Evaluating a mass balance model for soil trace metals using the historical data from the King's Kitchen Garden (Versailles, France).

The mass balance of reconstituted Cd, Cu, Pb and Zn fluxes from 1683 to 2021 was compared to the current levels of the soil used only for vegetable production in the King's Kitchen Garden in Versailles (France). This comparison was made on the basis of 4 scenarios of organic matter application in the 18th and 19th centuries and by an uncertainty analysis over the entire period. The topsoil contamination falls within that of French kitchen gardens. Modelling of past fluxes predicted the correct trend (an increase) and order of magnitude of the soil metal contents. It produced a relatively accurate evaluation of the Cu and Zn contents. The model underestimated the Pb contents by about 80%, revealing a large and unknown source of soil contamination by this metal. The calculation overestimated the current Cd levels by about 100%, probably due to various biases, for example on atmospheric fallout or the composition of organic amendments. This assessment shows that modelling the mass balance of trace metal fluxes can be used to predict the long-term trend in the levels of these elements in cultivated soils, providing the input data are chosen according to realistic scenarios.

How exogenous ligand enhances the efficiency of cadmium phytoextraction from soils?

Many experiments showed that exogenous ligands could enhance cadmium (Cd) phytoextraction efficiency in soils. Previous studies suggested that the dissociation and the apoplastic uptake of Cd complex could not fully explain the increase of root Cd uptake. Two hypotheses are evaluated to explain enhanced Cd uptake in the presence of ligand: i) enhanced apoplastic uptake of complex due to reduced apoplastic resistance and ii) complex internalization by membrane transporters. RESULTS: show that the ligand affinity for Cd is a key characteristic determining the potential mechanism for enhanced Cd uptake. When low molecular weight organic acids are applied, the complex dissociation could generally be fast (> 10-3.3 s-1) and result in the increased Cd uptake. When hydrophilic aminopolycarboxylic acids (APCAs) are applied in experiments without water or temperature stresses to the plant, the root water uptake flux could very likely be high (> 10-7.8 dm s-1), and the strong apoplastic complex uptake could enhance the root Cd uptake. When lipophilic APCAs are applied, the strong internalization of the complex by membrane transporters could result in the increased Cd uptake if the maximum internalization rate is high (> 10-12 mol dm-2 s-1). However, the complex internalization by membrane transporters must be experimentally confirmed.

Effect of calcium and trace metals (Cd, Cu, Mn, Ni, Zn) on root iron uptake in relation to chemical properties of the root-excreted ligands.

Interferences of major cations (Ca2+, Mg2+) and trace metals (TM, i.e. Cd2+, Cu2+, Mn2+, Ni2+ and Zn2+) in root Fe uptake were evaluated. Root Fe uptake was modelled including the reactions of the root exuded ligand with the soil major and trace cations. Fe uptake was simulated with different ligands representing various affinities for the cations, the latter varying in concentration. The stability constant of Fe complexes (KFeL) does not influence Fe uptake, contrarily to the ligand parameters for Fe-hydroxide dissolution. Fe uptake decreases when KCaL or Ca2+ in solution increases. Presence of TM has nearly no influence on Fe uptake when the TM complexes have low stability constants (KML), as in the case of oxalate and citrate complexes. When ligands have high KML, like EDTA, DFO-B or mugineic acid (MA), TM reduces Fe uptake by 51-55%, and much more in the case of TM contamination. Exudation of Fe ligands with low KML has no negative effect on TM uptake, which can increase if the dissociation rate is high, as for Cu complexes. Ligands with high KML (EDTA, DFO-B, MA) greatly reduce TM uptake, only if their hydrated cations can be absorbed. Calcium does not significantly reduce Fe uptake when Ca-complexes have KCaL < 104. Consequently, ligands like oxalate or MA should be efficient in most soils. TM should perturbate Fe uptake mediated by ligands with high KML such as MA, but not oxalate. Plants exuding phytosiderophores should also absorb TM complexes to avoid micronutrient deficiencies.

Assessing the future trends of soil trace metal contents in French urban gardens.

Soil trace metal concentrations (e.g., cadmium, copper, lead, zinc) in vegetable gardens have often been observed as exceeding the geochemical background levels. These metals are a threat both to soil and plant functioning and to human health through consumption of contaminated vegetables. We used a mass balance-based model to predict the four metal (Cd, Cu, Pb, Zn) concentrations in soils after a century's cultivation for 104 urban vegetable gardens, located in three French metropolises, Nancy, Nantes, and Marseille, based on a survey of gardening practices. If current gardening practices are maintained, an increase in soil Cd (35% on average), Cu (183%), and Zn (27%) contents should occur after a century. Soil Pb concentration should not vary consistently. Organic amendments are the major source of Cd, Pb, and Zn, followed by chemical fertilizer while fungicide application is the major source of Cu. Cessation of chemical fertilizer use would only slightly reduce the accumulation of the four metals. The solubility of the four metals would decrease significantly after a century, when pH increases by one unit. A liming practice of acidic soils should therefore be a feasible way to prevent any increase in the metal mobility and bioavailability.

Impact of phenanthrene on primary metabolite profiling in root exudates and maize mucilage.

This study was conducted to assess the impact of polycyclic aromatic hydrocarbon on the composition of rhizodeposits. Maize was submitted to increasing phenanthrene (PHE) concentrations in the substrate (0, 25, 50, and 100 mg PHE.kg-1 of dry sand). After 6 weeks of cultivation, two types of rhizodeposit solution were collected. The first one, called rhizospheric sand extract, resulted from the extraction of root adhering sand in order to collect mucilage and associated compounds. The second one, the diffusate solution, was collected by the diffusion of exudates from roots soaked in water. The impact of phenanthrene on maize morphology and functioning was measured prior to the analysis of the main components of the rhizodeposit solutions, by measuring total carbon, protein, amino acid, and sugars as well as by determining about 40 compounds using GC-MS and LC-MS. As maize exposure to PHE increased, different trends were observed in the two rhizodeposit solutions. In the diffusate solution, we measured a global increase of metabolites exudation like carbohydrates, amino acids, and proteins except for some monoglycerides and organic acids which exudation decreased in the presence of PHE. In the rhizospheric sand extract, we witnessed a decrease in carbohydrates and amino acids secretion as well as in fatty and organic acids when plants were exposed to PHE. Many of the compounds measured, like organic acids, carbohydrates, amino acids, or fatty acids, could directly or indirectly drive PAHs availability in soils with particular consequences for their degradation.

Biochar-assisted phytoextraction of Cd and Zn by Noccaea caerulescens on a contaminated soil: A four-year lysimeter study.

Amendments of biochar, the residual solid of biomass pyrolysis, have been shown to enhance metal phytoextraction from contaminated soils with hyperaccumulating plants in specific situations. In order to investigate this phenomenon over successive harvests in field conditions, two identical undisturbed soil cylinders (1-m2 section × 1.85-m height) were excavated from a contaminated agricultural plot and monitored with instrumented lysimeters. Wood-derived biochar was added at a rate of 5% (w/w) in the first 30 cm of one of the two lysimeters. The Cd/Zn-hyperaccumulator Noccaea caerulescens was then grown for the next four years on both lysimeters. Our results showed that the hyperaccumulating plant was able to remove about 2 g m-2 of Cd and 12-16 g m-2 of Zn within four years, representing about 40% and 4% of the initial Cd and Zn soil contamination, respectively. Biochar amendment improved plant germination and survival and increased root surface density. However, no significant effect of biochar on shoot metal content of N. caerulescens was observed. Mass balances suggested that up to 10% the metal contamination moved from the disturbed Ap horizon to the deeper horizons, particularly in the biochar-amended soil profile. Furthermore, shoot Cd and Zn concentration generally decreased over the successive harvests, together with soil metal availability. Depending on the way to account for this progressive decrease in efficiency, our estimations of the time necessary to remove the excess of metals in the topsoil in these conditions ranged from 11 to 111 years for Cd and from 97 years to an infinite time for Zn. In conclusion, the simultaneous use of N. caerulescens and biochar amendment can lead to a significant removal of specific metallic elements from the topsoil, but the risk of metal movement down the soil profile and the observed decrease in phytoextraction efficiency over time deserve further investigations.

Profiling of main metabolites in root exudates and mucilage collected from maize submitted to cadmium stress.

The aim of this study was to characterize qualitatively and quantitatively the composition of the main rhizodeposits emitted from maize (Zea mays) under Cd stress, in order to discuss their role in Cd availability and tolerance. Maize was grown for 6 weeks in sand at four Cd exposure levels (0, 10, 20, and 40 µM Cd in nutrient solution) and two types of rhizodeposits were collected at the end of cultivation period. Mucilage and other molecules adhering to rhizospheric sand were extracted with a buffer before root exudates were collected by diffusion into water. Total carbon, proteins, amino acids, and sugars were analyzed for both rhizodeposit types and about 40 molecules were identified using GC-MS and LC-MS. Cadmium effect on plant morphology and functioning was slight, but consistent with previous works on Cd toxicity. However, rhizodeposition did tend to be impacted, with a decrease in total carbon, sugars, and amino acids correlating with an increasing Cd content. Such a decrease was not noticeable for proteins in root exudates. These observations were confirmed by the same trends in individual compound contents, although the results were generally not statistically significant. Many of the molecules determined are well-known to modify, whether directly or indirectly, Cd speciation and dynamics in the soil and could play a role in Cd tolerance.

Breeding the hyperaccumulator Noccaea caerulescens for trace metal phytoextraction: first results of a pure-line selection.

To initiate the creation of phytoextraction cultivars, plants were selected from 60 populations of N. caerulescens for their high shoot biomass or Cd, Ni, and Zn concentrations. They were self-pollinated, and the selection and fixation were continued for three generations in greenhouse conditions. Selected plants showed a potential to produce 5-10 t dry matter ha-1, which is required to decontaminate soils which have been moderately contaminated with Cd. However, the high biomass genotypes could not be fixed, probably both because of their complexity and to the sensitivity of this trait to environmental conditions, and plant density in particular. The selection led to an improvement to the Cd and Zn accumulation capacities of the plants, yet caused a decrease in their Ni accumulation. This is most likely due to a decline in Ni availability in soil, rather than to a deleterious effect of inbreeding. Metal accumulation appeared to be more heritable than biomass production and fixation for the former trait should be quicker than for the latter. The accumulation capacities of the selected plants permitted offtakes representing around 25% of the soil Cd in a single cropping. This potential has to be confirmed in field conditions.

How could phytoextraction reduce Cd content in soils under annual crops? Simulations in the French context.

Human populations are threatened by chronic exposure to the Cd accumulated in foods after being taken up from soils by crops. To evaluate how phytoextraction with the hyperaccumulator Noccaea caerulescens as an annual crop or as a cover crop could modify the Cd mass balance in French agricultural soils, we simulated this process according to two scenarios. If current practices are maintained (first scenario), the average soil Cd content will increase by 2.9% after a century. If Cd content in P fertilizers is limited according to the European regulation project (second scenario), the decrease will be of about 4%. A phytoextraction crop with a 10 t dry matter (DM) ha-1 yield every 25 years would bring down the soil Cd content from 0.31 mg kg-1 to around 0.11 mg kg-1. However, this scenario is relatively unrealistic, because high dry matter yield is unlikely and the cost of the process is elevated. Phytoextraction as a cover crop every four to five years would decrease the soil Cd content more quickly. This requires a 2.5 t DM ha-1 yield, which appears realistic. This cover crop phytoextraction would be cheaper. It would need annual sowing of 4 million ha and the production of around 10 million t of dry biomass. To meet such a requirement, any breeding of the hyperaccumulator should favour traits allowing a 3-4 month cultivation period in the autumn. Processes also have to be developed to recover energy, metal or beneficial compounds from the biomass produced by phytoextraction.

Cadmium mass balance in French soils under annual crops: Scenarios for the next century.

Human populations are threatened by chronic exposure to the Cd accumulated in foods after being taken up from soils by crops. To decide whether and to what extent it is necessary to reduce the Cd content in cultivated soils, one needs to understand and predict its evolution. We therefore simulated the Cd mass balance in the soils under annual crops in France and in its 22 regions for the next century, following six scenarios of agricultural practices or regulatory conditions. If current cultivation practices are maintained, the average Cd content would increase by about 15% after a century, due to the input of Cd with P fertilizer applications. This represents around 85% of the soil Cd inputs and is nearly twice the Cd output caused by leaching and crop offtake. These results conflict with those recently obtained at the European level, due to three factors: the higher rate of P application in France than in Europe, a higher Cd content in the P fertilizers applied in France and a lower Cd leaching in French soils. Strict application of the good practices for P fertilization would stabilize the future soil Cd content at its present level. Assuming the current excessive P fertilization, the enforcement of a regulation limiting Cd content in the P fertilizers, as proposed by the European Union, would lead to a lesser increase in soil Cd, by 1.6% to 3.9% after a century. The combination of P fertilization good practices and Cd content limitation in P fertilizers would lead to a decrease in soil Cd content of between 3.0% to 5.2%. Organic agriculture would lead to an evolution of soil Cd content similar to that of conventional agriculture applying good practices. The accuracy of the mass balances could be ameliorated by a better assessment of Cd leaching.

Comparison between numeric and approximate analytic solutions for the prediction of soil metal uptake by roots. Example of cadmium.

The dissociation of metal complexes in the soil solution can increase the availability of metals for root uptake. When it is accounted for in models of bioavailability of soil metals, the number of partial differential equations (PDEs) increases and the computation time to numerically solve these equations may be problematic when a large number of simulations are required, for example for sensitivity analyses or when considering root architecture. This work presents analytical solutions for the set of PDEs describing the bioavailability of soil metals including the kinetics of complexation for three scenarios where the metal complex in solution was fully inert, fully labile, or partially labile. The analytical solutions are only valid i) at steady-state when the PDEs become ordinary differential equations, the transient phase being not covered, ii) when diffusion is the major mechanism of transport and therefore, when convection is negligible, iii) when there is no between-root competition. The formulation of the analytical solutions is for cylindrical geometry but the solutions rely on the spread of the depletion profile around the root, which was modelled assuming a planar geometry. The analytical solutions were evaluated by comparison with the corresponding PDEs for cadmium in the case of the French agricultural soils. Provided that convection was much lower than diffusion (Péclet's number<0.02), the cumulative uptakes calculated from the analytic solutions were in very good agreement with those calculated from the PDEs, even in the case of a partially labile complex. The analytic solutions can be used instead of the PDEs to predict root uptake of metals. The analytic solutions were also used to build an indicator of the contribution of a complex to the uptake of the metal by roots, which can be helpful to predict the effect of soluble organic matter on the bioavailability of soil metals.

Phytoremediation of urban soils contaminated with trace metals using Noccaea caerulescens: comparing non-metallicolous populations to the metallicolous 'Ganges' in field trials.

Urban soil contamination with trace metals is a major obstacle to the development of urban agriculture as crops grown in urban gardens are prone to accumulate trace metals up to toxic levels for human consumption. Phytoextraction is considered as a potentially cost-effective alternative to conventional methods such as excavation. Field trials of phytoextraction with Noccaea caerulescens were conducted on urban soils contaminated with Cd, Cu, Pb, and Zn (respectively around 2, 150-200, 400-500, and 400-700 µg g-1 of dry soil). Metallicolous (Ganges population) and non-metallicolous (NMET) populations were compared for biomass production and trace metal uptake. Moreover, we tested the effect of compost and fertilizer addition. Maximal biomass of 5 t ha-1 was obtained with NMET populations on some plots. Compared to Ganges- the high Cd-accumulating ecotype from South of France often used in phytoextraction trials- NMET populations have an advantage for biomass production and for Zn accumulation, with an average Zn uptake of 2.5 times higher. The addition of compost seems detrimental due to metal immobilization in the soil with little or no effect on plant growth. In addition to differences between populations, variations of growth and metal accumulation were mostly explained by soil Cd and Zn concentrations and texture. Our field trials confirm the potential of using N. caerulescens for both Cd and Zn remediation of moderately contaminated soils-with uptake values of up to 200 g Cd ha-1 and 47 kg Zn ha-1-and show the interest of selecting the adequate population according to the targeted metal.

Demographic history of the trace metal hyperaccumulator Noccaea caerulescens (J. Presl and C. Presl) F. K. Mey. in Western Europe.

Noccaea caerulescens (Brassicaceae) is a major pseudometallophyte model for the investigation of the genetics and evolution of metal hyperaccumulation in plants. We studied the population genetics and demographic history of this species to advance the understanding of among-population differences in metal hyperaccumulation and tolerance abilities. Sampling of seven to 30 plants was carried out in 62 sites in Western Europe. Genotyping was carried out using a combination of new chloroplast and nuclear neutral markers. A strong genetic structure was detected, allowing the definition of three genetic subunits. Subunits showed a good geographic coherence. Accordingly, distant metallicolous populations generally belonged to distinct subunits. Approximate Bayesian computation analysis of demographic scenarios among subunits further supported a primary isolation of populations from the southern Massif Central prior to last glacial maximum, whereas northern populations may have derived during postglacial recolonization events. Estimated divergence times among subunits were rather recent in comparison with the species history, but certainly before the establishment of anthropogenic metalliferous sites. Our results suggest that the large-scale genetic structure of N. caerulescens populations pre-existed to the local adaptation to metalliferous sites. The population structure of quantitative variation for metal-related adaptive traits must have established independently in isolated gene pools. However, features of the most divergent genetic unit (e.g. extreme levels of Cd accumulation observed in previous studies) question the putative relationships between adaptive evolution of metal-related traits and subunits isolation. Finally, admixture signals among distant metallicolous populations suggest a putative role of human activities in facilitating long-distance genetic exchanges.

Zinc Isotope Fractionation in the Hyperaccumulator Noccaea caerulescens and the Nonaccumulating Plant Thlaspi arvense at Low and High Zn Supply.

On the basis of our previous field survey, we postulate that the pattern and degree of zinc (Zn) isotope fractionation in the Zn hyperaccumulator Noccaea caerulescens (J. & C. Presl) F. K. Mey may reflect a relationship between Zn bioavailability and plant uptake strategies. Here, we investigated Zn isotope discrimination during Zn uptake and translocation in N. caerulescens and in a nonaccumulator Thlaspi arvense L. with a contrasting Zn accumulation ability in response to low (Zn-L) and high (Zn-H) Zn supplies. The average isotope fractionations of the N. caerulescens plant as a whole, relative to solution (Δ(66)Znplant-solution), were -0.06 and -0.12‰ at Zn-L-C and Zn-H-C, respectively, indicative of the predominance of a high-affinity (e.g., ZIP transporter proteins) transport across the root cell membrane. For T. arvense, plants were more enriched in light isotopes under Zn-H-A (Δ(66)Znplant-solution = -0.26‰) than under Zn-L-A and N. caerulescens plants, implying that a low-affinity (e.g., ion channel) transport might begin to function in the nonaccumulating plants when external Zn supply increases. Within the root tissues of both species, the apoplast fractions retained up to 30% of Zn mass under Zn-H. Moreover, the highest δ(66)Zn (0.75‰-0.86‰) was found in tightly bound apoplastic Zn, pointing to the strong sequestration in roots (e.g., binding to high-affinity ligands/precipitation with phosphate) when plants suffer from high Zn stress. During translocation, the magnitude of isotope fractionation was significantly greater at Zn-H (Δ(66)Znroot-shoot = 0.79‰) than at Zn-L, indicating that fractionation mechanisms associated with root-shoot translocation might be identical to the two plant species. Hence, we clearly demonstrated that Zn isotope fractionation could provide insight into the internal sequestration mechanisms of roots when plants respond to low and high Zn supplies.

Effect and localization of phenanthrene in maize roots.

Polycyclic aromatic hydrocarbons (PAHs) have a toxic effect on plants, which limits the efficiency of phytomanagement of contaminated soils. The mechanisms underlying their toxicity are not fully understood. A cultivation experiment was carried out with maize, used as model plant, exposed to sand spiked with phenanthrene (50 or 150 mg kg(-1) dw). Epi-fluorescence microscopic observation of root sections was used to assess suberization of exodermis and endodermis and phenanthrene localization along the primary root length. For 10 days of cultivation, exodermis and endodermis suberization of exposed maize was more extensive. However, after 20 days of exposure, exodermis and endodermis of non-exposed roots were totally suberized, whilst PHE-exposed roots where less suberized. Early extensive suberization may act as barrier against PHE penetration, however longer exposure inhibits root maturation. Phenanthrene patches were located only near suberized exodermis and endodermis, which may therefore act as retention zones, where the hydrophobic phenanthrene accumulates during its radial transport.

Root development of non-accumulating and hyperaccumulating plants in metal-contaminated soils amended with biochar.

Biochar may be used as an amendment in contaminated soils in phytoremediation processes. The mechanisms controlling plant metal uptake in biochar-amended soils remain however unclear. This work aimed at evaluating the influence of biochar on root development and its consequence on plant metal uptake, for two non-hyperaccumulating plants (Zea mays and Lolium perenne) and one hyperaccumulator of Cd and Zn (Noccaea caerulescens). We conducted rhizobox experiments using one acidic and one alkaline soil contaminated with Cd, Pb and Zn. Biochar was present either homogeneously in the whole soil profile or localized in specific zones. A phenomenon of root proliferation specific to biochar-amended zones was seen on the heterogeneous profiles of the acidic soil and interpreted by a decrease of soil phytotoxicity in these zones. Biochar amendments also favored root growth in the alkaline soil as a result of the lower availability of certain nutrients in the amended soil. This increase of root surface led to a higher accumulation of metals in roots of Z.mays in the acidic soil and in shoots of N. caerulescens in the alkaline soil. In conclusion, biochar can have antagonist effects on plant metal uptake by decreasing metal availability, on one hand, and by increasing root surface and inducing root proliferation, on the other hand.

Morphological and physiological responses of maize (Zea mays) exposed to sand contaminated by phenanthrene.

Phytoremediation is promising, but depends on clearly understanding contaminants' impact on plant functioning. We therefore focused on the impact of polycyclic aromatic hydrocarbons (PAH) on cultivated plants and understanding the impact of phenanthrene (PHE) on maize functioning (Zea mays). Cultivation was conducted under controlled conditions on artificially contaminated sand with PHE levels increasing from 50 to 750 mg PHE kg(-1). After four weeks, plants exposed to levels above 50 mg PHE kg(-1) presented decreased biomasses and reduced photosynthetic activity. These modifications were associated with higher biomass allocations to roots and lower ones to stems. The leaf biomass proportion was similar, with thinner blades than controls. PHE-exposed plant showed modified root architecture, with fewer roots of 0.2 and 0.4 mm in diameter. Leaves were potassium-deplete, but calcium, phosphorus, magnesium and zinc-enriched. Their content in nitrogen, iron, sulfur and manganese was unaffected. These responses resembled those of water-stress, although water contents in plant organs were not affected by PHE and water supply was not limited. They also indicated a possible perturbation of both nutritional functioning and photosynthesis.

Nickel and zinc isotope fractionation in hyperaccumulating and nonaccumulating plants.

Until now, there has been little data on the isotope fractionation of nickel (Ni) in higher plants and how this can be affected by plant Ni and zinc (Zn) homeostasis. A hydroponic cultivation was conducted to investigate the isotope fractionation of Ni and Zn during plant uptake and translocation processes. The nonaccumulator Thlaspi arvense, the Ni hyperaccumulator Alyssum murale and the Ni and Zn hyperaccumulator Noccaea caerulescens were grown in low (2 µM) and high (50 µM) Ni and Zn solutions. Results showed that plants were inclined to absorb light Ni isotopes, presumably due to the functioning of low-affinity transport systems across root cell membrane. The Ni isotope fractionation between plant and solution was greater in the hyperaccumulators grown in low Zn treatments (Δ(60)Ni(plant-solution) = -0.90 to -0.63‰) than that in the nonaccumulator T. arvense (Δ(60)Ni(plant-solution) = -0.21‰), thus indicating a greater permeability of the low-affinity transport system in hyperaccumulators. Light isotope enrichment of Zn was observed in most of the plants (Δ(66)Zn(plant-solution) = -0.23 to -0.10‰), but to a lesser extent than for Ni. The rapid uptake of Zn on the root surfaces caused concentration gradients, which induced ion diffusion in the rhizosphere and could result in light Zn isotope enrichment in the hyperaccumulator N. caerulescens. In high Zn treatment, Zn could compete with Ni during the uptake process, which reduced Ni concentration in plants and decreased the extent of Ni isotope fractionation (Δ(60)Ni(plant-solution) = -0.11 to -0.07‰), indicating that plants might take up Ni through a low-affinity transport system of Zn. We propose that isotope composition analysis for transition elements could become an empirical tool to study plant physiological processes.