Developmental toxicity of Louisiana crude oil-spiked sediment to zebrafish.

Embryonic exposures to the components of petroleum, including polycyclic aromatic hydrocarbons (PAHs), cause a characteristic suite of developmental defects and cardiotoxicity in a variety of fish species. We exposed zebrafish embryos to reference sediment mixed with laboratory weathered South Louisiana crude oil and to sediment collected from an oiled site in Barataria Bay, Louisiana in December 2010. Laboratory oiled sediment exposures caused a reproducible set of developmental malformations in zebrafish embryos including yolk sac and pericardial edema, craniofacial and spinal defects, and tissue degeneration. Dose-response studies with spiked sediment showed that total polycyclic aromatic hydrocarbons (tPAH) concentrations of 27mg tPAH/kg (dry weight normalized to 1 percent organic carbon [1 percent OC]) caused a significant increase in defects, and concentrations above 78mg tPAH/kg 1 percent OC caused nearly complete embryo mortality. No toxicity was observed in Barataria sediment with 2mg tPAH/kg 1 percent OC. Laboratory aging of spiked sediment at 4°C resulted in a nearly 10-fold decrease in sensitivity over a 40-day period. This study demonstrates oiled sediment as an exposure pathway to fish with dose-dependent effects on embryogenesis that are consistent with PAH mechanisms of developmental toxicity. The results have implications for effects on estuarine fish from oiled coastal areas during the Deepwater Horizon spill.

Prediction of aquatic toxicity mode of action using linear discriminant and random forest models.

The ability to determine the mode of action (MOA) for a diverse group of chemicals is a critical part of ecological risk assessment and chemical regulation. However, existing MOA assignment approaches in ecotoxicology have been limited to a relatively few MOAs, have high uncertainty, or rely on professional judgment. In this study, machine based learning algorithms (linear discriminant analysis and random forest) were used to develop models for assigning aquatic toxicity MOA. These methods were selected since they have been shown to be able to correlate diverse data sets and provide an indication of the most important descriptors. A data set of MOA assignments for 924 chemicals was developed using a combination of high confidence assignments, international consensus classifications, ASTER (ASessment Tools for the Evaluation of Risk) predictions, and weight of evidence professional judgment based an assessment of structure and literature information. The overall data set was randomly divided into a training set (75%) and a validation set (25%) and then used to develop linear discriminant analysis (LDA) and random forest (RF) MOA assignment models. The LDA and RF models had high internal concordance and specificity and were able to produce overall prediction accuracies ranging from 84.5 to 87.7% for the validation set. These results demonstrate that computational chemistry approaches can be used to determine the acute toxicity MOAs across a large range of structures and mechanisms.

Influence of taxonomic relatedness and chemical mode of action in acute interspecies estimation models for aquatic species.

Ecological risks to aquatic organisms are typically assessed using acute toxicity data for relatively few species and with limited understanding of relative species sensitivity. We developed a comprehensive set of interspecies correlation estimation (ICE) models based on acute toxicity data for aquatic organisms and evaluated three key sources of model uncertainty: taxonomic relatedness, chemical mode of action (MOA), and model parameters. Models are least-squares regressions of acute toxicity of surrogate and predicted species. A total of 780 models were derived from acute values for 77 species of aquatic organisms and over 550 chemicals. Cross-validation of models showed that accurate model prediction was greatest for models with surrogate and predicted taxa within the same family (91% of predictions within 5-fold of measured values). Recursive partitioning provided user guidance for selection of robust models using model mean square error and taxonomic relatedness. Models built with a single MOA were more robust than models built using toxicity values with multiple MOAs, and improve predictions among species pairs with large taxonomic distance (e.g., within phylum). These results indicate that between-species toxicity extrapolation can be improved using MOA-based models for less related taxa pairs and for those specific MOAs.

Augmenting aquatic species sensitivity distributions with interspecies toxicity estimation models.

Species sensitivity distributions (SSDs) are cumulative distribution functions of species toxicity values. The SSD approach is being used increasingly in ecological risk assessment but is often limited by available toxicity data needed for diverse species representation. In the present study, the authors evaluate augmenting aquatic species databases limited to standard test species using toxicity values extrapolated from interspecies correlation estimation (ICE) models for SSD development. The authors compared hazard concentrations at the 5th centile (HC5) of SSDs developed using limited measured data augmented with ICE toxicity values (augmented SSDs) with those estimated using larger measured toxicity datasets of diverse species (reference SSDs). When SSDs had similar species composition to the reference SSDs, 0.76 of the HC5 estimates were closer to the reference HC5; however, the proportion of augmented HC5s that were within 5-fold of the reference HC5s was 0.94, compared with 0.96 when predicted SSDs had random species assemblages. The range of toxicity values among represented species in all SSDs also depended on a chemical's mode of action. Predicted HC5 estimations for acetylcholinesterase inhibitors showed the greatest discrepancies from the reference HC5 when SSDs were limited to commonly tested species. The results of the present study indicate that ICE models used to augment datasets for SSDs do not greatly affect HC5 uncertainty. Uncertainty analysis of risk assessments using SSD hazard concentrations should address species composition, especially for chemicals with known taxa-specific differences in toxicological effects. This article is a US Government work and is in the public domain in the USA.

Development of aquatic toxicity benchmarks for oil products using species sensitivity distributions.

Determining the sensitivity of a diversity of species to spilled oil and chemically dispersed oil continues to be a significant challenge in spill response and impact assessment. We used standardized tests from the literature to develop species sensitivity distributions (SSDs) of acute aquatic toxicity values for several petroleum products and 2 Corexit oil dispersants. Fifth percentile hazard concentrations (HC5s) were computed from the SSDs and used to assess relative oil product toxicity and in evaluating the feasibility of establishing toxicity benchmarks for a community of species. The sensitivity of mysids (Americamysis bahia) and silversides (Menidia beryllina) were evaluated within the SSDs to determine if these common test species were appropriate surrogates for a broader range of species. In general, SSD development was limited by the availability of acute toxicity values that met standardization criteria for a diversity of species. Pooled SSDs were also developed for crude oil and Corexit dispersants because there was only small variability in the HC5s among the individual oil or dispersant products. The sensitivity of mysids and silversides varied across the oil and dispersant products, with the majority of toxicity values greater than the HC5. Application of SSDs appears to be a reasonable approach to developing oil product toxicity benchmarks, but additional toxicity data are needed for a larger range of species conducted under standardized test conditions.

Evaluation of in silico development of aquatic toxicity species sensitivity distributions.

Determining the sensitivity of a diversity of species to environmental contaminants continues to be a significant challenge in ecological risk assessment because toxicity data are generally limited to a few standard test species. This study assessed whether species sensitivity distributions (SSDs) could be generated with reasonable accuracy using only in silico modeling of toxicity to aquatic organisms. Ten chemicals were selected for evaluation that spanned several modes of actions and chemical classes. Median lethal concentrations (LC50s) were estimated using three internet-based quantitative structure activity relationship (QSAR) tools that employ different computational approaches: ECOSAR (Ecological Structure Activity Relationships), ASTER (Assessment Tools for the Evaluation of Risk), and TEST (Toxicity Estimation Software Tool). Each QSAR estimate was then used as input into the SSD module of the internet-based toxicity estimation program Web-ICE to generate an in silico estimated fifth percentile hazard concentration (HC5) for each of the ten chemicals. The accuracy of the estimated HC5s was determined by comparison to measured HC5s developed from an independent dataset of experimental acute toxicity values for a diversity of aquatic species. Estimated HC5s showed generally poor agreement with measured HC5s determined for all available aquatic species, but showed better agreement when species composition of the chemical specific SSDs were identical. These results indicated that LC50 variability and species composition were large sources of error in estimated HC5s. Additional research is needed to reduce uncertainty in HC5s using only in silico approaches and to develop computational approaches for predicting species sensitivity.