ORIGINAL RESEARCH published: 06 August 2021 doi: 10.3389/fpls.2021.689316 Genome-Wide Association Study Reveals Key Genes for Differential Lead Accumulation and Tolerance in Natural Arabidopsis thaliana Accessions Sílvia Busoms 1* , Laura Pérez-Martín 1 , Miquel Llimós 1,2 , Charlotte Poschenrieder 1 and Soledad Martos 1* 1 Plant Physiology Laboratory, Faculty of Bioscience, Universitat Autònoma de Barcelona, Barcelona, Spain, 2 Department of Biology, Healthcare and Environment, Faculty of Pharmacy and Food Science, Universitat de Barcelona, Barcelona, Spain Soil contamination by lead (Pb) has become one of the major ecological threats to the environment. Understanding the mechanisms of Pb transport and deposition in plants is of great importance to achieve a global Pb reduction. We exposed a collection of Edited by: 360 Arabidopsis thaliana natural accessions to a Pb-polluted soil. Germination rates, Mario Vallejo-Marin, growth, and leaf Pb concentrations showed extensive variation among accessions. University of Stirling, United Kingdom These phenotypic data were subjected to genome wide association studies (GWAs) Reviewed by: Timothy Paape, and we found a significant association on chromosome 1 for low leaf Pb accumulation. Brookhaven National Laboratory Genes associated with significant SNP markers were evaluated and we selected (DOE), United States EXTENSIN18 (EXT18) and TLC (TRAM-LAG1-CLN8) as candidates for having a role Sara V. Good, University of Winnipeg, Canada in Pb homeostasis. Six Pb-tolerant accessions, three of them exhibiting low leaf Pb *Correspondence: content, and three of them with high leaf Pb content; two Pb-sensitive accessions; Sílvia Busoms two knockout T-DNA lines of GWAs candidate genes (ext18, tlc); and Col-0 were Silvia.busoms@uab.cat Soledad Martos screened under control and high-Pb conditions. The relative expression of EXT18, TLC, Soledad.martos@uab.cat and other genes described for being involved in Pb tolerance was also evaluated. Analysis of Darwinian fitness, root and leaf ionome, and TEM images revealed that Pb- Specialty section: This article was submitted to tolerant accessions employ two opposing strategies: (1) low translocation of Pb and Plant Systematics and Evolution, its accumulation into root cell walls and vacuoles, or (2) high translocation of Pb and a section of the journal its efflux to inactive organelles or intracellular spaces. Plants using the first strategy Frontiers in Plant Science exhibited higher expression of EXT18 and HMA3, thicker root cell walls and Pb vacuolar Received: 31 March 2021 Accepted: 19 July 2021 sequestration, suggesting that these genes may contribute to the deposition of Pb in the Published: 06 August 2021 roots. On the other hand, plants translocating high amounts of Pb showed upregulation Citation: of TLC and ABC transporters, indicating that these plants were able to properly efflux Busoms S, Pérez-Martín L, Llimós M, Poschenrieder C and Pb in the aerial tissues. We conclude that EXT18 and TLC upregulation enhances Pb Martos S (2021) Genome-Wide tolerance promoting its sequestration: EXT18 favors the thickening of the cell walls Association Study Reveals Key Genes improving Pb accumulation in roots and decreasing its toxicity, while TLC facilitates the for Differential Lead Accumulation and Tolerance in Natural Arabidopsis formation of dictyosome vesicles and the Pb encapsulation in leaves. These findings are thaliana Accessions. relevant for the design of phytoremediation strategies and environment restoration. Front. Plant Sci. 12:689316. doi: 10.3389/fpls.2021.689316 Keywords: lead tolerance, cell wall, GWAS, extensins, TLC-domain Frontiers in Plant Science | www.frontiersin.org 1 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis INTRODUCTION proteins were the Extensins (EXTs), insoluble glycoproteins that strengthen the primary cell wall and also participate in defense Lead is a neurotoxic element that even at low concentrations is (Lamport et al., 2011). Furthermore, EXTs can interact with highly dangerous for humans and animals. Due to its former other cell wall components such as pectins (Qi et al., 1995). use as gasoline additive, this contaminant is distributed all over Many extensin genes have been cloned and their expression the world with urban and heavy-traffic roadside areas mostly can be regulated and induced by biotic and abiotic stressors affected (Levin-Schwartz et al., 2021). The primary sources of (Josè-Estanyol and Puigdomènech, 2000; Wei and Shirsat, Pb in urban soils are weathering paint, transportation, irrigation, 2006). Different studies have reported extensin induction after soil amendments and industrial waste/debris (Clark et al., 2006; wounding, however, to date they have not been associated with Ratul et al., 2018). In the case of agricultural soils, Pb release is Pb stress. On the contrary, soluble glycosylated proteins of cell generally attributed to metalliferous mining, smelting activities wall (AGPs, arabinogalactan proteins) have been related to metal and the use of fertilizers and pesticides (Reboredo et al., 2019; toxicity showing negative expression due to metal exposure Yuan et al., 2019). Plants, in contrast to humans, are relatively (Mareri et al., 2019). tolerant to soil Pb. Toxicity symptoms in the form of chlorosis, Molecular genetic approaches to identify Pb transport leaf browning, and stunted growth usually occur only in soils with mechanisms are still not conclusive. It has been argued that close to 1000 mg/kg. Nonetheless, phytotoxicity depends on the Pb, as a non-essential element, may not be moved in plants metal speciation and physicochemical properties of soil such as by specific transporters. In fact, HMA3, a metal transporter of pH, cation exchange capacity, and the presence of ligands notably the P-type ATPase family, seems to be involved in vacuolar affect the metal speciation and bioavailability (Smith, 2009). In transport not only of Pb, but also of Cd, Co, and Zn (Morel et al., fact, a Pb2+ activity as low as 0.06 µM was found to be highly 2009). Additional transporter genes involved in lead tolerance toxic to roots of cowpea (Kopittke et al., 2007). have been identified in Arabidopsis: the ABC genes ATM3, Pb enters the plant mainly by soil–root transfer. However, PDR8, and PDR12, and the ACBP1 (Acyl-CoA-binding domain foliar uptake is an important second pathway in areas with protein) (Lee et al., 2005; Kim et al., 2006, 2007; Xiao et al., high fallout of airborne particles (Uzu et al., 2010; Luo et al., 2008). Moreover, under Pb stress Arabidopsis thaliana induces 2019). Plants growing in Pb-contaminated soils accumulate the the expression of the cytoplasmatic protein PSE1 (Fan et al., metal mainly in the roots. Fortunately, only a small proportion 2016). The authors reported a certain connexion of this gene with is translocated to consumable tissues (Yousaf et al., 2016). the transporter PDR12 because PSE1 activates genes involved on Consumption of plants cultivated on Pb-rich soils is the greatest the phytochelatin synthesis which could induce the expression of potential health risk above exposure to soil/dust or drinking tap ABC transporters. water (Byers et al., 2020). Therefore, it is important to grow crops Recently developed genetic tools are being applied to discover with low ability for Pb accumulation in edible parts. new molecular markers associated to lead tolerance. One of Although non-hyperaccumulator species usually restrict Pb them is the genome-wide association study (GWAs), used to transport to the shoots, leaf Pb accumulation may considerably connect plant phenotypes with candidate marker genes. This vary among species (Bech et al., 2012; Ng et al., 2016; Zhou tool has lately been used to describe nine potential candidate et al., 2016) and even varieties (Liu et al., 2013). The mechanisms genes related to the Pb-tolerance in Brassica napus (Zhang behind such differential Pb accumulation in plants are still poorly et al., 2020). Using the GWAs approach, we aim to improve understood. One of the principal mechanisms to prevent the the fragmentary picture of genes responsible for Pb tolerance toxic effect of metals in plants is the absorption on cell wall taking advantage of the variability present in natural accessions components (Gupta et al., 2013). Pb2+ binds tightly to carboxyl of A. thaliana. In particular, our goals are (1) to elucidate the groups in cell walls and pectin has been proposed as a main mechanisms developed by Pb-tolerant accessions that differ in ligand for lead (Inoue et al., 2013; Chudzik et al., 2018). Methyl- their Pb sequestration strategies, and (2) to stablish the role of esterification of cell wall pectins reduces availability of carboxyl the novel candidate genes identified by GWAs in Pb tolerance. groups for metal immobilization. In fact, a substantial increase of low methyl-esterified pectins has been observed in Pb-induced cell wall thickenings in different plant species (Krzesłowska et al., MATERIALS AND METHODS 2010; Sumranwanich et al., 2018). Environmental signals produce modifications in the HapMap Phenotyping and GIS Data composition and chemical structure of cell walls. Related to Extrapolation metal stress, the thickening of xylem and cortical parenchyma Seeds of the Arabidopsis thaliana HapMap cohort (360 natural cell walls has been reported as a phytotoxic effect of metal populations) were obtained from Nottingham A. thaliana Stock exposition in Brachiaria decumbens (Gomes et al., 2011). After Centre (NASC, Nottingham, United Kingdom). The list and polysaccharides (cellulose, hemicelluloses, and pectins), proteins information of the accessions is detailed in Supplementary are the second major components of cell walls having structural Dataset 1. Seeds were surface sterilized using 3:10 bleach solution and enzymatic functions. Although proteins represent around in constant agitation for 10 min and rinsed with sterile water six 10% of cell wall, their activities are responsible for the cell wall times. Seeds were stratified for 4 days at 4◦ C in 0.1% agar solution alterations during plant development and environmental stress to synchronize germination, then five seeds per accession were (Cruz-Valderrama et al., 2019). The first described cell wall sown in two soil types with contrasted Pb concentrations into Frontiers in Plant Science | www.frontiersin.org 2 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis pot trays. The control soil was a mix of commercial potting determine the candidate genes in LD. GWAs output for each mix soil with perlite (2:1); the mine soil was a mix of soil phenotype are available in Supplementary Dataset 2. from an ancient lead mine (Mina de Can Vergeli; 41◦ 530 42.400 N To validate the significant SNPs obtained on the GWA- 3◦ 010 11.700 E), commercial potting mix soil and perlite (2:1:1). Portal, genotype data of the 210 A. thaliana accessions was Trays were placed in a growth chamber with 150 mmol/m2 s of obtained from 1001genomes.org/(1135_snp-short-indel.vcf.gz). light intensity, 12 h light/12 h dark photoperiod, and 25◦ C/20◦ C Indels were removed and SNPs were filtered using the criteria day/night temperature. Trays were bottom-watered twice weekly allele count < 100 and MAF < 4.8% using GATK-3.6.0 with deionized water. Germination and survival were monitored (McKenna et al., 2010). Principle components analysis (PCA) was and rosette diameter (RD) of three individuals per accession performed and a kinship matrix was calculated using the GAPIT and soil type was measured weekly for 4 weeks. Growth rate of package in R (Lipka et al., 2012). The compressed mixed linear each accession was calculated as mean(RDMine )/ mean(RDControl ) model (CMLM) was used for performing GWAs by incorporating of the last RD measure. Fifty-day old plants were harvested to K matrix along with PCAs employing the program TASSEL 5.0 measure its nutrient mineral contents. (Bradbury et al., 2007; Supplementary Table 1). In order to estimate the edaphic parameters of each A. thaliana All genes associated with the significant SNPs and their SNPs natural population native soil, coordinate locations and public in strong LD (r2 > 0.8) were explored (Supplementary Table 1). maps from the European Soil Data Centre (ESDAC) database Genes were annotated according to TAIR 10 (Araport11) and (Panagos et al., 2012) were combined using Q-GIS1 . Natural relative gene expression data for A. thaliana (Col−0) plants populations coordinates were extracted from GWAPP2 in exposed to lead was retrieved from a previous transcriptome WGS84 system (latitude and longitude). Maps of heavy metals study (unpublished data). in the soils of the EU, based on Lucas, 2009 HM data, were used to estimate the Pb, As, Cd, Cr, Cu, Hg, Mn, Sb, Co, and Plant Materials and Experimental Design Ni concentrations (mg/kg) of the European topsoils. Maps of Seeds from eight selected A. thaliana natural populations (Si-0, soil properties at European scale, based on Lucas, 2009/2012 Tiv-1, Wag-4, JEA, Kro-0, Rak-2, Appl-16, and Bsch-0), Col-0 topsoil data, were used to extract the following variables: pH and 2 T-DNA insertion lines (ext18 (SALK_201747C), and tlc (measured in H2 O), pH (in CaCl2 0.01 M solution), Cation (SALK_136500) mutants) were obtained from the A. thaliana Exchange Capacity (CEC), Calcium carbonates (CaCO3 ), C:N Stock Centre (NASC, Nottingham, United Kingdom). Figure 1 ratio, and N, P, and K concentrations (mg/kg). Native soil data details the workflow and the A. thaliana accessions and mutants was available for 305 of the 360 A. thaliana HapMap cohort. used in each experiment. A. thaliana natural accessions were classified as “High Leaf Lead” (HLL) or “Low Leaf Lead” (LLL) GWA Studies and Candidate Genes based on their leaf Pb concentrations, and as tolerant (T) or Selection sensitive (S) accessions based on their growth under high- Phenotypes obtained from the 210 A. thaliana natural accessions Pb conditions. that survived and grew on the mine soil were used for Genome Wide Association Studies (GWAs). These phenotypes Pb-Enriched Mine Soil Pot Experiment included: (1) Rosette Diameter mean (RDMine ) of three 50- Seeds were stratified for 4 days at 4◦ C in 0.1% agar solution days old plants cultivated in the mine soil, (2) growth rate to synchronize germination, then sown into the same soil [RDMine/Control = mean(RDMine )/ mean(RD Control )] of three 50- types (Mine/Control) used for the HapMap screening. Each tray day old plants, (3) native soil Pb concentration, and (4) leaf contained 11 accessions (Si-0, Tiv-1, Wag-4, JEA, Kro-0, Rak-2, mineral concentrations (Pb, Zn, Cu, B, Ca, K, Mg, Na, P, S, Fe, Appl-16, Bsch-0, Col-0, ext18 and tlc) randomized, with a total Mn) of three plants per accession cultivated in the mine soil of eight individuals per accessions and soil type. Plants were (Supplementary Dataset 1). All the phenotypes are detailed in allowed to germinate and grow at 12 h light/12 h dark and 25 Supplementary Dataset 1. to 20◦ C day/night temperature for 1 week before transferring The HapMap population has been genotyped for 250K them to a 4◦ C environment at 8h day length for 6 weeks of bi−allelic SNPs (Platt et al., 2010; Chao et al., 2012), and vernalization to synchronize bolting time between the accessions. this dataset is available in the GWA−Portal platform3 (Seren, After the vernalization period, trays were moved back to the 2018). GWAs were performed under the following settings: growth room at initial conditions. Trays were bottom-watered the accelerated mixed model, which considers the population twice weekly with 250 ml of deionized water per tray. Rosette structure; 250 k SNP dataset; and BoxCox transformation diameter was measured every week for 5 weeks and the number (normality was checked by applying the Shapiro–Wilk test). After of siliques was counted at maturity. For leaf ionome analysis, two running the analysis, the filtering of the minor allele frequency leaves from 5-week-old plants were collected from four plants per (MAF) was set to 0.1. SNPs above Bonferroni correction [– accession and soil type. log10(p) > 5] were selected, and the linkage disequilibrium (LD) region associated with the highest score SNP was explored to Hydroponic Experiment Seeds of the 11 accessions (Si-0, Tiv-1, Wag-4, JEA, Kro-0, Rak- 1 http://qgis.osgeo.org 2, Appl-16, Bsch-0, Col-0, ext18, and tlc) were surface sterilized 2 http://gwapp.gmi.oeaw.ac.at/ with bleach solution (30% commercial bleach + 0.02% Triton X- 3 http://gwas.gmi.oeaw.ac.at 100) for 15 min and washed five times with sterile water and Frontiers in Plant Science | www.frontiersin.org 3 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis FIGURE 1 | Methodology process flowchart. sown in plates with: half MS medium (pH 4.5, Control) or half from four plants per accessions were stored for ionomic and MS medium and 50 µM of PbNO3 (pH 4.5, Pb treatment). microscopy analysis. Plates were kept in the dark at 4◦ C for 48 h to synchronize germination. After that, plates were placed in a growth chamber with 150 mmol/m2 s of light intensity, 12 h light/12 h dark Elemental Composition of Soils and photoperiod, and 25◦ C/20◦ C day/night temperature. Radicle Plants emergence and root length were monitored for 15 days. Six 10- To analyze the composition of the control and mine soils, we took day old seedlings per accession and treatment were transferred to six independent samples of each soil type from the trays used individually hydroponic circular containers (50 mL) filled with for the pot experiment, three at the begging, and three at the 0.5-strength Hoagland solution (pH 4.5) and 0 µM (control) end of the experiment. The soil characterization was performed or 50 µM of PbNO3 . The hydroponic solution was changed on the 2-mm fraction samples following the extraction method every 3 days to maintain a constant concentration of nutrients described in Busoms et al. (2018). in the solution and PbNO3 was increased three times until a final Plant tissues were sampled by removing 2–3 leaves or cutting concentration of 100 µM of PbNO3 (pH 4.5, Pb treatment) when the roots (1–5 mg dry weight) and washed it with 18 M plants were 21 days old. Plants remained at these conditions for water before being placed in Pyrex digestion tubes. Sampled 2 weeks. We harvested 35-days old plants and we measured the plant material was dried for 2 days at 60◦ C and weighed before root length, rosette diameter and total fresh weight (biomass) open-air digestion in Pyrex tubes using 0.7 mL concentrated of six individuals per accession and treatment. Roots and leaves HNO3 at 110◦ C for 5 h in a hot-block digestion system Frontiers in Plant Science | www.frontiersin.org 4 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis (SC154-54-Well Hot Block, Environmental Express, Charleston, Statistical Analyses SC, United States). Concentrations of the selected elements All the statistical analyses were conducted using JMP SAS (Ca, K, Mg, Na, P, S, Mo, Cu, Fe, Mn, Zn, Pb, and Cd) were software (SAS Institute, Cary, NC, United States). Normality determined by ICP-MS (Perkin ElmerInc., ELAN 6000, MA, was checked and non-normal data were transformed before United States) or ICP-OES (Thermo Jarrell-Ash, model 61E applying any parametrical tests. Mean-standardized values (– Polyscan, England). 1 < value > 1) of elemental contents of soil and leaf material were used to represent the radar plots and compare between soil types, Gene Expression Analyses treatments or accessions. Outlier removal was implemented using Total RNA of about 100 mg of plant powder was extracted the algorithm described in Davies and Gather (1993). using the Maxwell RSC plant RNA kit (Promega Corporation, R The phenotypic responses to high Pb conditions were WI, United States) following the manufacturer’s instructions. quantified and compared to the respective control conditions Two micrograms of total RNA were used as a template to [relative measurements: mean(X Pb )/ mean(X Control )]. We used synthesize first-strand cDNA with the iScriptTM cDNA Synthesis a linear mixed effects model (LMM) to investigate changes in Kit (Bio-Rad, United States). The cDNA was used as a template rosette diameter (RD) and silique number (Fitness) in the pot for quantitative PCRs using iTaqTM Universal SYBR Green R experiment, and changes in root length (RL) and fresh weight Supermix (Bio-Rad, CA, United States). Real-time detection of (Biomass) in the hydroponic experiment. The models included fluorescence emission was performed on a CFX384 Real-Time a two-way interaction between ‘Pb treatment’ and ‘accession’ or System (Bio-Rad, CA, United States), and plates were edited ‘accession classification.’ using the CFX manager version 3.1 software. One-way ANOVA was used to test for significant differences Primers used for EXT18 (At1g26250), TLC (At1g26200), (P < 0.05) between means of phenotypic responses, gene ACBP1 (At5g53470), PDR8 (At1g59870), PDR12 (At1g15520), expression, and between means of elemental contents of soil and PSE1 (At5g06370), and HMA3 (At4g30120) transcript leaf material. To test for correlations between two variables a quantification are detailed in Supplementary Table 2. For Bivariate Fit was conducted. To perform multiple comparisons normalization across samples, the expression of the ACTIN of group means we used Tukey’s HSD. Statistical data analyses 2 gene (At3g18780) was analyzed. For each sample, the are specified in Supplementary Datasets 3, 4. average value from triplicate real-time PCRs was used to estimate transcript abundance. The mean Ct values Data Availability were normalized against Actin2 gene and Ct values were The data generated and/or analyzed in the current study calculated as (CtGene - CtActin1 ). The relative expression of are either included in this article as Supplementary Material each target gene was calculated with the 2−11Ct method (Pb or submitted to public repositories. Raw phenotype data are treatment – control) in four samples per accession and treatment available from AraPheno4 and full GWAs analysis are available (Livak and Schmittgen, 2001). at Gwa-portal under the study named ‘Leaf Pb Mine5 ’. TEM Imaging To determine ultrastructural changes and location of Pb deposition, plant material from four individuals of Si-0, Rak- RESULTS 2, Bsch-0, Col-0, ext18, and tlc accessions submitted to control (0 µM PbNO3 ) and high Pb conditions (100 µM PbNO3 ) Screening of Arabidopsis thaliana in hydroponics for 2 weeks was harvested. Roots and leaves HapMap Tolerance to Pb Stress were fixed using 3% (v/v) glutaraldehyde in 0.1M cacodylate To investigate the Pb tolerance of natural accessions of buffer overnight at 4◦ C. After washing three times in buffer, A. thaliana, the 360 accessions constituting the HapMap cohort plant material was post fixed with 1% (v/v) osmium tetroxide (Horton et al., 2012) were cultivated in a Pb-enriched soil in the medium buffer for 1 h and washed twice in distilled excavated from an old, abandoned metal mine in the NE of Spain water. Samples were dehydrated in a graded acetone series (Mine “Can Vergeli”; Supplementary Figure 2A). In parallel, followed by infiltration and embedded in Spurr’s resin. Roots the same cohort was cultivated in a control non-metalliferous and leaves were sectioned using a glass knife on Leica EM soil with similar characteristics (Supplementary Figure 1). Of UC7 ultramicrotome (Leica Microsystems, Wetzlar, Germany). the 360 accessions sown, 296 germinated and grew well under Roots were continuously oriented transversely to the root axis control conditions; 226 germinated in both soil types and to allow recording of the distance from the root tip by counting only 210 accessions survived more than 5 weeks in the mine sections. Ultrathin (80 nm) sections were made using a Diatome soil (Supplementary Dataset 1). Although the rosette diameter diamond knife on Leica EM UC7, collected onto copper grids, (RD) of 5 weeks old plants varied significantly among all the then treated with 5% uranyl acetate in water for 60 min. Sections accessions under both control and Pb stress conditions, the were examined in a Jeol 1200 TEM (Jeol, Tokyo, Japan). Cell RD of almost all the accessions was reduced under high Pb wall thickness was measured using ImageJ software. At least 30 conditions (RDMine/Control mean ± SD: 0.67 ± 0.18). However, cells per accession, organ and treatment were measured. For Pb identification, sections were examined using a TEM-EDX 4 https://arapheno.1001genomes.org/study/92/ spectroscopy system JEOL-1400 (Jeol, Tokyo, Japan). 5 https://arapheno.1001genomes.org/rest/study/12/values.csv Frontiers in Plant Science | www.frontiersin.org 5 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis we detected a group of accessions with great tolerance to Pb stress deposition of Pb in these cell structures. TLC is a protein (RDMine/Control > 0.9, Supplementary Dataset 1). with a TRAM/LAG1/CLN8 lipid-sensing domain involved in To inspect if this tolerance was associated with soil properties several lipid-related processes. Its role in Golgi transport vesicle from the native habitats of the tested accessions, we used the could enhance Pb encapsulation. To verify if these genes public maps from the European Soil Data Centre (ESDAC) to were responsible for the observed LLL association we obtained extrapolate the pH and trace metal content of all the European the knockout mutants of these genes (ext18, tlc) to conduct sites (Figure 2A). We observed that plants coming from sites further studies. with elevated soil Pb content (and acidic pH) had better capacity to tolerate this stress (Figure 2B). This confirms the Phenotypic Responses to High Pb importance of soil composition as a potential factor driving local To confirm the tolerance or sensitivity of the 11 selected adaptation of plants. accessions, we repeated the same pot experiment performed Leaf tissue of three plants grown under both conditions was to conduct the GWAs but with a higher number of replicates harvested for ionomic analysis. All the elements presented high (n = 8 individuals per accession and soil type). In parallel, variability depending on the accession and growth conditions. in order to conduct root analysis, the same accessions were However, we detected no significant correlation between the Pb cultivated in hydroponics and treated with 0 µM (Control) or tolerance of the accessions (growth rate) and the concentration 100 µM of PbNO3 (Pb treatment). The phenotypic responses of any of the elements analyzed (R2 < 0.1) (Supplementary to high Pb conditions were quantified and compared to the Dataset 1). The most relevant differences were found in leaf Pb respective control treatments. Both experimental settings verified accumulation from plants grown in the mine soil (RSD = 1.2). that accessions classified as tolerant (T), independently of being We observed two clear phenotypes among the most tolerant LLL or HLL, had larger rosettes, longer roots and produced more accessions: those accumulating high amounts of Pb in the shoots, seeds than the other accessions when they were cultivated under denoting a good Pb compartmentalization mechanism; and Pb stress (Figures 3A–D). Col-0 again showed an intermediate those not translocating the Pb to the aerial parts, suggesting tolerance and both mutants, tlc and ext18, performed similar to Pb exclusion or root accumulation strategies (Figure 2C). We the sensitive accessions, suggesting that the missing genes have classified the A. thaliana natural accessions as ‘High Leaf a role in Pb tolerance. We noticed that all the tested accessions Lead’ (HLL) or ‘Low Leaf Lead’ (LLL) based on their leaf Pb suffered a root length reduction under the Pb treatment, concentrations, and as tolerant (T) or sensitive (S) accessions but ext18 also had shorter roots under control conditions based on their growth rate (RDMine/Control ). (Supplementary Figure 2). The analysis of ionome denoted that our hydroponic experiment mimicked the mine soil conditions because in Detection of SNPs Associated With Low both settings we obtained similar patterns in the leaf nutrient Pb Content in Above-Ground Tissues composition of the different accessions (Supplementary To identify genomic regions underlying tolerance to Pb, we Figure 3). As observed by other authors, Pb reduces nutrient conducted GWAs on several phenotypes including growth concentration in shoots, especially of bivalent cationic elements (RDMine and RDMine/Control ) of 50-day old plants, native soil Pb such as Zn, Mn, Mg, Ca (Pourrut et al., 2011). In our pot content, and leaf mineral content of plants cultivated in the mine experiment we did not detected a decrease of tissue Zn levels, soil (see section “Materials and Methods”). The only trait that probably because the mine soil used was also rich in Zn, as gave significant associations was the leaf Pb content of natural well as in other trace elements like Cd and Cu (Supplementary accessions exposed to a Pb-enriched mine soil (Supplementary Figure 1). However, in the hydroponic experiment we could Dataset 2). Significantly associated SNPs [-log10(p) > 5] were clearly observe a Pb-induced reduction of most of the divalent identified in chromosomes 1 (7 SNPs), 3 (1 SNP), 4 (1 SNP), and nutrients in the shoots (Supplementary Figure 3). 5 (2 SNPs) (Figure 2D). The scores, positions, genes associated Leaf ionome analysis corroborated the LLL and HLL condition with these SNPs, and the LD of the regions are detailed in of the selected accessions (Figures 3E,F). We also detected Supplementary Table 1. Here, we focused on the most significant that ext18 mutants translocated high levels of Pb, hinting that peak of chromosome 1 but we consider that the peaks found in this mutant is not able to retain the Pb in the cell wall of Chr4 and Chr5 should be investigated in future studies. the roots (Figures 3E,F). Importantly, root ionome analysis We explored the 15 SNPs in strong LD (r2 > 0.8) with revealed that LLL accessions accumulate high amounts of Pb the top SNP of Chr1, obtaining a total of 12 candidate genes in the roots (Figure 3F), thus they cannot be considered (Figures 2D,E). For all the SNPs, the rare allele was associated Pb excluders and must have exceptional mechanisms of root with the ‘Low Leaf Lead’ (LLL) haplotype (Figures 2F,G). Pb deposition. On the other hand, both T-HLL and Pb- The function, localization, and expression of the candidate sensitive (S) accessions translocated similar Pb concentrations genes was examined. We selected EXT18 (At1g26250) and (Figure 3F), suggesting that (S) plants are not able to maintain TLC (At1g26200) because we considered they were the only the Pb out of the cytosol and it becomes toxic for them. target genes that could have a role in Pb sequestration. The tlc mutant, despite accumulating less Pb in the leaves EXT18 is a member of the 20 ‘classical EXTs’ considered than the HLL accessions, exhibited similar growth and fitness structural components of primary cell walls (Johnson et al., performance than the (S) plants (Figure 3), indicating the 2003). Its localization in the cell walls could contribute the costs of Pb toxicity. Frontiers in Plant Science | www.frontiersin.org 6 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis FIGURE 2 | GWAS of A. thaliana HapMap leaf Pb content. (A) Europe map of estimated soil Pb content with the geographic location of the European A. thaliana populations from the HapMap cohort (black dots). Selected accessions are highlighted with colored circles and names. (B) Estimated Pb concentration of the native soil of 141 European A. thaliana accessions and the relationship with the growth rate (Rosette diameter of 42-day-old plants) of each accession cultivated in common soil (control) or in Pb-enriched mine soil (mine). (C) Leaf Pb concentration of 210 accession cultivated in Pb-enriched mine soil for 6 weeks and the relationship with the growth rate (RDMine/Control ). We considered Pb-tolerant accessions (T) when growth rate is >0.9, and Pb sensitive accessions (S) when growth rate is <0.4. (D) Manhattan plot of all A. thaliana chromosomes and zoom of chromosome 1 region with a significant peak for SNP associations to leaf Pb content. The horizontal purple dashed line corresponds to a nominal 0.05 significance threshold after Bonferroni correction. SNPs are color-coded to show their LD relationships with the top SNP (yellow = 0.4 r 2 < 0.6; orange = 0.6 r 2 < 0.8; red = r 2 > 0.8). (E) Position, score (p-value), location, and allele description of the 15 SNPs in high LD; target genes of each SNP and its cellular localization. (F) Leaf Pb concentration of the two haplotypes identified: “low leaf lead” (LLL) associated with the rare allele, and “high leaf lead” (HLL) associated with the common allele. (G) Description of the nine selected A. thaliana natural accessions. (*) Indicates the number of SNPs in which they have the rare allele. Selected accessions are highlighted with colored dots in all the graphs: green dots = T-LLL; red dots = T-HLL; blue dots = (S) and gray dot = Col-0. Frontiers in Plant Science | www.frontiersin.org 7 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis FIGURE 3 | Phenotyping of representative accessions under Pb stress. Mean ± SE of (A) growth rate (RDMine/Control ) and (B) fitness (silique number) of 3 T-LLL (green), 3 T-HLL (red), 2 (S) (blue) accessions, Col-0 (gray), ext18 (dark purple), and tlc (light purple) cultivated in a pot experiment in control soil or in Pb-enriched mine soil (n = 8 plants per accession and soil type). Mean ± SE of (C) root length (mm) and (D) biomass (fresh weight, g) of 3 T-LLL, 3 T-HLL, 2 (S) accessions, Col-0, ext18, and tlc cultivated in hydroponic solution and treated with 0 or 100 µM of PbNO3 for 2 weeks (n = 6 plants per accession and treatment). Mean ± SE of (E) leaf Pb content from four plants per accession cultivated in the Pb-enriched mine soil; (F) leaf and root Pb content of four plants per accession cultivated in hydroponics and treated with 100 µM of PbNO3 for 2 weeks. Letters indicate significant differences (p < 0.05, Tukey’s HSD). Frontiers in Plant Science | www.frontiersin.org 8 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis FIGURE 4 | Transcript expression of GWAs target genes and other genes involved in Pb tolerance. Leaf and root relative expression (2- 11Ct ) of (A) TLC (At1g26200), (C) ACBP1 (At5g53470), and (D) PDR8 (At1g59870). Root relative expression (2- 11Ct ) of (B) EXT18 (At1g26250) and (G) HMA3 (At4g30120). Leaf relative expression (2- 11Ct ) of (E) PDR12 (At1g15520) and (F) PSE1 (At5g06370). Data represents the mean ± SE of four plants per accession and letters indicate significant differences (p < 0.05, Tukey’s HSD). Transcript Levels of Genes Involved in Pb plants exposed to 0 or 100 µM of PbNO3 for 2 weeks. We detected Tolerance that the TLC gene was induced by Pb in the leaves of T-HLL and Gene-expression followed by proteins synthesis is also affected by Col-0 plants and in the roots of the T-LLL accessions (Figure 4A). Pb stress. We compared the expression level of TLC and EXT18 in The transcription of TLC correlated with the amount of Pb found Frontiers in Plant Science | www.frontiersin.org 9 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis in the tissues of the tolerant plants, suggesting again that this gene both roots and leaves of Col-0 plants (Figures 6B,D). These can enhance Pb tolerance. depositions were not clearly immobilized in root cell walls or EXT18 only expresses in plants roots and we found that intercellular spaces, but we detected a high sequestration in the Pb treatment induced the expression of this gene in almost all leaf vacuoles (Figure 6D). the tested accessions (Figure 4B). However, EXT18 expression in roots of T-LLL plants was much higher than in the other plants. TLC and EXT18 expression of the respective mutants was DISCUSSION undetectable in any tissue under any condition (Supplementary Dataset 4), confirming that the T-DNA insertions used to Lead is a non-essential metal that does not have a role in generate the tlc and ext18 mutants are blocking its transcription any cell metabolism process but it is easily absorbed and (Figures 4A,B). accumulated in various plant tissues. Excessive Pb accumulation Several ABC (ATP-binding cassette) transporters, such as impairs morphological, physiological and biochemical functions ACBP1, PDR8, or PDR12, have been identified as being involved and has a range of deleterious effects, such as inhibition of in resistance to Pb (Kim et al., 2006, 2007; Xiao et al., 2008). In root growth, chlorosis or impaired uptake of essential elements our tested accessions these three genes were highly upregulated (Pourrut et al., 2013). Given this, metalliferous habitats exert clear in the leaves of T-HLL plants, especially in the Rak-2T−HLL selection pressures on plant communities (e.g., Sailer et al., 2018). accession (Figures 4C–E). ACBP1 and PDR8 were enhanced We observed that the HapMap cohort of A. thaliana natural by Pb in the roots of the T-LLL accession but also in the accessions come from habitats with remarkable differences in roots of the (S) plants (Figures 4C,D). Our results indicate that soil Pb content. Our screen of the 360 A. thaliana accessions the regulation of these transporters is more relevant in aerial in a Pb-enriched mine soil revealed a variety of responses to tissues, where they can be involved in Pb sequestration into Pb stress, from accessions that did not germinate, to some that inactive organelles. survived a few days, to accessions that grew even better in high PSE1 is a gene described for conferring Pb tolerance and Pb conditions than control soil. Among the tolerant accessions, root Pb accumulation in A. thaliana (Fan et al., 2016). Our we identified plants accumulating high levels of leaf Pb (termed T-LLL accessions displayed much higher expression of PSE1 than HLL accessions) and plants with low leaf Pb (LLL accessions). the rest of accessions (Figure 4F), supporting that this gene We observed that accessions coming from sites with acidic soils is induced under Pb stress and confers tolerance preventing and elevated levels of Pb were the most tolerant to this stress, Pb translocation. manifesting signals of local adaptation. Intracellular sequestration is a further relevant mechanism Numerous studies have documented local adaptation of to minimize Pb toxicity to vital functions of the cytosol. plants to environmental factors and soil characteristics (e.g., The HMA3 protein is involved in Pb detoxification by Antonovics, 2006; Busoms et al., 2015), but fewer have focused participating in its vacuolar sequestration (Morel et al., 2009). on adaptations to specific soil pollutants. An important condition We detected that Tiv-1T−LLL and the other T-LLL accessions for natural selection to act upon a trait is that this trait has also manifested higher levels of HMA3 expression in their phenotypic variation which is genetically determined. Pb uptake, roots (Figure 4G). translocation and accumulation in plants can be regulated by several genes that are still unknown. GWAs has become a Cell Wall Thickening and Pb Deposition standard procedure for dissecting complex genetic traits, with Cell wall thickening and lignification are important histological proven success in plants (e.g., Fusari et al., 2017). In order responses of the plants to avoid metal toxicity. To verify whether to detect novel genes related to Pb tolerance, we performed the presence of Pb or the upregulation of EXT18 stimulates the a GWA study on phenotypes obtained from the cultivation thickness of the cell walls, we measured the roots and leaves of 360 A. thaliana natural accessions in control and Pb- cells walls of Si-0T−LLL , Rak-2T−HLL , Bsch-0S , Col-0, ext18, and enriched mine soils. tlc plants exposed to control or 100 µM PbNO3 for 2 weeks. Leaf Pb content from plants cultivated in the Pb-enriched The major differences were identified in root cells walls. Pb mine soil was the only phenotype that showed significant stress caused a pronounced thinning of the root cell walls of associations. Significant SNPs from various chromosomes were the sensitive accessions Bsch-0S and ext18 (Figure 5). On the explored but we focus on the chromosome 1 region where contrary, Si-0T−LLL individuals exhibited wider root cell walls all the rare SNPs are associated with a unique phenotype: when plants were treated with Pb (Figure 5). accessions with low leaf Pb concentrations (LLL). Except for To determine the subcellular location of Pb, we performed hyperaccumulator species, metal tolerant plants tend to restrict TEM-EDX analysis in the same plants used for cell wall soil–root and root–shoot transfers. Ionomic analysis revealed that thickening measurements (Supplementary Figure 4). We T-LLL accessions were capable of accumulating Pb in the roots, observed large crystalline-like deposits that saturated the root restricting the translocation but not the uptake. cell walls of the tolerant Si-0T−LLL accession (Figure 6A). TEM Pb has strong affinity for cell wall, membrane constituents images also exposed that Rak-2T−HLL plants had few Pb deposits and detoxification systems (Meyers et al., 2009; Krzesłowska in the roots (Figure 6C) but in the leaves they were able to efflux et al., 2016). The two target genes, EXT18 and TLC, located almost all the translocated Pb to intercellular spaces (Figure 6E). in these structures were selected to study their role in Pb In comparison, we observed similar amount of Pb deposits in sequestration and tolerance. Frontiers in Plant Science | www.frontiersin.org 10 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis FIGURE 5 | Cell wall thickness of representative accessions under control and high Pb conditions. (A) Wall thickness of leaf cell (mesophyll) and root cells (cortex) of Si-0, Rak-2, Bsch-0, Col-0, ext18, and tlc plants cultivated in hydroponic solution and treated with 0 µM (control, gray bars) or 100 µM of PbNO3 (Pb treatment, black bars) for 2 weeks. Data represents the mean ± SE of 30 cells from three plants per accession. Significant differences between treatments are marked with asterisks (*p < 0.05, **p < 0.001; Student’s t-test). (B) Transmission electron microscopy (TEM) images of root cortex cells of Si-0, ext18 (50,000 magnification) and Bsch-0 (12,000 magnification) plants submitted to control or Pb treatment and showing the differences in wall thickness between accessions and treatment. Role of EXT18 in Pb Tolerance motifs (Lamport et al., 2011). Extensin deposition is considered Extensins are cell wall-located, basic, Hyp-rich structural important for the correct assembly and biophysical properties glycoproteins with alternating hydrophilic and hydrophobic of primary cell walls, with consequences for growth, cell Frontiers in Plant Science | www.frontiersin.org 11 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis FIGURE 6 | Lead localization in A. thaliana root and leaf cells by TEM-EDX. (A) Cross sections of tolerant Si-0LLL accession illustrating large Pb deposits saturating the root cell walls. Cross sections of root cortex cells from (B) Col-0 and (C) Rak-2HLL accessions showing Pb deposits not precipitated into cell walls. Cross section of leaf mesophyll cells from (D) Col-0 and (E) Rak-2HLL accessions exposing Pb deposits sequestered in vacuoles and intracellular spaces. adhesion, plant resistance to pathogens, and ion depositions Hydroponic cultivation of Pb tolerant (T-LLL and T-HLL) (Roberts and Shirsat, 2006; Pereira et al., 2011). Cannon and sensitive (S) natural accessions, the ext18 mutant and the et al. (2008) demonstrated that the loss of the structural wild type (Col-0) under control and high Pb conditions (100 µM glycoprotein network in A. thaliana is lethal and therefore PbNO3 ) revealed that Pb treatment enhanced EXT18 expression primary cell wall proteins are essential. Indeed, Choudhary in root of T-LLL and Col-0 accessions. Overexpression of EXTs et al. (2015) revealed that EXT18 is an important contributor promotes root hair growth (Marzol et al., 2018) and increases to cell wall integrity in rapidly extending walls, and it is cell wall biomass (Tan et al., 2014). T-LLL accessions had much essential for the normal vegetative growth of Arabidopsis longer roots and extended root cell walls farther than the rest plants. Extensins have been mainly associated with cell of accessions. Thicker cell walls are essential to accommodate Pb wall structure and biotic stress resistance. However, deposits. TEM images showed that root cell walls of T-LLL plants our results suggest that extensins are also relevant for exposed to high Pb were covered with Pb deposits, preventing the metal detoxification. translocating of Pb to the aerial tissues. Moreover, we observed Frontiers in Plant Science | www.frontiersin.org 12 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis FIGURE 7 | Model for the Pb tolerance mechanism of tolerant LLL (Low Leaf Lead) and HLL (High Leaf Lead) A. thaliana natural accessions. that ext18 mutants had thinner root cell walls than wild type (Col- and its molecular function is unknown (Araport11). However, 0) plants and the mutants accumulated higher Pb levels in the TLC was identified in a GWAs as a candidate gene with leaves when submitted to Pb stress, supporting the hypothesis known relationships to lipid metabolism and protection against that without this gene plants have a reduced capacity to retain peroxidation in maize (Riedelsheimer et al., 2013). metals in their roots. Early studies by Malone et al. (1974) and Wierzbicka Early studies revealed that oxygen peroxide causes enhanced et al. (2007) reported Pb sequestration in Golgi apparatus, transcription of extensin genes in tomato roots (Wisniewski et al., endoplasmatic reticulum, and plasma membrane. Results on Pb 1999). Peroxidation of cell wall EXTs is a fundamental process for compartmentation in the Pb-resistant signal grass (Brachiaria cell wall strengthening, and EXT18 is able to form iso-di-tyrosin decumbens) suggest that Pb locates to intracellular membrane bonds thus providing covalent cross-linking networks that play a structures, probably Golgi apparatus, prior to final deposition in fundamental role in cell wall integrity (Choudhary et al., 2015). cell walls (Kopittke et al., 2008). The endomembrane system is a We thus speculate that the enhanced expression of EXT18 found complex network of membrane-bound compartments. The post- in the T-LLL accessions is contributing to the expansion of the Golgi transport pathways are responsible for the deposition of root cell walls and the consequent Pb deposition (Figure 7). many of the structural components of cell walls in the apoplast (Sinclair et al., 2018). Lead deposits in Golgi-derived vesicles has been interpreted as a way to efflux Pb to the apoplast by exocytosis Potential Role of TLC-Domain in Pb (Malone et al., 1974). Sequestration TLC was upregulated in the leaves of T-HLL and Col-0 TRAM/LAG1/CLN8 lipid-sensing domain has been well accessions and in the roots of T-LLL accessions submitted to a described in humans and yeast, where it is associated with Pb treatment, in accordance with the Pb accumulation pattern ER-Golgi trafficking (Winter and Ponting, 2002). In plants, of each phenotype. TEM images revealed that Col-0 and Rak- TLC is cataloged as a ‘lipid-sensing domain containing protein’ 2T−HLL plants retained Pb in leaf vacuoles, inactive organelles, Frontiers in Plant Science | www.frontiersin.org 13 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis or intracellular spaces. The strong expression of TLC under Future studies with deeper sampling will surely discoverer Pb exposure in T-HLL plants suggests that this gene has an additional genes associated with the sequestration of Pb and important role in Pb compartmentalization and Pb efflux to the other environmental pollutants. Besides, given the success with apoplast (Figure 7). A. thaliana, our findings suggest a bright future for developing native plants to be used for phytoremediation strategies. Contribution of Other Genes Involved in Pb Tolerance DATA AVAILABILITY STATEMENT Most of the genes that have been identified in plants under Pb exposure are related to antioxidative enzymes, metal transporters The original contributions presented in the study are included in and Pb chelating compounds (Kumar and Prasad, 2018). Our the article/Supplementary Material or in the public repositories GWAs results were not confined into these categories. However, indicated in the Materials and Methods section. Further inquiries it is well known that metal tolerance in plants can be highly can be directed to the corresponding author/s. polygenic and therefore the enhancement of other genes may also contribute to the Pb tolerance observed in A. thaliana natural accessions. Pb entering the symplast is removed by AUTHOR CONTRIBUTIONS the activity of efflux pumps present in the plasma membrane. These transporters are constitutive, but their gene expression SB, ML, CP, and SM conceived the study. ML performed the is stimulated by Pb (Pourrut et al., 2013). The relative gene GWA study. LP-M and SB conducted the laboratory experiments expression of three plasma membrane ABC transporters (ACBP1, and data analyses. LP-M and SM performed the image analyses. PDR8, and PDR12) were assessed and we found that Pb exposure SB wrote the manuscript with primary input from all authors. All induced the transcription of these transporters especially in leaves authors edited and approved the final manuscript. of T-HLL plants. As noted by Jiang et al. (2017), these transporters may favor the active efflux of Pb followed by sequestration into inactive organelles or its transport to the cell exterior. FUNDING PSE1 encodes an unknown protein but was described for promoting Pb tolerance in A. thaliana plants by Fan et al. This research was funded by Spanish MICINN, grant number Sal- (2016). In this study, the authors showed that pse1-1 loss-of- CAL-MED (BFU2016- 75176-R) and IONPLAMIC (PID2019- function mutant was sensitive to Pb and the expression of 104000RB-100). ABC transporters was increased in PSE1-overexpressing plants subjected to Pb stress. Our expression analysis indicated that this gene was highly upregulated in the T-LLL accessions, supporting ACKNOWLEDGMENTS a role for it in Pb tolerance. Finally, we also evaluated the root expression of the vacuolar transporter HMA3. Morel et al. We would like to thank Marina Palmero from CREAF-UAB for (2009) suggested that HMA3 plays a role in Pb detoxification her help in GIS data analyses; Rosa Padilla for processing the by participating in their vacuolar sequestration. Pb exposure ionome data; Sharmila Gosth for her collaboration in laboratory significantly enhanced the expression of HMA3 in our T-LLL experiments; and Ümit Seren for his assistance in AraPheno accessions. It is possible that T-LLL plants take advantage of both and Gwa-Portal platforms. We are grateful to Microscopy strategies and the excess of Pb that is not deposited in the cell Service of the UAB for processing samples and assist in TEM walls gets trapped in the vacuole. images visualization. In summary, A. thaliana has evolved two main strategies to cope with high substrate Pb: (1) exclusion from the shoot by sequestration in the roots and (2) translocation to the shoots SUPPLEMENTARY MATERIAL and storage in inactive organelles or intracellular spaces. The first strategy involves upregulation of EXT18 and HMA3 in The Supplementary Material for this article can be found the roots favoring the thickening of the root cell walls and Pb online at: https://www.frontiersin.org/articles/10.3389/fpls.2021. vacuolar accumulation, improving Pb sequestration in roots and 689316/full#supplementary-material decreasing its toxicity (Figure 7). On the contrary, in plants Supplementary Figure 1 | Properties of the experimental soils. (A) Map of with strategy 2, the higher expression of ABC transporters may Catalonia indicating the location of the Pb/Zn mine where soil was excavated. (B) enhance the Pb translocation to the aerial tissues. However, Radial plot of normalized difference of 13 elements from control and mine soils. the upregulation of these transporters along with the TLC (C) Elemental composition and the pH of the control and mine soils. Data gene promotes efficient leaf Pb efflux to the apoplast and Pb represents the mean ± SE of six samples per soil type. Elements exhibiting significant differences are marked with an asterisk (∗ ). encapsulation (Figure 7). We believe these two strategies are not exclusive to A. thaliana and extensins and similar lipid-domains Supplementary Figure 2 | Growth and fitness of each A. thaliana accession should be explored in other plant species. cultivated under Pb stress conditions. Mean ± SE of (A) rosette diameter (cm) and (B) fitness (silique number) of each accession cultivated in a pot experiment in We conclude that the phenotypic and genomic variability control soil (gray) or in Pb-enriched mine soil (orange) (n = 8 plants per accession present in worldwide A. thaliana natural accessions is a valuable and soil type). Mean ± SE of (C) root length and (D) biomass (fresh weight, g) of resource to study molecular mechanism involved in Pb stress. each accession cultivated in hydroponic solution and treated with 0 µM (gray) or Frontiers in Plant Science | www.frontiersin.org 14 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis 100 µM of PbNO3 (orange) for 2 weeks (n = 6 plants per accession and Supplementary Table 1 | Significant SNPs information and associated genes. treatment). The score of each SNP was obtain from the GWA-Portal (AMM analysis) and from Tassel software (CMLM analysis). Supplementary Figure 3 | Ionomic profiles of each A. thaliana Pb-phenotype and ext18 and tlc mutants exposed to control and high Pb conditions. (A) Supplementary Table 2 | Primers list for transcript quantification of target genes. Radial plots of normalized differences of 12 elements from leaves of Supplementary Dataset 1 | HapMap cohort information and phenotypes 3 T-LLL, 3 T-HLL, 2 (S) accessions and Col-0 cultivated in a pot experiment analyzed of the 360 accessions cultivated in a Pb-enriched mine soil. in control soil (gray) or in Pb-enriched mine soil (orange). Radial plots of normalized differences of 12 elements from (B) leaf ionome and (C) roots Supplementary Dataset 2 | SNPs positions and scores of all the of 3 T-LLL, 3 T-HLL, 2 (S) accession and Col-0 cultivated in hydroponic phenotypes used for GWAs. solution and treated with 0 µM (gray) or 100 µM of PbNO3 (orange) for 2 weeks. Supplementary Dataset 3 | Statistical analysis of the pot and hydroponic experiments. Supplementary Figure 4 | TEM-EDX analysis of Pb deposits. (A) Example images of analyzed spectrums. (B) EDX weight and atomic ratios of spectrums Supplementary Dataset 4 | Statistical analysis of gene transcript analyses and focused in distinct areas. cell wall measurements. REFERENCES Fan, T., Yang, L., Wu, X., Ni, J., Jiang, H., Zhang, Q. A., et al. (2016). The PSE1 gene modulates lead tolerance in Arabidopsis. J. Exp. Bot. 67, 4685–4695. doi: Antonovics, J. (2006). Evolution in closely adjacent populations X: long term 10.1093/jxb/erw251 persistence of pre-reproductive isolation at a mine boundary. Heredity 97, Fusari, C. M., Kooke, R., Lauxmann, M. A., Annunziata, M. G., Enke, B., 33–37. doi: 10.1038/sj.hdy.6800835 Hoehne, M., et al. (2017). Genome-wide association mapping reveals that Bech, J., Duran, P., Roca, N., Poma, W., Sánchez, I., Roca-Pérez, L., et al. (2012). specific and pleiotropic regulatory mechanisms fine-tune central metabolism Accumulation of Pb and Zn in Bidens triplinervia and Senecio sp. spontaneous and growth in Arabidopsis. Plant Cell 29, 2349–2373. doi: 10.1105/tpc.17. species from mine spoils in Peru and their potential use in phytoremediation. 00232 J. Geochem. Explor. 123, 109–113. doi: 10.1016/j.gexplo.2012. Gomes, M. P., Nogueira, M. D. O. G., Castro, E. M. D., and Soares, ÂM. (2011). 06.021 Ecophysiological and anatomical changes due to uptake and accumulation of Bradbury, P. J., Zhang, Z., Kroon, D. E., Casstevens, T. M., Ramdoss, Y., heavy metal in Brachiaria decumbens. Sci. Agric. 68, 566–573. doi: 10.1590/ and Buckler, E. S. (2007). TASSEL: software for association mapping of S0103-90162011000500009 complex traits in diverse samples. Bioinformatics 23, 2633–2635. doi: 10.1093/ Gupta, D. K., Huang, H. G., and Corpas, F. J. (2013). Lead tolerance in plants: bioinformatics/btm308 strategies for phytoremediation. Environ. Sci. Pollut. Res. 20, 2150–2161. doi: Busoms, S., Paajanen, P., Marburger, S., Bray, S., Huang, X. Y., Poschenrieder, 10.1007/s11356-013-1485-4 C., et al. (2018). Fluctuating selection on migrant adaptive sodium transporter Horton, M. W., Hancock, A. M., Huang, Y. S., Toomajian, C., Atwell, S., Auton, alleles in coastal Arabidopsis thaliana. Proc. Natl. Acad. Sci. U. S. A. 115, A., et al. (2012). Genome-wide patterns of genetic variation in worldwide E12443–E12452. doi: 10.1073/pnas.1816964115 Arabidopsis thaliana accessions from the RegMap panel. Nat. Genet. 44, 212– Busoms, S., Teres, J., Huang, X. Y., Bomblies, K., Danku, J., Douglas, A., et al. 216. doi: 10.1038/ng.1042 (2015). Salinity is an agent of divergent selection driving local adaptation of Inoue, H., Fukuoka, D., Tatai, Y., Kamachi, H., Hayatsu, M., Ono, M., et al. (2013). Arabidopsis to coastal habitats. Plant Physiol. 168, 915–929. doi: 10.1104/pp.15. Properties of lead deposits in cell walls of radish (Raphanus sativus) roots. Int. 00427 J. Plant Res. 126, 51–61. doi: 10.1007/s10265-012-0494-6 Byers, H. L., McHenry, L. J., and Grundl, T. J. (2020). Increased risk for Jiang, L., Wang, W., Chen, Z., Gao, Q., Xu, Q., and Cao, H. (2017). A role for lead exposure in children through consumption of produce grown in APX1 gene in lead tolerance in Arabidopsis thaliana. Plant Sci. 256, 94–102. urban soils. Sci. Total Environ. 743:140414. doi: 10.1016/j.scitotenv.2020. doi: 10.1016/j.plantsci.2016.11.015 140414 Johnson, K. L., Jones, B. J., Schultz, C. J., and Bacic, A. (2003). “Non-enzymic Cannon, M. C., Terneus, K., Hall, Q., Tan, L., Wang, Y., Wegenhart, B. L., et al. cell wall (glyco) proteins,” in in Annual Plant Review vol 8 The Plant Cell (2008). Self-assembly of the plant cell wall requires an extensin scaffold. PNAS Wall, ed. J. K. C. Rose (Hoboken: John Wiley & Sons, Inc), 111–154. doi: 105, 2226–2231. doi: 10.1073/pnas.0711980105 10.1002/9781119312994.apr0070 Chao, D. Y., Silva, A., Baxter, I., Huang, Y. S., Nordborg, M., and Danku, J. (2012). Josè-Estanyol, M., and Puigdomènech, P. (2000). Plant cell wall glycoproteins and Genome−wide association studies identify heavy metal ATPase3 as the primary their genes. Plant Physiol. Biochem. 38, 97–108. doi: 10.1016/S0981-9428(00) determinant of natural variation in leaf cadmium in Arabidopsis thaliana. PLoS 00165-0 Genet. 8:e1002923. doi: 10.1371/journal.pgen.1002923 Kim, D. Y., Bovet, L., Kushnir, S., Noh, E. W., Martinoia, E., and Lee, Y. (2006). Choudhary, P., Saha, P., Ray, T., Tang, Y., Yang, D., and Cannon, M. C. AtATM3 is involved in heavy metal resistance in Arabidopsis. Plant Physiol. (2015). EXTENSIN18 is required for full male fertility as well as normal 140, 922–932. doi: 10.1104/pp.105.074146 vegetative growth in Arabidopsis. Front. Plant Sci. 6:553. doi: 10.3389/fpls.2015. Kim, D. Y., Bovet, L., Maeshima, M., Martinoia, E., and Lee, Y. (2007). 00553 The ABC transporter AtPDR8 is a cadmium extrusion pump conferring Chudzik, B., Szczuka, E., Leszczuk, A., and Strubińska, J. (2018). Modification of heavy metal resistance. Plant J. 50, 207–218. doi: 10.1111/j.1365-313X.2007. pectin distribution in sunflower (Helianthus annuus L.) roots in response to 03044.x lead exposure. Environ. Exp. Bot. 155, 251–259. doi: 10.1016/j.envexpbot.2018. Kopittke, P. M., Asher, C. J., Blamey, F. P. C., Auchterlonie, G. J., Guo, Y. N., 07.005 and Menzies, N. W. (2008). Localization and chemical speciation of Pb in Clark, H. F., Brabander, D. J., and Erdil, R. M. (2006). Sources, sinks, and exposure roots of signal grass (Brachiaria decumbens) and Rhodes grass (Chloris gayana). pathways of lead in urban garden soil. J. Environ. Qual. 35, 2066–2074. doi: Environ. Sci. Technol. 42, 4595–4599. doi: 10.1021/es702627c 10.2134/jeq2005.0464 Kopittke, P. M., Asher, C. J., Kopittke, R. A., and Menzies, N. W. (2007). Toxic Cruz-Valderrama, J. E., Gómez-Maqueo, X., Salazar-Iribe, A., Zúñiga-Sánchez, effects of Pb2+ on growth of cowpea (Vigna unguiculata). Environ. Pollut. 150, E., Hernández-Barrera, A., Quezada-Rodríguez, E., et al. (2019). Overview of 280–287. doi: 10.1016/j.envpol.2007.01.011 the role of cell wall DUF642 proteins in plant development. IJMS 20:3333. Krzesłowska, M., Lenartowska, M., Samardakiewicz, S., Bilski, H., and Woz’ny, A. doi: 10.3390/ijms20133333 (2010). Lead deposited in the cell wall of Funaria hygrometrica protonemata Davies, L., and Gather, U. (1993). The identification of multiple outliers. JASA 88, is not stable—a remobilization can occur. Environ. Pollut. 158, 325–338. doi: 782–792. doi: 10.2307/2290763 10.1016/j.envpol.2009.06.035 Frontiers in Plant Science | www.frontiersin.org 15 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis Krzesłowska, M., Rabȩda, I., Basińska, A., Lewandowski, M., Mellerowicz, Platt, A., Horton, M., Huang, Y. S., Li, Y., Anastasio, A. E., and Mulyati, N. W. E. J., Napieralska, A., et al. (2016). Pectinous cell wall thickenings (2010). The scale of population structure in Arabidopsis thaliana. PLoS Genet. formation–A common defense strategy of plants to cope with 6:e1000843. doi: 10.1371/journal.pgen.1000843 Pb. Environ. Pollut. 214, 354–361. doi: 10.1016/j.envpol.2016. Pourrut, B., Shahid, M., Douay, F., Dumat, C., and Pinelli, E. (2013). “Molecular 04.019 mechanisms involved in lead uptake, toxicity and detoxification in higher Kumar, A., and Prasad, M. N. V. (2018). Plant-lead interactions: transport, toxicity, plants,” in Heavy Metal Stress In Plants eds D. Gupta, F. Corpas, J. tolerance, and detoxification mechanisms. Ecotoxicol. Environ. Saf. 166, 401– Palma (Berlin, Heidelberg: Springer). 121–147. doi: 10.1007/978-3-642-384 418. doi: 10.1016/j.ecoenv.2018.09.113 69-1_7 Lamport, D. T., Kieliszewski, M. J., Chen, Y., and Cannon, M. C. Pourrut, B., Shahid, M., Dumat, C., Winterton, P., and Pinelli, E. (2011). Lead (2011). Role of the extensin superfamily in primary cell wall uptake, toxicity, and detoxification in plants. Rev. Environ. Contam. Toxicol. architecture. Plant Physiol. 156, 11–19. doi: 10.1104/pp.110. 213, 113–136. doi: 10.1007/978-1-4419-9860-6_4 169011 Qi, X., Behrens, B. X., West, P. R., and Mort, A. J. (1995). Solubilization Lee, M., Lee, K., Lee, J., Noh, E. W., and Lee, Y. (2005). AtPDR12 contributes to and partial characterization of extensin fragments from cell walls of lead resistance in Arabidopsis. Plant Physiol. 138, 827–836. doi: 10.1104/pp.104. cotton suspension cultures (evidence for a covalent cross-link between 058107 extensin and pectin). Plant Physiol. 108, 1691–1701. doi: 10.1104/pp.108. Levin-Schwartz, Y., Henn, B. C., Gennings, C., Coull, B. A., Placidi, D., 4.1691 Horton, M. K., et al. (2021). Integrated measures of lead and manganese Ratul, A. K., Hassan, M., Uddin, M. K., Sultana, M. S., Akbor, M. A., and Ahsan, exposure improve estimation of their joint effects on cognition in Italian M. A. (2018). Potential health risk of heavy metals accumulation in vegetables school-age children. Environ. Int. 146:106312. doi: 10.1016/j.envint.2020. irrigated with polluted river water. Int. Food Res. J. 25, 329–338. 106312 Reboredo, F., Simões, M., Jorge, C., Mancuso, M., Martinez, J., Guerra, M., Lipka, A. E., Tian, F., Wang, Q., Peiffer, J., Li, M., Bradbury, P. J., et al. (2012). et al. (2019). Metal content in edible crops and agricultural soils due to GAPIT: genome association and prediction integrated tool. Bioinformatics 28, intensive use of fertilizers and pesticides in Terras da Costa de Caparica 2397–2399. doi: 10.1093/bioinformatics/bts444 (Portugal). Environ. Sci. Pollut. Res. 26, 2512–2522. doi: 10.1007/s11356-018- Liu, J., Ma, X., Wang, M., and Sun, X. (2013). Genotypic differences among 3625-3 rice cultivars in lead accumulation and translocation and the relation with Riedelsheimer, C., Brotman, Y., Méret, M., Melchinger, A. E., and Willmitzer, L. grain Pb levels. Ecotoxicol. Environ. Saf. 90, 35–40. doi: 10.1016/j.ecoenv.2012. (2013). The maize leaf lipidome shows multilevel genetic control and high 12.007 predictive value for agronomic traits. Sci. Rep. 3:2479. doi: 10.1038/srep02479 Livak, K. J., and Schmittgen, T. D. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2−11CT method. Methods 25, Roberts, K., and Shirsat, A. H. (2006). Increased extensin levels in Arabidopsis 402–408. doi: 10.1006/meth.2001.1262 affect inflorescence stem thickening and height. J. Exp. Bot. 57, 537–545. doi: Luo, X., Bing, H., Luo, Z., Wang, Y., and Jin, L. (2019). Impacts of atmospheric 10.1093/jxb/erj036 particulate matter pollution on environmental biogeochemistry of trace metals Sailer, C., Babst-Kostecka, A., and Fischer, M. C. (2018). Transmembrane transport in soil-plant system: a review. Environ. Pollut. 255:113138. doi: 10.1016/j. and stress response genes play an important role in adaptation of Arabidopsis envpol.2019.113138 halleri to metalliferous soils. Sci. Rep. 8:16085. doi: 10.1038/s41598-018-33 Malone, C., Koeppe, D. E., and Miller, R. J. (1974). Localization of lead 938-2 accumulated by corn plants. Plant Physiol. 53, 388–394. doi: 10.1104/pp.53. Seren, Ü (2018). GWA-Portal: genome-wide association studies made easy. 3.388 Methods Mol. Biol. 1761, 303–319. doi: 10.1007/978-1-4939-7747-5_22 Mareri, L., Romi, M., and Cai, G. (2019). Arabinogalactan proteins: actors or Sinclair, R., Rosquete, M. R., and Drakakaki, G. (2018). Post-Golgi trafficking and spectators during abiotic and biotic stress in plants? Plant Biosyst. 153, 173–185. transport of cell wall components. Front. Plant Sci. 9:1784. doi: 10.3389/fpls. doi: 10.1080/11263504.2018.1473525 2018.01784 Marzol, E., Borassi, C., Bringas, M., Sede, A., Garcia, D. R. R., Capece, L., et al. Smith, S. R. (2009). A critical review of the bioavailability and impacts of heavy (2018). Filling the gaps to solve the extensin puzzle. Mol. Plant 11, 645–658. metals in municipal solid waste composts compared to sewage sludge. Environ. doi: 10.1016/j.molp.2018.03.003 Int. 35, 142–156. doi: 10.1016/j.envint.2008.06.009 McKenna, A., Hanna, M., Banks, E., Sivachenko, A., Cibulskis, K., Kernytsky, Sumranwanich, T., Boonthaworn, K., and Singh, A. (2018). The Roles of Plant Cell A., et al. (2010). The Genome Analysis Toolkit: a MapReduce framework for Wall as the First-line Protection Against Lead (Pb) Toxicity. Appl. Sci. Eng. Prog. analyzing next-generation DNA sequencing data. Genome Res. 20, 1297–1303. 11, 239–245. doi: 10.14416/j.ijast.2018.09.003 doi: 10.1101/gr.107524.110 Tan, L., Pu, Y., Pattathil, S., Avci, U., Qian, J., Arter, A., et al. (2014). Changes in Meyers, D. E., Kopittke, P. M., Auchterlonie, G. J., and Webb, R. I. cell wall properties coincide with overexpression of extensin fusion proteins (2009). Characterization of lead precipitate following uptake by roots of in suspension cultured tobacco cells. PLoS One 9:115906. doi: 10.1371/journal. Brassica juncea. Environ. Toxicol. Chem. 28, 250–255. doi: 10.1897/09- pone.0115906 131.1 Uzu, G., Sobanska, S., Sarret, G., Munoz, M., and Dumat, C. (2010). Foliar lead Morel, M., Crouzet, J., Gravot, A., Auroy, P., Leonhardt, N., Vavasseur, A., uptake by lettuce exposed to atmospheric fallouts. Environ. Sci. Techno. 44, et al. (2009). AtHMA3, a P1B-ATPase allowing Cd/Zn/Co/Pb vacuolar 1036–1042. doi: 10.1021/es902190u storage in Arabidopsis. Plant Physiol. 149, 894–904. doi: 10.1104/pp.108. Wei, G. U. O., and Shirsat, A. H. (2006). Extensin over−expression in Arabidopsis 130294 limits pathogen invasiveness. Mol. Plant Pathol 7, 579–592. doi: 10.1111/j.1364- Ng, C. C., Boyce, A., Rahman, M., and Abas, R. (2016). Phyto-assessment of soil 3703.2006.00363.x heavy metal accumulation in tropical grasses. J. Anim. Plant Sci. 26, 686–696. Wierzbicka, M. H., Przedpełska, E., Ruzik, R., Ouerdane, L., Połec-Pawlak, K., doi: 10.1177/1178622118777763 Jarosz, M., et al. (2007). Comparison of the toxicity and distribution of cadmium Panagos, P., Van Liedekerke, M., Jones, A., and Montanarella, L. (2012). European and lead in plant cells. Protoplasma 231, 99–111. doi: 10.1007/s11270-017- Soil Data Centre: response to European policy support and public data 3488-0 requirements. Land Use Policy 29, 329–338. doi: 10.1016/j.landusepol.2011.07. Winter, E., and Ponting, C. P. (2002). TRAM, LAG1 and CLN8: members of 003 a novel family of lipid-sensing domains? Trends Biochem. Sci. 27, 381–383. Pereira, C. S., Ribeiro, J. M., and Vatulescu, A. D. (2011). Extension doi: 10.1016/S0968-0004(02)02154-0 network formation in Vitis vinifera callus cells is an essential and Wisniewski, J. P., Cornille, P., Agnel, J. P., and Montillet, J. L. (1999). The extensin causal event in rapid and H2 O2 − induced reduction in primary cell multigene family responds differentially to superoxide or hydrogen peroxide wall hydration. BMC Plant Biol. 11:106. doi: 10.1186/1471-2229-11- in tomato cell cultures. FEBS Lett. 447, 264–268. doi: 10.1016/S0014-5793(99) 106 00315-4 Frontiers in Plant Science | www.frontiersin.org 16 August 2021 | Volume 12 | Article 689316 Busoms et al. Lead Tolerance Strategies in Arabidopsis Xiao, S., Gao, W., Chen, Q. F., Ramalingam, S., and Chye, M. L. (2008). Environ. Sci. Pollut. Res. 23, 21794–21804. doi: 10.1007/s11356-016- Overexpression of membrane−associated acyl−CoA−binding protein ACBP1 7362-1 enhances lead tolerance in Arabidopsis. Plant J. 54, 141–151. doi: 10.1111/j. 1365-313X.2008.03402.x Conflict of Interest: The authors declare that the research was conducted in the Yousaf, B., Liu, G., Wang, R., Imtiaz, M., Zia-ur-Rehman, M., Munir, absence of any commercial or financial relationships that could be construed as a M. A. M., et al. (2016). Bioavailability evaluation, uptake of heavy potential conflict of interest. metals and potential health risks via dietary exposure in urban-industrial areas. Environ. Sci. Pollut. Res. 23, 22443–22453. doi: 10.1007/s11356-016- Publisher’s Note: All claims expressed in this article are solely those of the authors 7449-8 and do not necessarily represent those of their affiliated organizations, or those of Yuan, Y., Xiang, M., Liu, C., and Theng, B. K. G. (2019). Chronic impact of an the publisher, the editors and the reviewers. Any product that may be evaluated in accidental wastewater spill from a smelter, China: a study of health risk of this article, or claim that may be made by its manufacturer, is not guaranteed or heavy metal(loid)s via vegetable intake. Ecotoxicol. Environ. Saf. 182:109401. endorsed by the publisher. doi: 10.1016/j.ecoenv.2019.109401 Zhang, F., Xiao, X., Xu, K., Cheng, X., Xie, T., Hu, J., et al. (2020). Genome-wide Copyright © 2021 Busoms, Pérez-Martín, Llimós, Poschenrieder and Martos. This association study (GWAS) reveals genetic loci of lead (Pb) tolerance during is an open-access article distributed under the terms of the Creative Commons seedling establishment in rapeseed (Brassica napus L.). BMC Gen. 21:139. doi: Attribution License (CC BY). The use, distribution or reproduction in other forums 10.1186/s12864-020-6558-4 is permitted, provided the original author(s) and the copyright owner(s) are credited Zhou, C., Huang, M., Li, Y., Luo, J., and Ping-Cai, L. (2016). Changes and that the original publication in this journal is cited, in accordance with accepted in subcellular distribution and antioxidant compounds involved academic practice. No use, distribution or reproduction is permitted which does not in Pb accumulation and detoxification in Neyraudia reynaudiana. comply with these terms. Frontiers in Plant Science | www.frontiersin.org 17 August 2021 | Volume 12 | Article 689316