Workpackage 5: Metabonomics




To identify metabolic biomarkers that are associated with ALS in a prospective study


To identify perturbed metabolic pathways associated with ALS phenotypes by systems biology approaches to molecular profile integration


To evaluate the translation of metabolic pathway effects between ALS models and man


To evaluate the predictive ability of metabolic biomarkers for ALS in a prospective manner


The systematic analysis of metabolism (metabolic profiling, metabonomics or metabonomics) in living organisms, alongside genomics, epigenomics, transcriptomics and proteomics, is increasingly being viewed as a vital part of the toolkit for global biomolecular modelling (systems biology) and biomarker discovery. Metabolic profiling have several potential advantages over genomic and proteomic counterparts: metabolites are a defined chemical entity irrespective of species, genotype, localisation and biological matrix, facilitating the translation of analytical procedures between models and man; changes in metabolism are a phenotypic and often functional endpoint. These factors, coupled with the fact that established and widely available analytical technologies such as NMR spectroscopy and mass spectrometry can be used, explain the explosion of research in the last decade using metabolic profiling as a strategy to identify pathways associated with disease.

NMR and MS in broad terms are highly fit-for-purpose as metabolic profiling technologies:  both can be used such that they are largely untargeted in the molecular structures that will be detected ; both have technical extensions that give further detailed structural information, e.g. multidimensional NMR, MS/MS.  However they differ in several key respects: MS is several orders of magnitude more sensitive than NMR; NMR spectroscopy is more analytically reproducible and robust across laboratories.  Hence in this project we will utilise both NMR and MS strategies depending on the sample type to maximise coverage of the metabolome and to provide the greatest range of structural information.

Task 1. Obtain metabonomic profiles of humans, animal and cellular models (partner 15, 11). Plasma samples from prospective cohorts will be analysed by LC-MS protocols, specifically UPLC chromatography and quadropole time-of-flight mass spectrometry. We will employ a procedure based on a protocol that has been shown to be suitable for long-term metabonomics studies. Briefly, we will inject a pooled sample and human serum/urine standard (Sigma) at multiple points during each batch as quality control (QC) samples. 

Samples from cell and animal models will be analysed by a combination of LC-MS and NMR protocols as appropriate. NMR spectroscopy will be performed using procedures as described previously (Beckonert et al, 2007). These data will be input for WP9.

Task 2. Multivariate models of metabolic profiling data for association with ALS (partner 15). ICL will support bioinformatic analysis by exploiting prior experience of modelling large-scale metabonomic databases (e.g. the COMET consortium database; Ebbels et al, 2007; Lindon et al, 2005; Sinha et al, 2005; INTERMAP human global phenotyping database, Holmes et al, 2008) and of mQTL mapping (Dumas et al, 2007).  Also we will use pathway enrichment techniques and number of other pattern recognition algorithms developed by ICL for analysis of metabolic phenotypes for data integration including but not limited to: dual-mode genetic algorithms (Cavill et al, 2009); geometric trajectory analysis (Keun et al, 2004); probability density mapping (CLOUDS, Ebbels et al, 2007); O2-PLS (Cloarec et al, 2005).