Understanding the control of cell metabolism and growth in Saccharomyces cerevisiae through manipulation of purine nucleotide concentrations

Lead Research Organisation: University of Cambridge
Department Name: Biochemistry


All cells require energy to survive and to grow. Green plants get this energy from the sun and convert it into biochemicals. The cells of most other organisms, including ourselves, get their energy from nutrients (biochemicals). However, the energy requirements of a cell vary, as does the supply of the nutrients that provide that energy. For instance, rapidly growing cells consume significantly more energy than cells that have stopped growing, and cells supplied with a high level of nutrients have more energy available for use than those kept on a starvation diet. Therefore, cells need control systems capable of both sensing the quantity and quality of nutrients available, and ensuring that the chemical processes which generate and consume energy are always balanced properly. What we don't understand is what it is that cells monitor to perform this balancing act. We have some clues that they record the amount or ratio of two families of high-energy biochemicals. Thus these biochemicals act not just as sources of energy but also as signals of nutrient status. We aim to investigate their role in the nutrient control systems using cells of the brewing and baking yeast, Saccharomyces cerevisiae. The structure and ways of working of yeast cells are pretty much the same as those of human cells. This means that what we learn about yeast provides important information about how our own cells, and those of our farm animals and crop plants, work. This can help us to understand how cells malfunction to produce disease. One of the primary characteristics of tumour cells, for instance, is that they grow and divide much more rapidly than a normal healthy cell of the same tissue type. Moreover, the energy metabolism of tumour cells is different from that of normal cells; in fact, it is more like that of yeast.
This project is not only important for our understanding of disease. Yeast is also increasingly used as a living factory for the sustainable production of renewable/carbon-neutral energy sources, and of biochemicals of pharmaceutical importance. Biotechnologists are introducing new metabolic pathways into yeast for the production of fuels or chemicals. This hijacks yeast metabolism away from its normal processes, and requires it to use its energy resources in a significantly different way. Additional burdens on metabolism may also arise as a result of nutrients being supplied to the yeast, to convert into these useful products, that are very different from its normal diet. This would typically be the case in commercial biotechnology processes where cheap waste materials, from forestry, agriculture, or the food industry are employed. A detailed understanding of the mechanisms which control and balance the supply and use of energy in yeast cells will help us increase the yields and efficiencies of novel biotechnological processes like these, thereby making it more likely that they can be implemented and commercially viable.

Technical Summary

The intracellular concentrations of the phosphorylated purine nucleotides ATP and GTP have been proposed to be important global signals for regulating the balance of catabolism and anabolism in eukaryotic cells. How the relative ratios of adenine (ATP, ADP, AMP) and guanine (GTP, GDP) nucleotides change under different nutritional conditions has been characterized in S. cerevisiae, but how these changes are sensed and translated into cellular signals for the regulation of metabolism is not understood. We will use the regulated expression of bacterial genes which encode enzymes that use ATP or GTP as substrates to develop methods for inducing controlled changes to in vivo nucleotide concentrations which are independent of nutrient supply, and for generating nucleotide analogues to interfere with the sensing of GTP/GDP levels. The consequences of the inductions on global cellular metabolism will be characterized using transcriptomics and metabolomics, thereby determining the relationship between nucleotide composition and metabolic status and signalling. To identify gene functions important for mediating the responses to the induced changes in purine nucleotide abundance, we will undertake: i) functional interaction screening using the yeast heterozygous diploid mutant collection to find suppressors/enhancers of the effects of induction; ii) a focused screening of a selected subset of the yeast homozygous deletion mutant collection to identify strains exhibiting an abnormal uncoupling of growth rate and intracellular nucleotide composition.
Knowledge of how S. cerevisiae cells control their metabolism and coordinate the metabolic processes that provide and consume energy is often transferable to human cells where the dysregulation of these processes has been implicated in a number of disease states including tumourigenesis and diabetes. It is also directly applicable to optimization of the biotechnological processes for which S. cerevisiae is being increasingly used.

Planned Impact

The results of this study will primarily be communicated through presentations in scientific conferences and through publications in the high-impact journals. We will communicate with relevant industrial partners both directly; for instance, with our collaborators - Novacta Biosystems of Welwyn Garden City, Syngenta and Unilever - and via the meetings of the Bioprocess Research Industry Club (BRIC). In year three of the Project, we will organise a half-day meeting to explain our research to interested industrial scientists. This meeting will be held in Cambridge; the Science Parks surrounding the city contain the UK's largest concentration of small and medium-sized companies involved in the biotechnological and biomedical industries, and major companies such as GSK, Unilever and Pfizer in easy reach. Two alumni from the Oliver laboratory are employed by Novacta, and another is at Pfizer, Abingdon. However, we will also provide a video link to facilitate the participation of those who are unable to travel to Cambridge. We will ask Cambridge Enterprise (the University's Technology Transfer company, see below) to sponsor this event.

We will also communicate with the industry via the web-site of the Cambridge Systems Biology Centre and data will be available there as well as at public repositories (e.g. ArrayExpress and MetaboLights at the EBI).

Our laboratories regularly contribute to Cambridge Science Week and we will communicate this project to the wider public by that route and through summer placements for 6th Form students. Currently, SGO has regular placement plans for summer students from the John Fisher School (Peaks Hill, Purley, Surrey, CR2 6BN). We will also exploit the Cambridge Press Office and the 'Naked Scientist' radio programme to communicate our work.

Cambridge Enterprise will be responsible for protecting any Intellectual Property arising from the research. The University's technology transfer company has been very successful at facilitating the application of research both through licensing and supporting spin-out companies, including three from our own Department - Kudos, Astex, and Biotica.
Description A set of yeast strains were successfully constructed in which it is possible to induce the futile consumption of ATP or GTP nucleotides. These have been used to generate large functional genomics datasets, which will be made publicly available to the research community, consisting of RNAseq transcriptome data and genome-wide genetic interaction data and characterising yeast adaptation to changes induced in purine nucleotide biosynthesis. Our analysis and interpretation of these data is ongoing, but it is anticipated that it will shortly produce two peer-reviewed publications in international journals. The initial results from the study have been presented as invited lectures at two international conferences (in Japan and South Korea) by the researcher co-PI. The research assistant employed on the project has subsequently progressed to studying for a PhD at King's College London.

We have engineered the inducible synthesis of the bacterial signalling nucleotides cdiGMP, cdiAMP and ppGpp by Saccharomyces cerevisiae in order to characterize the range of effects these nucleotides exert on the functioning of a eukaryotic cell. Synthetic genetic array (SGA) and transcriptome analyses indicate that, while these compounds elicit some common reactions in yeast, there are also complex and distinctive responses to each of the three nucleotides. All three are capable of inhibiting eukaryotic cell growth, with the guanine nucleotides exhibiting stronger effects than cdiAMP. Mutations compromising mitochondrial function and chromatin remodelling show negative epistatic interactions with all the nucleotides. In contrast, certain defects in chromatin modification and ribosomal protein function show positive epistasis, alleviating growth inhibition by at least two of the three nucleotides. Uniquely, cdiGMP is lethal both to cells growing by respiration on acetate and to cells of petite mutants which are incapable of respiration. This compound is also synthetically lethal with the ribonucleotide reductase (RNR) inhibitor, hydroxyurea. Heterologous expression of the human ppGpp hydrolase, Mesh1p, prevented the accumulation of ppGpp in the engineered yeast cells and restored cell growth. Extensive in vivo interactions between bacterial signalling molecules and eukaryotic gene function are therefore possible, resulting in outcomes ranging from the inhibition of cell growth to cell death. cdiGMP functions through a mechanism that must be compensated by unhindered RNR activity or by functionally competent mitochondria. Mesh1p may be required for abrogating the damaging effects of ppGpp in human cells subjected to bacterial infection.

During infections, pathogenic bacteria can release nucleotides into the cells of their eukaryotic hosts. These are recognised as a signal that contributes to the initiation of defensive immune responses which help the infected cells to recover. Despite the importance of this process, the broader impact of bacterial nucleotides on the functioning of eukaryotic cells remains poorly defined. To address this, we genetically modified cells of the eukaryote Saccharomyces cerevisiae (baker's yeast) to produce three of these molecules (cdiAMP, cdiGMP and ppGpp), and used the engineered strains as model systems to characterise the effects on the cells. In addition to demonstrating that the nucleotides are each capable of adversely affecting yeast cell function and growth, we also identified the cellular functions important for mitigating the damage caused, suggesting possible modes of action. This study expands our understanding of the molecular interactions that can take place between bacterial and eukaryotic cells.

Using the controlled nutritional environment provided by continuous (chemostat) culture, we observed a strong correlation between the abundance of ATP and GTP and the transcription of genes required for growth. By engineering short pathways for the inducible and futile consumption of ATP or GTP into S. cerevisiae, the effect on gene transcription of a reduction in the abundance of these nucleotides was analysed. During steady-state growth using the fermentable carbon source glucose, the futile consumption of ATP led to a decrease in intracellular ATP concentration, but an increase in that of GTP and the guanylate energy charge (GEC = [GTP] + 0.5[GDP] + [GTP + GDP + GMP]). The expression of transcripts encoding proteins involved in ribosome biogenesis, and those controlled by promoters subject to SWI/SNF-dependent chromatin remodelling, was correlated with these changes in nucleotide pools. Similar nucleotide abundance changes were observed in a parallel experiment using the non-fermentable carbon source acetate, but an effect on the growth-associated transcriptional programme was absent. Induction of the GTP-cycling pathway had only marginal effects on nucleotide abundance and gene transcription. The transcriptional response of respiring cells to glucose was dampened in chemostats induced for ATP-cycling, but not GTP cycling, and this was associated with altered ATP, AMP, GMP, cdiAMP and adenylate energy charge (AEC) levels

We conclude that gene transcription in Saccharomyces cerevisiae can be modulated by changes in intracellular nucleotide pools. The changes in transcription we observed are most consistently associated with changes in GTP and GEC levels, although the reprogramming in gene expression that takes place in respiring cells subjected to glucose repression is also sensitive adenine nucleotide levels. A comprehensive and integrative understanding of the control of eukaryotic gene expression thus requires the elucidation of how changes in guanine nucleotide abundance are sensed and transduced to alter the global pattern of transcription.
Exploitation Route The findings have been written up for publication and are about to be submitted.

The findings provided a valuable set of of data on purine nucleoside and nucleotide concentrations that revealed serious deficiencies in the genome-scale metabolic model of Saccharomyces cerevisiae in respect of iron metaboilism. This has allowed us to extend the model and use it to represent heritable human diseases that involve derangement of iron metabolism and the assembly and transport of Fe-S clusters.
Sectors Chemicals,Pharmaceuticals and Medical Biotechnology

Title ArrayExpress 
Description RNAseq data underpinning this paper entered into the ArrayExpress database under accession number E-MTAB-5174 
Type Of Material Database/Collection of data 
Year Produced 2017 
Provided To Others? Yes  
Impact N/A 
URL http://www.ebi.ac.uk/arrayexpress