Metabolites are intermediates of biochemical reactions and responsible for connecting the many different pathways operating in a living cell. The level of metabolites is determined by the properties and concentration of enzymes, and, therefore, their level is a consequence of many regulatory processes inside the cells such as the regulation of transcription, translation and protein-protein interactions. Thus, the level of metabolites represents the molecular phenotype of a cell or organism in response to the genetic or environmental factors.
|Figure 1. Omics technologies (adapted from "Villas-Bôas et al. 2007).|
Metabolomics is one of newest "omics" technology which is capable of screening a large number of metabolites in biological samples. Although sometimes considered as a young field of research, metabolomics has been evolving rapidly during recent years and it is now complementary to different "omics" technologies, such as transcriptomics and proteomics. As a consequence, metabolomics has been widely used as a functional genomics tool as well as playing essential role in systems biology studies. Today we can find metabolomics being applied to different areas of research in life sciences, from phenotypic characterization of microbial strains, gene-knockout mutants and plant cultivars, to diagnostic of diseases and metabolic disorders.
The analytical pipeline for metabolomics comprises multiple steps and there are different available protocols for each one of them. Some steps may be added or skipped depending on the biological samples, questions being asked and/or experiment design.
However, for all cases the first step is the “sampling”, which often requires the "quenching" of cellular metabolism. The time is a key factor during sampling, as some metabolites present a very fast intracellular turn-over (< 1 sec).Thus, the whole sampling and quenching process must be completed as fast as possible in order to avoid alterations in intracellular metabolite levels. In our laboratory we have developed a quenching method based on cold glycerol-saline solution as the quenching agent. It prevents significant leakage of intracellular metabolites during quenching of microbial cells and, therefore, permits more accurate measurement of intracellular metabolite levels.
|Figure 2. Metabolome analysis (adapted from "Villas-Bôas et al. 2004).|
For the analysis of intracellular metabolites (metabolic fingerprinting), it is necessary to extract metabolites from the cells. For that, we need to make the cell envelope permeable to an extracting agent, such as an organic solvent, thus releasing intracellular metabolites. In our laboratory we use freeze-thaw cycles combined to a solution of cold methanol-water. Freeze-thaw cycles increase the permeability of the cells and enhance the extraction of metabolites by the cold methanol-water solution. As a result, we have a simple and fast technique able to extract a broad range of polar metabolites, whilst maintaining the samples at low temperature to avoid chemical degradation of labile metabolites.
The analysis of the extracellular metabolites (metabolic footprinting) is simpler, as it often does not requires metabolite extraction. It is achieved by profiling the level of metabolites in the spent culture media. Every metabolite detected in the medium is either a compound secreted by the cell or part of the initial medium composition. Thus, using this technique we can easily determine the production and consumption of specific metabolites, which reflects the metabolic state of the cells and is specific to the genetic background of the cells (genotype) as well as the environmental conditions they were grown.
|Figure 3. Gas Chromatography - Mass Spectometry (adapted from "Villas-Bôas et al. 2007).|
Gas Chromatography – Mass Spectometry (GC-MS) is one of the most used analytical tools in metabolomics. Due to its high separation power, its capacity for reliable identification of hundreds of metabolites and its low cost, GC-MS is our “work-horse” for metabolite analysis. However, GC-MS systems are able to analyze only volatile compounds and, consequently, chemical derivatization of nonvolatile compounds is required.
There are mainly two types of chemical derivatization commonly used in metabolomics: silylation and alkylation. Silylation is efficient for derivatizing sugars and their derivatives. Although many research groups use silylation for analysis of different classes of metabolites, it is our experience that compounds at large concentration in the sample matrices such as sugars, urea, water, lipids and/or proteins strongly interferes with the derivatization reaction. Consequently, metabolites from microbial samples are generally poorly derivatized by silylation reactions. Therefore, our group developed a fast alkylation reaction based on methyl chloroformate (MCF). MCF converts amino and nonamino organic acids into volatile carbamates and esters. Although limited to compounds presenting amino and/or carboxyl groups, these include most metabolites of the central carbon metabolism, which are key intermediate of the cell metabolism. In addition, the MCF derivatization is very fast; it is performed at room temperature and uses relatively inexpensive reagents. Also, MCF uses chloroform for extracting alkylated derivatives from the sample matrix, thereby causing less damage to the GC-MS column than other commonly used derivatization reagents such as silylation reagents. We have built an MS library containing over 150 different metabolites.
|Figure 4. Example of graphical outputs to present metabolomics results (Figure produced by Xavier Duportet).|
After derivatization, the biological samples are then submitted to a GC-MS system and, as a result, the GC-MS produces data files containing chromatograms and mass spectra of metabolites detected in each sample. However, these files are not yet ready for analysis. They must be processed in order to correct some aspects of the data output, such as the GC-peak intensities (data mining). For that, we have developed a set of tools able to analyze metabolomics data sets in an automatic fashion way. These tools work in conjunction with the Automated Mass Spectral Deconvolution and Identification System (AMDIS) and the R software (www.r-project.gov). AMDIS is one of the freeware software most used in metabolomics. It is considered very efficient for deconvoluting chromatograms generated by mass spectrometers (MS) and also for identifying and quantifying metabolites in biological samples. However, because of some limitations, such as the use of different mass fragments for quantifying peak intensities, data sets processed by AMDIS often require extensive data correction. R software is an open source language environment developed for statistical analysis. However, R ended up being so powerful that it is now used for many other purposes such as analysis of chromatograms and mass spectra. Thus, using R software coupled to AMDIS we developed an R package called Metab. Metab is able to automatically normalize metabolomics data sets by internal standard, biomass and medium, tasks generally performed manually. Furthermore, Metab is able to delete compounds considered false positives and, in addition, it performs statistical tests such as t-test and ANOVA. As a consequence, Metab reduces considerably the amount of work spent for data mining and prevents human errors. In addition, Metab is able to extract the profile of non-identified mass fragments present in biological samples, which is widely applied for non-target analysis of metabolites.
Lastly, metabolomics is capable of providing important information about the cell metabolic state and about the activity of metabolic pathways. Recently, our group published a tool named PAPi. It is an R package which calculates what we called the "Activity Score" of metabolic pathways. As a result, it generates graphs highlighting the metabolic pathways that tend to be more important to define the differences between experimental conditions. This way, we can use metabolomics data sets to predict metabolic flux distribution and to characterize the metabolic response of organisms to different environment conditions.
For more information about the platform and methods used in our laboratory, please visit our Group publication section.
Villas-Bôas, S.G. Sampling and sample preparation. In Metabolome Analysis: An Introduction (eds. Villas-Bôas, S.G., Roessner, U., Hansen, M.A.E., Smedsgaard, J. & Nielsen, J) 39–82 (John Wiley & Sons, Hoboken, New Jersey, USA, 2007).