Metabolic footprinting and Systems Biology: The medium is the message
Kell et al. (2005): Metabolic footprinting and Systems Biology: The medium is the message
This paper deals with metabolomics.
Metabolomics is 'downstream' - changes in the metabolome are amplified relative to changes in the transcriptome and proteome, and are arguably numerically more tractable. There is no need for a whole genome sequence or for large expressed-sequence-tag databases to be available for each species. Metabolic profiling is cheaper and more high-throughput than proteomics and transcriptomics, making it feasible to examine large numbers of samples from organisms that have been 'grown' under a wide range of conditions. Finally, the technology involved in metabolomics is generic, as a given metabolite - unlike a transcript or protein - is the same in every organism that contains it (other than for secondary metabolites, as originally defined).
The authors "devised METABOLIC FOOTPRINTING as a novel method for the functional analysis and characterization of cells using the metabolome".
Metabolic footprinting as described relies not on the measurement of intracellular metabolites (a technique which is widely referred to as METABOLIC FINGERPRINTING) but on the monitoring of metabolites consumed from, and secreted into, the growth medium by batch cultures of yeast using directinjection MS, in which samples of culture media are injected directly into an electrospray ionization mass spectrometer. To maximize the excretion of metabolites, overflow metabolism is stimulated by adding to the fully defined medium various carbon compounds that 'probe' metabolically active networks in the same way that an engineer might probe an electrical circuit. [...] Metabolic footprinting has proved useful for detecting the patterns of metabolites in single-gene-knockout strains in functional genomic analyses as well as in mode-of-action studies, in which the pattern of metabolites excreted when strains are challenged with sub-lethal concentrations of growth inhibitors makes it possible to discriminate the site or mode of action of those inhibitors.
In the section "Future directions: data standards and curation", the authors write:
An important need for large-scale studies of the type facilitated by metabolic footprinting is the ability to store metabolomics data in well-designed and curated databases that can store, handle and disseminate large amounts of metabolomics data efficiently and readily lend themselves to data mining and machine learning. Various omics data models (colloquially 'data standards') have been proposed, focusing on specific research areas and analytical methods. For example, MIAME (minimum information about a microarray experiment) and MAGE-OM (microarray-geneexpression object model), and PEDRo/PSI (proteomics experimental data repository/proteomics standards initiative) are the emerging standards for transcriptomics and proteomics, respectively. Similar attempts are now coming through for metabolomics, and other MS data standards such as mzXML (in which MS data-acquisition files are converted into extensible markup language, XML) will be useful for both proteomics and metabolomics. [...] It usually takes time for a research community to agree on, and fully embrace, new standards. However, the systems biology experts correctly recognized the potential of XML. Systems biology markup language (SBML) is being developed as an XML format to represent models of biochemical reaction networks. Although far from complete, it has been widely accepted within the systems biology community as an information standard, allowing complex models to be shared, evaluated and developed cooperatively.