QSARs for aquatic toxicity: celebrating, extending and displaying the pioneering contributions of Ferguson, Konemann and Veith

Determining the toxicity of organic compounds to the nematode Caenorhabditis elegans based on aqueous concentrations

Caenorhabditis elegans is used for assessing the toxicity of chemicals in aqueous medium. However, chemicals can absorb to the bacterial food, which reduces the freely dissolved concentrations of the tested compounds. Thus, based on total or nominal concentrations, toxicity is underestimated, resulting in misleading assumptions on toxicity mechanisms or comparisons to other test organisms. As the verification of freely dissolved exposure concentrations (Cfree) is challenging in small test systems, simple partitioning models might by a good option for estimating Cfree. Therefore, C. elegans was exposed to seven differently acting organic chemicals with varying hydrophobicities, thus also different affinities to bind to the food of C. elegans. Measured concentrations of the dissolved aqueous and the bacterial-bound fraction allowed the calculation of binding constants (Kb). Experimental Kb were comparable to literature data of hydrophobic chemicals and correlated well with their hydrophobicity, expressed as log KOW. The chronic toxicity of the various compounds on C. elegans' reproduction, based on their aqueous concentration, was weakly related to their log KOW. Toxicity expressed based on chemical activity and comparisons with a baseline toxicity model, nevertheless, suggested a narcotic mode of action for most hydrophobic compounds (except methylisothiazolinone and trichlorocarbanilide). Although revealing a similar toxicity ranking than Daphnia magna, C. elegans was less sensitive, probably due to its ability to reduce its internal concentrations by means of its very impermeable cuticle or by efficient detoxification mechanisms. It could be shown that measured aqueous concentrations in the nematode test system corresponded well with freely dissolved concentrations that were modeled using simple mass-balance models from nominal concentrations. This offers the possibility to estimate freely dissolved concentrations of chemicals from nominal concentrations, making routine testing of chemicals and their comparison to other species more accurate.

Leveraging ToxCast data and protein sequence conservation to complement aquatic life criteria derivation

The USEPA's 1985 guidelines for the derivation of aquatic life criteria (ALC) are robust but data-intensive. For many chemicals, the extensive in vivo data sets required for ALC derivation are not available. Thus, alternative analyses and processes that can provide provisional values to guide states, tribes, and other stakeholders while data accumulate and more rigorous criteria are derived would be beneficial. The overarching purpose of this study was to assess the feasibility of using data from new approach methodologies (NAMs) like ToxCast to derive first-pass, provisional values to guide chemical prioritization and resource management as a complement to traditional ALC derivation. To address this goal, the study objectives were to (1) estimate chemical potency using data from NAMs for nine compounds with available aquatic benchmarks, (2) evaluate the utility of using NAM data to elucidate potential mechanisms of toxicity to guide problem formulation, and (3) determine the species relevance of toxicity pathways for compounds with clearly defined mechanisms of action as a means to evaluate whether minimum data requirements could potentially be waived when deriving a more formal ALC. Points of departure were derived from ToxCast data based on the fifth percentile of the distribution of activity concentration above cutoff values falling below the cytotoxic burst. Mechanistic inferences were made based on active target hits in ToxCast and, where applicable, assessed for taxonomic conservation using SeqAPASS. ToxCast-based point-of-departure aligned relatively closely (six of nine test chemicals within a factor of 10; eight of nine within a factor of 100) with aquatic benchmarks from the USEPA and US Department of Energy (DOE). Moreover, pathways of toxicity gleaned from NAM data were reflective of in vivo-based findings from the literature. These results, while preliminary, and based on a limited number of substances, support the potential application of NAM data to complement traditional ALC derivation approaches and prioritization. Integr Environ Assess Manag 2023;19:224-238. © 2022 Society of Environmental Toxicology & Chemistry (SETAC). This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.

Ethylene oxide derived glycol ethers: A review of the alkyl glycol ethers potential to cause endocrine disruption

The 'ethylene glycol ethers' (EGE) are a broad family of solvents and hydraulic fluids produced through the reaction of ethylene oxide and a monoalcohol. Certain EGE derived from methanol and ethanol are well known to cause toxicity to the testes and fetotoxicity and that this is caused by the common metabolites methoxy and ethoxyacetic acid, respectively. There have been numerous published claims that EGE fall into the category of 'endocrine disruptors' often without substantiated evidence. This review systematically evaluates all of the available and relevant in vitro and in vivo data across this family of substances using an approach based around the EFSA/ECHA 2018 guidance for the identification of endocrine disruptors. The conclusion reached is that there is no significant evidence to show that EGE target any endocrine organs or perturb endocrine pathways and that any toxicity that is seen occurs by non-endocrine modes of action.

Assessing the Toxicity of Individual Aromatic Compounds and Mixtures to American Lobster (Homarus americanus) Larvae Using a Passive Dosing System

Aquatic exposures to aromatic compounds (ACs) may be important contributors to biological effects of oil spills. The present study examined the acute toxicity of 11 ACs and 3 binary AC mixtures on stage 1 American lobster larvae using a passive dosing test design. The ACs investigated covered a range of classes and log octanol-water partition coefficient values (K_OW ; 2.5-5.5). Silicone O-rings were used to partition ACs into seawater and maintain stable exposures. Exposed lobster larvae were assessed for mobility and survival at 3, 6, 12, 24, 36, and 48 h. Fluorometry and gas chromatography-mass spectrometry measurements confirmed well-defined substance exposures. Expressing lethality in terms of chemical activities yielded values between 0.01 and 0.1, consistent with a baseline mode of action. Analysis of time-dependent median lethal/effect concentration (L/EC50) values were used to determine incipient values. An expected linear relationship between the incipient log L/EC50 and log K_OW was fit to the empirical toxicity data to derive critical target lipid body burdens for immobilization and lethality endpoints. These values indicate that American lobster larvae fall on the sensitive end of the acute species sensitivity distribution. We used AC toxicity data to successfully predict toxicity of binary mixtures assuming additive toxicity. The observed time-dependent toxicity was inversely related to log K_OW and occurred more quickly than reported previously. The results contribute to improving models for predicting oil spill impacts on American lobster larvae populations. Environ Toxicol Chem 2021;40:1379-1388. © 2021 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

A general linear free energy relationship for predicting partition coefficients of neutral organic compounds

Predicting the effects of organic compounds on environments and biological systems is an important objective for environmental chemistry and human health. The logarithm (to base 10) of the n-octanoll-water partition coefficient has been widely used to predict the mentioned properties. However, the suitability of this parameter for the purpose has been questioned, since the environments relating to the properties may be quite different from that of bulk n-octanol. In this study, we used a theoretical derivation approach to develop a model for predicting the partition coefficients of solutes between water and an organic solvent that may be similar to n-octanol or quite different from it. Our model relies on solute descriptors that can be calculated based on solute structures. It was used to predict the n-octanoll-water, hexadecanel-water and chloroforml-water partition coefficients of solutes. The calculated values of the examined parameters agreed well with their experimental counterparts. The model can find application in the accurate prediction of the effects of organic compounds on environments and the physicochemical properties of organic compounds by a full in-silico approach and can provide useful guidance for improving some properties of organic compounds.

A perspective on the role of fugacity and activity for evaluating the PBT properties of organic chemicals and providing a multi-media synoptic indicator of environmental contamination

The hazard and risk from organic chemicals present in the environment are routinely evaluated using P (persistence), B (bioaccumulation) and T (toxicity) criteria. We present a perspective on how models based on the equilibrium criteria of fugacity and chemical activity can contribute to all three evaluations, thus providing a consistent and seamless assessment process. Persistence and its closely related, but fundamentally different property, chemical residence time can be determined from degradation half-lives (typically obtained from monitoring data) and multi-media mass balance models describing chemical fate and transport in real or evaluative environments. Bioaccumulation is driven by equilibrium partitioning processes that can be estimated from fugacity models treating uptake by respiration and diet in single organisms and in food webs, most commonly for aquatic systems for which confirmatory monitoring data can be obtained. Biomagnification is readily evaluated both experimentally and as a fugacity (or activity) increase from prey to predator. Toxicokinetic evaluations of chemical uptake and disposition in a variety of organisms, including humans, can be obtained using fugacity- and physiologically-based pharmacokinetic models. Toxicodynamic processes of chemical interactions with organisms leading to adverse toxic outcomes are less obviously amenable to fugacity assessment with the notable exception of baseline toxicity (narcosis), the most common mode of action of commercial chemicals. It is shown that a range of fugacities can be defined and correlated for specific species thus enabling direct comparison of fugacities estimated or monitored in the environment with fugacities resulting in baseline toxicity. Finally, it is shown that fugacity and activity can serve as multi-media metrics of chemical contamination in ecosystems, thus enabling identification of specific regions and species at greatest risk in a pictorial format that is readily understandable to scientific and regulatory communities and to the general public.

Modulation of microbial growth and enzymatic activities in the marine environment due to exposure to organic contaminants of emerging concern and hydrocarbons

Organic pollutants are continuously being introduced in seawater with uncharacterized impacts on the engines of the marine biogeochemical cycles, the microorganisms. The effects on marine microbial communities were assessed for perfluoroalkyl substances, organophosphate esters flame retardants and plasticizers, polycyclic aromatic hydrocarbons, and n-alkanes. Dose-response experiments were performed at three stations and at three depths in the NW Mediterranean with contrasted nutrient and pollutant concentrations. In these experiments, the microbial growth rates, the abundances of the main bacterial groups, measured by Catalyzed Reporter Deposition Fluorescence In Situ Hybridization (CARD-FISH), and extracellular enzymatic activities, were quantified. Increasing concentrations of organic pollutants (OPs) promoted different responses in the communities that were compound, organism and nutrient availability (trophic status). The largest differences between OP treatments and controls in the growth rates of both heterotrophic and phototrophic microbial groups were observed in seawater from the deep chlorophyll maxima. Furthermore, there was a compound specific stimulation of different extracellular enzymatic activities after the exposure to OPs. Our results revealed that marine microbial communities reacted not only to hydrocarbons, known to be used as a carbon source, but also to low concentrations of organic pollutants of emerging concern in a complex manner, reflecting the variability of various environmental variables. Multiple linear regressions suggested that organic pollutants modulated the bacterial growth and extracellular enzymatic activities, but this modulation was of lower magnitude than the observed pronounced response of the microbial community to nutrient availability.

Application of an Iterative Fragment Selection (IFS) Method to Estimate Entropies of Fusion and Melting Points of Organic Chemicals

The main objective of this study is to develop and evaluate novel Quantitative Structure-Property Relationships (QSPRs) for predicting entropy of fusion (ΔS_M ) and melting point (T_M ) of organic chemicals from chemical structure. The QSPRs are developed using the Iterative Fragment Selection (IFS) method that requires only 2D structural information from the user (SMILES codes) for property prediction. The QSPRs also provide information on the applicability domain for each calculation and uncertainty estimates for the predictions. The root mean square error (RMSE) for the external validation sets are 11.8 J mol^-1  K^-1 and 46.9 K for the ΔS_M and T_M QSPRs, respectively. The performance of the new QSPRs is comparable to other predictive methods but has advantages with respect to availability and ease of use as well as the guidance on applicability domain for each prediction. Limitations of the new QSPRs are discussed. The QSPRs are coded as a user-friendly, freely available tool.

Linking algal growth inhibition to chemical activity: Excess toxicity below 0.1% of saturation

Chemical activity quantifies the energetic level of an organic compound relative to its pure liquid [0-1], and several studies have reported that baseline toxicity generally requires chemical activities of 0.01-0.1. The first aim was to challenge this chemical activity range for baseline toxicity. Algal growth inhibition data (median effective concentrations, EC₅₀) were compiled from two recent studies and included 108 compounds categorised as non-polar (mode of toxic action, MOA1) and polar (MOA2) narcotics. These data were linked to chemical activity by (1) plotting them relative to a regression for (subcooled) liquid solubility (S_L), which served as visual reference for chemical activity of unity and (2) determining EC₅₀/S_L ratios that essentially equal median effective chemical activity (Ea₅₀). Growth inhibition required chemical activity >0.01 for MOA1 and >0.001 for MOA2 compounds. The second aim was to identify compounds exerting excess toxicity, i.e., when growth inhibition occurred at chemical activity <0.001. From a recent review with 2323 data entries, 315 EC₅₀ values passed our selection criteria. 280 of these EC₅₀ values were within or near the baseline toxicity range (Ea₅₀>0.001), and 25 compounds were found to exert excess toxicity (Ea₅₀<0.001). Of these compounds, 16 are pesticides or precursors. Methodologically, this study includes two methods for translating EC₅₀ values into the chemical activity framework, each having advantages and limitations. Scientifically, this study confirms that baseline toxicity generally requires chemical activities of 0.01-0.1 and extends the application of the chemical activity approach beyond baseline toxicity, by demonstrating its utility to identify compounds that exert excess toxicity.

Chemical activity and distribution of emerging pollutants: Insights from a multi-compartment analysis of a freshwater system

Emerging pollutants are ubiquitous in the aquatic system and may pose risks to aquatic ecosystems. The quantification and prediction of environmental partitioning of these chemicals in aquatic systems between water, sediment and biota is an important step in the comprehensive assessment of their sources and final fates in the environment. In this multi-compartment field study, we applied equilibrium partitioning theory and chemical activity estimates to investigate the predictability of concentrations in Gammarus pulex as a model invertebrate from water and sediment in a typical small central European river. Furthermore, K_OW-based and LSER approaches were assessed for the calculation of sediment organic carbon-, lipid-, and protein-water partitioning coefficients and activity ratios between the different compartments. Gammarid-water activity ratios close to unity have been observed for many chemicals, while sediment-water and sediment-biota chemical activity ratios exceeded unity by up to six orders of magnitudes. Causes may be: disequilibrium due to slow desorption kinetics and/or an underestimation of partition coefficients due to the presence of strongly adsorbing phases in the sediments. Water concentrations, particularly when using LSER for prediction of partition coefficients were good predictors of internal concentrations in gammarids for most emerging pollutants. Some hydrophilic chemicals such as the neonicotinoid imidacloprid tend to accumulate more in G. pulex than expected from equilibrium partitioning. This conclusion holds both for K_OW as well as for LSER-based predictions and suggests previously unidentified mechanisms of bio-accumulation which may include binding to specific protein structures.

General baseline toxicity QSAR for nonpolar, polar and ionisable chemicals and their mixtures in the bioluminescence inhibition assay with Aliivibrio fischeri

The Microtox assay, a bioluminescence inhibition assay with the marine bacterium Aliivibrio fischeri, is one of the most popular bioassays for assessing the cytotoxicity of organic chemicals, mixtures and environmental samples. Most environmental chemicals act as baseline toxicants in this short-term screening assay, which is typically run with only 30 min of exposure duration. Numerous Quantitative Structure-Activity Relationships (QSARs) exist for the Microtox assay for nonpolar and polar narcosis. However, typical water pollutants, which have highly diverse structures covering a wide range of hydrophobicity and speciation from neutral to anionic and cationic, are often outside the applicability domain of these QSARs. To include all types of environmentally relevant organic pollutants we developed a general baseline toxicity QSAR using liposome-water distribution ratios as descriptors. Previous limitations in availability of experimental liposome-water partition constants were overcome by reliable prediction models based on polyparameter linear free energy relationships for neutral chemicals and the COSMOmic model for charged chemicals. With this QSAR and targeted mixture experiments we could demonstrate that ionisable chemicals fall in the applicability domain. Most investigated water pollutants acted as baseline toxicants in this bioassay, with the few outliers identified as uncouplers or reactive toxicants. The main limitation of the Microtox assay is that chemicals with a high melting point and/or high hydrophobicity were outside of the applicability domain because of their low water solubility. We quantitatively derived a solubility cut-off but also demonstrated with mixture experiments that chemicals inactive on their own can contribute to mixture toxicity, which is highly relevant for complex environmental mixtures, where these chemicals may be present at concentrations below the solubility cut-off.

Development of a general baseline toxicity QSAR model for the fish embryo acute toxicity test

Fish embryos have become a popular model in ecotoxicology and toxicology. The fish embryo acute toxicity test (FET) with the zebrafish embryo was recently adopted by the OECD as technical guideline TG 236 and a large database of concentrations causing 50% lethality (LC₅₀) is available in the literature. Quantitative Structure-Activity Relationships (QSARs) of baseline toxicity (also called narcosis) are helpful to estimate the minimum toxicity of chemicals to be tested and to identify excess toxicity in existing data sets. Here, we analyzed an existing fish embryo toxicity database and established a QSAR for fish embryo LC₅₀ using chemicals that were independently classified to act according to the non-specific mode of action of baseline toxicity. The octanol-water partition coefficient K_ow is commonly applied to discriminate between non-polar and polar narcotics. Replacing the K_ow by the liposome-water partition coefficient K_lipw yielded a common QSAR for polar and non-polar baseline toxicants. This developed baseline toxicity QSAR was applied to compare the final mode of action (MOA) assignment of 132 chemicals. Further, we included the analysis of internal lethal concentration (ILC₅₀) and chemical activity (La₅₀) as complementary approaches to evaluate the robustness of the FET baseline toxicity. The analysis of the FET dataset revealed that specifically acting and reactive chemicals converged towards the baseline toxicity QSAR with increasing hydrophobicity. The developed FET baseline toxicity QSAR can be used to identify specifically acting or reactive compounds by determination of the toxic ratio and in combination with appropriate endpoints to infer the MOA for chemicals.

Impact of temperature, pH, and salinity changes on the physico-chemical properties of model naphthenic acids

The effects of temperature, pH, and salinity change on naphthenic acids (NAs) present in oil-sands process wastewater were modeled for 55 representative NAs. COSMO-RS was used to estimate octanol-water (KOW) and octanol-air (KOA) partition ratios and Henry's law constants (H). Validation with experimental carboxylic acid data yielded log KOW and log H RMS errors of 0.45 and 0.55 respectively. Calculations of log KOW, (or log D, for pH-dependence), log KOA and log H (or log HD, for pH-dependence) were made for model NAs between -20 °C and 40 °C, pH between 0 and 14, and salinity between 0 and 3 g NaCl L(-1). Temperature increase by 60 °C resulted in 3-5 log unit increase in H and a similar magnitude decrease in KOA. pH increase above the NA pKa resulted in a dramatic decrease in both log D and log HD. Salinity increase over the 0-3 g NaCl L(-1) range resulted in a 0.3 log unit increase on average for KOW and H values. Log KOW values of the sodium salt and anion of the conjugate base were also estimated to examine their potential for contribution to the overall partitioning of NAs. Sodium salts and anions of naphthenic acids are predicted to have on average 4 log units and 6 log units lower log KOW values, respectively, with respect to the corresponding neutral NA. Partitioning properties are profoundly influenced by the by the relative prevailing pH and the substance's pKa at the relevant temperature.

Passive dosing versus solvent spiking for controlling and maintaining hydrophobic organic compound exposure in the Microtox® assay

Microbial toxicity bioassays such as the Microtox® test are ubiquitously applied to measure the toxicity of chemicals and environmental samples. In many ways their operation is conducive to the testing of organic chemicals. They are of short duration, use glass cuvettes and take place at reduced temperatures in medium lacking sorbing components. All of these are expected to reduce sorptive and volatile losses, but particularly for hydrophobic organics the role of such losses in determining the bioassay response remains unclear. This study determined the response of the Microtox® test when using solvent spiking compared to passive dosing for introducing the model hydrophobic compounds acenaphthene, phenanthrene, fluoranthene and benzo(a)pyrene. Compared to solvent spiking, the apparent sensitivity of the Microtox® test with passive dosing was 3.4 and 12.4 times higher for acenaphthene and phenanthrene, respectively. Furthermore, fluoranthene only gave a consistent response with passive dosing. Benzo(a)pyrene did not result in a response with either spiking or passive dosing even at aqueous solubility. Such differences in the apparent sensitivity of the Microtox® test can be traced back to the precise definition of the dissolved exposure concentrations and the buffering of losses with passive dosing. This highlights the importance of exposure control even in simple and short-term microbial bioassays such as the Microtox® test.

Linking algal growth inhibition to chemical activity: baseline toxicity required 1% of saturation

Recently, high-quality data were published on the algal growth inhibition caused by 50 non-polar narcotic compounds, of which 39 were liquid compounds with defined water solubility. In the present study, the toxicity data for these liquids were applied to challenge the chemical activity range for baseline toxicity. First, the reported effective concentrations (EC50) were divided by the respective water solubilities (S water), since the obtained EC50/S water ratio essentially equals the effective chemical activity (Ea50). The majority of EC50/S water ratios were within the expected chemical activity range of 0.01-0.1 for baseline toxicity, and none of the ratios were significantly below 0.01. On a practical level, these findings suggest EC50 values for baseline toxicity to be at or above 1% of liquid solubility, which would have been accurate or conservative for all 39 liquids with defined water solubility in the applied dataset. On an environmental risk assessment level, predicted no-effect concentrations (PNECs) for baseline toxicity could even be set as a percentage of saturation, which can easily be extended to mixtures. However, EC50 values well below 1% of liquid saturation can still occur and would be a direct indication of excess toxicity.

Application of mass balance models and the chemical activity concept to facilitate the use of in vitro toxicity data for risk assessment

Practical, financial, and ethical considerations related to conducting extensive animal testing have resulted in various initiatives to promote and expand the use of in vitro testing data for chemical evaluations. Nominal concentrations in the aqueous phase corresponding to an effect (or biological activity) are commonly reported and used to characterize toxicity (or biological response). However, the true concentration in the aqueous phase can be substantially different from the nominal. To support in vitro test design and aid the interpretation of in vitro toxicity data, we developed a mass balance model that can be parametrized and applied to represent typical in vitro test systems. The model calculates the mass distribution, freely dissolved concentrations, and cell/tissue concentrations corresponding to the initial nominal concentration and experimental conditions specified by the user. Chemical activity, a metric which can be used to assess the potential for baseline toxicity to occur, is also calculated. The model is first applied to a set of hypothetical chemicals to illustrate the degree to which test conditions (e.g., presence or absence of serum) influence the distribution of the chemical in the test system. The model is then applied to set of 1194 real substances (predominantly from the ToxCast chemical database) to calculate the potential range of concentrations and chemical activities under assumed test conditions. The model demonstrates how both concentrations and chemical activities can vary by orders of magnitude for the same nominal concentration.