Fractionation of Stable Cadmium Isotopes in the Cadmium Tolerant Ricinus communis and Hyperaccumulator Solanum nigrum

Cadmium (Cd) isotopes provide new insights into Cd uptake, transport and storage mechanisms in plants. Therefore, the present study adopted the Cd-tolerant Ricinus communis and Cd-hyperaccumulator Solanum nigrum, which were cultured under controlled conditions in a nutrient solution with variable Cd supply, to test the isotopic fractionation of Cd during plant uptake. The Cd isotope compositions of nutrient solutions and organs of the plants were measured by multiple collector inductively coupled plasma mass spectrometry (MC-ICPMS). The mass balance of Cd isotope yields isotope fractionations between plant and Cd source (δ114/110Cdorgans-solution) of −0.70‰ to −0.22‰ in Ricinus communis and −0.51‰ to −0.33‰ in Solanum nigrum. Moreover, Cd isotope fractionation during Cd transport from stem to leaf differs between the Cd-tolerant and -hyperaccumulator species. Based on these results, the processes (diffusion, adsorption, uptake or complexation), which may induce Cd isotope fractionation in plants, have been discussed. Overall, the present study indicates potential applications of Cd isotopes for investigating plant physiology.

However, to date there has been limited research on Cd isotopic composition in plants. In the present study, three Cd tolerant Ricinus communis cultivars (Zibo-5, Zibo-6, Zibo-8) and one Cd hyperaccumulator Solanum nigrum cultivar were used to study Cd uptake and translocation. These three R. communis cultivars were all high-Cd accumulators, and S. nigrum was a relatively fast-growing and high-biomass Cd-hyperaccumulator 39,40 used to develop new techniques for phytoextraction 41 . We conducted hydroponic culture experiments with these plant species and two nutrient solutions with differing Cd concentrations to 1) characterise the Cd isotope fractionation associated with Cd transfer in the Cd-tolerant and -hyperaccumulator species; and 2) explore possible mechanisms of Cd mobilisation from the solution to various physiological compartments.

Materials and Methods
Plant Growth. Seeds of three R. communis cultivars (Zibo.5, Zibo.6 and Zibo.8) and S. nigrum were obtained from the Zibo Academy of Agricultural Sciences (Shandong, China) and the Institute of Applied Ecology, Chinese Academy of Sciences (Shenyang, China), respectively. All seeds were washed in running deionized water before germination in the substrate for 14 d. The seedlings were then transferred into polycarbonate pots containing half strength Hoagland's solution 39 . The macronutrient solution consisted of 2 mmol·L −1 Ca(NO 3  Plants were cultivated under controlled conditions (16 h photoperiod with a white light intensity of 350 μ mol photons m −2 s −1 ; day: night temperatures 25 °C: 18 °C; relative humidity 60% ~ 70%). The isotopic composition (δ 114/110 Cd spex ) of the initial nutrient solution relative to Spex Cd standard solution was + 0.14 ± 0.08‰ (2SD, n = 3).

Sample Preparation.
Three plant samples as replicates were harvested 30 d after their transplantation, washed with tap water, and then rinsed thrice with deionized water. Each plant was divided into root, stem and leaf. Plant materials were freeze-dried and weighed prior analysis.
0.2 g of plant samples were digested in concentrated aristar grade HNO 3 (5 mL) and HF (1 mL) for 48 h in acid-cleaned Teflon beakers. The closed beakers were placed on a hot plate for 8 h at 80 °C and then at 160 °C until the plants were completely digested. Then 2-3 mL of HClO 4 was added to the digested solutions to remove organic materials. After evaporation at 165-180 °C, the samples were dried and redissolved in 5 mL 1% (v/v) HNO 3 (to convert the residue into the nitrate form). 2 mL of supernatant were transferred into pre-cleaned polyethylene bottles for the determination of the Cd content. The remaining fractions were evaporated to dryness, redissolved in 10 mol·L −1 HCl (to convert the residue into the chloride form), dried again, and finally taken up in 2 mL 2 mol·L −1 HCl for loading on columns.
Cd was purified by anionic exchange chromatography from nutrient solution (initial and final), root, stem and leaf following the procedure of Wei et al. 42 . Just prior to determination, the solutions were evaporated to near complete dryness and taken up in an appropriate volume of 1% HNO 3 to obtain the desired Cd concentration for mass spectrometric analysis. The recovery of Cd purification in this study was higher than 95%.
Cadmium Isotope Analysis. The Cd concentration of nutrient solutions, root, stem and leaf was measured prior to Cd purification by inductively coupled plasma quadrupole mass spectrometry (ICP-QMS) (Elan DRC-e, Perkin Elmer, USA). The Cd isotope ratios were measured by multiple collector inductively coupled plasma mass spectrometry (MC-ICPMS). Cd isotope ratios were measured by 30 cycles for each sample with an internal precision of ±0.01‰ ∼ ±0.02‰ (RSD). The Cd isotope values were expressed as permil deviation relative to the Spex Cd standard solution: Standard 1 and Standard 2 represented the standard solution measured before and after the sample. All concentration data were corrected for the procedural blank, which ranged from 8.8 ng to 13.2 ng during the course of this study. At this level, the blank has a negligible effect on the measured isotope compositions, because it constitutes less than 0.0132‰ of the indigenous Cd present in plant samples. Standard-sample bracketing was applied in this study to correct the mass bias. The instrumental reproducibility based on repetitive δ 114/110 Cd measurements of Spex Cd standard solution was 0.09‰ (2SD, N = 214). The accuracy of the measurements was verified by measuring the Münster Cd standard solution and the results (+ 4.53 ± 0.08‰ of δ 114/110 Cd) were in good agreement with previously published values 30,43 .
To express the isotope fractionation between two components A and B, we used δ 114/110 Cd A−B that equaled the difference δ 114/110 Cd A − δ 114/110 Cd B .
The isotope composition of the whole plant Cd δ 114/110 Cd WP has be established according to the following: The Cd isotope composition of shoot and the whole plant (WP) relative to Spex Cd standard solution, as well as the isotopic variation between the different organs are shown in Table 2.
Data Analysis. Cadmium bioconcentration factor (BCF) was defined as the ratio of Cd in shoot or root of the plant to that in the nutrient solution. Cadmium translocation factor (TF) was described as the ratio of Cd in the Scientific RepoRts | 6:24309 | DOI: 10.1038/srep24309 shoot to that in the root. Tolerance index (TI) was defined as the ratio of the plant biomass after Cd treatments to that of the control group. The indexes were defined as follows: where, C organ (mg·kg −1 ) and C medium (mg·L −1 ) represent the Cd concentration in the shoot or root and the Cd concentration in the nutrient solution, respectively.
where, C shoot (mg·kg −1 ) and C root (mg·kg −1 ) represent the Cd concentration in the shoot and the Cd concentration in the root, respectively.
where, W Cd (g) and W control (g) represent the biomass after Cd treatment and the biomass of the control group, respectively.

Results
Cd concentration and mass in organs of R. communis and S. nigrum. The Cd concentrations in different organs of R. communis and S. nigrum are shown in Fig. 1a,b. The Cd concentration of the leaf is much higher in S. nigrum than that in R. communis, whereas it is equal to or lower in stem and root of S. nigrum than that of R. communis. Cd concentrations in different organs of R. communis exhibit a significant gradient with a progressive increase from upper to lower organs, by the order of leaf < stem < root, independently of the Cd concentration in the nutrient solution. In contrast, the Cd concentration in the leaf of S. nigrum is higher than that in the stem under low Cd conditions. It is essential to precisely determine mass-balances for Cd in the different organs when Cd transfer in the plants is investigated 13,44 . The Cd mass is calculated using the dry weight and Cd concentrations of the plant organs as shown in Fig. 1c-f. The total Cd mass in R. communis is higher than that in S. nigrum. The Cd mass in the root of R. communis is higher than that in the shoot independently of low or high Cd conditions. In contrast, Cd mass in the root of S. nigrum is much lower than that in the shoot under low Cd conditions. The Cd mass in the two tested plant species exhibits a consistent gradient that progressively increase from upper to lower organs, by the order of leaf < stem< root.
Cd bioconcentration factor, translocation factor and tolerance index of R. communis and S. nigrum. All bioconcentration factors (BCFs) of the four plant cultivars are higher than 1 ( Table 1). The BCFs of the four cultivars under soil condition are lower than those under hydroponic conditions, considering that Cd in soil occurs in complicated forms because of its association with many physicochemical environments that impact Cd availability. The root BCFs of different cultivars increase by the order of Zibo-5 > Zibo-8 > Zibo-6 > S. nigrum under low Cd conditions, whereas they increase by the order of Zibo-5 > Zibo-8 > S. nigrum > Zibo-6 under high Cd conditions. The shoot BCFs of different cultivars increase by the order of S. nigru m > Zibo-8 > Zibo-5 > Zibo-6 under low Cd conditions, whereas they increase by the order of S. nigrum > Zibo -6 > Zibo-8 > Zibo-5 under high Cd conditions. Consequently, the root BCFs are highest in Zibo-5, followed by Zibo-8, whereas the shoot BCFs are highest in S. nigrum.
The translocation factors (TFs) of four plant cultivars are low, which indicates that the Cd concentration is higher in root than that in shoot. The TFs of different cultivars increase by the order of S. nigrum > Zibo-8 > Zibo -6 > Zibo-5 under low Cd conditions, whereas they increase by the order of S. nigrum > Zibo-6 > Zibo-8 > Zibo -5 under high Cd conditions. Thus, S. nigrum accumulates the highest Cd concentrations during Cd translocation from root to shoot, whereas Zibo-5 accumulates the least, regardless of Cd concentration in solution.
A tolerance index (TI) based on biomass exposed to heavy metals is used to evaluate the heavy metal toxicity in the plants 45  According to an independent samples T-test (p < 0.05), the Cd treatments exert significant effects on transport and accumulation in Zibo-6, Zibo-8, and S. nigrum but have no significant effects on shoot accumulation and transport in Zibo-5. Overall, R. communis is characterised by a higher Cd tolerance, whereas S. nigrum has a higher potential to translocate Cd from root to shoot. In the organs of these four plant cultivars, more Cd is  Table 2). The stem of Zibo-5 is enriched in lighter isotopes relative to the root, whereas Zibo-6 and S. nigrum are enriched in heavy isotopes relative to the root. In contrast, the δ 114/110 Cd Stem-Root values of Zibo-8 behave differently under low and high Cd conditions. In low Cd conditions, the stem of Zibo-8 is depleted of heavy isotopes relative to the root, which is consistent with Zibo-5, whereas, in high Cd conditions, the stem of Zibo-8 is enriched in heavy isotopes relative to the root, which is consistent with Zibo-6 and S. nigrum. The three R. communis cultivars show similar distributions of heavy and light Cd isotopes in stem and leaf, which are different to S. nigrum. The leaf of the three R. communis cultivars is all enriched in lighter isotopes relative to the stem, whereas those of S. nigrum are depleted of light isotopes relative to the stem (Fig. 2).
The Cd isotope compositions in the organs of Zibo-5, Zibo-6, and S. nigrum under low and high Cd conditions behave similarly, but differently to Zibo-8. The observed isotopic fractionations between the solution and organs increase by the order of δ 114/110 Cd Root-Solution > δ 114/110 Cd Stem-Solution > δ 114/110 Cd Leaf-Solution for Zibo-5, whereas they increase in the reverse order by δ 114/110 Cd Leaf-Solution > δ 114/110 Cd Stem -Solution > δ 114/110 Cd Root-Solution for S. nigrum. In contrast, the isotope value of δ 114/110 Cd Stem-Solution in Zibo-6 is larger than values of δ 114/110 Cd Root-Solution and δ 114/110 Cd Leaf-Solution . The Cd isotopic fractionation between the solution and organs of Zibo-8 under low conditions behave similar to Zibo-5, but those under high conditions behave similarly to Zibo-6.

Discussion
The average δ 114/110 Cd WP-Solution values observed from solution to plants for Zibo-5, Zibo-6, Zibo-8, and S. nigrum are − 0.36‰, − 0.40‰, − 0.30‰ and − 0.46‰, respectively ( Table 2). The observed enrichment of light Cd isotopes is consistent with previous studies on other metal isotopes (e.g. Cu, Fe, Zn, Ca) in plants, except for Mg exhibiting isotopically heavy plant biomass 6,[13][14][15]23,46 . The physiological and molecular mechanisms of Cd hyperaccumulation and tolerance include root proliferation in Cd-rich substrate, influx into cytosol or vacuole by specific and non-specific transporters, and complexation of Cd by certain ligands in cells 47 . Based on the physiological and molecular mechanisms of Cd in higher plants, the speciation and diffusion in solution, adsorption on the root cell walls, uptake by ZIP proteins (Zinc-regulated transporter, iron-regulated transporter protein), complexation by phytosiderophores in solution and uptake of the entire complex through the membrane may   affect the metal isotope fractionation 6,14,44 . The Cd isotopic composition in root and shoot possibly reflects a combination of all these processes. Two possible abiotic processes could lead to isotope fractionation at the solution-root interface: diffusion and adsorption. Rodushkin et al. 48 found that lighter isotopes diffused faster than heavier isotopes and free ions diffused faster than complex ions. Diffusion from solution to root could lead to an enrichment of the lighter isotopes at the root surface. In addition, adsorption could also result in Cd isotope fractionation. A previous study 49 showed a small Cd isotope fractionation occurred during sorption of Cd to synthetic birnessite from low ionic strength solution, with lighter isotopes sorbed and heavier isotopes remaining in solution. In the present study, the δ 114/110 Cd Root-Solution in R. communis and S. nigrum is − 0.51‰ to − 0.28‰ ( Table 2). The root is enriched in the lighter Cd isotope. Therefore, diffusion may be a dominant process, leading to Cd isotope fractionation at the solution-root interface of R. communis and S. nigrum.
Cd transport across the root cell and other cell membranes are possibly metabolically controlled 14 . In addition, within plants Cd can be transported along the electrochemical gradient via carrier proteins and ion channels or against the electrochemical gradient via electrogenic pumps 13,14 . Carrier-mediated transport favours heavy isotopes because it involves covalent binding to a carrier protein on the outer side of the membrane, with subsequent release on the inner side as a result of conformational changes in the carrier 13 . Conversely, transport through ion channels or via electrogenic pumps favours light isotopes because of its greater diffusion coefficient 14 . The observed net enrichment of the lighter isotopes in root and the differences between the plant cultivars, therefore, suggest that membrane transport is dominated by ion channels and electrogenic pumps rather than by carrier-mediated transport.
The differences in Cd isotopic fractionation from root to stem of four plant cultivars might be due to the different Cd supply limitation, which is associated with the tolerance of plants. Although the Cd mass in nutrient solutions is sufficiently supplied, the Cd mass translocated in the extracellular and cellular plant organs might be limited in different plant cultivars. The magnitudes of the isotopic shifts during the solution-to-organ transfer slightly increase with decreasing Cd concentrations in the organs (Fig. 3). Moreover, the plant biomass is higher under low Cd conditions than that under high Cd conditions (Fig. 1c,d). Therefore, the Cd stress affects the magnitude of the isotopic shift during the solution-to-organ transfer. Gault-Ringold et al. 34 proposed that Cd uptake of phytoplankton did not result in no net Cd isotopic fractionation under 'supply-limited' condition, but it could be kinetically driven resulting in Cd isotopic fractionation under sufficiently high Cd levels. This could explain the different Cd isotope fractionation from root to stem between the cultivars.
The variation in the Cd isotopic composition between stem and leaf in R. communis and S. nigrum is distinct. It may be attributed to the complexation with organic acids, phytochelatins (PCs), and metallothionein in the xylem of S. nigrum. Sun et al. 50 identified that complexation with organic acids, phytochelatins (PCs), and metallothionein was an important mechanism for Cd detoxification, transportation and storage in S. nigrum. In addition, previous work 14 also showed that complexation with organic ligands led to an enrichment of heavy isotope in the organs. Figure 1g,h show that a higher amount of Cd is stored in the root of R. communis, whereas more Cd is translocated to the stem and leaf of S. nigrum. This can be explained by the complexation of organic acids, phytochelatins (PCs), and metallothionein in S. nigrum with Cd, which catalyse the translocation of Cd from root to shoot. The observed difference between root and shoot in the Cd-tolerant and -hyperaccumulator species may reflect the different Cd transportation mechanisms of the species.

Conclusions
In the present study, the Cd isotope measurements show an isotopic shift to lighter isotopes during Cd transport from the nutrient solution to the plant organs of the Cd-tolerant R. communis and the Cd-hyperaccumulator S. nigrum. The observed isotope fractionation is enriched with the Cd isotope reservoirs in nature. In addition, the variation of the Cd isotopic compositions in leaf and stem differs between R. communis and S. nigrum implying different mechanisms of Cd translocation to the xylem in the Cd-tolerant and -hyperaccumulator species. Cd isotope fractionations of different organs provide new information to identify the chemical processes controlling Cd uptake and translocation in plants and organisms. Plant uptake is an important factor of isotopic variation in the Cd biogeochemical cycle. Thus, Cd isotope fractionation by plants needs to be taken into account in future investigations on environmental pollution using Cd isotopes. Overall, studies on Cd isotopes in plants lay the groundwork for understanding the biogeochemical Cd cycle and mechanisms of plant Cd acquisition and allocation.  24,26,28,29,[34][35][36][37][38][53][54][55][56][57][58] . GSS-1 and GSD-12 are geological reference materials. GSS-1 is a dark brown podzolitic soil, typical of the mountainous terrain of Northeast China. The underlying granitic bedrock is a part of a leadzinc mineralisation district. GSD-12 stream sediment is from a tributary draining river in the Yangchun ore field (Cu, W, Sn), Guangdong, China.