Role of Speciation in Organotin Toxicity to the Yeast - American

Jun 8, 2004 - JANE S. WHITE AND JOHN M. TOBIN*. School of Biotechnology, Dublin City University,. Dublin 9, Ireland. Assessment of organometal ...
4 downloads 0 Views 179KB Size
Download PDF
Environ. Sci. Technol. 2004, 38, 3877-3884

Role of Speciation in Organotin Toxicity to the Yeast Candida maltosa JANE S. WHITE AND JOHN M. TOBIN* School of Biotechnology, Dublin City University, Dublin 9, Ireland

Assessment of organometal pollution requires an understanding of the various processes that influence the bioavailability and toxicity of the contaminant. Organotins may exist as both cationic species and neutral hydroxides in aqueous solution, with the formation of chloride species in the presence of Cl-. Although these species have different chemical properties, there is very little information on the influence of speciation on organotin and microbial cell interactions. Tributyltin (TBT) and triphenyltin (TPT) interactions with the yeast Candida maltosa were investigated between pH 3.5 and 7.5 and in up to 0.5 M NaCl at pH 5.5. Toxicity increased with both pH and NaCl concentration and the mechanisms of interaction depended on the species present in solution. TBT and TPT interacted by different mechanisms, as evidenced by action on membrane fluidity. Furthermore, there was a strong correlation between toxicity and overall octanolwater distribution ratio (Dow) of organotin compounds. Triorganotin cations are less toxic than triorganotin hydroxides, which are in turn less toxic than triorganotin chlorides. These findings underline the importance of speciation effects on organotin interactions in the environment.

Introduction Tributyltin (TBT) and triphenyltin (TPT), as a consequence of their widespread use in antifouling paints, have been frequently detected in marine and freshwater ecosystems at concentrations exceeding toxicity levels (1). Triorganotin (TOT) pollution is also related to the use of TBT as preservative for timber, wood, textiles, paper, and leather (2), while TPT is often used as a fungicide to protect crops, including potato, celery, sugar beet, and rice and to prevent tropical diseases in peanuts, pecans, coffee, and cocoa (3). As organotin pollution occurs in a wide variety of ecosystems, including water, sediment, and biota of both freshwater and estuarine environments (4, 5), in soil (6, 7), and in wastewater and sewage sludge (7, 8), understanding the influence of the immediate environment on toxicity is of great importance. Solution pH, ionic composition and strength, and temperature affect the chemical speciation and solubility of organotins (9, 10). When dissolved in aqueous solution, triorganotins form positively charged diaquo complexes, R3Sn(H2O)2+, which can dissociate to the hydroxo complex, R3SnOH. In the presence of chloride ions, R3SnCl species may also form. The individual species may interact by different mechanisms, with complexation of charged species with cell ligands occurring in addition to or in place of * Corresponding author phone: +353 1 7005408; fax: +353 1 7005412; e-mail: [email protected]. 10.1021/es030099k CCC: $27.50 Published on Web 06/08/2004

 2004 American Chemical Society

hydrophobic partitioning of neutral species. For example, liposome-water partitioning coefficients (Dlipw) of TBT and TPT in phosphatidylcholine liposomes were higher for R3Sn+ than for R3SnOH species (11). In contrast, R3SnOH species have been reported to be more biologically active than R3Sn+ species. TBT uptake rates, bioaccumulation, and mortality are significantly higher in Daphnia magna at pH 8 than at pH 6 (12), while increased bioconcentration of TBT and TPT in Chironomus riparius at pH 8 compared to pH 5 was attributed to greater uptake of R3SnOH species (13). Similarly, TBT and TPT bioconcentration in carp increased between pH 6.0 and 7.8 (14). Overall, however, there is limited research specifically on the influence of pH and NaCl on the interaction of organotins with microorganisms. Extrapolated comparisons are hampered by the differing methodologies (15-18). Many studies use microorganisms exposed to organotins in the presence of glucose, so metabolism-dependent interactions may not be discounted, while the use of complex media may result in organotin complexation with media ligands. Also, reported reduction in TBT toxicity and uptake in the presence of NaCl (15, 16, 18) may be due to the formation of insoluble tributyltin chloride, which would reduce the availability of TBT in solution. In this work, the influence of pH and NaCl on the interactions of TBT and TPT with nonmetabolizing yeast cells at soluble organotin concentrations was assessed. A model developed by Arnold et al. (10) was used to predict TBT and TPT aqueous speciation and lipophilicity, as indicated by overall 1-octanol-water distribution ratios (Dow). Candida maltosa, an isolate from an asphalt refinery and its watershed (19), was selected as a model organism. Organotin contamination is often associated with hydrocarbon-polluted environments, where enrichment of C. maltosa occurs. Also, this organism was previously used to demonstrate the different interaction mechanisms between inorganic tin and organotins with yeast (20, 21). Toxicity of the compounds was reported as loss in cell viability, while membrane interactions were assessed by K+ release and alterations in membrane fluidity. The use of the fluorescent probes 1,6-diphenyl-1,3,5hexatriene (DPH) and 1-[4-(trimethylamino)phenyl]-6phenyl-1,3,5-hexatriene (TMA-DPH) in monitoring the effects of organotins on liposomes (22, 23) and C. maltosa membrane fluidity (21) has been documented. Change in anisotropy reflects alteration in probe rotational mobility and the degree of order of the phospholipid acyl chains in the immediate region, with an inverse relationship existing between membrane fluidity and anisotropy (24). The results presented here have important implications from an environmental perspective. The Cl-- and pH-dependent speciation of organotins has consequences for partitioning in microbial cells and plays a pivotal role in the physicochemical basis of their toxicity to microorganisms.

Experimental Section Reagents. All chemicals were obtained from Sigma-Aldrich (Dorset, U.K.) unless stated otherwise. Tributyltin chloride [(C4H9)3SnCl] and triphenyltin chloride [(C6H5)3SnCl] stock solutions were prepared in HPLC-grade methanol. Fluorescent probes 1,6-diphenyl-1,3,5-hexatriene (DPH) and 1-[4(trimethylamino)phenyl]-6-phenyl-1,3,5-hexatriene (TMADPH) were prepared as 2 × 10-4 M stock solutions in N,Ndimethylformamide. Organism and Culture Conditions. Candida maltosa DSM 15531 was routinely maintained on a yeast extractpeptone medium (YEPD) as described previously (21). For experimental purposes, C. maltosa was grown in YEPD (less VOL. 38, NO. 14, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3877

FIGURE 1. Effect of pH on the speciation of (a) TBT and (b) TPT and effect of Cl- concentration on (c) TBT and (d) TPT at pH 5.5. Speciation was predicted by use of the model of Arnold et al. (10). agar) and cells were harvested in the exponential phase by centrifugation (300g, 5 min) and washed twice with deionized water, followed by three washings with either 10 mM 2-(Nmorpholino)ethanesulfonic acid (MES) (pH 3.5, 4.5, and 5.5) or 5 mM piperazine-N,N′-bis(2-ethansulfonic acid) (PIPES) (pH 7.5) buffer. Where specified, NaCl was added to the desired concentration prior to washing of cells. Cells were used within 3 h of harvesting. Exposure to Organotins. Experiments were designed so that only soluble organotin concentrations were employed. TBT and TPT solubility was determined in pH 5.5 and 7.5 buffers and in 0.5 M NaCl buffer, pH 5.5, following the method of Inaba et al. (9). On the basis of the solubility levels, the highest organotin concentrations examined were 100 µM TBT between pH 3.5 and 7.5, 30 µM TPT up to pH 5.5, and 10 µM TPT at pH 7.5. The maximum organotin concentrations examined in the presence of NaCl were 50 µM TBT and 20 µM TPT. The use of soluble organotin concentrations under the specified experimental conditions was confirmed by the routine analysis of cell-free controls. Overall, the soluble concentrations used here were higher than those previously reported (9). The reason for this was the inclusion of methanol (up to 2%) as a cosolvent as the organotins were added from methanol stock solutions. To determine organotin uptake and toxicity, yeast cells, harvested and washed as described above, were suspended in 50 mL of 10 mM MES or 5 mM PIPES buffer to an approximate cell density of 1 × 107 cells mL-1. For experiments involving NaCl, cells were washed with buffer containing the corresponding NaCl concentration prior to exposure to organotins. After 30 min of equilibration by shaking at 150 rpm at room temperature, TBT or TPT (from stock solutions) was added to the desired concentration. Organotin controls, less cells, were included as reference to initial organotin concentrations. After 30 min of incubation, 5 mL samples were removed and the biomass separated by centrifugation (2500g, 5 min). Supernatants were retained for K+ and tin analysis. Cell suspension samples were also obtained for viability studies. Cell Viability and Metal Analysis. In all experiments, cell viability was monitored by growth on YEPD plates (21). Viability was defined as the ratio of experimental to control counts, expressed as a percentage. K+ was analyzed by atomic absorption spectrophotometry (AAS) and TBT by hydride generation AAS according to the methods of White and Tobin (21). TPT was analyzed on a model 394 electrochemical trace analyzer with a 303A static mercury drop electrode (EG&G Instruments, Princeton Applied Research). Differential pulse polarography (pulse amplitude 50 mV) and a hanging mercury drop working electrode were employed for analysis 3878

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 14, 2004

and the peak for TPT was obtained at a half-wave potential of approximately -0.69 V. Samples were diluted in supporting electrolyte consisting of 0.16 M NH4Cl in 40% (v/vaq) ethanol and adjusted to pH 2.5 by use of 2 M HCl. Samples were purged with N2 for 3 min and each sample was analyzed three times with a 30 s N2 purge between each reading. TPT concentrations were determined by reference to appropriate standard solutions. Fluorescence Anisotropy Measurements. Fluorescence anisotropy was monitored as described by White and Tobin (21) with organotin effects on membrane fluidity reported as the difference in anisotropy readings between cells exposed to organotin and organotin-free controls. TMA-DPH or DPH was added to cell suspensions to a final concentration of 0.5 µM, prior to the initial 30 min incubation period. Cells were exposed to organotins as described above. Fluorescence intensity and anisotropy readings were recorded after 30 min of exposure time on a Perkin-Elmer LS 50 luminescence spectrometer (Buckinghamshire, U.K.) fitted with horizontal and vertical polarizers. The excitation and emission wavelengths for both probes were 360 and 450 nm, respectively.

Results and Discussion Organotin Speciation and Dow. The variations in organotin speciation and Dow values with pH and NaCl concentration were predicted by use of the theoretical model of Arnold et al. (10). Appropriate chloride concentrations were included in the modeling to account for the addition of Cl- from organotin chloride stock solutions. Similarly, appropriate NO3- concentrations were included as HNO3 was used to adjust pH. The fractions of hydrated cationic species, R3Sn(H2O)2+, represented here as TBT+, TPT+, or R3Sn+, and neutral hydroxide species, TBTOH, TPTOH, or R3SnOH, present between pH 3.5 and 7.5 were predicted (Figure 1a,b). The fraction of TOT present as NO3- complex accounted for less than 0.2% of the total fraction of TOT species and was not included in Figure 1. The formation of neutral chloride species, TBTCl, TPTCl, or R3SnCl, in up to 0.5 M NaCl was predicted at pH 5.5 (Figure 1c,d). For TPT, the total fraction of neutral species remains relatively constant (63-65%) over the Cl- concentration range, with TPTCl being formed in place of TPTOH. For TBT, the fractions of both total neutral and chloride species increased with Cl-, with 15% and 38% neutral species present at 0 and 0.5 M Cl-, respectively. For both organotins, Dow, predicted at maximum concentrations of 100 µM TBT and 30 µM TPT, increased by approximately 2 orders of magnitude between pH 3.5 and 7.5 (Figure 2a). In calculating Dow the contribution of R3Sn+ species, negatively charged complexes with multivalent anions and possible triorganotin di- or oligomers is assumed

FIGURE 2. Variation in Dow values of TBT (b) and TPT (O) with (a) pH and (b) Cl- concentration (pCl) as predicted by use of the model of Arnold et al. (10).

FIGURE 3. Uptake of (a) TBT and (b) TPT at pH 3.5 (2), 4.5 (9), 5.5 (b), and 7.5 (4). to be insignificant (10). Dow values of TBT and TPT at pH 5.5 increased with Cl- (Figure 2b) corresponding to the increasing fraction of R3SnCl species. Log Dow of TBT was 3.27 and log Dow of TPT was 3.35 in the absence of NaCl, increasing to 4.60 and 3.94, respectively, in 0.5 M NaCl. Role of pH in Organotin Uptake and Toxicity. TBT and TPT interactions with yeast varied considerably with external pH. TBT uptake levels were similar between pH 4.5 and 7.5 but were reduced at pH 3.5, while uptake of TPT increased between pH 3.5 and 5.5 (Figure 3). Organotin effects on cell viability increased with pH (Figures 4 and 5). TBT-induced K+ leakage was greatest at pH 5.5 (Figure 4), while TPTinduced K+ release increased with pH (Figure 5). Increased toxicity was not necessarily due to higher uptake levels but was also dependent on the species present in solution. For example, TBT uptake was similar above pH 4.5, while toxicity increased. Cationic and neutral species interacted differently as evidenced by the differing effects on DPH and TMA-DPH fluorescence anisotropy (Figures 6 and 7). At pH 3.5,

membrane-damaging effects of both organotins were negligible. At pH 4.5, TBT caused a decrease in TMA-DPH anisotropy while TPT had no effect. Both organotins had a similar effect at pH 5.5 with increased DPH anisotropy. TPTinduced changes in membrane fluidity were similar at pH 7.5 compared to pH 5.5, while TBT had no effect at the higher pH. Changes in external pH may influence the uptake and toxicity of organotins in at least two ways: (i) by altering the state of the cell surface and (ii) by affecting organotin speciation. Uptake and toxicity of TBT and TPT were reduced considerably at pH 3.5. At this pH, greater than 98% R3Sn+ species are present. The absence of both K+ release and effects on membrane fluidity is consistent with organotin uptake via biosorption to the cell wall. The adsorptive capacity of the yeast cell wall for cationic species is determined by the degree of dissociation of negatively charged cell surface functional groups, which in turn is determined by the solution pH. For uptake of inorganic metals, a pH between 4.0 and 8.0 is widely accepted as being optimal for biosorptive uptake, while below this pH the protonation of possible binding sites becomes significant and biosorption is reduced (25). Here the reduction in uptake of both TBT and TPT indicates that competition with H+ ions was appreciable. Similarly, reduction in uptake of TBT by the cyanobacteria Synechocystis PCC6803 and Plectonema boryanum below pH 5.5 resulted from competition with H+ ions for binding sites (18). By contrast, protonation of binding sites was not a factor in partitioning of TBT and TPT in phosphatidylcholine liposomes, where uptake was greater at pH 3.0 than at higher pH (11). In this case speciation of the lipids was invariant to changes in pH and the greater sorption was attributed to the affinity of R3Sn+ for ligands in the phospholipids. TBT uptake levels were similar between pH 4.5 and 7.5 (Figure 3). In contrast, maximum uptake of 500 µM TBT by Synechocystis PCC6803 and P. boryanum occurred between pH 5.5 and 6.5 (18), and lower uptake above pH 6.5 was attributed to decreased biological activity of the hydroxide species. The apparent difference in behavior is likely due to insolubility of TBT at higher pH values, as the cyanobacteria were exposed to 500 µM TBT (18). TBT has a solubility of 134 µM in distilled water at pH 7.9, when added as a pure solution (9). Diminished competition from protons would lead to the expectation of increased uptake and toxicity above pH 3.5. For TBT, the difference in toxicity is not attributable only to greater uptake as cell death and K+ release differed even at similar uptake levels. Uptake mechanisms varied with speciation (and consequently pH) and so affected toxicity. For example, TBT had no effect on fluorescent anisotropy at pH 3.5, while TMA-DPH anisotropy decreased at pH 4.5 (Figure 6). An increase in membrane fluidity is consistent with partial insertion of the alkyl chains of TBT in the phospholipid bilayer. TMA-DPH is restrictively localized in the phospholipid head region of the cytoplasmic membrane (26), and compounds that adsorb at the lipid-water interface, with partial penetration of the lipid bilayer, are known to cause a marked fluidization effect (27). Thus, TBT uptake at pH 4.5 was not solely due to sorption at the cell wall, and TBT penetration to cytoplasmic membrane binding sites resulted in changes to membrane fluidity with consequent increased toxicity of TBT to cells. At pH 5.5, the anisotropy of DPH increased at concentrations above 20 µM TBT (Figure 6). In microbial cells, DPH is distributed between the hydrophobic region of lipidic membranes (26) and DPH anisotropy reflects the average fluidity of all cellular lipids (28). Thus, TBT disturbed the hydrophobic core of membranes, corresponding with greater cell death and extensive K+ release. At pH 5.5, 15% of TBT was present as neutral TBTOH, compared to 2% TBTOH at pH 4.5. The increased proportion of neutral species would VOL. 38, NO. 14, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3879

FIGURE 4. Influence of pH on the effect of TBT on cell viability (9) and K+ leakage (2) of C. maltosa. Cells were exposed to a range of TBT concentrations at pH (a) 3.5, (b) 4.5, (c) 5.5, and (d) 7.5.

FIGURE 5. Influence of pH on the effect of TPT on cell viability (9) and K+ leakage (2) of C. maltosa. Cells were exposed to a range of TPT concentrations at pH (a) 3.5, (b) 4.5, (c) 5.5, and (d) 7.5. favor greater membrane interactions. Surprisingly, the increase in TBT toxicity at pH 7.5 compared to pH 5.5 was less than that which would be predicted by aqueous speciation considerations. At pH 5.5, 15% TBTOH was present, increasing to 95% at pH 7.5. There was a marginal increase in toxicity at the higher pH, with complete loss in cell viability occurring at 70 and 50 µM TBT at pH 5.5 and 3880

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 14, 2004

7.5, respectively (Figure 4). However, membrane-damaging effects were actually reduced at pH 7.5. At concentrations above 50 µM TBT, K+ leakage was less at the higher pH (Figure 4), while there was no effect on membrane fluidity (Figure 6), indicating that TBTOH did not accumulate in the cytoplasmic membrane. These findings indicate that the combination of TBT+ and TBTOH species present at pH 5.5

FIGURE 6. Influence of pH on the effect of TBT on membrane fluidity of C. maltosa. Cells were exposed to a range of TBT concentrations at pH (a) 3.5, (b) 4.5, (c) 5.5, and (d) 7.5 and the effects on anisotropy of either DPH (b) or TMA-DPH (O) are shown. Change in anisotropy was calculated as the difference between the anisotropy of cells in the presence and absence of organotin.

FIGURE 7. Influence of pH on the effect of TPT on membrane fluidity of C. maltosa. Cells were exposed to a range of TPT concentrations at pH (a) 3.5, (b) 4.5, (c) 5.5, and (d) 7.5 and the effects on anisotropy of either DPH (b) or TMA-DPH (O) are shown. Change in anisotropy was calculated as the difference between the anisotropy of cells in the presence and absence of organotin. interacted differently with cells compared to TBTOH at pH 7.5. The absence of membrane effects at pH 7.5 would suggest that TBTOH diffused into the cell as discussed below. The effect of TPT on cell viability was enhanced with increasing pH and membrane interactions were greater, with

extensive K+ release at pH 5.5 compared to minimal changes in external K+ concentrations at lower pH (Figure 5). DPH anisotropy also increased with TPT concentration at pH 5.5, reflecting a reduction in fluidity at the hydrophobic core of membrane lipids (Figure 7). The change in TPT toxicity VOL. 38, NO. 14, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3881

corresponded with the fraction of neutral hydroxide species present in solution, which increased from 16% to 66% between pH 4.5 and 5.5. A similar increase in TPT toxicity occurred at pH 7.5, where 99% of TPT was present as TPTOH. However, comparisons were only possible at low organotin concentrations due to solubility constraints at pH 7.5. Evidently, TPTOH species are more toxic than TPT+ and uptake of the cation was restricted to cell wall biosorption, while TPTOH caused an increase in order of the fatty acyl chains in the hydrophobic core of the lipid bilayer. Similar localization of TPT in the phospholipid bilayer has been suggested, with TPT concentrations equivalent to those investigated here, causing an increase in DPH anisotropy in liposomes (23). This is similar to the effect of organophosphorus insecticides (29, 30) where the membrane location of the insecticides facilitates hydrogen bonds or dipoledipole interactions with fatty acyl chains, resulting in a decrease in lipid spacing and increased ordering of the membrane. In contrast, liposome-water partitioning of TPT+ was greater than that of TPTOH, with sorption of the cationic species governed by complex formation with phosphate groups (11). However, for cationic species direct comparisons between liposomes and cells may not be applicable, as effects of the yeast cell wall have to be taken into account. Overall, TBT and TPT cationic and hydroxide species interacted in contrasting ways with cells, as evidenced by action on membrane integrity. For the cationic species, this may be explained by the fact that TPT+ forms a stronger complex with oxygen ligands than TBT+ (11). Thus, under the conditions described here, TPT+ binds strongly to the yeast cell wall while TBT+ may interact by partial penetration in the cytoplasmic membrane. Also, membrane effects are concentration-dependent, and due to solubility constraints, cells were exposed to higher concentrations of TBT compared to TPT. At pH 4.5, changes in TMA-DPH anisotropy were only apparent after uptake of approximately 20 µmol of TBT (1010 cells)-1, while maximum TPT uptake was 15 µmol (1010 cells)-1. TPTOH caused a decrease in fluidity at the hydrophobic core, whereas the absence of TBTOH effects suggests that it does not accumulate in the cytoplasmic membrane. Diffusion of TBTOH across the cytoplasmic membrane may account for both this and reduced K+ leakage. Dissolution of lipophilic organic metal complexes in the membrane, followed by diffusion across the lipid bilayer and distribution among intracellular compartments, has been demonstrated (31). It is possible that the steric constraints of the larger phenyl groups of TPTOH may prevent diffusion through the membrane (32). Also, enhanced partitioning of TPTOH compared to TBTOH in liposomes was attributed to complex formation with the phospholipid groups (11). It appears that the greater lipophilicity of TBTOH, coupled with the additional complex formation of TPTOH with phospholipids, resulted in TBTOH diffusing into the cells while TPTOH was localized in the hydrophobic core of the lipid bilayer. Recently, TBT and TPT were found to have differing effects on the energy metabolism of bacterial chromatophores, with an OHuniport system observed only in the presence of TBT (33). The OH- uniport required membrane permeation of TBT, while the lack of TPT effects was attributed to its strong complexation to chromatophore ligands. Variation in Organotin Toxicity with Cl- Concentration. To further assess the influence of speciation on TBT and TPT toxicity, cells were exposed to organotin with NaCl at pH 5.5. Loss in cell viability following organotin exposure was enhanced considerably in NaCl (Figure 8). The effect of NaCl itself, at concentrations up to 0.5 M, on viability was negligible (results not shown). Organotin-induced K+ release also increased with NaCl concentration (Figure 8). Increased toxicity was not related to changes in organotin uptake levels as NaCl had very little effect on uptake (results not shown). 3882

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 14, 2004

FIGURE 8. Variation in the effects of TBT on (a) cell viability and (b) K+ leakage and effects of TPT on (c) cell viability and (d) K+ leakage with NaCl concentration at pH 5.5. Cells were exposed to TOT in 0 (b), 0.01 ([), 0.05 (9), 0.25 (1), and 0.5 (4) M NaCl. Note: the scales on the left and right panels are different. The limited reports on the influence of NaCl on organotin toxicity have suggested that organotin interactions are reduced in the presence of NaCl due to one or more of three main causes: competition between Na+ and organotin for cell surface binding sites, alteration in membrane lipid composition, and decrease in organotin solubility (15, 16). In this study, these effects can be discounted. First, Na+ did not inhibit organotin uptake. Second, cells were exposed to NaCl for only a short time period and in the absence of a carbon source so any NaCl-related membrane composition changes may be discounted. This was confirmed by the absence of either DPH or TMA-DPH fluorescence anisotropy effects after exposure of cells to 0.05 and 0.5 M NaCl for 1 h (results not shown). Finally, the maximum concentrations of organotins examined were selected so as not to exceed the solubility limit predetermined in 0.5 M NaCl buffer, pH 5.5. In contrast, uptake of 500 µM TBT by Synechocystis PCC 6803, Plectonema boryanum, and Chlorella emersonii at pH 5.5 was reduced by 55-65% in 0.5 M NaCl (18). In that work, however, reduced uptake may be due to the effect of Cl- on solubility, as the cells were exposed to 10 times the maximum soluble concentration used in the present study. Here, increased toxicity was associated with variations in speciation resulting from increasing NaCl levels present in solution. Enhanced toxicity associated with the speciation of organotins, as influenced by Cl-, may be due either to the overall increased proportion of neutral species or to the difference in biological activity of the individual species. For TPT, the fraction of neutral TPT species (TPTOH and TPTCl) remained relatively constant (63-65%) as TPTCl was formed in place of TPTOH (Figure 1d), which indicates that it was the increasing fraction of chloride species that enhanced toxicity. For TBT, both fractions of total neutral and chloride species increased with NaCl concentration (Figure 1c). However, TBT was more toxic in 0.5 M NaCl, pH 5.5 (Figure 8a), where the fraction of TBTOH was reduced to 4% with 34% TBTCl, than at pH 7.5 (Figure 3d), where 95% TBTOH was present, suggesting that TBTCl was more toxic than TBTOH. Evidently, for both TBT and TPT, R3SnCl species were more toxic than R3SnOH species. The variation in TBT and TPT toxicity with speciation is consistent with that proposed for mercuric compounds. Speciation of inorganic and methylmercury, as influenced by pH and salinity, governs toxicity in the diatom Thalassiosira weissflogii, with toxicity

toxicity is of primary significance. Furthermore, this work underscores the importance of the biological activity of the individual species as determined by the immediate environment as opposed to the total concentration of the contaminants. Understanding of the composition of the external environment and its influence on speciation is key in assessing organotin pollution.

Literature Cited

FIGURE 9. Relationship between (a) TBT and (b) TPT toxicity and Dow values. TBT toxicity was determined as IC50 and TPT as IC20, corresponding to the concentration of organotin causing 50% and 20% inhibition in cell viability, respectively. (O) Data from pH results (Figures 4 and 5); (b) data from NaCl results (Figure 8). The Dow values were determined at the corresponding experimental pH and salinity values. increasing with the fraction of neutral chloride species (HgCl2 or CH3HgCl) present in solution (34). As organotin pollution occurs in environments of varying salinity, including freshwater, seawater, and estuarine water and sediments (1, 35, 36), the increased toxicity associated with Cl- will have consequences for the assessment of fate and threat of organotin contamination in these ecosystems. Relationship between Organotin Toxicity and Dow. The Kow values of TBTCl (4.76) and TPTCl (4.19) are higher than those of TBTOH and TPTOH, 4.10 and 3.53, respectively (7). Consequently, differences in organotin toxicity may be interpreted not only in terms of speciation but also in relation to the overall lipophilicity of the combination of species expressed as log Dow. To further elucidate this relationship, TBT and TPT concentrations resulting in 50% (IC50) and 20% (IC20) inhibition of cell viability, respectively, were calculated from Figures 4, 5, and 8 and plotted versus log Dow (Figure 9). Two inhibitory concentrations were selected as TPT was less soluble and, under some experimental conditions, 50% inhibition of cell viability was not reached. For TBT, IC20 was not suitable as the concentration resulting in 20% inhibition of cell viability (i.e., 80% viability) could not be accurately determined from Figure 8. A linear correlation, comprising eight different experimental conditions for both organotins, was apparent (r2 ) 0.807 for TBT and 0.996 for TPT). This confirms the relationship between toxicity and the overall lipophilicity of the species present in solution. For both compounds there was a 30-fold difference in IC values between the highest and lowest Dow. The correlation of toxicity with lipophilicity indicates that TBT and TPT toxic effects are dependent on dissolution in cell membranes. Previously, lipophilicity, as reflected by Kow values of structurally distinct di- and trisubstituted organotins, was shown to correlate with toxicity toward fish cell cultures (37). Other physicochemical parameters such as total molecular surface area and steric characteristics have been related to toxicity toward aquatic organisms including algae and bacteria (38, 39) and mud crabs (40). However, this is the first time that the actual Dow of an individual compound as determined by the fraction of species present in solution has been directly correlated with toxicity. Organotins are usually present at low concentrations in polluted environments (8, 35, 41) and often at levels well below solubility limits. On the basis of the present findings, it is clear that speciation and lipophilicity, which in turn are determined by pH and Cl- concentrations, determine the interactions and consequent microbial toxicity of TBT and TPT in solution. Clearly, the influence of pH and NaCl on

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39)

Alzieu, C. Ocean Coastal Manage. 1998, 40, 23-36. Fent, K. Crit. Rev. Toxicol. 1996, 26, 1-117. Hoch, M. Appl. Geochem. 2001, 16, 719-743. Fent, K.; Hunn, J. Environ. Sci. Technol. 1991, 25, 956-963. Schebeck, L.; Andre`ae, M. O.; Tobschall, H. J. Environ. Sci. Technol. 1991, 25, 871-878. Loch, J. P. G.; Greve, P. A.; van der Berg, S. Water, Air, Soil Pollut. 1990, 53, 119-129. Fent, K.; Mu ¨ ller, M. Environ. Sci. Technol. 1991, 25, 489-483. Fent, K. Sci. Tot. Environ. 1996, 185, 151-159. Inaba, K.; Shiraishi, H.; Soma, Y. Water Res. 1995, 29, 14151417. Arnold, C. G.; Weidenhaupt, A.; David, M. M.; Mu ¨ ller, S. R.; Haderlein, S. B.; Schwarzenbach, R. P. Environ. Sci. Technol. 1997, 31, 2596-2602. Hunziker, R. W.; Escher, B. I.; Schwarzenbach, R. P. Environ. Sci. Technol. 2001, 35, 3899-3904. Fent, K.; Looser, P. W. Water Res. 1995, 29, 1631-1637. Looser, P. W.; Bertschi, S.; Fent, K. Appl. Organomet. Chem. 1998, 12, 601-611. Tsuda, T.; Aoki, S.; Kojima, M.; Harada, H. Comp. Biochem. Physiol. 1990, 95C, 151-153. Cooney, J. J.; de Rome, L,; Laurence, O.; Gadd, G. M. J. Ind. Microbiol. 1989, 4, 279-288. Laurence, O. S.; Cooney, J. J.; Gadd, G. M. Microb. Ecol. 1989, 17, 275-285. Gadd, G. M.; Gray, D. J.; Newby, P. J. Appl. Microbiol. Biotechnol. 1990, 34, 116-121. Avery, S. V.; Codd, G. A.; Gadd, G. M. Appl. Microbiol. Biotechnol. 1993, 39, 812-817. Turner, W. E.; Ahearn, D. G. Recent Trends in Yeast Research; Georgia State University: Atlanta, GA, 1978; Vol. 31, pp 113123. Tobin, J. M.; Cooney, J. J. Environ. Contam. Toxicol. 1999, 36, 7-12. White, J. S.; Tobin, J. M. Appl. Microbiol. Biotechnol. 2004, 63, 445-451. Ambrosini, A.; Bertoli, E.; Zolese, G.; Tanfani, F. Chem. Phys. Lipids 1991, 58, 73-90. Ambrosini, A.; Bertoli, E.; Zolese, G. Appl. Organomet. Chem. 1996, 10, 53-59. Shinitsky, M.; Barenholz, Y. Biochim. Biophys. Acta 1978, 515, 367-394. Blackwell, K. J.; Singleton, I.; Tobin, J. M. Appl. Microbiol. Biotechnol. 1995, 43, 579-584. Kuhry, J.-G.; Fonteneau, P.; Duportail, G.; Maechling, C.; Laustriat, G. Cell Biophys. 1983, 5, 129-140. Ro´z˘ ycka-Roszak, B.; Pruchnik, H.; Kaminski, E. Appl. Organomet. Chem. 2000, 14, 465-472. Swan, T. M.; Watson, K. Can. J. Microbiol. 1997, 43, 70-77. Blasiak, J. Comp. Biochem. Physiol. 1995, 110c, 15-21. Antunes-Madeira, M. C.; Videira, R.; Lopes, V.; Madeira, V. M. C. Med. Sci. Res. 1996, 24, 753-756. Phinney, J. T.; Bruland, K. W. Environ. Sci. Technol. 1994, 28, 1781-1790. Langner, M.; Gabrielska, J.; Pruchnik, H. Appl. Organomet. Chem. 2000, 14, 25-33. Hunziker, R. W.; Escher, B. I.; Shwarzenbach, R. P. Environ. Toxicol. Chem. 2002, 21, 1191-1197. Mason, R. P.; Reinfelder, J. R.; Morel, F. M. M. Environ. Sci. Technol. 1996, 30, 1835-1845. Fent, K.; Hunn, J. Environ. Toxicol. Chem. 1995, 14, 1123-1132. Michel, P.; Averty, B. Mar. Pollut. Bull. 1999, 38, 268-275. Bruschweiler, B. J.; Wu ¨ rler, F. E.; Fent, K. Aquat. Toxicol. 1995, 32, 143-160. Wong, P. T. S.; Chau, Y. K.; Kramar, O.; Bengert, G. A. Can. J. Fish. Aquat. Sci. 1982, 39, 2793-2789. Eng, G.; Tierney, E. J.; Olson, G. J.; Brinckman, F. E.; Bellama, J. M. Appl. Organomet. Chem. 1991, 5, 33-37.

VOL. 38, NO. 14, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3883

(40) Laughlin, R. B., Jr.; Johanessen, R. B.; French, W.; Guard, H. E.; Brinckman, F. E. Environ. Toxicol. Chem. 1985, 4, 343-351.

Received for review July 25, 2003. Revised manuscript received March 8, 2004. Accepted May 3, 2004.

(41) Evans, S. M. Mar. Pollut. Bull. 1995, 30, 14-21.

ES030099K

3884

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 14, 2004