Metabolomics in Yeast
- 1Donnelly Centre for Cellular and Biomolecular Research, University of Toronto, Toronto, Ontario M5S3E1, Canada;
- 2Department of Molecular Genetics, University of Toronto, Toronto, Ontario M5S3E1, Canada;
- 3Department of Biochemistry and Cambridge Systems Biology Centre, University of Cambridge, Cambridge CB2 1GA, United Kingdom;
- 4The Francis Crick Institute, Mill Hill Laboratory, London NW7 1AA, United Kingdom
Abstract
Budding yeast has from the beginning been a major eukaryotic model for the study of metabolic network structure and function. This is attributable to both its genetic and biochemical capacities and its role as a workhorse in food production and biotechnology. New inventions in analytical technologies allow accurate, simultaneous detection and quantification of metabolites, and a series of recent findings have placed the metabolic network at center stage in the physiology of the cell. For example, metabolism might have facilitated the origin of life, and in modern organisms it not only provides nutrients to the cell but also serves as a buffer to changes in the cellular environment, a regulator of cellular processes, and a requirement for cell growth. These findings have triggered a rapid and massive renaissance in this important field. Here, we provide an introduction to analysis of metabolomics in yeast.
METABOLOMICS IN YEAST
Research on metabolism and metabolic enzymes dominated the early days of molecular biology. Interest in the topic decreased with the appearance of polymerase chain reaction (PCR) and molecular genetics during the 1980s and 1990s, although the importance of budding yeast in biotechnology ensured continued progress in the understanding of its physiology. Yeast has been used in winemaking, brewing, and baking since ancient times. The economic importance of fermentation triggered the first attempts to breed and manipulate yeast species and led to the purification, crystallization, and characterization of the first enzymes in the early 20th century (with major contributions from Sumner and Northrop in the United States and Warburg in Germany). Yeast experiments were so important in the development of biochemistry that the word “enzyme” is derived from the Greek word for in leaven (yeast).
The full set of metabolic reactions in the cell is referred to as the metabolic network. Although complex, the basic structure of the metabolic network is largely conserved among organisms, indicating that it is of a common evolutionary origin (Jeong et al. 2000; Ravasz et al. 2002). Compared with the large number of chemical reactions and possible mechanisms, metabolism operates with only a subset of the reactions and prefers short paths in its functionality (Noor et al. 2010). Overall, this conservation means that metabolomics is a particularly attractive technique, as in principle any analytical method can be applied to a large variety of different species. However, sample preparation methods are species specific, because composition of the cell membrane, presence of a cell wall, cellular resistance to physical parameters, and relative and absolute metabolite content differ from species to species.
In principle, two values are important for a metabolic intermediate: its concentration and its turnover rate. Both values provide different information about metabolism. Whereas the concentration values can point to perturbations in the pathway, allow conclusions on the functionality of individual enzymes, and thus serve as a functional parameter for the regulatory cross talk of metabolism, the turnover rate (referred to as metabolic flux on a pathway and network scale) is informative about the activity of a metabolic pathway and its yield and consumption of cofactors and intermediate metabolites (Grüning et al. 2010; Heinemann and Sauer 2010).
Certain parameters are crucial in a yeast metabolomics experiment, and they determine not only the appropriate methods for sample preparation and analysis but also which yeast strains and growth conditions can be applied. These parameters include the following.
-
Nature of application: Targeted versus shotgun approach. Although it sounds attractive to quantify a large number of metabolites at the same time, quantitative accuracy declines in broad experiments. Different metabolite classes have different stabilities, chemical properties, and turnover rates. There is therefore no universal sample-quenching procedure that works equally well for all metabolites. Our accompanying protocols (see Protocol: Metabolite Extraction from Saccharomyces cerevisiae for Liquid Chromatography–Mass Spectrometry [Rosebrock and Caudy 2015], Protocol: A High-Throughput Method for the Quantitative Determination of Free Amino Acids in Saccharomyces cerevisiae by Hydrophilic Interaction Chromatography–Tandem Mass Spectrometry [Mülleder et al. 2015], and Protocol: Spectrophotometric Analysis of Ethanol and Glucose Concentrations in Yeast Culture Media [Caudy 2015]) suggest several widely applicable methods, but it is important to define the biological questions of interest in advance to select the optimal experimental method.
-
Turnover rates. Turnover is fast, and concentration changes of metabolites occur rapidly. For instance, applying hydrogen peroxide to yeast cells to cause oxidative stress leads to an inactivation of the metabolic enzyme GAPDH within 2–15 sec, while equally fast, an increase in the concentration of pentose phosphate pathway metabolites is observed (Ralser et al. 2009). Sampling procedures need to take this speed of metabolism into account. Washing and centrifugation steps in water or chemical buffers quickly cause starvation phenotypes and should be avoided. To preserve the state of metabolism, cells should be quenched as quickly as possible (i.e., by rapid freezing in cold organic solvents [Rabinowitz 2007; de Koning and van Dam 1992]), as described in Protocol: Metabolite Extraction from Saccharomyces cerevisiae for Liquid Chromatography–Mass Spectrometry (Rosebrock and Caudy 2015).
-
Auxotrophies and media supplements. Auxotrophic markers are very attractive tools in yeast genetics, but be aware that they represent a major problem in yeast metabolism research. For example, the common laboratory strains BY4741 or W303 possess four to five autotrophies; in practice, this means that highly active biosynthetic pathways are simultaneously perturbed. The physiological effects are only partially complemented by media supplementation, and different nutrients are depleted at different rates during the growth phase. Thus, matching autotrophies between strains within an experiment is essential, and where possible, prototrophic cells are preferred in a metabolomics experiment (Pronk 2002; Mülleder et al. 2012; VanderSluis et al. 2014).
-
Growth. Yeast metabolism strongly differs between growth phases. Some metabolite concentrations differ between early-, mid-, and late-exponential phases, and stationary phase before or beyond the diauxic shift. In batch cultures, these effects are amplified for auxotrophic strains, because essential supplements are consumed at a different rate. It is important to carefully control growth conditions for cell harvesting.
ANALYSIS OF METABOLOMICS
Metabolites have very different physicochemical properties and can span a concentration range of many orders of magnitude. Therefore, appropriate sample preparation and analytical methods must be used to reliably determine the metabolites of interest and provide a comprehensive snapshot of the entire metabolome. Currently, metabolomics is mostly conducted by nuclear magnetic resonance (NMR) and mass spectrometry (MS).
Nuclear Magnetic Resonance
NMR is a widely used analytical platform for metabolism research and structural elucidation of compounds. Its application in metabolomics is limited by sensitivity (typically micromolar to nanomolar) compared with MS (femtomolar to attomolar) range. However, it has the advantage of allowing in vivo measurements and direct observation of metabolic reactions when hyperpolarization is used to increase sensitivity (Meier et al. 2011; Rodrigues et al. 2014).
Mass Spectrometry
Different platforms and strategies exist for MS, and they have advantages and limitations according to the question to be answered. Mass analyzers with high accuracy, acquisition speed, and resolution are the main requirement for untargeted approaches; this application is currently dominated by time-of-flight (TOF) and Fourier transform–based mass spectrometry (FT-MS) using Orbitrap instruments. In targeted analyses, sensitivity and dynamic range are the important parameters. These analyses are typically performed on (triple) quadrupole platforms or hybrid or tandem platforms combining quadrupoles with high-resolution mass analyzers (i.e., Quadrupole-TOF [qTOF]). Quantification of metabolites is achieved with external or internal standards (stable isotope-labeled standard compounds) or metabolic labeling (extracts from cells grown in isotope-enriched media). Samples can be measured directly by surface-based methods or by flow injection analysis (FIA); however, to increase specificity and quantitative reliability, the complexity of the samples is usually reduced by separation techniques such as gas chromatography (GC), liquid chromatography (LC), capillary electrophoresis (CE), or ion mobility. GC-MS is very useful for volatile, thermostable, and energetically stable molecules and achieves high selectivity and reproducibility. Many polar metabolites, however, are poorly volatile in GC and therefore need prior derivatization to ionize. In contrast, LC-MS allows separation of a broad range of metabolite classes and is therefore the most widely used platform for metabolomics. The dominant LC technique coupled to MS is reversed phase high-performance liquid chromatography (HPLC). Hydrophilic interaction liquid chromatography (HILIC) and ion pairing reagents have proven very useful for the separation of hundreds of polar hydrophilic compounds of primary metabolism (Bajad et al. 2006; Buescher et al. 2010). Although surface-based MS techniques (e.g., matrix-assisted laser desorption/ionization [MALDI]) are limited to abundant metabolites, the ability to measure metabolites at single cell concentrations promises to drive future analysis (Amantonico et al. 2008; Ibáñez et al. 2013).
CONCLUSION
Metabolism is involved in many human diseases, including diabetes, cancer, and neurodegeneration (Hsu and Sabatini 2008; Grüning et al. 2010; Hanahan and Weinberg 2011; Buescher et al. 2012; Keller et al. 2014). Yeast as a model organism continues to be of central importance to the study of metabolism. With the availability of the yeast genome sequence and the wealth of knowledge on the biochemistry of the yeast cell, researchers have generated a reconstruction of its metabolic network. The reconstruction is under continuous refinement and is currently the best eukaryotic model available (Mo et al. 2009). It not only describes the network topology but also allows for simulation and prediction of phenotypes, such as growth, gene knockouts, production of important compounds, and nutrient uptake. Genome-scale models can be combined with transcriptomic, proteomic, and/or metabolomic profiles to better understand an observation, and they have also been successfully applied for metabolic engineering and strain optimization (Oberhardt et al. 2009). Furthermore, they allow for estimation of the metabolic flux under defined steady-state conditions, providing valuable information about pathway activities and aiding in the determination of in vivo flux rates from isotope tracer distributions in metabolic flux analyses (Sauer 2006). It is important to note that because the models are based on our knowledge, they are prone to errors due to incorrect or missing information; therefore, predictions must always be verified experimentally (Österlund et al. 2012).
