Functional Characterization of NRAMP3 and NRAMP4 from the Metal Hyperaccumulator Thlaspi Caerulescens

Research

FunctionalBlackwellOxford,NPHNew0028-646X1469-8137©269410.1111/j.1469-8137.2008.02694.xNovember0637???650???OriginalXX The Phytologist Authors UKArticle Publishing 2008 (2008). Ltd Journal compilation © New Phytologist (2008) characterization of XX NRAMP3 and NRAMP4 from the metal hyperaccumulator Thlaspi caerulescens

Ronald J. F. J. Oomen1,7*, Jian Wu2,3*, Françoise Lelièvre1, Sandrine Blanchet1, Pierre Richaud4,5,6, Hélène Barbier-Brygoo1, Mark G. M. Aarts2 and Sébastien Thomine1 1Institut des Sciences du Végétal, CNRS, Avenue de la Terrasse, Gif-sur-Yvette, France; 2Laboratory of Genetics, Wageningen University, Arboretumlaan 4, NL–6703 BD Wageningen, The Netherlands; 3Institute of Vegetables and Flowers, Chinese Academy of Agricultural Sciences, Zhongguancun South Street 12, Beijing 100081, China; 4CEA, DSV, IBEB, Lab Bioenerget Biotechnol Bacteries & Microalgues, Saint-Paul-lez-Durance, F–13108, France; 5CNRS, UMR Biol Veget & Microbiol Environ, Saint-Paul-lez-Durance, F–13108, France; 6Aix-Marseille Université, Saint-Paul-lez-Durance, F-13108, France; 7(current address) Biochimie & Physiologie Moléculaire des Plantes, UMR5004, CNRS/INRA/SupAgro/UM II, F-34060 Montpellier Cedex 1, France

Summary

Author for correspondence: • The ability of metal hyperaccumulating plants to tolerate and accumulate heavy Ronald J. F. J. Oomen metals results from adaptations of metal homeostasis. NRAMP metal transporters Tel: +33499613152 were found to be highly expressed in some hyperaccumulating plant species. Fax: +33467525737 TcNRAMP3 TcNRAMP4 Email: [email protected] • Here, we identified and , the closest homologues to AtNRAMP3 and AtNRAMP4 in Thlaspi caerulescens and characterized them by 29 August 2008 Received: expression analysis, confocal imaging and heterologous expression in yeast and Accepted: 21 October 2008 Arabidopsis thaliana. • TcNRAMP3 and TcNRAMP4 are expressed at higher levels than their A. thaliana New Phytologist (2009) 181: 637–650 homologues. When expressed in yeast TcNRAMP3 and TcNRAMP4 transport the doi: 10.1111/j.1469-8137.2008.02694.x same metals as their respective A. thaliana orthologues: iron (Fe), manganese (Mn) and cadmium (Cd) but not zinc (Zn) for NRAMP3; Fe, Mn, Cd and Zn for NRAMP4. Key words: cadmium (Cd), They also localize at the vacuolar membrane in A. thaliana protoplasts. Inactivation of hyperaccumulator, metal transport, AtNRAMP3 and AtNRAMP4 in A. thaliana results in strong Cd and Zn hypersensitivity, NRAMP, Thlaspi caerulescens, zinc (Zn). which is fully rescued by TcNRAMP3 or TcNRAMP4 expression. However, metal tolerance conferred by TcNRAMP expression in nramp3nramp4 mutant does not exceed that of wild-type A. thaliana. • Our data indicate that the difference between TcNRAMP3 and TcNRAMP4 and their A. thaliana orthologues does not lie in a different protein function, but probably resides in a different expression level or expression pattern.

Küpper et al., 2000; Lombi et al., 2000; Yang et al., 2003; Introduction Vogel-Mikus et al., 2006). Hyperaccumulation implies an Among the plant species adapted to environments containing adapted metal homeostasis allowing enhanced accumulation of high levels of transition metals, some species called hyperac- heavy metals and tolerance to both exposure and accumulation. cumulators have developed the exceptional ability to accumulate Therefore, functional and/or transcriptional modifications of metals to levels 10–100 times higher than other species. Among metal transporters, metal chelators and other proteins are these, only c. 15 species display zinc (Zn) hyperaccumulation needed to maintain a balance between plant metal uptake, (Baker et al., 1992; Brooks, 1994; Meerts & van Isacker, distribution and storage (reviewed by Clemens, 2001; Mäser 1997; Bert et al., 2000; Schat et al., 2000). Cadmium (Cd) et al., 2001; Cobbett & Goldsbrough, 2002). Elevated transcript hyperaccumulation is even less frequent and only identified in levels for some genes encoding metal transporters and some Thlaspi caerulescens and Arabidopsis halleri accessions, in chelators in T. caer ulescens and Thlaspi goesingense provided Thlaspi praecox and in Sedum alfredii (Brown et al., 1995; the first hints about their involvement in metal tolerance and accumulation (Pence et al., 2000; Assunção et al., 2001; *These authors contributed equally to this work. Persans et al., 2001). Further global transcriptomic studies,

using A. thaliana microarrays to compare hyperaccumulating Materials and Methods species with A. thaliana, identified several candidate genes encoding metal transporters and chelators putatively involved Library screening in Zn and Cd tolerance and hyperaccumulation (Becher et al., 2004; Weber et al., 2004; Talke et al., 2006; van de Mortel cDNA libraries made from roots of T. caer ulescens J. & C. et al., 2006, 2008). Presl (accession La-Calamine), as described in Rigola et al. In particular, transcriptomic studies identified NRAMP as (2006), were used for full-length cDNA cloning. The partial one of the gene families highly expressed in hyperaccumulating cDNA clone RR23nr019 has been identified as an AtNRAMP3 plants. Weber et al. (2004) showed that NRAMP3 displays homologue (Rigola et al., 2006). A 1-kb cDNA fragment cut constitutively high expression in A. halleri compared with from the pAD-GAL4-2.1 vector plasmid using EcoRI and A. thaliana and micro-array data indicated that, in T. caerulescens, XhoI was used as probe in the cDNA library screening. The NRAMP3 expression is generally higher than in A. thaliana probe was labelled with [α-32P]dATP using the Hexalabel (van de Mortel et al., 2006). NRAMP genes have been DNA labelling kit (Fermentas; http://www.fermentas.de). identified in many organisms, from bacteria to man. They Eight positive clones were obtained. The three longest cDNA play important roles in metal ion homeostasis, especially clones were sequenced by ABI PRISM BigDye terminator iron uptake and recycling in mammals and manganese cycle sequencing technology v2.0, according to the manufacturer’s (Mn) uptake in yeast and bacteria. Most NRAMP proteins, instruction (Applied Biosystems; http://www3.appliedbiosystems. however, are able to transport multiple metal ions such as com), using an ABI3700 DNA analyser. Sequence analysis Mn, Zn, copper (Cu), iron (Fe), Cd, nickel (Ni) and cobalt was performed using standard blast method (http://www. (Co) (reviewed in Nevo & Nelson, 2006). NRAMP genes ncbi.nlm.nih.gov/BLAST/) and revealed that one of the three have been identified in many other plant species including positive clones, containing a 1539-bp open reading frame rice, tomato, soybean and Thlaspi japonicum (Belouchi et al., (ORF), was homologous to AtNRAMP3 and one, with a 1997; Bereczky et al., 2003; Kaiser et al., 2003; Mizuno 1494-bp ORF, was homologous to AtNRAMP4 and that both et al., 2005). The first studies on NRAMP functions in plants ORFs were full length. indicated a role in Fe homeostasis (Curie et al., 2000; Thomine et al., 2000, 2003; Lanquar et al., 2005). In addition, Southern blot analysis expression of AtNRAMP1, AtNRAMP3 and AtNRAMP4 in yeast showed that these proteins are able to transport Cd Genomic DNA was extracted from four T. caerulescens accessions and that AtNRAMP4 is able to transport Zn (Curie et al., (La Calamine, Monte Prinzera, Ganges and Austria) and four 2000; Thomine et al., 2000; Lanquar et al., 2004). A recent Thlaspi species (T. japonicum, T. praecox, T. minimum and study of a NRAMP3 homologue in T. caerulescens showed T. perfoliatum) using the cetyltrimethylammonium bromide that TcNRAMP3 is also able to transport Fe and Cd (Wei (CTAB) method. One microgram of DNA, digested with et al., 2008). XbaI (TcNRAMP3) or EcoRI (TcNRAMP4) was separated by The role of NRAMP genes in metal homeostasis and their gel electrophoresis using a 1% agarose-TAE (Tris-Acetate strong expression in metal hyperaccumulators led us to 40 mm, ethylenediaminetetraacetic acid (EDTA) 1 mm) gel investigate the differences between the NRAMP genes in and vacuum blotted onto Hybond N+ nylon membrane (GE hyperaccumulating and nonaccumulating plants. At the Healthcare Bio-Sciences Corp., Piscataway, NJ, USA) according moment it is unclear whether this is restricted to a difference to standard procedures. cDNA fragments from digested at the transcriptional level (plant, tissue or cell-type specific) full-length cDNA clones using XhoI for TcNRAMP3 (462 bp) or whether NRAMP proteins are functionally divergent and SalI for TcNRAMP4 (248 bp) were used as probes and between hyperaccumulating and nonhyperaccumulating plants. labelled as described earlier. Prehybridization and hybridization In this study, we investigated the function of NRAMP3 and were performed in hybridization solution (10% dextran sulphate, NRAMP4 proteins from the hyperaccumulator T. caerulescens. 1 m NaCl, 1% sodium dodecyl sulphate (SDS)) supplemented TcNRAMP3 and TcNRAMP4 cDNAs (Rigola et al., 2006) with denatured calf thymus DNA (100 µg ml−1). After an highly similar to AtNRAMP3 and AtNRAMP4, respectively, overnight incubation at 65°C the membranes were rinsed were isolated from T. caerulescens La Calamine. We have twice in 2 × standard saline citrate (SSC: 300 mm NaCl, confirmed that these two NRAMP genes are differently 30 mm Na citrate, pH 7.0) for 2 min at room temperature, expressed between A. thaliana and T. caerulescens and com- twice in 2 × SSC, 1% SDS for 20 min at 65°C, and twice in pared the metal transport abilities of TcNRAMPs with their 0.1 × SSC, 1% SDS for 20 min at 65°C. A. thaliana homologues by heterologous expression in yeast. We studied TcNRAMP3 and TcNRAMP4 intracellular Plant material for expression analysis localization in plant cells and tested whether their expression is able to rescue the Cd- and Zn-hypersensitive phenotype of Arabidopsis thaliana Columbia-0 and T. caerulescens accession the A. thaliana nramp3 nramp4 double mutant. La Calamine seeds were germinated on garden peat soil

New Phytologist (2009) 181: 637–650 © The Authors (2008) www.newphytologist.org Journal compilation © New Phytologist (2008) Research 639

(Jongkind BV, Aalsmeer, the Netherlands). Three-week-old at 68°C for 10 min. The PCR products were recombined seedlings were transferred to hydroponics, three plants per pot into pDONOR207 (Invitrogen) in a 10-µl BP Clonase filled with 1 l modified half-strength Hoagland nutrient solution (Invitrogen) reaction following the manufacturer’s instruction. (van de Mortel et al., 2006). After 3 wk, the T. caerulescens A similar procedure was used for AtNRAMP3 and AtNRAMP4 plants were transferred to the same nutrient solution with a from the plasmids pGEM-T-easy AtNRAMP3 and pGEM-T deficient (0 µm), sufficient (100 µm), or excess (1000 µm) easy AtNRAMP4 (Thomine et al., 2000) using the primers ′ ZnSO4 concentration for 7 d. The A. thaliana plants were AtNRAMP3 (forward 5 -GGAGATAGAACCATGCCA- transferred to the same nutrient solution with deficient CAACTCGAGAACAACG-3′ and reverse 5′-CAAGAA- ′ (0 µm), sufficient (2 µm) or excessive (25 µm) ZnSO4. The AGCTGGGTCTCAATGACTAGACTCCGCTTTG-3 ) nutrient solution was replaced once a week during the first and AtNRAMP4 (forward 5′-GGAGATAGAACCATGT- 3 wk and thereafter twice a week. CGGAGACTGATAGAGAGC and reverse 5′-CAAGA- AAGCTGGGTCTCACTCATCATCCCTCTGTGG-3′) followed by nested PCR (on the underlined region of the Quantitative reverse transcriptase polymerase chain primers) with universal Gateway extension primers attB1 reaction (qRT-PCR) analysis 5′-GGGACAAGTTTGTACAAAAAAGCAGGCTTCGA- Leaves or roots of one pot containing three A. thaliana or AGGAGATAGAACCATG-3′ and attB2 5′-GGGGAC- three T. caerulescens plants per treatment were pooled and CACTTTGTACAAGAAAGCTGGGTC-3′. For plant homogenized in liquid nitrogen. Total RNA of leaves or transformation, the TcNRAMP fragments were transferred from roots was extracted with Trizol (Invitrogen, Groningen, the their donor constructs to the binary overexpression vector Netherlands) following the manufacturer’s instructions. For pGD625 (de Folter et al., 2006), containing the double 35S qRT-PCR, first-strand cDNA was synthesized from 5 µg of CaMV enhancer, in a 10-µl LR Clonase (Invitrogen) reaction total RNA using the SUPERSCRIPT III first-strand synthesis following the manufacturer’s instruction. The binary constructs system (Invitrogen). One-twentieth of the cDNAs was used as were introduced into the A. tumefaciens strain AGL0 by electro- a template in 10-µl PCR reactions. The PCR was performed poration. For the generation of yeast expression vectors, a similar with a Light Cycler apparatus and the ‘LC FastStart DNA LR reaction was performed using the AtNRAMP and TcNRAMP Master SYBR Green IR’ (Roche Diagnostics GmbH, Mannheim, fragments to generate pDR195gtw constructs (pDR195 adapted Germany) in a standard PCR reaction according to the to Gateway cloning by Andéol Falcon de Longevialle, Unité manufacturer’s instructions. Gene-specific PCR primers were de Recherche en Génomique Végétale, Evry, France). designed according to the cDNA sequences (maximum size, 300 bp; melting temperature, 60°C). The following gene- Expression in yeast specific primers were used (forward and reverse, respectively) for both T. caerulescens and A. thaliana, using conserved The yeast strains used in this study were DEY1453 (fet3fet4, domains: 5′-ATGGTTTTGTGGGTTATGGC-3′ and Eide et al., 1996), ZHY3 (zrt1zrt2, Zhao & Eide, 1996a,b) 5′-CTCGAGCTTCCTTATTCCGT-3′ for Nramp3; 5′- and smf1 (Supek et al., 1996; Thomine et al., 2000). Yeast CCAGGACTATCAAACAAGCTGT-3′ and 5′-CAATG- cells were grown (at 30°C) on yeast extract–peptone–dextrose ′ GAGTAGTACTTGAGAGCTTC-3 for Nramp4. Parallel (supplemented with 0.2 mm FeCl3, 0.1 mm ZnSO4 or 0.1 mm ′ reactions to amplify actin (5 -GGTAACATTGTGCTC- MnCl2 for the three strains respectively) before transformation AGTGGTGG-3′ and 5′-AACGACCTTAATCTTCATGC- and on synthetic dextrose-URA (with the same metal additions) TGC-3′) were performed as reference. Primer combinations after transformation. Yeast cells were transformed according had at least 85% efficiency. to standard procedures (Invitrogen). The fet3fet4 and smf1 complementation was tested by drop spotting assays, spotting diluted cultures of individual transformants on synthetic dextrose- Construction of expression vectors URA agar plates with 80 µm BPS and 30 µm FeCl3 or 200 µm TcNRAMP3 and TcNRAMP4 were amplified from cDNA FeCl3 for fet3fet4 and plates with 10 mm ethylene glycol- clones by PCR using the following Gateway recombination bis(beta-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA) ′ site extended primers: 5 -GGGGACAAGTTTGTACAA- with or without 0.1 mm MnCl2 for smf1. zrt1zrt2 complemen- AAAAGCAGGCTGATCGAATTAGGATCCTCTGC-3′, tation was tested by measuring the optical density at 600 nm containing the attB1 sequence (underlined) and BamHI site (OD600) in 5 ml cultures of Low Zinc Medium (Zhao & Eide, ′ (italic), and 5 -GGGGACCACTTTGTACAAGAAAGC- 1996b) with 10 µm ZnSO4 inoculated with 50 µl synthetic ′ TGGGTCTAATGGGCTCGAGAGTCGAC-3 , containing dextrose-URA pre-culture at OD600 = 1. Cadmium sensitivity the attB2 sequence (underlined) and an XhoI site (italic). was tested by measuring the OD600 of fet3fet4 transformants PCRs were performed with the Pfu polymerase (Fermentas) at grown in 5 ml cultures of synthetic dextrose-URA (pH 5.5) ° ° ° 93 C for 5 min, followed by 30 cycles 93 C for 1 min, 55 C with or without 10 µm CdCl2 after inoculation with 150 µl ° for 1 min and 68 C for 3 min, and completed by an extension synthetic dextrose-URA pre-culture at OD600 =1.

(N, N′-Di(2-hydroxybenzyl)″ ethylenediamine-N, N′-diacetic Green fluorescent protein (GFP) fusion and intracellular acid monohydrchloride)). localization The expression levels of transgenic lines were checked by To construct a cDNA encoding a translational fusion between RT-PCR using the total RNA extracted from Ws-nramp3nramp4

TcNRAMP3 or TcNRAMP4 and the GFP, RSsm-GFP T4 plants (grown in vitro on ABIS medium (2.5 mm H3PO4, (Davis & Vierstra, 1998), the TcNRAMP fragments were 5 mm KNO3, 2 mm MgSO4, 1 mm Ca(NO3)2, Murashige & transferred from their donor constructs, by a 10-µl LR Skoog microelements, 1% sucrose, 0.7% agar, 1 mm 2- Clonase (Invitrogen) reaction following the manufacturer’s (Nmorpholino) ethanesulfonic acid (MES) adjusted with KOH instruction, into the plant transient expression vector CD3-327 to pH 6.1) with 50 µM FeHBED), with Trizol (Invitrogen) (Davis & Vierstra, 1998) adapted to Gateway cloning by following the manufacturer’s instructions. A 5-µg sample of L. Gissot (INRA, Versailles, France). This generated total RNA was used to synthesize cDNA with MLV reverse TcNRAMP3::GFP or TcNRAMP4::GFP fusion proteins transcriptase (Invitrogen) and oligo(dT) as a primer (Invitrogen). with the GFP as an additional carboxy-terminal domain, under PCR amplification (20 cycles) was performed with a cDNA the control of the 35S CaMV promoter. The TcNRAMP::GFP aliquot (2 µl) using the following primers; TcNRAMP3F fusion constructs or the empty vector CD3-327 were transiently 5′-TCTCTGGGCTGGTGTCAT-3′ and TcNRAMP3R expressed in A. thaliana protoplasts through polyethylene 5′-AGCCGCCTCCATACTTG-3′; TcNRAMP4F 5′-TCT- glycol (PEG) mediated transformation. Free DsRed2 was GGGCTGGAGTTGTAATCACC-3′ and TcNRAMP4R co-transformed in the same cells. The walls of A. thaliana 5′-TCCGCTATGTCCGTCCCGTAAAAG-3′. Primer pairs suspension cells were digested in Gamborg’s B5 medium for actin were used as a control for similar cDNA quantities supplemented with 0.17 m glucose, 0.17 m mannitol, 1% between the samples: ActinF 5′-GGTAACATTGTGCTCA- cellulase and 0.2% macerozyme. Protoplasts were purified by GTGGTGG-3′ and ActinR 5′-AACGACCTTAATCTTC- floatation in Gamborg’s B5 medium supplemented with ATGCTGC-3′. Twenty microlitres from the 50-µl reaction was 0.28 m sucrose. For transformation, 0.2 × 106 cells were mixed separated on an ethidium bromide-stained 1% agarose gel. with 5 µg of plasmid DNA in a solution containing 25% PEG The tolerance and/or sensitivity of transgenic lines was

6000, 0.45 m mannitol, 0.1 m Ca(NO3)2 pH 9 and incubated compared with wild type and nramp3nramp4 mutant plants. in the dark for 20 min. The PEG was then washed twice with Arabidopsis thaliana (T3 or T4) seedlings were grown as

0.275 m Ca(NO3)2 and the protoplasts were resuspended in previously described by Lanquar et al. (2005) using the same Gamborg’s B5 medium supplemented with 0.17 m glucose media with concentrations of FeHBED, ZnSO4 and CdCl2 as and 0.17 m mannitol in which they were maintained until indicated in the figure legends. microscopic observation. After 24–48 h, fluorescent cells were imaged by confocal microscopy (Leica TCS SP2) with Metal content excitation at 488 nm and the fluorescence emission signal recovery between 495 nm and 535 nm for GFP fusion proteins Seedlings were grown vertically for 12 d on plates containing and excitation at 543 nm and the fluorescence emission ABIS medium (1% agar), supplemented or not with 30 µm

signal recovery between 578 nm and 707 nm for DsRed2 CdCl2 or 300 µm ZnCl2. Roots and shoots were harvested (Thomine et al., 2003). separately and their metal content was measured as described in Lanquar et al. (2005). The dry weight (DW) of the samples was measured after drying at 60°C for 3 d. The dried samples Plant transformation and metal tolerance/sensitivity were digested in a mixture of 1 ml of 65% nitric acid and 0.5 ml growth assays ultrapure H2O in a MARS5 microwave (CEM GmbH, The constructs with full-length cDNAs of TcNRAMP3 Kamp-Lintford, Germany) at 200°C at 15 bar for 10 min. and TcNRAMP4 were used to transform A. thaliana After dilution in trace-metal-free water, the metal content of the nramp3nramp4 double mutants (in accession Wassilewskija samples was determined by inductively coupled plasma optical (Ws) background) by the flower dip method (Clough, 2005). emission spectroscopy (ICP-OES) using an IRIS Advantage Independent homozygous T2 A. thaliana Ws-nramp3nramp4 Duo ER/S (Thermo Jarrell Ash, Franklin, MA, USA). transformants with a single insertion locus were obtained from heterozygous T1 plants selected on the basis of Results complementation of the nramp3nramp4 phenotype on low Fe medium. Seeds were sown on plates containing ABIS medium Cloning of T. caerulescens NRAMP3 and with adjusted Fe concentration and 50 µg ml−1 kanamycin NRAMP4 cDNAs (2.5 mm H3PO4, 5 mm KNO3, 2 mm MgSO4, 1 mm Ca(NO3)2, Murashige & Skoog microelements, 1% sucrose, The expressed sequence tag (EST) collection of T. caerulescens 0.7% agar, 1 mm 2-(N-morpholino) ethanesulfonic acid accession La Calamine (Rigola et al., 2006) contained one (MES) adjusted with KOH to pH 6.1 and 0.3 µm FeHBED AtNRAMP3 homologous partial cDNA (RR23nr019).

New Phytologist (2009) 181: 637–650 © The Authors (2008) www.newphytologist.org Journal compilation © New Phytologist (2008) Research 641

Fig. 1 Expression analysis of NRAMP3 and NRAMP4 in Arabidopsis thaliana and Thlaspi caerulescens in response to different zinc (Zn) supplies. Quantitative reverse transcriptase polymerase chain reaction (RT-PCR) analysis of NRAMP3 (a), NRAMP4 (b) expression relative to actin in leaves and roots of T. caerulescens accession La Calamine (closed circles) and A. thaliana (open circles). Plants were grown in hydroponic culture for 3 wk and subsequently transferred for 7 d to nutrient solutions with deficient (0 µM; 0), sufficient (100 (Tc) or 2 (At) µM; +), or excess (1000

(Tc) or 25 (At) µM; +++) ZnSO4 concentrations before separation of roots and leaves followed by RNA extraction.

Screening of a T. caerulescens cDNA library using this partial and 25 µm (excess) ZnSO4 and 100 µm (sufficient), and cDNA as probe (Assunção et al., 2001) allowed the cloning of 1000 µm (excess) ZnSO4 were used for A. thaliana and two NRAMP homologues. These TcNRAMPs show a 96.5% T. caerulescens, respectively (van de Mortel et al., 2006). Based and 95.1% similarity to AtNRAMP3 and AtNRAMP4 at on the available Arabidopsis cDNA sequences and the newly the amino acid level. As they are the closest T. caerulescens identified NRAMP3 and NRAMP4 T. caerulescens cDNA homologues of AtNRAMP3 and AtNRAMP4, we annotated sequences, we designed common sets of primers that allow them as TcNRAMP3 (EF639294) and TcNRAMP4 efficient amplification of NRAMP3, NRAMP4 and actin in (EF639295), respectively. both T. caerulescens and A. thaliana. Independent of Zn status, TcNRAMP3 and TcNRAMP4 show elevated expression levels compared with AtNRAMP3 and AtNRAMP4 respectively. NRAMP3 and NRAMP4 display elevated expression For example, under Zn-sufficient conditions, NRAMP3 was levels in T. caerulescens expressed 10 times more in roots and nine times more in To confirm the differential expression of NRAMP genes shoots of T. caerulescens compared with A. thaliana; NRAMP4 between T. caerulescens and A. thaliana observed in micro was expressed three times more in roots and nine times more array experiments (van de Mortel et al., 2006), we compared in shoots of T. caerulescens compared with A. thaliana (Fig. 1; see their expression levels in roots and shoots of A. thaliana and the Supporting Information, Table S1). The NRAMP–actin T. caerulescens La Calamine exposed to deficient, sufficient expression ratios, as shown in Fig. 1, also suggest a regulation and excess Zn in hydroponic solutions (Fig. 1). Note that of NRAMP3 and NRAMP4 expression upon Zn deficiency physiologically equivalent sufficient and excess Zn concentra- and excess. However, actin expression level is modified upon tions, resulting in Zn sufficiency and Zn excess symptoms, variations in Zn concentration (Table S1). Using tubulin as respectively, are different for the two species: 2 µm (sufficient) an alternative control gene, TcNRAMP3 and TcNRAMP4

showed a complex hybridization pattern of multiple bands and a higher band intensity, despite the equal loading of genomic DNA, for NRAMP3 as well as NRAMP4 suggesting the presence of multiple NRAMP copies.

Thlaspi caerulescens and A. thaliana NRAMP3 and NRAMP4 show similar metal transport specificities To identify putative differences in transport properties between NRAMP3 and NRAMP4 from A. thaliana and T. caerulescens, we expressed them in yeast metal uptake deficient mutants allowing determination of their metal transport specificity (Pinner et al., 1997; Curie et al., 2000; Thomine et al., 2000; Lanquar et al., 2004). The ability of TcNRAMP3 and TcNRAMP4 to transport Fe and Mn was investigated by testing for the complementation of fet3fet4 and smf1 yeast mutant phenotypes. The fet3fet4 strain is defective in its low- and high-affinity Fe uptake systems and is unable to grow on a low Fe medium (Dix et al., 1994). The smf1 strain is disrupted in SMF1, a yeast NRAMP gene essential for high affinity Mn uptake. smf1 is unable to grow on synthetic media containing high concentrations of the divalent cation chelator Fig. 2 Genomic DNA blot analysis of NRAMP3 and NRAMP4 in Thlaspi caerulescens accessions and related Thlaspi species. Genomic EGTA (Supek et al., 1996) and this growth defect is specifically DNA was digested with XbaI (a) and EcoRI (b), and blots were rescued by Mn supplementation (Fig. 3b). We tested the hybridized with a TcNRAMP3 (a) or TcNRAMP4 (b) cDNA probe. The ability of fet3fet4 strains expressing AtNRAMP3, AtNRAMP4, numbers above the lanes designate the different T. caerulescens TcNRAMP3, TcNRAMP4 and AtIRT1, a metal transporter accessions: 1, La Calamine; 2, Monte Prinzera; 3, Ganges; 7, Austria; with a broad substrate range (Korshunova et al., 1999), to or related Thlaspi species: 4, T. japonicum; 5, T. praecox; 6, T. minimum; 8, T. perfoliatum. grow on low Fe medium (Fig. 3a). All four NRAMPs as well as the IRT1 control complement the fet3fet4 phenotype without any marked difference. Expression of TcNRAMPs, show different apparent regulation by Zn status, owing to AtNRAMPs and IRT1 also rescued growth of the smf1 strain different regulation of tubulin by Zn (Table S1, Fig. S1). on synthetic medium with 10 mm EGTA, with no major Hence, because of the lack of an appropriate control gene to difference between the different transporters (Fig. 3b). These data study Zn-dependent gene regulation by quantitative PCR in indicate that, like their A. thaliana homologues, TcNRAMP Thlaspi caerulescens, our experiments do not allow a conclusion proteins are able to transport Fe and Mn. on the regulation of TcNRAMP3 and TcNRAMP4 expression Since T. caerulescens and A. thaliana differ greatly in their by Zn. By contrast, whatever the control gene used, we found tolerance and accumulation of Zn and Cd, we used a more clearly increased expression of NRAMP3 and NRAMP4 genes quantitative experimental set-up to assess the Zn and Cd in Thlaspi caerulescens compared with Arabidopsis thaliana, transport capabilities of the NRAMP3 and NRAMP4 proteins regardless of the plants Zn status. from these two species. The zrt1zrt2 double mutant yeast To determine whether increased expression is associated strain is deficient in high- (ZRT1) and low- affinity (ZRT2) with the presence of multiple copies of TcNRAMP3 and Zn transport systems (Zhao & Eide, 1996a,b), and cannot TcNRAMP4 in the Thlaspi genome, we performed two grow on synthetic media without high Zn supplementation. genomic DNA blot analyses in four T. caerulescens accessions The zrt1zrt2 strain was transformed with the four NRAMPs and three other Thlaspi species using genomic DNA digested or an empty vector (control), and grown in low Zn medium

with EcoRI and XbaI (Fig. 2, Fig. S2). In most T. caer ulescens (Zhao & Eide, 1996b) containing 10 µm ZnSO4. Only accessions, we identified single bands hybridizing with the AtNRAMP4 and TcNRAMP4 were able to significantly rescue TcNRAMP probes for both enzyme digestions suggesting that zrt1zrt2 growth (Fig. 3c). By contrast, zrt1zrt2 strains expressing NRAMP3 (Fig. 2a) and NRAMP4 (Fig. 2b) are present as AtNRAMP3 and TcNRAMP3 do not grow significantly better single-copy genes in T. caer ulescens (as in A. thaliana), although than zrt1zrt2 empty vector control strain. When these yeast we cannot rule out the possibility of several very closely linked strains were grown in the same medium supplemented with

copies. For the other Thlaspi species, T. minimum and T. praecox 100 µm ZnSO4 they showed equal growth (data not shown). seem to contain both genes as single copies. In T. perfoliatum, The Cd transport activity of NRAMP proteins was tested by there appear to be two copies of both genes. Thlaspi japonicum using their ability to enhance Cd sensitivity in yeast, revealed

New Phytologist (2009) 181: 637–650 © The Authors (2008) www.newphytologist.org Journal compilation © New Phytologist (2008) Research 643

by a reduced growth on Cd-containing medium (Thomine et al., 2000). The four NRAMPs and IRT1 (Rogers et al., 2000) were introduced into the fet3fet4 yeast strain and twin cultures,

with or without 10 µm CdCl2, were inoculated from pre-cultures of individual transformants. Compared with the empty-vector control, expression of all four NRAMPs leads to a significant increase in Cd sensitivity (Fig. 3d). After 18 h, NRAMP expressing cultures showed growth inhibitions to 32–48% in Cd- containing medium compared with 62% for the empty-vector control and 21% for IRT1. This result indicates that, as for Fe and Mn, all four NRAMP proteins are able to transport Cd.

TcNRAMP3 and TcNRAMP4 proteins are targeted to the vacuolar membrane To test whether the TcNRAMP3 and TcNRAMP4 intracellular localization is similar to that of AtNRAMP3 and AtNRAMP4, we expressed TcNRAMP transporters fused to GFP in protoplasts prepared from an Arabidopsis cell suspension. Free DsRed2 was co-transformed in the same cells. Green fluorescent protein and DsRed2 fluorescence was imaged by confocal microscopy. Figure 4 shows that while DsRed2 fluorescence was homogeneous in the cytosol and nucleus, GFP fluorescence consistently lined the vacuolar membrane when GFP was fused to the carboxy-terminal end of either TcNRAMP3 or TcNRAMP4. These results indicate that, like their A. thaliana counterparts, NRAMP3 and NRAMP4 from T. caerulescens reside at the vacuolar membrane.

TcNRAMP expression complements the A. thaliana nramp3nramp4 mutant Zn and Cd hypersensitivity As their heterologous expression in yeast and their cellular localization indicated similar functions for T. caerulescens

or no added Mn (−Mn). Yeast pre-cultures were diluted –2 –3 to OD600 =10 or 10 . (a) and (b) 10 µl of each sample was spotted and incubated at 30°C for 4 d. (c) Growth of transformed zrt1zrt2 yeast cells in low Zn medium (pH 6)

supplemented with 10 µM ZnSO4. The OD600 of 5 ml cultures inoculated at OD600 = 0.01 was measured after 13, 16, 19 and 22 h. Mann–Whitney tests indicate that AtNRAMP4 and TcNRAMP4 significantly rescue zrt1zrt2 growth (P < 0.001) while AtNRAMP3 Fig. 3 Thlaspi caerulescens NRAMP3 and NRAMP4 transport and TcNRAMP3 do not. (d) The ability of the NRAMP proteins abilities in yeast. AtNRAMP3, AtNRAMP4, TcNRAMP3, TcNRAMP4 to transport cadmium (Cd) was tested by their effect to enhance and IRT1 (control) were expressed in the fet3fet4, smf1, zrt1zrt2 Cd sensitivity of transformed fet3fet4 yeast cells. The 5-ml cultures yeast mutants (deficient in iron (Fe), manganese (Mn) and zinc (Zn) of synthetic dextrose-URA (pH 5.5) with or without 10 µM CdCl2 were ° uptake, respectively) and subsequently studied for their ability to inoculated with 150 µl preculture (OD600 = 1) and grown at 30 C. rescue the mutant phenotype. An empty vector transformant (Ø) was The OD600 was measured after 12, 15 and 18 h and the growth in the used as control for vector effects. (a) Growth of transformed fet3fet4 Cd-containing culture was expressed as a percentage of the growth yeast cells on synthetic dextrose-URA (pH 5.5) supplemented with in its control culture without Cd. Mann–Whitney tests indicate that − 80 µM BPS and 200 µM FeCl3 (+Fe) or 30 µM FeCl3 ( Fe). Yeast AtNRAMP3 and 4, and TcNRAMP3 and 4 significantly increase yeast –3 –4 pre-cultures were diluted to OD600 =10 or 10 . (b) Growth of Cd sensitivity compared with control (P < 0.001). (c,d) Representative transformed smf1 yeast cells on synthetic dextrose-URA (pH 6) results of one of four experiments that gave qualitatively similar supplemented with 10 mM ethylene glycol-bis(beta-aminoethyl results. Bars show the mean of duplicate cultures and the arrow ′ ′ ether)-N,N,N ,N -tetraacetic acid (EGTA) and 0.1 mM MnCl2 (+Mn) bars represent the SD.

Fig. 4 TcNRAMP3::GFP and TcNRAMP4::GFP fusion proteins are targeted to the vacuolar membrane. Arabidopsis thaliana protoplasts transformed with TcNRAMP3::GFP (a–d) or TcNRAMP4::GFP (e–h) plasmids and free DsRed2. Confocal cross sections show green fluorescent protein (GFP) fluorescence (a,e; excitation 488 nm, emission measured between 495 nm and 535 nm) DsRed2 fluorescence (b,f; excitation 543 nm, emission measured between 578 nm and 707 nm) and the overlay of NRAMP– and free DsRed2 fluorescence (c,g). In (d) and (h) transmission images of the corresponding protoplasts are shown. The GFP fluorescence delimitates the vacuole lumen (black) and the cytosol (red) indicating vacuolar membrane localization. Bars, 10 µm.

and A. thaliana NRAMP proteins, we wondered whether associated with a decreased ability to detoxify Zn and Cd or a expression of TcNRAMP3 and TcNRAMP4 in the A. thaliana general disturbance of metal homeostasis. nramp3nramp4 double mutant could rescue its phenotypes. To test whether TcNRAMP3 and TcNRAMP4 can perform This double mutant has been shown to display arrested the same functions as AtNRAMP3 and AtNRAMP4 in planta, seedling development under Fe-deficient conditions (Lanquar we expressed them individually in the nramp3nramp4 mutant et al., 2005), but response to excess Zn or Cd had not been under control of the constitutive CaMV 35S promoter. reported. We tested Zn and Cd sensitivity in media containing Homozygous single insertion lines were selected and their 50 µm FeHBED, which is sufficient to rescue the iron deficiency sensitivity to Fe starvation, Zn excess and Cd toxicity was ana- phenotype of the double mutant. Figure 5a shows that addi- lysed (Fig. 6). Semi-quantitative RT-PCR analysis showed a tion of Zn or Cd inhibited plant growth in a concentration- strong expression of TcNRAMP3 in three independently dependent manner in A. thaliana. Under the same conditions, transformed lines with little variation in expression level nramp3nramp4 plants displayed dramatically enhanced Zn (Fig. 6a). The three transformants as well as untransformed and Cd sensitivity (Fig. 5b,c). The addition of 300 µm Zn nramp3nramp4 and wild type were germinated on ABIS or 30 µm of Cd reduced wild-type root growth to 74% and medium supplemented with 0.3 µm FeHBED and grown for 67%, respectively, compared with 19 and 14% for the 11 d. The nramp3nramp4 mutant was unable to develop nramp3nramp4 mutant plants. Zinc or Cd hypersensitivity under these conditions while TcNRAMP3 expression fully was not observed in single nramp3 or nramp4 mutants rescued the nramp3nramp4 mutant growth defect on low (Thomine et al., 2000; V. Lanquar et al., unpublished). To Fe in all three transgenic lines (Fig. 6b). The ability of determine if altered Zn and Cd sensitivity is related to modi- TcNRAMP3 expression to rescue the nramp3nramp4 Zn and fied Zn or Cd uptake and or storage, we analysed the metal Cd sensitivity phenotype was also tested by growing the trans- content of wild type, single and double mutant plants grown formants on Zn- and Cd-containing medium (Fig. 6c,d). on control medium as well as on 300 µm Zn or 30 µm Cd This showed that TcNRAMP3 expression fully restored the (Fig. 5d). These measurements did not reveal any significant nramp3nramp4 sensitivity to Zn and Cd to wild-type levels. difference in Cd and Zn levels between wild type and nramp In addition, we found that expression of TcNRAMP4 in the mutant plants, so the enhanced sensitivity is likely to be nramp3nramp4 mutant background also fully complemented

New Phytologist (2009) 181: 637–650 © The Authors (2008) www.newphytologist.org Journal compilation © New Phytologist (2008) Research 645

Fig. 5 The Arabidopsis thaliana nramp3nramp4 double mutant shows enhanced sensitivity to zinc (Zn) and cadmium (Cd). A. thaliana wild type (a) and nramp3nramp4 double mutant (b) seedlings were tested for root growth sensitivity to Zn and Cd by growing them vertically for 11 d on ABIS agar medium containing 50 µM FeHBED and 30 µM ZnSO4 or on the same medium supplemented with 3, 10 or 30 µM CdCl2 or 100 or 300 µM ZnSO4. Pictures represent the result of one representative experiment out of four. (c) Average root length of seedlings, grown in the same conditions as in (a,b), was determined by measuring c. 70 plants per genotype. Arrow bars represent the SD. *, Significant difference between wild type and nramp3nramp4 double mutant according to a Mann–Whitney test with a 0.1% threshold. (d) Zinc and Cd contents in shoots and roots of wild type or nramp3, nramp4 and nramp3nramp4 mutant plants grown on control ABIS medium containing 30 µM ZnSO4 and ABIS supplemented with 300 µM ZnSO4 or 30 µM CdCl2. Results are shown as mean ± SE of three independent experiments, except the bar labelled * for which two independent experiments were performed.

mutant growth on low Fe (data not shown) and its Cd hyper- construct was introduced in A. thaliana accessions Ws and sensitivity (Fig. 6e). We noted, however, that expression of Columbia. However, the transgenic lines tested did not show 35S::TcNRAMP3 or 35S::TcNRAMP4 in the nramp3nramp4 any significant alteration in metal tolerance, nor consistent background did not lead to heavy metal tolerance that alteration in Cd or Zn accumulation (Fig. S3). exceeded that of the wild type. This coincides with the results previously obtained with 35S::AtNRAMP lines, which also Discussion did not display increased Cd or Zn tolerance (Thomine et al., 2003; Lanquar et al., 2004). To further test whether In this study, we report the identification and functional TcNRAMP4 expression in Arabidopsis wild-type background characterization of the homologues of A. thaliana NRAMP3 and could alter metal tolerance or content, a 35S TcNRAMP4 NRAMP4 in the metal hyperaccumulating plant T. caerulescens.

Fig. 6 Expression of TcNRAMP proteins in the nramp3nramp4 double mutant complements zinc (Zn)- and cadmium (Cd)-hypersensitivity. (a) Semi-quantitative reverse transcriptase polymerase chain reaction (RT-PCR) of TcNRAMP3 and actin (RNA concentration control) expression in the three nramp3nramp4 35S::TcNRAMP3 lines (nr3nr4_3-1, 2 and 7 ), wild type (WT; accession Ws) and the nramp3nramp4 control (nr3nr4). Polymerase chain reactions of 22 (TcNRAMP3) and 20 (actin) cycles were performed. (b) Arabidopsis thaliana seeds of the same three nramp3nramp4 35S::TcNRAMP3 lines were germinated on iron (Fe)-deficient medium (ABIS with 0.3 µM FeHBED), together with wild type as positive control and nramp3nramp4 as negative control and grown vertically for 11 d. (c) Root length of the same genotypes grown vertically

for 11 d on ABIS containing 50 µM FeHBED and 30 (normal), 100 or 300 µM ZnSO4. (d) Root length of the same genotypes grown vertically for 11 d on ABIS containing 50 µM FeHBED and 0, 3, 10 or 30 µM CdCl2. (e) Root length of three nramp3nramp4 35S::TcNRAMP4 lines (nr3nr4_4-#), wild type (WT; accession Ws) and the nramp3nramp4 control (nr3nr4) grown vertically for 12 d on ABIS containing 50 µM FeHBED a,b,c and 0, 3, 10 or 30 µM CdCl2. (c–e) Values are the average of measurements on 10–12 roots and bars represent SD. , Bars with different letters are significantly different according to a Mann–Whitney test with a 0.1% threshold.

Compared with AtNRAMP3 and AtNRAMP4, TcNRAMP3 35S::TcNRAMP3 or 35S::TcNRAMP4 in the nramp3nramp4 and TcNRAMP4 showed enhanced expression levels. Genomic mutant background led to complete restoration of its Southern blots indicated that TcNRAMP3 and TcNRAMP4 phenotypes, similar to the effect of 35S::AtNRAMP3 and are present at single loci. Expression of TcNRAMP3 and 35S::AtNRAMP4 expression (Thomine et al., 2003; Lanquar TcNRAMP4 in yeast mutants impaired in metal transport et al., 2005). revealed functional properties similar to their A. thaliana counterparts, and imaging of TcNRAMP–GFP fusions in plant Thlaspi caerulescens shows enhanced NRAMP cell protoplasts showed that they are also localized on the expression vacuole membrane. Furthermore, we showed that A. thaliana nramp3nramp4 mutant is hypersensitive to Cd and Zn in Thlaspi caerulescens and A. thaliana NRAMP3 and NRAMP4 addition to its previously described sensitivity to low iron display differences related to gene expression levels. The nutrition (Lanquar et al., 2005). Individual expression of upregulation of TcNRAMP3 in comparison with AtNRAMP3

New Phytologist (2009) 181: 637–650 © The Authors (2008) www.newphytologist.org Journal compilation © New Phytologist (2008) Research 647 shown here is in agreement with the T. caerulescens transcriptome the Ni hyperaccumulator T. japonicum specifically enhances analysis (van de Mortel et al., 2006). Interestingly, a similarly Ni sensitivity and Ni accumulation when expressed in yeast. enhanced expression was also found for A. halleri NRAMP3 TjNRAMP4 is however unable to complement yeast strains (Becher et al., 2004; Weber et al., 2004; Filatov et al., 2006; deficient in Fe, Mn and Zn uptake, in contrast to AtNRAMP4 Talke et al., 2006) and other metal homeostasis genes such as (Thomine et al., 2000; Lanquar et al., 2004). We did not find ZNT1, ZNT2, ZTP1, HMA4 and MT3 in T. caerulescens a similar enhanced Ni sensitivity when expressing the A. thaliana (Assunção et al., 2001; Bernard et al., 2004; Roosens et al., and T. caer ulescens NRAMP3 and NRAMP4 in yeast (data not 2004). In addition, we show that NRAMP4 also displays shown). Interestingly, our DNA blot experiments show the higher expression levels in T. caerulescens compared with presence of multiple NRAMP3 and NRAMP4 gene copies in A. thaliana. Hence NRAMP3 and NRAMP4 belong to a the Ni hyperaccumulator T. japonicum. This may be an example common set of genes involved in metal homeostasis, which of gene duplication similar to what has been described for display higher expression in both Zn- and Cd-hyperaccumulating several A. halleri metal homeostasis genes (Dräger et al., species. 2004; Talke et al., 2006). It is possible that specialization in Our genomic southern blots indicated that both NRAMP3 Ni transport in the particular TjNRAMP4 cDNA reported and NRAMP4 are present as single copy genes in T. caerulescens, by Mizuno et al. (2005) resulted in the loss of its ability to as suggested recently for TcNRAMP3 by Wei et al. (2008). transport other metals. Hence, as for ZNT1, ZNT2, ZTP1 (Assunção et al., 2001), YSL3, YSL5, YSL7 (Gendre et al., 2007) and the NAS genes A. thaliana nramp3nramp4 mutant is hypersensitive to (van de Mortel et al., 2006), their increased expression is most Cd and Zn likely caused by differences in regulatory cis or trans elements or in metal perception and signalling and not because of The nramp3nramp4 double mutant shows strong hypersensitivity extensive gene duplications as has been found for several metal to Cd and Zn compared with wild type. This phenotype is homeostasis genes in A. halleri (Talke et al., 2006; Hanikenne comparable to the strongest Cd hypersensitive mutants reported et al., 2008). However, our experiments do not exclude the to date, such as cad1 and cad2 (Howden et al., 1995a,b). We possibility of local duplication of NRAMP3 and NRAMP4 previously reported that 35S::AtNRAMP3 and 35S::AtNRAMP4 genes in T. caerulescens as found for HMA4 in Arabidopsis overexpressing plants also showed an enhanced Cd sensitivity halleri (Hanikenne et al., 2008). (Thomine et al., 2000; Lanquar et al., 2004). In addition, Wei et al. (2008) have shown a slight increase in Cd sensitivity of root growth in a 35S::TcNRAMP3-expressing tobacco line. Functional equivalence between A. thaliana and However, the phenotypes of NRAMP overexpressing plants T. caerulescens NRAMP proteins are weak compared with the hypersensitivity described Our results show that, when expressed as GFP fusion in here for the nramp3nramp4 mutant. The hypersensitivity A. thaliana protoplasts, the T. caerulescens NRAMP proteins associated with NRAMP overexpression in A. thaliana have the same intracellular localization as their A. thaliana differs from the hypersensitivity resulting from NRAMP3 and homologues. We further found by heterologous expression in NRAMP4 inactivation in its dependence on iron nutrition yeast that they also display strong similarity in their metal and in its specificity. While nramp3nramp4 hypersensitivity to transport abilities, in agreement with Wei et al. (2008). This Cd and Zn is observed even under high Fe nutrition, suggests that the T. caerulescens NRAMP3 and NRAMP4 hypersensitivity of 35S::AtNRAMP4 plants to Cd is apparent proteins have thus not developed a specific function different only under low Fe nutrition and is abolished at elevated Fe from A. thaliana. nutrition (Lanquar et al., 2004). In addition, 35S::AtNRAMP4 In agreement with Lanquar et al. (2004), we show here that plant hypersensitivity to metal is specific to Cd (Lanquar AtNRAMP4 or TcNRAMP4 complement zrt1zrt2 while et al., 2004), while nramp3nramp4 displays hypersensitivity AtNRAMP3 or TcNRAMP3 do not, indicating that NRAMP3 to both Cd and Zn. and NRAMP4 differ in their ability to transport Zn. Such These different features of metal hypersensitivity observed differences in metal selectivity in closely related metal trans- in nramp3nramp4 double mutants or triggered by AtNRAMP3, porters have also been reported for members of the ZIP family: AtNRAMP4 or TcNRAMP3 overexpression suggest that A. thaliana IRT1 transports Mn and Cd while IRT2 does not they originate from different mechanisms. Cadmium (Korshunova et al., 1999; Vert et al., 2001) and rice ZIP3 hypersensitivity induced by AtNRAMP3 or AtNRAMP4 transports Cd while ZIP2 does not (Ramesh et al., 2003). overexpression was proposed to be caused by increased Cd The conserved metal transport properties between remobilization from the vacuole (Thomine et al., 2003). By TcNRAMP3 and AtNRAMP3 on one hand, and AtNRAMP4 contrast, nramp3nramp4 pleiotropic metal hypersensitivity and TcNRAMP4 on the other hand, contrast strongly with may be the result of a globally unbalanced metal homeostasis. the highly divergent function reported for TjNRAMP4 by We propose that nramp3nramp4 hypersensitivity to Zn and Mizuno et al. (2005). They reported that TjNRAMP4 from Cd is related to a general defect in mobilization of essential

metals from the vacuole, which would be required for tolerance References to metals. Assunção AGL, Da Costa Martins P, de Folter S, Vooijs R, Schat H, Aarts MGM. 2001. Elevated expression of metal transporter genes in three TcNRAMP3 and TcNRAMP4 are functional orthologues accessions of the metal hyperaccumulator Thlaspi caerulescens. Plant, Cell & Environment 24: 217. of AtNRAMP3 and AtNRAMP4 Baker AJM, Proctor J, van Balgooy MMJ, Reeves DR 1992. The heterologous expression in yeast and the localization of Hyperaccumulation of nickel by the flora of the ultramafics of Palawan, republic of the Philippines. In: Baker AJM, Proctor J, Reeves DR, TcNRAMPs in A. thaliana did not reveal functional differences eds. The vegetation of ultramafic (serpentine) soils. Proceedings of the first with their A. thaliana homologues. Functional similarity international conference on serpentine ecology. Andover, UK: Intercept, was confirmed by expression of TcNRAMP genes in the 291–304. nramp3nramp4 double mutant. Expression of TcNRAMP3 or Becher M, Talke IN, Krall L, Krämer U. 2004. Cross-species microarray TcNRAMP4 in nramp3nramp4 restores the sensitivity of this transcript profiling reveals high constitutive expression of metal homeostasis genes in shoots of the zinc hyperaccumulator Arabidopsis mutant to Fe starvation, Zn excess and Cd toxicity back to halleri. Plant Journal 37: 251–268. wild type levels. Belouchi A, Kwan T, Gros P. 1997. Cloning and characterization of the In conclusion, our study of the T. caerulescens and the A. Osnramp family from Oryza sativa, a new family of membrane proteins thaliana NRAMP3 and NRAMP4 genes strongly indicates possibly implicated in the transport of metal ions. Plant Molecular Biology that they are true functional orthologues, with identical protein 33: 1085–1092. Bereczky Z, Wang HY, Schubert V, Ganal M, Bauer P. 2003. Differential function. On the other hand, we confirm the transcriptional regulation of nramp and irt metal transporter genes in wild type and differences between these NRAMP genes, which are expressed iron uptake mutants of tomato. The Journal of Biological Chemistry at much higher levels in T. caerulescens than in A. thaliana. 278: 24697–24704. From these results, it is difficult to conclude on the functional Bernard C, Roosens N, Czernic P, Lebrun M, Verbruggen N. 2004. A novel significance of the high expression of NRAMP3 and NRAMP4 cpx-atpase from the cadmium hyperaccumulator Thlaspi caerulescens. FEBS Letters 569: 140–148. in the metal tolerance of T. caerulescens. The Zn- and Cd-sensitive Bert V, Macnair MR, de Laguerie P, Saumitou-Laprade P, Petit D. 2000. phenotype of the nramp3nramp4 mutant suggests that Zinc tolerance and accumulation in metallicolous and nonmetallicolous mobilization of essential metals from the vacuole is necessary populations of Arabidopsis halleri (Brassicaceae). New Phytologist for tolerance to Zn and Cd. Although high expression of 146: 225–233. TcNRAMP4 in A. thaliana does not modify tolerance to Zn Brooks RR 1994. Plants that hyperaccumulate heavy metals. In: Farago ME, ed. Plants and the chemical elements. Weinheim, Germany: or Cd, it is possible that elevated expression of TcNRAMP3 VCH, 87–107. and TcNRAMP4 is necessary to tolerate high Zn and Cd Brown SL, Chaney RL, Angle JS, Baker AJM. 1995. Zinc and cadmium concentrations associated with metal hyperaccumulation in uptake by hyperaccumulator Thlaspi caerulescens grown in nutrient T. caerulescens. The elevated expression of TcNRAMPs may solution. Soil Science Society of America Journal 59: 125–133. also be a consequence rather than a cause of the increased Zn Clemens S. 2001. Molecular mechanisms of plant metal tolerance and homeostasis. Planta 212: 475–486. and Cd levels, as described for IRT3 and ZIP4 in A. halleri Clough SJ. 2005. Floral dip: Agrobacterium-mediated germ line roots (Hanikenne et al., 2008), but this elevated expression transformation. Methods in Molecular Biology 286: 91–102. may nevertheless be essential to maintain a balanced metal Cobbett C, Goldsbrough P. 2002. Phytochelatins and metallothioneins: homeostasis. It will be interesting to study the origin of these roles in heavy metal detoxification and homeostasis. Annual Review of expression level differences and whether the spatial expression Plant Biology 53: 159–182. Curie C, Alonso JM, Le Jean M, Ecker JR, Briat JF. 2000. Involvement pattern of NRAMP3 and NRAMP4 is also modified in of nramp1 from Arabidopsis thaliana in iron transport. The Biochemical hyperaccumulating species. Journal 347 (Pt 3): 749–755. Davis SJ, Vierstra RD. 1998. Soluble, highly fluorescent variants of green fluorescent protein (gfp) for use in higher plants. Plant Molecular Biology Acknowledgements 36: 521–528. Dix DR, Bridgham JT, Broderius MA, Byersdorfer CA, Eide DJ. 1994. The We acknowledge Bettine Aigner, Paula Pongrac, Andrea Pirondini fet4 gene encodes the low affinity Fe(II) transport protein of Saccharomyces and Takafumi Mizuno for providing us with seeds of different cerevisiae. The Journal of Biological Chemistry 269: 26092–26099. T. caerulescens accessions and Thlaspi species and Judith van Dräger DB, Desbrosses-Fonrouge AG, Krach C, Chardonnens AN, Meyer de Mortel for her assistance in preparing the T. caerulescens cDNA. RC, Saumitou-Laprade P, Krämer U. 2004. Two gen es enc odi ng We thank Susanne Bolte and Marie-Noëlle Solers for help with Arabidopsis halleri mtp1 metal transport proteins co-segregate with zinc tolerance and account for high mtp1 transcript levels. Plant Journal confocal microscopy at ‘The Imaging and Cell Biology facility’ 39: 425–439. of the IFR87 (FR-W2251) ‘La plante et son environnement’, Eide D, Broderius M, Fett J, Guerinot ML. 1996. A novel iron-regulated supported by Action de Soutien à la Technologie et la Recherche metal transporter from plants identified by functional expression in yeast. en Essonne, Conseil de l’Essonne. We thank Ute Krämer Proceedings of the National Academy of Sciences, USA 93: 5624–5628. for assistance with metal content analyses. This work was Filatov V, Dowdle J, Smirnoff N, Ford-Lloyd B, Newbury HJ, Macnair MR. 2006. Comparison of gene expression in segregating families supported by the European Union (Research Training identifies genes and genomic regions involved in a novel adaptation, Network METALHOME, HPRN-CT-2002-00243). zinc hyperaccumulation. Molecular Ecology 15: 3045–3059.

New Phytologist (2009) 181: 637–650 © The Authors (2008) www.newphytologist.org Journal compilation © New Phytologist (2008) Research 649 de Folter S, Shchennikova AV, Franken J, Busscher M, Baskar R, Pence NS, Larsen PB, Ebbs SD, Letham DL, Lasat MM, Garvin DF, Grossniklaus U, Angenent GC, Immink RG. 2006. A bsister mads-box Eide D, Kochian LV. 2000. The molecular physiology of heavy metal gene involved in ovule and seed development in petunia and Arabidopsis. transport in the Zn/Cd hyperaccumulator Thlaspi caerulescens. Proceedings Plant Journal 47: 934–946. of the National Academy of Sciences, USA 97: 4956–4960. Gendre D, Czernic P, Conéjéro G, Pianelli K, Briat JF, Lebrun M, Persans MW, Nieman K, Salt DE. 2001. Functional activity and role Mari S. 2007. TcYSL3, a member of the YSL gene family from the of cation-efflux family members in Ni hyperaccumulation in Thlaspi hyper-accumulator Thlaspi caerulescens, encodes a nicotinamine-Ni/Fe goesingense. Proceedings of the National Academy of Sciences, USA transporter. Plant Journal 49: 1–15. 98: 9995–10000. Hanikenne M, Talke IN, Haydon MJ, Lanz C, Nolte A, Motte P, Pinner E, Gruenheid S, Raymond M, Gros P. 1997. Functional Kroymann J, Weigel D, Krämer U. 2008. Evolution of metal complementation of the yeast divalent cation transporter family smf hyperaccumulation required cis-regulatory changes and triplication by nramp2, a member of the mammalian natural resistance-associated of HMA4. Nature 453, 391–395. macrophage protein family. The Journal of Biological Chemistry Howden R, Andersen CR, Goldsbrough PB, Cobbett CS. 1995a. 272: 28933–28938. A cadmium-sensitive, glutathione-deficient mutant of Arabidopsis Ramesh SA, Shin R, Eide DJ, Schachtman DP. 2003. Differential metal thaliana. Plant Physiology 107: 1067–1073. selectivity and gene expression of two zinc transporters from rice. Howden R, Goldsbrough PB, Andersen CR, Cobbett CS. 1995b. Plant Physiology 133: 126–134. Cadmium-sensitive, cad1 mutants of Arabidopsis thaliana are Rigola D, Fiers M, Vurro E, Aarts MGM. 2006. The heavy metal phytochelatin deficient. Plant Physiology 107: 1059–1066. hyperaccumulator Thlaspi caerulescens expresses many species-specific Kaiser BN, Moreau S, Castelli J, Thomson R, Lambert A, Bogliolo S, genes, as identified by comparative expressed sequence tag analysis. Puppo A, Day DA. 2003. The soybean nramp homologue, gmdmt1, New Phytologist 170: 753–765. is a symbiotic divalent metal transporter capable of ferrous iron transport. Rogers EE, Eide DJ, Guerinot ML. 2000. Altered selectivity in an Plant Journal 35: 295–304. arabidopsis metal transporter. Proceedings of the National Academy Korshunova YO, Eide D, Clark WG, Guerinot ML, Pakrasi HB. of Sciences, USA 97: 12356–12360. 1999. The irt1 protein from Arabidopsis thaliana is a metal Roosens NH, Bernard C, Leplae R, Verbruggen N. 2004. Evidence transporter with a broad substrate range. Plant Molecular Biology for copper homeostasis function of metallothionein (mt3) in the 40: 37–44. hyperaccumulator Thlaspi caerulescens. FEBS Letters 577: 9–16. Küpper H, Lombi E, Zhao FJ, McGrath SP. 2000. Cellular Schat H, Llugany M, Bernhard R 2000. Metal-specific patterns of tolerance, compartmentation of cadmium and zinc in relation to other uptake, and transport of heavy metals in hyperaccumulating and elements in the hyperaccumulator Arabidopsis halleri. Planta nonhyperaccumulating metallophytes. In: Terry N, Banuelos G, eds. 212: 75–84. Phytoremediation of contaminated soils and water. Boca Raton, FL, USA: Lanquar V, Lelievre F, Barbier-Brygoo H, Thomine S. 2004. Regulation CRC Press LLC, 171–188. and function of Atnramp4 metal transporter protein. Soil Science and Plant Supek F, Supekova L, Nelson H, Nelson N. 1996. A yeast manganese Nutrition 50: 1141–1150. transporter related to the macrophage protein involved in conferring Lanquar V, Lelievre F, Bolte S, Hames C, Alcon C, Neumann D, Vansuyt resistance to mycobacteria. Proceedings of the National Academy of Sciences, G, Curie C, Schroder A, Kramer U et al. 2005. Mobilization of vacuolar USA 93: 5105–5110. iron by Atnramp3 and Atnramp4 is essential for seed germination on low Talke IN, Hanikenne M, Krämer U. 2006. Zinc-dependent global iron. EMBO Journal 24: 4041–4051. transcriptional control, transcriptional deregulation, and higher gene Lombi E, Zhao FJ, Dunham SJ, McGrath SP. 2000. Cadmium copy number for genes in metal homeostasis of the hyperaccumulator accumulation in populations of Thlaspi caerulescens and Thlaspi Arabidopsis halleri. Plant Physiology 142: 148–167. goesingense. New Phytologist 145: 11–20. Thomine S, Lelievre F, Debarbieux E, Schroeder JI, Barbier-Brygoo H. Mäser P, Thomine S, Schroeder JI, Ward JM, Hirschi K, Sze H, 2003. Atnramp3, a multispecific vacuolar metal transporter involved Talke IN, Amtmann A, Maathuis FJ, Sanders D et al. 2001. in plant responses to iron deficiency. Plant Journal 34: 685–695. Phylogenetic relationships within cation transporter families Thomine S, Wang R, Ward JM, Crawford NM, Schroeder JI. 2000. of Arabidopsis. Plant Physiology 126: 1646–1667. Cadmium and iron transport by members of a plant metal transporter Meerts P, van Isacker N. 1997. Heavy metal tolerance and accumulation family in Arabidopsis with homology to nramp genes. Proceedings of the in metallicolous and nonmetallicolous populations of Thlaspi caerulescens National Academy of Sciences, USA 97: 4991–4996. from continental europe. Plant Ecology 133: 153–159. Vert G, Briat JF, Curie C. 2001. Arabidopsis irt2 gene encodes a Mizuno T, Usui K, Horie K, Nosaka S, Mizuno N, Obata H. 2005. root-periphery iron transporter. Plant Journal 26: 181–189. Cloning of three zip/nramp transporter genes from a Ni hyperaccumulator Vogel-Mikus K, Pongrac P, Kump P, Necemer M, Regvar M. 2006. plant Thlaspi japonicum and their ni2+-transport abilities. Plant Physiology Colonisation of a Zn, Cd and Pb hyperaccumulator Thlaspi praecox wulfen and Biochemistry 43: 793–801. with indigenous arbuscular mycorrhizal fungal mixture induces changes in van de Mortel JE, Almar Villanueva L, Schat H, Kwekkeboom J, Coughlan heavy metal and nutrient uptake. Environmental Pollution 139: 362–371. S, Moerland PD, Ver Loren van Themaat E, Koornneef M, Aarts MGM. Weber M, Harada E, Vess C, Roepenack-Lahaye E, Clemens S. 2004. 2006. Large expression differences in genes for iron and zinc homeostasis, Comparative microarray analysis of Arabidopsis thaliana and Arabidopsis stress response, and lignin biosynthesis distinguish roots of Arabidopsis halleri roots identifies nicotinamine synthase, a zip transporter and other thaliana and the related metal hyperaccumulator Thlaspi caerulescens. genes as potential metal hyperaccumulation factors. Plant Journal Plant Physiology 142: 1127–1147. 37: 269–281. van de Mortel JE, Schat H, Moerland PD, Ver Loren van Themaat E, van Wei W, Chai T, Zhang Y, Han L, Xu J, Guan, Z. 2008. The Thlaspi der Ent S, Blankestijn H, Ghandilyan A, Tsiatsiani S, Aarts MGM. 2008. caerulescens NRAMP homologue TcNRAMP3 is capable of divalent Expression differences for genes involved in lignin, glutathione and cation transport. Molecular Biotechnology. DOI 10.1007/x12033-008- sulphate metabolism in response to cadmium in Arabidopsis thaliana and 9088-x. the related Zn/Cd-hyperaccumulator Thlaspi caerulescens. Plant, Cell & Yang XE, Long XX, Ye HB, He ZL, Calvert DV, Stoffella PJ. 2003. Environment 31: 301–324. Cadmium tolerance and hyperaccumulation in a new Zn- Nevo Y, Nelson N. 2006. The nramp family of metal-ion transporters. hyperaccumulating plant species (Sedum alfredii Hance). Plant and Soil Biochimica et Biophysica Acta 1763: 609–620. 66: 1–9.

Zhao H, Eide D. 1996a. The yeast zrt1 gene encodes the zinc transporter Fig. S2 Genomic DNA blot analysis of NRAMP3 and protein of a high-affinity uptake system induced by zinc limitation. NRAMP4 in Thlaspi aerulescens accessions and related Thlaspi Proceedings of the National Academy of Sciences, USA 93: 2454–2458. species. Zhao H, Eide D. 1996b. The zrt2 gene encodes the low affinity zinc transporter in Saccharomyces cerevisiae. The Journal of Biological Chemistry Fig. S3 Expression of TcNRAMP4 in Arabidopsis thaliana 271: 23203–23210. accessions Ws and Columbia does not modify Zn or Cd sensitivity nor their content. Supporting Information Table S1 Expression analysis of NRAMP3 and NRAMP4 Additional supporting information may be found in the expression levels in Arabidopsis thaliana and Thlaspi caerulescens online version of this article. Please note: Wiley-Blackwell are not responsible for the con- Fig. S1 Expression analysis of NRAMP3 and NRAMP4 in tent or functionality of any supporting information supplied Arabidopsis thaliana and Thlaspi caerulescens in response to by the authors. Any queries (other than missing material) different Zn supply. should be directed to the New Phytologist Central Office.

About New Phytologist

• New Phytologist is owned by a non-profit-making charitable trust dedicated to the promotion of plant science, facilitating projects from symposia to open access for our Tansley reviews. Complete information is available at www.newphytologist.org.

• Regular papers, Letters, Research reviews, Rapid reports and both Modelling/Theory and Methods papers are encouraged. We are committed to rapid processing, from online submission through to publication ‘as-ready’ via Early View – our average submission to decision time is just 29 days. Online-only colour is free, and essential print colour costs will be met if necessary. We also provide 25 offprints as well as a PDF for each article.

• For online summaries and ToC alerts, go to the website and click on ‘Journal online’. You can take out a personal subscription to the journal for a fraction of the institutional price. Rates start at £139 in Europe/$259 in the USA & Canada for the online edition (click on ‘Subscribe’ at the website).

• If you have any questions, do get in touch with Central Office ([email protected]; tel +44 1524 594691) or, for a local contact in North America, the US Office ([email protected]; tel +1 865 576 5261).

New Phytologist (2009) 181: 637–650 © The Authors (2008) www.newphytologist.org Journal compilation © New Phytologist (2008)