SEARCH

SEARCH BY CITATION

Keywords:

  • Metal mixtures;
  • Plants;
  • Biotic ligands;
  • Toxic equivalency factor;
  • Stability constant

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

The biotic ligand model (BLM) was applied to predict metal toxicity to lettuce, Lactuca sativa. Cu2+ had the lowest median effective activity (EA50M), compared with Ag+ and Zn2+ (EA50Cu = 2.60 × 10−8 M, EA50Ag = 1.34 × 10−7 M, EA50Zn = 1.06 × 10−4 M). At the 50% response level, the fraction of the total number of biotic ligands occupied by ions (f50M) was lowest for Ag+ among the metals (f50Ag = 0.22, f50Cu = 0.36, f50Zn = 0.42). Cu2+ had the highest affinity for biotic ligands compared with Ag+ and Zn2+, as shown by stability constants of the cation–biotic ligand binding, expressed as log KMBL (log KCuBL = 7.40, log KAgBL = 6.39, log KZnBL = 4.00). Furthermore, the BLM was combined with the toxic equivalency factor approach in predicting toxicity of mixtures of Cu2+–Zn2+ and Cu2+–Ag+. The fraction of biotic ligands occupied by ions was used to determine the relative toxic potency of metals and the toxic equivalency quotient (TEQ) of mixtures. This approach allowed for including interactions in estimating mixture toxicity and showed good predictive power (r2 = 0.64–0.84). The TEQ at the 50% response level (TEQ50, Cu2+ equivalents) for Cu2+–Zn2+ mixtures was significantly lower than the value for Cu2+–Ag+ mixtures. Joint toxicity depended on both TEQ and specific composition of the mixture. The present study supports the use of the accumulation of metal ions at the biotic ligands as a predictor of toxicity of single metals and mixtures. Environ. Toxicol. Chem. 2013;32:137–143. © 2012 SETAC


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

The biotic ligand model (BLM) is usually applied to predict toxicity of single metals, taking into account the effects of common cations, for example, H+, Ca2+, and Mg2+. According to the BLM concept, ions compete with each other for transport sites at the biotic ligands, and this competition acts as a mechanism for ion–ion interactions 1, 2. This assumption is based on physiological findings indicating that toxic cations, such as Cu2+ and Ag+, may inhibit the uptake of Na+ or Ca2+ for specific binding sites at the fish gill, leading to adverse effects 3–5. Furthermore, the assumption potentially allows taking into account interactions between different metal ions in assessment of mixture toxicity 1, 6, 7. In particular, it is possible to predict how different metals interact with one another if their stability constants are known. If two metals compete for binding to the same site of toxic action, the total amount of the metals bound to the site would be a key property, determining mixture toxicity. Alternatively, if competitive binding does not occur following exposure to metal mixtures, bioavailability of each component estimated by the BLM can be a reliable predictor of mixture toxicity through the effects addition model.

In the present study, the assumption of competitive binding—that is, metals after exposure to their mixtures may compete for transport sites at the biotic ligands—was applied to metal mixtures. This assumption is supported by the observation that the uptake of metals usually involves transporter proteins 8. Furthermore, based on the functions of the transporters, the physiological mechanism of metal binding can be classified into three categories 9. According to this classification, Cu2+ and Zn2+ may bind to the same transporters that are responsible for the uptake of divalent cations, whereas the uptake of Ag+ is related to the participation of transporters for monovalent cations. Consequently, mixtures of Cu2+–Zn2+ and Cu2+–Ag+ were chosen in the present study in an attempt to model toxicity of both competitive and noncompetitive mixtures with the assumption that Cu2+ and Zn2+ compete for binding sites at the biotic ligands, whereas this competition does not occur between Cu2+ and Ag+.

According to the BLM concept, toxic effects result from binding of metal ions to biotic ligands. In other words, the extent of toxic effects is determined by the fraction of the biotic ligands occupied by metal ions in the total number of biotic ligands. This fraction was used as a basic unit in the toxic unit approach for estimating toxicity of metal mixtures to the duckweed Lemna paucicostata by Hatano and Shoji 10 and to the bacteria Vibrio fischeri by Jho et al. 11. This modeling approach was found to result in higher predictive potential than the free ion activity model and the total metal concentration model. In other words, the accumulation of metal ions at the biotic ligands was a better indicator of toxicity of metal mixtures compared with metal concentrations or activities in the solution. Both the BLM and the free metal ion activity model are based on the assumption that free ions are the main reactive species of metals determining metal toxicity, so the authors mentioned above further suggested that the advantage of the BLM over the other models was a result of the integration of competitive binding of metal ions to biotic ligands.

The toxic equivalency factor (TEF) approach has been widely used in assessment of mixture toxicity. In this approach, a substance is used as a reference compound (TEF = 1) to which others are related 12. The TEF of each mixture component reflects its toxic potency relative to the reference compound. This approach is usually performed based on concentrations of substances in mixtures. Particularly, the toxic equivalency quotient (TEQ) of mixtures is calculated as the sum of the equivalency concentration of each component, which is a product of the concentration and the TEF of the compound. Therefore, the exposure level can be expressed by a single concentration of the reference substance, which represents the overall toxicity of the mixture, assuming no interactions between different components. However, the concentration of metals in the solution might not be a reliable predictor of their toxicity, because chemical properties of the exposure medium, such as chelators and pH, affect binding of the metals with biotic and abiotic ligands, influencing metal uptake 13.

The present study sought to predict toxicity of binary metal mixtures (Cu2+–Zn2+ and Cu2+–Ag+) to lettuce, Lactuca sativa, by combining the BLM and the TEF approach for the first time. In particular, the accumulation of metals at the biotic ligands, which determines toxicity of the metals following single exposure according to the BLM principle, was used to determine the TEF of the metals in mixtures. The accumulation of metals at the biotic ligands in turn is influenced by interactions between the metals and other competing cations. Therefore, interactions between different metal ions in mixtures at the biotic ligands can be integrated in estimating the toxicity of metal mixtures, overcoming the disadvantages of the conventional TEF approach based on metal concentrations.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Toxicity assays

Metal toxicity was assessed in terms of relative root elongation (RRE; percentage) in hydroponic experiments using Steiner solution as the test medium 14. Cu2+, Zn2+, and Ag+ were added to the Steiner solution as nitrate salts while the concentrations of other cations in the solution were kept at the background level of the default medium. The solution pH was kept at 7 by using 3-[N-morpholino]propane sulfonic acid buffering 14.

Metal measurements and speciation

Free Zn2+ activities in solutions were derived from the total Zn2+ concentrations in the exposure solution, which were the sum of the Zn2+ concentration in Steiner solution and the Zn2+ concentration added. The determination of free Zn2+ activities was performed by the speciation model Windermere humic–aqeous model VI, with Steiner solution as the default medium 15. The chemical composition of the Steiner solution used for chemical speciation is given in Supplemental Data, Table S1. In addition, free ion activities of H+, Cu2+, and Ag+ were measured by using H, Cu, and Ag sulfide ion-selective electrodes (Metrohm), which had been calibrated at different concentrations of these cations in the solution 14. The free ion activities of Cu2+, Zn2+, and Ag+ in all test solutions are presented in Supplemental Data, Table S2.

Toxicity of single metals

Free metal ions were considered the main reactive species determining toxicity of the metals. As reviewed in a previous study, different results have been reported on the effects of common cations, for example, Na+, K+, Ca2+, and Mg2+ on toxicity of Cu2+ 14. This difference may be related to the species and the exposure levels investigated. The present study and the study of Le et al. 14 were carried out on the same plant species and with the same test medium. As a result, the findings reported from the study of Le et al. 14 were applied in the present research. According to these authors, Cu2+ toxicity to L. sativa was significantly inhibited by protons 14. The study also indicated that effects of Na+, Ca2+, Mg2+, and K+ on Cu2+ toxicity to this plant species could not be quantified by the BLM, because the pattern of these impacts was inconsistent at the concentration range studied. Therefore, in the present study, we assumed that H+ competes with toxic cations—that is, Cu2+, Ag+, and Zn2+—for binding sites at the biotic ligand while keeping concentrations of all other cations at the background level of the default medium. In other words, the accumulation of these toxic ions, which determines their toxicities according to the BLM principle, is influenced by binding of H+ with biotic ligands. Accordingly, the faction of the total number of biotic ligands occupied by metal ion Mn+—that is, Cu2+, Ag+, or Zn2+ (fM)— is determined as follows

  • equation image(1)

where [BL]T (mol/L) is the total number of biotic ligands; KHBL and KMBL (L/mol) are stability constants of binding of H+ and Mn+ to biotic ligands, respectively; and {H+} and {Mn+} (mol/L) are free ion activities of H+ and Mn+ in the solution, respectively.

The free ion activity of metal ion Mn+ in the solution that results in a 50% reduction in the growth of lettuce roots is termed as median effective activity (EA50M). A detailed description of the derivation of a BLM for single metals, for example, determination of the stability constant KMBL and the fraction of the total number of biotic ligands occupied by metal ions at the 50% response level f50M, has been presented in a previous article 14. Specifically, f50M was determined by fitting a sigmoid curve to the relationships between fM and the RRE in GraphPad Prism, according to the following equation

  • equation image(2)

Toxicity of binary metal mixtures (Cu2+–Zn2+ and Cu2+–Ag+)

Toxicity of mixtures of Cu2+–Zn2+ and Cu2+–Ag+ were modeled from the toxicological data for the single metals, that is, KMBL and f50M, by combining the BLM and the TEF approach. Particularly, the TEF of metals in mixtures and the TEQ of mixtures were determined based on the fraction of the total number of biotic ligands occupied by metal ions. Cu2+ was considered the reference metal ion to which toxicity of Zn2+ and Ag+ was related. This selection was used because of the presence of Cu2+ in both mixtures studied and a high level of concern about its environmental effects and subsequent wide availability of toxicological data. Accordingly, the TEF of metals in mixtures (TEFM; M denotes Cu2+, Zn2+, or Ag+) was determined according to the following equation

  • equation image(3)

where f50M and f50Cu are the fractions of the biotic ligands occupied by Mn+ and Cu2+ in the total number of biotic ligands at the 50% response level following exposure to these metal ions individually, respectively 14. The TEFM represents the comparative toxic potency of metal ion Mn+ in mixtures. Furthermore, the TEQ of mixtures (Cu2+ equivalents), which reflects the overall toxicity of the mixture, was calculated from the TEFM of mixture components and the fraction of the total number of biotic ligands occupied by metal ions in the mixture fM according to the following equation.

  • equation image(4)

Equations 3 and 4 were derived from the common expression of the TEF approach based on substance concentrations 6. According to the BLM, fM is determined based on assumed independence of the complexation capacity of ions on water quality characteristics 14, 16. The calculation of fM following exposure to metal mixtures was based on the stability constant KMBL determined from the toxicological data following exposure to single metals as described below. Toxic effects, in terms of root growth represented by RRE (%), were expressed in relation to the TEQ (Cu2+ equivalents) according to the following equation:

  • equation image(5)

where TEQ50 (Cu2+ equivalents) is the TEQ of the mixture at the 50% response level, and β is the slope parameter. Estimates of these coefficients and statistical parameters, for example, 95% confidence interval (CI), were determined by fitting the empirical data on toxicity of metal mixtures to Equation 5 in GraphPad Prism.

Mixtures of Cu2+ and Zn2+

Taking into account the competition between Cu2+ and Zn2+ and between Cu2+/Zn2+ with H+, the fraction of the total number of biotic ligands occupied by Cu2+ and Zn2+ (fCu and fZn, respectively) can be determined as a function of their stability constants and their free ion activities in the solution as follows:

  • equation image(6)
  • equation image(7)

where KHBL, KCuBL, and KZnBL (L/mol) are stability constants of binding of H+, Cu2+, and Zn2+ to biotic ligands, respectively, and were determined by toxicological data for single metals; and {H+}, {Cu2+}, and {Zn2+} (mol/L) are free ion activities of H+, Cu2+, and Zn2+ in the exposure solution, respectively.

Mixtures of Cu2+ and Ag+

It was assumed that Cu2+ and Ag+ bind to different transporters, so binding of these metal ions to transport sites at the biotic ligands is influenced only by H+ (Eqns. 8 and 9):

  • equation image(8)
  • equation image(9)

where KHBL, KCuBL, and KAgBL (L/mol) are stability constants of binding of H+, Cu2+, and Ag+ to biotic ligands, respectively, and were estimated by experimental data on toxicity of single metals; and {H+}, {Cu2+}, and {Ag+} (mol/L) are free ion activities of H+, Cu2+, and Ag+ in the exposure solution, respectively.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Toxicity of Cu2+, Ag+, and Zn2+ individually

Generally, Zn2+ had the lowest affinity for binding sites at the biotic ligands as shown by its lowest stability constant compared with Cu2+ and Ag+ (that is, log KZnBL < log KAgBL < log KCuBL; Table 1). This indicates that the formation of complexes of the biotic ligands with Zn2+ occurred to a lesser extent than the formation of complexes with Cu2+ or Ag+. At the same time, the highest fraction of the total number of biotic ligands occupied by Zn2+ was required to result in a 50% inhibition of the root growth (Table 1 and Fig. 1A); that is, at the same concentration, the Zn2+-biotic ligand complex resulted in the lowest toxic effects in comparison with the complexes of Cu2+ and Zn2+. Consequently, Zn2+ had the highest value of the median effective activity among the three metal ions studied (Table 1 and Fig. 1B). By contrast, Cu2+ had the highest affinity for binding sites at the biotic ligands and the lowest median effective activity compared with Zn2+ and Ag+ (Table 1 and Fig. 1B). Additionally, the fraction of the total number of biotic ligands occupied by Ag+ to inhibit the root growth by 50% was lower than the corresponding fraction occupied by Cu2+ or Zn2+ (that is, f50Ag < f50Cu < f50Zn; Table 1 and Fig. 1A). In other words, the Ag+-biotic ligand complex led to the highest toxic effects compared with the complexes of Cu2+ and Zn2+, contributing to the highest slope of the curve describing the relationship between the RRE and fM for Ag+ (Fig. 1A and Supplemental data, Table S3).

Table 1. Toxicity of Cu2+, Zn2+, and Ag+ individually to plants as expressed by the median effective activity (EA50M, M) and biotic ligand model (BLM) parameters, that is, the fraction of the total number of biotic ligands occupied by metal ions at the 50% response level (f50M) and the stability constant (log KBML) as found in the present study and in other literature reports
Plant speciesVigna unguiculataLactuca sativaHordeum vulgare
MetalsEA50M (M)EA50M (M; 95% CI)f50M (95% CI)log KMBLTEFMaf50Mlog KMBLlog KMBLf50Mlog KMBL
  • a

    Toxic equivalency factor (TEFM, Cu2+ equivalents) is determined according to Equation 3 based on the determined values of f50M.

Cu2+2.90 × 10−72.60 × 10−8 (1.87 × 10−8−3.61 × 10−8)0.36 (0.29–0.43)7.401  7.4 ± 0.20.446.28
Ag+2.4 × 10−81.34 × 10−7 (1.19 × 10−7−1.50 × 10−7)0.22 (0.20–0.24)6.391.64     
Zn2+1.6 × 10−51.06 × 10−4 (9.11 × 10−5−1.24 × 10−4)0.42 (0.38–0.44)4.000.860.384.06   
SourceKopittke et al. 17Present studyWang et al. 18Thakali et al. 19Luo et al. 20
thumbnail image

Figure 1. Dose–response curves describing toxicity of Cu2+, Ag+, and Zn2+ individually are expressed by the relationship between the relative root elongation (RRE, %) and the fraction of the total biotic ligands occupied by metal ions (fM; A) and the free metal ion activity in the solution (log{Mn+}, M; B).

Download figure to PowerPoint

Toxicity of binary metal mixtures (Cu2+–Zn2+ and Cu2+–Ag+)

Generally, the combination of the BLM and the TEF approach using the TEF values for Cu2+, Ag+, and Zn2+ calculated based on f50M (Table 1) performed equally well in estimating the toxicity of mixtures of Cu2+–Zn2+ and of Cu2+–Ag+, as indicated by a negligible difference between the values of r2 (Fig. 2 and Table 2). Approximately 70% of the variability in the toxicity of mixtures of Cu2+–Zn2+ and Cu2+–Ag+ could be explained by the TEQ based on the fraction of the total number of biotic ligands occupied by metal ions (r2 = 0.65–0.69). Moreover, the TEQ50 (Cu2+ equivalents) of the Cu2+–Zn2+ mixture was lower than the corresponding value for the Cu2+–Ag+ mixture (Fig. 2 and Table 2). This difference is significant because the 95% CIs of the TEQ50 for the two mixtures statistically deviated from each other (Table 2). These results indicate that mixtures of Cu2+ and Zn2+ were significantly more toxic than mixtures of Cu2+ and Ag+ based on Cu2+ equivalents. Additionally, no significant difference was found between the slopes of the dose–response curves describing the toxicity of the mixtures of Cu2+–Zn2+ and of Cu2+–Ag+ (Fig. 2 and Table 2). This similarity in the slopes for the two mixtures was explained by the similar meaning of the two curves, that is, representing the changes in the RRE with the changes in Cu2+ equivalents. This explanation is supported by the overlapping 95% CIs of these slopes with the slope of the curve describing the relationship between fCu and RRE (Table 2 and Supplemental Data, Table S3).

thumbnail image

Figure 2. Toxic effects of the mixtures of Cu2+–Zn2+ (A) and Cu2+–Ag+ (B) expressed as the relative root elongation (RRE, %). The RRE is plotted as a function of the toxic equivalency quotient (TEQ; Cu2+ equivalents) according to Equation 5.

Download figure to PowerPoint

Table 2. Estimations of coefficients, that is, toxic equivalency quotient at the 50% response level (TEQ50; Cu2+ equivalents) and slope parameter (β), in Equation 5 and statistical parameters representing the toxicity of mixtures of Cu2+–Zn2+ and of Cu2+–Ag+
ParametersCu2+–Zn2+Cu2+–Ag+Cu2+–Ag+ ({Cu2+} > 2 × 10−8 M)Cu2+–Ag+ ({Cu2+} < 2 × 10−8 M)
  1. CI = confidence interval.

TEQ500.560.820.880.61
95% CI0.51–0.600.76–0.870.82−0.930.58−0.64
β−1.534−1.599−1.896−3.701
95% CI−1.817 to −1.252−1.928 to −1.270−2.458 to −1.334−4.560 to −2.842
n1111076245
r20.650.690.640.84

Notably, substantial deviations were found in measurements from predictions for a number of mixtures of Cu2+ and Ag+ (Fig. 2B). These deviated data points correspond to mixtures in which the free Cu2+ activities in these mixtures are all below 2 × 10−8 M. Additionally, a shift in the trend of toxic effects over this exposure level of Cu2+ was shown in the dose–response curve describing toxicity of Cu2+ following single exposure (Fig. 1B). Particularly at the free Cu2+ activities below 2 × 10−8 M, toxic effects of Cu2+ did not increase with an increase in the exposure level of Cu2+. These results may be related to the fact that Cu is an essential element, and, apparently, 2 × 10−8 M is the lower level of the optimal activity range at which the growth of lettuce roots is not inhibited. Therefore, we divided the exposure solutions containing Cu2+–Ag+ mixtures into two categories on the basis of the free ion activity of Cu2+ in the solution: {Cu2+} > 2 × 10−8 M (Fig. 3A) and {Cu2+} < 2 × 10−8 M (Fig. 3B). A similar cutoff value was not found for the mixtures of Cu2+ and Zn2+.

thumbnail image

Figure 3. Response of lettuce roots exposed to mixtures of Cu2+ and Ag+ expressed as the relative root elongation (RRE, %) plotted as a function of the toxic equivalency quotient of mixtures (TEQ; Cu2+ equivalent) at two levels of free ion activity of Cu2+ in the exposure solution: {Cu2+} > 2 × 10−8 M (A) and {Cu2+} < 2 × 10−8 M (B).

Download figure to PowerPoint

The method of classifying the mixtures of Cu2+ and Ag+ based on the cutoff value of 2 × 10−8 M led to substantial improvement in estimating toxicity of the Cu2+–Ag+ mixtures at low free ion activities of Cu2+ in the exposure solution (r2 = 0.84; Fig. 3B and Table 2). Moreover, a statistically significant difference was found between the estimated toxicities of the two classified mixture groups as shown by nonoverlapping 95% CIs of the TEQ50 (Table 2). This indicates that the toxicity of Cu2+–Ag+ mixtures depended not only on their TEQ but also on the exact amount of their components (Fig. 3 and Table 2). With the same TEQ below approximately 0.8, mixtures with lower free ion activities of Cu2+ in the solution (that is, {Cu2+} < 2 × 10−8 M) were statistically significantly more toxic than mixtures with higher {Cu2+}. This is additionally indicated by a statistically significantly steeper dose–response curve describing toxicity of the mixture with {Cu2+} < 2 × 10−8 M compared with the curve with {Cu2+} > 2 × 10−8 M (Fig. 3A and B, and Table 2). This supports the suggestion above that the value of 2 × 10−8 M is the lower level of the optimal activity range.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Toxicity of Cu2+, Zn2+, and Ag+ individually

The results of the present study indicate that the BLM parameters are better indicators of the intrinsic toxicity of single metals than EA50M. The BLM parameters provide a mechanistic explanation for the difference in comparison of toxicity of Cu2+ and Ag+ based on EA50M and f50M as shown in Results. In particular, because the Ag+-biotic ligand complex resulted in higher effects than the Cu2+-biotic ligand complex, the lower affinity of Ag+ for binding sites at the biotic ligands accounts for the higher median effective activity of Ag+ compared with the corresponding value for Cu2+. In other words, the inclusion of interactions between ions and biotic ligands at the environment–organism interface contributes to the different orders of toxicity of Cu2+ and Ag+ based on EA50M and on f50M. These observations strongly indicate that the BLM parameters provide better insight into mechanisms of metal binding and toxicity compared with a single value of EA50M.

Based on the median effective activity, EA50M, substantial differences were found between the sensitivities of cowpea, Vigna unguiculata, reported by Kopittke et al. 17 and of lettuce, L. sativa, found in the present study (Table 1). On the basis of the estimations of EA50M, L. sativa had higher tolerance to Zn2+ and Ag+ but was more sensitive to Cu2+ than V. unguiculata. Moreover, the BLM parameters reflecting the toxicity of Cu2+ and Zn2+ individually obtained in the present study on L. sativa were, in general, within the ranges reported from other studies on barley, Hordeum vulgare 18–20 (Table 1). This indicates negligible differences in the sensitivities of L. sativa and H. vulgare to Cu2+ and Zn2+ based on the BLM parameters.

Toxicity of binary metal mixtures (Cu2+–Zn2+ and Cu2+–Ag+)

The present study shows a strong dependence of mixture toxicity on the composition and proportion of the metal mixture, that is, the TEQ of the mixture and the specific amount of Cu2+ in the mixture. This observation was previously reported by Sharma et al. 21 and Hamm et al. 22. This dependence is potentially attributed to physiological processes that are highly specific, depending on the exposure level. For example, for essential metals such as Cu2+, their presence above certain exposure levels is vital and beneficial for plant growth. By contrast, exposure to extremely low or extremely high concentrations of these elements is toxic to the growth of plants. This explanation potentially accounts for the observations on Cu2+ toxicity found in the present study, for example, higher toxic effects caused by the mixtures with lower activities of Cu2+ among mixtures with the same TEQ below 0.8. Similarly, physiological responses of plants exposed to Zn2+ vary greatly depending on the exposure level, ranging from changes in the plant cell vacuolization or in membrane permeability to damage to enzyme systems, respiration, or the photosynthetic apparatus 23, 24.

Integration of ion–ion interactions in estimating metal toxicity

In the present study, 64 to 84% of the variability in toxicity of Cu2+–Zn2+ and Cu2+–Ag+ could be explained by TEQ (r2 = 0.64–0.84). Together with the studies of Hatano and Shoji 10 and Jho et al. 11, findings in the present study support the assumption that the fraction of the biotic ligands bound to metal ions in mixtures in the total number of biotic ligands might be indicative of toxicity of metal mixtures. As presented here, the incorporation of the BLM into the extended TEF approach allows integration of interactions between different metal ions in estimating their joint toxicity based on particular assumptions about metal binding. However, an exact understanding of metal binding is usually lacking, causing difficulties in applying the BLM to predict toxicity of metal mixtures. Diverse binding of metals to a variety of biotic sites further complicates the issue. Moreover, in the BLM, ion–ion interactions are interpreted in terms of competition for binding sites at the biotic ligands. However, previous studies indicate that effects of the interactions on bioaccumulation and toxicity of single metals and mixtures could not be interpreted completely in terms of competitive binding to biotic ligands 14, 19, 25. Apart from competition for binding sites, joint toxicity of multiple metals is influenced by a number of other mechanisms, for example, the production of metal-binding proteins such as metallothionein, changes in the permeability of the plasma membrane induced by exposure to metal mixtures, and interactions between essential and nonessential metals 26–32. These mechanisms are not taken into account in estimating toxicity of metal mixtures by the BLM approach, potentially contributing to deviations of predictions from measurements.

In summary, the present study supports the BLM principle that the fraction of the total number of biotic ligands occupied by metal ions is a key indicator determining metal toxicity. The BLM parameters provide a better understanding of metal binding and intrinsic toxicity of single metals. More importantly, the present study indicates the potential applicability of the BLM principle to metal mixtures. This was shown by a good predictive power of the combination of the BLM and the TEF approaches, using the TEQ based on the fraction of the total number of biotic ligands bound to metal ions in estimating toxicity of metal mixtures. This modeling approach additionally allows integrating assumed ion–ion interactions in predicting joint toxicity of multiple metals, which are usually excluded in models estimating mixture toxicity. Furthermore, this method of integrating the fraction of biotic ligands occupied by metal ions with the TEF approach is applicable to mixtures of more than two substances, consisting of metals binding to the same or different sites at the biotic ligands.

SUPPLEMENTAL DATA

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Table S1. Chemical composition of the Steiner solution used for solution speciation.

Table S2. Free ion activities of Cu2+, Zn2+, and Ag+ in the test solutions.

Table S3. Slopes (β) of the curves describing the relationships between the fraction of the total number of biotic ligands occupied by Cu2+, Zn2+, and Ag+ (fCu, fZn, and fAg, respectively) and the growth of lettuce roots expressed by the relative root elongation (RRE, %), with 95% confidence intervals. (291 KB DOC).

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

We thank the reviewers. M.G. Vijver is supported by Netherlands Organization for Scientific VENI grant project 863.08.023.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
etc_2039_sm_SupplTabsS1-S3.doc283KSupplementary Tables S1-S3

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.