Multiscale and age-dependent leaf nickel in the Ni-hyperaccumulator Leptoplax emarginata
- 108 Downloads
Nickel-hyperaccumulator plants are of interest due to their potential use in agromining. We aimed to characterize leaf traits and Ni concentration variabilities occurring between individual plants, leaves of differing age or between various leaf tissues, in a single Greek population of the Ni-hyperaccumulator Leptoplax emarginata (Boiss.) O.E. Schulz. We linked these results to ecophysiological characteristics and other element concentrations at leaf and leaf tissue scales. We measured leaf gas exchanges, stomatal density, and we carried out rapid freezing and freeze-drying processes on leaf sections before microanalysis with scanning electron microscopy and energy-dispersive spectrometry. Leaf or leaf-tissue Ni concentrations were influenced by a combination of individual plant and leaf age factors. The greatest Ni concentrations were found in the highly transpiring young and thin leaves with the greatest stomatal densities. Indeed, Ni was statistically seven times more concentrated in both epidermis layers than in their bulk neighbour leaf counterparts, whatever the leaf age. In both epidermis layers, increases in the Ni–S and Mg–S correlations from the oldest leaves to the youngest ones were observed. The Mg:Ni, Ca:Ni and P:Ni mass ratios decreased from the oldest leaves to the mature leaves. We would recommend time-series characterization of leaf traits belonging to at least three plant replicates in order to take into account the allogamous character of many Ni-hyperaccumulator plants. Long-distance Ni transport via the xylem is predominant in the Ni-hyperaccumulator L. emarginata whereas a complementary redistribution via the phloem should also occur. The perspectives of this study are the validation and refinement of this process over shorter periods using relevant tracers.
KeywordsNickel hyperaccumulation Age-dependent leaves Age-dependent leaf-tissues Allogamy Transpiration
Most metal-hypertolerant plants (e.g. Ernst et al. 2008) are non-accumulating excluders, limiting the root-to-shoot metal translocation via the xylem (Baker and Brooks 1989). In contrast, a minority of plants growing on metalliferous soils accumulate metals in their leaves to such high levels (e.g. see reviews by Verbruggen et al. 2009 and Krämer 2010) that they are called hyperaccumulators (Brooks et al. 1977). These are plants of interest due to their potential use in the phytoremediation of toxic metals, such as cadmium (e.g. McGrath et al. 2002; Koopmans et al. 2007; Tang et al. 2012), or in the phytomining or agromining of precious metals such as nickel (Nick and Chambers 1995; Li et al. 2003, Robinson et al. 2003a; Chaney et al. 2007; Bani et al. 2015; van der Ent et al. 2015).
At the shoot, leaf and leaf-tissue scales, Ni detoxification in Ni-hyperaccumulator plants has mainly been attributed to (1) Ni complexation with carboxylate ligands (e.g. Montargès-Pelletier et al. 2008; McNear et al. 2010; van der Ent et al. 2017) and (2) Ni sequestration in both epidermis layers (walls and/or vacuoles) instead of the mesophylls (Krämer et al. 2000; Küpper et al. 2001; Robinson et al. 2003b; Kachenko et al. 2008a, 2008b). As for Ni-hyperaccumulators from the Brassicaceae family (Montargès-Pelletier et al. 2008), it was demonstrated that malate is responsible for Ni tolerance in the non-accumulator Arabidopsis thaliana (Agrawal et al. 2012) and in the phloem sap of the hyperaccumulator Noccaea caerulescens (Deng et al. 2016). In general, both citrate and malate seem to be involved in Ni chelation in the majority of Ni-hyperaccumulators (van der Ent et al. 2017). These detoxification mechanisms also occur for other metals (Küpper et al. 2004; Cosio et al. 2005; Freeman et al. 2006; Xu et al. 2006; Vogel-Mikus et al. 2008; Tian et al. 2009; Fernando et al. 2010). The preferential storage in the epidermis suggests that Ni, as other metals, is transported along with the transpiration stream, as also indicated by its intense localization in the leaf marginal areas and edges, and in the leaf vascular bundles and veins (Mesjasz-Przybylowicz et al. 2001; Kachenko et al. 2008b; Mesjasz-Przybylowicz and Przybylowicz 2011). Also, phloem redistribution of Ni seems to be a common process in Ni hyperaccumulating Brassicaceae that explains why flowers and seeds are so rich in Ni (Estrade et al. 2015; Deng et al. 2016). These processes could also show a more complex behaviour of Ni in leaves, in which the sink might not be an irreversible one.
Huge leaf Ni concentration variabilities occur between individual plants in a single population of a considered Ni-hyperaccumulator plant (Nick and Chambers 1995; Boyd et al. 1999; Boyd and Jaffré 2009; Richau and Schat 2009) or between branches and leaves in an individual metal-hyperaccumulator plant (Boyd et al. 1999; Perronnet et al. 2003; Robinson et al. 2003b). However, leaf-tissue Ni analyses were often carried out on a single leaf and rarely on an age-dependent leaf series. Moreover, this leaf or these leaves were always sampled from a single individual hyperaccumulator plant. The leaf-age effect was studied for 11 New-caledonian woody Ni-hyperaccumulator species and no real variation in Ni concentration in leaves (Boyd and Jaffré 2009). Only on Geissois pruinosa and Homalium kanaliense—two hypernickelophore species (i.e. species that accumulate more than 1% Ni in their aboveground biomass as defined by Jaffré and Schmid 1974)—there were significant variations, which did not seem to correlate with leaf age. The leaf trait approach for metal hyperaccumulators has been developed recently (Lange et al. 2017a, b). Following such approaches, they found that functional leaf traits such as the specific leaf area (SLA) seemed to exhibit a homeostatic response to Co. We therefore wanted to specifically investigate this aspect to better understand the impact of physiological leaf traits (transpiration and leaf growth) on Ni hyperaccumulation in leaves in a herbaceous semi-perennial hipernickelophore from the Brassicaceae family, Leptoplax emarginata.
Thus, the purpose of the present study was threefold: (1) to study the respective impact of individual plant replicates and leaf age on leaf or leaf-tissue Ni concentration for the Ni-hyperaccumulator L. emarginata cultivated on a fertilized and Ni-contaminated sandy topsoil; (2) to link leaf or leaf-tissue Ni concentration to leaf ecophysiological characteristics. We sought to validate the hypothesis that leaf or leaf-tissue Ni concentrations are highest in young leaves, where leaf biomasses are lowest and transpiration rates at their highest; and (3) to study, in the leaf tissues, the relationships between Ni concentration and concentrations of key nutriments (Mg, Ca, P, K), which are also relevant physiologic tracers.
Materials and methods
Plants, porous media and solutions
The hyperaccumulator plant selected was the Ni-hyperaccumulator Leptoplax emarginata (Boiss.) O.E. Schulz (Brassicaceae), endemic to serpentine soils in Greece (Reeves et al. 1980; Bani et al. 2009; Cecchi et al. 2010). Seeds of L. emarginata were sampled in July 2006 in the Trigona village, Pindus Mountains, Central Greece (830 m a.s.l.) (Bani et al. 2009). These were removed from their siliques before use. The mean Ni content of a single seed without its outer shell was 16.4 μg per air-dried seed. The porous media were (1) a fertilized calibrated sedimentary sand (particle size distribution ranging from 0.4 to 0.8 mm, Sibelco, France) from Hostun, South-Eastern France for the pre-cultures and (2) a fertilized and Ni-contaminated sandy topsoil from a cultivated podzol (Food and Agriculture Organization of the United Nations (FAO) 2001), South-Western France, for the cultures. More details are available in Coinchelin et al. (2012).
The porous media were both supplied with a nutritive solution containing 998 μM Ca(NO3)2, 4H2O; 823 μM MgSO4, 7H2O; 3570 μM NH4NO3; 2580 μM KH2PO4; 71.6 μM FeSO4, 7H2O; 9.1 μM MnSO4, H2O; 4.59 μM ZnSO4, 7H2O; 9.25 μM H3BO3, 4H2O; 0.157 μM CuSO4, 5H2O; 0.104 μM Na2MoO4, 2H2O. For the sandy topsoil, Ni contamination (NiSO4 7H2O) in the initial fertilizing solution was 1700 μM Ni but this initial Ni concentration greatly reduced to 10.3 ± 0.4 μM Ni (pH 6.44 ± 0.02) after 1 week of soil solution ageing in the medium and just before plant transplantation (Coinchelin et al. 2012).
Each pot was lined with a polyethylene bag to prevent leakage and filled with 1.37 kg (1.5 l pot) or 0.91 kg (1 l pot) of sandy topsoil, and 1 week before the culture this was gently mixed with, respectively, 200 or 133 ml of the fertilizing solution or the fertilizing and Ni-contaminated solution, in order to obtain a volumetric porous medium solution of 0.2 cm3 cm−3. A white polyethylene film covered the top of each pot, with a single 1-cm diameter hole for all the planted and unplanted pots.
Plant growth, gas exchanges and plant samplings
Leptoplax emarginata seeds were germinated in the fertilized sand for a 2-week pre-culture, except for the cultures for 35 days in 1 l pots (pre-culture of 5 weeks). The growing conditions were as follows: a 16-h photoperiod, with a photon flux density of 325 μmol photons m−2 s−1 in the Parabolic Anodized Reflector (PAR) range, 20 °C day/18 °C night temperatures and 50% relative humidity. The daily vapour pressure deficit was 1.17 kPa.
For the 1 l pots, the culture time was 35 days both for those plant replicates cultivated in the reference fertilized sandy topsoil, and those in the fertilized and Ni-contaminated sandy topsoil (two plant replicates per treatment). For the 1.5 l pots, the culture time was 43 days (P1, P2 and P3 plant replicates) or 51 days (P4, P5 and P6 plant replicates) for the plants cultivated in the fertilized and Ni-contaminated sandy topsoil (three plant replicates per culture age: Fig. S1). The leaf age referred to was always the starting time of each culture (30 March 2009), all the dated leaves being developed thereafter. We established the relationship between leaf position and age, in order to sample one of the few very old (51 day-old) leaves, which appeared before Ni contamination, and representative leaves which appeared after Ni contamination: the old (43 day-old), mature (26 day-old) and young (16 day-old) leaves. These sampled leaves were used for scanning electron microscopy (SEM) and energy-dispersive spectrometry (EDS) analyses. For this, the leaves which appeared on April 1, 7, 13, 24 April and May 4 2009 were well-identified, with complementary observations on the leaf development throughout the culture.
The plants were irrigated daily with deionized water, to maintain the volumetric porous medium solution of 0.2 cm3 cm−3 constant throughout the culture, whilst assuming a sufficient soil buffer power favoring a constant Ni concentration in the solution.
The day before each sampling period, leaf gas exchanges were recorded on 55% of the leaves of the three plant replicates cultivated for 43 days (50, 55 and 60% for the P1, P2 and P3 plant replicates, respectively) and on 44% of the three plant replicates cultivated for 51 days (47, 38 and 47% for the P4, P5 and P6 plant replicates, respectively), using a CIRAS-1 portable photosynthesis system (PP SYSTEMS Inc., USA). The photon flux density and temperature inside the leaf chamber delivered to the leaf area measured were normalized using a lamp (at 260 μmol photons m−2 s−1 and temperature of 24.3 °C).
Plant biomass and leaf traits
Roots were collected and washed with deionized water in order to remove all the root-adhering particles. Leaf moisture, stem biomass, root biomass, as well as total and individual leaf biomass (LB) were determined after oven-drying at 70 °C till attaining a constant weight. The total and individual leaf area (LA) were determined using the WinFOLIA® Software, the total or individual specific leaf area (SLA = LA to LB ratio) being calculated thereafter. These parameters were systematically determined on the three plant replicates cultivated for 43 days, but not on the other plant replicates. Each oven-dried single leaf of the three plant replicates cultivated for 43 days and each of the leaves neighbouring those sampled for the SEM–EDS analyses was carefully crushed using an agate mortar to a 0.5 g aliquot. They were then dissolved and mineralized by a HNO3:H2O2 solution (1:1 v/v) in open small reactors. The mixture was then heated in a microwave oven (Mars 5, CEM Corporation Inc., USA). The Ni contents in resulting solutions were determined by inductively coupled plasma optical emission spectrometry (ICP-OES, Liberty RL, Varian, Inc., USA).
Age-dependent SEM-image analyses of stomatal density
Twelve leaf discs of 1 cm2 were harvested on the very old (51 day-old), old (43 day-old), mature (26 day-old) and young (16 day-old) leaves to record stomatal density. They were immediately frozen in liquid nitrogen (van der Ent et al. 2017) and stored at − 80 °C thereafter. Because leaves are amphistomatous, each disc was split for the separate analysis of the leaf’s adaxial and abaxial sides. Sample discs were then stuck to aluminium stubs on a Peltier stage (− 50 °C), before being examined under a controlled-pressure scanning electron microscope (model 1450VP, Leo, Cambridge, UK; 40 Pa inside the chamber, accelerating voltage 15 kV, working distance 8 mm). Nine microphotographs at × 300 were then taken of each sub-disc (upper or lower epidermis) and the number of stomata were counted.
Age-dependent leaf-tissue SEM–EDS microanalyses on freeze-dried leaf sections
Absolute concentrations of Ni and other elements cannot be measured directly using energy-dispersive spectrometry (EDS) on fully hydrated, fractured surfaces. In addition, the careful preparation of external standards is required, using the typical cocktail of elements likely to be found in the growing fresh leaf. Moreover, to acquire the most meaningful data from EDS systems, specimens should ideally be flat, polished and homogeneous. It follows that fractured, frozen or freeze-dried biological specimens are unable to meet these criteria. However, reasonable measures in instrument set-up and careful specimen preparation can go some way to satisfying the first two requirements. The material to be analysed was prepared according to the method used in Le Thiec et al. (1994). Sections of plant material were examined under a scanning electron microscope (model 1450VP, Leo, Cambridge, UK) at 15 kV, equipped with a dispersive energy microanalysis system (EDX, diode Si–Li; the INCA Energy 300 software, Oxford Instruments). The Point and Identify Module was used to target cell types across the freeze-dried surface. The spectra acquired are semi-quantitative results based on calculations using the system’s internal standards for each element and can be used to compare relative concentrations. The upper epidermis, palissadic parenchyma, lacunar (spongy) parenchyma and lower epidermis were analysed in the different-aged leaves. Nine analysis replicates per tissue were carried out.
Treatment- and leaf age-dependent SEM–EDS–WDS microanalyses of ground leaf pellets
Ground leaf pellets were put on an aluminium stub and coated with carbon. We used the energy-dispersive spectrometry (EDS) system for sulfur, magnesium, calcium, potassium and phosphorus; for Ni we used the wavelength-dispersive spectrometry (WDS) system with a pure metallic Ni as reference. Both EDS and WDS were linked to the same scanning electron microscope. The analyses were performed at 20 kV, sample current 50 nA, time 30 s. Six analyses replicates per leaf pellet were carried out.
Statistical analyses and curve-fitting were carried out using the XLSTAT 2010 Excel package software or the KaleidaGraph™ 3.52 software package. Principal component analysis (PCA) and Spearman correlation matrices (because some Ni or nutrient concentration distributions were non-normal) were carried out using the XLSTAT 2010 Excel software package. The respective effect of individual plant and leaf age, and their interaction (type III two-way ANOVA analysis) on the studied parameters were tested by performing a two-way ANOVA analysis (and the Tukey’s HSD post hoc test with α = 0.05 and P < 0.05), using also the XLSTAT 2010 Excel software package.
Age-dependent ecophysiological leaf traits
Stomata (Fig. S3) occurred on both sides of the leaf epidermis (Fig. S4). A positive stomatal density gradient occurred from the oldest leaves to the youngest ones, with a significantly greater stomatal density for the young leaves than for the other leaves whatever the plant replicate (Fig. S4; Fig. 2d). Stomatal density was similar for the upper epidermis and the lower epidermis and an individual plant effect on stomatal density was only identified for the mature leaves. Age-dependent distributions of both stomatal density (Fig. 2d) and leaf transpiration flux (Fig. 2c) were fitted by similar sigmoidal curves.
Age-dependent leaf Ni and nutrient concentrations
The same treatment- and leaf age-dependent trends occurred between the leaf Ni contents determined by SEM–WDS analysis on ground leaf pellets leaves and those determined by ICP–OES after mineralization of the same leaves (Table S2). However, for those plants cultivated in the fertilized and Ni-contaminated topsoil, the leaf Ni contents determined by SEM–WDS analysis were overestimated. Leaf Ca or Mg contents were constant, whatever the leaf age or the treatment, except for the old leaves, which were poorer in Mg when the fertilizing solution was contaminated with Ni (Table S2). Consequently, the mean Ca:Mg mass ratios of 7.8–9.8 were similar. The young leaves were significantly richer in S than the old leaves, whatever the treatment, whereas they were richer in Ni and poorer in P when the fertilizing solution was contaminated with Ni (Table S2). The variabilities of leaf Ni, S and P concentrations were explained by a combination of leaf age and treatment effects. The mean Ca:Ni, K:Ni and P:Ni mass ratios were significantly greater for the old leaves (mean Ca:Ni, K:Ni and P:Ni mass ratios of 11.4, 19.3 and 4.3, respectively) than for the young leaves (mean Ca:Ni, K:Ni and P:Ni mass ratios of 9.2, 9.5 and 1.6, respectively).
Thus, the ecophysiological characteristics, order of magnitude and age-dependent trends of leaf Ni concentrations were similar for the leaves of both the 43 day-old plants and the 51 day-old plants (Fig. 1g, e). Leaf Ni concentration was always significantly greater for the young leaves (leaf Ni concentration of 1.38 ± 0.16 or 1.25 ± 0.13 mg Ni g−1 for the 43 or 51 day-old plant replicates, respectively), than for the mature leaves (leaf Ni concentration of 1.00 ± 0.06 or 0.89 ± 0.12 mg Ni g−1 for the 43 day-old or 51 day-old plant replicates, respectively), with combined leaf age and individual plant effects. The age-dependent distribution of leaf Ni concentration was significantly (P < 0.01) fitted by a downward bell curve for the 43 day-old plant replicates (Fig. 1 g), whereas it was significantly (P < 0.001) fitted by a power law curve for the 51 day-old plant replicates (Fig. 2e). Ni concentration of the old leaves was significantly lower for the 51 day-old plant replicates than for the 43 day-old plant replicates.
Age-dependent leaf-tissue nickel and nutrient concentrations
In the whole tissue series, the spatial distribution of Ca concentration in leaves was mostly opposite to Ni, S or Mg concentrations, with parenchymas being richer than their epidermis in Ca (Fig. 4; Figs. S6 and S7). Consequently, almost no spatial Ca–Ni, Ca–S and Ca–Mg correlations were observed within the tissue series. Indeed, negative Ca–Ni and Ca–S correlations were visible for two of the three young leaf replicates (Fig. 5d). Ni and Ca concentrations in the two parenchymas were significantly (P < 0.001) positively correlated, except in the case of the young leaves. However, the Ni–Ca correlation was less significant or did not exist for either of the epidermis (Tables S4, S5, S6 and S7).
Principal-component analysis showed the wide variations in element concentrations occurring between the three plant replicates. The most visible differences were between both epidermis layers and both parenchymas, upper and lower (Figs. S8a, S8c and S8e). This analysis also revealed a difference between, on the one hand, the young and mature leaves and, on the other hand the old and very old leaves (Figs. S8b, S8d and S8f). For the P4 and P6 plant replicates, the main contributions for the first axis were the leaf tissue Ni, S and Mg mass concentrations (17.6–18.1, 17.2–25.3 and 27.5–29.2%, respectively). In contrast, the key contributions for the second axis were always the leaf tissue K and P mass concentrations (21.8–52.9 and 15.6–29.7%, respectively).
Phenotypic-based variability of leaf or leaf-tissue Ni concentration
For the Ni-hyperaccumulator L. emarginata, variations in leaf or leaf-tissue Ni concentrations were always explained by a combination of individual plant and leaf age effects, with interactions between these two main factors. Variations in leaf transpiration flux, WUE, leaf moisture, leaf-tissue S, Mg, Ca, K and P concentrations were also explained by a combination of individual plant and leaf age effects, except for the upper epidermis, the palissadic parenchyma Mg concentrations, and for the lower epidermis K concentrations. These wide phenotypic-based variations were attributed to the outcrossing (allogamous) fertilisation system of the Ni-hyperaccumulator L. emarginata (formerly Peltaria emarginata) previously determined by Peer et al. (2006). Leptoplax emarginata is also biannual, with 2n = 14 chromosomes and a genome size (2C) of 1.8–2.0, about three times greater than most other metal hyperaccumulating plants such as N. caerulescens (formerly Thlaspi caerulescens), Noccaea fendleri (formerly Thlaspi montanum) or Arabidopsis halleri (Peer et al. 2003, 2006).
Underlying mechanisms of age-dependent leaf Ni accumulation
In this study, leaf Ni concentration values were 1.38–1.25 mg Ni g−1 (young leaves), whereas for the same population of L. emarginata collected at seeding stage in the field (on Ni-rich ultramafic soils in Trigona, Pindus mountains, Greece), leaf Ni concentration was ten times greater (leaf Ni concentration of 13.6 ± 0.1 mg Ni g−1, Bani et al. 2009). These results highlight that the classical leaf-tissue Ni analysis carried out on a single leaf belonging to a single individual plant is open to question. There is a need for interpretation of leaf or leaf-tissue concentrations in the light of leaf traits as it has already been pointed out for Co hyperaccumulators (Lange et al. 2017b).
Our objectives were achieved through relevant methodologies, i.e. the use of rapid freezing and freeze-drying of leaf samples before energy-dispersive spectrometry (EDS) analyses linked to SEM, in order to avoid elemental Ni (and nutrients) redistribution in leaf tissues (van der Ent et al. 2017). The hypothesis that Ni is transported along with the transpiration stream (Coinchelin et al. 2012) is supported by the facts that (1) Ni was predominantly located in the epidermis, Ni being seven times more concentrated in both epidermis layers than in their bulk neighbouring leaf counterparts and (2) the greatest Ni concentrations occurred in the highly transpiring young and thin leaves with the greatest stomatal densities. However, there is also evidence, as recently shown for N. caerulescens, that Ni is strongly redistributed from old to young leaves by phloem stream under the form of malate complexes (Deng et al. 2016).
These highly-transpiring young leaves, and the mature leaves, were also characterized by strong positive spatial or time-series Ni–S correlations at the leaf-tissue scale. Moreover, principal-component analysis revealed an opposition between, on the one hand, both epidermis layers, which are rich in Ni, S and Mg, particularly those of the young and mature leaves, and, on the other hand, the two parenchymas, which are rich in K, P and Ca. The question arises as to Ni speciation in specific tissues of the leaves. Peer et al. (2006) observed that concentrations of O-acetyl-l-serine in shoots of 18 hyperaccumulating plants from Brassicacae family collected from natural sites, including two Greek L. emarginata populations, are strongly correlated with their ability to hyperaccumulate Ni in their shoots. Plants are highly capable of coupling photosynthesis to a reduction in sulfate, assimilation into cysteine, and further metabolism into methionine, glutathione, and many other compounds (Leustek et al. 2000; Leustek 2002), key sulfur ligands for metals (Küpper et al. 2004; Freeman et al. 2004, 2006; Haydon and Cobbett 2007; Mijovilovich et al. 2009; Richau et al. 2009). Such S assimilation is regulated during development, where the greatest activity of sulfur assimilation enzymes takes place at the growing points, i.e. the youngest leaves and the root tips (Leustek et al. 2000; Leustek 2002). Indeed, we observed that young leaves were richer in S than mature or old leaves, even without Ni in the substrate. The question arises as to whether Ni complexation occurs with sulfonate ligands. A negative answer can be formulated because ligands responsible for the Ni transfer within 4 month-old Ni-hyperaccumulator L. emarginata were citrate and malate (Montargès-Pelletier et al. 2008), rendering Ni much more mobile, whereas malate was found as the predominant ligand of leaf Ni (X-ray absorption spectroscopy). It was also shown that malate is involved in phloem re-translocation of Ni in hyperaccumulating Brassicaceae (Deng et al. 2016). Moreover, CaSO4 crystals were observed around the stomata on air-dried leaves (SEM–EDS observations and analyses, data not shown). This suggests that Ni is partly accumulated in leaf epidermis along with sulfate ions, also bearing in mind that Ni was already supplied in the form of NiSO4 7H2O. Moreover, the speciation of Ni in the xylem sap was proved to be mostly under the free ion in hyperaccumulating Alyssum species (Alves et al. 2011; Centofanti et al. 2013) and it was also shown that none of the chelating molecules (e.g. malate, citrate, nicotianamine) were at comparable concentrations to Ni in xylem sap. We suggest that sulfate, which is uploaded through the xylem with Ni, forms the weak Ni(SO4)0 complex that is transported along the transpiration stream towards the leaf epidermis, although this remains to be demonstrated.
The age-dependent S:Ni, Ca:Ni, P:Ni and K:Ni mass ratios in the whole leaves or in the leaf tissues may be attributed to the relative phloem Ni mobility (Deng et al. 2016), as Ca, K and P seem to be rather phloem-immobile compared to Ni. Unfortunately, Ca, Mg, P and S were not monitored in their study (Deng et al. 2016). The fact that Mg concentration in the epidermis tended to be higher in the very old leaves than in the young leaves (in contrast to Ni and S concentrations), would suggest a joint transport of Ni and Mg in the transpiration stream (xylem to stomata and epidermis) and possibly the role of sulfate ions at least as counter-ions to facilitate their mobility. However, phloem mobility of Ni would induce Ni redistribution in newer leaves, whereas Mg would remain phloem-immobile, thus further increasing the Ni concentrations in active transpiring young leaves. The re-mobilisation of Ni from mature to young leaves has certainly implications on the Ni re-cycling through leaf litter decay, although Ni biogeochemical recycling in soil-hyperaccumulator systems has been shown to be extremely intense in the case of Alyssum murale (Echevarria 2018).
Conclusion and perspectives
The classical leaf-tissue Ni analyses, which are carried out on a single leaf belonging to a single individual hyperaccumulating plant, is open to question in the light of our results. The characterization of time-series leaves belonging to at least three plant replicates, as in this study, appeared to be particularly relevant as variations in leaf or leaf-tissue Ni concentrations were partly explained by individual plant effects. We attributed this individual effect to the allogamous character of the Ni-hyperaccumulator L. emarginata.
The accumulation of Ni appears to be mediated by both the transpiration stream and phloem redistribution towards physiologically active (with high-transpiration rates) young leaves. Long-distance Ni transport via the xylem is predominant in the Ni-hyperaccumulator L. emarginata, whereas a complementary redistribution via the phloem from old to young leaves should also occur, as observed for N. caerulescens. A strong correlation and co-occurrence of S and Ni suggests that there might be some processes that involve both elements. The combined use of radioactive or stable tracers of Ni, S, Mg, Ca or of H 2 18 O, would help to unravel further issues concerning the spatio-temporal dynamics of the elements from xylem to leaf tissues over short periods and to evaluate the role of phloem redistribution in element segregation. Moreover, synchrotron techniques could also (micro-EXAFS) help understand the different mechanisms involved after Ni is downloaded from the xylem and during its further redistribution through phloem flux, including the potential interactions with sulfate ions. Also, some unpublished results on the Ni-hyperaccumulating Brassicaceae A. murale showed that leaf traits could vary (e.g. SLA) among individuals of the same population if they are grown on either ultramafic or non-ultramafic soils. Therefore, it would be necessary for a better understanding of the mechanisms of Ni homeostasis to conduct such a study on L. emarginata and other hyperaccumulating Brassicaceae by comparing leaf treats and leaf Ni accumulation in a gradient of Ni availability from non-ultramafic to ultramafic soils.
Trust and financial support from the Université de Lorraine, ADEME and Lorraine Regional Council given to Dr. David Coinchelin for his PhD Grant were greatly appreciated. We thank Helen Selliez for improving the English. The authors deeply acknowledge their co-author, the late Dr. François Bartoli, for the many years shared in developing creative research together and for his valuable friendship.
- Baker AJM, Brooks RR (1989) Terrestrial higher plants which hyperaccumulate metallic elements—a review of their distribution, ecology and phytochemistry. Biorecovery 1:81–126Google Scholar
- Echevarria G (2018) Genesis and behaviour of ultramafic soils and consequences for nickel biogeochemistry. In: van der Ent A, Echevarria G, Baker AJM, Morel JL (eds) Agromining: farming for metals. Extracting unconventional resources using plants. Mineral Resources Series, Springer International Publishing AG, Cham, Switzerland, pp 135–156Google Scholar
- Food and Agriculture Organization of the United Nations (FAO) (2001) Lecture notes on the major soils of the world. In: Driessen P, Deckers J, Spaargaren O, Nachtergaele F (eds) World soil resources report, no. 94. FAO, RomeGoogle Scholar
- Jaffré T, Schmid M (1974) Accumulation du nickel par une Rubiacée de Nouvelle-Calédonie, Psychotria douarrei (G. Beauvisage) Däniker. CR Acad Sci D Nat 278:1727–1730Google Scholar
- Küpper H, Mijovilovich A, Meyer-Klaucke W, Kroneck MH (2004) Tissue- and age-dependent differences in the complexation of cadmium and zinc in the cadmium/zinc hyperaccumulator Thlaspi caerulescens (Ganges ecotype) revealed by X-Ray absorption spectrometry. Plant Physiol 134:748–757CrossRefPubMedPubMedCentralGoogle Scholar
- Leustek T (2002) Sulfate metabolism. The Arabidopsis book 1. https://doi.org/10.1199/tab.0017
- Mesjasz-Przybylowicz J, Przybylowicz WJ, Pineda CA (2001) Nuclear microprobe studies of elemental distribution in apical leaves of the Ni hyperaccumulator Berkheya coddii. S Afr J Sci 97:591–593Google Scholar
- Mijovilovich A, Leitenmaier B, Meyer-Klaucke W, Kroneck MH, Götz B, Küpper H (2009) Complexation and toxicity of copper in higher plants. II. Different mechanisms for copper versus cadmium detoxification in the copper-sensitive cadmium/zinc hyperaccumulator Thlaspi caerulescens (Ganges ecotype). Plant Physiol 151:715–731CrossRefPubMedPubMedCentralGoogle Scholar
- Nick LJ, Chambers MF (1995) Farming for metals? Mining Environ Manag 3:15–18Google Scholar