TY - JOUR
AU1 - Craciun, Adrian Radu
AU2 - Courbot, Mikael
AU3 - Bourgis, Fabienne
AU4 - Salis, Pietrino
AU5 - Saumitou-Laprade, Pierre
AU6 - Verbruggen, Nathalie
AB - Abstract Cadmium (Cd) tolerance seems to be a constitutive species-level trait in Arabidopsis halleri. In order to identify genes potentially implicated in Cd tolerance, a backcross (BC1) segregating population was produced from crosses between A. halleri ssp. halleri and its closest non-tolerant relative A. lyrata ssp. petraea. The most sensitive and tolerant genotypes of the BC1 were analysed on a transcriptome-wide scale by cDNA-amplified fragment length polymorphism (AFLP). A hundred and thirty-four genes expressed more in the root of tolerant genotypes than in sensitive genotypes were identified. Most of the identified genes showed no regulation in their expression when exposed to Cd in a hydroponic culture medium and belonged to diverse functional classes, including reactive oxygen species (ROS) detoxification, cellular repair, metal sequestration, water transport, signal transduction, transcription regulation, and protein degradation, which are discussed. Arabidopsis halleri, cadmium, cDNA-AFLP, heavy metal tolerance, hyperaccumulator Abbreviations Abbreviations AFLP amplified fragment length polymorphism BC backcross CTR-TDF Cd tolerance-related transcript-derived fragment EST expressed sequence tag GST glutathione S-transferase ROS reactive oxygen species TDF transcript-derived fragment TRX thioredoxin VHA V-type ATPase Introduction Cadmium (Cd) is a heavy metal with highly toxic effects on organisms, and Cd pollution is recognized as an environmental problem worldwide. Although not essential for plant growth, Cd2+ is readily taken up by roots and can be translocated into aerial organs, where it affects photosynthesis and ‘consequently’ root and shoot growth. No Cd-specific transporter has been identified, and Cd seems to be transported via several classes of Ca2+, Fe2+, or Zn2+ transporters, affecting their uptake and distribution in plants and inducing deficiency (Clemens et al., 1998, 2002). Targets of Cd toxicity include zinc metalloenzymes and membrane phospholipids, causing inhibition of enzyme activities, protein denaturation, disruption of cell transport processes, alteration of RNA synthesis (Clemens, 2001; Suzuki et al., 2001; Hall, 2002; Schützendübel and Polle, 2002; Deckert, 2005; Kovalchuk et al., 2005), and DNA mismatch repair system inhibition (Jin et al., 2003; Banerjee and Flores-Rozas, 2005). Cd also induces an oxidative stress via indirect actions such as the disruption of the electron transport chain, the induction of lipid peroxidation, the depletion of glutathione or the displacement of Fe atoms from proteins resulting in increased production of reactive oxygen species (ROS) through the Haber–Weiss reaction (Smeets et al., 2005). Furthermore, Cd is considered as a mutagen. Plants have a variety of potential mechanisms at the cellular level that might be involved in the tolerance to heavy metal stress. These are involved primarily in avoiding the build-up of toxic concentrations at sensitive sites within the cell. Within the cytosol, tolerance mechanisms can include the chelation of metals by organic acids, amino acids, or peptides, their sequestration in the vacuole, their efflux to the apoplasm, and the repair of damaged proteins. In the case of Cd, the main mechanism of cellular tolerance in plants described up to now is its chelation by high affinity low molecular weight ligands, such as glutathione, phytochelatins, and metallothioneins (Clemens, 2001; Cobbet and Goldsbrough, 2002; Hall, 2002, Clemens and Simm, 2003), to limit uncontrolled binding to physiologically important functional groups. However, it has been demonstrated that the physiological mechanism of Cd tolerance is not based on an enhanced synthesis of phytochelatins (Ebbs et al., 2002; Schat et al., 2002) and that in tolerant plants only a small proportion of Cd seems to be co-ordinated by sulphur donor atoms (Küpper et al., 2004; Ueno et al., 2005). Rather, the majority of Cd was co-ordinated with malate in leaves of Thlaspi caerulescens, which was by far the most concentrated vacuolar ligand (Ueno et al., 2005). In Schizosaccharomyces pombe or Saccharomyces cerevisiae, PC-Cd or GS-Cd sequestration in the vacuole is essential for tolerance. Although vacuolar Cd-binding complexes have been measured (Rauser, 2000), the transport pathways of free or complexed Cd through the tonoplast have not been discovered in plants yet. AtCAX2 might function as a Cd2+/H+ antiporter at the tonoplast (Hirschi et al., 2000), and ABC members are probably involved in the transport of Cd chelates, as they are in yeast (Bovet et al., 2003). Interestingly 17 of the 127 putative ABC Arabidopsis thaliana genes were induced after Cd treatments (Bovet et al., 2005). Most plant species are not tolerant to Cd and most of the metallophytes on Cd-contaminated soils are Cd excluders, inhibiting its entry and favouring its retention in roots. However, a rare class of plants, called hyperaccumulators, can accumulate >100 μg of Cd g−1 of their shoot dry weight, without showing any sign of phytotoxicity (Brooks, 1998). Arabidopsis halleri (L.) O'Kane and Al-Shehbaz has been shown to hyperaccumulate Zn and tolerate Cd, and for some populations to hyperaccumulate Cd. Zn and Cd tolerance traits seem to be constitutive, as revealed by a study on a large number of European populations of A. halleri growing on both metal-contaminated and non-contaminated sites (Bert et al., 2002). Arabidopsis halleri is an emerging model species for the molecular elucidation of plant metal hyperaccumulation (Küpper et al., 2000; Dräger et al., 2004; Weber et al., 2004). Thlaspi caerulescens and A. halleri are the only two Cd hyperaccumulator species. The active accumulation in the above-ground parts of hyperaccumulator plants provides a promising approach both for cleaning anthropogenically contaminated soils (phytoremediation) and for commercial extraction (phytomining) of metals from naturally metal-rich (serpentine) soils (McGrath et al., 1993, 2002). To evaluate the potential of hyperaccumulator-mediated remediation, the genetics and physiology of tolerance and hyperaccumulation have first to be investigated. The Zn- and Cd-tolerant and hyperaccumulator species A. halleri ssp. halleri (population of Auby, France) was crossed with the Cd- and Zn-sensitive species A. lyrata ssp. petraea. One F1 plant was backcrossed with the sensitive parent species, giving a backcross progeny. The grandparents, the F1, and some of the backcross genotypes were previously characterized for their Cd and Zn tolerance and hyperaccumulation properties, which appeared to be independent characters (Bert et al., 2003). This unique plant material was analysed with an improved cDNA-amplified fragment length polymorphism (AFLP) protocol (Breyne et al., 2003) for identifying candidate genes responsible for Cd tolerance, by comparing the transcript profile of the tolerant and sensitive genotypes and selecting the co-segregating transcript-derived fragments (TDFs) expressed only/mostly/differently in the tolerant genotypes. Analysis has been conducted on roots. Root is the first organ to perceive Cd stress and to adapt to its presence in the soil. Moreover, Cd stress signals are first communicated by the root to the shoot. Although A. halleri is considered a Cd-hyperaccumulator species, Cd accumulates mainly in the roots (Bert et al., 2003). Roots of this population thus have exceptional mechanisms of cellular tolerance allowing them to grow on toxic concentrations (up to 300 μM) of Cd. Therefore, as a first step to unravel the mechanisms involved in Cd tolerance, the transcriptome of roots was studied. Materials and methods Plant material, growth conditions A single cross was performed between one individual from the Zn-tolerant species A. halleri ssp. halleri (Ah: pollen donor) and one from the non-tolerant species A. lyrata ssp. petraea (Alp 1: pollen recipient). The Ah individual (2n=16) originated from a site highly contaminated with Zn, Cd, and Pb (Auby, France) (Van Rossum et al., 2004). The Alp 1 individual (2n=16) originated from an uncontaminated site in the Czech Republic (Unhost, Central Bohemia) (Macnair et al., 1999). Both species are self-incompatible and usually outcrossing. One randomly selected F1 individual was used as male parent to fertilize a second A. lyrata ssp. petraea individual (Alp 2), generating the interspecific backcross progeny (BC1) of >350 individuals. Sixty-six of these genotypes were evaluated for their characters of Cd tolerance and hyperaccumulation (Bert et al., 2003). Between seven and 10 cuttings were generated per parental genotype (Ah, Alp 1 and 2, and F1), and per genotype of the BC1 already characterized for tolerance to both Cd and Zn. The 11 most sensitive (BCS) and the 18 most tolerant (BCT) genotypes of the BC1 were selected. Cuttings from mother plants were grown on sand for 6 weeks in a greenhouse. Hydroponic plant culture was performed in a modified MS solution consisting of K2SO4 (0.88 mM), KH2PO4 (0.25 mM), NaCl (10 μM), Ca(NO3)2 (2 mM), MgSO4 (1 mM), FeEDDHA (20 μM), H3BO3 (10 μM), ZnSO4 (1 μM), MnSO4 (0.6 μM), CuSO4 (0.1 μM), and (NH4)6Mo7O24 (0.01 μM) adjusted to pH 5.8, in a climate-controlled growth chamber (temperature cycle of 20/17 °C and a light (100 μmol m−2 s−1) cycle of 16 h light/8 h dark). The hydroponic solutions used were continuously aerated and changed every week. After 4 weeks in nutrient solution, 10 μM CdSO4 was added to half of the cuttings belonging to the same genotype. Roots were collected after 72 h of treatment and immediately frozen in liquid nitrogen until use. RNA extraction and cDNA synthesis Total RNA was prepared by LiCl precipitation (Sambrook and Russel, 2001). Four pools of RNA were made, by mixing 15–20 μg of total RNA extracted from the sensitive genotypes not treated or treated with 10 μM CdSO4 for 72 h, and tolerant genotypes not treated or treated with 10 μM CdSO4 for 72 h. A 6 μg aliquot of total RNA was used to synthesize first-strand cDNA by reverse transcription with a biotinylated oligo(dT)25 primer and Superscript II (N.V. Invitrogen SA, Merelbeke, Belgium). Second-strand synthesis was performed by strand displacement with Escherichia coli ligase (New England Biolabs, Westburg, Leusden, The Netherlands), E. coli polymerase I (New England Biolabs), and RNase H (Invitrogen). The resulting double-stranded cDNA was purified using a Qiaquick PCR purification kit (Qiagen, Westburg, Leusden, The Netherlands), quantified, and analysed by electrophoresis. cDNA-AFLP analysis A 500 ng aliquot of double-stranded cDNA was used for AFLP analysis as described (Vos et al., 1995; Bachem et al., 1996; Breyne et al., 2003) with the following modifications. The restriction enzymes used were BstYI and MseI (New England Biolabs), and the digestion was performed in two separate steps. After digestion with BstYI, the 3′ end fragments were collected on Dyna beads (Dynal, Invitrogen). After digestion with MseI, the restriction fragments released from the beads were collected and used as templates in the subsequent AFLP steps. The following adaptors were used: BstYI-F, 5′-CTCGTAGACTGCGTAGT-3′; BstYI-R, 5′-GATCACTACGCAGTCTAC-3′; MseI-F, 5′-GACGATGAGTCCTGAG-3′; and MseI-R, 5′-TACTCAGGACTCAT-3′; the primers for BstYI and MseI were 5′-GACTGCGTAGTGATC(T/C)N-3′ and 5′-GATGAGTCCTGAGTAANN-3′, respectively, where N represents the selective nucleotides. For pre-amplifications, an MseI primer without selective nucleotides was combined with a BstYI primer containing either a T or a C at the 3′ end. The use of either BstYI+T or BstYI+C primer in combination with the MseI primer without a selective nucleotide during the non-specific polymerase chain reaction (PCR) amplification reduced by 2-fold the complexity of the mixture of TDFs. The sensitivity of detection was enhanced by reducing the complexity of the mixture of amplified fragments, by the use of primers containing one or two additional selective nucleotides for the second amplification. A 5 μl aliquot of the amplification mixtures obtained following the non-selective amplification was used for final selective amplifications. Selective [33P]ATP-labelled amplification products were separated on a 5% polyacrylamide gel run at 100 W for 3 h. Gels were dried onto 3MM Whatmann paper, and positionally marked before being exposed for 2 d to Fuji Super RX Medical X-Ray film (Fujifilm Medical Systems, Benelux N.V., Belgium) to obtain the autoradiographs, or 1 d to PhosphorImager screens and scanned in a Storm 860 PhosphorImager (Molecular Dynamics). Amplifications using all the 128 possible BstYI +T/C+N+MseI +NN primer combinations were performed on each genotype and condition. Isolation and sequencing of amplified cDNA products The bands of interest were cut from the gels. DNA fragments were extracted from denaturing gels according to Frost and Guggenheim (1999). The gel fragments were rehydrated in 100 μl of 2× PCR buffer (100 mM KCl, 20 mM Tris–HCl, pH 9.0 at 25 °C, 3 mM MgCl2 and 0.2% Triton® X-100) for 10 min at room temperature. The remaining buffer was removed and replaced with a fresh 100 μl aliquot of buffer, samples were incubated at 94 °C for 90 min, and the gel slices were completely crushed by pipetting. A 20 μl aliquot was used for re-amplification using primers with selective nucleotides under the same PCR conditions as in the cDNA-AFLP analysis. Eventually, subsequent PCR amplification was performed to produce sufficient DNA for cloning. Purified DNA fragments were then cloned in pTZ57R/T vector (InsT/Aclone™ PCR Product Cloning Kit; Fermentas GmbH, Germany). Two or three individual clones, corresponding to the major insert and of the expected size, were isolated and sequenced using one plasmid-specific primer (Lac1 pUC18/19: 5′-AGTCACGACGTTGTAAAACGACGGCCAGT-3′). The nucleotide sequences obtained were compared with NCBI (http://www.ncbi.nlm.nih.gov/BLAST/) and TAIR (http://www.arabidopsis.org/Blast/) non-redundant databases using Blastn, Blastx, and tBlastx sequence alignment programs (Altschul et al., 1997). A total of 307 A. halleri expressed sequence tag (EST) sequences were submitted to the dbEST National Center for Biotechnology Information, Bethesda, MD, USA. They were assigned GenBank accession numbers, from DV752011 to DV752315, DV935717, and DV935718, and dbEST Id numbers, from 33963327 to 33963631, 34147988, and 34147989, respectively. Reverse transcriptase–PCR analysis Reverse primers (Table 3) were designed in the 3′ part of the selected TDF sequences, and the forward primers (Table 3) in regions homologous to the A. thaliana corresponding coding sequences. PCR primers were tested on genomic DNA extracted from Ah, Alp 1, and 2 using the Wizard Genomic DNA Purification Kit (Promega Benelux, Leiden, The Netherlands) to check the specificity and the efficiency. Only PCR primers giving similar results on both Ah and Alp DNA were used further for reverse transcription (RT)–PCRs. Reverse transcription was performed for 23 candidate genes, using the first-strand cDNA synthesis kit (Fermentas GmbH, Germany) following the manufacturer's instructions (2 μg of total RNA). cDNA fragments were amplified by PCR using the GoTaq DNA polymerase Master Mix (Promega) with the corresponding gene-specific primers, with an annealing temperature of 62 °C or 58 °C. PCR samples were taken at successive cycles; the number of cycles at which an optimum signal was observed is indicated in Table 3. The quantification was performed using the ImageQuant software (Amersham Biosciences, GE Healthcare Europe GmbH, Benelux, Belgium) before saturation. To ensure that equal amounts of cDNA were used in each PCR procedure, two cDNA fragments from housekeeping genes (that showed no expression changes in the cDNA-AFLP analysis), encoding elongation factor 1a and the ADP ribosylation factor (primers given in Table 3), were amplified simultaneously and quantified. Three independent experiments were conducted. Analysis of nine candidate genes (homologous to At3g06130, At2g40140, At1g61640, At5g64350, At2g17480, At4g11010, At5g63470, At5g47030, and At5g56030) did not produce results similar to the cDNA-AFLP. Results Selection of the Cd tolerance-related TDF The objective of this study was the detection of differences in the root transcript profile associated with the Cd tolerance character. Analysis was performed on plants belonging to A. halleri ssp. halleri, A. lyrata ssp. petraea, the F1, the 11 most sensitive (BCS) and 18 most tolerant (BCT) to Cd2+ BC1 genotypes, in control conditions, or following an exposure to CdSO4 (10 μM for 72 h). This treatment was compatible with adaptation of both the sensitive and the tolerant plants (Bert et al., 2003). The improved cDNA-AFLP protocol of Breyne et al. (2003) was used that permits transcript profiling for genome-wide screening of differentially expressed genes. The 128 possible BstYI+T/C+N+MseI+NN primer combinations were performed on each genotype or pool of BC genotypes and conditions. A total of 125–175 AFLP bands have been detected per combination on autoradiography, ranging from 70 to 600 bp (Fig. 1). Fig. 1 View largeDownload slide Left: autoradiography of a representative polyacrylamide gel. Amplicons resulting from three specific primer combinations were separated electrophoretically. There are 15 lanes for one specific primer combination, corresponding to A. petraea (Ap), A. halleri (Ah), F1, backcross-sensitive (BCS), and backcross-tolerant (BCT) genotypes, exposed (+) or not (–) to 10 μM CdSO4. Plants were cultivated for 4 weeks in hydroponic conditions and half of the plants 10 μM CdSO4. Roots were harvested after 72 h of were treated with treatment. Right: magnification of a region of the autoradiograph where differentially expressed TDFs between tolerant and sensitive genotypes are clearly visible. Encircled TDFs were cut out of the gel, re-amplified by PCR using the same primer combination, subcloned in the pTZ57R vector, sequenced, and the sequence analysed in silico. Fig. 1 View largeDownload slide Left: autoradiography of a representative polyacrylamide gel. Amplicons resulting from three specific primer combinations were separated electrophoretically. There are 15 lanes for one specific primer combination, corresponding to A. petraea (Ap), A. halleri (Ah), F1, backcross-sensitive (BCS), and backcross-tolerant (BCT) genotypes, exposed (+) or not (–) to 10 μM CdSO4. Plants were cultivated for 4 weeks in hydroponic conditions and half of the plants 10 μM CdSO4. Roots were harvested after 72 h of were treated with treatment. Right: magnification of a region of the autoradiograph where differentially expressed TDFs between tolerant and sensitive genotypes are clearly visible. Encircled TDFs were cut out of the gel, re-amplified by PCR using the same primer combination, subcloned in the pTZ57R vector, sequenced, and the sequence analysed in silico. In order to select TDFs associated with Cd tolerance, a comparison was made of the profiles of BCS and BCT. The TDFs that were differentially abundant between the two types of pools, with or without 10 μM CdSO4 treatment, were selected only when the same tendency was also observed between the original parental species. As a result of cDNA-AFLP analysis, a total of 194 TDFs were identified as differentially expressed and segregating along with the Cd tolerance between the Cd-sensitive and -tolerant genotypes, which were named Cd tolerance-related TDFs (CTR-TDFs). Of these, 97% were more abundant in tolerant than in sensitive genotypes, while the remaining 3% were only detected in sensitive genotypes. Among the overexpressed CTR-TDFs (188), all were already more abundant in the absence of CdSO4, and most of them (149) were not affected by the CdSO4 treatment. In the presence of CdSO4, 20 CTR-TDFs were further induced, seven only in tolerant and 13 in tolerant genotypes, sensitive genotypes, or pools of genotypes. Nineteen were repressed, five only in tolerant and 14 in both tolerant and sensitive genotypes. The 188 overexpressed CTR-TDFs, which all correspond to A. halleri genes, were further analysed. Bands were cut out in duplicate from lanes of tolerant genotypes, and PCR amplified (Fig. 1). The recovered fragments, ranging in length between 90 and 500 bp (250 bp on average), were cloned before sequencing in order to prevent problems associated with direct sequencing of PCR products (Durrant et al., 2000; Ditt et al., 2001) and the correct sequences were obtained for 156 TDFs. The remaining 32 CTR-TDFs could not be amplified from the polyacrylamide gels. Based on pairwise comparisons of the 156 sequences of recovered fragments, 134 (86%) showed similarity to reported Arabidopsis thaliana genes, but 22 of them did not show homology to any nucleotide or amino acid sequence in databases. The 134 sequences of TDFs showing homology with A. thaliana open reading frames (ORFs), that were isolated following the transcript profiling analysis, are listed in Table 1. Table 1 Properties of cloned TDFs expressed only in tolerant genotypes or with a higher expression in the tolerant genotypes, identified by cDNA-AFLP     View Large While most CTR fragments were identified only in one primer combination, 5% were present in several. Redundancy seemed to be due to the presence of highly homologous sequences, resulting from mispriming during PCR amplification, or PCR products representing either different alleles from the same gene or different instances of multicopy genes. Bachem et al. (1996) reported the occurrence of TDFs with the same mobility in fingerprints obtained with primers having similar sequence extensions and showed this to be due to mismatched primed PCR of highly abundant transcripts. Among these CTR-TDFs are those homologous to A. thaliana genes coding for AAA-type ATPases At3g15120 and At2g27600, for transcription factors At2g40140 and At1g78280, for histone H1 At2g18050, enolase At2g36530, mitochondrial inner membrane translocase At2g37410, and heat shock protein (HSP) At5g56030. The 134 sequences of CTR-TDFs corresponding to A. thaliana genes were classified according to functional class based upon MIPS classification (http://mips.gsf.de/proj/thal/db/tables/tables_func_frame.html): transcription, transport, cellular metabolism, signal transduction, cell rescue/stress-related/defence, and unknown function. The largest category of identified genes was in cellular metabolism (53%). CTR-TDFs showed 65–100% nucleotide identity to A. thaliana coding sequences, with an average of 91%. Identification of candidate genes and confirmation by semi-quantitative RT–PCR In order to validate the results presented above, candidates with possible roles in adaptation to toxic metal concentration were selected for further characterization. All the selected genes were more highly expressed in pools of tolerant genotypes than in the sensitive ones (Table 1). Expression was reassessed by semi-quantitative RT–PCR, with gene-specific primers, on RNA coming from the same genotypes obtained from experiments independent of those analysed by cDNA-AFLP (Fig. 2). Fig. 2 View largeDownload slide RT–PCR analysis of the steady-state transcript levels of candidate genes in A. petraea (Ap), A. halleri (Ah), F1, backcross-sensitive (BCS), and backcross-tolerant (BCT) genotypes. Plants were cultivated for 4 weeks in hydroponic conditions. Roots were harvested 72 h after being exposed (+) or not (–) to 10 μM CdSO4. Total RNA was extracted and used as a template for cDNA synthesis. PCR products were amplified using the specific primers given in Table 3. The presented photographs are representative of three independent extractions. Fig. 2 View largeDownload slide RT–PCR analysis of the steady-state transcript levels of candidate genes in A. petraea (Ap), A. halleri (Ah), F1, backcross-sensitive (BCS), and backcross-tolerant (BCT) genotypes. Plants were cultivated for 4 weeks in hydroponic conditions. Roots were harvested 72 h after being exposed (+) or not (–) to 10 μM CdSO4. Total RNA was extracted and used as a template for cDNA synthesis. PCR products were amplified using the specific primers given in Table 3. The presented photographs are representative of three independent extractions. Among the 19 A. halleri (Ah) genes belonging to the transcription and chromatin remodelling category, putatively involved in the tolerance to Cd, two were chosen for the confirmation by semi-quantitative RT–PCR: histone H1-3 and bZIP23 transcription factor homologues. Transcript levels of the H1-3 gene (homologous to At2g18050) were ∼2.5-fold higher in Ah than in A. petraea (Ap) and 3-fold higher in BCT compared with BCS. The CdSO4 treatment repressed the expression of this gene both in Ah (3-fold) and in Ap (2-fold). The bZIP23 gene (homologous to At2g16770) was expressed ∼2.5-fold more in Ah than in Ap and 2-fold more in BCT versus BCS. Gene expression was induced by Cd in Ap (2-fold), in Ah (1.5-fold), and in F1 (4-fold). Several genes identified by this approach have a putative function in transport. Expression profiles of five of them were tested and confirmed as being constitutively up-regulated in the tolerant genotypes. The Ah gene homologous to AtMTP1 (At2g47830) was selected; it belongs to the CDF family and is directly implicated in metal transport. The observed expression was maximal in Ah where it was 5-fold higher than in Ap. A 2-fold difference was confirmed between the sensitive and tolerant pools of BC1 plants. Expression was stimulated by Cd in BCT plants, but did not reach the level of expression observed in Ah. Furthermore, an up-regulation (5–6-fold) of the proton-ATPase 16 kDa proteolipid AVA-P1 or VHA-c gene (homologous to At2g16510) in the tolerant genotypes compared with the sensitive genotypes was confirmed, with no particular regulation by Cd exposure. A gene belonging to the H+/Ca2+ exchangers was also selected; this is homologous to AtCAX8 (At5g17850). The CAX8 homologue was clearly overexpressed (10-fold) in Ah compared with Ap and in BCT compared with BCS (5-fold). As water transport was reported to be a target of Cd, a gene homologous to PIP2A (At3g53420; plasma intrinsic protein 2) was selected; this gene encodes a member of the plasma membrane aquaporins. A 2-fold overexpression in Cd-tolerant plants was confirmed between BCT and BCS, and between Ah and Ap after Cd treatment. As the last member in the transport category, one mitochondrial protein import component, TIM17 (homologous to At2g37410), was found to be expressed 2–3-fold more in Ah and in BCT, especially after Cd treatment. Out of 134 CTR-TDFs, 71 fell into the class of cellular metabolism. Five genes have been chosen: two in the lipid modification class (homologous to At2g34690 and At5g65110), two of the AAA-ATPase family (homologous to At2g27600 and to At3g15120), and one in protein turnover (homologous to At5g67250). Several genes putatively implicated in the regulation of lipid composition were found to be overexpressed. Two genes implicated in this control were selected: accelerated-cell-death11 (ACD 11, homologous to At2g34690) and acyl-coenzymeA oxidase (ACX2, homologous to At5g65110). A 2-fold difference in expression of ACD11 was observed between Ah and Ap and between BCT and BCS, and Cd treatment further enhanced expression in Ah. A similar difference was observed between Ah+Cd and Ah+Cd. ACX2 expression was >10-fold higher in Ah and in BCT compared with Ap and BCS, but was not regulated by Cd treatment. Three members of the AAA-type ATPase family (homologous to At2g27600, At3g15120, and At3g09840) were found and two of them were selected for confirmation. The transcript abundance of both genes, homologous to At2g27600 and to At3g15120, was 2-fold higher in Ah and in F1 compared with Ap. Expression was 1.5-fold higher in BCT compared with BCS as well as in BCT+Cd compared with BCS+Cd. The expression of the putative SKP-interacting partner 2 gene (homologous to SKIP2, At5g67250), which is a part of the ubiquitin-protein ligase, was reassessed. A 2-fold induction in Ah versus Ap and in BCT compared with BCS was confirmed. In the signal transduction category, two candidates, the CLE41 and the CXIP4 homologous genes, were re-analysed. Expression levels of CLE41 (homologous to At3g24770) showed a 5-fold overexpression in Ah and BCT compared with Ap and BCS, and a >10-fold overexpression for F1 versus Ap. The presence of Cd induced a 2-fold increase in CLE41 expression in Ap, Ah, and F1, but no significant change in BCS and BCT. The transcript abundance of the homologue of the A. thaliana gene encoding the CAX-interacting protein 4 (CXIP4; At2g28910), which is implicated in the activation of CAX1, was 3-fold more abundant in Ah and in F1 than in Ap. Overexpression in BCT compared with BCS was only visible upon Cd treatment. The expression of five genes confirmed by semi-quantitiative RT–PCR in four sensitive and four tolerant individual BC1 genotypes was analysed further (Fig. 3). EC100 values associated with those genotypes and previously determined by Bert et al. (2003) are given in Table 2. The expression of none of the candidate genes strictly co-segregated, but was, on average, more abundant in the BCT than in the BCS. Fig. 3 View largeDownload slide Expression of six genes confirmed by RT–PCR in four tolerant (BCT; BC11, BC66, BC176, and BC323) and four sensitive (BCS; BC108, BC121, BC177, and BC179) backcross genotypes. Plants were cultivated during 4 weeks in hydroponic conditions. Roots were harvested after 72 h of 10 μM CdSO4 exposure. Total RNA was extracted and used as a template for cDNA synthesis. PCR products were amplified using the specific primers given in Table 3. The presented photographs are representative of three independent extractions. Fig. 3 View largeDownload slide Expression of six genes confirmed by RT–PCR in four tolerant (BCT; BC11, BC66, BC176, and BC323) and four sensitive (BCS; BC108, BC121, BC177, and BC179) backcross genotypes. Plants were cultivated during 4 weeks in hydroponic conditions. Roots were harvested after 72 h of 10 μM CdSO4 exposure. Total RNA was extracted and used as a template for cDNA synthesis. PCR products were amplified using the specific primers given in Table 3. The presented photographs are representative of three independent extractions. Table 2 EC100 values of four tolerant (BC11, BC66, BC176, and BC323) and four sensitive (BC108, BC121, BC177, and BC179) backcross genotypes Backcross genotypes  EC100 (in μM Cd)  BC11  150  BC66  133  BC176  133  BC323  150  BC108  25  BC121  25  BC177  25  BC179  25  Backcross genotypes  EC100 (in μM Cd)  BC11  150  BC66  133  BC176  133  BC323  150  BC108  25  BC121  25  BC177  25  BC179  25  View Large Table 3 PCR primers, annealing temperature, and cycle numbers used in this study to obtain quantitative signals from each gene tested Genes  Forward primer 5′–3′  Reverse primer 5′–3′  T (°C)  No. of cycles  At1g07920  CAACATTGTGGTCATTGGCCACGTCGA  CTCCTTCTCAATCTCCTTACCAGAAC  62  24  At3g15120  AAAGTTCTGGAGCTCTTAGTTGGA  TTGTTCAACTGGTTCAGGCATGTT  62  35  At2g16510  CTTCAATGGGAGTGATGAGACCTGAGT  CCAACAAAATAACATCCTACACTC  62  39  At2g16770  TTTGCTTGTTGATATAAGAGGAAG  CACATGAATGATGATACTTCAAAC  58  36  At5g17850  CATTTGCTTCTTCTGTGCTGCT  ACCAAGCCTCATACGGTTACT  62  47  At2g18050  AACCGAAAACCACCACTCATCCTCCAT  GCTGTAGAAAAAGTATTCCTACGG  62  47  At3g24770  CCAGCTTCATCAACAATGGATCTAC  CTCACCGGTAAAAAACCAGATGT  62  30  At2g27600  GGAAACGAGAGTGAAGCTTCAGACGTA  TCAACCTTCTTCTCCAAACTCCTGTGTG  62  27  At2g28910  GATGTTGATTCTGAGATGGAGAGGA  GACCGCTTAGAAGAACCACGTCCA  62  30  At2g34690  ATGGCGGATTCGGAAGCAGATAAGCCA  ATGGTGCCGATGAATTGACATAGCTT  62  27  At2g37410  GGGAACACCAGAGACATCTCG  GGAGGAGCATCAAAACTCTCC  62  27  At2g47830  GGAGTGGTTTGATGATGGCAAATGCT  AAGCTGAATGCTGACACGAGCCACA  62  33  At3g53420  GTGGAAGCCGTTCCCGGAGAAGG  TTAGACGTTGGCAGCACTTCTGAAAGATCA  62  24  At3g62290  GGTCTCGATGCAGCTGGTAAGACTAC  TGTTAGAGAGCCAGTCAAGTCCCTCA  62  24  At5g65110  GCAATTGGTTGCTGATGTCCA  GCTTCGATGAACTTGGCAAGA  62  30  At5g67250  TCGAGGAATTCATGAAGCAGC  ACCACTAGTGTCCTCCTTTGT  62  35  Genes  Forward primer 5′–3′  Reverse primer 5′–3′  T (°C)  No. of cycles  At1g07920  CAACATTGTGGTCATTGGCCACGTCGA  CTCCTTCTCAATCTCCTTACCAGAAC  62  24  At3g15120  AAAGTTCTGGAGCTCTTAGTTGGA  TTGTTCAACTGGTTCAGGCATGTT  62  35  At2g16510  CTTCAATGGGAGTGATGAGACCTGAGT  CCAACAAAATAACATCCTACACTC  62  39  At2g16770  TTTGCTTGTTGATATAAGAGGAAG  CACATGAATGATGATACTTCAAAC  58  36  At5g17850  CATTTGCTTCTTCTGTGCTGCT  ACCAAGCCTCATACGGTTACT  62  47  At2g18050  AACCGAAAACCACCACTCATCCTCCAT  GCTGTAGAAAAAGTATTCCTACGG  62  47  At3g24770  CCAGCTTCATCAACAATGGATCTAC  CTCACCGGTAAAAAACCAGATGT  62  30  At2g27600  GGAAACGAGAGTGAAGCTTCAGACGTA  TCAACCTTCTTCTCCAAACTCCTGTGTG  62  27  At2g28910  GATGTTGATTCTGAGATGGAGAGGA  GACCGCTTAGAAGAACCACGTCCA  62  30  At2g34690  ATGGCGGATTCGGAAGCAGATAAGCCA  ATGGTGCCGATGAATTGACATAGCTT  62  27  At2g37410  GGGAACACCAGAGACATCTCG  GGAGGAGCATCAAAACTCTCC  62  27  At2g47830  GGAGTGGTTTGATGATGGCAAATGCT  AAGCTGAATGCTGACACGAGCCACA  62  33  At3g53420  GTGGAAGCCGTTCCCGGAGAAGG  TTAGACGTTGGCAGCACTTCTGAAAGATCA  62  24  At3g62290  GGTCTCGATGCAGCTGGTAAGACTAC  TGTTAGAGAGCCAGTCAAGTCCCTCA  62  24  At5g65110  GCAATTGGTTGCTGATGTCCA  GCTTCGATGAACTTGGCAAGA  62  30  At5g67250  TCGAGGAATTCATGAAGCAGC  ACCACTAGTGTCCTCCTTTGT  62  35  View Large Discussion The two Cd-hyperaccumulator species, T. caerulescens and A. halleri have been previously characterized to a certain extent; both of them belong to the Brassicaceae family and are also Zn hyperaccumulators. Arabidopsis halleri is considered as a model species for metal tolerance studies, mostly because of its close relationship to A. thaliana. Arabidopsis halleri species seems to be constitutively tolerant to Cd since all accessions of A. halleri tested to date showed Cd tolerance (Bert et al., 2002). Cd hyperaccumulation is, however, not constitutive in the species but is present in the Auby ecotype, which was used in this study. While hyperaccumulators are usually good translocators of heavy metals from the root to the shoot, A. halleri population Auby retains most of the absorbed Cd in the roots. Zn, however, is translocated much more than Cd, with reported shoot:root concentration ratios always higher than 1 (Küpper et al., 2000; Bert et al., 2003). However, Zn and Cd tolerance characters seem to co-segregate partially in A. halleri, underlying the fact that common mechanisms are responsible for both Cd and Zn tolerance (Bert et al., 2003). Those mechanisms are thought to rely on detoxification and compartmentation. Previous transcriptomic studies of A. halleri (of the non-Cd accumulator population of Langelsheim) have been performed, using the A. thaliana microarray chips containing probes for ∼8300 genes, to identify genes highly expressed in the non-treated roots (Weber et al., 2004) or Cd2+-exposed roots (Weber et al., 2006), and shoots of A. halleri grown under low or high Zn supply (Becher et al., 2004). For the Cd/Zn hyperaccumulator T. caerulescens, a new microarray chip was developed, containing 1900 cDNAs isolated from roots. This array was hybridized with reverse-transcribed RNA from two accessions of Zn-treated T. caerulescens plants originating from two different soil types (Plessl et al., 2005). This comparative transcript profiling resulted in the identification of genes that are constitutively highly expressed in the hyperaccumulators and genes affected by heavy metals. The current microarray technique, despite its strengths, still has limitations. The principal limitation for many organisms is that only a fraction of the genes for which either the DNA sequence or a cDNA clone is available can be investigated. Another additional disadvantage for microarrays is the difficulty in distinguishing among different transcripts of genes belonging to the same family, when using spotted arrays. Finally, there are two additional technical limitations that restrict the broad use of the current microarray technology, namely the requirement for large amounts of RNA and the sensitivity of hybridization; 40% (Breyne and Zabeau, 2001), 53% (Weber et al., 2004), or 60% (Becher et al., 2004) of the transcripts can be detected, whereas in the case of hybridization of A.halleri cDNA to A. thaliana chips, 25% (Weber et al., 2004) to 35% (Becher et al., 2004) of the genes were detected. In this study, a genome-wide comparative transcriptomic analysis was carried out by cDNA-AFLP between closely related genotypes mainly differentiated by the Cd tolerance character. The principal advantage of cDNA-AFLP compared with microarrays is that it allows genome-wide expression analysis in any species without prior sequence knowledge. In addition, it has the inherent advantage of identifying and assessing new genes. One drawback of this technique is that generating a global overview of gene expression patterns involves a time-consuming series of PCRs and data cannot readily be compared and merged. However, this technique offers the best alternative for performing transcript profiling in non-model plant species, and for in-depth analyses of gene expression, overcoming the problems caused by plant gene redundancy during microarrays studies (Reijans et al., 2003; Vuylsteke et al., 2006). In this study, genotypes coming from interspecific crosses were compared. Arabidopsis halleri and A. petraea are thought to have diverged ∼2.9 million years ago (X Vekemans, personal communication), which is half the time for the divergence between A. halleri and A. thaliana (Koch et al., 2001). The polymorphism between A. halleri and A. petraea probably introduces a bias in the comparison of transcript profiles. However, the comparison of pools of BC plants is thought to have reduced the impact of this problem and favoured the identification of TDFs associated with Cd tolerance. In strong support of this, the proportion of TDFs that were validated (61%; 14 out of 23) was similar to other cDNA-AFLP analyses that compared the same genotype in different growth conditions (Yang et al., 2003; Mao et al., 2004; Fusco et al., 2005). Lack of confirmation may be due to polymorphism problems, but also to difference in growth conditions and physiological status of the plants. From an estimated 80–85% of the transcriptome, corresponding to ±19 000 transcript tags per genotype and treatment condition scanned by cDNA-AFLP, 188 CTR-TDFs overexpressed in the tolerant genotypes were discovered. Sequences were obtained for 156 CTR-TDFs. Of these, 134 presented significant homology with A. thaliana genes with known or putative functions, whereas no homology was found to any nucleotide or amino acid sequences in the database for the remaining 22 CTR-TDFs. Similar findings were reported in cDNA-AFLP studies, even in recent articles on A. thaliana (De Paepe et al., 2004). These 22 sequences probably belong to the 3′ untranslated region (UTR) of genes, which is frequently more divergent between the genes of A. thaliana and A. halleri, or to regions close to the centromeric regions, for which even in A. thaliana the whole genomic sequence is as yet unknown. Several categories of genes expressed more in the tolerant genotypes and possibly involved in Cd tolerance are discussed below. Cellular detoxification and repair Cd is not a redox active metal ion; however, it can cause oxidative stress (Smeets et al., 2005). Detoxification can occur via proteins that either scavenge the ROS produced by Cd, or repair the cellular components damaged directly by Cd or derived ROS. In this category, sequences of thioredoxin-like 3 (homologous to At5g61440), thioredoxin H type 5 (homologous to At1g45145), glutathione S-transferase (GST, homologous to At1g65820), S-adenosylmethionine decarboxylase (SAMdc; homologous to At3g02470), a putative methionine sulphoxide reductase (homologous to At4g21850), an HSP (homologous to AtHSP81-2), and AAA-ATPases were identified. Thioredoxins (TRXs) play direct roles in the antioxidative system by regenerating peroxiredoxins oxidized by peroxides (Foyer and Noctor, 2005). In S. cerevisiae, Cd inactivates TRX (Vido et al., 2001), and TRX expression is up-regulated during Cd treatment in S. pombe (Bae and Chen, 2004). GSTs are considered to play an important role in cellular protection against oxidative stress by synthesizing glutathione-S-conjugates, and several members have been shown to be up-regulated during Cd treatment (Marrs and Walbot, 1997; Bae and Chen, 2004; Fusco et al., 2005). Dihydrolipoamide dehydrogenase 1 (At1g48030) may also play a role in cellular detoxification. It belongs to a family of enzymes with an oxidoreductase activity, which, besides established roles in the regulation of enzymes of carbon metabolism, may modulate the activity of key antioxidative enzymes (Lutziger and Oliver, 2001). Another gene associated with antioxidative defences is SAMdc, involved in polyamines synthesis. Higher levels of SAMdC were previously reported in plants during stress conditions (saline, drought, or external abcisic acid application). These plants also had a higher content of polyamines. Their antioxidant properties have been studied in various plant species under Cd-induced oxidative stress (Groppa et al., 2001). HSPs are involved in repair mechanisms of proteins damaged directly by Cd or by ROS indirectly produced by Cd stress (Bae and Chen, 2004). The homologue of HSP81-2 (At5g46030) was constitutively overexpressed in A. halleri and in BCT. Methionine sulphoxide reductase repairs oxidative damage to methionine residues arising from ROS and reactive nitrogen intermediates (Taylor et al., 2003). Several predicted genes for AAA-ATPases, which are implicated in the folding of proteins, have also been identified in the present screen (homologous to At2g27600, At3g15120, and At3g09840) and may also play a role in protein repair. Membranes are a main target of ROS, and many genes implicated in the synthesis of fatty acids or in the metabolism of lipids are regulated during stress, such as acyl-CoA oxidase in response to dehydration (Grossi et al., 1995), or to UV (Logemann and Hahlbrock, 2002), or β-hydroxyacyl-ACP dehydratase in response to Cd (Fusco et al., 2005). In this study, several genes possibly involved in fatty acid biosynthesis or modifications have been identified, such as an acyl-CoA oxidase homologue (At5g65110), a putative phospholipase (At4g38550), FAD2 and LCB2 homologues, an oleoyl-acyl-carrier hydrolase, and a homologue to ACD11, which encodes a sphingosine transfer protein (Brodersen et al., 2002). Metal sequestration There is evidence in the Cd/Zn hyperaccumulator T. caerulescens that a considerable proportion of the accumulated Cd is sequestered within the cell vacuole in the form of malate complexes (Ueno et al., 2005). In A. halleri, the forms of accumulated Cd have not been determined as yet, but malate has been identified as a main Zn ligand (Sarret et al., 2002). Active transport is required to sequester Cd from the cytoplasm into the vacuolar compartment. This might occur via primary active transport, i.e. through an ATP-energized transporter, or by secondary active transport through coupling to the gradient of protons. Several genes possibly involved in Cd compartmentation that are constitutively overexpressed in tolerant genotypes were identified, such as AhMTP1 (initially named ZAT1 in A. thaliana; Van Der Zaal et al., 1999), which encodes a vacuolar transporter of the CDF family. The gene has been identified, and is associated with Zn, Mn, Ni, and Co tolerance in a number of plant species (Van der Zaal et al., 1999; Assunção et al., 2001; Mäser et al., 2001; Persans et al., 2001; Delhaize et al., 2003; Desbrosses-Fonrouge et al., 2005), including in A. halleri (Dräger et al., 2004), where it has been associated with Zn compartmentalization, but not with Cd transport. Because most of the Cd-tolerant BC1 genotypes were also Zn tolerant, MTP1 might be associated with Zn tolerance. A novel and interesting gene candidate is the 16 kDa V-ATPase component c gene (homologous to At2g16510), for which constitutively higher transcript levels were found in the tolerant genotypes. The H+-translocating V-type ATPase (VHA) plays a central role in the growth and development of plant cells and has been localized not only in the tonoplast but also in other endomembranes, such as the endoplasmic reticulum (ER) or the Golgi apparatus. The proton electrochemical gradient formed by the VHA provides the primary driving force for the transport of numerous ions and metabolites against their electrochemical gradients, and thus may participate in intracellular membrane trafficking, sorting, and secretion. The 16 kDa proteolipid subunit (component c) is the principal integral membrane protein of the VHA complex that forms the proton channel responsible for translocating protons across lipid bilayers (Sze et al., 2002). No other component of this complex has been identified in the present analysis, meaning that they either are not limiting or that they escaped in the present analysis. Another candidate gene which may be involved in Cd compartmentation is the homologue to AtCAX8 (At5g17850). The transcript of this gene is constitutively overexpressed in A. halleri and induced in A. petraea following Cd treatment. CAX8 belongs to the H+/cation exchangers family but, to our knowledge, has not been characterized up to now. CAX2, another member of the family, was however characterized as a high affinity and high capacity H+/Cd2+ antiporter of the tonoplast (Hirschi et al., 2000), involved in metal sequestration in the vacuoles (Hirschi et al., 2000). In addition, CAX2 is constitutively overexpressed in the shoots of A. halleri compared with A. thaliana (Becher et al., 2004). A homologue to AtCXIP4 (At2g28910) was also identified, which was identified in A. thaliana as encoding a CAX1-interacting protein 4 (AtCXIP4). The CXIP4 homologue was overexpressed in A. halleri and further induced by the addition of Cd in BCT. The fact that members from the CDF and CAX transporter families are constitutively overexpressed in A. halleri roots and in tolerant genotypes derived from A. halleri supports their important role in the sequestration of metal. Hydric balance Cd has been previously reported to inhibit root water transport, and one of the targets of Cd toxicity is the PIP subfamily of aquaporins. In the present study on root transcriptome, an A. halleri gene homologous to the aquaporin AtPIP2A (At3g53420) was identified. Its expression was induced in sensitive and tolerant plants treated with Cd. RT–PCR confirmed a higher expression in the tolerant genotypes, and an induction by Cd in both A. petraea and A. halleri. In A. thaliana, PIP2A is expressed specifically in the vascular bundles, and the protein localizes to the plasma membrane (Kammerloher et al., 1994). Interestingly a PIP2 homologue was also induced by Cd treatment in Brassica juncea (Fusco et al., 2005), supporting that Cd affects water transport. Signal transduction Crucial regulatory roles in many aspects of plant development are played by members of the receptor-like kinase (RLK) gene family, but the ligands to which they bind are largely unknown. Two putative RLK genes (homologues to At1g73080 and At5g61560) have been identified in this study, as well as a CLE family member. However, none of the identified sequences have previously been associated with a heavy metal stress response. A growing body of evidence suggests that the CLE genes constitute an ancient, functionally conserved gene family that plays important roles in plant growth and development (Strabala et al., 2006). CLE ligands interact with members of the leucine-rich repeat (LRR)-RLK receptor family. Recently, CLE41 was partially characterized in A. thaliana (Strabala et al., 2006). The CLE41 (At3g24770) homologous gene was overexpressed in tolerant genotypes (A. halleri, F1, BCT). Transcriptional regulation In the present study, the histone AhH1-3 gene homologous to At2g18050, which was constitutively overexpressed in the tolerant genotypes in control growth conditions but repressed by Cd in both tolerant and non-tolerant species, was identified. H1-3 was previously associated with the stress response (Scippa et al., 2004). Polyamines and histones can influence the DNA structure by causing local DNA bending, which is an effective mechanism for promoting transcriptionally induced stresses (Peng and Jackson, 2000). Several genes belonging to the bZIP, zinc finger, and basic helix–loop–helix (bHLH) transcription factor families were also identified in the present study. Among them, the gene AhbZIP23 (homologous to At2g16770) is constitutively overexpressed in the tolerant genotypes compared with the sensitive ones, and induced by Cd, in both A. halleri and A. petraea. There are 75 distinct members of the bZIP family in A. thaliana, most of which are not functionally characterized. In fungi, bZIP transcription factors CAD1 and ZIP1 mediate resistance to Cd in S. cerevisiae and S. pombe, respectively (Wu et al., 1993; Harrison et al., 2005). The 19 predicted transcription factor genes, which were constitutively overexpressed in A. halleri when compared with A. petraea, suggest an important up-regulation of transcription. However, post-transcriptional regulations may also present modifications between A. halleri and A. thaliana. Protein degradation Proteolysis often plays a crucial role in the feedback control underlying transcription-dependent stress responses. The ubiquitin−proteasome pathway is one of the major pathways for targeted proteolysis in the eukaryotic cell (Hochstrasser, 1996; Hershko and Ciechanover, 1998). Several genes involved in the ubiquitin–proteasome pathway proteolysis were identified as constitutively more highly expressed in the Cd-tolerant genotypes. Some of them belong to the proteasomal complexes 26S and 20S, others encode F-box family proteins, ubiquitin ligases, polyubiquitin, and other ubiquitin-associated proteins, and three members of the AAA-type ATPase family, similar to those present in the 26S proteolitic complex (Table 1). Such an important number of genes involved in protein degradation processes led to the hypothesis that Cd2+ tolerance probably depends on efficient recycling of abnormal proteins. Some of the reported A. halleri genes were previously identified by microarray analysis with A. thaliana GeneChips, such as the AhMTP1 (At2g47830), the homologue to At2g48020, encoding a major facilitator superfamily putative sugar transporter (Becher et al., 2004), AhMlo8 (homologous to At2g17480), or the Ah gene encoding a 40S ribosomal protein homologous to At2g16360 (Weber et al., 2004, 2006). Others match A. thaliana homologues of the same family as those reported in the transcriptomic comparisons published by Becher et al. (2004) and Weber et al. (2004, 2006), such as the GST At1g65820, HSP At5g56030, zinc finger (AN1-like) family protein At2g36320, C2H2 zinc finger At1g04850, cytochrome P450 family protein At4g37370, zinc finger (C3HC4-type RING finger) family proteins At1g11020 and At2g42350, calcium-binding EF hand family proteins At2g41410 and At2g45380, peroxidase At5g64120, lipoxygenase 1 (At1g55020), or CAX8 (At5g17850). Many other genes were also revealed by this study. There are reasons for those differences; an important reason is that the population of A. halleri and the growth conditions are different. For example, the Langelsheim population used for the microarray analysis does not hyperaccumulate Cd, whereas the population from Auby does. Furthermore, the comparisons presented in this study are different and genes have been further investigated only if their corresponding TDF was more abundant in the pool of the tolerant BC1 genotypes than in the sensitive ones. For example, NAS2 was previously published as being overexpressed in the roots of A. halleri compared with A. thaliana (Weber et al., 2004), but a significant difference in expression between the pools of contrasting BC genotypes was not observed (data not shown). In addition, genes directly implicated in Cd chelation were not identified in the screen. This is consistent with the observation that there was also no up-regulation of the genes encoding the histidine biosynthetic pathway at the transcript level in Alyssum lesbiacum upon exposure to Ni, despite the fact there was a large increase in histidine biosynthesis (Ingle et al., 2005). Finally, because the cDNA-AFLP technology is based on a double digest, it can generate short fragments, which are lost in the analysis. This was, for example, the case for AhHMA4, which is however expressed more in tolerant genotypes than in sensitive ones (C Courbot, G Willems, P Motte, S Arvidsson, P Saumitou-Laprade and N Verbruggen, unpublished data). Conclusion The presented analysis has allowed the identification of new A. halleri genes, which are thought to be involved in the adaptation of the roots to tolerate high Cd concentrations. The transcriptome analysis revealed a large variation in genome expression, in controlled growth conditions, resulting from several million years of evolution between A. halleri and A. lyrata, and profound adaptation of the metabolism to high metal contents present in the environment of the A. halleri species. Many of the genes which were identified as overexpressed in the roots of Cd-tolerant genotypes (A. halleri, F1, and BCT) compared with the sensitive genotypes (A. lyrata ssp. petraea and BCS) are involved in cellular detoxification, repair, and sequestration, but the largest category belongs to the cellular metabolism pathways. Despite its limitations, the cDNA-AFLP method has been proven to be a convenient and effective technique for the transcriptomic comparison of different genotypes, without prior knowledge of the genome sequence. Genes implicated in the Cd tolerance of the Zn/Cd hyperaccumulator plant A. halleri were revealed and confirmed by the use of an exceptional plant material, and a collection of A. halleri ESTs has been generated (dbEST Id numbers, from 33963327 to 33963631, 34147988, and 34147989). This analysis has identified many putative target genes of Cd toxicity, which may be important to characterize further at the protein level, and it has confirmed the importance of detoxification and metal sequestration. However, the expression of none of the selected genes strictly co-segregates with the tolerance character, but was on average higher in tolerant genotypes of the backcross. Questions remain as to how this large array of genes is overexpressed in the tolerant genotypes. What is the sensing mechanism of the metal concentration in plants, which seems to be deregulated/different in hyperaccumulators? In the future, it will be crucial to answer these questions and to pinpoint the major genes in order to understand tolerance to a toxic environment and to engineer efficient plants for phytoremediation of Cd-contaminated soils. Quantitative trait locus analysis will be a major tool to address that task. We thank Valerie Bert, Ali Barkat, and Fabrizio Vankerkhoven for their help in the experimental techniques or data analyses. We acknowledge Sam Mugford for helpful discussion. The present investigation was supported by grants from the European Union (Metalhome RTN, HPRN-CT-2002-00243). This research was also supported by a grant from the Interuniversity Attraction Pole Programme: Belgian Science Policy (ProjectV/13) and the FNRS (project FRFC 2.4565.02). References Altschul SF,  Madden TL,  Schaffer AA,  Zhang JH,  Zhang Z,  Miller W,  Lipman DJ.  Gapped BLAST and PSI-BLAST: a new generation of protein database search programs,  Nucleic Acids Research ,  1997, vol.  25 (pg.  3389- 3402) Google Scholar CrossRef Search ADS PubMed  Assunção AGL,  Da Costa MP,  De Folter S,  Vooijs R,  Schat H,  Aarts MGM.  Elevated expression of metal transporter genes in three accessions of the metal hyperaccumulator Thlaspi caerulescens,  Plant, Cell and Environment ,  2001, vol.  24 (pg.  217- 226) Google Scholar CrossRef Search ADS   Bachem CWB,  van der Hoeven RS,  de Bruijn SM,  Vreugdenhil D,  Zabeau M,  Visser RGF.  Visualization of differential gene expression using a novel method of RNA fingerprinting based on AFLP: analysis of gene expression during potato tuber development,  The Plant Journal ,  1996, vol.  9 (pg.  745- 753) Google Scholar CrossRef Search ADS PubMed  Bae W,  Chen X.  Proteomic study for the cellular responses to Cd2+ in Schizosaccharomyces pombe through amino acid-coded mass tagging and liquid chromatography tandem mass spectrometry,  Molecular and Cellular Proteomics ,  2004, vol.  3 (pg.  596- 607) Google Scholar CrossRef Search ADS PubMed  Banerjee S,  Flores-Rozas H.  Cadmium inhibits mismatch repair by blocking the ATPase activity of the MSH2–MSH6 complex,  Nucleic Acids Research ,  2005, vol.  33 (pg.  1410- 1419) Google Scholar CrossRef Search ADS PubMed  Becher M,  Talke IN,  Krall L,  Krämer U.  Cross-species microarray transcript profiling reveals high constitutive expression of metal homeostasis genes in shoots of the zinc hyperaccumulator Arabidopsis halleri,  The Plant Journal ,  2004, vol.  37 (pg.  251- 268) Google Scholar CrossRef Search ADS PubMed  Bert V,  Bonnin I,  Saumitou-Laprade P,  de Laguerie P,  Petit D.  Do Arabidopsis halleri from nonmetallicolous populations accumulate zinc and cadmium more effectively than those from metallicolous populations? ,  New Phytologist ,  2002, vol.  155 (pg.  47- 57) Google Scholar CrossRef Search ADS   Bert V,  Meerts P,  Saumitou-Laprade P,  Salis P,  Gruber W,  Verbruggen N.  Genetic basis of Cd tolerance and hyperaccumulation in Arabidopsis halleri,  Plant and Soil ,  2003, vol.  249 (pg.  9- 18) Google Scholar CrossRef Search ADS   Bovet L,  Eggmann T,  Meylan-Bettex M,  Polier J,  Kammer P,  Marin E,  Feller U,  Martinoia E.  Transcript levels of AtMRPs after cadmium treatment: induction of AtMRP3,  Plant, Cell and Environment ,  2003, vol.  26 (pg.  371- 381) Google Scholar CrossRef Search ADS   Bovet L,  Feller U,  Martinoia E.  Possible involvement of plant ABC transporters in cadmium detoxification: a cDNA sub-microarray approach,  Environment International ,  2005, vol.  31 (pg.  263- 267) Google Scholar CrossRef Search ADS PubMed  Breyne P,  Dreesen R,  Cannoot B,  Rombaut D,  Vandepoele K,  Rombauts S,  Vanderhaeghen R,  Inzé D,  Zabeau M.  Quantitative cDNA-AFLP analysis for genome-wide expression studies,  Molecular Genetics and Genomics ,  2003, vol.  269 (pg.  173- 179) Google Scholar PubMed  Breyne P,  Zabeau M.  Genome-wide expression analysis of plant cell cycle modulated genes,  Current Opinion in Plant Biology ,  2001, vol.  4 (pg.  136- 142) Google Scholar CrossRef Search ADS PubMed  Brodersen P,  Petersen M,  Pike HC,  Olszak B,  Skov S,  Odum N,  Jørgensen LB,  Brown RE,  Mundy J.  Knockout of Arabidopsis accelerated-cell-death11 encoding a sphingosine transfer protein causes activation of programmed cell death and defense,  Genes and Development ,  2002, vol.  16 (pg.  490- 502) Google Scholar CrossRef Search ADS PubMed  Brooks RR.  Brooks RR.  Geobotany and hyperaccumulators,  Plants that hyperaccumulate heavy metals ,  1998 Wallingford, UK CAB International(pg.  55- 94) Clemens S.  Molecular mechanisms of plant metal tolerance and homeostasis,  Planta ,  2001, vol.  212 (pg.  475- 486) Google Scholar CrossRef Search ADS PubMed  Clemens S,  Antosiewicz DM,  Ward JM,  Schachtman DP,  Schroeder JI.  The plant cDNA LCT1 mediates the uptake of calcium and cadmium in yeast,  Proceedings of the National Academy of Sciences, USA ,  1998, vol.  95 (pg.  12043- 12048) Google Scholar CrossRef Search ADS   Clemens S,  Palmgreen MG,  Kraemer U.  A long way ahead: understanding and engineering plant metal accumulation,  Trends in Plant Science ,  2002, vol.  7 (pg.  309- 315) Google Scholar CrossRef Search ADS PubMed  Clemens S,  Simm C.  Schizosaccharomyces pombe as a model for metal homeostasis in plant cells: the phytochelatin-dependent pathway is the main Cd detoxification mechanism,  The New Phytologist ,  2003, vol.  159 (pg.  323- 329) Google Scholar CrossRef Search ADS   Cobbett C,  Goldsbrough P.  Phytochelatins and metallothioneins: roles in heavy metal detoxification and homeostasis,  Annual Review of Plant Biology ,  2002, vol.  53 (pg.  159- 182) Google Scholar CrossRef Search ADS PubMed  De Paepe A,  Vuylsteke M,  Van Hummelen P,  Zabeau M,  Van Der Straeten D.  Transcriptional profiling by cDNA-AFLP and microarray analysis reveals novel insights into the early response to ethylene in Arabidopsis,  The Plant Journal ,  2004, vol.  39 (pg.  537- 559) Google Scholar CrossRef Search ADS PubMed  Deckert J.  Cadmium toxicity in plants: is there any analogy to its carcinogenic effect in mammalian cells?,  Biometals ,  2005, vol.  18 (pg.  475- 481) Google Scholar CrossRef Search ADS PubMed  Delhaize E,  Kataoka T,  Hebb DM,  White RG,  Ryan PR.  Genes encoding proteins of the cation diffusion facilitator family that confer manganese tolerance,  The Plant Cell ,  2003, vol.  15 (pg.  1131- 1142) Google Scholar CrossRef Search ADS PubMed  Desbrosses-Fonrouge AG,  Voigt K,  Schroder A,  Arrivault S,  Thomine S,  Kramer U.  Arabidopsis thaliana MTP1 is a Zn transporter in the vacuolar membrane which mediates Zn detoxification and drives leaf Zn accumulation,  FEBS Letters ,  2005, vol.  579 (pg.  4165- 4174) Google Scholar CrossRef Search ADS PubMed  Ditt RF,  Nester EW,  Comai L.  Plant gene expression response to Agrobacterium tumefaciens,  Proceedings of the National Academy of Sciences, USA ,  2001, vol.  98 (pg.  10954- 10959) Google Scholar CrossRef Search ADS   Dräger DB,  Desbrosses-Fonrouge AG,  Krach C,  Chardonnens AN,  Meyer RC,  Saumitou-Laprade P,  Kramer U.  Two genes encoding Arabidopsis halleri MTP1 metal transport proteins co-segregate with zinc tolerance and account for high MTP1 transcript levels,  The Plant Journal ,  2004, vol.  39 (pg.  425- 439) Google Scholar CrossRef Search ADS PubMed  Durrant WE,  Rowland O,  Piedras P,  Hammond-Kosack KE,  Jones JDG.  cDNA-AFLP reveals a striking overlap in race-specific resistance and wound response gene expression profiles,  The Plant Cell ,  2000, vol.  12 (pg.  963- 977) Google Scholar CrossRef Search ADS PubMed  Ebbs S,  Lau I,  Ahner B,  Kochian L.  Phytochelatin synthesis is not responsible for Cd tolerance in the Zn/Cd hyperaccumulator Thlaspi caerulescens (J. & C. Presl),  Planta ,  2002, vol.  214 (pg.  635- 640) Google Scholar CrossRef Search ADS PubMed  Foyer CH,  Noctor G.  Redox homeostasis and antioxidant signalling: a metabolic interface between stress perception and physiological responses,  The Plant Cell ,  2005, vol.  17 (pg.  1866- 1875) Google Scholar CrossRef Search ADS PubMed  Frost MR,  Guggenheim JA.  Prevention of depurination during elution facilitates the re-amplification of DNA from differential display gels,  Nucleic Acids Research ,  1999, vol.  27 pg.  e6  Google Scholar CrossRef Search ADS PubMed  Fusco N,  Micheletto L,  Dal Corso G,  Borgato L,  Furini A.  Identification of cadmium-regulated genes by cDNA-AFLP in the heavy metal accumulator Brassica juncea L,  Journal of Experimental Botany ,  2005, vol.  56 (pg.  3017- 3027) Google Scholar CrossRef Search ADS PubMed  Groppa MD,  Tomaro ML,  Benavides MP.  Polyamines as protectors against cadmium or copper-induced oxidative damage in sunflower leaf discs,  Plant Sciences ,  2001, vol.  161 (pg.  481- 488) Google Scholar CrossRef Search ADS   Grossi M,  Gulli M,  Stanca AM,  Cattivelli L.  Characterization of two barley genes that respond rapidly to dehydration stress,  Plant Sciences ,  1995, vol.  105 (pg.  71- 80) Google Scholar CrossRef Search ADS   Hall JL.  Cellular mechanisms for heavy metal detoxification and tolerance,  Journal of Experimental Botany ,  2002, vol.  53 (pg.  1- 11) Google Scholar CrossRef Search ADS PubMed  Harrison C,  Katayama S,  Dhut S,  Chen D,  Jones N,  Bahler J,  Toda T.  SCF(Pof1)-ubiquitin and its target Zip1 transcription factor mediate cadmium response in fission yeast,  EMBO Journal ,  2005, vol.  24 (pg.  599- 610) Google Scholar CrossRef Search ADS PubMed  Hershko A,  Ciechanover A.  The ubiquitin system,  Annual Review of Biochemistry ,  1998, vol.  67 (pg.  425- 479) Google Scholar CrossRef Search ADS PubMed  Hirschi KD,  Korenkov VD,  Wilganowski NL,  Wagner GJ.  Expression of Arabidopsis CAX2 in tobacco. Altered metal accumulation and increased manganese tolerance,  Plant Physiology ,  2000, vol.  124 (pg.  125- 133) Google Scholar CrossRef Search ADS PubMed  Hochstrasser M.  Ubiquitin-dependent protein degradation,  Annual Review of Genetics ,  1996, vol.  30 (pg.  405- 439) Google Scholar CrossRef Search ADS PubMed  Ingle RA,  Mugford ST,  Rees JD,  Campbell MM,  Smith JA.  Constitutively high expression of the histidine biosynthetic pathway contributes to nickel tolerance in hyperaccumulator plants,  The Plant Cell ,  2005, vol.  17 (pg.  2089- 2106) Google Scholar CrossRef Search ADS PubMed  Jin YH,  Clark AB,  Slebos RJ,  Al-Refai H,  Taylor JA,  Kunkel TA,  Resnick MA,  Gordenin DA.  Cadmium is a mutagen that acts by inhibiting mismatch repair,  Nature Genetics ,  2003, vol.  34 (pg.  326- 329) Google Scholar CrossRef Search ADS PubMed  Kammerloher W,  Fischer U,  Piechottka GP,  Schaffner AR.  Water channels in the plant plasma membrane cloned by immunoselection from a mammalian expression system,  The Plant Journal ,  1994, vol.  6 (pg.  187- 199) Google Scholar CrossRef Search ADS PubMed  Koch M,  Haubold B,  Mitchell-Olds T.  Molecular systematics of the Brassicaceae: evidence from coding plastidic matK and nuclear CHS sequences,  American Journal of Botany ,  2001, vol.  88 (pg.  534- 544) Google Scholar CrossRef Search ADS PubMed  Kovalchuk I,  Titov V,  Hohn B,  Kovalchuk O.  Transcriptome profiling reveals similarities and differences in plant responses to cadmium and lead,  Mutation Research ,  2005, vol.  570 (pg.  149- 161) Google Scholar CrossRef Search ADS PubMed  Küpper H,  Lombi E,  Zhao FJ,  McGrath SP.  Cellular compartmentation of cadmium and zinc in relation to other elements in the hyperaccumulator Arabidopsis halleri,  Planta ,  2000, vol.  212 (pg.  75- 84) Google Scholar CrossRef Search ADS PubMed  Küpper H,  Mijovilovich A,  Meyer-Klaucke W,  Kroneck PMH.  Tissue- and age-dependent differences in the complexation of cadmium and zinc in the cadmium/zinc hyperaccumulator Thlaspi caerulescens (Ganges ecotype) revealed by X-ray absorption spectroscopy,  Plant Physiology ,  2004, vol.  134 (pg.  748- 757) Google Scholar CrossRef Search ADS PubMed  Logemann E,  Hahlbrock K.  Crosstalk among stress responses in plants: pathogen defense overrides UV protection through an inversely regulated ACE/ACE type of light-responsive gene promoter unit,  Proceedings of the National Academy of Sciences, USA ,  2002, vol.  99 (pg.  2428- 2432) Google Scholar CrossRef Search ADS   Lutziger I,  Oliver DJ.  Characterization of two cDNAs encoding mitochondrial lipoamide dehydrogenase from Arabidopsis,  Plant Physiology ,  2001, vol.  127 (pg.  615- 623) Google Scholar CrossRef Search ADS PubMed  Macnair MR,  Bert V,  Huitson SB,  Saumitou-Laprade P,  Petit D.  Zinc tolerance and hyperaccumulation are genetically independent characters,  Proceedings of the Royal Society B ,  1999, vol.  266 (pg.  2175- 2179) Google Scholar CrossRef Search ADS PubMed  Mao C,  Yi K,  Yang L,  Zheng B,  Wu Y,  Liu F,  Wu P.  Identification of aluminium-regulated genes by cDNA-AFLP in rice (Oryza sativa L.): aluminium-regulated genes for the metabolism of cell wall components,  Journal of Experimental Botany ,  2004, vol.  55 (pg.  137- 143) Google Scholar CrossRef Search ADS PubMed  Marrs KA,  Walbot V.  Expression and RNA splicing of themaize glutathione S-transferase Bronze2 gene is regulated by cadmium and other stresses,  Plant Physiology ,  1997, vol.  113 (pg.  93- 102) Google Scholar CrossRef Search ADS PubMed  Mäser P,  Thomine S,  Schroeder JI, et al.  Phylogenetic relationships within cation transporter families of Arabidopsis,  Plant Physiology ,  2001, vol.  126 (pg.  1646- 1667) Google Scholar CrossRef Search ADS PubMed  McGrath SP,  Sidoli CMD,  Baker AJM,  Reeves RD.  Eijsackers HJP,  Hamers T.  The potential for the use of metal-accumulating plants for the in situ decontamination of metal-polluted soils,  Integrated soil and sediment research: a basis for proper protection ,  1993 Dordrecht, The Netherlands Kluwer Academic Publishers(pg.  673- 677) McGrath SP,  Zhao FJ,  Lombi E.  Phytoremediation of metals, metalloids, and radionuclides,  Advances in Agronomy ,  2002, vol.  75 (pg.  1- 56) Peng HF,  Jackson V.  In vitro studies on the maintenance of transcription-induced stress by histones and polyamines,  Journal of Biological Chemistry ,  2000, vol.  275 (pg.  657- 668) Google Scholar CrossRef Search ADS PubMed  Persans MW,  Nieman K,  Salt DE.  Functional activity and role of cation-efflux family members in Ni hyperaccumulation in Thlaspi goesingense,  Proceedings of the National Academy of Sciences, USA ,  2001, vol.  98 (pg.  9995- 10000) Google Scholar CrossRef Search ADS   Plessl M,  Rigola D,  Hassinen V,  Aarts MG,  Schat H.  Transcription profiling of the metal-hyperaccumulator Thlaspi caerulescens (J. & C. PRESL),  Zeitschrift für Naturforschung ,  2005, vol.  60 (pg.  216- 223) Google Scholar PubMed  Rauser WE.  Roots of maize seedlings retain most of their cadmium through two complexes,  Journal of Plant Physiology ,  2000, vol.  156 (pg.  545- 551) Google Scholar CrossRef Search ADS   Reijans M,  Lascaris R,  Groeneger AO, et al.  Quantitative comparison of cDNA-AFLP, microarrays, and GeneChip expression data in Saccharomyces cerevisiae,  Genomics ,  2003, vol.  82 (pg.  606- 618) Google Scholar CrossRef Search ADS PubMed  Sambrook J,  Russel T. ,  Molecular cloning: a laboratory manual ,  2001 2nd edn. Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press Sarret G,  Saumitou-Laprade P,  Bert V,  Proux O,  Hazemann JL,  Traverse A,  Marcus MA,  Manceau A.  Forms of zinc accumulated in the hyperaccumulator Arabidopsis halleri,  Plant Physiology ,  2002, vol.  130 (pg.  1815- 1826) Google Scholar CrossRef Search ADS PubMed  Schat H,  Llugany M,  Vooijs R,  Hartley-Whitaker J,  Bleeker P.  The role of phytochelatins in constitutive and adaptative heavy metal tolerances in hyperaccumulator and non-hyperaccumulator metallophytes,  Journal of Experimental Botany ,  2002, vol.  53 (pg.  2381- 2392) Google Scholar CrossRef Search ADS PubMed  Schutzendubel A,  Polle A.  Plant responses to abiotic stresses: heavy metal-induced oxidative stress and protection by mycorrhization,  Journal of Experimental Botany ,  2002, vol.  53 (pg.  1351- 1365) Google Scholar CrossRef Search ADS PubMed  Scippa GS,  Di Michele M,  Onelli E,  Patrignani G,  Chiatante D,  Bray EA.  The histone-like protein H1-S and the response of tomato leaves to water deficit,  Journal of Biological Chemistry ,  2004, vol.  55 (pg.  99- 109) Smeets K,  Cuypers A,  Lambrechts A,  Semane B,  Hoet P,  Van Laere A,  Vangronsveld J.  Induction of oxidative stress and antioxidative mechanisms in Phaseolus vulgaris after Cd application,  Plant Physiology and Biochemistry ,  2005, vol.  43 (pg.  437- 444) Google Scholar CrossRef Search ADS PubMed  Strabala TJ,  O'Donnell PJ,  Smit AM,  Ampomah-Dwamena C,  Martin EJ,  Netzler N,  Nieuwenhuizen NJ,  Quinn BD,  Foote HC,  Hudson KR.  Gain-of-function phenotypes of many CLAVATA3/ESR genes, including four new family members, correlate with tandem variations in the conserved CLAVATA3/ESR domain,  Plant Physiology ,  2006, vol.  140 (pg.  1331- 1344) Google Scholar CrossRef Search ADS PubMed  Suzuki N,  Koizumi N,  Sano H.  Screening of cadmium-responsive genes in Arabidopsis thaliana,  Plant, Cell and Environment ,  2001, vol.  24 (pg.  1177- 1188) Google Scholar CrossRef Search ADS   Sze H,  Schumacher K,  Muller ML,  Padmanaban S,  Taiz L.  A simple nomenclature for a complex proton pump: VHA genes encode the vacuolar H(+)-ATPase,  Trends in Plant Science ,  2002, vol.  7 (pg.  157- 161) Google Scholar CrossRef Search ADS PubMed  Taylor AB,  Benglis DMJr,  Dhandayuthapani S,  Hart PJ.  Structure of Mycobacterium tuberculosis methionine sulfoxide reductase A in complex with protein-bound methionine,  Journal of Bacteriology ,  2003, vol.  185 (pg.  4119- 4126) Google Scholar CrossRef Search ADS PubMed  Ueno D,  Ma JF,  Iwashita T,  Zhao FJ,  McGrath SP.  Identification of the form of Cd in the leaves of a superior Cd-accumulating ecotype of Thlaspi caerulescens using 113Cd-NMR,  Planta ,  2005, vol.  221 (pg.  928- 936) Google Scholar CrossRef Search ADS PubMed  Van der Zaal BJ,  Neuteboom LW,  Pinas JE,  Chardonnens AN,  Schat H,  Verkleij JAC,  Hooykaas PJJ.  Overexpression of a novel Arabidopsis gene related to putative zinc-transporter genes from animals can lead to enhanced zinc resistance and accumulation,  Plant Physiology ,  1999, vol.  119 (pg.  1047- 1055) Google Scholar CrossRef Search ADS PubMed  Van Rossum F,  Bonnin I,  Fenart S,  Pauwels M,  Petit D,  Saumitou-Laprade P.  Spatial genetic structure within a metallicolous population of Arabidopsis halleri, a clonal, self-incompatible and heavy-metal-tolerant species,  Molecular Ecology ,  2004, vol.  13 (pg.  2959- 2967) Google Scholar CrossRef Search ADS PubMed  Vido K,  Spector D,  Lagniel G,  Lopez S,  Toledano MB,  Labarre J.  A proteome analysis of the cadmium response in Saccharomyces cerevisiae,  Journal of Biological Chemistry ,  2001, vol.  276 (pg.  8469- 8474) Google Scholar CrossRef Search ADS PubMed  Vos P,  Hogers R,  Bleeker M, et al.  AFLP: a new technique for DNA fingerprinting,  Nucleic Acids Research ,  1995, vol.  23 (pg.  4407- 4414) Google Scholar CrossRef Search ADS PubMed  Vuylsteke M,  Van Den Daele H,  Vercauteren A,  Zabeau M,  Kuiper M.  Genetic dissection of transcriptional regulation by cDNA-AFLP,  The Plant Journal ,  2006, vol.  45 (pg.  439- 446) Google Scholar CrossRef Search ADS PubMed  Weber M,  Harada E,  Vess C,  Roepenack-Lahaye EV,  Clemens S.  Comparative microarray analysis of Arabidopsis thaliana and Arabidopsis halleri roots identifies nicotianamine synthase, a ZIP transporter and other genes as potential metal hyperaccumulation factors,  The Plant Journal ,  2004, vol.  37 (pg.  269- 281) Google Scholar CrossRef Search ADS PubMed  Weber M,  Trampczynska A,  Clemens S.  Comparative transcriptome analysis of toxic metal responses in Arabidopsis thaliana and the Cd2+-hypertolerant facultative metallophyte Arabidopsis halleri,  Plant, Cell and Environment ,  2006, vol.  29 (pg.  950- 963) Google Scholar CrossRef Search ADS   Wu A,  Wemmie JA,  Edgington NP,  Goebl M,  Guevara JL,  Moye-Rowley WS.  Yeast bZip proteins mediate pleiotropic drug and metal resistance,  Journal of Biological Chemistry ,  1993, vol.  268 (pg.  18850- 18858) Google Scholar PubMed  Yang L,  Zheng B,  Mao C,  Yi K,  Liu F,  Wu Y,  Tao Q,  Wu P.  cDNA-AFLP analysis of inducible gene expression in rice seminal root tips under a water deficit,  Gene ,  2003, vol.  314 (pg.  141- 148) Google Scholar CrossRef Search ADS PubMed  © The Author [2006]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
TI - Comparative cDNA-AFLP analysis of Cd-tolerant and -sensitive genotypes derived from crosses between the Cd hyperaccumulator Arabidopsis halleri and Arabidopsis lyrata ssp. petraea
JF - Journal of Experimental Botany
DO - 10.1093/jxb/erl062
DA - 2006-08-17
UR - https://www.deepdyve.com/lp/oxford-university-press/comparative-cdna-aflp-analysis-of-cd-tolerant-and-sensitive-genotypes-q48Dta0Dnq
SP - 2967
EP - 2983
VL - 57
IS - 12
DP - DeepDyve
ER -