Community Research and Development Information Service - CORDIS

Final Report Summary - MRM-YEAST METABOLISM (A Systems Biology Approach to Elucidate the Yeast Metabolic Network)

The ultimate goal of systems biology is to generate mathematical models to comprehensively describe a biological system. Among all biochemical and informational systems that operate in cells, metabolism is unique because the composition and topology of the network of metabolic reactions is almost completely known. Most of the reactions, the enzymes catalyzing them, the enzyme-encoding genes, and the involved chemicals are known. What we do not understand currently is how this system operates, is controlled and how it adapts to changing conditions of supply and demand. To study these system properties, it would be crucial to measure all the elements that constitute and regulate it. However, quantitative and comprehensive measurement remains technically difficult even in a simple model organism, such as yeast, especially at the proteome level. Classical proteomic methods are based on the analysis by liquid chromatography coupled-tandem mass spectrometry (MS) of complex peptide mixtures, generated by proteolysis of protein samples. In the most successful set-ups they allow to identify and quantify, in a single experiment, up to few thousand proteins. However, these methods are non-targeted, i.e. in each measurement they quasi-randomly sample a fraction of a proteome. Each analysis will thus sample only a subset of a list of proteins of interest (e.g. those that constitute a metabolic or signaling network), thus hindering the generation of complete datasets. Extensive fractionation at the protein or peptide level allows to increase the coverage. However, the time and the effort it requires strongly limit the number of differentially treated samples and replicates that can be analyzed in a reasonable time.
In the context of this project, I applied an alternative targeted proteomic approach, based on a MS technique called selected reaction monitoring, to the study of the dynamics of the yeast metabolic network. The essence of the approach is the generation of specific, quantitative mass spectrometric assays for each protein of interest and their subsequent application to multiple biological samples. The approach starts with the generation of a list of target proteins. For each protein proteotypic peptides (PTPs, peptides that are unique to the protein and preferentially detectable by MS) are selected. For each PTP, precursor/fragment ion relationships are then established, that specifically identify the PTP. These relationships, commonly termed SRM (or MRM) transitions, therefore effectively constitute mass spectrometric assays that identify and quantify a specific peptide and, by inference, the corresponding protein, in a complex protein digest1.
First, I tested the applicability of the proposed targeted proteomics approach to the analysis of a whole metabolic proteome in S. cerevisiae. I used the approach to detect and quantify a set of proteins expressed in yeast across the whole range of cellular abundances, down to a concentration below 50 copies per cell2. I also demonstrated that i) proteins can be measured in total cell lysates, without the need for sample fractionation or enrichment, making the use of the technique fast and practical; ii) the technique is multiplexed, supporting the detection and quantification of different proteins, deliberately chosen and spanning all abundances, in a single analysis; iii) the technique has the capability to detect proteins which were undetectable by classical MS or affinity-based methods2. Overall these achievements demonstrated that SRM allows to comprehensively monitor protein networks in yeast in about one hour of instrument time, and thus to analyze in a reasonable time the effects on the system under study of different perturbing conditions. The characteristics of the technique satisfy in an ideal way the growing demand of systems biology and biomedical research for consistent, complete, and quantitative data sets from differently stimulated cells.
The development of the first set of SRM assays for yeast proteins inspired me to develop a database that supports the collection, organization and dissemination of the developed assays in a centralized, publicly accessible data base. This resulted in the construction of the MRMAtlas3, a web-based resource to store the final coordinates of SRM assays and share them between different laboratories. The MS coordinates required to reproduce the assays can be downloaded in a spreadsheet format and used to quantify the proteins of interest in any yeast sample of interest. These achievements demonstrated the feasibility of developing reproducible MRM assays for any yeast protein and sharing them with the scientific community.
Next, we focused on the development of SRM assays for metabolic proteins. This comprised all enzymes in the central carbon and amino acid metabolism of yeast, including known or hypothetical central metabolism isoenzymes, as derived from the currently most advanced stoichiometric model of yeast metabolism. The assays were applied to quantifying the metabolic subproteome in yeast cells grown under a set of conditions inducing radically different metabolic setups (i.e. yeast growth in glucose, galactose, ethanol-based or complex media, and under anaerobic conditions) and in a growth time-course of yeasts transiting through a complex series of metabolic phases (transition from glucose- to ethanol- growth, through the diauxic shift and until the stationary phase). These unique quantitative dataset, interpreted through flux balance modeling, indicates that S. cerevisiae expressed superfluous proteins, not necessarily used in a particular metabolic condition4. Further, the data allowed to suggest differential functionality for several metabolic isoenzymes. The quantitative proteomic dataset was also coupled to transcriptomic and phosphoproteomic analyses of the same set of metabolic enzymes. This highlighted enzymes that change their degree of phosphorylation, under the same set of conditions mentioned above, and are therefore potential sites of post translational regulation within the network. This resulted in the most complete quantitative dataset to date illustrating how the yeast metabolic network reacts to changing conditions of supply and demand of nutrients. Ultimately, it provides an ideal blueprint for the biological and biomedical community for the advancement of current models of metabolism and for their mathematical formulation.
During the course of the project, I became aware of the significant effort required to establish a SRM assay for a protein. This is based on a lengthy series of operations, mostly on a trial and error basis. This can result in several days to establish an SRM assay for a single protein. To overcome this limitation, I developed a strategy to greatly facilitate and accelerate the generation of SRM assays. It is based on the use of libraries of low-cost crude, unpurified synthetic peptides as a reference for deriving the final coordinates of a SRM assay5. This technical breakthrough enabled facing the challenge of developing SRM assays for the complete yeast proteome. In the last months of my work, using a set of ~25,000 synthetic peptides I could successfully develop SRM assays for 97% of proteins associated to all protein-coding open reading frames in the S. cerevisiae genome6. This includes ORFs for which no experimental evidence demonstrated yet that an expressed protein is encoded. The assays will be made public via the MRMAtlas interface (expected, fall 2010) and enable the scientific community to attempt the detection and quantification of any protein in yeast cells grown under any condition of interest6. The approach was also applied to the development of SRM assays for a set of about 8,500 human proteins.
Overall the targeted proteomic approach, developed and applied in this project, revealed its potential as a robust platform for studying the effects on the cell’s physiology of any internal or external stimulus and thus to find broad application in basic and applied biological and biomedical research.

Related information

Reported by

Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top