However, it is important to bear in mind that even if minimising the fertiliser input had a less relevant impact on costs, it is key to preventing leaching nutrients and preserving the over-polluted aquifers in the region.Industrialization during the last centuries has led to extended areas of contamination with increased metal concentration levels. During the last three decades, eco-friendly, gentle soil remediation techniques using different plant species have been developed to address this problem. Plant characteristics such as metal tolerance, accumulation behaviour, biomass production, and root morphology determine metal extraction capacities. A number of plant species known as “hyperaccumulators” share the ability to grow on metalliferous soils and to accumulate large amounts of heavy metals in their aerial organs, to concentration levels exceeding that of normal plants >100 times without suffering phytotoxic effects. One species, Noccaea rotundifolia ssp. cepaeifolia, was found to accumulate up to 21,000 mg kg1 Zn. Reeves and Brooks found Zn concentrations up to 17,300 mg kg1 in leaves of N. rotundifolia in the vicinity of a PbeZn mine in the Cave del Predile area, Northern Italy. Reeves and Brooks reported that this species was also able to accumulate Pb. The highest Pb concentrations were found in shoots of N. rotundifolia in Cave del Predile. The same authors reported N. rotundifolia being a Cd hyperaccumulator.
In contrast, Koubova et al. using pot and hydroponic experiments showed that the potential of N. rotundifolia to accumulate Pb in the plant shoots biomass was negligible and did not confirm the potential Pb hyperaccumulation ability of this species.For this experiment a soil from Arnoldstein with moderate Zn, Pb and Cd contamination and loamy texture was used as control. As this soil was moderately contaminated with Cd, we also studied Cd accumulation in shoots as well as the labile fractions and mobilisation of Cd in the rhizosphere. Part of the soil material was spiked to obtain a highly Zn, Pb and Cd contaminated soil. The desired amount of metals was dissolved in deionized water,25 liter pot then sprayed on the soil in polyethylene bags and subsequently thoroughly mixed by shaking the plastic bag. For equilibration, soils were incubated at 40 C for 3 weeks with several wetting and drying cycles using deionized water. Selected characteristics of the experimental soils are shown in Table 1. To study changes of labile metal fractions, pH and DOC concentration in the N. rotundifolia rhizosphere we conducted a rhizobox experiment using the design of Wenzel et al. in 4 replicates. Eight seedlings were planted on top of a soileroot compartment with an opening slit at the bottom. After 6 weeks the number of plants was reduced to four, and the plant compartment was transferred on top of a rhizosphere soil compartment attached to a vertical root-only compartment that is separated from rhizosphere soil compartment by a nylon membrane and a transparent acrylic window. After another 7 weeks, sufficient root coverage of the nylon membrane separating roots from rhizosphere soil was achieved.
Subsequently,the rhizosphere compartment was separated from the root layer. Roots adjacent to the membrane were washed in an ultrasonic bath and rinsed using bi-distilled water.The experimental soil used in this experiment was obtained from an arable field near Wiesen, Austria ; selected characteristics of this loamy soil are given in Table 2. Airdried soil was passed through a 2-mm sieve. Cation exchange capacity,organic matter content,plant available P based on the Olsen method and plant available K using 1 M ammonium acetate were determined. Part of the sieved soil was maintained to provide non-polluted control soil,other batches were enriched with Zn or Pb to yield nominal concentrations of 100, 500, 1000, 2000 and 3000 mg kg1 or 250, 500, 1000 and 2000 mg kg1 by mixing the soil with ZnO or PbO, respectively. Metal oxides were used for spiking to avoid mass flow of metals between compartments through irrigation. The near-total Zn and Pb concentrations in Wiesen soil were 31.9 and 4.89 mg kg-1 respectively. To meet the presumable high zinc requirement of N. rotundifolia, we also spiked the control soil to obtain 100 mg kg-1. The metal spiking procedure was carried out as described in chapter 2.2. The near total,ammonium nitrate-extractable,and calcium nitrate extractable concentrations of Zn and Pb in the soils, soluble P,soil pH and dissolved organic carbon were determined after three weeks equilibration and are presented in Table S1. A split-root system technique was used to study the effect of localized Zn and Pb patches on the root foraging behaviour of N. rotundifolia according to a modified method of Langer et al.. Half-liter pots, which were separated into 2 equal compartments, were filled with homogeneous or heterogeneous pairs of control or Pb- and Zn-enriched soil treatments as shown in Table 3 and Fig. S1.
The metalenriched patches in heterogeneous treatments were placed on the right-hand side of the pots. Experimental pots were equipped with a mesh permeable for root growth but stabilizing the paired substrate patches supporting proper separation during the harvesting procedure. The free-draining pots were placed on a grid to avoid mixing of percolating water from different patches. Twelve day-old seedlings were placed individually at the contact zone of the paired soil patches. The experiments were carried out at a soil moisture of 80e85% of maximum water holding capacity in a greenhouse with a completely randomized design and 4 replications. At harvest,fresh root biomass that developed in each split substrate patch as well as shoots were harvested. Root morphological characteristics were measured on fresh root material. Fresh roots were weighed and cut into pieces. An aliquot of fresh roots from each part of the pot was randomly selected and kept in ethanol for root scanning; in addition, a sub-sample was dried at 65 C for 2 days to determine the dry weight factor. Detection of root morphological characteristics in different diameter classes was done by scanning the roots using a modified Optical Scanner ; images were processed by using WinRHIZO®. To increase the contrast of the roots against the background on the scanned images, they were stained with 5% Giemza solution for 10 min at about 40 C on a heating plate. Data analyses were performed with ANOVA in SAS software, followed by LSD and Duncan-test. The differences in root characteristics between patches were evaluated using a paired t-test.N. rotundifolia tended to produce less shoot and root dry matter in the homogeneous as compared to the heterogeneous Zn and Pb treatments.
Comparison of root allocation patterns revealed no significant difference in the partitioning of root mass in any of the homogeneous Zn and Pb soil treatments, with approximately 50% allocation into the left and right sides of all pots. We found preferred partitioning of root mass into the low Zn and Pb halves of heterogeneous soil treatments with a large difference of metal concentrations between low and high metal compartments. In such Zn treatments,approximately 35% of the total root mass was found in the Znenriched half patch with higher Zn concentration;similarly, in the highly heterogeneous Pb treatments 38 and 32% of total root mass respectively, allocated to the Pb-enriched half. In heterogeneous treatments with less pronounced Zn concentration differences between the high and low metal compartments,we found no difference in root dry matter allocation, however, increased root length and related larger surface area in the compartment with lower Zn concentration. The average root diameter decreased with increasing Zn concentration in soil in heterogeneous treatments except in 100/ 3000. We found no significant differences in root length and surface area between the compartments of heterogeneous Pb treatments. However, specific root area and specific root length were significantly higher in the Pbenriched patch in two treatments with the largest Pb concentration difference between compartments. Lead and Zn concentrations in N. rotundifolia shoots increased with increasing soil metal concentration, following a linear function for Pb but approaching a plateau in the case of Zn, with only small differences between homogeneous and heterogeneous treatments. The only considerable deviation from this pattern was found for the heterogeneous Zn treatments 100/3000 and 500/3000 where the shoot Zn concentrations in the heterogeneous treatments were clearly below the fit expected from all other data, and lower than the corresponding homogenous treatments.
In the homogenous Zn treatments, the plant Zn contents increased with increasing Zn supply, approaching a plateau at high soil Zn concentrations. In the heterogeneous treatments, the Zn contents follow a more irregular pattern, with lower than expected values in the treatments 100/1000 and 100/3000.The shoot Pb contents showed a linear positive response to increasing Pb concentrations in soil, with generally larger contents in the heterogeneous treatments.In heterogeneous treatments with smaller differences in Zn concentration between compartments, toxicity stress was indicated by changes in root morphology, including decreased root length and smaller root surface area in the Zn-enriched patches with higher Zn concentrations even though allocation of total root dry matter was not different. Toxicity was also indicated by decreasing average root diameter as the overall Zn concentration in the split-pot study was increased. In contrast, spiking the moderately polluted ARN-B soil with Zn,Cd,and Pb resulted in a ~32% increase of shoot dry matter. Overall, our findings provide evidence for high tolerance of N. rotundifolia to Zn in hydroponic solution and even on soil from its natural habitat spiked with a metal cocktail, whereas this species appeared to be more sensitive to Zn-only spikes in an initially low-metal agricultural soil. Taking all evidence together we classify this species as highly tolerant Zn accumulator plant. N. rotundifolia exhibits similar tolerance to Zn exposure as N. caerulescens, raspberry cultivation pot albeit at lower levels of Zn accumulation in shoots. In the hydroponic Pb dose-response experiment, N. rotundifolia accumulated enormous Pb concentrations in its roots, with almost 81.8 g kg1 at 1000 mM Pb in solution. However, only a small proportion of Pb was transferred to shoots, with ~1250 mg kg1 in the 1000 mM Pb treatment.Only at this largest concentration the commonly accepted hyperaccumulator threshold of Pb was exceeded. Apart from the high Pb concentrations in the soils of the split-pot experiment, Pb concentrations in shoots did not exceed 100 mg kg-1 thus remaining one order of magnitude below the hyperaccumulation threshold.Even in the spiked soil the Pb shoot concentrations in the rhizobox experiment did not exceed 86 mg kg-1. This is in contrast to previous reports of Pb hyperaccumulation by N. rotundifolia, with Pb shoot concentrations up to 8200 mg kg-1 in field-collected specimen from the Southern Alps in adjacent regions of Italy and Austria.
In all experiments of our study, the shoot: root concentration quotient of Pb was clearly <1, decreasing to values near 0.01 in the highest Pb treatments of the hydroponic study. Similarly, the bioconcentration factor for Pb remained clearly below 1 in all experimental setups. N. rotundifolia responded to Pb with decreased root and shoot dry matter production, starting from the lowest level of exposure. On the other hand, we found increased dry matter production upon spiking a moderately metal-polluted soil from the plant’s native environment with a cocktail containing Cd, Zn and Pb, resulting in a soil concentration of 1650 mg Pb kg-1. As for Zn, the root biomass of N. rotundifolia was evenly distributed in all homogenous and heterogeneous treatments unless the concentration difference between the compartments was very high. We found no significant differences in root surface area and root length between the compartments of heterogeneous Pb treatments. However, specific root surface area and specific root length were significantly higher in the Pb-enriched patch of the two treatments with the largest soil Pb concentration difference between compartments. Taking together all evidence obtained in our study,we challenge earlier reports on Pb hyperaccumulation properties of field-collected N. rotundifolia specimen from the same site. Our findings support the doubts regarding the identification of hyperaccumulation solely based on field-collected specimen related to potentially ineffective washing procedures prior to analysis. In that case aerial contamination from emissions or soil dust derived in the vicinity of the collected plant specimen cannot be excluded and may bias the results towards meeting hyperaccumulation thresholds.