Biotic Ligand Model Does Not Predict the Bioavailability of Rare Earth

Jan 22, 2015 - Earth Elements in the Presence of Organic Ligands ... M Tm3+. The bioavailability of the metal complexes could not be explained by a ...
0 downloads 0 Views 637KB Size
Download PDF
Article pubs.acs.org/est

Biotic Ligand Model Does Not Predict the Bioavailability of Rare Earth Elements in the Presence of Organic Ligands Chun-Mei Zhao and Kevin J. Wilkinson* Biophysical Environmental Chemistry Group, Department of Chemistry, University of Montreal, C.P. 6128 Succursale Centre-Ville, Montreal, Quebec H3C 3J7, Canada S Supporting Information *

ABSTRACT: Due to their distinct physicochemical properties, rare earth elements (REEs) are critical to high-tech and cleanenergy industries; however, their bioavailability is still largely unexplored. In this paper, the bioavailability of several REEs has been carefully examined for the freshwater alga, Chlamydomonas reinhardtii. In the presence of organic ligands (L), the biouptake of REEs was much higher than that predicted by the biotic ligand model (BLM). Enhancement of the biouptake flux was observed for six ligands (metal = thulium) and six REEs (ligand = citric acid), indicating that this could be a common feature for these metals. In order to explore the mechanism for the enhanced uptake, Tm internalization was carefully evaluated. The Tm internalization flux (Jint) followed first-order (Michaelis−Menten) kinetics with a calculated maximum internalization flux (Jmax) of (1.1 ± 0.08) × 10−14 mol·cm−2·s−1 and an affinity constant for the reaction of the metal with the transport sites (KTm−R) of 107.1 M−1. In the presence of citric acid, malic acid, or NTA, the Jint for Tm was more than 1 order of magnitude higher than that predicted by the BLM when algae were exposed to a constant 10−9 M Tm3+. The bioavailability of the metal complexes could not be explained by a piggyback internalization (through an anion channel) or the contribution of labile complexes. The enhanced biouptake was attributed to the formation of a ternary Tm complex {L−Tm−R} at the metal transport site. In the natural environment where organic ligands are ubiquitous, classic models are unlikely to predict the bioavailability of REEs to aquatic organisms.

■

INTRODUCTION Over the past several decades, a consensus has formed that a metal’s bioavailability is not dependent on its total concentration but rather on its free ion.1,2 The free-ion activity (FIAM) and biotic ligand models (BLM) assume that the metal of interest is in thermodynamic equilibrium with both ligands in the external medium and biotic ligands, such as trace metal transporters on the cell membrane. By employing (equilibrium) stability constants for the interaction of free ions in the solution with the sensitive sites at the biological surface (i.e., the biotic ligands), the BLM has often been shown to predict metal toxicity and bioaccumulation under conditions of varying hardness and complexation, both in the laboratory and in the field.3−5 A number of exceptions to the BLM have been documented in cases where the free ion activity underestimated the biological effects in the presence of metal complexes.6−8 For example, when biouptake is limited by mass transport, the dissociation of metal complexes can contribute to metal bioavailability by increasing the supply of free ions.6,9 Such a situation is most likely to occur in soils, sediments, and biofilms6,10 or under conditions of enhanced biouptake.11 Under equilibrium conditions, metal complexes can only contribute to biouptake if the metal complex itself is © 2015 American Chemical Society

internalized (e.g., conditions of facilitated transport over a transporter meant for the ligand)7,12 or for cases where an equilibrium ternary complex can be formed at the biological uptake site.13−15 While the predictive capacity of the BLM has been validated for a number of divalent metals and their complexes,3,5 its validity for trivalent metals is still in question. Indeed, it was observed that in the presence of fluoride it was not possible to explain Al bioavailability to juvenile Atlantic salmon without invoking the contribution of AlFx complexes.13 More recently, the uptake of Sc by Chlamydomonas reinhardtii could only be predicted from free ion concentrations below pH 6.5. Above pH 6.5, it was necessary to invoke the contribution of hydroxo complexes (including Sc(OH)2+, Sc(OH)2+, Sc(OH)3).16 Similarly, Sc(F)x(3−x)+ complexes had to be taken into account in order to model Sc uptake by the same organism.17 While only limited data are available on the application of the BLM to the trivalent metals, these intriguing examples suggest that it will be necessary to carefully verify the effects of complex formation on their bioavailability. Received: Revised: Accepted: Published: 2207

November 6, 2014 January 19, 2015 January 22, 2015 January 22, 2015 DOI: 10.1021/es505443s Environ. Sci. Technol. 2015, 49, 2207−2214

Article

Environmental Science & Technology

Determination of REE Internalization Fluxes. In order to precisely quantify the metal internalization fluxes, biouptake experiments were first conducted using short-term exposures with data points collected at 1, 20, 40, and 60 min using REE concentrations ranging from 10−9 to 10−4 M. A 5 mL amount of 0.1 M EDTA was added to 45 mL of the exposure medium in order to stop biouptake and simultaneously wash the weakly surface-adsorbed metal from the cell surface. After 1 min in the EDTA, the solution was filtered through two stacked nitrocellulose filters (pore size 3.0 μm, Millipore). The filters were rinsed three more times with 5 mL of 0.01 M EDTA. The upper filter retained the algae, while the lower one was used to quantify adsorptive losses; metal biouptake was determined from the difference between two filters. The filters and washed algae were transferred into 0.3 mL of 69% HNO3 and digested at 85 °C overnight. Metal concentrations in the exposure solutions were monitored in samples: (i) before the addition of the algae, (ii) during the exposure (dissolved metal < 0.45 μm, Millipore), and (iii) after the filtration of the algae. Metal concentrations in the filter digests and in the exposure solutions were analyzed by inductively coupled plasma mass spectrometry (PerkinElmer, NexION 300x, Supporting Information). The internalization flux, Jint (mol·cm−2·s−1), was calculated from the slope of a linear regression between the measured REE concentration in the algae and the exposure time. All experiments were performed in triplicate and repeated at least twice. Metal speciation, in the presence or absence of ligands, was calculated using WHAM 7.0 with thermodynamic parameters (Table S1, Supporting Information) obtained from NIST Standard Reference Database 46.25 REE Internalization in the Presence of Citric Acid. The biouptake of six REEs (La, Nd, Sm, Eu, Tm, and Y) was measured in the presence of citric acid. Free ion concentrations of 10−9 M were obtained by adding different concentrations of citric acid to 10−6 M of the REEs. The controls corresponded to 10−6 M of the REEs without citric acid. In these experiments, biological variability was reduced by using a single algal culture and by estimating biouptake fluxes from the single 60 min time point (rather than the 4 time points). Otherwise, the exposure and sample treatment protocols were identical to measurements above. Tm Internalization in the Presence of Ligands. In order to examine the influence of complexation on Tm bioavailability, internalization fluxes were measured in the presence of several hydrophilic, organic ligands using three different scenarios. (i) Constant total Tm. [Tm] was maintained at 10−6 M, while concentrations of citric acid (7 × 10−7−5.2 × 10−6 M), malic acid (4.5 × 10−6−4.8 × 10−4 M), and nitrilotriacetic acid (NTA, 6.9 × 10−7−1.6 × 10−6 M) were adjusted. In the presence of ligand, the concentration of Tm3+ varied from 10−9 to 3 × 10−7 M. (ii) Constant Tm3+. [Tm3+] was kept constant at 10−8 M. By changing the concentration of both Tm (3 × 10−8−1 × 10−5 M) and the organic ligands (citric acid 2.8 × 10−8− 1.4 × 10−5 M, malic acid 3.2 × 10−6−5.1 × 10−4 M, and NTA 2.1 × 10−8−1.1 × 10−5 M), simultaneously, it was possible to vary the concentrations of the Tm complexes (10−8−10−5 M). For a given ligand, experiments were performed using a unique algal stock solution. Algae were filtered and digested for a single-exposure time point (60 min), which was used to estimate Tm internalization fluxes.

The rare earth elements (REEs) include the lanthanide series metals (15 elements) plus scandium and yttrium.18 Because of their unique optical, magnetic, and catalytic properties, they are required in numerous high-tech and clean-energy industries, including the fabrication of liquid crystal displays, permanent magnets, and catalytic converters.19 China currently controls >90% of the global market for REEs; however, with recent restrictions on China’s exports and a continual increase in global demand, numerous other countries are looking to exploit REE deposits. While REE concentrations in natural waters are normally lower than nanomolar levels,20 micromolar concentrations have been observed near mining sites.21 Only limited studies are available on the toxicity of the REEs,22,23 and few studies have evaluated their bioavailability. Pałasz and Czekaj (2000) speculated that the REEs could be internalized through a Ca channel.45 Due in part to a current lack of model parameters for the BLM, it is currently impossible to predict their potential ecological risk. The goal of the present study was to validate the use of the BLM for predicting the bioavailability of REEs in the presence of organic ligands. Our hypothesis was that complexation would reduce bioaccumulation in direct proportion to the concentration of the free ion. We initially investigated the biouptake of several REEs in the presence of citrate. Following an initial observation that REE−citrate complexes contributed to the biouptake of six different REE (La, Nd, Sm, Eu, Tm, and Y), we focused the study on a single element Tm in order to quantitatively explain this apparent contradiction of the BLM and to better understand the mechanism leading to this important observation. If this observation is as widespread as we believe that it might be, it will have important implications into the ecological risk assessment of the REEs or perhaps more generally the trivalent metals.

■

MATERIALS AND METHODS Algal Culture. A unicellular green alga, Chlamydomonas reinhardtii (wild-type CL2010), was maintained on a solid trisacetate-phosphate (TAP) medium (1.5% agar24) prior to its transfer into a diluted (4×) and autoclaved TAP solution (dTAP). Algae were grown in an incubator (Multitron, Infors) at 20 °C under a 12 h light:12 h dark regime at 100 rpm. Once the cells attained midexponential growth (1−2 × 106 cells/mL; ca. 3 days), they were transferred into a larger volume of dTAP. Milli-Q water (R > 18 MΩ cm, total organic carbon < 2 μg/L) was used to prepare all solutions. All of the chemicals were at least analytical grade. Algal Exposures. After again reaching midexponential growth, algae were harvested (1872g; 2 min) and washed three times with 50 mL of exposure media containing no metal. The cell concentrate from the final wash (10 mL) was used to spike the exposure media in order to give a final concentration of 8 × 104 cells·mL−1, corresponding to a cell surface area of 0.15 cm2· mL−1. Algal densities and surface areas were measured by a Multisizer 3 particle counter (50 μm aperture, Beckman Coulter). Exposure media consisted of 0.01 M NaMES (2-[Nmorpholino]ethanesulfonic acid sodium) and 0.01 mM Ca(NO3)2, adjusted to pH 6.0 using HNO3 or NaOH. pH was verified following the addition of the metal and ligand and following an equilibration of at least 18 h. The polymerware used for the metal exposures was soaked in 1% HNO3 (≥18 h) and then rinsed at least 7 times with Milli-Q water before being dried under laminar flow (Hesse, Germany). 2208

DOI: 10.1021/es505443s Environ. Sci. Technol. 2015, 49, 2207−2214

Article

Environmental Science & Technology (iii) Constant Tm, constant Tm3+, different ligands. Biouptake was examined in the presence of six different ligands, including citric acid, diglycolic acid, iminodiacetic acid (IDA), malic acid, malonic acid, and NTA. Tm concentrations were maintained at 10−6 M, and the concentrations of the different ligands were adjusted in order to obtain a constant [Tm3+] = 10−9 M. Internalization of 14C-Labeled Ligands. Experiments were designed to determine whether Tm complexes (Tm−L) were internalized by the algae. 14C-Labeled citric and malic acids were purchased from PerkinElmer (specific activities of the stock solutions 46.7 μCi/mL for citric acid and 178.6 μCi/ mL for malic acid). Ligand concentrations ([L]) were held constant at 10−7 M, while [Tm] was adjusted from 0 to 2 × 10−6 M (for citric acid) or from 0 to 10−4 M (for malic acid) in order to vary the proportion of [Tm−L]/[L] from 0 to 99.8% (citric acid) and from 0 to 98.3% (malic acid). Exposure conditions were identical to those described above; algae were sampled after 60 min. Details on the digestion and analysis of the 14C-labeled samples have been provided in the Supporting Information. Data Analysis. The relationship between Jint and [Tm3+] can be described by the Michaelis−Menten equation Jint =

Figure 1. Internalization fluxes of several rare earth elements (La, Nd, Sm, Eu, Tm, and Y) to C. reinhardtii with (solid bars) and without (shaded bars) citric acid. The total REE concentration was 10−6 M, while the concentration of free ion was constant at 10−9 M (different concentrations of citric acid were used for each of the metals). (A) Jint in the presence and absence of citric acid; (B) Jint normalized for [REE3+] (i.e., permeability).

1B). The increased permeability (Jint/[REE3+]) of the organism to REEs in the presence of the REE complexes was a strong indication that the citric acid complexes were contributing to biouptake. Therefore, in order to gain a better mechanistic understanding of the role of the REE complexes, we focused the remaining work on the bioavailability of a single REE (thulium, Tm) using several hydrophilic ligands. Tm Internalization without Ligands. At pH 6, in the absence of ligand, equilibrium calculations predicted that >96% of Tm in the simple MES medium was found as Tm3+. Tm biouptake by the algae increased linearly with time over a 60 min exposure to 1.1 × 10−6 M Tm (Figure S1, Supporting Information), indicating that the concentration of Tm remained constant in the experimental solutions and that Tm efflux was negligible. It is of note that a significant, nonzero y intercept was consistently observed following the washing of the cells by 0.01 M EDTA (1 min) and rinsing of the filter three times with the same EDTA concentration (see Figure S2, Supporting Information, for EDTA washing efficiency). Given that the y intercept was always positive and since it could account for a significant quantity of the cell-associated metal, Tm internalization fluxes were determined from triplicate measurements of Tm exposure data using 4 time points. Internalization fluxes were obtained in the range from 10−9 to 10−4 M Tm3+. The Jint increased linearly (slope of 0.99 on the log−log graph) up to 3 × 10−8 M Tm3+ (Figure 2A). The observation of a single plateau indicated that single type of transporter was likely involved in Tm internalization. A weighted, nonlinear regression of the Michaelis−Menten equation (eq 1) gave a maximum internalization flux, Jmax, of (1.1 ± 0.08) × 10−14 mol·cm−2·s−1 and half saturation constant, Km, of (7.7 ± 1.0) × 10−8 M. For equilibrium models such as the BLM, one critical assumption that must be satisfied is that metal internalization is the rate-limiting process, i.e., Jint is lower than the physical transport flux.5 Indeed, Jint was more than 2 orders of magnitude lower than the maximum diffusive flux (dashed line in Figure 2A), indicating that Tm was in thermodynamic equilibrium with the metal transport sites. It follows that the affinity constant of Tm with the uptake sites on algal surface (KTm−R) could be determined from the reciprocal of Km, i.e., 107.1 M−1. It is of note that this value is very similar to values

Jmax [Tm 3 +] K m + [Tm 3 +]

(1)

−2 −1

where Jmax (mol·cm ·s ) is the maximum internalization flux and Km (mol·L−1) is the half saturation constant (i.e., the Tm3+ concentration at 50% Jmax). The parameters of the Michaelis− Menten relationship were obtained from a weighted nonlinear regression (weighted by 1/y2). The affinity constant, KTm−R (L· mol−1), describing the binding of Tm3+ to the transport site (R) on the cell membrane, was calculated from the reciprocal of Km under the assumption that Jint was the rate-limiting flux.5 In order to validate this basic assumption of the BLM, the maximum diffusive flux of Tm3+ was obtained from a Fick’s second-law calculation (Supporting Information).26 Statistical analyses were performed using SPSS (version 16.0) or R project software (http://www.r-project.org/).

■

RESULTS AND DISCUSSION Internalization of the REEs with and without Ligand. The biouptake of six REEs (La, Nd, Sm, Eu, Tm, and Y) was initially analyzed in the presence and absence of citric acid. For 10−6 M REE, free ion concentrations were adjusted to 10−9 M by varying the concentration of citric acid, i.e., 99.9% of the REEs were complexed by the citric acid. In the absence of citric acid, internalization fluxes for the different REEs were similar, indicating that not only do the metals have a physicochemical similarity, but also they also appear to have similar bioavailability toward the algae (Figure 1A). Indeed, a similar toxicity of the REEs has been observed for the growth of a marine microalga Skeletonema costatum, where EC50 values ranged from 28.27 μM for Dy to 30.34 μM for Nd (for all REEs except Sc and Y).23 In all cases, citric acid decreased the internalization flux of the metals (Figure 1A, p < 0.05, Student t- test); however, the reduction in Jint (2.1−4.6 times decrease) was not in line with the reduction of free ion (1000 times decrease). In other words, when normalized for the concentration of free ion (REE3+), biouptake fluxes were more than 2 orders of magnitude higher in the presence of the citric acid than in its absence (Figure 2209

DOI: 10.1021/es505443s Environ. Sci. Technol. 2015, 49, 2207−2214

Article

Environmental Science & Technology

capricornutum, Cd was shown to be accumulated in the form of a Cd−citrate complex at one-fourth the speed of the noncomplexed citrate.7 Fortin and Campbell have also shown that Ag complexes could be internalized through a sulfate/ thiosulfate channel in C. reinhardtii.12 Our data do not appear to be consistent with the internalization of the metal via a transport site meant for the ligand. For a constant concentration of 14C labeled citric or malic acid (10−7 M), the fraction of Tm complexes could be manipulated by increasing the concentration of Tm (up to 99.8% complexation for citric acid and 98.3% for malic acid). For both ligands, the 14C internalization flux decreased with increasing [Tm−L] (p < 0.05, one-way ANOVA; Figure 3).

Figure 2. (A) Tm internalization fluxes as a function of the concentration of Tm3+, which ranged from 10−9 to 10−4 M. Solid line was determined by a weighted nonlinear regression of the Michaelis−Menten equation (eq 1, weight 1/y2). (B) Tm internalization flux in the presence of organic ligands (citric acid, malic acid, and NTA). [Tm] was constant at 10−6 M, and ligand concentrations were adjusted in order to vary [Tm3+]. In B the solid line represents the regression obtained in A. In both figures, the dashed lines correspond to the calculated maximum diffusive flux of Tm3+ (eq 1, Supporting Information). Each value of Jint represents the mean and standard deviation of 3 replicates.

found for Eu and Sm, 27 again suggesting that the physicochemical similarities among the REEs may result in the metals having similar bioavailabilities. The value of the constants indicates relatively strong binding with the algal transport sites when compared to constants determined for the uptake of other metals to C. reinhardtii under similar conditions (Cd log K = 6.0 at pH 7.0;28 Ni log K = 5.1 at pH 6.0;29 Zn log K1 =5.1 and log K2 7.4 at pH 7.0;30 Cu log K = 5.8 at pH 6.0; Pb log K = 5.9 at pH 6.031). Tm Internalization in the Presence of Organic Ligands. By varying the ligand concentrations at a constant [Tm] = 10−6 M, [Tm3+] could be adjusted in the range from 10−9 to 3 × 10−7 M. From 10−9 to 3 × 10−8 M Tm3+, biouptake fluxes determined in the presence of citric acid, malic acid, and NTA were all higher than those predicted from the free ion concentrations (Figure 2B). For example, at the lowest [Tm3+], i.e., 10−9 M, Jint was more than 1 order of magnitude higher than predicted. Due to the use of low cell densities, Tm depletion in the exposure media was minimized (generally ≤20%) and could not account for the large observed differences. For all three ligands, Tm biouptake increased linearly with exposure time, indicating that the enhanced Tm uptake by the algae was not caused by adsorption (Figure S3, Supporting Information). Furthermore, measured Tm concentrations, as opposed to nominal ones, were used to plot the data. The observation that in the presence of ligands biouptake could not be predicted from the Tm3+ concentrations suggested that the Tm complexes may also be bioavailable. Four hypotheses were therefore examined to explain the enhanced biouptake observed in the presence of the organic ligands. Hypothesis I: Tm Complexes Are Internalized. The uptake of metal complexes through cell membrane has been observed for several lipophilic ligands, such as diethyldithiocarbamate, 8-hydroxyquinoline, and ethylxanthate.32−34 While the uptake of lipophilic complexes is thought to occur by passive diffusion,35 citric acid, malic acid, and NTA (and other ligands used below) all form hydrophilic complexes. In some cases, hydrophilic metal complexes have been shown to enter cells through anion channels. Indeed, for hydrophilic ligands with biological function, such as amino acids or citrate, specialized anion channels are thought to be involved in their biouptake.35 For example, for the green alga, Selenastrum

Figure 3. Internalization flux of 14C-labeled ligands in the presence of an increasing Tm concentration. (A) In the presence of 10−7 M citric acid (CA), Tm was varied from 0 to 2 × 10−6 M Tm such that [Tm− CA]/[CA] varied from 0 to 99.8%. (B) In the presence of 10−7 M malic acid (MA), Tm was varied from 0 to 10−4 M Tm such that [Tm−MA]/[MA] varied from 0 to 98.3%. Solid lines represent the predicted fluxes if it is assumed that only free ligand is accumulated by the cells. Asterisks (*) represent significant differences of the means with respect to the control (0%, p < 0.05, Student t test). Each value of 14 C Jint was the mean and standard deviation of 3 replicate measurements.

Furthermore, ligand uptake was very well predicted by assuming that only the free ligand crossed the membrane (solid lines, Figure 3). Even if it is assumed that the metal complexes could enter the cells via the ligand transporter, the extremely low Jint of the complexes (measured by 14C) could not explain the greater than 10-fold increase in metal uptake fluxes that was observed in the presence of ligands (Figure 2B), i.e., Jint for the metal complexes was kd[LTm−R]). Calculation of dissociation rate constants was performed according to Buffle et al.38 and summarized in the Supporting Information. By using the experimentally determined affinity constants for the reaction of Tm3+ or Tm−L with the transporters (Table 1), it is possible to show that kd[L−TmR] ≥ kd[LTm−R]. Indeed, for each of the ligands, kd[L−TmR] were at least 1 order of magnitude higher than kd[LTm−R] (Table S4, Supporting Information), indicating that the organic ligands would have a higher tendency to dissociate from the ternary complexes and leave Tm−R behind, which could subsequently be internalized.

Figure 6. Relationship between measured internalization flux and BLM predicted internalization flux using the experimentally determined constants. (A) Jint calculated based only on [Tm3+] (with Jmax and Km obtained from Figure 2A); (B) Jint calculated based on the interaction of both Tm3+ and Tm−L (with Jmax and Km obtained from Table 1). The solid line indicated a 1:1 ratio between calculated Jint and measured Jint. Dashed lines were placed at 2:1 and 1:2, respectively.

out of a 2-fold zone of prediction, especially at low [Tm3+]. This observation strongly suggests that bioavailability of the Tm (and REE) complexes needs to be included in the models of metal uptake (such as the BLM). When Jint were recalculated as the sum of contributions from the free ion and its complexes, using the parameters listed in Table 1 (eq 2), the results demonstrated a much improved relationship between the predicted and the measured biouptake (Figure 6B) 3+

Jint =

Tm Jmax KTm[Tm 3 +]

1 + KTm[Tm 3 +]

+

TmL Jmax KTmL[TmL]

1 + KTmL[TmL]

(2)

where and are the maximum internalization fluxes for Tm3+ and Tm−L and KTm and KTmL are the affinity constants for Tm3+ or Tm−L with the transport sites (Table 1). Environmental Implications. In the present study, the simple BLM failed to provide a reasonable prediction of REE uptake, underestimating biouptake by over an order of magnitude in the presence of hydrophilic complexes. Improved model predictions were obtained when the formation of a JTm3+ max

2212

JTmL max

DOI: 10.1021/es505443s Environ. Sci. Technol. 2015, 49, 2207−2214

Article

Environmental Science & Technology

(7) Errecalde, O.; Seidl, M.; Campbell, P. G. C. Influence of a low molecular weight metabolite (citrate) on the toxicity of cadmium and zinc to the unicellular green alga Selenastrum capricornutum: An exception to the free-ion model. Water Res. 1998, 32 (2), 419−429. (8) Fortin, C.; Campbell, P. G. C. Silver uptake by the green alga Chlamydomonas reinhardtii in relation to chemical speciation: Influence of chloride. Environ. Toxicol. Chem. 2000, 19 (11), 2769− 2778. (9) Van Leeuwen, H. P. Metal speciation dynamics and bioavailability: Inert and labile complexes. Environ. Sci. Technol. 1999, 33 (21), 3743−3748. (10) Zhang, H.; Zhao, F. J.; Sun, B.; Davison, W.; McGrath, S. P. A new method to measure effective soil solution concentration predicts copper availability to plants. Environ. Sci. Technol. 2001, 35 (12), 2602−2607. (11) Hassler, C. S.; Wilkinson, K. J. Failure of the biotic ligand and free-ion activity models to explain zinc bioaccumulation by Chlorella kesslerii. Environ. Toxicol. Chem. 2003, 22 (3), 620−626. (12) Fortin, C.; Campbell, P. G. C. Thiosulfate enhances silver uptake by a green alga: Role of anion transporters in metal uptake. Environ. Sci. Technol. 2001, 35 (11), 2214−2218. (13) Wilkinson, K. J.; Campbell, P. G. C.; Couture, P. Effect of fluoride complexation on aluminum toxicity towards juvenile atlantic salmon (Salmo Salar). Can. J. Fish. Aquat. Sci. 1990, 47 (7), 1446− 1452. (14) Lamelas, C.; Slaveykova, V. I. Comparison of Cd(II), Cu(II), and Pb(II) biouptake by green algae in the presence of humic acid. Environ. Sci. Technol. 2007, 41 (11), 4172−4178. (15) Aristilde, L.; Xu, Y.; Morel, F. M. M. Weak organic ligands enhance zinc uptake in marine phytoplankton. Environ. Sci. Technol. 2012, 46 (10), 5438−5445. (16) Crémazy, A.; Campbell, P. G. C.; Fortin, C. The biotic ligand model can successfully predict the uptake of a trivalent ion by a unicellular alga below pH 6.50 but not above: Possible role of hydroxo-species. Environ. Sci. Technol. 2013, 47 (5), 2408−2415. (17) Crémazy, A.; Campbell, P. G. C.; Fortin, C. In the presence of fluoride, free Sc3+ is not a good predictor of Sc bioaccumulation by two unicellular algae: Possible role of fluoro-complexes. Environ. Sci. Technol. 2014, 48 (16), 9754−9761. (18) Cornell, D. H. Rare earths from supernova to superconductor. Pure Appl. Chem. 1993, 65 (12), 2453−2464. (19) Bauer, D.; Diamond, D.; Li, J.; Sandalow, D.; Telleen, P.; Wanner, B. Critical materials strategy; U.S. Department of Energy: Washington, DC, 2010. (20) Noack, C. W.; Dzombak, D. A.; Karamalidis, A. K. Rare earth element distributions and trends in natural waters with a focus on groundwater. Environ. Sci. Technol. 2014, 48 (8), 4317−4326. (21) Miekeley, N.; Dejesus, H. C.; Dasilveira, C. L. P.; Linsalata, P.; Morse, R. Rare-earth elements in groundwaters from the Osamu Utsumi mine and Morro do Ferro analogue study sites, Poços de Caldas, Brazil. J. Geochem. Explor. 1992, 45 (1−3), 365−387. (22) Weltje, L.; Verhoof, L.; Verweij, W.; Hamers, T. Lutetium speciation and toxicity in a microbial bioassay: Testing the free-ion model for lanthanides. Environ. Sci. Technol. 2004, 38 (24), 6597− 6604. (23) Tai, P. D.; Zhao, Q.; Su, D.; Li, P. J.; Stagnitti, F. Biological toxicity of lanthanide elements on algae. Chemosphere 2010, 80 (9), 1031−1035. (24) Harris, E. H., A comprehensive guide to biology and laboratory use. The Chlamydomonas sourcebook; Academic Press: San Diego, 1989. (25) Critically Selected Stability Constants of Metal Complexes, Version 8.0; NIST Standard Reference Database 46; NIST: Gaithersburg, MD, 2004. (26) Wilkinson, K. J.; Buffle, J. Critical evaluation of physicochemical parameters and processes for modelling the biological uptake of trace metals in environmental (aquatic) systems. In Physicochemical kinetics and transport at biointerfaces; van Leeuwen, H. P., Köster, W., Eds.; John Wiley & Sons, Ltd.: West Sussex, England, 2004; pp 445−533.

ternary complex was included, suggesting that REE complexes will need to be taken into account when attempting to evaluate the environmental risk of these emerging metal contaminants. As organic ligands are ubiquitous in natural waters, the majority of REEs will be found in the form of complexes and there is thus a high risk that simple models/understanding will greatly underestimate REE bioavailability. It follows that improved speciation analysis of the REEs will be critical in predicting their bioavailability, especially for natural waters where the stability constants of the REE-NOM (natural organic matter) complexes are currently poorly defined.

■

ASSOCIATED CONTENT

S Supporting Information *

Detailed methodology on the analysis of metal concentrations and the digestion and analysis of 14C-labeled samples; Calculation of maximum diffusive flux of Tm3+; Theoretical calculation of the dissociation rate constant of a metal complex; Table S1 Equilibrium constants of rare earth element complexes used to calculate metal speciation; Table S2 Water loss rate constants of divalent and trivalent metal ions; Table S3 Dissociation rate constants; Table S4 The proportion of Tm complexes for each of the six ligands; Figure S1 Tm bioaccumulation in C. reinhardtii as a function of exposure time; Figure S2 Efficiency of the EDTA extraction; Figure S3 Tm bioaccumulation in the presence of organic ligands as a function of exposure time. This material is available free of charge via the Internet at http://pubs.acs.org/.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +1-514 343 6741. Fax: +1-514 343 7586. E-mail: kj. [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Project funding from the Natural Sciences and Engineering Council of Canada (Strategic project grant), the Fonds de recherche du Québec (Strategic initiative grant) and Environment Canada is greatly appreciated.

■

REFERENCES

(1) Morel, F. M. M.; Hering, J. G. Principle and Applications of Aquatic Chemistry; Wiley-Interscience: New York, 1993. (2) Campbell, P. G. C. Interactions between trace metals and aquatic organisms: A critique of the free-ion activity model. In Metal speciation and bioavailability in aquatic systems; Tessier, A., Turner, D., Eds.; John Wiley: New York, 1995; pp 45−102. (3) Di Toro, D. M.; Allen, H. E.; Bergman, H. L.; Meyer, J. S.; Paquin, P. R.; Santore, R. C. Biotic ligand model of the acute toxicity of metals. 1. Technical basis. Environ. Toxicol. Chem. 2001, 20 (10), 2383−2396. (4) Campbell, P. G. C.; Errecalde, O.; Fortin, C.; Hiriart-Baer, W. R.; Vigneault, B. Metal bioavailability to phytoplankton-applicability of the biotic ligand model. Comp. Biochem. Physiol., C: Toxicol. Pharmacol. 2002, 133 (1−2), 189−206. (5) Slaveykova, V. I.; Wilkinson, K. J. Predicting the bioavailability of metals and metal complexes: Critical review of the biotic ligand model. Environ. Chem. 2005, 2 (1), 9−24. (6) McLaughlin, M. J.; Andrew, S. J.; Smart, M. K.; Smolders, E. Effects of sulfate on cadmium uptake by Swiss chard: I. Effects of complexation and calcium competition in nutrient solutions. Plant Soil 1998, 202 (2), 211−216. 2213

DOI: 10.1021/es505443s Environ. Sci. Technol. 2015, 49, 2207−2214

Article

Environmental Science & Technology (27) Yang, G.; Tan, Q.-G.; Zhu, L.; Wilkinson, K. J. The role of complexation and competition in the biouptake of europium by a unicellular alga. Environ. Toxicol. Chem. 2014, 33 (11), 2609−2615. (28) Kola, H.; Wilkinson, K. J. Cadmium uptake by a green alga can be predicted by equilibrium modelling. Environ. Sci. Technol. 2005, 39 (9), 3040−3047. (29) Worms, I. A. M.; Parthasarathy, N.; Wilkinson, K. J. Ni uptake by a green alga. 1. Validation of equilibrium models for complexation effects. Environ. Sci. Technol. 2007, 41 (12), 4258−4263. (30) Lavoie, M.; Campbell, P. G. C.; Fortin, C. Extending the biotic ligand model to account for positive and negative feedback interactions between cadmium and zinc in a freshwater alga. Environ. Sci. Technol. 2012, 46 (21), 12129−12136. (31) Chen, Z. Z.; Zhu, L.; Wilkinson, K. J. Validation of the biotic ligand model in metal mixtures: Bioaccumulation of lead and copper. Environ. Sci. Technol. 2010, 44 (9), 3580−3586. (32) Poldoski, J. E. Cadmium bioaccumulation assays - Their relationship to various ionic equilibria in lake-superior water. Environ. Sci. Technol. 1979, 13 (6), 701−706. (33) Phinney, J. T.; Bruland, K. W. Uptake of Lipophilic Organic Cu, Cd, and Pb Complexes in the Coastal Diatom Thalassiosira weissf logii. Environ. Sci. Technol. 1994, 28 (11), 1781−1790. (34) Boullemant, A.; Lavoie, M.; Fortin, C.; Campbell, P. G. C. Uptake of hydrophobic metal complexes by three freshwater algae: Unexpected influence of pH. Environ. Sci. Technol. 2009, 43 (9), 3308−3314. (35) Simkiss, K.; Taylor, M. L. Transport of metals across membranes. In Metal Speciation and Bioavailability in Aquatic Systems; Tessier, A., Turner, D., Eds.; John Wiley: New York, 1995; pp 1−44. (36) Pasciak, W. J.; Gavis, J. Transport limitation of nutrient uptake in phytoplankton. Limnol. Oceanogr. 1974, 19 (6), 881−898. (37) Bayen, S.; Ter Laak, T. L.; Buffle, J.; Hermens, J. L. M. Dynamic exposure of organisms and passive samplers to hydrophobic chemicals. Environ. Sci. Technol. 2009, 43 (7), 2206−2215. (38) Buffle, J.; Zhang, Z.; Startchev, K. Metal flux and dynamic speciation at (Bio)interfaces. Part 1: Critical evaluation and compilation of physicochemical parameters for complexes with simple Ligands and Fulvic/Humic substances. Environ. Sci. Technol. 2007, 41 (22), 7609−7620. (39) van Leeuwen, H. P.; Town, R. M.; Buffle, J. Impact of ligand protonation on Eigen-type metal complexation kinetics in aqueous systems. J. Phys. Chem. A 2007, 111 (11), 2115−2121. (40) Hudson, R. J. M.; Morel, F. M. M. Iron transport in marine phytoplankton: Kinetics of cellular and medium coordination reactions. Limnol. Oceanogr. 1990, 35 (5), 1002−1020. (41) Slaveykova, V. I.; Wilkinson, K. J.; Ceresa, A.; Pretsch, E. Role of fulvic acid on lead bioaccumulation by Chlorella kesslerii. Environ. Sci. Technol. 2003, 37 (6), 1114−1121. (42) Lamelas, C.; Wilkinson, K. J.; Slaveykova, V. I. Influence of the composition of natural organic matter on Pb bioavailability to microalgae. Environ. Sci. Technol. 2005, 39 (16), 6109−6116. (43) Wilkinson, K. J.; Bertsch, P. M.; Jagoe, C. H.; Campbell, P. G. C. Surface complexation of aluminum on isolated fish gill cells. Environ. Sci. Technol. 1993, 27 (6), 1132−1138. (44) Slaveykova, V. I.; Wilkinson, K. J. Physicochemical aspects of lead bioaccumulation by Chlorella vulgaris. Environ. Sci. Technol. 2002, 36 (5), 969−975. (45) Pałasz, A.; Czekaj, P. Toxicological and cytophysiological aspects of lanthanides action. Acta Biochim. Polym. 2000, 47, 1107−1114.

2214

DOI: 10.1021/es505443s Environ. Sci. Technol. 2015, 49, 2207−2214