250
lM) prepared by microwave acid digestion
Element Concentration (ppm)
K 26100 Æ 700
Ca 28200 Æ 700
Ti 7 Æ 3
Cr 4 Æ 1
Mn78Æ 3
Fe 81 Æ 3
Ni 5 Æ 1
Cu 14 Æ 1
Zn 41 Æ 2
Cd 240 Æ 60
F.J. Ager et al. / Nucl. Instr. and Meth. in Phys. Res. B 189 (2002) 494–498 495
low pile-up background and dead time. Backscat-
tered protons were detected using a surface barrier
detector of an active area of 300 mm
2
at an angle of
37° to the beam, in Cornell geometry. Emitted
electrons were detected using a channeltron detec-
tor. An electron gun generating 50 mA was used to

avert electrostatic charging of the insulating bio-
logical material. All signals were recorded together
with the beam position using the OM_DAQ data
acquisition system [8].
Macro-PIXE analyses were performed with a
2.4 MeV proton beam from the 3 MV Van de
Graaff accelerator of the ITN (Portugal), colli-
mated up to 5 mm diameter with a current of 100
nA and a total accumulated charge of 100 lC. The
incoming proton beam angle was 15°. X-rays were
detected with a Linke Si(Li) detector at an angle
of 55°,8lm Be window, a 350 lm Mylar
â
filter
and a 2.7 mm diameter collimator placed in front
of the detector specially for improvement of the
line shape.
Data analysis was carried out using GUPIX [9]
for PIXE spectra and RUMP [10] and SIMNRA
[11] for RBS spectra.
Fig. 1. PIXE elemental maps of a leaf edge of an Arabidopsis plant grown in CdCl
2
(250 lM), showing Ca, K, P, Si, Mn and Fe. Area
of scan is 100 Â 100 lm
2
. The maps correspond to two consecutive scans so that the lower right end of the first row of elements
coincides with the central area of the second row.
496 F.J. Ager et al. / Nucl. Instr. and Meth. in Phys. Res. B 189 (2002) 494–498
3. Results and discussion
Macro-PIXE analysis of plant leaves treated

with CdCl
2
(250 lM) and prepared by acid di-
gestion in microwave oven gives a Cd concentra-
tion of 0:024 Æ 0:006 wt.% as shown in Table 1.
Micro-PIXE elemental maps and point analyses
of different leaves were recorded. Si, P, S, Cl, K,
Ca, Fe, Mn, Zn and Cd were all detected by PIXE.
For Cd-treated leaves, Cd mapping is unpractical
because of the long acquisition times needed for
this purpose. However, once the leaf structure is
known by means of the other maps (electron
imaging, Ca, K, etc.), those maps can be comple-
mented with analyses at selected points or regions
of interest.
Fig. 1 shows elemental maps (side view,
100 Â 100 lm
2
) of a leaf of an Arabidopsis plant
grown in CdCl
2
(250 lM). Thus, the trichome can
be divided in three different areas according to
their composition: the base, richer in Si, K and P; a
central zone, richer in metals such as Mn and Fe;
and the head, richer in Ca. The main elements
detected in the leaf are K, S, Ca and Cl (not shown
in Fig. 1).
A front view of a trichome emerging from the
leaf surface is presented in Fig. 2 (250 Â 250 lm

2
),
showing the typical impact points for analysis in
trichome (T) and leaf (L).
Cd concentration can be ideally computed from
the K
a
line of the PIXE spectrum because it lies in
a very clean region far from other peaks and from
pile-up effects from the main detected elements
(Ca, K, etc.). Fig. 3 depicts the Cd contents found
by PIXE microanalysis in different leaves. Cad-
mium is present in both trichome and leaf, but the
Fig. 2. Secondary electron image and maps of elemental distribution of Ca, K, Cl, P and Mn (250 Â 250 lm
2
) of a leaf of Arabidopsis
grown in CdCl
2
(2500 lM). The image corresponds to a front view of a trichome emerging from the leaf surface. The typical impact
points for analysis in trichome (T) and leaf (L) are presented.
Fig. 3. Bar chart of cadmium contents obtained by PIXE mi-
croanalysis in A. thaliana leaves. Question marks indicate
points where there is no data available. Error bars are also
shown.
F.J. Ager et al. / Nucl. Instr. and Meth. in Phys. Res. B 189 (2002) 494–498 497
highest amount of Cd is found in the trichomes.
For plants treated with CdCl
2
(250 lM), Cd was
detectable in the trichomes but it required a high

integrated charge in order to be quantified with
low errors. To circumvent this difficulty, a higher
Cd dose (2500 lM CdCl
2
) was used. This treat-
ment would result in the increase of the metal
uptake by the plant and the enhancement in the
Cd concentration in the plant tissues. In effect,
analyses show that Cd concentration in leaf epi-
dermis and trichome is increased when the plants
are treated with CdCl
2
(2500 lM), although the
growth was affected by such toxic levels of Cd
(smaller plants with shorter purplish leaves). The
Cd content also depends on the point of analysis
and the orientation of the trichome, being higher
when the trichome is analysed in the head with the
ion beam coming from its growth direction.
4. Conclusions
The preliminary results presented in this study
suggest that Cd is preferentially accumulated in
the epidermal trichomes of the cadmium accumu-
lator plant A. thaliana. However, further analyses
are being performed at present to establish the
distribution of Cd along the trichomes, because
there are evidences [2,5] that the base could con-
centrate even more Cd than the head or the stem.
This work also contributes to progress in the
decontamination of metal polluted soils by means

of phytoremediation techniques. Present investi-
gations by the same authors are also aimed at
comparing the wild variety of A. thaliana with ge-
netically modified specimens produced to be more
resistant to heavy metal contamination.
Acknowledgements
We thank Dr. Teresa Pinheiro for her assistance
during the preparation and analysis of samples at
the ITN.
References
[1] R.R. Brooks, B.H. Robinson, in: R.R. Brooks (Ed.),
Plants that Hyperaccumulate Heavy Metals, CAB Inter-
national, Wallingford, UK, 1998.
[2] J.R. Dom

ıınguez-Sol

ııs, G. Guti

eerrez-Alcal

aa, L. Romero,
C. Gotor, J. Biol. Chem. 276 (2001) 9297.
[3] H. K
€
uupper, E. Lombi, F.J. Zhao, S.P. McGrath, Planta
212 (2000) 75.
[4] U. Kr
€
aamer, G.W. Grime, J.A.C. Smith, C.R. Hawes,

A.J.M. Baker, Nucl. Instr. and Meth. B 130 (1997)
346.
[5] G. Guti

eerrez-Alcal

aa, C. Gotor, A.J. Meyer, M. Fricker,
J.M. Vega, L.C. Romero, Proc. Natl. Acad. Sci USA 97
(20) (2000) 11108.
[6] M. Stoeppler, Sampling and Sample Preparation, Springer,
Berlin, 1997.
[7] J. Garc

ııa L

oopez, F.J. Ager, M. Barbadillo Rank, M.

AA.
Ontalba, M.

AA. Respaldiza, M.D. Ynsa, Nucl. Instr. and
Meth. B 161–163 (2000) 1137.
[8] G.W. Grime, M. Dawson, Nucl. Instr. and Meth. B 104
(1995) 107.
[9] J.A. Maxwell, J.L. Campbell, W.J. Teesdale, Nucl. Instr.
and Meth. B 43 (1989) 218.
[10] L.R. Doolitle, Nucl. Instr. and Meth. B 9 (1985) 344.
[11] M. Mayer, SIMNRA User’s Guide, Technical Report IPP
9/113, Max-Planck-Institut f
€

uur Plasmaphysik, Garching,
Germany, 1997.
498 F.J. Ager et al. / Nucl. Instr. and Meth. in Phys. Res. B 189 (2002) 494–498