Evolutionary patterns and physicochemical properties explain macroinvertebrate sensitivity to heavy metals

Area of Science:

• Ecotoxicology
• Evolutionary Biology
• Computational Biology

Background:

• Ecological risk assessment relies heavily on species sensitivity data, often obtained from laboratory toxicity bioassays.
• Performing exhaustive bioassays for all species and chemicals is practically constrained.
• Bilinear models integrating species phylogeny and chemical properties offer a predictive approach.

Purpose of the Study:

• To develop and validate bilinear models for predicting invertebrate species sensitivity to heavy metals.
• To assess the contribution of phylogenetic information versus physicochemical properties in predicting heavy metal toxicity.
• To establish a statistical framework for forecasting heavy metal toxicity in untested invertebrate species.

Main Methods:

• Combined DNA sequences from 31 invertebrate species with physicochemical properties of six bivalent metals.
• Constructed bilinear models to explain variability in species sensitivity.
• Validated model predictions against experimental toxicity values.

Main Results:

• Bilinear models explained 70-80% of the variability in species sensitivity to heavy metals.
• Phylogenetic information was the most significant factor, accounting for over 40% of the explained variance.
• Predicted values from the models showed good agreement with experimental data (over 50%).

Conclusions:

• Bilinear modeling integrating phylogenetic and chemical data is a valuable tool for predicting heavy metal toxicity in invertebrates.
• The approach provides a strong foundation for developing statistical models to forecast toxicity for unassayed species.
• Further improvements can be achieved with enhanced phylogenetic and toxicological datasets, linking evolutionary biology with ecotoxicology.