Uptake of silver by a unicellular alga:

exceptions to the free-ion model

Claude Fortin and Peter G.C. Campbell

Université du Québec, INRS-Eau, C.P. 7500, Sainte-Foy, QC, Canada G1V 4C7

fortinc@uquebec.ca

ABSTRACT

Short-term (< 1 h) silver uptake by the green alga Chlamydomonas reinhardtii was measured in the laboratory in defined inorganic media in the presence or absence of ligands (chloride and thiosulfate). In contradiction to the Free-Ion Model of metal uptake, silver accumulation by the alga proved to be sensitive to the choice of ligand used to buffer the free silver concentration. For a low fixed free Ag⁺ concentration of 10 nM, silver uptake in the presence of thiosulfate (0.11 µM) was 2X greater than in the presence of chloride (4 mM). When sulfate was removed from the exposure medium, silver uptake in the presence of thiosulfate was even more markedly enhanced (more than 4X greater than in the presence of chloride). Varying the sulfate concentration in the exposure medium only affected silver uptake if thiosulfate was present. We conclude that silver-thiosulfate complexes are transported across the plasma membrane via sulfate / thiosulfate transport systems, and that sulfate acts as a competitive inhibitor of this uptake mechanism.

Introduction

It is generally accepted that the total aqueous concentration of a metal is not a good predictor of its <bioavailability>, i.e. the metal's speciation will affect its availability to aquatic organisms. Qualitatively, complexation of a metal normally leads to a decrease in its bioavailability – in effect, most dissolved ligands that bind metals form hydrophilic complexes, ML_n^±, and in such systems metal uptake, nutrition and toxicity normally vary as a function of the concentration of the free-metal ion in solution (Morel and Hering 1993). However, a number of intriguing experiments have been reported in the literature where the metal's "residual" bioavailability in the presence of hydrophilic ML_n^± complexes has been found to exceed that which would have been predicted on the basis of the free-metal ion concentration at equilibrium. Most of these apparent exceptions to the Free-Ion Model (FIM) of metal toxicity involve organic ligands that are assimilable in their own right, and this has led to the suggestion that "accidental" metal transport may occur in their presence (i.e., the ligand is assimilated as a metal-ligand complex and the metal "comes along for the ride") (Campbell 1995). In principle, the assimilation of intact hydrophilic metal-ligand complexes could also occur with inorganic ligands such as sulfate. Uptake systems for such essential nutrient anions exist at biological interfaces; if these transport systems could be "fooled" into binding and transporting the intact metal-anion complex, then the metal would find its way into the cell "accidentally". The binding of Ag by thiosulfate (AgS₂O₃^-, Ag(S₂O₃)₂^3‑: log K₁ = 8.82, log b₂ = 13.50) will reduce the free Ag⁺ concentration and thus, according to the FIM, should reduce silver bioavailability. However, exactly the opposite results have been reported for Ag accumulation by rainbow trout, Oncorhynchus mykiss, in laboratory exposure experiments (Hogstrand et al. 1996; Wood et al. 1996). We postulated that silver could cross biological membranes as the silver-thiosulfate complex, via an anion transporter, and set out to test this "molecular mimicry" hypothesis (Clarkson 1993) using a unicellular alga as our biological model. Since algae are known to possess membrane-bound transport systems for the assimilation of sulfate (Hodson et al. 1968; Pérez-Castiñeira et al. 1998), they should be appropriate models for testing the hypothesis that thiosulfate (and 1:1 silver-thiosulfate complexes) can mimic sulfate and enter the cells via the same pathway.

METHODS

The experiments were carried out with a euryhaline unicellular green alga, Chlamydomonas reinhardtii, in defined inorganic media. Cells were grown axenically in 100 mL of modified high salt medium (Fortin and Campbell 2000) with an ionic strength of 6 meq·L^-1. Silver uptake experiments were done with radiolabeled ^110mAg (136 mCi·mmol^-1; Amersham Canada). For each experiment, cells were initially inoculated at a density of 2,500 cells·mL^-1, allowed to grow for 48 h to reach mid-exponential growth, and then gently harvested on a 2-µm polycarbonate filter membrane (Poretics) using a vacuum pressure of £ 10 cm Hg. Harvested cells were rinsed five times with 10 mL of sterile simplified culture medium containing neither phosphate nor trace metals, and then re-suspended in ~ 10 mL of the same simplified medium. Size distribution, average surface area and density were rapidly determined using a Coulter Multisizer II particle counter (70-µm orifice tube) and recorded. Cells were then exposed under the conditions outlined in Table 1 for a short period of time (15 min). Short exposure times were used to avoid release by the algal cells of metal-binding peptides that could affect silver speciation in solution, and to minimize cell division that would increase cell density during the exposure. Experiments were conducted under ambient laboratory conditions and with low cell numbers (10,000 cells·mL^‑1), to minimize silver depletion through uptake and adsorption by the algal cells (e.g., minimal decrease in dissolved silver concentration, < 5 % after 15 minutes). Finally, cells were recovered and rinsed to remove the surface adsorbed silver as described in Fortin and Campbell (2000). All uptake experiments were performed at neutral pH (7.0 ± 0.1) without any buffers, with a minimum of three replicates and uptake values were then normalized for the total algal surface area. Silver speciation in the exposure solutions was calculated with the chemical speciation model MINEQL+ (Schecher and McAvoy 1994) with an updated thermodynamic data base prepared from a reliable source of thermodynamic data (Martell et al. 1998). The data base is available at http://www.inrs-eau.uquebec.ca/activites/groupes/biogeo/personal.htm.

Table 1: Exposure conditions for the silver uptake experiment.

Medium	[Ag]T (nM)	[Ag^+] (nM)	[Cl^-] (mM)	[NO3^-] (mM)	[SO4^2-] (µM)	[S2O3^2-] (µM)
A	104	10	0.05	5.23	0.0	0.114
B	104	104	0.05	5.23	0.0	0.000
C	104	10	0.05	5.23	120	0.114
D	104	10	4.00	1.23	0.0	0.000

Silver uptake (15 min) was determined in the absence of complexing ligands (medium B) and compared with uptake in three contrasting media: (A) 114 nM S₂O₃, 0 µM SO₄; (C) 114 nM S₂O₃, 120 µM SO₄; (D) 4 mM Cl, 0 µM SO₄. Total silver was held constant at 104 nM in all exposures and the free silver concentration was also constant 10 nM, except in the non-complexing medium B that contained neither chloride nor thiosulfate and thus had a free silver concentration equal to the total silver concentration.

RESULTS AND DISCUSSION

As expected, silver uptake in the presence of chloride (4 mM) was lower than in the non-complexing medium (0.53; Fig. 1, column D ¸ column B), consistent with the anticipated protective effect of chloride complexation. Uptake in the thiosulfate / sulfate medium was also less than in the non-complexing medium, but only slightly so (0.85; Fig. 1, column C ¸ column B; t‑test, p < 0.05), but much less than in the sulfate-free medium (0.44; Fig. 1, column C ¸ column A), suggesting a role for sulfate in silver uptake. Even more remarkably, silver uptake in the sulfate-free thiosulfate medium A was higher than in the ligand-free medium B (1.9; Fig. 1, column A ¸ column B), revealing an enhancement of silver uptake even though the free Ag⁺ concentration for column A was 10X lower than for column B.

Fig. 1: Comparison of silver uptake after 15 min of exposure from four exposure media at either 10 or 104 nM Ag⁺ (t) or Ag_T (). Error bars represent standard deviations from the average of three measurements.

In previous experiments we demonstrated that silver uptake by C. reinhardtii was enhanced in the presence of chloride – e.g., for a fixed free Ag⁺ concentration (10 nM), silver uptake increased markedly when the external chloride concentration was increased from 5 µM to 4 mM. The enhanced uptake observed in the presence of chloride was related to the very high silver uptake rates demonstrated by the test alga (e.g. ~ 1000X those observed for Cd²⁺ and Mn²⁺), which led to diffusion limitation in the boundary layer surrounding the algal cell (Fortin and Campbell 2000). In such a situation, metal accumulation is proportional to the total metal concentration (i.e., to the concentration gradient between the bulk solution and the algal surface). This diffusion limitation dissipated at total Ag concentrations greater than 10^‑7 M.

A similar but greater increase in silver uptake was observed in the current experiments in the presence of thiosulfate (Fig. 1, compare media A and C with medium D). In this case, however, changes in total silver concentration cannot be invoked to explain the enhanced metal uptake in media A, C and D, since both total and free silver concentrations were equal in all three media (104 and 10 nM, respectively – see Table 1). Instead, we conclude (i) that the enhanced uptake observed in the presence of thiosulfate is the result of silver-thiosulfate complexes being transported across the plasma membrane via sulfate / thiosulfate transport systems, and (ii) that this membrane transport mechanism is affected by the external sulfate concentration.

There are several indications in the literature that sulfate and thiosulfate share a common membrane transport system in bacteria (Sirko et al. 1995) and algae (Hodson et al. 1968). A competitive effect between sulfate and thiosulfate has been noted in sulfate uptake experiments with C. reinhardtii (Pérez-Castiñeira et al. 1998), thiosulfate being an efficient inhibitor of sulfate uptake. Several algal species can grow on thiosulfate as a sole sulfur source (Hodson et al. 1968; Pérez-Castiñeira et al. 1998). Removal of sulfate lead to a greater than 4‑fold increase in silver uptake compared to the chloride exposure medium (Fig. 1, compare media A and D). This result clearly supports our contention that a sulfate / thiosulfate transporter is involved in silver uptake in the presence of thiosulfate. Progressive addition of sulfate resulted in a gradual decrease in silver uptake (results not shown), as would be expected from sulfate / thiosulfate competition for a membrane transport system. Note, however, that addition of excess sulfate did not depress silver uptake to the levels measured in the thiosulfate-free medium. We conclude that in the absence of thiosulfate (media B and D) silver is taken up via a cation transporter (probably via a Cu(I) transport system: Fortin and Campbell 2000) and that this transporter is unaffected by changes in ambient sulfate concentrations. In media A and C, however, a second parallel pathway for silver uptake is introduced, involving the accidental transport of silver-thiosulfate complexes via one or more sulfate / thiosulfate transporters.

The quantitative importance of the silver-thiosulfate uptake pathway can be deduced from Fig. 1 (comparison of columns A and B). For equal total silver concentrations (104 nM), silver uptake after 15 min exposure was 2X higher in the presence of thiosulfate (in a sulfate-free medium) than in its absence. Silver uptake is thus not only "greater than would have been expected" on the basis of the free Ag⁺ concentration, but is in fact truly enhanced by the presence of thiosulfate. Residual uptake due to the presence of 10 nM Ag⁺ in medium A was estimated to be of 0.33 µmol·m^-2 (10 % of column B) whereas uptake due to 94 nM AgS₂O₃^- complexes in medium A was 6.1 µmol·m^-2 (column A less 0.33 µmol·m^-2). On an equimolar basis, silver uptake after 15 minutes would be 32 nmol·m^‑2·nM for Ag⁺ compared to 65 nmol·m^-2·nM for AgS₂O₃^-. Silver uptake rates via the thiosulfate transport system are thus twice those of silver through the cation transporter.

The prevailing paradigm for metal uptake by aquatic organisms, i.e. the Free Ion Model or its derivative the Biotic Ligand Model, assumes that metals enter living cells via facilitated cation transport. Most known exceptions to the FIM involve either ligands that form lipophilic complexes, M‑L_n^o, which can bypass normal metal transport mechanisms and cross biological membranes by simple diffusion, or "chaperone" ligands that are synthesized by living (micro)organisms specifically to complex essential metals and facilitate their eventual uptake (e.g., the role of siderophores in iron nutrition). In contrast, evidence of metal uptake through anion transport systems is scarce to non-existent. To our knowledge the present results represent the first hard evidence for metal transport into cells via an inorganic anion transport system.

The environmental implications of our findings will depend on two factors: (i) how likely are metal-thiosulfate complexes to exist in the exposure medium (e.g., in natural waters), and (ii) how widespread are the membrane transport systems involved in the movement of sulfate / thiosulfate across biological membranes?

REFERENCES

(1)Campbell, PGC (1995), In: Metal Speciation and Bioavailability in Aquatic Systems (A Tessier, DR Turner, Editors). New York, John Wiley & Sons Ltd, pp 45-102.

(2)Clarkson TW (1993), Ann. Rev. Pharmacol. Toxicol. 32: 545-571.

(3)Fortin C, Campbell PGC (2000), Environ. Toxicol. Chem. In press.

(4)Hodson RC, Schiff JA, Scarsella AJ (1968), Plant Physiol. 43: 570-577.

(5)Hogstrand C, Galvez F, Wood CM (1996), Environ. Toxicol. Chem. 15: 1102-1108.

(6)Martell AE, Smith RM, Motekaitis RJ (1998), NIST Standard Reference Database 46. U.S. Dept. of Commerce: Gaithersburg, MD.

(7)Morel FMM, Hering JG (1993), Principles and Applications of Aquatic Chemistry. New York, John Wiley & Sons.

(8)Pérez-Castiñeira JR, Prieto JL, González-Arroyo JG, Vega JMJ (1998), Plant Physiol. 153: 324-331.

(9)Schecher WD, McAvoy D (1994), Environmental Research Software. Hallowell, ME.

(10)Sirko A, Zatyka, M, Sadowy E, Hulanicka D (1995), J. Bacteriol. 177: 4134-4136.

(11)Wood CM, Hogstrand C, Galvez F, Munger RS (1996), Aquat. Toxicol. 35: 111-125.