Research Areas > Contaminants > Molecular Tools for Toxicity Identification Evaluation

Project: Molecular Tools for Toxicity Identification Evaluation

Background and Objectives

The toxicity identification evaluation (TIE) process uses a variety of chemical/physical treatments coupled with toxicity testing to identify which compounds may be causing toxicity in water or sediment samples. TIEs have generally been successful in differentiating broad classes of toxicants in a sample, but identification of specific contaminants and characterization of the effects of contaminant mixtures are more difficult to achieve, often requiring extensive chemical fractionation and analysis. Conducting TIEs with sediments is even more challenging because fewer treatment methods are available, the methods are less reliable for samples with low levels of toxicity, and the use and interpretation of chemical fractionation methods are limited by the effect of sediment partitioning on contaminant bioavailability. For these reasons, sediment TIEs are used infrequently, forcing environmental managers to rely on incomplete or inaccurate information in determining which constituents are responsible for impaired sediment quality.

Recent advances in molecular biotechnology have resulted in new methods to measure the response of organisms to toxicants. These could result in more effective TIE methods. Gene microarrays are now available that can simultaneously measure changes in the expression of thousands of genes in aquatic organisms. Research has demonstrated that aquatic organisms produce distinctive patterns of gene expression in response to different types of contaminant stress, and that these patterns can be used in a diagnostic manner to investigate the cause of toxicity. Combining these molecular techniques with traditional toxicity test methods may enable the development of a molecular TIE approach, which could potentially greatly enhance the success and applicability of sediment TIEs. Still, the accuracy and reliability of such a molecular TIE method is unknown.

The goal of this project is to develop and evaluate a new suite of TIE tools based on genomics (e.g., analysis of gene expression or protein production).

Status

This project was initiated in 2009 with anticipated completion in 2014.

Methods

SCCWRP is developing a molecular TIE approach for the marine amphipod Eohastorius estuarius, which is frequently used for sediment toxicity testing in California monitoring programs. The RNA from amphipods exposed to a variety of contaminant types will be extracted and sequenced in order to develop a gene microarray that contains targets for several thousand different genes. This microarray will then be calibrated by testing amphipods exposed to different types of contaminants, in order to identify contaminant-specific patterns of gene expression. A series of evaluation experiments will also be conducted, including analyses of gene expression in amphipods exposed to different combinations and doses of sediment-borne contaminants and potential confounding factors (e.g., salinity, temperature). The results will be used to evaluate the performance of a molecular TIE method in comparison to a traditional sediment TIE approach.
 

Findings

To date, this project has achieved several important milestones in the effort to develop a molecular TIE method for evaluating sediment toxicity. First, RNA from E. estuarius was sequenced for the first time, and a gene microarray was successfully developed. The microarray was used to identify a subset of candidate genes that responded differentially (ranging from 40-fold down-regulation to 100-fold up-regulation) in response to 11 different types of chemical exposure. Several of the single chemical treatments and chemical types were shown to produce distinctive patterns of differential expression.

In addition, some of the modeling approaches tested on microarray results correctly identified the cause of toxicity in independent evaluation samples (one spiked and two field sediment samples). Cluster analysis had limited success, whereas a class prediction model using random forest analysis provided more accurate classification results. Several remaining research needs include (1) gene expression information for contaminants of concern, (2) standardization of sample preparation and analysis methods for use across multiple laboratories, and (3) further development and evaluation of statistical classification models.

Partners

Environment Canada
San Francisco Estuary Institute
UC Davis
UC Berkeley

Presentation

• Development of molecular tools for stressor identification in sediment toxicity tests - January 2012 presentation to SCCWRP member agencies summarizing progress on developing molecular methods for toxicity identification evaluation.