Selenate-resistant mutants of Arabidopsis thaliana identify Sultr1;2, a sulfate transporter required for efficient transport of sulfate into roots


  • Nakako Shibagaki,

    1. Department of Botany, Iowa State University, Ames, Iowa 50011, USA,
    2. Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, the University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan, and
    Search for more papers by this author
  • Alan Rose,

    1. Section of Molecular and Cellular Biology, Division of Biological Sciences, University of California, Davis, 1 Shields Avenue, Davis, California 95616, USA
    Search for more papers by this author
  • Jeffrey P. McDermott,

    1. Department of Botany, Iowa State University, Ames, Iowa 50011, USA,
    Search for more papers by this author
  • Toru Fujiwara,

    1. Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, the University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan, and
    Search for more papers by this author
  • Hiroaki Hayashi,

    1. Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, the University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan, and
    Search for more papers by this author
  • Tadakatsu Yoneyama,

    1. Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, the University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan, and
    Search for more papers by this author
  • John P. Davies

    Corresponding author
    1. Department of Botany, Iowa State University, Ames, Iowa 50011, USA,
    Search for more papers by this author
    • Current address: Exelixis Plant Sciences, 16160 SW Upper Boones Ferry Road, Portland, Oregon 97224-7744, USA.

* For correspondence (fax +1 503 670 7703; e-mail


To investigate how plants acquire and assimilate sulfur from their environment, we isolated and characterized two mutants of Arabidopsis thaliana deficient in sulfate transport. The mutants are resistant to selenate, a toxic analogue of sulfate. They are allelic to each other and to the previously isolated sel1 (selenate-resistant) mutants, and have been designated sel1-8 and sel1-9. Root elongation in these mutants is less sensitive to selenate than in wild-type plants. Sulfate uptake into the roots is impaired in the mutants under both sulfur-sufficient and sulfur-deficient conditions, but transport of sulfate to the shoot is not affected. The sel1 mutants contain lesions in the sulfate transporter gene Sultr1;2 located on the lower arm of chromosome 1. The sel1-1, sel1-3 and sel1-8 mutants contain point mutations in the coding sequences of Sultr1;2, while the sel1-9 mutant has a T-DNA insertion in the Sultr1;2 promoter. The Sultr1;2 cDNA derived from wild-type plants is able to complement Saccharomyces cerevisiae mutants defective in sulfate transport, but the Sultr1;2 cDNA from sel1-8 is not. The Sultr1;2 gene is expressed mainly in roots, and accumulation of transcripts increases during sulfate deprivation. Examination of transgenic plants containing the Sultr1;2 promoter fused to the GUS-reporter gene indicates that Sultr1;2 is expressed mainly in the root cortex, the root tip and lateral roots. Weaker expression of the reporter gene was observed in hydathodes, guard cells and auxiliary buds of leaves, and in anthers and the basal parts of flowers. The results indicate that Sultr1;2 is primarily involved in importing sulfate from the environment into the root.


Plants import sulfur from the soil through the roots as sulfate. Inside the plant, sulfate is metabolized in the plastids where it is incorporated into amino acids and a variety of other sulfo-organic compounds. Sulfate is a relatively inert compound and must be activated by ATP sulfurylase prior to being metabolized. Adenosine 5′-phosphosulfate, the activated form of sulfate, is reduced and incorporated into the amino acids cysteine and methionine, and used in sulfation reactions to produce sulfated lipids and polysaccharides.

As sulfate is imported into the roots and metabolized in plastids, a network of transporters is required to move it throughout the plant. Sulfate is imported from the soil into the root symplast and transported across the root to the central stele. It is then imported into the xylem, transported to the shoot, discharged into leaf cells, and finally transported into the chloroplast, the predominant site of sulfur reduction. Sulfate is also transported across the tonoplast into the vacuole where it can be stored at relatively high concentrations (Saito, 2000). Furthermore, the observed increase in sulfate uptake when plants are deprived of sulfate (Smith et al., 1997; Takahashi et al., 1997) may result from induced expression of specific sulfate transporters. Fungi and algae synthesize new high-affinity sulfate transport systems when placed in sulfur-deficient conditions (Marzluf, 1970; Yildiz et al., 1994).

Sequence analysis indicates that the Arabidopsis thaliana genome has 14 putative sulfate transporter genes. These transporters fall into four closely related phylogenetic groups (Grossman and Takahashi, 2001; Takahashi et al., 2000), 12 have 12 membrane-spanning domains and a STAS domain at their carboxy-terminus (Aravind and Koonin, 2000) the other 2 have 12 membrane spanning domains but lack the STAS domain. Plant sulfate transporters function as proton/sulfate co-transporters, transporting three protons with each sulfate ion. The driving force for sulfate transport is the pH gradient across the membrane set up by a proton pumping ATPase (Hawkesford et al., 1993; Lass and Ullrich-Eberius, 1984).

Several sulfate transporters from Arabidopsis have been characterized. Kinetic analysis of two Arabidopsis sulfate transporters in yeast cells has identified Sultr1;1 as a high-affinity transporter and Sultr2;1 as a low-affinity transporter (Takahashi et al., 2000). Gene expression patterns of four sulfate transporters have been analysed in transgenic Arabidopsis plants expressing reporter genes from the promoters of these genes. The results indicate that Sultr1;1 is expressed in the root tips and external cell layers of the roots, Sultr2;1 in the vascular tissue of both the leaves and the roots, Sultr2;2 in the phloem of roots and vascular bundle sheath in leaves (Takahashi et al., 2000), and Sultr4;1 in the chloroplast (Takahashi et al., 1999). Because it has a high affinity for sulfate and is expressed in the root tip, Sultr1;1 has been proposed to be the transporter primarily responsible for sulfate uptake from the soil solution. Sultr2;1 is thought to be primarily responsible for transport of sulfate inside the plant from one tissue to another (Takahashi et al., 2000). Multiple sulfate transporters with unique kinetic properties and gene-expression patterns appear to be necessary for plants to import sulfate from the soil solution and transport it throughout the plant under a variety of environmental conditions (Saito, 2000).

Although progress has been made in identifying sulfate transporters and examining patterns of gene expression, the role of most of the sulfate transporters in sulfur metabolism remains unclear. To begin to dissect how these transporters work together, we have used a genetic approach to identify Arabidopsis plants containing lesions in a specific sulfate transporter.

To identify sulfate transport mutants, we screened for plants resistant to toxic analogues of sulfate. Selenate and chromate are thought to be assimilated into modified forms of cysteine and methionine via the sulfate assimilation pathway. Incorporation of these modified amino acids into proteins may disrupt protein structure and enzymatic activity, and inhibit growth. We reasoned that plants resistant to these compounds might contain lesions in a sulfate transporter or an enzyme involved in sulfate assimilation, and enable us to begin to dissect the sulfate assimilation pathway of Arabidopsis. Identification of selenate- and chromate-resistant micro-organisms has been invaluable for elucidating sulfur transport or assimilation mechanisms in these organisms (Breton and Surdin-Kerjan, 1977; Cherest et al., 1997; Marzluf, 1970; Smith et al., 1995a).

Here we report on the identification and characterization of Arabidopsis mutants with lesions in the Sultr1;2 gene. Our results indicate that Sultr1;2 is primarily expressed in the root and is mainly involved in transporting sulfate from the soil into the root. It does not appear to be involved in transporting sulfate to the shoot.


Isolation and genetic characterization of selenate-resistant mutants

Selenate-resistant mutants of Arabidopsis were identified by germinating M2 seeds of a mutagenized population on solid medium containing 20 µm selenate and 40 µm chromate, allowing the seedlings to grow for 2 weeks and comparing root length. Under these conditions, the roots of wild-type plants were very short (<1 mm). Plants from the mutagenized population with longer roots (≈3–5 mm) were selected for further analysis. From ten thousand EMS mutagenized M2 lines of the Columbia (Col-0) accession, seven putative selenate-resistant mutants were identified. From one hundred thousand T4 seedlings of 9510 T-DNA tagged lines of the Wassilewskija (Ws) accession, 10 putative selenate-resistant mutants were identified. After re-screening progeny of the putative mutants, we identified one mutant line from the EMS mutagenized population, and another from the T-DNA mutagenized population.

Genetic analysis indicates that the selenate-resistant phenotypes of mutants are caused by single recessive mutations (Table 1). F1 progeny of the both mutant lines crossed to parental wild-type strains are sensitive to selenate, indicating that the lesions are recessive. Selenate sensitivity in F2 progeny of both mutant lines segregates in a 3 : 1 ratio (sensitive : resistant), indicating that the mutant phenotypes are caused by a mutation at a single locus.

Table 1.  Genetic segregation of the selenate-resistant phenotype
Strain or crossGenerationnSenaResaχ2 b
  • a Sen, sensitive; Res, resistant on medium lacking sulfate and containing 20 µm selenium and 40 µm chromate.

  • bχ2

    value calculated based on an expected ratio of 3 : 1 segregation.

  • c

    P > 0.1.

  • d

    sel1-8 corresponds to mutant line from EMS mutagenized population (see text).

  • e

    sel1-9 corresponds to mutant line from T-DNA mutagenized population (see text).

  • f

    All crosses were performed reciprocally. There was no significant difference in the ratio of segregation between the reciprocal crosses (data not shown). The above figures are the sum of all crosses performed.

WT Col-0 85850 
WT Ws 78780 
Col-0 × sel1-8fF2300228720.124c
Ws × sel1-9fF28964250.45c
sel1-8 × sel1-9fF122022 
sel1-8d × sel1-1fF1606 
sel1-9e × sel1-3fF1606 

Genetic complementation tests indicate that the two selenate-resistant mutants identified are in the same complementation group (Table 1). The mutant line from the EMS mutagenized population and the line from the T-DNA mutagenized population were crossed to each other, and all F1 and F2 progeny tested were resistant to selenate. The results indicate that these mutants contain lesions at the same genetic locus. Rose (1997) previously isolated seven allelic selenate-resistant mutants (sel1-1 to sel1-7) by screening for plant growth on medium containing 10 µm selenate. To determine whether the mutants that we isolated are allelic with sel1, genetic complementation tests were performed by crossing the mutant from the EMS mutagenized population with sel1-1, and the mutant from the T-DNA mutagenized population with sel1-3. All F1 and F2 progeny from these crosses were selenate-resistant (Table 1), indicating that the mutants are all alleles of sel1. We have named the mutant isolated from the EMS mutagenized population sel1-8, and the mutant from the T-DNA mutagenized population sel1-9.

The selenate-resistant phenotype of sel1-9 co-segregates with a T-DNA insertion. The sel1-9 mutant line was isolated from a T-DNA tagged population generated by transformation of wild-type Arabidopsis with a Ti plasmid (Forsthoefel et al., 1992). This plasmid carries the neomycine phosphotransferase (NPTII) gene that confers kanamycin resistance. Kanamycin resistance in the F2 progeny of wild type and the sel1-9 mutant segregates 61 : 26 (resistant : sensitive), suggesting that the line sel1-9 carries an NPTII gene at a single genetic locus (P > 0.1). In addition, all 30 selenate-resistant F2 plants tested were kanamycin-resistant, and all 12 kanamycin-sensitive F2 plants examined were sensitive to selenate, suggesting that the T-DNA insertion caused the mutation conferring selenate resistance.

Phenotypic characterization of the mutants

The selenate-resistant mutants sel1-8 and sel1-9 were isolated based on their ability to grow longer roots than wild-type plants on sulfur-deficient medium containing selenate and chromate. To test if the mutants were able to grow roots more efficiently than the wild-type strains under other conditions, root growth was compared on medium lacking selenate and chromate, or containing various concentrations of these compounds, both in the presence and absence of sulfate. Root growth of mutant and wild-type plants was identical on sulfate-replete and sulfate-deficient media in the absence of selenate and chromate (Figure 1b). However, both sel1-8 and sel1-9 grew longer roots than the wild-type plants on sulfate-deficient medium containing selenate (Figure 1a,c). Chromate did not significantly affect root growth of wild-type and mutant plants at the concentrations used for screening (data not shown), suggesting that selenate was the major cause for the defect in root elongation. Although root growth in sel1-8 and sel1-9 is sensitive to selenate under sulfur-deficient conditions, it is less sensitive than in wild-type plants (Figure 1c). Root growth in sel1-9 was also less sensitive to selenate than the Ws wild-type strain in the presence of 0.7 mm sulfate (data not shown).

Figure 1.

Root growth in wild-type and mutant plants.

The sel1-8 and sel1-9 mutants are resistant to selenate.

(a,b) show pictures of seedlings germinated on solid medium. Wild-type (Col-0, Ws) and mutant (sel1-8 and sel1-9) seedlings were germinated on medium lacking sulfate containing 5 or 10 µm selenate (a), or containing 1600 or 0 µm sulfate with no selenate. (b) Seedlings were grown for 7 days and photographed through a dissecting microscope. Bars are 0.5 cm (a) and 1 cm (b).

(c) Root length of 9-day-old-seedlings. Root length (mean standard deviation, n = 20) of Col-0 (◆) and sel1-8 (▪) (upper panel) and Ws (▴) and sel1-9 (•) (lower panel) was measured 9 days after sowing on solid medium lacking sulfate and containing 0, 2.5, 5.0, 10, 20 or 40 µm selenate.

The selenate-resistant phenotype of the mutants could be caused by a defect in sulfate uptake or assimilation. To determine if sulfate transport was disrupted in the mutants, we compared sulfate uptake in sel1-9 and Ws wild type. Plants were grown under sulfate-replete conditions or starved for sulfate for 72 h and sulfate uptake was measured at several different sulfate concentrations. The results shown in Figure 2 indicate that in both mutant and wild-type plants, the rate of sulfate transport is higher in plants deprived of sulfate for 72 h than in plants maintained in sulfur-replete medium. However under both sulfur-replete and sulfur-deficient conditions, transport of sulfate into the roots of mutant plants was slower than into the roots of wild-type plants (Figure 2a). Similar results were obtained when sulfate transport in sel1-8 and Col-0 was compared (data not shown). The results suggest that these mutants are deficient in a sulfate transporter that contributes to sulfate transport into roots at a broad range of concentrations under both sulfate-sufficient and sulfate-deficient conditions.

Figure 2.

Sulfate uptake rate is impaired in the sel1 mutants.

Sulfate uptake rates in Ws (▵, ▴) and sel1-9 (○, •) into roots (a)and shoots (b)of plants grown in sulfur-replete medium (filled symbols, ▴, •) and sulfur-deficient medium (open symbols, ▵, ○). (a) The total amount of sulfate (pmol) imported into the entire plant through roots during 1 hour uptake is divided by the root fresh weight (mg). (B) The amount of sulfate imported into the shoot (pmol) is divided by the shoot fresh weight (mg). All the values represent the means of three to five experiments, and each experiment was composed of three to six measurements.

The sel1-9 mutant shows no defect in the rate of transport of sulfate into the shoot. These rates were calculated by dividing the amount of radioactive sulfate transported into the shoot by the fresh weight of the shoot (Figure 2b). There was no difference in the mass of the wild-type and mutant plants (data not shown). The sel1-9 mutant appears to be able to maintain sulfate transport into the shoot despite lower rates of sulfate transport into the roots.

Mapping and identification of the mutations

To identify the gene responsible for the mutant phenotype, the lesion in sel1-8 was genetically mapped. The mutant (Col-0 accession) was crossed with a wild-type plant from the Ler accession. The F1 plants were allowed to self-fertilize and the resulting F2 plants used to genetically map the mutation. F2 seeds were germinated on solid medium containing 20 µm selenate and scored for selenate resistance. Total DNA was isolated from F2 plants, and segregation of molecular genetic markers was analysed. Phenotypes were confirmed by examining the F3 plants.

The mutation was mapped to the lower arm of chromosome 1 between the markers nF22K20 and SGCSNP253 by examining 197 F2 progeny. Seven recombinations among 372 chromosomes were observed between the mutation and nF22K20, and five recombinations among 394 chromosomes between the mutation and SGCSNP253. This mapped the mutation to a region of ≈400 kbp (Figure 3a).

Figure 3.

The sel1 mutations map to the bottom of chromosome I in the Sultr1;2 gene.

(a)Map of chromosome I. Seven recombinations in 372 chromosomes and five recombinations in 394 chromosomes were observed between the sel1-8 mutation and the markers F22K20 and SGCSNP253, respectively. The horizontal black bars represent BAC clones that map between F22K20 and SGCSNP23.

(b)The BAC F28K19 contains genes encoding the sulfate transporters Sultr1;2 (hatched arrow) and Sultr2;2 (open arrow). The direction of the arrow indicates the direction of transcription. Asterisks represent the position of the lesions in sel1-1, sel1-3 and sel1-8. T-DNA was integrated at 434 bp upstream of translation initiation site of Sultr1;2 in sel1-9. This map is based on data from the Arabidopsis Information Resource (

Seven BAC clones spanning the region between these two markers had been sequenced and partially annotated by the Arabidopsis Genome Initiative at the time of this analysis. One of these, F28K19 (GenBank accession no. AC009243), contained two genes encoding sulfate transporters: Sultr2;2 (GenBank accession no. AB012047; Takahashi et al., 1996) and Sultr1;2 (GenBank accession no. AB042322; Yoshimoto et al., 2002). Because the mutants were impaired in sulfate transport, we investigated whether either of these genes contained a mutation. Genomic DNA fragments corresponding to these genes from sel1-8 were PCR-amplified and sequenced. To examine the Sultr2;2 gene, a DNA fragment spanning the region from 1.7 kbp upstream of the start codon to 0.9 kbp downstream of the stop codon was amplified by PCR. No difference in DNA sequence between the amplified fragment from the sel1-8 mutant and the corresponding region registered in GenBank was observed. To examine the Sultr1;2 gene, the region from 0.7 kbp upstream of the start codon to 0.2 kbp downstream of the stop codon was amplified and sequenced. A mutation was found in the ninth exon at position +1532 relative to the start codon; the Col-0 wild-type DNA contains a T while sel1-8 has a C. This change in the nucleotide sequence alters codon 511 from coding for Ile in the wild-type to Thr in sel1-8 (Figure 3b). Ile 511 is located in the cytosolic extension of the transporter located between of the 12th membrane spanning domain (MSD) and the STAS domain. The change was also observed in cDNA fragments of Sultr1;2 generated from RNA isolated from the sel1-8 mutant plant; cDNA fragments from selenate-sensitive lines did not have this change.

Mutations were also detected in PCR products containing the Sultr1;2 genes of the selenate-resistant mutants sel1-1 and sel1-3 (Rose, 1997). In sel1-1 the C at position +287 is changed to a T, and in sel1-3 the G at +1526 is changed to an A. The lesion in the sel1-1 mutant changes Ser 96 to Phe and the lesion in sel1-3 changes Gly 509 to Glu (Figure 3b). Interestingly, the lesions in sel1-3 and sel1-8 cause changes at amino acids 509 and 511, respectively. These amino acids are both within the region linking the MSDs and the STAS domain (Aravind and Koonin, 2000).

Since the sel1-9 mutant phenotype co-segregates with a T-DNA insert and is allelic with the other sel1 mutants, we tested whether the sel1-9 mutant contains a T-DNA insertion in the Sultr1;2 gene. We attempted to PCR amplify DNA fragments from sel1-9 genomic DNA using several sets of primers specific for the T-DNA and the Sultr1;2 gene. One set, containing one primer specific for the left border of the T-DNA and the other for the 3′ end of the Sultr1;2 gene, amplified a 4.1 kbp fragment of genomic DNA from sel1-9. The DNA sequence of this fragment determined that the T-DNA integrated into the promoter region of the Sultr1;2 gene, 434 bp upstream of the translation initiation site and 374 bp upstream of the putative transcription initiation site (Figure 3b). DNA gel-blot and PCR analysis showed that the adjacent sulfate transporter gene, Sultr2;2, was not disrupted in sel1-9 (data not shown). Furthermore, Sultr1;2 transcripts could not be detected by either RNA gel blot or RT–PCR in sel1-9, while transcripts of Sultr2;2 were readily detected (Figure 4; Table 2). These results indicate that transcription of the Sultr1;2 gene but not the Sultr2;2 gene is disrupted by the T-DNA insertion in sel1-9.

Figure 4.

Transcript accumulation of the Sultr1;2 gene.

RNA gel-blot analysis of Sultr1;2 and α-tubulin in roots and aerial tissue. Plants were grown for 5 weeks with 0.7 mm sulfate and transferred to 0 mm sulfate at time 0. Total RNA was isolated at time 0 and 72 h. Total RNA (9 µg) was separated by electrophoresis, blotted onto a membrane, and hybridized with probes specific for Sultr1;2 and α-tubulin.

Table 2.  Relative accumulation of Sultr1;2 and Sultr2;2 transcripts
Plant parts Sultr1;2/βTUBSultr2;2/βTUB
Time after –S treatment (h)
72Time after –S treatment (h)
  1. Plants were grown for 4 weeks with 1.5 mm sulfate and transferred to 0 mm sulfate at time 0. Total RNA was isolated at 0 and 72 h, and RT–PCR was performed in SmartCyclerTM. to monitor amplification of each cDNA. All values represent relative accumulation of Sultr1;2 or Sultr2;2 transcripts to that of the β-tubulin (βTUB) in three independent reverse transcriptase reactions followed by PCR (mean ± SD, n = 3).

  2. nd, not detected.

AerialCol-0ndnd4.0 ± 1.82.9 ± 0.8
sel1-8ndnd2.5 ± 0.65.5 ± 1.2
Wsndnd1.4 ± 0.12.7 ± 0.4
sel1-9ndnd1.6 ± 0.55.2 ± 1.0
RootsCol-00.4 ± 0.12.1 ± 1.00.5 ± 0.11.3 ± 0.2
sel1-81.0 ± 0.33.3 ± 1.60.8 ± 0.22.0 ± 0.3
Ws0.6 ± 0.34.8 ± 2.40.8 ± 0.30.9 ± 0.4
sel1-9ndnd0.9 ± 0.32.9 ± 1.0

Functional complementation of a yeast mutant by Sultr1;2 cDNA

The deduced amino acid sequence of Sultr1;2 is 72% identical to Sultr1;1 and 54% identical with Sultr2;1, two genes that have been demonstrated to have sulfate transport activity (Takahashi et al., 2000; Vidmar et al., 2000). Analysis of Sultr1;2 hydrophobicity using MEMSAT 2 (, 1998) predicted 12 putative MSDs. This topology is known to be common in all the eukaryotic proton/sulfate transporters identified to date.

To confirm that the Sultr1;2 gene encodes a functional sulfate transporter, a cDNA clone was used to functionally complement the Saccharomyces cerevisiae methione auxotrophic strain CP154-7B (Matα, his3, leu2, ura3, ade2, trp1, sul1::LEU2, sul2::URA3). This strain carries insertions in the sulfate transporter genes SUL1 and SUL2 (Cherest et al., 1997) and is therefore a methionine auxotroph. A cDNA fragment of the Sultr1;2 gene including start and stop codons was amplified by RT–PCR using total RNA isolated from the roots of the wild-type Col-0 plants as a template. The PCR fragments were cloned into the yeast expression vector pYX222x (a gift from Dr Beom-Seok Seo, Iowa State University, unpublished), generating pYSultr1;2WT. In this plasmid, transcription of the Sultr1;2 cDNA is driven by the S. cerevisiae triose phosphate isomerase promoter. The plasmid also contains the HIS3 gene to serve as a selectable marker for transformation of his3 mutants. pYSultr1;2WT was introduced into CP154-B, and transformed cells were tested for methionine prototrophy. Figure 5 shows that CP154-B cells carrying pYSultr1;2WT can grow on minimal medium lacking methionine and containing 0.5 mm sulfate as the sole source of sulfur. Yeast cells carrying the vector alone or mock-transformed cells (no DNA) were unable to grow on this medium. This demonstrates that the Arabidopsis Sultr1;2 gene encodes a functional sulfate transporter.

Figure 5.

Functional complementation of yeast mutant CP154-7B.

(a) Yeast strain CP154-7B and three transformants, carrying either an empty vector, or the Sultr1;2 cDNA derived from wild type Arabidopsis (Col-0) or from sel1-8 streaked on minimal medium containing 0.5 mm sulfate supplemented with 0.5 mm methionine.

(b) The same cells streaked on minimal medium lacking methionine. pYX222x designates cells containing the empty yeast expression cassette; pYSultr1;2WT designates cells containing the expression cassette with the Sultr1;2 cDNA from wild-type (Col-0), and pYSultr1;2mut cells containing the expression cassette with the Sultr1;2 cDNA from the sel1-8 mutant. ‘No DNA’ indicates where cells that were not transformed were streaked.

To examine if the mis-sense mutation in the Sultr1;2 gene in sel1-8 disrupts the ability of Sultr1;2 protein to transport sulfate, Sultr1;2 cDNAs derived from sel1-8 were cloned into pYX222x to form pYSultr1;2mut and introduced into the S. cerevisiae strain CP154-7B. Unlike transformants carrying the wild-type copy of the Sultr1;2 gene, transformants carrying the mutated Sultr1;2 gene were unable to grow on medium lacking methionine (Figure 5). Ten independent His+ transformants with pYSultr1;2mut were tested and none could grow without methionine in the medium. Sequence analysis of pYSultr1;2mut confirmed that the only difference with the wild-type gene was the mis-sense mutation that alters Ile 511 to Thr in pYSultr1;2mut. These results indicate that the substitution of Thr for Ile at amino acid at position 511 in Sultr1;2 severely disrupts the sulfate transport activity of this protein.

Accumulation of the Sultr1;2 transcripts under sulfur deficiency

As sulfate transport increases when plants are deprived of sulfate, we investigated whether accumulation of the Sultr1;2 transcript increased when plants were moved from sulfur-replete to sulfur-deficient medium. Total RNA was isolated from plants grown in sulfate containing hydroponic medium for 5 weeks and then transferred to medium lacking sulfate for 72 h. RNA was extracted independently from roots and aerial parts of sel1-9 and wild-type (Ws) plants. Transcripts of Sultr1;2 were detected in both non-starved and sulfur-starved roots of wild-type plants. No Sultr1;2 transcript was detected in the aerial parts (Figure 4). Thus the Sultr1;2 gene appears to encode a root-specific transporter that is expressed under both sulfur-sufficient and sulfur-deficient conditions.

Table 2 shows the relative accumulation of the Sultr1;2 and Sultr2;2 transcripts in wild-type and sel1 mutant plants grown in sulfur-replete medium and starved for sulfur for 72 h. RT–PCR was performed and relative rates of amplification monitored in a SmartCycler (Cepheid, Sunnyvale, CA). The Sultr1;2 transcript was five to eight times more abundant in the roots of sulfur-starved than non-starved wild-type plants. Prior to sulfur starvation, the Sultr1;2 transcript is approximately twice as abundant in sel1-8 plants as in wild-type Col-0 plants. The Sultr1;2 transcript is also more abundant in sulfur-starved sel1-8 plants. Accumulation of the Sultr2;2 transcript is greater in the sel1-8 and sel1-9 mutants compared to the corresponding wild-type plants, especially under sulfate starvation, both in aerial parts and roots. These results indicate that both sulfur starvation and mutations in the Sultr1;2 gene cause increased accumulation of transcripts encoding Sultr1;2 and Sultr2;2.

Expression of a Sultr1;2 promoter–GUS fusion gene

Analysis of transgenic plants expressing Sultr1;2 promoter-β-glucuronidase (GUS) fusion constructs indicate that Sultr1;2 is primarily expressed in root tissue. The Sultr1;2::GUS gene was constructed by fusing 1.6 kbp of sequence upstream of the Sultr1;2 start codon to the uidA reporter gene in pCB308, a mini-binary vector plasmid designed for promoter analysis (Xiang et al., 1999). Seeds of transgenic plants were germinated on sulfate-containing solid medium, transferred to sulfate-containing liquid medium for 5 days, and transferred to medium lacking sulfate for 3 days. Plants were stained for GUS activity either before or after sulfur starvation. No significant difference in the staining pattern was observed between the 10 lines analysed.

GUS activity was visualized by staining with 5-bromo-4-chloro-3-indolyl-β-d-glucronic acid (X-gluc) and observed under a microscope. Figure 6a shows that there was abundant GUS activity in cortical tissue of primary roots in the zone of elongation. Root tips of primary and lateral roots (Figure 6b,c) were also stained. In aerial portions, weak GUS staining was observed in hydathodes, guard cells in leaves (Figures 6d,e). Weak staining was also observed in auxiliary buds, anthers and basal parts of flowers (data not shown).

Figure 6.

Tissue-specific expression of Sultr1;2.

(a–d) Seedlings of transgenic plants carrying pSultr1;2proGUS were germinated and grown for 7 days on medium containing 1.6 mm sulfate, then transferred to medium containing 0 mm for 3 days and stained with X-gluc as described in Experimental procedures. Staining in (a )root cortex; (b) in root tip; (c) in lateral root; (d) in guard cells.

(e) Leaves from plants grown for 20 days in soils fed with Hyponex solution show staining in hydathodes.

(f) Seedlings grown for 7 days on sulfate-containing medium then transferred to medium containing 1.6 mm (left) or 0 mm (right) sulfate. 3 days after transfer, seedlings were GUS-stained.

Bars are 25 µm (a, b, d), 0.5 mm (c), 0.5 cm (e) and 1 cm (f).

There was no significant difference in spatial pattern of GUS-staining between plants grown in sulfate-sufficient and sulfate-deficient media. However, the roots of sulfur-starved plants appeared to have more GUS activity (Figure 6f), suggesting that the regulation of transcript accumulation observed by RT–PCR (Table 2) is exerted mainly at the level of transcription.


Sulfate, the primary source of sulfur imported from the environment by vascular plants, is taken up through the roots and distributed throughout the plant. No single transporter is likely to be able to carry out all of the transport processes required. Genomic DNA sequence indicates that Arabidopsis has at least 14 putative sulfate transporter genes. Some of these transporters are synthesized in specific tissues and are probably localized to specific membranes within the cells of these tissues (Takahashi et al., 2000). Our results show that transcripts encoding Sultr1;2 are present in the roots of sulfur-replete grown plants, and increase in abundance when plants are deprived of sulfate (Figure 4; Table 2). Reporter gene studies indicate that Sultr1;2 is expressed primarily in the root cortex and cap where plant cells interface directly with the soil solution. We also detected expression in the guard cells, hydathodes and auxiliary buds of leaves (Figure 6). The significance of expression of Sultr1;2 in aerial tissues is not understood at this time. Similar patterns of gene expression were observed by Yoshimoto et al. (2002) using Sultr1;2–GFP fusion.

We have isolated and characterized selenate-resistant mutants of Arabidopsis defective in the sulfate transporter Sultr1;2. Based on phylogenetic analysis of the amino acid sequence, Sultr1;2 falls into group 1 of the sulfate transporters (Takahashi et al., 2000). Among group 1 sulfate transporters, Shst1, Shst2 (of Stylosanthes hamata), Hvst1 (of Hordeum vulgare) and Sultr1;1 (of Arabidopsis) have been shown to be high-affinity sulfate transporters when expressed in yeast cells (Smith et al., 1995b; Smith et al., 1997; Takahashi et al., 2000). Yoshimoto et al. (2001) measured the kinetics of sulfate transport into S. cerevisiae expressing Sultr1;2 and determined that it is also a high-affinity sulfate transporter with a Km of 6.9 µm.

The sel1 mutants were identified by screening for plants resistant to toxic analogues of sulfate (Figure 1). At least one of these mutants, sel1-9, contains no detectable Sultr1;2 transcript and appears to be completely defective in this sulfate transporter, but this mutant is able to grow to maturity in sulfur-sufficient conditions (1.5 mm sulfate) indicating that the gene is not essential. However, direct measurement of sulfate transport into the roots of this mutant demonstrated that sulfate transport is disrupted (Figure 2a). The sel1-9 mutant, which contains a T-DNA insertion in the promoter region of the Sultr1;2 gene, imports sulfate at approximately half of the rate of wild-type plants. It is unclear whether the defect in sulfate transport in this mutant is compensated for by an increase in activity or expression of other sulfate transporters. Accumulation of the Sultr2;2 transcript was twice as high in the mutant as in the wild type (Table 2), suggesting that increased expression of other sulfate transporters may compensate for the loss of Sultr1;2.

Although sel1-9 is clearly defective in transport of sulfate into the root, there is no difference in the rate of transport of sulfate into the shoots of mutant and wild-type plants (Figure 2b). As most sulfate reduction and assimilation is thought to occur in leaves, plants may have mechanisms of maintaining high sulfate concentrations in the shoot, even when sulfate levels in the root are low. When plants are deprived of sulfur, sulfate content in the shoot decreases less rapidly than in the root (Clarkson et al., 1983; Lappartient and Touraine, 1996), supporting the hypothesis that plants maintain sulfate in the shoot at the expense of sulfate in the root. An alternative explanation for the lack of an effect of the sel1 mutant on shoot sulfate transport is that Sultr1;2 does not play a role in the transport of sulfate to the shoot. There may be separate transport systems for supplying sulfate to the root and shoot, and Sultr1;2 may function solely to provide sulfate to the root and may not participate in loading sulfate into the xylem for transport to the shoot.

The sulfate transporters of plants are proton/sulfate co-transporters, and their amino acid sequences are similar to those of sulfate transporters found in animal cells. Sequence analysis of the 12 of the sulfate transporters of Arabidopsis predicts that they contain 12 MSDs and a carboxy-terminal cytoplasmic STAS domain (Aravind and Koonin, 2000). STAS domains are found in sulfate transporters of both plants and animals, as well as in bacterial antisigma-factor antagonists (ASAs) such as the Bacillus subtilis SPOIIAA protein. ASAs physically interact with antisigma factors to prevent them from binding to sigma factors and inhibiting transcription. SPOIIAA binds GTP and ATP, and possesses a weak NTPase activity (Najafi et al., 1996). As ASAs are known to interact with other proteins, STAS domains may facilitate the interaction of sulfate transporters with other proteins and could regulate sulfate transporter activity.

Two of the three mis-sense mutations identified in this work are located between the twelfth MSD and the STAS domain (Figure 3b). This portion of the protein is conserved among sulfate transporters, but has no similarity with any known functional domain. In sel1-3 Gly 509 is changed to a Glu, and in sel1-8 Ile 511 is a Thr. These positions are highly conserved in the Arabidopsis sulfate transporters. Gly 509 is conserved in all the Arabidopsis sulfate transporters, while Ile 511 is either an Ile or Leu in all the Arabidopsis transporters with the exception of Sultr2;1 where it is a Met. The high level of conservation in this region among plant sulfate transporters and the mutations described here indicate that this domain is critical for function of the sulfate transporter. The mis-sense mutation in sel1-1 changes Ser 96 to a Phe. This region of the protein is in the first MSD (based on MEMSAT2 prediction) that is conserved in sulfate transporters of both plants and animals.

Selenate resistance has provided a useful tool for identifying mutations in the Sultr1;2 gene. Nine of the 10 selenate-resistant mutants of Arabidopsis isolated to date have lesions in this gene. The identification of these mutants, along with their functional characterization, indicates a critical role for this transporter in sulfate uptake into the roots of plants.

Experimental procedures

Plant growth

Plants were grown in a growth chamber on plant-nutrition (PN) medium containing 5.0 mm KNO3, 2.35 mm KH2PO4, 0.15 mm K2HPO4, 2.0 mm Ca(NO3)2, 1.3 mm MgCl2, 70 µm H3BO3, 14 µm MnCl2, 0.5 µm CuCl2, 1 µm ZnCl2, 0.2 µm Na2MoO4, 10 µm NaCl and 0.01 µm CoCl2 (final pH 5.5). Sulfate was provided in the media as MgSO4. MgSO4 was replaced with MgCl2 in PN-S medium. The growth chamber was maintained at 22°C under continuous fluorescent illumination (75 µmol m−2 sec−1). Hydroponic cultures were grown according to Hirai et al. (1995). Soil-grown plants were grown in vermiculite and periodically fertilized with Hyponex (N-P-K = 8-12-6, Hyponex Japan Corp., Ltd, Osaka, Japan).

Screening for selenate-resistant mutants

Ten thousand ethyl methanesulfonate (EMS) mutagenized M2 seeds (Columbia accession) and 100 000 T4 seeds of 9510 T-DNA tagged lines (Wassilewskija accession, kindly provided by ABRC, Ohio State University, Columbus, OH, USA) were surface-sterilized and germinated on PN-S medium containing 20 µm selenate and 40 µm chromate. Petri plates containing PN-S medium solidified with 0.6% agarose were oriented vertically to allow seedlings to grow roots on the surface of the medium. After 2 weeks, root lengths were compared. Seedlings with roots 3–5 mm long were selected for rescreening. Putative mutants were transferred to PN medium lacking selenate and allowed to recover. Plants were then moved to soil and allowed to self-fertilize. Seeds from the putative mutants were rescreened under the same conditions.

Sulfate uptake assay

Surface-sterilized seeds were sown on 300 µm mesh nylon screen embedded vertically in PN media containing 0.6% agarose, 700 µm sulfate and 0.5% sucrose in parafilm sealed Petri dishes. The plates were oriented vertically at 22°C with continuous fluorescent illumination (60 µmol m−2 sec−1). Roots of the seedlings grew through the mesh into the surface of medium. After 14 days the parafilm seals on the Petri dishes were removed. Twenty-four hours later the dish covers were opened to let the plants adjust to ambient room humidity. After 4 hours, the seedlings were transferred to sulfate-replete or sulfate-deficient liquid medium with aeration for 3 days. The liquid media was changed once during this period.

At the start of the sulfate uptake experiments, the plants were supported on the rim of a 5.5 ml plastic tubes containing 5.4 ml of the PN-S medium with 0.6–1.25 µCi ml−1[35S] sulfate. At the end of the 1 h labelling period, the roots were rinsed in 300 ml of the ice-cold PN medium containing 1.5 mm sulfate for 5 sec, then incubated in 1 l of PN medium containing 1.5 mm sulfate at room temperature for 1 h. The PN medium was changed once during the incubation.

The fresh weight of shoots and roots were measured separately and the tissues transferred into scintillation vials containing 2 ml of scintillation cocktail (Safety-Solve, RPI Corp., Mount Prospect, IL, USA), and the incorporated radioactivity was measured in a scintillation counter. The uptake of [35S] sulfate to plants was linear for at least 1 h under these conditions.

Mapping of sel1-8 mutation

M3 plants of the selenate-resistant sel1-8 line were crossed with Landsberg erecta (Ler). The F1 progeny were allowed to self-fertilize to generate a segregating population. F2 seeds were scored for selenate resistance by germinating them on 0.6% agarose-solidified PN-S medium containing 20 µm SeO42– and measuring root length. Resistant plants had longer roots than sensitive plants. The selenate-resistant plants were transferred to agarose-solidified PN media lacking selenate and then to soil. Genomic DNA was isolated from shoots of individual F2 plants as described in Liu et al. (1995) and used as a PCR template in genetic mapping experiments. To confirm the phenotypes of individual F2 plants, F3 seeds from self-fertilized plants were germinated on selenate-containing medium and scored for selenate sensitivity.

All primers for molecular marker-based mapping were purchased from Research Genetics, Inc. (Huntsville, AL, USA), except for SGCSNP142 and SGCSNP253, which were synthesized by the DNA-sequencing facility at Iowa State University (Ames, IA, USA). The single nucleotide polymorphism in SGCSNP142 and SGCSNP253 between Col and Ler was assayed by PCR. For SGCSNP142, the primers SNP142for (5′ aaggtgatgaccgatccaaa 3′) and SNP142rev (5′ ccgatactgaactcgtggct 3′) were used and for SGCSNP253, the primers SNP253for (5′ tgggcgtgaagagttcgtat 3′) and SNP253rev (5′ gattccggagagttccatct 3′) were used. The PCR products for both reactions were 288 bp. They were digested with SspI or ClaI, respectively. The SNP142 fragments derived from Ler but not from Col were cleaved with SspI and the SNP253 fragment derived from Col but not from Ler was cleaved with ClaI.

Identification of T-DNA insertion in sel1-9

Six primers both in forward and reverse orientation were designed in the region containing Sultr1;2 and Sultr2;2. PCR was carried out using these primers in combination with a primer designed in left border (LB) or right border (RB) of T-DNA with sel1-9 genome DNA as template. Among the combination tested, only the following primers amplified a fragment of 4060 bp; LB (5′ tctgggaatggcgtaacaaaggc 3′) and rev3 (5′ agatgtcgacttgaccccttggtgtgat 3′). The amplified fragment was sequenced to determine the insertion site.

Northern hybridization

Total RNA was isolated from hydroponically grown Arabidopsis plants using SDS/phenol as described in Pawlowski et al. (1994). Total RNA (10 µg) was separated in 1% formaldehyde agarose gel and blotted onto Zeta-Probe GT membrane (Bio-Rad, Hercules, CA, USA). RNA was fixed to the membrane by UV irradiation and hybridized according to the manufacturer's recommendations. A genomic clone containing coding sequence of Sultr1;2 was PCR-amplified from Arabidopsis (Col-0) genomic DNA using primers Sultr1;2FB (5′ gtatggatccaacccaaaacgatgat 3′) and Sultr1;2RSal (5′ agatgtcgacttgaccccttggtgtgat 3′). The clone was truncated by digestion with KpnI and the resulting 1.85 kbp fragment containing the 3′-half of the Sultr1;2 gene was used as a gene-specific probe. The α-tublin cDNA was RT–PCR-amplified with α-TUBfor (5′ aaattagggtttctactgagagaag 3′) and α-TUBrev (5′ acgaatattttacaggatttaaaca 3′) and used as a probe.

Expression of Sultr1;2 in yeast

RNA was reverse-transcribed using AMV reverse transcriptase purchased from Gibco BRL (Rockville, MD, USA). The products of this reaction were used as a template for PCR using the Expand· Long Template PCR system (Boehringer Mannheim, Mannheim, Germany) following the manufacturer's instruction.

cDNA clones containing the coding region of the Sultr1;2 gene from the wild-type (Col-0) and sel1-8 mutant strains were amplified by RT–PCR using the primers Sultr1;2FE (5′ atagaattcatgtcgtcaagagctcaccc 3′) and Sultr1;2RXh (5′ actctcgagtcagacctcgttggagaga 3′). The amplified DNA fragments were cloned into the EcoRI–XhoI site of the yeast-expression vector pYX222x, which contains the HIS3 gene which serves as a selectable marker for transformation of his3 mutants, and the triose phosphate isomerase promoter to drive constitutive expression of the inserted gene. The resulting plasmids, pYSultr1;2WT and pYSultr1;2mut were separately transformed into the S. cerevisiae mutant strain CP154-7B (Matα, his3, leu2, ura3, ade2, trp1, su1l::LEU2, sul2::URA3; Cherest et al., 1997) using the lithium acetate transformation protocol (Rose et al., 1990). Cells were incubated at 30°C on synthetic medium containing 0.5 mm sulfate, 20 g l−1 glucose, 0.5 mm methionine and the required amino acids. After 2 days, His+ transformants were detectable. Transformants were tested for methionine auxotrophy on synthetic medium lacking methionine.

Quantification of transcripts by RT–PCR

RNA isolated by RNeasy Plant Mini Kit (Qiagen, Germany) and first-strand cDNA was synthesized using MuLV Reverse Transcriptase (Perkin Elmer, Norwalk, CT, USA) by priming with oligo-d(T)16. The cDNA was amplified by PCR in a SmartCycler (Cepheid, Sunnyvale, CA, USA) with SYBR Green PCR Master Mix (Applied Biosystem, Warrington, UK). The primers used in RT–PCR were: β-TUBfor 5′ gctcgctaatcctacctttgg 3′; β-TUBrev 5′ agccttgggaatgggataag 3′; Sultr1;2for2 (5′ ataggatccattcaacagtatcctgaagcc 3′); Sultr1;2rev2 (5′ atactcgaggaaactgaatcctaggtaggc 3′); Sultr2;2for2 (5′ cgacatgtcttgcgtgatgggcg 3′); and Sultr2;2rev2 (5′ gctcgcttcaatttgtgaagtaccct 3′). The size of amplified fragment (200 bp) was confirmed by gel electrophoresis.

Arabidopsis transformation and GUS staining

A genomic fragment of the Sultr1;2 promoter corresponding nucleotide segment from 1393 bp upstream to 154 bp downstream of the transcription initiation site was PCR-amplified using the primers F (5′ gfgtctagaggctaaaaagcgagatcgaa 3′) and R (5′ tgaggatccagctatgtaactctgcaaac 3′), and cloned into the XbaI–BamHI site of a binary vector pCB308 (Xiang et al., 1999) to generate pSultr1;2proGUS. The Agrobacterium tumefaciens strain GV3101 (Koncz and Schell, 1986) was transformed with pSultr1;2proGUS by electroporation. Arabidopsis plants were transformed using the floral dip method (Clough and Bent, 1998). Transformed T1 plants were selected in soil for herbicide resistance, and seeds of the T2 generation were geminated on agarose-solidified plates prior to staining for GUS activity. Seedlings were vacuum-infiltrated with staining solution containing 100 mm Na2HPO4 pH 7, 0.1% Triton X-100, 2 mm K3Fe[CN]6, 2 mm K4Fe[CN]6 and 0.5 mg ml−1 5-bromo-4-chloro-3-indolyl-β-d-glucronic acid (Wako, Japan) for 20 min followed by overnight incubation at 37°C. Pigments were cleared from GUS-stained seedlings by treatment with 70% and 100% ethanol. Root cross-section was made from root tissues embedded in plastic resin Technovit7100 (Heraeus Kulzer GmbH, Wehrheim, Germany) and microsliced into 10 µm by microtome LR-85 (Yamato-kouki, Japan).


We thank ABRC (Ohio State University, Columbus, OH, USA) for providing seeds of T-DNA tagged lines, Dr Yolande Surdin-Kerjan (Centre National de la Recherche Scientifique, France) for the yeast mutant strain CP154-7B, Dr Beom-Seok Seo (Iowa State University, Ames, IA, USA) for the yeast vector pYX222x, and Dr Chenbin Xiang (Iowa State University) for the binary vector pCB308. Funds to finance this research were graciously provided by grants from the USDA (Grant no. 9900622 to J.P.D.), the Ministry of Education, Culture, Sports, Science and Technology of Japan for the Japanese Junior Scientists (no. 5230 to N.S.), and the Scientific Research on Priority Areas (B), Molecular Mechanisms of Storage Activity in Plants (no. 12138201 to T.F.).