NICHE · Network for Integrated Cellular Homeostasis

Network for Integrated Cellular Homeostasis

Cellular metabolism takes place in a complex and dynamic environment in which enzymes perform at highly crowded conditions, low water activity and relatively constant ion concentrations (pH, ionic strength). The mechanisms underlying these homeostatic processes are crucial for the performance of the cell, and quantitative understanding of the physicochemistry of the cell is required for any rationale engineering of a metabolic pathway. NICHE aims at a comprehensive understanding of homeostatic mechanisms that control the volume, internal pH and ion concentrations of the cell. The major objectives of WP1 and WP4 were to develop methods to determine the position and dynamics of proteins inside live Escherichia coli cells. State-of-the-art microscopy-based methods have been employed to determine the locations of molecules inside the cell and cellular membranes, and single molecule tracking has been used to probe the traffic of molecules. Fluorescent proteins with high photostability and high molecular brightness (for single molecule tracking) and photochemical switching and fast maturation (for super-resolution microscopy) are used and variants with improved properties have been developed. Mechanosensitive channels are important regulators of cell turgor volume by acting as osmolyte release valves. The Groningen (partner 1) and Aberdeen (partner 3) groups have used super-resolution optical microscopy to determine the dynamics, localization and cluster formation of the major mechanosensitive channels MscS, MscL and MscK in Escherichia coli. We show that protein abundance and number of fluorescent-labeled subunits per channel impact clustering of MscL but do not affect MscS and MscK. Clustering affects the mobility of MscL in the membrane but the proteins remain functional and display wild type conductance and tension dependence. We furthermore present a quantitative relationship between MS channel number and cell survival under osmotic downshock conditions. Our results contribute to outstanding questions on how bacteria regulate the activity and distribution of MS channels. ATP-Binding Cassette (ABC) transporters play crucial roles in cellular processes such as nutrient uptake, drug resistance, cell-volume regulation and others. Despite their importance, all proposed molecular models for transport are based on indirect evidence, i.e., functional interpretation of static crystal structures and ensemble measurements of function and structure. Thus, classical biophysical and biochemical techniques do not readily visualize dynamic structural changes. The Groningen group together with associated partner Cordes have used single-molecule fluorescence techniques to study conformational states and changes of ABC transporters in vitro, in order to observe directly how the different steps during transport are coordinated. The conformational states and transitions of ABC-associated substrate-binding domains (SBDs) were visualized with single-molecule Förster resonance energy transfer, permitting a direct correlation of structural and kinetic information of SBDs. We also delineated the different steps of the transport cycle. The major objective of WP2 is to determine the effect of homeostatic perturbations on gene expression. For this we use massively parallel deep-sequencing approaches to determine the copy numbers of each mRNA, the secondary structure of the whole cellular mRNAome and the position of translating ribosomes on each mRNA. Another parameter we determine on the cell-wide global level is the tRNA concentration and its charging and provided training to the ESRs and ERs on these methods. The comprehensive information our whole-cell analysis delivered was integrated in several mathematical models (WP5). Our data (1) provide the first genome-wide map of reallocation of translation resources to maintain osmotic stress or when simultaneously facing multiple forms of stress and (2) highlight the contribution of mRNA structure in as a direct effector of a variety of processes, including translation and mRNA degradation. WP3 aims to understand the mechanisms regulating homeostatic K+ transport complexes by determining their molecular mechanisms. We focussed on the Kef system. We created an assay system to provide quantitative measurements of ligand binding and this data has been written into 2 papers for publication, one published 2014. Partner 3 also identified, in collaboration with Partner 6 (Oxford), a novel nucleotide-binding site, which confers structural stability on the system and we investigated the role of residues, which bind this ligand (paper in preparation). Mathematical models are a key component in systems biology, and their interplay with experiments allowed us to generate new testable hypothesis regarding cellular processes. The main objectives in WP3 and WP5 are to develop model building and analysis techniques and to apply them to model the dynamics of potassium homeostasis in E. coli. Partner 5 has developed novel optimisation-based techniques, which allow a more efficient model-building loop, from model inference to optimal experimental design and parameter estimation. We have also incorporated regularization techniques in order to obtain more predictive models (reducing over-fitting). Regarding the dynamic modelling of homeostasis (WP3) and macromolecular diffusion in the cytoplasm (WP1), we have applied the novel model development methods developed in WP5 to the system regulating potassium homeostasis in E. coli. Partner 2 and 4 have refined statistical and modeling tools to be applied to the analysis of proteins and mRNA stability. The final aim of this analysis approach is to understand how proteins and mRNA expression levels are regulated through their degradation. The analysis of massively parallel deep-sequencing techniques for ribosomes has also highlighted a high frequency of unspecific ribosome drop-off, which is clearly enhanced under acute stress conditions A mathematical model of tumor growth of partner 8 incorporates both the various populations of cells, tumor vs stroma, in different points of the cell cycle and spatial effects of nutrient and drug delivery (WP1). Attention was paid to how phenotypic and drug effect biomarkers can support the parameter estimation of these models when applied to data from in vivo xenograft experiments. The tumor growth model (WP5) has been expanded from a single tumor mass surrounded by vasculature to clusters of tumour cells. This has been validated against experimental data generated at AstraZeneca using deterministic ODE approaches and stochastic models. Significant progress has also been made in image analysis of IHC slides to inform the parameterization of the model. Plans are being made to extend the model to incorporate drug effects to support drug discovery projects. Research on a photometric method to measure tumour volume in animal systems has demonstrated the feasibility of this approach and partner 8 has entered into a development agreement with a vendor of such a device to make the technique useable across animal xenograft studies. This has the potential to refine and reduce the usage of animals in such experiments. All results have been published in peer review journals. Website: https://www.chemie.uni-hamburg.de/niche/index.html Contact details: Prof. Dr. B. Poolman Faculty of Mathematics and Natural Sciences, Gron. Inst. Biomolecular Sciences & Biotechnology - Enzymology Nijenborgh 4 9747 AG Groningen The Netherlands

Data: CORDIS, © European Union