Natural Variation at the MATE Transporter Locus Reveals Cross-Talk between Fe Homeostasis and Zn Tolerance in


Zinc (Zn) is essential for the optimal growth of plants but is toxic if present in excess, so Zn homeostasis needs to be finely tuned. Understanding Zn homeostasis mechanisms in plants will help in the development of innovative approaches for the phytoremediation of Zn-contaminated sites. In this study, Zn tolerance quantitative trait loci (QTL) were identified by analyzing differences in the Bay-0 and Shahdara accessions of Arabidopsis thaliana. Fine-scale mapping showed that a variant of the Fe homeostasis-related FERRIC REDUCTASE DEFECTIVE3 (FRD3) gene, which encodes a multidrug and toxin efflux (MATE) transporter, is responsible for reduced Zn tolerance in A. thaliana. Allelic variation in FRD3 revealed which amino acids are necessary for FRD3 function. In addition, the results of allele-specific expression assays in F1 individuals provide evidence for the existence of at least one putative metal-responsive cis-regulatory element. Our results suggest that FRD3 works as a multimer and is involved in loading Zn into xylem. Cross-homeostasis between Fe and Zn therefore appears to be important for Zn tolerance in A. thaliana with FRD3 acting as an essential regulator.


Published in the journal: . PLoS Genet 8(12): e32767. doi:10.1371/journal.pgen.1003120
Category: Research Article
doi: https://doi.org/10.1371/journal.pgen.1003120

Summary

Zinc (Zn) is essential for the optimal growth of plants but is toxic if present in excess, so Zn homeostasis needs to be finely tuned. Understanding Zn homeostasis mechanisms in plants will help in the development of innovative approaches for the phytoremediation of Zn-contaminated sites. In this study, Zn tolerance quantitative trait loci (QTL) were identified by analyzing differences in the Bay-0 and Shahdara accessions of Arabidopsis thaliana. Fine-scale mapping showed that a variant of the Fe homeostasis-related FERRIC REDUCTASE DEFECTIVE3 (FRD3) gene, which encodes a multidrug and toxin efflux (MATE) transporter, is responsible for reduced Zn tolerance in A. thaliana. Allelic variation in FRD3 revealed which amino acids are necessary for FRD3 function. In addition, the results of allele-specific expression assays in F1 individuals provide evidence for the existence of at least one putative metal-responsive cis-regulatory element. Our results suggest that FRD3 works as a multimer and is involved in loading Zn into xylem. Cross-homeostasis between Fe and Zn therefore appears to be important for Zn tolerance in A. thaliana with FRD3 acting as an essential regulator.

Introduction

Zinc (Zn) is an essential micronutrient for a wide range of cellular functions [1], yet if present in excess, it causes drastic toxicity symptoms resulting in yield reduction and stunted growth [1]. Soil mineral nutrient content varies greatly and plants have adapted to such varied environments in different ways. In highly Zn-contaminated soils only a small range of plants can develop. These Zn-tolerant plants may also hyper-accumulate Zn, like Arabidopsis halleri and Noccaea caerulescens [2]. For phytoremediation, for instance replanting in Zn-contaminated soils, it is crucial to understand the molecular bases of such adaptative processes. A. halleri and N. caerulescens have received extensive attention and various Zn transporters, such as HMA4 and MTP1, and the Zn chelator nicotianamine were shown to be important for Zn tolerance and accumulation [2][6]. However, genetic approaches are not straightforward in these species, mainly because of the lack of available molecular tools, so it may be difficult to find additional genes involved in Zn tolerance in these species.

To identify new genes involved in Zn tolerance, an alternative genetic approach is to benefit from how one species, Arabidopsis thaliana, has adapted to the environment and characterize the genetic factors underlying this natural variation. In A. thaliana, the numerous available natural accessions and RIL populations facilitate traditional linkage analyses and genome wide association studies. In particular QTL mapping analyses have already been successful in identifying genetic factors controlling metal tolerance such as aluminum and copper tolerance [7], [8]. In A. thaliana, the natural variation in Zn accumulation in different organs and growth conditions has been characterized [9][15]. Although no Zn tolerance mechanism has been identified yet, A. thaliana accessions display substantial natural variation in their response to excess Zn [16].

In the presence of an excess of a given mineral element in the medium, from a nutritional point of view, there is some competition between this element and related elements; plants need to coordinate homeostasis to avoid ion imbalances. For instance, in A. thaliana, in the presence of an excess of Zn, plants reduce iron (Fe) accumulation in shoots becoming prone to Fe-deficiency [17]. In A. halleri, a fine regulation of Fe homeostasis contributes to Zn tolerance [17]. Many studies have revealed the importance of the coordination between homeostatic mechanisms for different metals, but the identification of the corresponding molecular components is rare [18].

Here, we describe the mapping of a QTL for the Zn root response which led to the identification of the AtFRD3 gene, which plays a role in Fe nutrition [19], as being responsible for the natural variation in the root response to excess of Zn between the Bay-0 and Shahdara A. thaliana accessions. Results of this study confirmed the importance of the coordination between Fe and Zn homeostasis for the tolerance to Zn excess, with FRD3 being a major regulator of the cross homeostasis.

Results/Discussion

Identification of Zn tolerance QTLs in A. thaliana

To explore the molecular bases of Zn tolerance variation, a QTL analysis was performed in the Bay-0×Sha recombinant inbred line (RIL) population [20] where the trait considered was the relative primary root length (RelPR150) between high Zn (150 µM Zn) and low Zn (1 µM Zn) control conditions (Figure 1A, 1B, Table S1, Figure S1). Three QTLs named ZnT1, ZnT2, and ZnT3 were localized on chromosomes 1, 3 and 5, respectively. No epistasis was observed either between the identified QTLs or between the QTLs and other regions of the genome. The ZnT1 and ZnT2 QTLs were confirmed by analyzing relative primary root length in the progeny of heterogeneous inbred families in which either the Sha or Bay allele was fixed (Figure S2). ZnT2 was the main-effect QTL explaining 29% of the total phenotypic variation (Table S1); the recessive Sha allele (Figure S3) was responsible for the greater reduction in primary root length observed in response to 150 µM Zn compared to the Bay-0 (Bay) allele (Table S1).

Fig. 1. QTL and fine-mapping of the ZnT2 locus.
QTL and fine-mapping of the ZnT2 locus.
(A) Root phenotype of 10-day-old Bay-0 and Sha plants grown on agar plates containing 1 µM or 150 µM Zn. Bay-0 is more Zn-tolerant than Sha. Scale bar, 1 cm. (B) QTL map for the response of the primary root length to Zn (relative primary root length of plants grown at 150 µM and 1 µM Zn), evaluated in a sub-set of 165 RILs from the Bay-0×Sha RIL population. The map was obtained by composite interval mapping and a LOD threshold of 2.5 obtained from permutation. (C) Fine mapping of the ZnT2 QTL. ZnT2, initially mapped between markers NGA172 and MSAT3.19, was localized between MSAT302503 and MSAT2630717 and narrowed down to between CAPS2560983 and CAPS2587730 by recombinant screening. This 23 kb-long region encompasses 7 full-length genes including FRD3. Zn tolerance phenotypes of recombinant plants with informative genotypes are given. Black and grey bars represent Bay and Sha genomes, respectively.

AtFRD3 underlies natural variation in root response to excess of Zn

A mapping population of 2,296 plants derived from a near-isogenic line (NIL) bearing a Sha introgression at the ZnT2 QTL in an otherwise Bay background (NIL[Sha]; Figure S4) was analyzed and ZnT2 was mapped to a 23-kb region spanning 7 genes (Figure 1C). Among these genes, the FERRIC REDUCTASE DEFECTIVE3 (FRD3) gene was considered a good micronutrient-related candidate as it encodes a citrate efflux transporter involved in Fe nutrition [19], [21][23].

To confirm a functional role of FRD3 in the Zn response by genetic complementation, wild-type Col-0 and the frd3-7 knock-out mutant were both crossed separately with Bay-0 and NIL[Sha] plants and the primary root lengths of F1 progeny were measured. The primary root growth of NIL[Sha], inhibited in the presence of 150 µM Zn, was complemented with the FRD3Col allele but not with the frd3-7 mutant allele. The primary root length of Bay-0 was little affected by the presence of either allele. This indicates that the ZnT2 QTL is likely to be allelic to the FRD3 candidate gene (Figure S1, Figure S5). Independently, the dominant FRD3Bay allele was introduced into NIL[Sha] plants by genetic transformation. This resulted in the conversion of the recessive Zn-sensitive phenotype of NIL[Sha] plants into a more Zn-tolerant phenotype (Figure 2). Together, these two approaches demonstrate that FRD3 has a functional role in determining the Zn-responsive ZnT2 trait.

Fig. 2. Functional complementation.
Functional complementation.
Relative primary root lengths are shown for NIL[Sha] plants and for transgenic NIL[Sha] plants expressing either the FRD3Bay or the FRD3Sha allele. Relative primary root length is the primary root length of plants grown at 150 µM Zn as a percentage of the primary root length of plants grown at 1 µM Zn. Error bars represent confidence intervals calculated after logarithmic transformation of data [41] (P<0.05; n = 7 to 21). Similar increased relative root lengths were observed in 2 additional independent NIL[Sha] lines complemented with FRD3Bay (data not shown).

Modifications in AtFRD3 explain variation in gene expression and transport activity

Sequencing the FRD3 gene in the Bay-0 and Shahdara accessions revealed differences due to multiple nucleotide substitutions, deletions and insertions (Figure 3A, Figure S6). The main variations differentiating the FRD3Bay and FRD3Sha alleles were 27-bp and 28-bp indels in the promoter region, two non-synonymous substitutions in exon 2 in the FRD3Sha allele and a 12-bp indel in the last intron. Other differences were a 4-bp indel in the 3′ UTR, ten 1- or 2-bp indels in non-coding regions, 5 synonymous SNPs in coding regions and 48 SNPs in non-coding regions. The non-synonymous substitutions in exon 2 induce N116S and L117P substitutions in the FRD3Sha polypeptide compared to FRD3Bay and FRD3Col polypeptides. Based on the analysis of the five main allelic variations in 109 A. thaliana accessions, five haplotypes were found (Figure S7, Table S2). Interestingly, the 28-bp deletion present in the promoter and the two non-synonymous substitutions in exon 2 were always associated with each other and this haplotype was found in 6% of the 109 accessions (Figure S7). No haplotype was found in which the promoter deletion and non-synonymous substitutions were separate. We hypothesized that the 28-bp deletion present in the promoter and/or the two non-synonymous substitutions present in exon 2, as found in the FRD3Sha allele, are the allelic variations responsible for the Zn-sensitive phenotype. A review of the types of nucleotide polymorphisms that underlie QTLs revealed that such a combination of nucleotide polymorphism in both promoter and coding regions, altering both gene expression and protein function, is not rare [24].

Fig. 3. Natural variation in the FRD3 sequence, FRD3 transport activity, and FRD3 transcript accumulation.
Natural variation in the <i>FRD3</i> sequence, FRD3 transport activity, and <i>FRD3</i> transcript accumulation.
(A) Gene organization and allelic variations identified at the FRD3 locus in Col-0 (reference sequence), Bay-0 and Sha. In non-coding regions, only insertions and deletions are shown. In the coding region, synonymous SNPs are not indicated. (B) 13[C]-citrate efflux activity resulting from the expression of FRD3Bay (Bay), FRD3Col (Col) and FRD3Sha (Sha) proteins or the co-expression of FRD3Sha and FRD3Col (Col+Sha) in Xenopus oocytes. In control oocytes (Ø), no RNA was injected. Data are the means ± S.D.M. for n = 3 sets of 14 oocytes. (C) FRD3 transcript accumulation under Fe deficiency (Fe 0), control conditions (Zn 1) and excess Zn (Zn 200) in parental and NIL lines. FRD3 transcript levels are shown relative to the transcript level of ACT2/ACT8. Values represent the means ± S.D.M. for n = 3 biologically independent experiments. Different letters above bars indicate significantly different values within a treatment set (P<0.05) according to a non-parametric test for Fe0, and Zn200 and a Tukey test for Zn1. The table indicates FRD3 transcript levels for the Fe0 and Zn200 conditions relative to those for the control Zn1 condition.

AtFRD3 is reported to be a citrate efflux transporter [22], so the functionality of the different protein variants was tested using a citrate efflux assay in Xenopus oocytes. The FRD3Bay or FRD3Col proteins were found to mediate citrate efflux, but the N116S and L117P substitutions abrogated this function in the FRD3Sha protein (Figure 3B). The FRD3Bay and FRD3Sha genes are also expressed differently. Expression of FRD3Bay was induced by an excess of Zn and by Fe shortage (FRD3 plays a role in Fe deficient conditions [19], [21][23]), while FRD3Sha was expressed at a markedly lower level under the same conditions (Figure 3C). Genotype-dependent expression of FRD3 under Fe deficiency has already been observed in natural accessions [25]. Thus the FRD3Sha gene, the Zn-sensitive allele, not only encodes a non-functional transporter but is also weakly expressed in the presence of high concentrations of Zn in the medium.

FRD3 belongs to the MATE family, one of the multidrug transporter families encountered in all living organisms [26]. Little is known of how MATE transporters work. When FRD3Col was co-expressed with FRD3Sha in Xenopus oocytes the ability of the FRD3Col protein to efflux citrate was completely lost (Figure 3B). This would suggest first that FRD3Sha interacts with FRD3Col, indicating that the FRD3 transporter functions as a multimer, and second that FRD3Sha is a dominant negative isoform. The discovery that N116 and L117 are crucial for FRD3 function and that FRD3 likely functions as a multimer is thus an important step in understanding the general mode of action of MATE transporters.

The apparent dominant negative effect of FRD3Sha in the citrate efflux assay in Xenopus oocytes may seem inconsistent with the observed recessive character of the Zn-sensitive phenotype associated with the FRD3Sha allele in planta. Compared to the control condition, transcript levels of the FRD3Sha allele are 2.5 to 5 times lower in response to Zn excess, while transcript levels of the FRD3Bay allele are 1.5 to 2 times higher in the same condition (Figure 3C). We therefore favor the hypothesis that in planta, the effect of down regulation of the FRD3Sha transcript predominates over the dominant negative effect of the FRD3Sha protein. On this point, no accession harboring only the non-synonymous substitutions leading to the non-functional FRD3Sha isoform was found after screening 109 A. thaliana accessions. This suggests that these mutations are strongly selected against and that a mutation that reduces transcription is needed to overcome the dominant negative effect of the non-synonymous substitutions.

Towards the identification of a cis-regulatory element that controls FRD3 induction in response to Fe deficiency

FRD3 transcript accumulation was compared in NILs and natural accessions in relation to the haplotype. While FRD3 expression was induced by Fe deficiency and by Zn excess in Bay-0, NIL[Bay] and Col-0, regulation occurred in the opposite sense in Sha and NIL[Sha] (Figure 3C). When correlated to the presence of certain polymorphisms in the gene particularly within the FRD3 promoter, the results point to the likely existence of local Zn- and Fe-responsive, possibly cis-acting, regulatory elements (Figure 3C, Figure S7). The differential induction of FRD3Bay and FRD3Sha transcripts by Fe deficiency was tested through allele-specific expression assays in F1 individuals. The differential induction was maintained in the presence of the contrasting allele indicating that the response is controlled in cis (Figure S8A). The cis-acting differential regulation was confirmed in a cross between Ct-1 and Ita-0 (Figure S8B), where Ct-1 is a Bay-like accession in terms of FRD3 transcriptional induction and sequence (Table S2), and Ita-0 only differs from Ct-1 at the 27bp-polymorphism (Figure S7) and is not transcriptionally responsive to Fe deficiency. Therefore, although we cannot totally exclude the possibility that polymorphisms outside the sequenced region play a role, the indel sequences identified are probably important. None of the known Zn- or Fe-responsive cis-elements from plants is present in the Bay or Col alleles of the FRD3 promoter. The region around the 27-bp sequence in the FRD3 promoter region may therefore represent a new type of metal-responsive cis-regulatory element.

Impact of the FRD3 polymorphisms in Bay-0 and Shahdara on Fe nutrition

FRD3 is known to be involved in Fe homeostasis [19], [21][23]. More precisely, FRD3 releases citrate into the xylem so Fe can be solubilized, transported to the shoots and loaded into leaf cells. Variation at the FRD3 locus might therefore be expected to affect the Fe content of xylem sap. As expected, the Fe concentration in xylem sap of Sha plants and NIL[Sha] plants, which have the non-functional FRD3Sha allele, was much lower than in plants harboring the FRD3Bay-0 allele (Figure 4A). However, unlike frd3 mutants in the Col-0 background [19], [23], [27], they were neither dwarf nor chlorotic, and under control conditions, the shoot and root Fe content was normal. Also no Fe overload was observed in the root stele of plants carrying the FRD3Sha allele (Figure 4B, Figure S9, Figure S10). No indication of Fe deficiency, such as increased mRNA levels or constitutive activity of the root ferric reductase oxidase FRO2, was detected in plants harboring the FRD3Sha allele grown in control conditions, in contrast to what has been observed in frd3Col mutants [19] (Figure S11, Figure S12). Xylem exudates of NIL[Sha] plants did not contain less citrate than Bay-0 plants (Figure S13).

Fig. 4. Fe and Zn homeostasis in NILs and parental lines.
Fe and Zn homeostasis in NILs and parental lines.
Fe (A) and Zn (C) concentrations in xylem exudates. Plants were grown in a growth chamber for 6 weeks and xylem exudates were collected during 30 min after removing aerial parts of the plant at the hypocotyl. Fe (B) and Zn (D) shoot content. Plants were grown on agar plates under control (contr.), 200 µM Zn (+ Zn), and 200 µM Fe plus 200 µM Zn (+ Zn Fe) conditions for 10 days. Percentages given above bars in (B) refer to the difference in shoot Fe content in ‘+ Zn’ conditions as a percentage of the control value. Measurements were performed on sets of 5 to 20 plants. Values are the means ± S.D.M. Different letters above bars refer to significantly different values (P<0.05), according to non-parametric tests for (A), (B) and (D) and Tukey tests for (C) where n = 3–8.

Altogether these observations may suggest that FRD3Sha is a leaky allele of FRD3. Even an inefficient transport activity may be sufficient to avoid the severe phenotypes that are observed with the frd3 knock-out mutation. However, this hypothesis is not in agreement with the oocyte assay data showing that FRD3Sha is not functional. An alternative hypothesis could be that FRD3Sha is a strong mutant allele and that another mechanism is responsible for loading citrate or another Fe-chelating agent into the xylem to compensate for the lack of FRD3-driven citrate delivery in Shahdara and Bay-0. To test the latter hypothesis, F2 plants issuing from a cross between Shahdara and frd3-7 (KO mutant in Col-0 background), were assayed for Fe overaccumulation in the root stele using Perls' stain as a rapid assay of the functionality of FRD3 [21]. A quarter of the F2 population showed Fe overload in the root stele, indicating that this trait was under the control of one recessive locus (Table S3). Importantly, some of the F2 plants showing Fe overaccumulation in the root stele were not homozygous for the frd3 knock-out and two plants showing no Fe overload in the root stele were homozygous for the frd3 knock-out (Table S3). This result means that it is unlikely that FRD3Sha is a leaky allele of FRD3 and that a still uncharacterized mechanism can compensate, at least partially, for the lack of functionality of FRD3, and is thus also involved in the translocation of Fe and Zn ions from the roots to the shoot. This mechanism is present in Shahdara, but it is missing in Col-0, thus explaining why the frd3 mutant phenotype is stronger in the Col-0 background than in the Shahdara background. This compensatory mechanism is also most likely active in the four other accessions that harbor the FRD3Sha haplotype, two of which (Hiroshima and 9481B) being even more tolerant to Zn excess than Bay-0 (Table S2).

Impact of FRD3 on Zn tolerance and homeostasis

We investigated whether FRD3 has a direct impact on Zn homeostasis. The Zn concentration in the xylem exudates of NIL[Sha] plants was lower than that of Bay-0 plants in control conditions (Figure 4C), indicating that FRD3 plays a role in the loading of Zn into the xylem. In addition, in the presence of high concentrations of Zn in the culture medium, shoots of NIL[Sha] plants contained less Zn than Bay-0 shoots (Figure 4D). This clearly indicated that FRD3 is involved in the translocation of Zn from the roots to the shoot in A. thaliana. The mechanism by which FRD3 is involved in Zn tolerance in roots is in contrast more difficult to infer from the data. A positive correlation between translocation of Zn from root to shoots and Zn tolerance in roots has already been established from the analysis of heavy metal atpase 4 (hma4) mutant lines and HMA4-overexpressing lines [28][30]. The interpretation was that reducing Zn translocation from the roots to the shoot resulted in an increase in the Zn content in roots that would be the cause of an increased sensitivity to Zn in roots. This interpretation may not be of great help to interpret our data. Whatever the Zn concentration in the medium, the presence of the FRD3Sha allele does not induce any increase in the Zn content in roots compared to the FRD3Bay allele (Figure S9). It is possible that although the Zn content is similar in roots of plants harboring the FRD3Sha or FRD3Bay allele the Zn distribution is different within the roots and that this difference would results in a difference in the root sensitivity to Zn. An alternative hypothesis could however be that the impact of FRD3 on Zn tolerance results from an impairment in Fe homeostasis.

We therefore investigated the phenotypic relationship between Zn and Fe homeostasis. High concentrations of Zn in the medium induced a decrease in the shoot Fe content (Figure 4B), mimicking Fe limiting conditions (Figure 3C, Figure S11). Vice versa increasing the Fe concentration in the medium resulted in a decrease in shoot Zn content (Figure 4D). Similar data have already been reported [17]. In response to the Zn constraint, the shoot Fe content in plants harboring the FRD3Sha allele was more reduced than in plants harboring the FRD3Bay-0 allele (Figure 4B). Also, in the presence of excess Zn, more FRO2 (Ferric Reduction Oxidase) transcripts were expressed in NIL[Sha] plants than in Bay-0 plants (Figure S11), indicating that FRD3Sha confers sensitivity to Zn-induced Fe deficiency. Therefore we have shown that FRD3 acts at an intersection between Fe and Zn nutrition.

As shown above, FRD3 controls the loading of both Zn and Fe into the xylem and these two metals appear to compete for root-to-shoot transport. This conclusion leads to the hypothesis that the primary reason for which FRD3 has an impact on Zn tolerance would be that excess Zn impairs Fe nutrition and that FRD3 is an important control point in the Zn-Fe relationship. This is in agreement with, and may explain, recent observations on how the cross-homeostasis between Fe and Zn deals with excess Zn in A. halleri and A. thaliana [17]. In particular, exposing A. halleri plants to high Zn concentrations induced neither a marked alteration in Fe root-to-shoot transport nor Fe deficiency [17], which can be related to the fact that FRD3 expression is 45 times higher in A. halleri than in A. thaliana [31], and thus would not be a limiting factor in the transport of either Zn or Fe from the roots to the shoot.

In conclusion, we show that the cross-homeostasis between Fe and Zn is important in the A. thaliana response to excess Zn and that FRD3 is an essential regulator of this cross homeostasis. This new perspective on FRD3, including the potential multimeric topology of the FRD3 transporter and the presence of an Fe deficiency-responsive element in the promoter, provides clear directions for further study of how FRD3 contributes to regulating plant micronutrient status.

Materials and Methods

Plant materials and growth conditions

QTL were identified from 165 Bay-0×Shahdara RIL lines [20] and validated using the HIF004, HIF044 and HIF338 lines available from INRA Versailles Genomic Resource Centre (http://dbsgap.versailles.inra.fr/vnat/). NIL[Bay] and NIL[Sha] were obtained from RIL112 and RIL070 following three successive back-crosses with the Shahdara and Bay-0 parental lines, respectively. After each backcrossing step, lines were selected with 38 microsatellite markers [20]. At the end of the process, NIL[Bay] plants had a Bay allele at marker ATHCHIB2 and Sha alleles at the other 37 markers. The reverse was true for NIL[Sha] plants. Seeds of frd3-7 plants were kindly provided by C. Curie (BPMP, Montpellier, France).

Zn tolerance phenotypes were determined as previously described [16]. Plants were grown in the presence of 1 µM ZnSO4 (control condition) and in the presence of 150 µM ZnSO4 (high Zn condition). For each genotype the primary root length was measured in both conditions to obtain the relative primary root length (RelPR150 = (PR150/PR1)×100%). For Fe-response assays, plants were grown on control medium for 7 days and transferred to either Fe-deficient (Fe 0) or Fe-sufficient (Fe 50) medium for 4 days before collecting roots for RNA preparation or ferric reductase assays. The Fe-deficient medium was control medium without NaFeEDTA but with 300 µM ferrozine.

For the sampling of xylem exudates and Perls staining, plants were grown in compost for 6 or 3 weeks respectively in a growth chamber (20°C, 180 µmol.m−2.s−1 and an 8-h light/16-h dark photoperiod). Xylem exudates were collected by removing rosettes with a scalpel then placing a glass capillary tube on the root after discarding the first droplet exuded. After 30 min or 2 h of sap collection, xylem exudates were placed on ice then stored frozen at −20°C.

Citrate, Fe, and Zn measurements

The citrate content of xylem exudates (50 µl to 100 µl from the 2-h collection) was analyzed by high performance ionic chromatography (LC20, Ionex) using an IonPac AS11 column and a 1 mM to 22 mM NaOH gradient.

The Zn and Fe contents of xylem exudates (5 µl from the 30-min collection) were estimated by atomic absorption spectrophotometry using graphite tube atomizers GTA220 (Varian) with omega platform tubes for Zn and partitioned tubes for Fe. Zn concentration in plant tissues was assessed as previously described [16] and Fe concentration in plant tissues was measured by the absorbance of Fe2+-o-phenanthroline at 510 nM [32].

QTL analysis and fine-mapping of ZnT2

Core-Pop165 plants from the Bay-0×Shahdara RIL population [20] were grown on agar plates as previously described [16]. Primary root lengths were determined for ten plants from each of control (1 µM ZnSO4; PR1) and excess Zn (150 µM ZnSO4; PR150) agar plates. Relative primary root length (RelPR150 = (PR150/PR1)×100%) was analyzed using QTLCartographer (http://statgen.ncsu.edu/qtlcart/). Composite interval mapping (CIM) was performed using model 6 and the LOD significance threshold was obtained from permutation analyses. The percentage of variance explained by each QTL and its predicted allelic effect were obtained from QTLCartographer.

Fine-mapping populations were obtained by crossing NIL[Sha] plants with Bay-0 plants, producing an F1 population which was self-pollinated to produce an F2 population. DNA was extracted from plants (grown on soil in the greenhouse) by freeze-drying and grinding cotyledons in 300 µl buffer (100 mM Tris HCl, 1.5 M NaCl, 20 mM EDTA, 2% (w/v) mixed alkyltrimethyl ammonium bromide (Sigma), 0.5% (w/v) sodium sulfite, 1% (v/v) PEG6000). After chloroform extraction, DNA was precipitated using isopropanol, dried then dissolved in 50 µl water. Fine-mapping was done in two steps; first 624 plants were screened for recombination between MSAT302422 and CAPS7012599, then 1,672 plants were screened for recombination between MSAT302503 and MSAT2630717. Markers used for genotyping are described by INRA Versailles Genomic Resource Centre (http://dbsgap.versailles.inra.fr/vnat/) or in Table S4, for those newly identified here.

FRD3 sequencing

FRD3 alleles were amplified from genomic DNA of Bay-0 and Shahdara using KlenTaq LA DNA Polymerase Mix (#D5062, Sigma Aldrich) with primers promo1F, promo2F, exon2R, intron4F, exon12R and post3 (see Table S5 for primer sequences) and three independent PCR products were sequenced.

Transgenic plants

The genomic sequences of FRD3 including the promoter were amplified from Bay-0 and Shahdara genomic DNA (Table S5), cloned in the pGEM-T vector, verified by resequencing, then cloned in the pGREEN0179 vector [33]. Binary recombinant vectors were introduced into Agrobacterium tumefaciens strain GV3101, which was used to transform NIL[Sha] plants with either FRD3Bay or FRD3Sha by the floral dip method [34]. Transformants were selected on MS/2 agar plates with 50 mg L−1 hygromycin. Single T-DNA insertion and homozygous lines were successively selected during segregation analysis.

Quantitative gene expression

RNA extraction, cDNA preparation and real-time quantitative RT–PCR were done as previously described [16]. Primers used are listed in Table S5. Three independent biological experiments were done to analyze transcript levels, once relative to ACT2/ACT8 [35] transcript levels and once relative to transcript levels of ACT2/ACT8 [35], clathrin [36] (At4g24550), At5g12240 [37] and PP2A [37] (At1g13320). Similar results were observed in the two technical replicates.

Allele-specific expression assays

Two pairs of accessions were analyzed using different SNPs, Bay-0 versus Shahdara and Ct-1 versus Ita-0. Parents and F1 individuals (from reciprocal crosses) were grown in vitro as described above and transferred onto Fe-deficient media. RNA was extracted from roots and cDNA prepared as described above. Pyrosequencing reactions were set up around SNPs in the parental coding sequence of FRD3 to assess the relative contribution of each allele to the mRNA population of mRNA [38]. Pyrosequencing was performed on F1 cDNA, on 1∶1 mixtures of parental cDNA, and on F1 genomic DNA as a control to normalize the ratios against possible pyrosequencing biases. Anything significantly driving allele-specific expression in hybrids is, by definition, acting in cis as F1 nuclei contain a mix of all trans factors [39]. In the Bay/Sha experiment, SNP1 and SNP2 interrogate ACA[A/G]GA[T/C]TGG with primers PyroSNP1-2_F and PyroSNP1-2_R-biotin for the PCR and PyroSNP1-2_Seq for the pyrosequencing reaction. In the Ct-1/Ita-0 experiment, SNP3 interrogates AA[C/T]GAT with primers PyroSNP3_F-biotin, PyroSNP3_R and PyroSNP3_Seq; and SNP4 interrogates TC[G/A]TTA with primers PyroSNP4_F, PyroSNP4_R-biotin and PyroSNP4_Seq (Table S5).

Citrate efflux experiment

The predicted cDNAs of the different FRD3 alleles (Figure 4, Figure S6) were obtained from FRD3Col cDNA (G14324; pENTR223_FRD3Col) by site-directed mutagenesis using the QuikChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA). We obtained FRD3Bay cDNA by introducing a single base modification in exon 9 of FRD3Col cDNA and FRD3Sha cDNA by introducing the two SNPs in exon2 of FRD3Bay cDNA (see Table S5 for primer sequences). All mutated cDNAs were sequenced to verify the mutation(s) and the complete sequence. These cDNAs were transferred from pENTR to the pGEM-GWC vector by recombination using LR-clonase (Life Technologies, Grand Island, USA). The different FRD3 cDNA clones were linearized with PstI then transcribed using mMessage mMachine T7® Ultra Kit (Life Technologies). Xenopus laevis oocytes (CRBM, CNRS, Montpellier, France) were prepared as previously described [40] and injected with 20 ng of RNA coding for FRD3Col, FRD3Bay or FRD3Sha using a micropipette (10–15 µM tip diameter) and a pneumatic injector. Three days after injection, 14 oocytes (either injected with FRD3 cRNAs or uninjected) were placed in modified Ringer solution (in which HEPES was replaced by 1 mM Tris) supplemented with 10 mM citrate. Each batch of oocytes was injected with 25 nl of 100 mM 13[C]-citrate (Sigma). After 5 min of recovery, oocytes were washed five times in 15 ml of cold modified Ringer solution (pH 6.5) and placed in 1 ml of modified Ringer buffer for efflux measurement. After 20 min, 100 µl of efflux buffer was sampled in three replicates and the 13[C] abundance (atom %) was analyzed by continuous-flow mass spectrometry using the Euro-EA Eurovector elemental analyzer coupled to an IsoPrime mass spectrometer (GV Instruments).

Statistical analysis

Once the normality of residues had been tested, either one-way ANOVA and Tukey tests (for parametric comparison of means) or Kruskal-Wallis test followed by a non-parametric comparison of means were used. All these tests used an α-value of 0.05 and were done with the R software (using shapiro, bartlett, aov, TukeyHSD, kruskal and nparcomp functions of R; R Development Core Team). Two-way ANOVA was used to test the interaction between genotypes in F1 progeny using the aov function of R software. To test differences in relative primary root growth values, confidence intervals (α-value of 0.05) were calculated after ln transformation of data [41].

Accession numbers

The genomic DNA sequences of FRD3 reported in this paper have been deposited in the EMBL Nucleotide Sequence Database: HE803766 (Bay-0) and HE803767 (Shahdara).

Supporting Information

Attachment 1

Attachment 2

Attachment 3

Attachment 4

Attachment 5

Attachment 6

Attachment 7

Attachment 8

Attachment 9

Attachment 10

Attachment 11

Attachment 12

Attachment 13

Attachment 14

Attachment 15

Attachment 16

Attachment 17

Attachment 18


Zdroje

1. BroadleyMR, WhitePJ, HammondJP, ZelkoI, LuxA (2007) Zinc in plants. New Phytol 73: 677–702.

2. KramerU (2010) Metal hyperaccumulation in plants. Annu Rev Plant Biol 61: 517–34.

3. HanikenneM, NouetC (2011) Metal hyperaccumulation and hypertolerance: a model for plant evolutionary genomics. Curr Opin Plant Biol 14: 252–259.

4. MarquèsL, OomenR (2011) On the way to unravel zinc hyperaccumulation in plants: a mini review. Metallomics 3: 1265–1270.

5. DeinleinU, WeberM, SchmidtH, RenschS, TrampczynskA, et al. (2012) Elevated nicotianamine levels in Arabidopsis halleri roots play a key role in zinc hyperaccumulation. Plant Cell 24: 708–723.

6. WillemsG, DragerDB, CourbotM, GodeC, VerbruggenN, et al. (2007) The genetic basis of zinc tolerance in the metallophyte Arabidopsis halleri ssp. halleri (Brassicaceae): an analysis of quantitative trait loci. Genetics 176: 659–674.

7. RyanPR, TyermanSD, SasakiT, FuruichiT, YamamotoY, et al. (2011) The identification of aluminium-resistance genes provides opportunities for enhancing crop production on acid soils. J Exp Bot 62: 9–20.

8. KobayashiY, KurodaK, KimuraK, Southron-FrancisJL, FuruzawaA, et al. (2008) Amino acid polymorphisms in strictly conserved domains of a P-type ATPase HMA5 are involved in the mechanism of copper tolerance variation in Arabidopsis. Plant Physiol 148: 969–980.

9. BuescherE, AchbergerT, AmusanI, GianniniA, OchsenfeldC, et al. (2010) Natural genetic variation in selected populations of Arabidopsis thaliana is associated with ionomic differences. PLoS ONE 5: e11081 doi:10.1371/journal.pone.0011081.

10. BaxterI, HermansC, LahnerB, YakubovaE, TikhonovaM, et al. (2012) Biodiversity of mineral nutrient and trace element accumulation in Arabidopsis thaliana. PLoS ONE 7: e35121 doi:10.1371/journal.pone.0035121.

11. GhandilyanA, BarbozaL, TisnéS, GranierC, ReymondM, et al. (2009a) Genetic analysis identifies quantitative trait loci controlling rosette mineral concentrations in Arabidopsis thaliana under drought. New Phytol 184: 180–192.

12. GhandilyanA, IlkN, HanhartC, MbengueM, BarbozaL, et al. (2009b) A strong effect of growth medium and organ type on the identification of QTLs for phytate and mineral concentrations in three Arabidopsis thaliana RIL populations. J Exp Bot 60: 1409–1425.

13. VreugdenhilD, AartsMGM, KoornneefM, NelissenH, ErnstWHO (2004) Natural variation and QTL analysis for cationic mineral content in seeds of Arabidopsis thaliana. Plant Cell Environ 27: 828–839.

14. WatersBM, GrusakMA (2008) Quantitative trait locus mapping for seed mineral concentrations in two Arabidopsis thaliana recombinant inbred populations. New Phytol 179: 1033–1047.

15. PrinzenbergAE, BarbierH, SaltDE, StichB, ReymondM (2010) Relationships between growth, growth response to nutrient supply, and ion content using a recombinant inbred line population in Arabidopsis. Plant Physiol 154: 1361–1371.

16. RichardO, PineauC, LoubetS, ChaliesC, VileD, et al. (2011) Diversity analysis of the response to Zn within the Arabidopsis thaliana species revealed a low contribution of Zn translocation to Zn tolerance and a new role for Zn in lateral root development. Plant Cell Environ 34: 1065–1078.

17. ShanmugamV, LoJC, WuCL, WangSL, LaiCC, et al. (2011) Differential expression and regulation of iron-regulated metal transporters in Arabidopsis halleri and Arabidopsis thaliana – the role in zinc tolerance. New Phytol 190: 125–137.

18. ShanmugamV, TsedneeM, YehKC (2012) ZINC TOLERANCE INDUCED BY IRON1 reveals the importance of glutathione in the cross-homeostasis between zinc and iron in Arabidopsis thaliana. Plant J 69: 1006–1017.

19. RogersEE, GuerinotML (2002) FRD3, a member of the multidrug and toxin efflux family, controls iron deficiency responses in Arabidopsis. Plant Cell 14: 1787–1799.

20. LoudetO, ChaillouS, CamilleriC, BouchezD, Daniel-VedeleF (2002) Bay-0×Shahdara recombinant inbred line population: a powerful tool for the genetic dissection of complex traits in Arabidopsis. Theor Appl Genet 104: 1173–1184.

21. GreenLS, RogersEE (2004) FRD3 controls iron localization in Arabidopsis. Plant Physiol 136: 2523–2531.

22. DurrettTP, GassmannW, RogersEE (2007) The FRD3-mediated efflux of citrate into the root vasculature is necessary for efficient iron translocation. Plant Physiol 144: 197–205.

23. RoschzttardtzH, Séguéla-ArnaudM, BriatJF, VertG, CurieC (2011) The FRD3 citrate effluxer promotes iron nutrition between symplastically disconnected tissues throughout Arabidopsis development. Plant Cell 23: 2725–2737.

24. Alonso-BlancoC, AartsMGM, BentsinkL, KeurentkesJJB, ReymondM, et al. (2009) What has natural variation taught us about plant development, physiology, and adaptation? Plant Cell 21: 1877–1896.

25. SteinRJ, WatersBM (2012) Use of natural variation reveals core genes in the transcriptome of iron-deficient Arabidopsis thaliana roots. J Exp Bot 63: 1039–1055.

26. OmoteH, HiasaM, MatsumotoT, OtsukaM, MoriyamaY (2006) The MATE proteins as fundamental transporters of metabolic and xenobiotic organic cations. Trends Pharmacol Sci 27: 587–593.

27. DelaizeE (1996) A metal-accumulator mutant of Arabidopsis thaliana. Plant Physiol 111: 849–855.

28. HanikenneM, TalkeIN, HaydonMJ, LanzC, NolteA, et al. (2008) Evolution of metal hyperaccumulation required cis-regulatory changes and triplication of HMA4. Nature 453: 391–395.

29. VerretF, GravotA, AuroyP, LeonhardtN, DavidP, et al. (2004) Overexpression of AtHMA4 enhances root-to-shoot translocation of zinc and cadmium and plant metal tolerance. FEBS Letters 576: 306–312.

30. HussainD, HaydonMJ, WangY, WongE, ShersonSM, et al. (2004) P-type ATPase heavy metal transporters with roles in essential zinc homeostasis in Arabidopsis. Plant Cell 16: 1327–1339.

31. TalkeIN, HanikenneM, KrämerU (2006) Zinc-dependent global transcriptional control, transcriptional deregulation, and higher gene copy number for genes in metal homeostasis of the hyperaccumulator Arabidopsis halleri. Plant Physiol 142: 148–167.

32. LobreauxS, BriatJF (1991) Ferritin accumulation and degradation in different organs of pea (Pisum sativum) during development. Biochem J 274: 601–606.

33. HellensRP, EdwardsEA, LeylandNR, BeanS, MullineauxPM (2000) pGreen: a versatile and flexible binary Ti vector for Agrobacterium-mediated plant transformation. Plant Mol Biol 42: 819–832.

34. CloughSJ, BentAF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16: 735–743.

35. CharrierB, ChampionA, HenryY, KreisM (2002) Expression profiling of the whole Arabidopsis shaggy-like kinase multigene family by real-time reverse transcriptase-polymerase chain reaction. Plant Physiol 130: 577–590.

36. SéguélaM, BriatJF, VertG, CurieC (2008) Cytokinins negatively regulate the root iron uptake machinery in Arabidopsis through a growth-dependent pathway. Plant J 55: 289–300.

37. CzechowskiT, StittM, AltmannT, UdvardiMK, ScheibleWR (2005) Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiol 139: 5–17.

38. WittkoppPJ, HaerumBK, ClarkAG (2004) Evolutionary changes in cis and trans gene regulation. Nature 430: 85–88.

39. ZhangX, RichardsEJ, BorevitzJO (2007) Genetic and epigenetic dissection of cis regulatory variation. Curr Opin Plant Biol 10: 142–148.

40. KroukG, LacombeB, BielachA, Perrine-WalkerF, MalinskaK (2010) Nitrate-regulated auxin transport by NRT1.1 defines a mechanism for nutrient sensing in plants. Dev Cell 18: 927–937.

41. HedgesLV, GurevitchJ, CurtisPS (1999) The meta-analysis of response ratios in experimental ecology. Ecology 80: 1150–1156.

Štítky
Genetika Reprodukční medicína

Článek vyšel v časopise

PLOS Genetics


2012 Číslo 12
Nejčtenější tento týden
Nejčtenější v tomto čísle
Kurzy

Zvyšte si kvalifikaci online z pohodlí domova

Současné pohledy na riziko v parodontologii
nový kurz
Autoři: MUDr. Ladislav Korábek, CSc., MBA

Svět praktické medicíny 3/2024 (znalostní test z časopisu)

Kardiologické projevy hypereozinofilií
Autoři: prof. MUDr. Petr Němec, Ph.D.

Střevní příprava před kolonoskopií
Autoři: MUDr. Klára Kmochová, Ph.D.

Aktuální možnosti diagnostiky a léčby litiáz
Autoři: MUDr. Tomáš Ürge, PhD.

Všechny kurzy
Kurzy Podcasty Doporučená témata Časopisy
Přihlášení
Zapomenuté heslo

Zadejte e-mailovou adresu, se kterou jste vytvářel(a) účet, budou Vám na ni zaslány informace k nastavení nového hesla.

Přihlášení

Nemáte účet?  Registrujte se