Environmentally relevant responses in different Escherichia coli pathotypes: a functional genomics study of motility and associated regulons

In nature bacteria may act as friends, like 'good' bacteria in our digestive systems, or foes, sometimes as life-threatening pathogens. Escherichia coli can be either friend or foe, and is also a 'model' bacterium, much studied to understand bacterial structure and function. This knowledge is based mainly on a harmless strain named K-12, isolated in 1922. Today, K-12 is so well adapted to the laboratory that it cannot even survive in the human digestive tract. But there are other distinctive 'pathotypes' of E. coli that can cause disease, including E. coli O157. Complete genomic DNA sequences have been determined for K-12 and several other pathotypes. In each, about 20% of its genes are unique, a big surprise since it had been expected that E. coli as a species would be genetically quite homogeneous. So, is K-12 a good model representing the species as a whole? To find out, we will analyse and compare the ways K-12 and other pathotypes respond genetically to conditions that induce them to change their behaviour, focusing on their ability to 'swim' in liquids (motility) and related properties that enable them to survive and persist in the environment. New 'functional genomics' techniques, based on knowledge of genome sequences, enable the expression of every gene in the genome to be monitored simultaneously. This shows how each gene is involved in the behaviour of the bacteria under given conditions. It would be tempting to use these techniques just to fill the gaps in our knowledge of how K-12 functions. But this would neglect the potential to analyse different pathotypes, of known genome sequence, in relation to their behaviour in nature. We propose to exploit that potential to the full. Some transcription networks in E. coli, comprising subsets of genes ('regulons') controlled by proteins called transcription factors or TFs, have received much attention due to their involvement in the basic biochemistry and physiology of the cell. This has been echoed in the choice of networks that have been analysed by functional genomics methods. For example, 3 independent such studies of the FNR regulon involved in adaptation to the absence of oxygen have already been published. In contrast, there has been far less experimental investigation of the regulon that controls motility, governed by the TF complex FlhDC. We believe this TF forms a significant node or 'crossing point' in the global cellular transcription network, interacting with, among others, those that mediate responses to the presence of metals and to oxidative stress. The FlhDC regulon also responds to an important and little-explored signal molecule made within the cell, called cyclic di-GMP. This molecule is known to be involved in the ability of E. coli and related species to colonise solid surfaces in a growth mode known as a 'biofilm' in which motility is strongly repressed. We will use functional genomics methods to analyse the genome-wide transcription activity controlled by FlhDC and associated TFs named Fur and SoxS. We will impose conditions in which the TFs would normally be activated, and monitor global changes in gene expression in representative pathotypes. In parallel experiments we will study gene expression in mutants in which the TFs have been deleted. We also propose to use new methods to study TF binding to different regions of DNA, modulating expression of the genes in those regions. This will provide a definitive picture of how the regulons respond to stimuli and what cellular functions they then modulate to allow the cell to adapt to environmental challenges. We will thus gain exciting insights into the diversity of responses mediated by FlhDC and related TFs in different pathotypes of this uniquely important bacterium, and generate for the first time a global comparison of patterns of gene expression and their control among different representatives of the same bacterial species.

Known genome sequences of multiple E. coli pathotypes are a unique resource for investigation of gene regulation in biologically distinctive relatives, by the powerful methods of functional genomics. We will compare a carefully selected set of pathotypes using phenotypic and functional genomics methods, focusing on regulons significant in environmental survival and persistence. We will thus comprehend the diversity of function of transcription networks within 'wild' members of the species, and gain a clear view of the relevance of the K-12 lab model as a paradigm for the species as a whole. Among less investigated transcription networks relevant to survival in nature is the FlhDC regulon, named for its control of motility but increasingly implicated in other cellular responses, for example to metals such as copper and to oxidative stress, through cross talk with the Fur and SoxS regulons among others. It is also a key target for modulation by the cellular signal molecule cyclic-di-GMP. This signal is intimately involved in the transition from planktonic phenotypes (swimming) to biofilm (non-flagellar e.g. twitching motility) in many bacteria including wild-type E. coli, a phenomenon also favoured at ambient temperatures. We propose that FlhDC and associated regulons have a key role in less explored aspects of adaptation of diverse wild E. coli pathotypes to their natural environments. We will define both the broad phenotype, using Biolog phenotypic arrays, and the FlhDC, Fur and SoxS regulons, in selected pathotypes (ETEC, EHEC, UPEC) representing the breadth of the species and in K-12, both by conventional transcriptomics and by ChIP-on-chip analysis. We will analyse and compare global gene expression in wild types and mutants in these transcription factors, with and without exposure to cognate stimuli. Thus we will advance understanding of the regulation of motility and related behaviours of E. coli in nature.