| Summary: | Dioxin exposure is linked to a range of toxic effects in humans including cancer and irreversible developmental effects. Exposure is primarily through the ingestion of animal-derived foods so the ability to detect and eliminate dioxin from the food chain is paramount to ensuring public health. This study was designed to investigate the potential of analytical profiling tools to detect and identify live animals which have been exposed to dioxin based upon patterns of changes in analytes present in accessible biofluids such as blood and urine. In vitro cell culture studies identified a range of additive and non-additive interactions between dioxin and non-dioxin like co-toxins across a spectrum of endpoints including EROD activity, reduction of Alamar Blue and SELDI-TOF peptidomic profiling. Rats were then exposed to dioxin in their diet as 2,3,7,8-TCDD, but also in the form of incinerator soot or the PCB oil Aroclor 1254. The plasma and urine were screened for metabolomic changes which resulted from exposure using liquid id chromatography-high resolution mass spectrometry methods. This data was processed using multivariate statistical analysis techniques to identify potential biomarkers and to look for predictive statistical models which may be able to identify exposure. Furthermore the level of expression of a range of proteins were determined by the use of 2-dimensional gel electrophoresis to help understand the pathways affected by dose to real-world dioxin exposure condition s. A predictive statistical model was validated based upon metabolomic profiling of the plasma of the exposed rats which was able to correctly identify the source of dioxin exposure with a >97.2% accuracy rate. A novel method for validation of the accuracy of OPLS-DA models was developed and tested. Patterns of changes to amino acid metabolism such as tyrosine and tryptophan were identified as being important in exposure, indicating novel modes-of-action not previously associated with dioxin exposure.
|
|---|
