Search

Ecological risk assessment can be enhanced with predictive models for metal toxicity. Modelings of published data were done under the simplifying assumption that intermetal trends in toxicity reflect relative metal-ligand complex stabilities. This idea has been invoked successfully since 1904 but has yet to be applied widely in quantitative ecotoxicology. Intermetal trends in toxicity were successfully modeled with ion characteristics reflecting metal binding to ligands for a wide range of effects. Most models were useful for predictive purposes based on an F-ratio criterion and cross-validation, but anomalous predictions did occur if speciation was ignored. In general, models for metals with the same valence (i.e., divalent metals) were better than those combining mono-, di-, and trivalent metals. The softness parameter (sigma p) and the absolute value of the log of the first hydrolysis constant ([symbol: see text] log KOH [symbol: see text]) were especially useful in model construction. Also, delta E0 contributed substantially to several of the two-variable models. In contrast, quantitative attempts to predict metal interactions in binary mixtures based on metal-ligand complex stabilities were not successful.

Predictive models for relative toxicity of divalent metal ions using ion characteristics have been produced with both Microtox®, a 15 min microbial bioassay, and the 24 h Caenorhabditis elegans bioassay. Relative toxicity of mono-, di- and trivalent metal ions has also been successfully modeled using ion characteristics with the Microtox® bioassay. This study extends this approach to include longer exposure durations (24 h) and a more complex organism (metazoan). Twenty-four-hour LC50s (expressed as total and free ion concentrations) for the free-living soil nematode, C. elegans , were determined for Li, Na, Mg, K, Ca, Cr, Mn, Fe, Co, Ni, Cu, Zn, Sr, Cd, Cs, Ba, La, and Pb in an aqueous medium. Relative metal toxicity was predicted with least squares linear regression and several ion characteristics. Toxicity was most effectively predicted ( r 2 =0.85) with a two-variable model containing |log K OH | (where K OH is the first hydrolysis constant) and χ 2 m r (the covalent index). The first hydrolysis constant reflects a metal ion's tendency to bind to intermediate ligands such as biochemical groups with O donor atoms, while χ 2 m r reflects binding to soft ligands such as those with S donor atoms. The use of LC50s based on free ion concentrations did not significantly improve model fit. The results of this study are consistent with earlier models generated with Microtox® data, with the exception of barium, which was much more toxic to C. elegans than would be predicted from the model. We conclude that, with thoughtful application, ion characteristics can be used to predict the relative toxicity of metal ions that vary widely in both valence and binding tendency.

Quantitative Structure Activity Relationships (QSAR) predict relative toxicity of a family of chemicals from fundamental and surrogate molecular qualities. Most QSARs are developed for organic toxicants, with inorganic toxicants (metals) being under-represented. Successful predictive models for relative toxicity of divalent metal ions using ion characteristics have been produced using Microtox®, a 15 min microbial bioassay. The present study extends this approach to longer exposure durations (24 h), and a more complex organism (metazoan). Twenty-four hour LC50s (expressed as total metal concentration) for the free-living soil nematode, C. elegans were determined for Ca, Cd, Cu, Hg, Mg, Mn, Ni, Pb, and Zn in an aqueous medium. Relative metal toxicity was predicted with least squares linear regression and several ion characteristics. Toxicity was most effectively predicted (r2 = 0.89) with ¦ log K OH ∣ (where KOH is the first hydrolysis constant), which reflects a metal ion's tendency to bind to intermediate ligands such as biochemical functional groups with O donor atoms. The best fitting model was obtained using LC50 metameters based on total metal concentration, indicating that the identification of the bioactive species of metals can be ambiguous, and does not necessarily aid in the prediction of relative metal toxicity with ion characteristics. The modelling of relative metal toxicity using ion characteristics was successful for 24 h exposure durations using this more complex organism.

Quantitative structure-activity relationships have been used to predict the relative toxicity of organic compounds. Although not as common, ion characteristics have also proven useful for predicting the relative toxicity of metal ions. The purpose of this study was to determine if the relative toxicity of metal ions using the Microtox® bioassay was predictable using ion characteristics. Median effect concentrations (EC50s) were determined for 20 metals in a NaNO 3 medium, which reflected freshwater speciation conditions, using the Microtox bacterial assay. The log of EC50 values was modeled using several ion characteristics, and Akaike's Information Criterion was calculated to determine which ion characteristics provided the best fit. Whether modeling total ion (unspeciated) or free ion (speciated) EC50 values, the one variable which best modeled EC50s was the softness index (σ p , i.e., [coordinate bond energy of the metal fluoride - coordinate bond energy of the metal iodide]/[coordinate bond energy of the metal fluoride]), while a combination of χ m 2 r (χ m = electronegativity, r = Pauling ionic radius) and |log K OK | (absolute value of the log of the first hydrolysis constant, K OH for M n+ + H 2 O → MOH n-1 + H + ) was the best two-variable model. Other variables, including ΔE o and χ m 2 r (one-variable models) and (AN/ΔIP, ΔE o ) and (χ m 2 r, Z 2 /r) (two-variable models), also gave adequate fits. Modeling with speciated (free ion) versus unspeciated (total ion) EC50 values did not improve fits. Modeling mono-, di-, and trivalent metal ions separately improved the models. We conclude that ion characteristics can be used to predict the relative toxicity of metal ions whether in freshwater (NaNO 3 medium) or saltwater (NaCl medium) speciation conditions and that this approach can be applied to metal ions varying widely in both valence and binding tendencies.

Both relative toxicity and interactions between paired metal ions were predicted with least-squares linear regression and various ion characteristics. Microtox{reg_sign} 15 min EC50s (expressed as free ion) for Ca(II), Cd(II), Cu(II), Hg(II), Mg(II), Mn(II), Ni(II), Pb(II), and Zn(II) were most effectively modeled with the constant for the first hydrolysis (K{sub H} for M{sup n+} + H{sub 2}O {yields} MOH{sup a{minus}1} + H{sup +}) although other ion characteristics were also significant in regression models. The {vert_bar}log K{sub H}{vert_bar} is correlated with metal ion affinity to intermediate ligands such as many biochemical functional groups with O donor atoms. Further, ordination of metals according to ion characteristics, e.g., {vert_bar}log K{sub H}{vert_bar} facilitated prediction of paired metal interactions. Pairing metals with strong tendencies to complex with intermediate or soft ligands such as those with O or S donor atoms resulted in strong interactions.

Anthracene, a polycyclic aromatic hydrocarbon, is acutely toxic to fish when they are exposed simultaneously to solar ultraviolet radiation (SUVR). However, the physiological mechanism of acute anthracene photo-induced toxicity is not known. The purpose of this study was to investigate possible modes of action associated with simultaneous anthracene and SUVR exposure using blood and gill biochemistry in bluegill sunfish (Lepomis macrochirus). Fish were exposed to anthracene (< 0.01 and 7.04 μg/l) and SUVR in a flow-through system. Following 96 h of anthracene and SUVR exposure, fish exhibited significant increases in hematocrit and significant decreases in whole blood hemoglobin content when compared to no-anthracene controls. Evidence of hemolysis was also observed in anthracene and SUVR exposed fish. Using in vitro enzyme analysis, both Na,K-ATPase and Mg-ATPase were signficantly inhibited in gill tissue homogenates exposed to anthracene and SUVR when compared to homogenates exposed to SUVR alone. These blood and gill measurements provide evidence of osmotic stress in exposed fish. Given the results presented in this and in previous studies, we conclude that there are numerous sites of acute toxic action with anthracene and SUVR exposure in fish and that this toxicity appears to be associated with a general disruption of cell membrane function.

Polycyclic aromatic hydrocarbons (PAHs) are widespread pollutants in the aquatic environment and demonstrate a wide variety of toxic effects in freshwater and marine organisms. We have previously documented photo-induced toxic effects of a model PAH (i.e. anthracene) in fish. The goal of our ongoing studies is to examine environmental factors which may enhance or mitigate this toxicity in order to better estimate the risks of phototoxic PAHs to aquatic organisms. The purpose of the present study was to determine the effect of water temperature and oxygen concentration on the photo-induced toxicity of anthracene to juvenile bluegill sunfish (Lepomis macrochirus). Bluegill sunfish were exposed to anthracene ( μg l ) and solar ultraviolet radiation (SUVR) at different water temperatures (20 and 30°C) and oxygen concentrations (5.0, 6.9 and 8.1 mg O2/1 for 120 h in a laboratory flow-through system. A trend toward an inverse relationship was found between water temperature and LC50 values, although no significant temperature effect was observed. A nonlinear relationship was found between oxygen concentration and LC50 values, with significantly increased toxicity at the middle oxygen concentration. Opercular ventilation rates were inversely related to oxygen concentration and directly related to anthracene concentration. A statistical relationship was developed on the basis of oxygen concentration to predict LC50 values. We conclude that dissolved oxygen concentration is an important environmental factor in the assessment of photo-induced toxicity of anthracene to fish.

Predominant methods for analyzing dose- or concentration-effect data (i.e., probit analysis) are based on the concept of individual tolerance or individual effective dose (IED, the smallest characteristic dose needed to kill an individual). An alternative explanation (stochasticity hypothesis) is that individuals do not have unique tolerances: death results from stochastic processes occurring similarly in all individuals. These opposing hypotheses were tested with two types of experiments. First, time to stupefaction (TTS) was measured for zebra fish (Brachydanio rerio) exposed to benzocaine. The same 40 fish were exposed during five trials to test if the same order for TTS was maintained among trials. The IED hypothesis was supported with a minor stochastic component being present. Second, eastern mosquitofish (Gambusia holbrooki) were exposed to sublethal or lethal NaCl concentrations until a large portion of the lethally exposed fish died. After sufficient time for recovery, fish sublethally exposed and fish surviving lethal exposure were exposed simultaneously to lethal NaCl concentrations. No statistically significant effect was found of previous exposure on survival time but a large stochastic component to the survival dynamics was obvious. Repetition of this second type of test with pentachlorophenol also provided no support for the IED hypothesis. We conclude that neither hypothesis alone was the sole or dominant explanation for the lognormal (probit) model. Determination of the correct explanation (IED or stochastic) or the relative contributions of each is crucial to predicting consequences to populations after repeated or chronic exposures to any particular toxicant.

In classical pharmacology, oral bioavailability of a toxicant is defined as that fraction of an orally administered dose reaching the systemic circulation of the animal. The present study estimates the bioavailability of methylmercury in channel catfish (Ictalurus punctatus) by comparing concentrations in the blood through time after oral and intra-arterial (IA) administration. Catfish were cannulated in the dorsal aorta and gavaged a pelleted feed that had been spiked with methylmercury. Each catfish was gavaged an increasing amount of spiked feed. Following oral dosing, serial blood samples were removed for more than 1,500 h. One month after removal of the last blood sample, the same fish were injected IA with methylmercury and serial blood samples were removed for more than 3,000 h. The area under the curve of the blood concentration-time curve extrapolated to infinity (AUC0→∞) was calculated from fish dosed orally and IA using both noncompartmental (trapezoidal) and compartmental methods. Bioavailability was estimated as the ratio of the dose-corrected oral AUC0→∞ to the IA AUC0→∞. Average bioavailability estimates from this approach were 33% using noncompartmental (range 14–55%) and 29% using compartmental (range 12–42%) methods and were correlated with the amount of food gavaged to the fish (r2 = 0.95, p = 0.026). Bioavailability estimates using the present methods were much lower than estimates using more conventional methods (i.e., assimilation efficiency estimates using mass balance), suggesting that conventional methods may overestimate the true bioavailability of toxicants in fish.