T2 PM 'Omic' Technologies: Current and Future Applications to Environmental Toxicology (Part 2)
Tuesday, 15 November 2005: 1:50 PM - 5:30 PM in Ballroom 2

(VIA-1117-921075) Comparison of experimental strategies in NMR based metabolomics: An application to environmental disease diagnosis.

Viant, M¹, Southam, A¹, Ludwig, C¹, Slupsky, C², Günther, U¹, Stentiford, G³, Lyons, B³, Feist, S³, ¹ University of Birmingham, Birmingham, U.K.² CHENOMX Inc., Edmonton, Alberta, Canada³ Centre for Environment, Fisheries & Aquaculture Science, Weymouth, U.K.

ABSTRACT- As the interest in metabolomics increases, in particular as a tool in ecotoxicology, it is important to critically evaluate the data acquisition and analysis strategies that are currently available, and to determine which may be of most use to the environmental scientist. Here we focus upon ¹H nuclear magnetic resonance (NMR) based metabolomics, and compare several different methodologies for characterising the metabolic differences between cancerous and healthy liver tissue from fish. Specifically, histologically defined hepatocellular adenomas were compared with control liver in dab (Limanda limanda), a marine flatfish, that were collected as part of the National Marine Monitoring Programme of the United Kingdom. Following extraction of polar metabolites, ¹H NMR spectra were measured at both 500 and 800 MHz using a one-dimensional experiment, as well as at 800 MHz using two-dimensional J-resolved spectroscopy. Spectra were then analysed using two contrasting approaches; the first comprised of extensive pre-processing using ProMetab software (Birmingham) followed by principal components analysis (PCA) of the spectral datapoints. Second, Profiler (Chenomx NMR Suite) was used to identify and quantify multiple metabolites that were then analysed using PCA. All datasets yielded similar PCA scores plots, which identified consistent and significant metabolic differences between the cancerous and healthy liver tissue. The advantages and disadvantages of these data acquisition and analysis strategies will be discussed. In particular, the advantages of Profiler include the ease of metabolic interpretation and instrument-independent data archiving. The advantages of J-resolved spectral projections and ProMetab pre-processing include unprecedented spectral resolution, and multivariate analyses which include all peaks within the spectra. This presentation will conclude with recommendations for researchers interested in using metabolomics as a complementary tool to genomics, transcriptomics and proteomics in ecotoxicogenomics.

Key words: metabolomics, ecotoxicogenomics, metabolites, cancer