Caesium-affected gene expression in Arabidopsis thaliana

Authors

  • Tobias Sahr,

    1. GSF – National Research Center for Environment and Health: 1Institute of Biochemical Plant Pathology; 3Institute of Radiation Protection; 4Institute of Ecological Chemistry, D-85764 Neuherberg, Germany; 2International Atomic Energy Agency, Agency's Laboratories Seibersdorf, PO Box 100, A-1400 Vienna, Austria; *Present address: Commissariat à l’Energie Atomique (CEA) Grenoble, Laboratoire de Physiologie Cellulaire Végétale (PCV), 17 Rue des Martyrs, F-38000 Grenoble, France
    Search for more papers by this author
  • 1* Gabriele Voigt,

    1. GSF – National Research Center for Environment and Health: 1Institute of Biochemical Plant Pathology; 3Institute of Radiation Protection; 4Institute of Ecological Chemistry, D-85764 Neuherberg, Germany; 2International Atomic Energy Agency, Agency's Laboratories Seibersdorf, PO Box 100, A-1400 Vienna, Austria; *Present address: Commissariat à l’Energie Atomique (CEA) Grenoble, Laboratoire de Physiologie Cellulaire Végétale (PCV), 17 Rue des Martyrs, F-38000 Grenoble, France
    Search for more papers by this author
  • 2 Herwig G. Paretzke,

    1. GSF – National Research Center for Environment and Health: 1Institute of Biochemical Plant Pathology; 3Institute of Radiation Protection; 4Institute of Ecological Chemistry, D-85764 Neuherberg, Germany; 2International Atomic Energy Agency, Agency's Laboratories Seibersdorf, PO Box 100, A-1400 Vienna, Austria; *Present address: Commissariat à l’Energie Atomique (CEA) Grenoble, Laboratoire de Physiologie Cellulaire Végétale (PCV), 17 Rue des Martyrs, F-38000 Grenoble, France
    Search for more papers by this author
  • 3 Peter Schramel,

    1. GSF – National Research Center for Environment and Health: 1Institute of Biochemical Plant Pathology; 3Institute of Radiation Protection; 4Institute of Ecological Chemistry, D-85764 Neuherberg, Germany; 2International Atomic Energy Agency, Agency's Laboratories Seibersdorf, PO Box 100, A-1400 Vienna, Austria; *Present address: Commissariat à l’Energie Atomique (CEA) Grenoble, Laboratoire de Physiologie Cellulaire Végétale (PCV), 17 Rue des Martyrs, F-38000 Grenoble, France
    Search for more papers by this author
  • and 4 Dieter Ernst 1

    Corresponding author
    1. GSF – National Research Center for Environment and Health: 1Institute of Biochemical Plant Pathology; 3Institute of Radiation Protection; 4Institute of Ecological Chemistry, D-85764 Neuherberg, Germany; 2International Atomic Energy Agency, Agency's Laboratories Seibersdorf, PO Box 100, A-1400 Vienna, Austria; *Present address: Commissariat à l’Energie Atomique (CEA) Grenoble, Laboratoire de Physiologie Cellulaire Végétale (PCV), 17 Rue des Martyrs, F-38000 Grenoble, France
    Search for more papers by this author

Author for correspondence: Dieter Ernst Tel: +49 89 3187 4440 Fax: +49 89 3187 3383 Email: ernst@gsf.de

Summary

  • • Excessive caesium can be toxic to plants. Here we investigated Cs uptake and caesium-induced gene expression in Arabidopsis thaliana.
  • • Accumulation was measured in plants grown for 5 wk on agar supplemented with nontoxic and up to toxic levels of Cs. Caesium-induced gene expression was studied by suppression-subtractive hybridization (SSH) and RT–PCR.
  • • Caesium accumulated in leaf rosettes dependent upon the external concentration in the growth media, whereas the potassium concentration decreased in rosettes. At a concentration of 850 µm, Cs plants showed reduced development, and withered with an increase in concentration to 1 mm Cs. SSH resulted in the isolation of 73 clones that were differentially expressed at a Cs concentration of 150 µm. Most of the genes identified belong to groups of genes encoding proteins in stress defence, detoxification, transport, homeostasis and general metabolism, and proteins controlling transcription and translation.
  • • The present study identified a number of marker genes for Cs in Arabidopsis grown under nontoxic Cs concentrations, indicating that Cs acts as an abiotic stress factor.

Introduction

Caesium can be taken up easily by plant roots and then transferred to above-ground plant parts. This is a key process for 137Cs entering the food chain (Witherspoon & Brown, 1965; Guivarch et al., 1999) and thus leading to internal exposure of humans in the long term after contamination of soils. Genetic variation in 137Cs uptake is evident from the wide range of transfer factors observed in different plants (Broadley & Willey, 1997; Zhu & Smolders, 2000), in addition to its physicochemical form and its behaviour in the soil (migration and binding in soil clay components, depending on soil properties). In one species growing on the same soil, the developmental stage and mycorrhizal symbiosis are important for the uptake efficiency of Cs (Smolders & Shaw, 1995; Zhu & Shaw, 2000; Berreck & Haselwandter, 2001). However, under natural conditions normally < 3% of the total Cs in the soil is absorbed by plants (Lasat et al., 1998).

Potassium is an abundant cation found in plants (Raven et al., 1976), and the uptake into roots occurs via three different types of ion channel (Schachtman, 2000). The alkali cation Cs acts as a K analogue and is also toxic to plants (White & Broadley, 2000). It is well known that Cs is a competitive inhibitor of K and acts as a K-channel blocker (White & Broadly, 2000; Zhu & Smolders, 2000), and the accumulation of Cs in plants is decreased by increasing K concentrations (Smolders et al., 1996; Tsukada et al., 2002). This fact has been used in the past with the application of K fertilizers to reduce radiocaesium uptake in agricultural systems. However, short-term K starvation experiments resulted in increased Cs influx rates, indicating the importance of internal and external K status (Zhu et al., 2000).

Because of the large hydrated ion radius of Cs, the free mobile single electron can react with water and oxygen, resulting in the formation of reactive oxygen species. This results in activation of the anti-oxidative defence system. In plants, Cs application resulted in induction of peroxidases and catalases, and an increased amount of anti-oxidative metabolites such as glutathione (Ghosh et al., 1993). Caesium interferes with internal K ions, thus influencing the osmotic potential and regulation of the ion balance. Application of Cs to erythrocytes of frogs resulted in a disturbed water balance (Ermakova, 1970). In mice, chromosomal aberrations and a reduction in mitotic cell division was found (Ghosh et al., 1993). Small amounts of Cs inactivated the 60S ribosomal subunit in liver tissue of rats (Arpin et al., 1972). In plants, Cs has a negative effect on growth and photosynthesis (Ghosh et al., 1993). At higher concentrations, Cs interferes with the biosynthesis of chlorophyll by inactivation of uroporphyrinogen decarboxylase (Shalygo et al., 1997). However, in general little is known about the effects of 133Cs on plants. It can be assumed that Cs, like other alkali cations, will interact with negatively charged groups of amino acids, nucleic acids and cell-wall compounds. The structure and activity of proteins will be affected, and the fluidity of membranes can be changed by binding to membrane proteins and phospholipids (Ghosh et al., 1993).

Diverse methods have been described of selecting plants for phytoremediation of 137Cs or to minimize its uptake (White et al., 2003). In the present study we analysed Cs uptake by the model plant Arabidopsis thaliana. The combination of suppression-subtractive hybridization (SSH) and microarray technology resulted in the identification of numerous genes, which showed a changed transcriptional profile on application of nontoxic Cs concentrations.

Materials and Methods

Plant growth conditions

Arabidopsis thaliana ecotype Columbia (Col-0) wild-type seeds were surface sterilized and transferred into 0.1% agar. After cold treatment in the dark for 1 d at 4°C, seeds were sown on 25% MS agar (pH 5.7, 2 mm K) containing 0.5% saccharose (w/v) (Serva, Heidelberg, Germany) and different concentrations of CsCl. Phytatray II pots (Sigma, Taufkirchen, Germany) containing the seeds were transferred into a controlled-environment cabinet, and plants were grown for 5 wk as described (Heidenreich et al., 1999). For additional Cs and K uptake studies, plants were grown in a medium according to Gibeaut et al. (1997) containing 0.7 mm K. For molecular biological analysis, seedlings 3 d old were transferred to Erlenmeyer flasks containing 30 ml liquid medium (Gibeaut et al., 1997) and cultivated under the conditions described above on a gyratory shaker at 60 rpm. Plants were separated into root and leaf tissue, frozen in liquid nitrogen, homogenized with a mortar and pestle, and stored at −80°C for further analysis.

Caesium and potassium quantification

Total Cs was determined using a sector field ICP-MS ELEMENT 1 (Finnigan Mat, Bremen, Germany) after pressurized wet digestion of the sample material (Schramel et al., 1980; Schramel & Wendler, 1998). From the same sample solution, K was analysed using an ICP-OES (JY70, Jobin-Yvon, Longjumeau, France).

RNA isolation and cDNA synthesis

Total root RNA was isolated with TRIzol reagents (Invitrogen, Karlsruhe, Germany). Poly(A)+ was isolated from total RNA using the Oligotex mRNA kit (Qiagen, Hilden, Germany). The RNA yield and quality were determined by spectral photometry at 260 and 280 nm.

Total RNA (2 µg) was reverse transcribed using Superscript II reverse transcriptase according to the manufacturer's instructions (Invitrogen).

Suppression-subtractive hybridization (SSH)

SSH was performed with the PCR Select cDNA subtraction kit (Clontech, Heidelberg, Germany) using 2 µg poly(A)+ RNA (Heidenreich et al., 2001). The cDNA fragments obtained by the SSH method were cloned into a pGEM-T vector (Promega, Mannheim, Germany). Bacterial clones were stored at −80°C as stocks for PCR amplification of the cDNA inserts. Three different subtractive cDNA libraries were constructed. Control (no Cs) and treated plants (100 µm Cs added from the beginning, plus 50 µm additional Cs 3 d before harvest) were used alternately as driver and tester.

Array production, hybridization and data analysis

Isolated cDNA clones were amplified using flanking vector DNA sequences. PCR products were concentrated using multiscreen plates (Millipore, Schwalbach, Germany), resuspended in 50 µl H2O, and spotted in duplicate onto Hypond-N+ membranes (Amersham, Freiburg, Germany) using the Microgrid II robot (400 µm pins; BioRobotics, Cambridge, UK). After spotting the membranes were UV cross-linked (Stratagene, Amsterdam, the Netherlands), denatured and hybridized against radiolabelled forward- and reverse-subtracted, respectively, unsubtracted driver and tester cDNAs. Nucleic acids were radiolabelled with α-33P-dATP (37 MBq ml−1) using a Random Priming Kit (Invitrogen). DNA arrays were hybridized overnight at 68°C in 5 × saline-sodium citrate (SSC) buffer containing 5× Denhardt's solution, 0.5% SDS and 100 µg ml−1 denatured salmon sperm DNA. After final washings at 68°C with 0.2 × SSC/0.1% SDS, membranes were exposed to imaging plates and data were acquired using an FLA-3000 imaging system (Fuji, Düsseldorf, Germany). Primary data were processed and normalized using the arrayvision 6.0 software (InterFocus, Mering, Germany). Clones that hybridize in all replicates to the forward-subtracted and unsubtracted tester probes, but not (or at least with ratios of > 2 for induced or < 0.5 for suppressed genes) to the reverse-subtracted or unsubtracted driver probes, were considered as an indication of a significant change of expression. Positive clones were sequenced and a database search (Munich Information Center for Protein Sequences (MIPS), http://mips.gsf.de; National Center for Biotechnology Information (NCBI), http://www.ncbi.nlm.nih.gov) assigned the putative function of the different genes.

Semiquantitative PCR

Semiquantitative RT–PCR was carried out in the presence of GeneAmp RNA pAW109 (Applied Biosystems, Darmstadt, Germany). Amplification of cDNA was carried out as follows: 95°C for 5 min; then 33–39 cycles at 95°C for 1 min; 53°C for 30 s; 72°C for 5 min. The artificial cDNA was amplified using the primer pair DM151/DM152. After separation on 2% agarose gels, the intensity of the corresponding bands was analysed under a transilluminator and quantification carried out using the multianalyst software (Biorad, München, Germany). An induction factor was calculated using the ratio of the specific gene/internal artificial standard.

Real-time RT–PCR

The RT–PCR was performed in a 25 µl reaction mixture of the QuantiTect SYBR Green PCR kit (Quiagen), including 2 µl cDNA, using a Taqman ABI 7700 system (Perkin Elmer, Weiterstadt, Germany). Amplification of PCR products was monitored via intercalation of the fluorescence dye SYBR Green. The following program was applied: initial polymerase activation at 95°C for 15 min; then 50 cycles at 94°C for 1 min; 53°C for 30 s; 72°C for 1 min. PCR conditions were optimized for high amplification efficiency for all primer pairs used. The efficiency of each PCR reaction was determined by plotting a defined dilution series of the same cDNA (1 : 1; 1 : 10; 1 : 100; 1 : 1000) vs the corresponding cycle threshold value, resulting in a straight regression line. As an internal control standard an artificial pAW109 RNA (Applied Biosystems) or the housekeeping gene CyclinB1 was used.

The quality of PCR products was inspected visually by electrophoresis, and the generation of a single band of the expected size was taken as a criterion for specificity. The identity of PCR products was confirmed by direct DNA sequencing.

To compare the means of statistical series, a one-way anova test was carried out; significance was defined as P < 0.05.

Results and Discussion

Caesium uptake and morphological changes of leaves

Growth of A. thaliana Col-0 on 25% MS medium (2 mm K) with increasing external CsCl concentration up to 1 mm resulted in increased Cs accumulation in leaves (Fig. 1). The bioaccumulation ratio of Cs ([Cs]leaves : [Cs]medium) was in the range 70–75. Comparable values were reported for several other plant species, ranging from 38 to 165 (Lasat et al., 1997). This indicates that Cs is highly mobile in Arabidopsis, as in other plant species (Resnik et al., 1969). Germination of plants was unaffected up to 700 µm CsCl; at 850 µm germination was retarded and the rate was drastically reduced; and at a concentration of 1 mm seedlings died after a few days (Fig. 2). After 3 wk growth at 500–700 µm the first visible symptoms such as necrotic leaf areas were visible, and root growth was also affected (Fig. 2). Such phytotoxic effects have also been reported for other metal ions, and might be attributed to a general disturbance of the physiological and biochemical status of the plant, probably caused by oxidative stress (Heidenreich et al., 1999, 2001). Lower concentrations of CsCl had no influence on the morphology and development of plants. Similarly, hydroponically grown plants showed no change in leaf morphology at 150 µm CsCl. Increasing the external K concentration to 8 mm (100% MS medium) resulted in normal growth even at a CsCl concentration of 2 mm. Under decreased K concentrations (≈ 0.7 mm) plants were not affected up to 500 µm CsCl. Increasing external Cs at this K concentration resulted in increased Cs accumulation and consequently in decreased K uptake (Fig. 3a,b). Decreasing the external K concentration from 2 mm to 0.7 mm resulted in an elevated bioaccumulation ratio of ≈ 150. This antagonistic effect of K on Cs accumulation is well documented in the literature (Zhu & Smolders, 2000).

Figure 1.

Caesium concentration in leaf rosettes of Arabidopsis thaliana Col-0. Plants were grown for 5 wk on 25% MS agar containing different CsCl concentrations. Increasing the external Cs concentration resulted in increased Cs accumulation in leaf rosettes.

Figure 2.

Effect of CsCl on Arabidopsis thaliana Col-0. Plants were grown for 5 wk on 25% MS agar containing the indicated CsCl concentration in µm. Germination was unaffected up to 700 µm CsCl. After 3 wk growth at 500–700 µm, the first visible symptoms such as necrotic leaf areas were visible, and root growth was also affected. Necrotic leaf areas are shown in more detail in the panel inset at 700 µm CsCl compared with the control panel inset.

Figure 3.

Caesium (a) and potassium (b) concentration in leaf rosettes of Arabidopsis thaliana Col-0. Plants were grown for 5 wk on a medium according to Gibeaut et al. (1977) containing 0.7 mm K and the indicated CsCl concentrations. Increasing the external Cs concentration resulted in increased Cs accumulation in leaves (a) and consequently in decreased uptake of K (b). Error bars, ± SD.

Isolation of CsCl-affected genes

Stress responses detected at up-regulated or down-regulated transcripts are often not specific, as identical genes might also be affected by other stress factors (Heidenreich et al., 1999; Reymond et al., 2000; Broschéet al., 2002). Oxidative stress effects are well known, caused by quite different stress factors (Inzé & Van Montagu, 2002 and articles therein). This often results in a cross-talk of different stress factors affecting identical genes (Sandermann et al., 1998). In addition the dose, as well as the duration of stress, might affect the stress specificity. To identify stress-responsive genes, the enrichment of responsive transcripts by the SSH method is a powerful tool (Diatchenko et al., 1996; Gurskaya et al., 1996; Heidenreich et al., 2001; Sävenstrand et al., 2002). To isolate Cs-responsive genes, Arabidopsis plants were grown for 5 wk on a nontoxic micromolar concentration of Cs that occurs naturally in highly exposed soils (up to 25 µg g−1 soil; White & Broadley, 2000). Investigation of plant uptake of Cs from Cs-treated peat soil (up to 40 mg kg−1) resulted in the accumulation of ≈ 2000 mg Cs per kg leaf material (Campbell & Davies, 1997). This value is comparable with our results of ≈ 2000–3000 mg kg−1 at an external Cs concentration of 100–150 µm CsCl (Fig. 3a). Nevertheless, it is important to note that the bioavailability of Cs under natural conditions is in the low micromolar range in soil solutions (White & Broadley, 2000). Under our growth conditions, plants had enough time to adapt to the Cs concentration applied. Thus it is expected that changes at transcriptional level are near the detection limit. However, for selected up- and down-regulated transcripts isolated by SSH, small transcriptional changes could be verified by quantitative RT–PCR and corresponding statistical analysis.

To avoid toxic effects of Cs, plants were grown at an external Cs concentration of up to 150 µm. After liquid cultivation for 5 wk, roots were harvested for the isolation of Cs-affected transcripts using the SSH method.

The 1500 clones obtained by the three SSHs were first analysed by DNA-microarray analysis. All clones were spotted onto nylon membranes and hybridized against radiolabelled subtracted and nonsubtracted cDNAs. This was repeated three times for each probe. Because the membranes contained each clone twice, each single gene was analysed six times. Only clones that showed a difference in signal intensities in all replicates (subtracted vs nonsubtracted) with a factor > 2 or < 0.5 were taken as positive. In addition, spots showing no visible intensity above local background after the normalization procedure were rejected. After sequencing ≈ 300 clones, redundant clones were eliminated, finally resulting in 73 annotated genes (Tables 1 and 2; Supplementary material). These genes could be grouped into nine classes according to their function (Table 1). Of the 73 affected genes, 57 turned out to be up-regulated and 16 to be down-regulated (Table 2; Supplementary material). As a control for the validity of microarray results, real-time PCR amplification was carried out for selected genes (Table 2). Plants were grown in the presence of 150 µm CsCl and roots were harvested after 3 d and after 1.5 and 4 wk. These kinetic experiments were repeated up to five times to verify even very weakly induced or repressed genes, respectively (Table 2).

Table 1.  Composition of clusters of the isolated 133Cs-affected ESTs in roots of Arabidopsis thaliana Col-0
Major functional categoryUpDown
Cellular metabolism 95
Energy 2
Cell growth, division and development 12
Transcription and translation 72
Protein synthesis, folding and modification 62
Transport and homeostasis114
Cellular communication and signalling 2
Defence, stress response and detoxification16
Unknown genes111
Table 2.  Up-regulated and down-regulated transcripts in roots of Arabidopsis thaliana isolated by the SSH method and verified by semi-quantitative or real-time PCR
Accession numberAnnotationRT-PCR
  1. Induction (< 1.0) and repression (> 1.0) indicated, according to anova: ***, highly significant; **, very significant; *, significant; ns, not significant.

At3g22890ATP sulfurylase/APS kinase-like protein1.3*
At4g16260β-1,3-glucanase class I precursor1.9**
At1g12520Cu/Zn SOD copper chaperone-like protein1.3**
At2g41430Dehydration-induced protein (ERD15)1.5*
At3g55610δ-1-pyrroline-5-carboxylate synthetase3.7**
At5g02500DNAK-type molecular chaperone hsc70.12.7***
At1g02920Glutathione S-transferase1.3*
At1g08090High-affinity nitrate transporter NRT21.8**
At4g11650Osmotin precursor-like protein2.2***
At4g34870Peptidylprolyl isomerase (cyclophilin)1.3*
At3g56150Probable eukaryotic translation-initiation factor 3 subunit 81.3*
At1g72360Putative AP2 domain-transcription factor2.2**
At2g43570Putative endochitinase1.6**
At1g08830Superoxide dismutase – copper/zinc signatures1.6**
At2g37170Aquaporin (plasma membrane intrinsic protein 2b)0.7**
At1g73330Dr4 putative protease inhibitor0.6*
At3g01220Putative homeobox–leucine zipper protein, HAT70.9ns
At3g25830Putative limonene cyclase0.6**
At1g72050Transcription factor IIIA – C2H2-type zinc finger domain0.8ns

Functional classification of the isolated genes

One group of transcripts is involved in general metabolism (Table 1), including S-adenosyl-l-homocysteinase, anthralinate synthase, acetylserine thiolyase and 1-pyrroline-5-carboxylate synthase, respectively, which are important for amino acid synthesis. ATP sulfurylase and adenosyl phosphosulfate (APS) kinase are important enzymes for the assimilation of sulphur. It has been shown that Cs changes the stability of SH groups (Sgarrella et al., 1983). As the above-mentioned enzymes are important for the formation of sulphuric compounds, they may counteract the negative effect of Cs. RT–PCR for pyrroline-5-carboxylate synthase and ATP sulfurylase resulted in an induction of 3.7 and 1.3 after 1.5 wk growth on 150 µm Cs (Table 2). Other transcripts of this group, such as limonene cyclase and cycloartenol synthase, involved in terpene biosynthesis, were reduced with Cs stress.

Cytochrome c and the β-chain of mitochondrial ATP synthase were induced upon Cs application. These genes are important for redox reactions and the mitochondrial respiration chain, and thus were classified into the energy group (Tables 1 and 2; Supplementary material).

The third group, belonging to cell growth, cell division and plant development, showed a Cs-induced membrane protein (annexin), a phospholipid-binding protein (Barton et al., 1991). The gene actin2, a main component of the cytoskeleton, was repressed. The gene for the homeobox–leucin zipper protein Hat7 showed a slight reduction (factor of 0.8) after growth for 4 wk on CsCl. Proteins with a homeobox domain are often important in developmental processes (Scott et al., 1998). The Hat7 transcription factor leads up to the next group of genes, involved in transcription and translation.

The initiation factor IF3 and elongation factor EF1 are important for the translation machinery, and shown to be induced upon Cs application (Table 2; Supplementary material). An induced polyA-binding protein also belongs in this class. Motifs for mRNA recognition of polyA-binding proteins are present in numerous proteins that regulate splicing and translation processes (Bandziulis et al., 1989; Birney et al., 1993). A transcription factor of the AP2 family binds to specific consensus sequences of specific promoters, thus stimulating transcription, and was induced by a factor of 2.6 after 4 wk of Cs treatment (Table 2; Supplementary material). AP2-TF are often involved in defence and developmental processes of plants (Liu et al., 1998). Recently it was shown that AP2-TF with the same accession number as given in this paper is induced by K deprivation (Shin & Schachtman, 2004). This suggests similarities in gene regulation between Cs stress and K deprivation. As the internal K concentration is reduced by 150 µm Cs (Fig. 3b), this reduction might be responsible for the up-regulation of AP2-TF. Transcripts of a ribosomal protein L19 and a protein with a C2H2-type zinc finger domain were reduced (Table 2; Supplementary material).

Genes involved in protein synthesis, folding and modification were also affected by Cs (Table 1). Hsp70 and a DNAJ-like protein were induced by Cs (Supplementary material). Heat-shock proteins are well known to be induced by several stress factors, and DNAJ-like proteins may interact in this process. Similarly a copper chaperone and a cyclophilin, involved in protein folding and activation, were induced by Cs (Supplementary material) (Matouschek et al., 1995; Torres et al., 2001). The strong reducing effect of Cs results in a destabilization of proteins, e.g. a cleavage of disulfide bridges. In addition, an interaction with OH and COO groups is possible, and binding of Cs to bacterial cell membranes has been reported (Ghosh et al., 1993). Thus the induction of chaperones could protect the plant against protein-damaging effects of Cs. A leucine aminopeptidase and an acetylglucosamine phosphotransferase, involved in the transfer of glycosyl groups, also belong to the induced transcripts of this group. A reduced transcript amount was found for two protease inhibitors, a Dr4 as well as a trypsin inhibitor (Supplementary material).

Induced transcripts involved in transport and homeostasis are given in Tables 1 and 2. Potassium plays an important role in the regulation of the osmotic potential and pH gradient of cells. Internal K concentrations changed by an external application of Cs (Fig. 3b) may result in altered osmotic equilibrium, which can be balanced by transcripts coding for proteins involved in osmotic regulation. Cs-induced transcripts were a vacuolar-sorting receptor (Paris et al., 1997); a proton pump H+-ATPase type 2; and a high-affinity nitrate transporter NRT2 (factor 1.8 after 1.5 wk) coupled to the degradation of proton gradients (Table 2; Okamoto et al., 2003). A carnitine/acylcarnitine translocase, a carrier protein involved in mitochondrial energy transfer, was also induced. Dehydration protein ERD15 and an osmotin precursor, both important in the regulation of osmotic stress, were up-regulated by a factor of 1.5 and 2.3, respectively (Table 2; Kiyosue et al., 1994; Capelli et al., 1997). The formation of proline by the induced pyrroline-5-carboxylate synthase is given above, but its function as an osmotically active substance is also important (Yoshiba et al., 1997). Genes for water-selective channels such as the aquaporine PIP1b, PIP2b and TIP, important for bidirectional H2O transport (Baiges et al., 2002), as well as the K transporter HAK5, were reduced with growth on Cs (Table 2; Supplementary material).

Only two transcripts for cellular communication and signal transduction (Table 1), a kinase-like and a calcium-binding protein, were induced by Cs (Supplementary material).

Genes coding for proteins important in cellular defence, stress response and detoxification (Table 1) include a β-1,3-glucanase precursor (1.9), which has been shown to be induced by several stress factors (Simmons, 1994). A putative endochitinase (1.6) and PR1 protein were also induced by CsCl application (Table 2; Supplementary material). Pathogenesis-related (PR) proteins are synthesized upon pathogen infection, but are also induced by abiotic stress factors such as ozone or heavy metal ions (Eckey-Kaltenbach et al., 1997; Datta & Muthukrishnan, 1999 and articles therein; Heidenreich et al., 1999). Transcripts for several isoforms for glutathione S-transferases, important for detoxification processes, as well as a superoxide dismutase (2.7), belong to induced transcripts in this classification (Table 2; Supplementary material). As all these transcripts are induced with oxidative stress (Inzé & Montagu, 2002 and articles therein), Cs can also be classified as an abiotic oxidative stress factor. Finally, two peroxidases and a cytochrome P450 mono-oxygenase are also presented in this Cs-induced gene cluster (Table 2; Supplementary material). This is in agreement with increased peroxidase and catalase activities upon Cs application (Ghosh et al., 1993). Two peroxidases and a glutathione S-transferase (GST) were also induced upon K deprivation (Shin & Schachtman, 2004). Thus, as mentioned above for AP2-TF, reduction of the internal K concentration by Cs application might be responsible for these induced transcripts. It was shown by Shin & Schachtman (2004) that hydrogen peroxide mediates the plant root-cell response to K deprivation. As reactive oxygen species can also be produced by Cs, this indicates again that Cs can act as an oxidative stress factor. Induced transcripts for extensin 3, for a glycine- and proline-rich protein as well as for a pectin esterase, were also grouped in this class. They are important in cell-wall formation and rigidity, and were shown to be induced with biotic, as well as abiotic, stress factors such as ozone or mercuric ions (Schneiderbauer et al., 1995; Heidenreich et al., 2001; Yoshiba et al., 2001). In the latter group, genes are summarized that cannot be exactly defined or classified. However, known motifs and domains within their sequences are given (Supplementary material).

Conclusion

The present study identified a number of marker genes for Cs at an external nontoxic concentration of Cs. Many of these genes have been shown also to be induced by other stress factors, indicating that truly stress-specific gene regulation is scarce. The up-regulation of anti-oxidative defence genes and the similarities to K deprivation suggest an oxidative stress factor of nontoxic Cs concentrations. Further studies will focus on the analysis of selective K-ion channels in wild type and mutants of Arabidopsis, as well as on a direct influence of radiocaesium on gene expression. This will contribute to the identification of crops for cultivation with a reduced uptake of radiocaesium, which might be an alternative to expensive countermeasure applications in contaminated areas.

Supplementary material

The following material is available as supplementary material at http://www.blackwellpublishing.com/products/journals/suppmat/NPH/NPH1282/NPH1282sm.htm

Table S2 (Supplementary to Table 2) Up-regulated and down-regulated transcripts in roots of Arabidopsis thaliana isolated by the suppression subtractive hybridization (SSH) method.

Ancillary