Article pubs.acs.org/est
Gene Expression Differences between Noccaea caerulescens Ecotypes Help to Identify Candidate Genes for Metal Phytoremediation Pauliina Halimaa,*,‡ Ya-Fen Lin,⊥ Viivi H. Ahonen,‡ Daniel Blande,‡ Stephan Clemens,§ Attila Gyenesei,† Elina Haï kiö,# Sirpa O. Kar̈ enlampi,‡ Asta Laiho,† Mark G. M. Aarts,⊥ Juha-Pekka Pursiheimo,† Henk Schat,¶ Holger Schmidt,§ Marjo H. Tuomainen,‡ and Arja I. Tervahauta‡ ‡
Department of Biology, University of Eastern Finland, P.O. Box 1627, Kuopio, 70210, Finland Turku Centre for Biotechnology, University of Turku and Åbo Akademi University, Tykistökatu 6, Turku, 20520, Finland ⊥ Laboratory of Genetics, Wageningen University, P.O. Box 309, Wageningen, 6700 AH, The Netherlands § Department of Plant Physiology, University of Bayreuth, Universitaetsstrasse 30, Bayreuth, D-95440, Germany # Department of Environmental Science, University of Eastern Finland, P.O. Box 1627, Kuopio, 70210, Finland ¶ Department of Genetics, Vrije Universiteit Amsterdam, De Boelelaan 1085, Amsterdam, 1081 HV, Netherlands †
S Supporting Information *
ABSTRACT: Populations of Noccaea caerulescens show tremendous differences in their capacity to hyperaccumulate and hypertolerate metals. To explore the differences that could contribute to these traits, we undertook SOLiD highthroughput sequencing of the root transcriptomes of three phenotypically well-characterized N. caerulescens accessions, i.e., Ganges, La Calamine, and Monte Prinzera. Genes with possible contribution to zinc, cadmium, and nickel hyperaccumulation and hypertolerance were predicted. The most significant differences between the accessions were related to metal ion (di-, trivalent inorganic cation) transmembrane transporter activity, iron and calcium ion binding, (inorganic) anion transmembrane transporter activity, and antioxidant activity. Analysis of correlation between the expression profile of each gene and the metal-related characteristics of the accessions disclosed both previously characterized (HMA4, HMA3) and new candidate genes (e.g., for nickel IRT1, ZIP10, and PDF2.3) as possible contributors to the hyperaccumulation/tolerance phenotype. A number of unknown Noccaea-specific transcripts also showed correlation with Zn2+, Cd2+, or Ni2+ hyperaccumulation/tolerance. This study shows that N. caerulescens populations have evolved great diversity in the expression of metal-related genes, facilitating adaptation to various metalliferous soils. The information will be helpful in the development of improved plants for metal phytoremediation.
■
INTRODUCTION
Filling the knowledge gaps would facilitate the improvement of plants for metal phytoextraction. The cruciferous plant Noccaea caerulescens (previously Thlaspi caerulescens) has been used as a model to study the mechanisms of metal hyperaccumulation and hypertolerance.12 Compared to the majority of metal hyperaccumulators, N. caerulescens is unusual for its ability to accumulate several metals in the shoots at very high concentrations.13 Populations of N. caerulescens show tremendous intraspecific differences in their capacity to tolerate and accumulate metals. At the phenotypic level, this is rather well documented.13−15 However, the mechanisms by which one population accumulates more Ni2+,
Over 450 plant species have been characterized as metal hyperaccumulators. These plants open up the possibility for plant-assisted extraction (phytoextraction) of metals from contaminated or metal-rich soils. Of the naturally hyperaccumulating plants, progress in developing commercial technology for Ni2+ phytoextraction has been made using Alyssum murale.1 For Cd2+ phytoextraction, only Noccaea caerulescens from southern France has shown the ability to phytoextract useful amounts of Cd2+.2 However, high-biomassyielding crops would be central for the commercial success of phytoextraction technology. This creates a need to develop genetically improved hyperaccumulator plants. As the understanding of metal hyperaccumulation and hypertolerance mechanisms remains fragmentary, there is a great need to gain an overview of the metabolism of the hyperaccumulator plants with such extraordinary properties. © 2014 American Chemical Society
Received: Revised: Accepted: Published: 3344
September 26, 2013 January 14, 2014 February 21, 2014 February 21, 2014 dx.doi.org/10.1021/es4042995 | Environ. Sci. Technol. 2014, 48, 3344−3353
Environmental Science & Technology
Article
Figure 1. Data analysis workflow. Top row: data analysis for SOLiD transcriptomes. Second row: analysis of N. caerulescens isotigs obtained from Roche GS-FLX sequencing. Bottom: functional analysis. DAVID, The Database for Annotation, Visualization and Integrated Discovery.
and another one more Cd2+ or Zn2+, while probably having the same metal transporters with limited metal specificity, are poorly known. Most of the existing data are related to Zn2+, and to a lesser extent to Cd2+ accumulation or tolerance, but the knowledge about Ni2+ is generally lacking. The objective of this study was to uncover differences in the transcriptomes of three N. caerulescens accessions with pronounced differences and well-defined phenotypes in metal tolerance and accumulation, i.e., Ganges (GA), La Calamine (LC), and Monte Prinzera (MP) (Assunçaõ et al.;15 Table S1, Figure S1, Supporting Information). Accessions GA and LC translocate both Cd2+ and Zn2+ efficiently from the roots to the shoots, but GA is superior in accumulating Cd2+, as LC is poor in taking up the metals from the soil. Both accessions hypertolerate Cd2+ and Zn2+. Accession MP shows excellent Ni2+ hyperaccumulation and hypertolerance capacity, besides an intermediate tolerance and high accumulation of Zn2+, and a low capacity to tolerate and accumulate Cd2+. Several genes are proposed to contribute to metal accumulation and tolerance in the processes of uptake in the root, detoxification, vacuolar sequestration, translocation to the shoots (xylem loading/ unloading), homeostasis of other nutrients, and stress protection (reviewed by, e.g., Verbruggen et al.; Hassan and Aarts; Na and Salt; Rascio and Navari-Izzo).16−19 In this paper, we focus on the root processes. As a number of genes that have been proposed to contribute to metal hyperaccumulation are constitutively more highly expressed in the hyperaccumulators than in the related nonhyperaccumulators,4,5,16,20,21 we hypothesized that also the N. caerulescens accessions show differences in the expression of genes accounting for metal accumulation and tolerance traits, even when grown at rather low metal concentrations. A great number of candidate genes have been revealed by us and others with nontargeted techniques such as cross-species microarray,3−6Noccaea-specific microarray,7 comparative EST analysis,8 differential display (DD),9 and proteomics.10,11 All these methods have limitations that can be at least partially overcome with new sequencing technologies. Here, we used high-throughput sequencing (RNA-seq) as a tool to identify differences in the gene expression between the closely related N. caerulescens accessions. The massive-scale short-tag RNA sequencing overcomes several limitations of array-based analysis, such as reliance on probe design and limits of sensitivity and dynamic range. Our data provide the most comprehensive sequence resource available today for N.
caerulescens. We show here that the N. caerulescens accessions differ significantly in the expression of a number of genes that potentially contribute to metal hyperaccumulation. Furthermore, we found 1109 transcripts that are specific for N. caerulescens; i.e., no orthologs were found in the genome of A. thaliana or of any other species. The expression pattern of a number of those genes correlated with Zn2+, Cd2+, or Ni2+ hyperaccumulation or hypertolerance. The findings are highly relevant for the understanding of Zn2+, Cd2+, and Ni2+ hyperaccumulation/tolerance mechanisms, as they establish a basis for targeted studies.
■
MATERIALS AND METHODS Full descriptions are available in Supporting Information Materials and Methods. Plant Material. The seeds of three inbred N. caerulescens accessions, GA, LC, and MP (Table S1, Supporting Information), were germinated in soil, and the plants were transferred at eight- to ten-leaf stage into 10 L containers filled with a half-strength Hoagland nutrient solution (modified from Schat et al.22). ZnSO4 was increased to 10 μM in LC and GA, and 10 μM NiSO4 was added to MP, as these accessions are known to suffer from low Zn2+ and Ni2+, respectively. The LC accession consistently develops chlorosis after a couple of weeks at 2 μM Zn2+ but remains healthy at 10 μM Zn2+. The MP accession performs clearly better with Ni2+ addition but does not need extra Zn2+, which it would primarily hyperaccumulate instead of Ni2+. The MP accession primarily hyperaccumulates Ni2+ in nature, because the soil at Monte Prinzera is extremely rich in Ni2+ but very poor in Zn2+ (about 40 ppm); MP is hypersensitive to Cd2+ (0.5 μM is already toxic), which may be due to the fact that the soil at Monte Prinzera does not contain Cd2+. This experimental arrangement guarantees healthy plants of all three accessions. The plants were grown in three climate chambers under the following conditions: 20/15 °C day/night, 250 μmol/m−2/s, 75% RH, and light period 14 h/day. Continuously aerated solutions were changed twice a week. After three weeks, the roots of twelve plants of uniform appearance were pooled from each chamber to obtain three independent biological replicates, frozen in liquid N2 and stored at −80 °C. cDNA Library Preparation and SOLiD Sequencing. Nine cDNA libraries were prepared: three biological replicates for three N. caerulescens accessions. Total RNA was extracted with RNeasy Plant Mini Kit (Qiagen). Barcoded libraries were 3345
dx.doi.org/10.1021/es4042995 | Environ. Sci. Technol. 2014, 48, 3344−3353
Environmental Science & Technology
Article
functional categorization tool at TAIR, using the biological process categories (Figure S2, Supporting Information). The differentially expressed gene IDs were uploaded to the DAVID Functional Annotation Tool.30,31 The Functional Annotation Chart was used to obtain GO terms enriched in the differentially expressed genes. The Chart was filtered for GO terms related to molecular function and at a false discovery rate of 5% (Figure 3). In order to search the functionality for the remaining isotigs, they were translated into protein sequences using OrfPredictor.32 These predicted protein sequences were annotated using Blast2GO,33−36 but most remained unknown (Table S2, Supporting Information). The Blast2GO run included blastp and interproscan search. Parameters set for Blast2GO were an e-value of 0.001 for blastp searches and an HSP length cutoff of 20. Validation of Gene Expression Data. The expression levels of three genes were analyzed by quantitative real-time reverse transcriptase (RT) PCR using LightCycler 480 SYBR Green I Master Mix (Roche). RNA was extracted with RNeasy Plant Mini Kit (Qiagen) and reverse transcribed using DyNAmo cDNA Synthesis Kit (Finnzymes) with oligo(dT)15. The HMA3 primers were as in Ueno et al.,37 except that the forward primer was adjusted to match with all accessions: HMA3 F 5′-AAA GCT GGA GAA AGT ATA CCG ATC-3′. The HMA4 primers were qPCR_HMA4_1F and qPCR_HMA4_1R.38 The NRAMP1 primers were: NRAMP1 F 5′-GCT CTA ATG GTG GCC TTT CTC-3′ and NRAMP1 R 5′-CGA AGC CTT GTT TAA GTC CAA A-3′. Actin 1 was used as the reference gene.39 Correlation Analysis. A model was built upon the known characteristics of the three N. caerulescens accessions (Assunçaõ et al.;15 Table S1, Supporting Information). Codes −1, 0, and 1 represent low, middle, and high Zn 2+ , Cd 2+ , or Ni 2+ accumulation or tolerance, respectively, when ranking the accessions. For each gene, the expression profiles, i.e., the normalized read counts, from each of the nine samples were compared to the models to see how well the expression pattern for each gene correlated to a particular model. Correlation was calculated using the Spearman rank correlation, which does not presume a linear difference between the different model categories but takes into account only the relative order. For each correlation, the statistical significance (p-value) and correlation estimate were recorded. In addition, p-values adjusted for multiple testing using Benjamini-Hochberg false discovery rate (FDR) were calculated.27 All analyses were performed with the R version 2.15.1.40 Low-Molecular-Weight Chelators and Elements. Citric acid and malic acid were analyzed with GC-MS according to the metabolite profiling protocol described by Erban et al.41 Quantification of Fmoc-derivatized nicotianamine was performed via UPLC-ESI-QTOF-MS, and stable isotope dilution analysis was described by Schmidt et al.42 Elements were analyzed with plasma emission spectrometer (ICP-OES, IRIS Intrepid ll XSP) (data presented in Figure S3, Supporting Information).
prepared according to SOLiD 4 system library preparation guidelines (Applied Biosystems). The samples were sequenced on the SOLiD 4 system (Applied Biosystems) on one flow cell with SOLiD 4 chemistry for 50 and 35 bp reads. Total read numbers recovered from different samples were LC: 21 531 999; 26 617 588; 33 283 677; GA: 31 884 261; 19 807 681; 26 240 893; MP: 35 863 030; 31 147 922; 22 250 294. The raw sequence reads were processed according to Figure 1. Alignment of Reads to Arabidopsis thaliana Reference Genome. The reads were aligned against A. thaliana reference genome available at TAIR (version 9; Arabidopsis.org) using the standard whole transcriptome pipeline and the color space alignment tool provided by Applied Biosystems (bioscope v1.3.1). Only the uniquely aligned reads were used for downstream analysis. The alignment was started by locating short matches between the read and reference sequence and was run using the default parameters. For 50- and 35-base reads, the default seed is 25 bases long, with up to two mismatches allowed. For 50-base reads the seed is anchored at positions 0 and 20, for 35-base reads at positions 0 and 10. Extension only occurred when a read passed the initial seeding phase. After aligning the reads to A. thaliana genome, they were associated with the corresponding mRNAs. Data Normalization. R/Bioconductor package DESeq23 was used to normalize the samples for library size. There was extremely high correlation between the replicate samples (0.996 to 0.999). Using Euclidean metrics (range −1 to 1), the samples clustered according to the accessions. To be able to compare the expression levels of different genes with each other, the read count values were divided with the median exonic length of all isoforms of each gene as known from A. thaliana. Differentially Expressed Genes. Downstream analysis of the reads focused on genes differentially expressed among the three N. caerulescens accessions. Gene expression data were analyzed using R statistical software version 2.11 24 with libraries from Bioconductor project version 2.6.25 In order to detect differentially expressed genes, the normalized data were analyzed using linear model fitting-based statistical testing with empirical Bayes variance smoothing procedure contained in the Linear Models for Microarray Data (limma) R-package.26 Pairwise comparisons were made between the accessions (LCGA, LC-MP, and GA-MP) using a moderated t test (limma). Multiple testing adjustments were performed for the obtained p-values using the Benjamini and Hochberg procedure.27 To identify differentially expressed genes not found from A. thaliana, the reads that did not align to the A. thaliana genome were subsequently aligned to a set of N. caerulescens isotigs, i.e., transcript variants generated from contigs of assembled reads (Y. -F. Lin et al., in prep.). Read alignment was conducted by using SHRiMP2 v2.2.2.28 The parameters set for SHRiMP2 were read orientation “opp-in” and -all-contigs and -single-bestmapping flags. Properly paired reads were counted using bioconductor GenomicRanges.29 Pairwise comparisons between each accession were made using the DEseq package. First, isotigs that had an A. thaliana ortholog were removed from the analysis. Read counts were normalized using the estimateSizeFactors function and dispersion estimated using the estimateDispersions function. Tests were conducted between each accession using exactTest. An adjusted p-value of 0.05 was used as the level for significance. Functional Annotation. The differentially expressed genes were first categorized by their GO terms with the GO
■
RESULTS AND DISCUSSION RNA-Seq Analysis of Noccaea Root Transcriptomes. Limited information is available about the mechanisms responsible for the distinct differences in the metal hyperaccumulation and hypertolerance capacities of N. caerulescens populations. We therefore undertook an RNA-seq analysis of three phenotypically well-characterized N. caerulescens acces3346
dx.doi.org/10.1021/es4042995 | Environ. Sci. Technol. 2014, 48, 3344−3353
Environmental Science & Technology
Article
Figure 2. Expression of HMA4, HMA3, and NRAMP1 verified by real-time PCR quantification. Top row: quantitative PCR. Actin was used for normalizing gene expression between samples. Bottom row: expression levels determined by SOLiD transcriptome sequencing.
Figure 3. The enriched molecular function GO terms were searched among the differentially expressed genes using DAVID functional annotation tool.30 An FDR p-value 0.8. Genes predicted to be linked to the metal-related traits are listed in Table S6, Supporting Information, and a selection is given in Table 1. Expression levels of some of the genes are shown in Figure S6, Supporting Information. In the following sections, we will discuss the molecular mechanisms of Zn2+, Cd2+, and Ni2+ accumulation and tolerance in N. caerulescens in the light of our gene expression data and current literature. Genes Associated with Ni Hypertolerance or Hyperaccumulation. The only candidate previously proposed, based on yeast complementation, to mediate the uptake of Ni2+, as a Ni2+−NA complex, is NcYSL3.43 Apparently Ni2+ is mostly chelated by histidine in N. caerulescens roots, thus inhibiting vacuolar sequestration.44 The genes predicted by our model to be associated with Ni2+ accumulation included, e.g., those encoding putative metal transporters IRT1, ZIP10, IREG2, PDR6, transcription factor ARR11, and an unannotated isotig 05798. Many genes were associated with Ni2+ tolerance, e.g., those encoding the metal transporters IRT2, ALMT12, and PCR1, the defensins PDF2.1, PDF2.2, and PDF2.3, and the isotigs 00203, 17695, and 05123. The expression pattern observed here strongly suggests that NcIRT1 contributes to Ni2+ uptake and accumulation in MP accession. AtIRT1 has previously been shown to have low substrate specificity, transporting primarily iron but also Zn2+, Cd2+, and Ni2+.45,46 Small differences in transporter sequence can alter the substrate specificity,45 and it remains to be shown if IRT1 in MP accession favors Ni2+ over other metals. The same holds for ZIP10, which was previously identified as a candidate in the transport of Zn2+ or other metals across the membranes in A. halleri. 47 Our data showed a strong correlation between Ni2+ accumulation and IREG2 transcript levels. AtIREG2 confers Ni2+ tolerance in yeast, and A. thaliana T-DNA insertion lines lacking IREG2 expression are sensitive to Ni2+, especially under Fe deficiency.48 The iron-regulated AtIREG2 has been proposed to have a role in the sequestration of excess Ni2+ into the root vacuole to counterbalance Ni2+ uptake by low-specificity transporters like IRT1.48 This could be the role of IREG2 also in MP accession especially at high Ni2+ concentrations, in which increasingly higher fraction of Ni2+ is retained in the roots. To conclude, the strongest candidates for Ni2+ accumulation in N. caerulescens based on our results are IRT1 and ZIP10, and the possible dependence on their metal specificity on protein structure is of high interest. 3350
dx.doi.org/10.1021/es4042995 | Environ. Sci. Technol. 2014, 48, 3344−3353
Environmental Science & Technology
Article
accumulation. High expression may be needed because Cd2+ competes with Mn2+ for uptake. Another candidate gene, ZIP9, had its highest expression in GA and is up-regulated in N. caerulescens both by Cd2+ exposure and Zn2+ deficiency.71 The apparent link between Zn2+ and Cd2+ suggests that the uptake of Zn2+ by a nonspecific transporter, active also in Cd2+ transport, initiates the need to sequester Cd2+ in the vacuole. In conclusion, major differences in the expression of genes were found in the three N. caerulescens accessions with an approach that is more far-reaching than any other transcriptome analysis reported so far on metal hyperaccumulators. New candidates were predicted for Zn2+, Cd2+, and Ni2+ tolerance and accumulation, some of them specific to N. caerulescens. Still, the limitation remains that some of these genes will not contribute to metal hyperaccumulation but are related to other phenotypic differences between the N. caerulescens accessions. For a number of genes, the possibility also exists that they are involved in the tolerance or accumulation of more than one metal. These need to be explored in the future. Detailed knowledge of the genes conferring metal hyperaccumulation traits will provide tools for breeding plant varieties for metal phytoremediation.
hyperaccumulation. In our metabolite analyses, citric acid concentration (Figure S7, Supporting Information) was shown to be significantly higher in LC and GA than in MP, linking citric acid to Zn2+ and Cd2+ tolerance. This is in agreement with the expression profile of citrate synthase 5. Genes Associated with Cd Hyperaccumulation. The genes predicted by our model to be associated with Cd2+ hyperaccumulation included those encoding the metal transporters ZIP6, ACA8, MTP2, MTP8, FRO8, the phosphate transporter PHT1;9 as well as the unannotated isotigs 02262, 08052, and 02715. Cadmium has been proposed to use Ca2+ and Zn2+ and iron transporters, e.g., members of ZIP and NRAMP families to enter the cells.61 ZIP6 is located in the stele62,63 and could thus be involved in loading apoplastic Cd2+ across the plasma membrane. Supporting this, the expression of TcZNT6 (AtZIP6 ortholog) from the LC accession enhanced Cd sensitivity of A. thaliana, apparently due to the lack of Cd2+ detoxification mechanisms. No phenotype was observed in ZIP6 knockout mutant.64 As a member of the cation efflux protein family, MTP8 could have a role in Cd2+ efflux across the plasma membrane. In A. thaliana, the gene is expressed in all root cell layers but is highest in procambium.62,63 Our model predicted Ca2+-transporting P-type ATPase ACA8 as a novel candidate for Cd2+ accumulation. As Ca2+transporting P-type ATPases are generally considered to be involved in cation efflux from the cytosol,65 it is possible that ACA8, together with HMA4, moves Cd2+ to the xylem. However, it is also possible that ACA8 functions in Ca2+ signaling during metal hyperaccumulation. Most Cd2+ is chelated in the root cytoplasm, but our data do not indicate major differences among the N. caerulescens accessions in the expression of genes for the biosynthesis of potential Cd2+ detoxification ligands, like metallothioneins, glutathione, and phytochelatins. Some of the Cd2+ can be transported into the vacuole (reviewed by Verbruggen et al.).66 The HMA3 transporter is a major player in vacuolar Cd2+ sequestration.37 The more than 10-fold higher level of HMA3 transcripts in GA compared to LC and MP, reported here, might explain the lower shoot-to-root Cd2+ concentration ratio in GA compared to LC, previously observed by Assunçaõ et al.15 It might also explain the differences in Cd2+ tolerance between the accessions; extremely high Cd2+ uptake is tolerated by GA but not by MP (Assuncao et al.).15 In LC, Cd2+ tolerance seems to derive from lower uptake and efficient translocation to the shoots, perhaps by HMA4. On the basis of our data, NRAMP6 is a candidate for effluxing Cd2+ from vacuoles, with 5-fold expression in GA compared to LC and MP. Also Cailliatte et al. proposed that NRAMP6 affects Cd2+ distribution/availability within the cell, as A. thaliana overexpressing NRAMP6 was Cd2+ hypersensitive, and the null allele is more Cd2+ tolerant, whereas Cd2+ concentration remained unchanged.67 The high expression of NRAMP1 in GA suggests a role in Cd2+ transport. AtNRAMP1 is present in all root layers in A. thaliana and is primarily considered as high-affinity manganese transporter,68 but the metal specificity of NcNRAMP1 is not known. The elemental analysis showed that Mn2+ concentration in GA is ca. 10-fold compared to that in LC and MP (Figure S3, Supporting Information), supporting the role for NcNRAMP as Mn2+ transporter. In rice, however, NRAMP1 and NRAMP5 were recently reported as major proteins in Cd2+ accumulation and uptake, respectively.69,70 We conclude that NcNRAMP1 is also a good candidate for Cd2+ uptake or
■
ASSOCIATED CONTENT
S Supporting Information *
Detailed materials and methods. Supporting figures: characteristics of the accessions, elemental analysis, GO-analysis, Domain Database search for isotigs, Venn diagrams, examples of differentially expressed genes, and metabolite analysis. Supporting tables: Blast2GO analysis, differentially expressed genes, GO-analysis, metal transporter expression among accessions, correlation analysis, and examples of correlation analysis. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: pauliina.halimaa@uef.fi. Tel.: +358 40 3553835. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was financially supported by the Academy of Finland (projects No. 122338, 260552), COST FA0905, and EnSTe Graduate School. We thank Kuopio Campus Research Garden for maintaining and growing the plant material. We acknowledge CSC − IT Center for Science Ltd. for the allocation of computational resources. The RNA-Seq data are deposited at the NCBI Gene Expression Omnibus as GSE35900. The transcriptome shotgun assembly data are deposited at DDBJ/ EMBL/GenBank under the accession GASZ00000000.
■
REFERENCES
(1) Li, Y.; Chaney, R.; Brewer, E.; Roseberg, R.; Angle, J. S.; Baker, A.; Reeves, R.; Nelkin, J. Development of a technology for commercial phytoextraction of nickel: Economic and technical considerations. Plant Soil 2003, 249 (1), 107−115. (2) Chaney, R. L.; Angle, J. S.; Broadhurst, C. L.; Peters, C. A.; Tappero, R. V.; Sparks, D. L. Improved understanding of hyperaccumulation yields commercial phytoextraction and phytomining technologies. J. Environ. Qual. 2007, 36 (5), 1429−1443.
3351
dx.doi.org/10.1021/es4042995 | Environ. Sci. Technol. 2014, 48, 3344−3353
Environmental Science & Technology
Article
(3) Hammond, J. P.; Bowen, H. C.; White, P. J.; Mills, V.; Pyke, K. A.; Baker, A. J. M.; Whiting, S. N.; May, S. T.; Broadley, M. R. A comparison of the Thlaspi caerulescens and Thlaspi arvense shoot transcriptomes. New Phytol. 2006, 170 (2), 239−260. (4) Weber, M.; Harada, E.; Vess, C.; Roepenack-lahaye, E. V.; Clemens, S. Comparative microarray analysis of Arabidopsis thaliana and Arabidopsis halleri roots identifies nicotianamine synthase, a ZIP transporter and other genes as potential metal hyperaccumulation factors. Plant J. 2004, 37 (2), 269−281. (5) Becher, M.; Talke, I. N.; Krall, L.; Krämer, U. Cross-species microarray transcript profiling reveals high constitutive expression of metal homeostasis genes in shoots of the zinc hyperaccumulator Arabidopsis halleri. Plant J. 2004, 37 (2), 251−268. (6) van de Mortel, J. E.; Villanueva, L. A.; Schat, H.; Kwekkeboom, J.; Coughlan, S.; Moerland, P. D.; Ver Loren Van Themaat, E.; Koornneef, M.; Aarts, M. G. M. Large expression differences in genes for iron and zinc homeostasis, stress response, and lignin biosynthesis distinguish roots of Arabidopsis thaliana and the related metal hyperaccumulator Thlaspi caerulescens. Plant Physiol. 2006, 142 (3), 1127−1147. (7) Plessl, M.; Rigola, D.; Hassinen, V. H.; Tervahauta, A.; Kärenlampi, S.; Schat, H.; Aarts, M. G. M.; Ernst, D. Comparison of two ecotypes of the metal hyperaccumulator Thlaspi caerulescens (J. & C. PRESL) at the transcriptional level. Protoplasma 2010, 239 (1−4), 81−93. (8) Rigola, D.; Fiers, M.; Vurro, E.; Aarts, M. G. M. The heavy metal hyperaccumulator Thlaspi caerulescens expresses many species-specific genes, as identified by comparative expressed sequence tag analysis. New Phytol. 2006, 170 (4), 753−766. (9) Hassinen, V. H.; Tervahauta, A. I.; Halimaa, P.; Plessl, M.; Peräniemi, S.; Schat, H.; Aarts, M. G. M.; Servomaa, K.; Kärenlampi, S. O. Isolation of Zn-responsive genes from two accessions of the hyperaccumulator plant Thlaspi caerulescens. Planta 2007, 225 (4), 977−989. (10) Tuomainen, M. H.; Nunan, N.; Lehesranta, S. J.; Tervahauta, A. I.; Hassinen, V. H.; Schat, H.; Koistinen, K. M.; Auriola, S.; McNicol, J.; Kärenlampi, S. O. Multivariate analysis of protein profiles of metal hyperaccumulator Thlaspi caerulescens accessions. Proteomics 2006, 6 (12), 3696−3706. (11) Tuomainen, M.; Tervahauta, A.; Hassinen, V.; Schat, H.; Koistinen, K. M.; Lehesranta, S.; Rantalainen, K.; Häyrinen, J.; Auriola, S.; Anttonen, M.; Kärenlampi, S. Proteomics of Thlaspi caerulescens accessions and an inter-accession cross segregating for zinc accumulation. J. Exp. Bot. 2010, 61 (4), 1075−1087. (12) Milner, M. J.; Kochian, L. V. Investigating heavy-metal hyperaccumulation using Thlaspi caerulescens as a model system. Ann. Bot. 2008, 102 (1), 3−13. (13) Roosens, N.; Verbruggen, N.; Meerts, P.; Ximénez-Embrún, P.; Smith, J. Natural variation in cadmium tolerance and its relationship to metal hyperaccumulation for seven populations of Thlaspi caerulescens from western Europe. Plant Cell Environ. 2003, 26 (10), 1657−1672. (14) Zhao, F.; Hamon, R. E.; Lombi, E.; McLaughlin, M. J.; McGrath, S. P. Characteristics of cadmium uptake in two contrasting ecotypes of the hyperaccumulator Thlaspi caerulescens. J. Exp. Bot. 2002, 53 (368), 535−543. (15) Assunçaõ , A. G.; Bookum, W. M.; Nelissen, H. J.; Vooijs, R.; Schat, H.; Ernst, W. H. Differential metal-specific tolerance and accumulation patterns among Thlaspi caerulescens populations originating from different soil types. New Phytol. 2003, 159 (2), 411−419. (16) Verbruggen, N.; Hermans, C.; Schat, H. Molecular mechanisms of metal hyperaccumulation in plants. New Phytol. 2009, 181 (4), 759−776. (17) Hassan, Z.; Aarts, M. G. M. Opportunities and feasibilities for biotechnological improvement of Zn, Cd or Ni tolerance and accumulation in plants. Environ. Exp. Bot. 2011, 72 (1), 53−63. (18) Na, G.; Salt, D. E. The role of sulfur assimilation and sulfurcontaining compounds in trace element homeostasis in plants. Environ. Exp. Bot. 2011, 72 (1), 18−25.
(19) Rascio, N.; Navari-Izzo, F. Heavy metal hyperaccumulating plants: How and why do they do it? And what makes them so interesting? Plant Sci. 2011, 180 (2), 169−181. (20) van de Mortel, J. E.; Aarts, M. G. M. Comparative transcriptomics - Model species lead the way. New Phytol. 2006, 170 (2), 199−201. (21) Roelofs, D.; Aarts, M. G. M.; Schat, H.; Van Straalen, N. M. Functional ecological genomics to demonstrate general and specific responses to abiotic stress. Funct. Ecol. 2008, 22 (1), 8−18. (22) Schat, H.; Vooijs, R.; Kuiper, E. Identical major gene loci for heavy metal tolerances that have independently evolved in different local populations and subspecies of Silene vulgaris. Evolution 1996, 50, 1888−1895. (23) Anders, S.; Huber, W. Differential expression analysis for sequence count data. Genome Biol. 2010, 11 (10), R106. (24) R Development Core Team. R: A language and environment for statistical computing; R Foundation for Statistical Computing: Vienna, Austria, 2009. (25) Gentleman, R. C.; Carey, V. J.; Bates, D. M.; Bolstad, B.; Dettling, M.; Dudoit, S.; Ellis, B.; Gautier, L.; Ge, Y.; Gentry, J. Bioconductor: Open software development for computational biology and bioinformatics. Genome Biol. 2004, 5 (10), R80. (26) Smyth, G. K. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat. Appl. Genet. Mol. Biol. 2004, 3 (1), 3. (27) Benjamini, Y.; Hochberg, Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 1995, 57, 289−300. (28) David, M.; Dzamba, M.; Lister, D.; Ilie, L.; Brudno, M. SHRiMP2: Sensitive yet practical short read mapping. Bioinformatics 2011, 27 (7), 1011−1012. (29) Aboyoun, P.; Pages, H.; Lawrence, M. GenomicRanges: Representation and manipulation of genomic intervals. R package version 1.9.28; Bioconductor: 2012, http://www.bioconductor.org/. (30) Huang, D. W.; Sherman, B. T.; Lempicki, R. A. Bioinformatics enrichment tools: Paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 2009, 37 (1), 1−13. (31) Huang, D. W.; Sherman, B. T.; Lempicki, R. A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc. 2009, 4 (1), 44−57. (32) Min, X. J.; Butler, G.; Storms, R.; Tsang, A. OrfPredictor: Predicting protein-coding regions in EST-derived sequences. Nucleic Acids Res. 2005, 33 (suppl 2), W677−W680. (33) Conesa, A.; Götz, S.; García-Gómez, J. M.; Terol, J.; Talón, M.; Robles, M. Blast2GO: A universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 2005, 21 (18), 3674−3676. (34) Conesa, A.; Götz, S. Blast2GO: A comprehensive suite for functional analysis in plant genomics. Int. J. Plant Genomics 2008, DOI: 10.1155/2008/619832. (35) Götz, S.; García-Gómez, J. M.; Terol, J.; Williams, T. D.; Nagaraj, S. H.; Nueda, M. J.; Robles, M.; Talón, M.; Dopazo, J.; Conesa, A. High-throughput functional annotation and data mining with the Blast2GO suite. Nucleic Acids Res. 2008, 36 (10), 3420−3435. (36) Götz, S.; Arnold, R.; Sebastián-León, P.; Martín-Rodríguez, S.; Tischler, P.; Jehl, M.; Dopazo, J.; Rattei, T.; Conesa, A. B2G-FAR, a species-centered GO annotation repository. Bioinformatics 2011, 27 (7), 919−924. (37) Ueno, D.; Milner, M. J.; Yamaji, N.; Yokosho, K.; Koyama, E.; Clemencia Zambrano, M.; Kaskie, M.; Ebbs, S.; Kochian, L. V.; Ma, J. F. Elevated expression of TcHMA3 plays a key role in the extreme Cd tolerance in a Cd-hyperaccumulating ecotype of Thlaspi caerulescens. Plant J. 2011, 66 (5), 852−862. (38) Craciun, A. R.; Meyer, C.; Chen, J.; Roosens, N.; De Groodt, R.; Hilson, P.; Verbruggen, N. Variation in HMA4 gene copy number and expression among Noccaea caerulescens populations presenting different levels of Cd tolerance and accumulation. J. Exp. Bot. 2012, 63 (11), 4179−4189. 3352
dx.doi.org/10.1021/es4042995 | Environ. Sci. Technol. 2014, 48, 3344−3353
Environmental Science & Technology
Article
(39) Xing, J.; Jiang, R.; Ueno, D.; Ma, J.; Schat, H.; McGrath, S.; Zhao, F. Variation in root-to-shoot translocation of cadmium and zinc among different accessions of the hyperaccumulators Thlaspi caerulescens and Thlaspi praecox. New Phytol. 2008, 178 (2), 315−325. (40) R Development Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2007. (41) Erban, A.; Schauer, N.; Fernie, A. R.; Kopka, J. Nonsupervised construction and application of mass spectral and retention time index libraries from time-of-flight gas chromatography-mass spectrometry metabolite profiles. Metabolomics 2007, 358, 19−38. (42) Schmidt, H.; Böttcher, C.; Trampczynska, A.; Clemens, S. Use of recombinantly produced 15N3-labelled nicotianamine for fast and sensitive stable isotope dilution ultra-performance liquid chromatography/electrospray ionization time-of-flight mass spectrometry. Anal. Bioanal.Chem. 2011, 399 (3), 1355−1361. (43) Gendre, D.; Czernic, P.; Conéjéro, G.; Pianelli, K.; Briat, J.; Lebrun, M.; Mari, S. TcYSL3, a member of the YSL gene family from the hyper-accumulator Thlaspi caerulescens, encodes a nicotianamineNi/Fe transporter. Plant J. 2007, 49 (1), 1−15. (44) Richau, K. H.; Kozhevnikova, A. D.; Seregin, I. V.; Vooijs, R.; Koevoets, P. L.; Smith, J. A. C.; Ivanov, V. B.; Schat, H. Chelation by histidine inhibits the vacuolar sequestration of nickel in roots of the hyperaccumulator Thlaspi caerulescens. New Phytol. 2009, 183 (1), 106−116. (45) Rogers, E. E.; Eide, D. J.; Guerinot, M. L. Altered selectivity in an Arabidopsis metal transporter. Proc. Natl. Acad. Sci. U. S. A. 2000, 97 (22), 12356−12360. (46) Nishida, S.; Tsuzuki, C.; Kato, A.; Aisu, A.; Yoshida, J.; Mizuno, T. AtIRT1, the primary iron uptake transporter in the root, mediates excess nickel accumulation in Arabidopsis thaliana. Plant Cell Physiol. 2011, 52 (8), 1433−1442. (47) Talke, I. N.; Hanikenne, M.; Krämer, U. Zinc-dependent global transcriptional control, transcriptional deregulation, and higher gene copy number for genes in metal homeostasis of the hyperaccumulator Arabidopsis halleri. Plant Physiol. 2006, 142 (1), 148−167. (48) Schaaf, G.; Honsbein, A.; Meda, A. R.; Kirchner, S.; Wipf, D.; von Wirén, N. AtIREG2 encodes a tonoplast transport protein involved in iron-dependent nickel detoxification in Arabidopsis thaliana roots. J. Biol. Chem. 2006, 281 (35), 25532−25540. (49) Montargès-Pelletier, E.; Chardot, V.; Echevarria, G.; Michot, L. J.; Bauer, A.; Morel, J. Identification of nickel chelators in three hyperaccumulating plants: An X-ray spectroscopic study. Phytochemistry 2008, 69 (8), 1695−1709. (50) Agrawal, B.; Lakshmanan, V.; Kaushik, S.; Bais, H. P. Natural variation among Arabidopsis accessions reveals malic acid as a key mediator of Nickel (Ni) tolerance. Planta 2012, 236 (2), 477−489. (51) Meyer, S.; Mumm, P.; Imes, D.; Endler, A.; Weder, B.; AlRasheid, K. A.; Geiger, D.; Marten, I.; Martinoia, E.; Hedrich, R. AtALMT12 represents an R-type anion channel required for stomatal movement in Arabidopsis guard cells. Plant J. 2010, 63 (6), 1054− 1062. (52) Feng, W.; Benz, F. W.; Cai, J.; Pierce, W. M.; Kang, Y. J. Metallothionein disulfides are present in metallothionein-overexpressing transgenic mouse heart and increase under conditions of oxidative stress. J. Biol. Chem. 2006, 281 (2), 681−687. (53) Kerkeb, L.; Krämer, U. The role of free histidine in xylem loading of nickel in Alyssum lesbiacum and Brassica juncea. Plant Physiol. 2003, 131 (2), 716−724. (54) Dräg er, D. B.; Desbrosses-Fonrouge, A.; Krach, C.; Chardonnens, A. N.; Meyer, R. C.; Saumitou-Laprade, P.; Krämer, U. Two genes encoding Arabidopsis halleri MTP1 metal transport proteins co-segregate with zinc tolerance and account for high MTP1 transcript levels. Plant J. 2004, 39 (3), 425−439. (55) Gustin, J. L.; Loureiro, M. E.; Kim, D.; Na, G.; Tikhonova, M.; Salt, D. E. MTP1-dependent Zn sequestration into shoot vacuoles suggests dual roles in Zn tolerance and accumulation in Znhyperaccumulating plants. Plant J. 2009, 57 (6), 1116−1127.
(56) Milner, M. J.; Craft, E.; Yamaji, N.; Koyama, E.; Ma, J. F.; Kochian, L. V. Characterization of the high affinity Zn transporter from Noccaea caerulescens, NcZNT1, and dissection of its promoter for its role in Zn uptake and hyperaccumulation. New Phytol. 2012, 195 (1), 113−123. (57) Courbot, M.; Willems, G.; Motte, P.; Arvidsson, S.; Roosens, N.; Saumitou-Laprade, P.; Verbruggen, N. A major quantitative trait locus for cadmium tolerance in Arabidopsis halleri colocalizes with HMA4, a gene encoding a heavy metal ATPase. Plant Physiol. 2007, 144 (2), 1052−1065. (58) Hanikenne, M.; Talke, I. N.; Haydon, M. J.; Lanz, C.; Nolte, A.; Motte, P.; Kroymann, J.; Weigel, D.; Krämer, U. Evolution of metal hyperaccumulation required cis-regulatory changes and triplication of HMA4. Nature 2008, 453 (7193), 391−395. (59) Wojas, S.; Hennig, J.; Plaza, S.; Geisler, M.; Siemianowski, O.; Skłodowska, A.; Ruszczyńska, A.; Bulska, E.; Antosiewicz, D. M. Ectopic expression of Arabidopsis ABC transporter MRP7 modifies cadmium root-to-shoot transport and accumulation. Environ. Pollut. 2009, 157 (10), 2781−2789. (60) Deinlein, U.; Weber, M.; Schmidt, H.; Rensch, S.; Trampczynska, A.; Hansen, T. H.; Husted, S.; Schjoerring, J. K.; Talke, I. N.; Krämer, U.; Clemens, S. Elevated nicotianamine levels in Arabidopsis halleri roots play a key role in zinc hyperaccumulation. Plant Cell 2012, 24 (2), 708−723. (61) Clemens, S. Toxic metal accumulation, responses to exposure and mechanisms of tolerance in plants. Biochimie 2006, 88 (11), 1707−1719. (62) Birnbaum, K.; Shasha, D. E.; Wang, J. Y.; Jung, J. W.; Lambert, G. M.; Galbraith, D. W.; Benfey, P. N. A gene expression map of the Arabidopsis root. Science 2003, 302 (5652), 1956−1960. (63) Brady, S. M.; Orlando, D. A.; Lee, J.; Wang, J. Y.; Koch, J.; Dinneny, J. R.; Mace, D.; Ohler, U.; Benfey, P. N. A high-resolution root spatiotemporal map reveals dominant expression patterns. Science 2007, 318 (5851), 801−806. (64) Wu, J.; Zhao, F.-J.; Ghandilyan, A.; Logoteta, B.; Guzman, M. O.; Schat, H.; Wang, X.; Aarts, M. G. M. Identification and functional analysis of two ZIP metal transporters of the hyperaccumulator Thlaspi caerulescens. Plant Soil 2009, 325 (1), 79−95. (65) Bonza, M.; De Michelis, M. The plant Ca2+-ATPase repertoire: Biochemical features and physiological functions. Plant Biol. 2011, 13 (3), 421−430. (66) Verbruggen, N.; Hermans, C.; Schat, H. Mechanisms to cope with arsenic or cadmium excess in plants. Curr. Opin. Plant Biol. 2009, 12 (3), 364−372. (67) Cailliatte, R.; Lapeyre, B.; Briat, J.; Mari, S.; Curie, C. The NRAMP6 metal transporter contributes to cadmium toxicity. Biochem. J. 2009, 422, 217−228. (68) Cailliatte, R.; Schikora, A.; Briat, J.; Mari, S.; Curie, C. Highaffinity manganese uptake by the metal transporter NRAMP1 is essential for Arabidopsis growth in low manganese conditions. Plant Cell 2010, 22 (3), 904−917. (69) Takahashi, R.; Ishimaru, Y.; Senoura, T.; Shimo, H.; Ishikawa, S.; Arao, T.; Nakanishi, H.; Nishizawa, N. K. The OsNRAMP1 iron transporter is involved in Cd accumulation in rice. J. Exp. Bot. 2011, 62 (14), 4843−4850. (70) Sasaki, A.; Yamaji, N.; Yokosho, K.; Ma, J. F. Nramp5 is a major transporter responsible for manganese and cadmium uptake in rice. Plant Cell 2012, 24 (5), 2155−2167. (71) van de Mortel, J. E.; Schat, H.; Moerland, P. D.; Ver Loren Van Themaat, E.; Van Der Ent, S.; Blankestijn, H.; Ghandilyan, A.; Tsiatsiani, S.; Aarts, M. G. M. Expression differences for genes involved in lignin, glutathione and sulphate metabolism in response to cadmium in Arabidopsis thaliana and the related Zn/Cd-hyperaccumulator Thlaspi caerulescens. Plant Cell Environ. 2008, 31 (3), 301−324.
3353
dx.doi.org/10.1021/es4042995 | Environ. Sci. Technol. 2014, 48, 3344−3353
