Final Activity Report Summary - GENES AND GROWTH (Bacterial gene networks and the cell cycle)
Living cells need to be able to turn on and off the production of different proteins in response to changes in their internal and external environments. Proteins are produced from genes, and the expression of a particular gene can be regulated by protein 'transcription factors' which bind to the DNA and prevent or enhance gene expression. Genes may be regulated by complex networks of protein-protein and protein-DNA interactions involving positive or negative feedback loops. There has been much recent interest in the dynamical properties of these gene regulatory networks, and artificial networks have been constructed in the bacterium Escherichia coli which can function as oscillators or bistable switches. However, the bacterial cell is a highly complex environment. Not only is it densely packed with DNA and protein, but it is also strongly influenced by cell growth and division. E. coli is able to replicate itself as fast as every 20 minutes, and this is likely to have a profound effect on the physical environment inside the cell. Despite the obvious importance of such effects, no systematic study of the effect of growth on gene regulation has yet been carried out.
In this project, we have used a combination of computer simulations and wet lab experiments to predict the effects of changes in bacterial growth rate on various gene regulatory networks. The questions that we would like to address are: Are some regulatory networks more robust to changes in growth rate than others? What are the consequences of changes in growth rate for the functioning of different types of gene network?
On the simulation side, we have written codes to simulate a variety of different gene networks, including the effects of cell growth, DNA replication and cell division. This allows us to plot predicted gene expression, or the predicted 'noise' in gene expression, as a function of growth rate. We have produced both stochastic and deterministic versions of this code. Thus far, our results concern a 'constitutive' (unregulated) gene network and networks in which one gene represses another ('A represses B'), or one gene represses itself ('autorepressor').
On the experimental side, we have investigated different ways of changing the growth rate of E. coli and have started to build a chemostat. This is a vessel in which bacteria can be grown with constant inflow and outflow of nutrients, maintaining a steady state. We have made initial measurements using a bacterial strain previously used to measure 'noise' in gene expression. In this strain, two constitutive genes are inserted onto the chromosome, one encoding yellow fluorescent protein and the other cyan fluorescent protein. By measuring the correlations between yellow and cyan fluorescence intensities, one can deduce the contributions to the variation between cells ('noise') due to intrinsic fluctuations in gene expression, and due to external factors.
We plan to measure these contributions as a function of growth rate. Population heterogeneity among bacteria is very important in antibiotic and stress resistance, pathogenicity and biofilm formation, so these results should be industrially and clinically relevant as well as leading to fundamental understanding of the way that gene expression is affected by changes in the physiology of the cell.
In this project, we have used a combination of computer simulations and wet lab experiments to predict the effects of changes in bacterial growth rate on various gene regulatory networks. The questions that we would like to address are: Are some regulatory networks more robust to changes in growth rate than others? What are the consequences of changes in growth rate for the functioning of different types of gene network?
On the simulation side, we have written codes to simulate a variety of different gene networks, including the effects of cell growth, DNA replication and cell division. This allows us to plot predicted gene expression, or the predicted 'noise' in gene expression, as a function of growth rate. We have produced both stochastic and deterministic versions of this code. Thus far, our results concern a 'constitutive' (unregulated) gene network and networks in which one gene represses another ('A represses B'), or one gene represses itself ('autorepressor').
On the experimental side, we have investigated different ways of changing the growth rate of E. coli and have started to build a chemostat. This is a vessel in which bacteria can be grown with constant inflow and outflow of nutrients, maintaining a steady state. We have made initial measurements using a bacterial strain previously used to measure 'noise' in gene expression. In this strain, two constitutive genes are inserted onto the chromosome, one encoding yellow fluorescent protein and the other cyan fluorescent protein. By measuring the correlations between yellow and cyan fluorescence intensities, one can deduce the contributions to the variation between cells ('noise') due to intrinsic fluctuations in gene expression, and due to external factors.
We plan to measure these contributions as a function of growth rate. Population heterogeneity among bacteria is very important in antibiotic and stress resistance, pathogenicity and biofilm formation, so these results should be industrially and clinically relevant as well as leading to fundamental understanding of the way that gene expression is affected by changes in the physiology of the cell.