Relative abundance of chemical forms of Cu(II) and Cd(II) on soybean roots as influenced by pH, cations and organic acids

Little information is available on chemical forms of heavy metals on integrate plant roots. KNO3 (1 M), 0.05M EDTA at pH6 and 0.01 M HCl were used sequentially to extract the exchangeable, complexed and precipitated forms of Cu(II) and Cd(II) from soybean roots and then to investigate chemical form distribution of Cu(II) and Cd(II) on soybean roots. Cu(II) and Cd(II) adsorbed on soybean roots were mainly exchangeable form, followed by complexed form, while their precipitated forms were very low under acidic conditions. Soybean roots had a higher adsorption affinity to Cu(II) than Cd(II), leading to higher toxic of Cu(II) than Cd(II). An increase in solution pH increased negative charge on soybean and thus increased exchangeable Cu(II) and Cd(II) on the roots. Ca2+, Mg2+ and NH4+ reduced exchangeable Cu(II) and Cd(II) levels on soybean roots and these cations showed greater effects on Cd(II) than Cu(II) due to greater adsorption affinity of the roots to Cu(II) than Cd(II). L-malic and citric acids decreased exchangeable and complexed Cu(II) on soybean roots. In conclusion, Cu(II) and Cd(II) mainly existed as exchangeable and complexed forms on soybean roots. Ca2+ and Mg2+ cations and citric and L-malic acids can potentially alleviate Cu(II) and Cd(II) toxicity to plants.

uptake to and by plants were rarely found in literature 26,27 . The relative abundance of the various chemical forms of heavy metal cations on plant roots affects the toxicity of metals to plants and the uptake of these metals by plant roots. While, few studies have investigated this topic. The mechanisms by which heavy metals are adsorbed by plant roots and the relative abundance of the chemical forms of adsorbed heavy metals on the roots need to be better understood.
Cu(II) is mainly adsorbed by soils and soil components specifically 28 , while Cd(II) is adsorbed mainly through electrostatic mechanisms and is found in the exchangeable form on soil particles 29 . The different chemical forms of heavy metals exhibit differing bioavailability 30 . Similarly, Nishizono et al. 31 found that the cell walls of plant roots showed a higher affinity to Cu 2+ than that to Zn 2+ and Cd 2+ 31 . The cations of the various heavy metals may also be adsorbed by plant roots through different mechanisms. Cu(II) and Cd(II) were selected for this study to examine the relative abundance of their various chemical forms on soybean roots after adsorption.
Carboxyl and phenolic groups are commonly found on plant roots 32 . The deprotonation of these acidic groups gives plant roots a negative charge and enables roots to adsorb heavy metals through electrostatic interaction. In addition, these acidic groups have a strong complexing ability and thus can adsorb heavy metals specifically through the formation of surface complexes with metal cations on the plant root surfaces 33 . The degree of dissociation of weak acids increases with increasing system pH, so that pH not only affects the charge characteristics of a root surface, but can also affect the hydrolysis of heavy metal cations in solution, thereby affecting their reaction with functional groups on the root surface.
Environmental stress causes plant roots to actively or passively release soluble organic acids into the rhizosphere. Organic acids contain a large quantity of carboxyl functional groups and form soluble complexes with heavy metal cations 34 . Consequently, organic acids may affect adsorption and combinations of heavy metals form on the root surface. Exogenous cations of Ca 2+ , Mg 2+ and NH 4 + effectively inhibit absorption of Mn(II) by barley and thus reduce Mn(II) toxicity to the plant 35 . Cations such as Ca 2+ , Mg 2+ and NH 4 + compete with Al 3+ for adsorption sites and inhibit adsorption of Al 3+ by the roots of soybean and maize 33 . Exogenous cations may also inhibit the adsorption of heavy metal cations by plant roots and affect the relative abundance of the chemical forms of heavy metals on plant roots. However, the effects of cations and organic acids on the relative abundance of the chemical forms of heavy metals on plant roots requires further research.
The objectives of this study were: (1) to investigate the relative abundance of the various chemical forms of Cu(II) and Cd(II) on soybean roots after adsorption by the roots; and (2) to examine the effects of exogenous cations and organic acids on the relative abundance of the chemical forms of Cu(II) and Cd(II) on soybean roots. The findings should improve our understanding of the chemical behaviors of these heavy metals at the root/soil interface, and suggest means of reducing the toxicity of heavy metals to plants.
Neutral salts were previously used to extract exchangeable heavy metals adsorbed on soils and exchangeable Al from plant roots 10,33 . EDTA was used to extract heavy metal cations adsorbed specifically by soil organic matter and crop straw biochars in previous studies 12,13 . The mechanism for specific adsorption of Cu(II) and Cd(II) on soybean roots was formation of surface complexes of the metals with functional groups on the roots, similar to the specific adsorption of heavy metals by organic matter and biochars. HCl was used to extract the precipitates of Al hydroxides from plant roots in previous studies 33,36 . Cu(II) and Cd(II) mainly formed the precipitates of hydroxides on soybean roots in present study. Therefore, the neutral salt of KNO 3 , EDTA, and HCl were used sequentially to extract the exchangeable, complexed and precipitated forms of Cu(II) and Cd(II) from soybean roots in this study.

Results and Discussion
The relative abundance of the chemical forms of Cu(II) and Cd(II) on soybean roots. Cu(II) and Cd(II) adsorbed onto soybean roots could be differentiated into exchangeable, complexed, and precipitated forms. Under strong acidic conditions of pH 4.2, Cu(II) and Cd(II) was mainly found on soybean roots in the exchangeable forms, followed by complexed forms, while the amounts of precipitated Cu(II) and Cd(II) were very low (Fig. 1). For example, when the initial concentration of Cu(II) and Cd(II) was 0.4 mM, the exchangeable Cu(II) and Cd(II) accounted for 68.9% and 86.8% of total adsorbed Cu(II) and Cd(II), and the complexed Cu(II) and Cd(II) accounted for 30.5% and 13.2% of total adsorbed Cu(II) and Cd(II), respectively. Both the exchangeable and complexed Cu(II) increased significantly with increasing initial Cu(II) concentration (P < 0.05). Similarly, the exchangeable Cd(II) was significantly increased (P < 0.05) as the initial concentration of Cd(II) rose. The complexed Cd(II) increased significantly as its initial concentration increased from 0.05 mM to 0.4 mM (P < 0.05), while the increase was not significant when the initial concentration increased from 0.4 mM to 1.0 mM. Comparing the two heavy metals, the amount of exchangeable Cu(II) was almost the same as that of the exchangeable Cd(II) at initial concentrations of 0.05 and 0.4 mM. However, the amount of complexed Cu(II) was much higher than that of complexed Cd(II) (Fig. 1). For example, at initial concentrations of 0.05, 0.4 and 1.0 mM, the amount of exchangeable Cu was 1.02, 0.99, and 1.15 times the content of exchangeable Cd(II) respectively, while the content of complexed Cu(II) was 2.43, 2.89, 3.82 times the content of complexed Cd(II), respectively. These results suggest that the adsorption affinity of soybean roots to Cu(II) was much greater than to Cd(II), leading to the higher specific adsorption capacity of the roots to Cu(II) than to Cd(II). This resembles the Cu(II) and Cd(II) adsorption capacity of soil organic matter and crop straw biochars 13,37 , caused by the stronger complexing ability of organic functional groups with Cu(II) than with Cd(II) 38 .
The effect of Cu(II) and Cd(II) on the zeta potential of soybean roots provides evidence for the differences between the formation of surface complexes of Cu(II) and Cd(II) with organic functional groups on soybean roots. The zeta potential of soybean roots was negative (Fig. 2), suggesting that soybean roots carry a negative charge. The adsorption of Cu(II) and Cd(II) by soybean roots reduced the negativity of the zeta potential, suggesting that surface complexes of Cu 2+ and Cd 2+ with functional groups form on the root surfaces. This resembles the effect of the formation of surface complexes of Cu 2+ with surface functional groups on the zeta potential observed on crop straw biochars 39 . The adsorption of Cu(II) led to a greater change in zeta potential of soybean roots than did the adsorption of Cd(II) (Fig. 2), suggesting that a greater number of Cu(II) complexes formed than those of Cd(II) on soybean roots. This was consistent with the relative abundance of complexed Cu(II) and Cd(II) on soybean roots (Fig. 1).
The abundance of the chemical forms of Cu(II) and Cd(II) on soybean roots may be related to their toxicities to the plant. The effects of different concentrations of Cu(II) and Cd(II) on soybean root elongation were compared and the results are shown in Fig. 3. Root elongation was significantly inhibited with the increasing concentrations of these heavy metals, suggesting that increasing amounts of exchangeable and complexed Cu(II) and Cd(II) on soybean roots increased their toxicities to the plant and the resulting inhibition of soybean root growth. The inhibiting effect of Cu(II) on root elongation was greater than that of Cd(II) at the same concentration, suggesting that Cu(II) was more toxic to the plant than Cd(II). This was consistent with previous observations by other researchers 40 . The amount of exchangeable Cu(II) on soybean roots was almost the same as exchangeable  Cd(II), but complexed Cu(II) was much more abundant than complexed Cd(II), which may account for the greater toxicity of Cu(II) to soybean than Cd(II).
Effect of pH on the relative abundance of the chemical forms of Cu(II) and Cd(II). As the pH increased, the amount of exchangeable Cu(II) on soybean roots increased significantly (P < 0.05) (Fig. 4), suggesting that the increased negative charge on the roots induced by increasing pH led to an increase in exchangeable Cu(II). The amount of complexed Cu(II) also increased with increasing pH, but the increase was not significant (Fig. 4). Increasing pH enhanced dissociation of acidic functional groups on the roots and hence the complexing ability of the functional groups with Cu(II). The amount of exchangeable Cd(II) also increased with increasing pH, although not to a significant degree. The increase in pH did not affect the complexed Cd(II) on the roots. The effect of pH on Cu(II) adsorption by soybean roots was greater than on Cd(II) adsorption, suggesting that Cu(II) adsorption was more sensitive to pH change than Cd(II) adsorption. Figure 5 indicates that the deprotonation of the acidic functional groups on soybean roots increased with increasing pH, increasing the negative charge on the roots and resulting in increased negative zeta potential of the roots. This provides evidence for the increases of exchangeable Cu(II) and Cd(II) on soybean roots with increasing pH, as shown in Fig. 4.   Figure 4 also shows a greater amount of precipitated Cu(II) on soybean roots than precipitated Cd(II), due to the greater hydrolyzing ability of Cu 2+ compared with Cd 2+ . The hydrolysis constant (pK) is an important parameter in characterizing the hydrolyzing ability of heavy metal cations. Smaller pK values indicate stronger hydrolyzing potential. The pK values of Cu 2+ and Cd 2+ reported in the literature are 6.5 and 9.7, respectively 41 , indicating that the hydrolyzing ability of Cu 2+ is greater than Cd 2+ . This accounts for the greater amount of precipitated Cu(II) at pH 5 than at pH 4.2 and 4.6.
Cation effects on the relative abundance of the chemical forms of Cu(II) and Cd(II). The addition of exogenous Ca 2+ , Mg 2+ and NH 4 + significantly reduced the exchangeable Cu(II) on soybean roots by 13.2%, 8.8% and 11.0 %, respectively, compared with the control (Fig. 6), but the decrease in exchangeable Cu(II) was not statistically significant. Ca 2+ and Mg 2+ significantly reduced the exchangeable Cd(II) on soybean roots by 76.1% and 51.1% (P < 0.05). NH 4 + decreased the exchangeable Cd(II) by 17.7% compared with the control, but this decrease was not significant. Cations of Ca 2+ , Mg 2+ and NH 4 + decreased exchangeable Cu(II) and Cd(II) on soybean roots through competition with Cu(II) and Cd(II) for exchangeable sites on negatively charged roots.  The presence of these cations led to a greater decrease in exchangeable Cd(II) than exchangeable Cu(II) due to the greater adsorption affinity of the roots to Cu(II) than Cd(II), as mentioned above. Of the three cations, Ca 2+ and Mg 2+ had a much greater effect on exchangeable Cd(II) than did NH 4 + due to the higher competitive ability for adsorption sites of the bivalent cations of Ca 2+ and Mg 2+ compared with the monovalent cation NH 4 + . Ca 2+ , Mg 2+ and NH 4 + significantly reduced the amount of complexed Cu(II) by 23.5%, 18.4% and 24.3%, respectively, compared with the control (P < 0.05). Ca 2+ significantly reduced the amount of complexed Cd(II), while Mg 2+ and NH 4 + had no significant effect compared with the control (Fig. 6). The presence of cations of Ca 2+ , Mg 2+ and NH 4 + inhibited the adsorption of Cu(II) and Cd(II) by soybean roots and may therefore decrease the toxicities of these metals to the plant and their absorption by soybean roots. The observations of Hardiman et al. 42 support this conclusion 42 . They found that the presence of base cations inhibited the uptake of Cd(II) by soybean roots and the effects of these cations were ranked: Ca 2+ > Mg 2+ > K + > Na + .

Effects of organic acids on the relative abundance of the chemical forms of Cu(II) and
Cd(II). The effects of low-molecular-weight (LMW) organic acids on the relative abundance of the chemical forms of Cu(II) and Cd(II) are shown in Fig. 7. The presence of L-malic and citric acids significantly decreased the amounts of exchangeable and complexed Cu(II) on soybean roots compared with the control (P < 0.05). Lactic acid also reduced the amounts of exchangeable and complexed Cu(II) on soybean roots, but the decrease in exchangeable and complexed Cu(II) was not significant. The three organic acids also decreased the amount of precipitated Cu(II) on soybean roots (Fig. 7). During adsorption experiments of Cu(II) and Cd(II), the solution pH was kept constant at 4.2. Therefore, the effects of these organic acids on the relative abundance of the chemical forms of Cu(II) and Cd(II) on soybean roots were mainly attributed to their complexation with both metal cations in solutions. The stability constants (logK) for complexes of citric, L-malic and lactic acids with Cu 2+ are 5.9, 3.33 and 3.02, respectively 43,44 . Compared with lactic acid, the stronger complexing ability of citric and L-malic acids led to higher levels of Cu(II) remaining in the nutrient solution through the formation of more complexes with Cu(II) and thus increased their inhibitions on Cu(II) adsorption by soybean roots. The formation of complexes between the organic acids and Cu 2+ also inhibited the precipitation of Cu(II) on soybean roots (Fig. 7).
The presence of citric acid significantly decreased the amount of exchangeable Cd(II) (P < 0.05), but did not reduce the amount of complexed Cd(II) significantly, compared with the control (Fig. 7). Both L-malic and lactic acids did not significantly decrease the amounts of exchangeable and complexed Cd(II) on soybean roots. These organic acids showed less effect on the adsorption and relative abundance of the chemical forms of Cd(II) compared with Cu(II), due to their relative lower complexing ability with Cd(II) than with Cu(II). The stability constants (logK) for complexes of citric, L-malic and lactic acids with Cd 2+ are 3.15, 1.34 and 1.07, respectively 43,45 ,

Figure 7. Effects of organic acids on adsorption of Cu(II) and Cd(II) and their chemical forms on 14-dayold roots of soybean at pH 4.2.
much lower than these for the complexes with Cu 2+ . The logK of citric acid with Cd 2+ is greater than those of L-malic and lactic acids with Cd 2+ , and is responsible for the significant decrease in exchangeable Cd(II) caused by citric acid. LMW organic acids arise mainly from the exudates of plant roots and the various stages of decomposition of plant residues in soils 34 . These organic acids can form complexes with heavy metal cations in soil solutions and thus inhibit the adsorption of heavy metals onto plant roots, which may also decrease the availability and toxic effect of these metals on plants, especially in the case of Cu(II) due to the greater logK of organic acid complexes with Cu 2+ .

Conclusions
Cu(II) and Cd(II) adsorbed on soybean roots mainly occurred in the exchangeable forms, followed by complexed forms, while the precipitated forms occurred at low levels under acidic conditions. Soybean roots had a higher adsorption affinity to Cu(II) than Cd(II), leading to greater toxicity of Cu(II) to the plant than Cd(II). The formation of complexes of Cu 2+ and Cd 2+ with functional groups on soybean roots reduced the negative level of the zeta potentials and the adsorption of Cu(II) showed a greater effect on the zeta potential of soybean roots than did the adsorption of Cd(II). The increase in solution pH increased the negative charge on soybean roots and thus increased the amounts of exchangeable Cu(II) and Cd(II) on the roots. Cu(II) adsorption by soybean roots was more affected by pH than was Cd(II) adsorption. The presence of Ca 2+ , Mg 2+ and NH 4 + decreased the amounts of exchangeable Cu(II) and Cd(II) on soybean roots through competition for exchangeable sites on negatively charged roots with Cu 2+ and Cd 2+ and the cations showed greater effects on Cd(II) than Cu(II) due to the greater adsorption affinity of the roots to Cu(II) than Cd(II). Ca 2+ showed the greatest effect on exchangeable Cd(II), followed by Mg 2+ , while NH 4 + showed the least effect on exchangeable Cd(II). The presence of L-malic and citric acid decreased the amounts of exchangeable and complexed Cu(II) on soybean roots. Citric acid decreased the amount of exchangeable Cd(II) on soybean roots. In summary, the cations Ca 2+ and Mg 2+ and the LMW organic acids of citric and L-malic acids can decrease the availability of the heavy metals Cu(II) and Cd(II) and potentially alleviate their toxicity to plants.
The differences in adsorption affinity of soybean roots to Cu(II) and Cd(II) led to the differences in the relative abundance of the chemical forms of both metals and thus affected their toxicity to the plant. The exogenous cations of Ca 2+ , Mg 2+ and NH 4 + and organic acids of citric, L-malic and lactic acids decreased the exchangeable and complexed forms of Cu(II) and Cd(II) on soybean roots to various extents, mainly depending on the competing ability of these cations for adsorption sites with the heavy metals and the complexing ability of these organic acids with Cu 2+ and Cd 2+ . These findings are of fundamental significance in understanding the chemical behaviors of heavy metals at the root/soil interface, and suggest means of alleviating the toxicity of heavy metals to plants.

Materials and Methods
Plant materials and cultivation. Soybean (Glycine max cv. Xudou 14) was used in this study. The soybean seeds were surface-sterilized with 10% H 2 O 2 for 10 min, washed with deionized water three times, soaked for 4 h in deionized water, and then germinated in sterilized sand in the dark. From the fourth day, the plant seedlings were cultivated in a nutrient solution (initial pH 4.2) and grown in a controlled environment growth chamber at a temperature of 27 °C, a day-length of 14 h, a light intensity of 375 μ mol photonm −2 s −1 , and a relative humidity of 70%. The composition of the nutrient solution was as follows: (mM) 2.0 (NH 4

Cu(II) and Cd(II) adsorption/desorption experiments with entire roots.
Before the adsorption experiments, the plant roots were placed in deionized water for 1 h to remove excess nutrient ions from the root surfaces. The plant roots were then separated from the stems, washed with deionized water three times, and then carefully blotted dry. About 8 g of fresh roots were placed in a 300-mesh nylon bag (preliminary experiments showed that the nylon bag did not adsorb Cu(II) and Cd(II)), then immersed in 1-L Cu(II) and Cd(II) solutions and magnetically stirred. After 2 h, the bag containing the plant roots was removed from the adsorption solution and washed with distilled water three times, and any excess water was removed with filter paper. Then the roots were sequentially placed in 1 L 1 M KNO 3 , 0.05 M EDTA at pH 6, and 0.01 M HCl for 1 h each to extract the exchangeable, complexed, and precipitated Cu(II) and Cd(II) from the roots. After each extraction, the roots were washed with distilled water three times and the excess water was blotted off in preparation for the next extraction.
The exchangeable heavy metals were absorbed by the soybean roots by electrostatic attraction and can be exchanged with cations of indifferent electrolytes. High concentration KCl has previously been used to extract exchangeable heavy metals from soils 46 and from plant roots. The complexed heavy metals comprised the binding state of the metals through the formation of surface complexes with functional groups on the plant roots. The precipitated heavy metals bound to the roots as hydroxyl polymers or oxides formed on the root surfaces. The concentrations of Cu(II) and Cd(II) in the extractants were determined by means of atomic adsorption spectroscopy.
To investigate the effect of the initial concentrations of Cu(II) and Cd(II) on their adsorption and the relative abundance of their chemical forms, 0.05, 0.4 and 1.0 mM of Cu(II) and Cd(II) solutions were used. Similar experimental procedures were then conducted. The solution was kept at pH 4.2 throughout the experiment.
To investigate the effect of pH on the adsorption of Cu(II) and Cd(II) and the relative abundance of their chemical forms, the solution was kept at pH 4.2, 4.6 and 5.0. The initial concentration of Cu(II) and Cd(II) was 0.4 mM. Similar experimental procedures were then conducted.
To investigate the effects of cations on the adsorption of Cu(II) and Cd(II) and the relative abundance of their chemical forms, we prepared solutions containing mixtures of 0.4 mM of Cu(II) or Cd(II) and 1 mM each of Ca 2+ Scientific RepoRts | 6:36373 | DOI: 10.1038/srep36373 (CaCl 2 ), Mg 2+ (MgCl 2 ) or NH 4 + (NH 4 Cl) and carried out similar experimental procedures. The solution pH was kept at 4.2 throughout the experiment.
The effect of organic acid on the adsorption of Cu(II) and Cd(II) and the relative abundance of their chemical forms was investigated by preparing mixed solutions containing 0.4 mM of Cu(II) or Cd(II) and 1 mM each of L-malic acid, lactic acid or citric acid and carrying out similar experimental procedures. The solution pH was kept at 4.2 throughout the experiment.
The determination of the zeta potential of the root surface. About 10 g of fresh roots were reacted with 0.4 mM solutions of Cu(II) and Cd(II) for 2 h, at a constant solution pH of 4.2. After adsorption of Cu(II) and Cd(II), the roots were washed with distilled water three times and then air-dried. A streaming potential apparatus was used to measure the zeta potential of soybean roots with and without adsorbed Cu(II) and Cd(II) 47 . The air-dried soybean roots were cut to a length of 2 cm, and 0.25 g of the roots were horizontally loaded into a sample cell. NaCl (1.5 L, electrolyte solution) with a conductivity of 80 μ S cm −1 and pH 4.2 was put into an electrolyte container and then flowed through the soybean roots for 40 min using a magnetic pump as a driver. The used electrolyte solution was then discarded, another 1.5 L of the electrolyte solution was put into the electrolyte container and flowed through the soybean roots for 10 min until equilibrium was reached, after which the flow rate of electrolyte was increased stepwise by adjusting the valve and the streaming potential Δ E and pressure difference Δ P were recorded. The zeta potential of the soybean roots was calculated using the Helmholtz-Smoluchowski equation.
Root elongation measurement. The roots of 4-day-old seedlings were exposed for 48 h to 0. Where RLC1 is the initial root length of untreated controls; RLC2 is the final root length of untreated controls after 48 h (control); RLT1 is the initial root length before exposure; RLT2 is the final root length after exposure to Cu(II) and Cd(II) for 48 h.