ABSTRACT Two biotic ligand models (BLMs) predicting copper accumulation and toxicity to heavy metal tolerant moss (Funaria hygrometrica Hedw.) were developed in the present study. Although BLMs developed in this study could predict the effects of pH and calcium on copper accumulation and toxicity, stability constants representing the binding strength between cations and the biotic ligand for the toxicity model were different than those for the accumulation model. Stability constants of copper, calcium and proton for the toxicity model were logKCu = 2.98, logKCa = 3.38, and logKH = 4.17, respectively, and for the accumulation model were logKCu = 4.46, logKCa = 3.28, and no effect, respectively. Proton competition with copper accumulation was not observed. Difference in logKCu between the two models indicated that tolerance with copper results in the decrease in KCu. In addition, it was indicated that decreases in copper toxicity due to increases in proton concentration was not caused by the competitive effects of proton but by the changes in internalization flux.
Journal of Water and Environment Technology, Vol. 8, No.4, 2010 Address correspondence to Ryo Shoji, Department of Chemical Science and Engineering, Tokyo National College of Technology, Email: shoji@tokyo-ct.ac.jp Received May 8, 2010, Accepted August 24, 2010. - 339 - Application and Comparison of Two Biotic Ligand Models Predicting Copper Toxicity and Accumulation in Heavy Metal Tolerant Moss Hiroki NAKANISHI*, Ryo SHOJI*, Misao ITOUGA**, Hitoshi SAKAKIBARA** *Department of Chemical Science and Engineering, Tokyo National College of Technology, 1220-2 Kunugida-cho, Hachioji, Tokyo 193-0997, Japan **RIKEN Plant Science Center, 1-7-22 Suehiro-cho, Tsurumi, Yokohama, Kanagawa 230-0045, Japan ABSTRACT Two biotic ligand models (BLMs) predicting copper accumulation and toxicity to heavy metal tolerant moss (Funaria hygrometrica Hedw.) were developed in the present study. Although BLMs developed in this study could predict the effects of pH and calcium on copper accumulation and toxicity, stability constants representing the binding strength between cations and the biotic ligand for the toxicity model were different than those for the accumulation model. Stability constants of copper, calcium and proton for the toxicity model were logK Cu = 2.98, logK Ca = 3.38, and logK H = 4.17, respectively, and for the accumulation model were logK Cu = 4.46, logK Ca = 3.28, and no effect, respectively. Proton competition with copper accumulation was not observed. Difference in logK Cu between the two models indicated that tolerance with copper results in the decrease in K Cu . In addition, it was indicated that decreases in copper toxicity due to increases in proton concentration was not caused by the competitive effects of proton but by the changes in internalization flux. Keywords: non-competitive binding, phytoremediation, stability constant. INTRODUCTION Environmental pollution by heavy metals due to technological development is one of the most serious problems today. Heavy metals are hazardous to the environment because these metals cause significant toxicity to organisms. Heavy metal removal from the environment is necessary and is considered a challenging task. Because the removal of heavy metals using conventional physiochemical methods can be very costly, phytoremediation has gained increased attention in the last decades. Estimations of heavy metal toxicity and accumulation in heavy metal tolerant plants are necessary in order to apply the phytoremediation for heavy metal removal. However, it has been recognized that physicochemical characteristics such as pH, water hardness, and dissolved organic carbon concentration can decrease the metal toxicity and accumulation in organisms (Cheng and Allen, 2001; Di Toro et al., 2001). The Biotic Ligand Model (BLM) (Di Toro et al., 2001) was developed to incorporate metal speciation and protective effects of competing cations into the prediction of metal bioavailability and toxicity. The main assumption of the BLM is that metal toxicity is caused by free metal ions reacting with the binding sites at the organism-water interface to form a metal-biotic ligand (BL) complex. In the BLM, the stability constants for metal- and competing cation-BL complexes are required to predict the heavy metal toxicity and accumulation. However, no literatures Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 340 - about the relationship between the toxicity model and the accumulation model can be found so far though it is necessary to compare these two models for appropriate modeling and prediction of heavy metal toxicity and accumulation. Funaria hygrometrica Hedw. is one of the most tolerant mosses to heavy metals (Itouga et al., 2006). Therefore, F. hygrometrica was selected as a test organism in this study. Copper was selected as a typical heavy metal because copper was well used in the development of the BLM. The objectives of this study were 2-fold: (1) to develop two BLMs for the prediction of copper toxicity and copper accumulation in F. hygrometrica and (2) to investigate the relationship between two BLMs in terms of each stability constants. MATERIALS AND METHODS Test organism and cultures The protonema of F. hygrometrica provided from the RIKEN Plant Science Center, Japan, was cultured in BCD medium (Ashton and Cove, 1977) which is widely used for the moss incubation. After 2 weeks of incubation, the moss was treated with a Polytron homogenizer PT2100 (Kinematica, Switzerland). The suspended moss was transferred into the medium and then cultured. All incubations including toxicity test were performed at 25.0 ± 1.0ºC under continuous light and aseptic condition. Toxicity test The toxicity tests were carried out by imitating the standard procedure for the duckweed growth inhibition test based on the image analysis for the frond area (Organization for Economic Cooperation and Development, 1999). The suspended moss was transferred onto each solidified agar medium for the preculture and incubated for 2 days. The moss and dialysis membrane were transferred onto the agar medium for the main culture. Instead of CuSO 4 ·5H 2 O, CuCl 2 ·2H 2 O was added to the main culture medium for the toxicity test. Concentrations of Cu and Ca were varied from 1.0 × 10 -6 to 8.0 × 10 -5 mol/L and from 8.0 × 10 -5 to 8.0 × 10 -4 mol/L, respectively. By adding small volumes of 0.1 mol/L HCl and 0.1 mol/L NaOH, pH was adjusted within the range from 4.4 to 6.5. After the transfer of the moss, the exposure medium which consists of the same components as the main culture, except for agar, was added onto the moss layer. The photograph of the moss was taken by a charge-coupled device (CCD) camera having a resolution of 832 × 624 pixels (DXC-108; SONY, Japan) immediately after the exposure. The moss was incubated for 3 days, and then the photograph of the moss was taken again. All incubations were performed under aseptic condition. The area covered by the moss was measured by an image analysis. Only the saturated portion of the photograph was extracted and the brightness was adjusted using a commercial image analysis software (SimpleVmpViewer). The transformed image was binarized, and the area covered by the moss was measured using Image J ver. 1.38 (National Institute of Health, USA). The threshold gray-scale levels to distinguish between the moss and the background agar were determined by trial and error. By using the measured area (A N ) of the moss after N days from exposure, the growth rate of the moss was calculated using eq. 1. Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 341 - )ln()ln( )ln()ln( (%) rategrowth control0,control3, 03 AA AA − − = Accumulation test The moss incubated for more than 2 weeks in a 3 L narrow-necked bottle was collected and rinsed with distilled water. It was transferred into 100 mL of exposure medium containing CuCl 2 ·2H 2 O (varied from 1.0 × 10 -5 to 8.0 × 10 -5 mol/L) and CaCl 2 ·2H 2 O (varied from 8.1 × 10 -6 to 8.1 × 10 -4 mol/L) in a 300 mL Erlenmeyer flask. The density of the moss was 3 mg/L (wet weight). Small volumes of 0.1 mol/L HCl and 0.1 mol/L NaOH were added to adjust the pH within 4.4 to 6.5. The moss suspension in the Erlenmeyer flask was shaken using a shaking apparatus (110 rpm) at 25.0 ± 1.0ºC under continuous light. After 24 hours, the moss was collected by suction filtration using membrane filter (ADVANTEC, Japan) with pore size of 0.45 μm. The moss was then rinsed with distilled water and dried at 80ºC for 6 hours. The dried moss was digested with 65% HNO 3 , and the digested moss was transferred to a 100 mL volumetric flask and diluted with distilled water after filtration using membrane filter (IWAKI, Japan) with pore size of 0.45 μm. Copper concentration in the diluted solution was analyzed by ICP-AES (SPS-7800, SII, Japan). Models Growth rate of the moss was fitted to the logistic equation (eq. 2) using KaleidaGraph ver. 4.0 (Synergy Software, USA). })(1{100 50 β x/x/y += where y (%) is the growth rate of the moss, x is the copper dose, x 50 is the dose to reduce 50% of growth rate, and β is a parameter showing the inverse of the slope of the dose response curve. Total copper concentration (total metal model), and the fraction of BL bound by Cu 2+ , f Cu , (BLM) were substituted for the dose term x in eq. 2. According to the BLM, f Cu determines the bioavailability and the toxicity of copper. The cation-BL binding is treated as a complexation reaction with stability constant, K xi , and cations such as calcium and proton bind onto BL competitively with copper. Thus, f Cu can be calculated by eq. 3. }M{}{Cu1 }{Cu n iM 2 Cu 2 Cu Cu i + + + ∑ ++ = KK K f where {Cu 2+ } (mol/L) is the free ion activity of the cupric ion, and M i n+ is the competing cation that inhibits copper toxicity and accumulation. In this study, calcium ion and proton were assumed to be the only competing cations. WHAM VI (The Windermere Humic Aqueous Model, Natural Environment Research Council 2001) was used to calculate the free cupric ion activity and free calcium ion activity. The parameters β and the stability constants were estimated by minimizing the root mean squared error (RMSE) of the predicted growth rate using the SOLVER program in Microsoft Excel 2003 (Thakali et al., 2006). To obtain a unique solution, f 50 was fixed to 0.01 according to Thakali’s results for the BLM optimization. Standard errors of fitted parameters were calculated using the Visual Basic for Applications (VBA) macro (1) (2) (3) × 100 Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 342 - “SolverAid” (De Levie, 2004). In order to predict the heavy metal accumulation in the moss, Michaelis-Menten-type equation (eq. 4) was used (Vázquez et al., 1999), [M] [M] [M] m moss + = K n where [M] moss (mg/g d.w.) is the heavy metal accumulation in the moss, n (mg/g d.w.) is the maximum heavy metal accumulation, [M] (mol/L) is the total concentration of heavy metal, and K m (mol/L) is Michaelis constant. The observed data of copper accumulation were fitted to eq. 3 using KaleidaGraph ver. 4.0. In order to predict the copper accumulation in the moss while taking the metal speciation and the effects of competing cations into consideration, f Cu was calculated using eq. 2 and copper accumulation can also be predicted by employing eq. 5 (Cheng and Allen, 2001), }M{}{Cu1 } [Cu] predicted n iM 2 Cu 2 Cu Cumoss i + + + ∑ ++ == KK nK nf {Cu The parameters n and the stability constants were estimated by minimizing the RMSE of the predicted copper accumulation using the SOLVER program. RESULTS AND DISCUSSION Estimation of the stability constants Relationship between growth rate or copper accumulation and total copper concentration or f Cu is shown in Fig. 1. The BLM fitted with the observed data better than the total metal model, indicating that the fraction of the BL bound by copper represent the bioavailability much better than the total metal concentration. Table 1 shows the stability constants for the toxicity model and the accumulation model. The stability constants of the toxicity model except copper for F. hygrometrica were similar with those for Daphnia magna. However, stability constant of copper for F. hygrometrica was extremely lower than that of D. magna. It is indicated that tolerance to copper should affect the stability constant. On the other hand, every stability constant of the accumulation model for F. hygrometrica was lower than those for lettuce root. However, n of F. hygrometrica was much higher than that of lettuce root, indicating that F. hygrometrica has higher potential to accumulate copper. Basile et al. also suggested that F. hygrometrica has high potential to accumulate zinc and lead (Basile et al., 2001). Comparison between the toxicity model and the accumulation model As shown in Table 1, the stability constants of the toxicity model were different from those of the accumulation model except for calcium. Stability constant of copper for the toxicity model was significantly lower than that of the accumulation model (p < 0.001). It was because of the detoxification processes in the moss. Since copper toxicity should occur after the uptake of heavy metal into moss cells, the stability constant of copper calculated for the toxicity model should be lower due to the detoxification processes. The stability constant of copper for the organisms which are sensitive to copper toxicity (4) (5) Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 343 - 0 20 40 60 80 100 120 10 -6 10 -5 10 -4 10 -3 growth rate (%) Total Cu concentration (mol/L) A R 2 =0.89 RMSE=11 % 0 20 40 60 80 100 120 10 -4 10 -3 10 -2 10 -1 growth rate (%) f Cu B R 2 =0.96 RMSE=7.2 % 0 5 10 15 20 25 30 020406080 Cu accumulation (mg/g d.w.) Total Cu concentration (μmol/L) C R 2 =0.76 RMSE=3.1 mg/g d.w. 0 5 10 15 20 25 30 0 0.2 0.4 0.6 0.8 Cu accumulation (mg/g d.w.) f Cu D R 2 =0.93 RMSE=1.7 mg/g d.w. Fig. 1 - Relationship between growth rate and total copper concentration (A) or f Cu (B) and relationship between copper accumulation and total copper concentration (C) or f Cu (D) Table 1 - Stability constants (95% confidence limits) and optimized parameters for the toxicity model and the accumulation model Toxicity model organisms logK Cu logK Ca logK H f 50 β F. hygrometrica 2.98 (2.76-3.20) 3.38 (2.91-3.85) 4.17 (3.78-4.56) 0.01 5.3 D. magna d 8.02 a 3.31 (3.07-3.80) ～5.4 a 0.39 ND b Accumulation model organisms logK Cu logK Ca logK H n (mg/g d.w.) F. hygrometrica 4.46 (4.28-4.64) 3.28 (3.07-3.50) No effects 42 Lettuce root e 12.29 8.49 (8.19-8.79 c ) 6.80 (6.68-6.92 d ) 6.4 a 95% confidence limits are not shown. b Not determined. c Standard error d De Schamphelaere and Janssen, 2002, e Cheng and Allen, 2001 should be higher because toxicity occurrence requires lesser level of copper in such organisms. Indeed, K Cu depends on the effective concentration to reduce 50% of biotic activity (EC 50 ) as shown in Fig. 2. On the other hand, the stability constant of proton could be calculated only for the toxicity model, indicating that increases in proton concentration cause decreases in copper toxicity though proton is not a competitive cat- ion with copper binding onto BL. Heijerick et al. (2002) developed a BLM to predict zinc toxicity to Pseudokirchneriella subcapitata, and they concluded that the effects of pH on zinc toxicity could not be described by the BLM and that pH affects the physiology of the BL. Francois et al. (2007) investigated the effects of pH on short-term uptake of manganese and cadmium into Chlamydomonas reinhardtii, and they concluded that the increase in proton concentration inhibits the metal internalization flux non-competitively by binding to a binding site other than BL. In addition, changes in the conformation of BL caused by this binding of proton result in decreases in the number of BL. The conformation change of BL may decrease the internalization flux of Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 344 - 10 2 10 3 10 4 10 5 10 6 10 7 10 8 10 9 10 -8 10 -7 10 -6 10 -5 10 -4 K Cu (L/mol) EC 50 (mol/L) Fig. 2 - Relationship between K Cu and EC 50 for D. magna (■) (De Schamphelaere and Janssen, 2002), barley (○) (Shoji, 2006), wheat (▼) (Luo et al., 2008), L. paucicostata (×) (Hatano and Shoji, 2008) and F. hygrometrica (▲). In the selection of EC 50 , pH and free calcium ion activity are kept as constant as possible. copper without changes in f Cu and potential of copper accumulation. Therefore, copper accumulation should not be affected by varying pH. However, because copper toxicity should depend on the integration of copper accumulation until each exposure duration, decreases in the internalization flux of copper should result in decreases in copper toxicity. This is a plausible explanation on the difference in the stability constant of proton for the toxicity model and for the accumulation model. The BLM in the present state does not consider these differences between the toxicity model and the accumulation model so far. More detailed modeling of the BLM about heavy metal detoxification and conformation change of BL leads to more appropriate prediction of heavy metal toxicity and accumulation in organisms. CONCLUSIONS Two BLMs predicting copper toxicity and copper accumulation in F. hygrometrica were developed to investigate the relationship between the stability constants for toxicity model and those for accumulation model. Stability constant of copper for toxicity model was lower than that for accumulation model, and the difference seemed to be because of the detoxification processes of copper in the moss. The stability constant of proton could be calculated only for toxicity model because proton should not compete with copper ion for binding on BL. The effect of proton on copper toxicity seemed to be dependent on the conformation change of BL. More detailed modeling of the BLM can lead to more extensive and appropriate prediction of copper toxicity and copper accumulation in organisms. REFERENCES Ashton N. W. and Cove D. J. (1977). The isolation and preliminary characterization of Auxotrophic and resistant mutants of the moss Physcomitrella patens, Mol. and Gen. Genet., 154, 87-95. Basile A., Cogoni A. E., Bassi P., Fabrizi E., Sorbo S., Giordano S. and Castaldo Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 345 - Cobianchi R. (2001). Accumulation of Pb and Zn in Gametophytes and Sporophytes of the Moss Funaria hygrometrica (Funariales), Ann. Bot., 87, 537-543. Cheng T. and Allen H. E. (2001). Prediction of uptake of copper from solution by lettuce (Lactuca sativa Romance), Environ. Toxicol. Chem., 20(11), 2544-2551. De Levie R. (2004). Advanced excel for scientific data analysis, Oxford University Press, New York. De Schamphelaere K. A. C. and Janssen C. R. (2002). A Biotic Ligand Model predicting acute copper toxicity for Daphnia magna: The effects of calcium, magnesium, sodium, potassium, and pH, Environ. Sci. Technol., 36 (1), 48-54. Di Toro D. M., Allen H. E., Bergman H. L., Meyer J. S., Paquin P. R. and Santore R. C. (2001). Biotic Ligand Model of the acute toxicity of metals. 1. Technical basis, Environ. Toxicol. Chem., 20(10), 2382-2396. Francois L., Fortin C. and Campbell P. G. C. (2007). pH modulates transport rates of manganese and cadmium in the green alga Chlamydomonas reinhardtii through non-competitive interactions: Implications for an algal BLM, Aquat. Toxicol., 84, 123-132. Hatano A. and Shoji R. (2008). Toxicity of Copper and Cadmium in Combinations to Duckweed Analyzed by the Biotic Ligand Model, Environ. Toxicol., 23, 372-378. Heijerick D. G., De Schamphelaere K. A. C. and Janssen C. R. (2002). Biotic ligand model development predicting Zn toxicity to the alga Pseudokirchneriella subcapitata: possibilities and limitations, Comp. Biochem. Physiol. Part C, 133, 207-218. Itouga M., Suzuki T., Komatsu K., Yamaguchi I., Shiraishi T., Ono Y. and Sakakibara H. (2006). Effect of some leaching waters from ashes of municipal wastes on protonemal cell division of Funaria hygrometrica Hedw, Bryol. Res., 9(3), 78-83. Luo X. S., Li L. Z. and Zhou D. M. (2008). Effect of cations on copper toxicity to wheat root: Implications for the biotic ligand model, Chemosphere, 73, 401-406. Organization for Economic Cooperation and Development (1999). Proposal for a New Guideline: Lemna sp. Growth Inhibition Test, 1-20. Shoji R. (2006). Terrestrial Biotic Ligand Model to predict bioavailability of heavy metals in rhizosphere, Jpn. J. Environ. Toxicol., 9(2), 149-156. Thakali S., Allen H. E. and Di Toro D. M., Ponizovsky A. A,. Roonty C. P, Zhao F. and McGrath S. P. (2006) A Terrestrial Biotic Ligand Model. 1. Development and Application to Cu and Ni Toxicities to Barley Root Elongation in Soils, Environ. Sci. Technol., 40(22), 7085-7093. Vázquez M. D., López J. and Carballeira A. (1999). Uptake of heavy metals to the extracellular and intracellular compartments in three species of aquatic bryophyte, Ecotoxicol. Environ. Saf., 44, 12-24. . d.w.) F. hygrometrica 4.46 (4. 28- 4.64) 3. 28 (3.07-3.50) No effects 42 Lettuce root e 12.29 8. 49 (8. 19 -8. 79 c ) 6 .80 (6. 68- 6.92 d ) 6.4 a 95% confidence. logK H f 50 β F. hygrometrica 2. 98 (2.76-3.20) 3. 38 (2.91-3 .85 ) 4.17 (3. 78- 4.56) 0.01 5.3 D. magna d 8. 02 a 3.31 (3.07-3 .80 ) ～5.4 a 0.39 ND b Accumulation