Phase variable epigenetic control in firmicutes

All the cells of a species contain the same DNA; what distinguishes them is the way in which that DNA is activated ('transcribed'). One method of regulating DNA transcription is through direct chemical modification of the DNA itself, most commonly through a process of methylation. This is termed 'epigenetic' regulation. There has been extensive study of epigenetic regulation in humans and other complex life forms, but comparatively little in bacteria. Bacterial genomes are often extensively methylated, as a consequence of 'restriction modification' (RM) systems, which modify the cell's DNA at particular sites to allow it to be distinguished from the DNA of infecting viruses. The many sites of methylation across the genome have the potential to substantially affect the way in which genes are regulated. However, as most RM systems are stable and therefore cannot serve as a regulatory mechanism. This is not the case for two sets of genes we have recently independently characterised in the bacterium Streptococcus pneumoniae, the pneumococcus. The pneumococcus is a commensal bacterium, typically carried by between a quarter and a third of young children asymptomatically, that is a major cause of diseases including middle ear infections, pneumonia and meningitis. The sets of genes we found are RM systems that vary over the course of hours or days through a specific set of DNA rearrangements. This results in the patterns of methylation caused by the RM systems also changing over short timescales. Experimental data found that different forms of the inverting RM system caused different patterns of methylation, each of which was associated with a distinct pattern of gene expression. This epigenetic regulation of bacterial genes was found to change the virulence of the bacterium, with some patterns of methylation making the pneumococcus more likely to cause disease. This could be an important factor in the transition from the pneumococcus being a harmless commensal, to becoming a dangerous pathogen. This project is designed to test this hypothesis through studying whether the second variable RM system in the pneumococcus affects the same processes, or has a different effect on cell physiology, and whether such systems regulate the virulence of other pathogenic bacteria. Searching of the thousands of publically available bacterial DNA sequences has allowed us to identify hundreds of species that harbour similar systems. These include bacterial species that are very common in the human gut, some that are present in probiotic drinks and others involved in the production of cheese. Perhaps most importantly, they are also present in many pathogenic bacteria. This project is designed to investigate whether these variable RM systems might also regulate virulence in three bacterial species that each represent major threats to public health. The first is Streptococcus suis, a species normally associated with pigs that is emerging as a major pathogen capable of causing serious infections, such as meningitis. The second is Listeria monocytogenes, a foodborne bacterium that causes potentially fatal infections. The third is Enterococcus faecalis, a major cause of highly antibiotic-resistant infections, particularly in a hospital setting. In the three species, the variable RM loci are present with lineages that are associated with causing high levels of disease in humans, and absent from those that are asymptomatically found in animals or humans. The overall aim of the project is to work out how these systems may play a role in regulating genes involved in the bacteria's virulence, as well as how they evolved and how diverse they are. Such information will allow us to understand why these unusual genes are distributed, and why bacteria progress from being harmlessly carried to causing disease. This would better inform our strategies as to how to prevent this transition, and thereby tame these common, but potentially dangerous, bacteria.

Two distinct phase variable type I restriction modification systems, able to change their specificity through shuffling of sequences facilitated by a recombinase, have been described by us in the pathogen Streptococcus pneumoniae in two independent papers (Manso et al., Nat Commun. 2014, 5:5055; Croucher et al., Nat Commun. 2014. 5:5471). One of these was found to be responsible for phase variable epigenetic modification of the microbial chromosome with impact on gene expression and virulence. Both loci are absent from the related non-pathogenic species, S. mitis. We now have mined available genome information and detected four families of recombinases associated with Type I RM loci across many species, all of which are predicted to be able to vary the patterns of methylation over short timescales. Although phase variable RM loci can be found in many phyla, they are particularly highly represented among Firmicutes. In particular we have collected detailed information on likely phase variable systems in Listeria monocytogenes, Enterococcus faecalis and Streptococcus suis. In all three cases the lineage or clonal complex most frequently found to cause human invasive infection is associated with the presence of a phase variable Type I RM locus. The aim of the project is to analyse the overall evolution of these systems, and then isolate clones expressing different allelic variants in the four species of interest on which to perform quantification of phase variation, methylome analysis, gene expression profiling associated to metabolome assays and virulence tests in experimental infection models. These tests are designed understand the biological mechanism underpinning the epigenetic regulation of important phenotypes. Demonstration of phase variable epigenetic control in these four species is predicted to allow us to propose phase variable epigenetic control by type I RM systems as global and novel paradigm of bacterial evolution, physiology and virulence.

This research project is designed to test if the novel epigenetic control mechanism discovered in Streptococcus pneumoniae is an unicum of this species, as published independently by us in two high impact papers (Manso et al., Nat Commun. 2014, 5:5055; Croucher et al., Nat Commun. 2014. 5:5471), or if holds true to be a paradigm of epigenetic control for a wide number of bacteria. We have clear preliminary data that show that these phase variable type RM systems are broadly distributed, and likely to be active, in many bacterial phyla.

One of the main beneficiaries is the wider public, since the project aims to unravel basic mechanisms involved in bacterial population dynamics and more importantly showing that phase variable epigenetic control is widespread also in bacteria. In the specific case we address four bacterial species of substantial interest to the BBSRC, industry and health care professionals.

Interest in food safety is a primary goal for all players in the long chain and the species Listeria monocytogenes has been selected as an important food borne pathogen. Data on phase variable RM systems in Listeria are hoped to yield novel epidemiological markers, markers associated to disease severity but also shed light on the virulence of this species. The concept linking presence of phase variable RM systems to highly pathogenic lineages will be tested in also S. suis, a zoonotic pathogen and first cause of human meningitis in South-east Asia. In this species the clear association of the phase variable type I RM system to serotype 2 strains, could well turn out to be one of the important virulence determinant.

The UK has recently launched the 5 Year Antimicrobial Resistance Strategy 2013 to 2018. One of the important aspects in this strategy is to limit drug resistance to develop, but more importantly to spread. Of particular concern is the fact that many more antibiotics are used for industrial animal husbandry than directly for humans, which in turn has driven an alarming rise in antimicrobial drug resistance in bacteria carried by animals. Hence the final model species included in this project is Enterococcus faecalis, among the main candidates responsible for transfer of antimicrobial drug resistance from animals to humans.

The widespread nature of these gene modules will allow transferring information more widely, as these gene clusters are found, in the two main phyla found within the human intestinal tract, in many probiotic and dairy species. Given the clear preliminary experimental data and the broad range of bacteria being investigated, this scientifically novel project is highly relevant to the study of bacteria globally. The industrial or health related impact may not be immediate, but given the above considerations we envisage potentially positive measurable effects in terms of public health and the field of food safety and production within few years after the project's completion.

The interdisciplinary and multi-centre nature of this project means it will be effective in training researchers in a wide range of skills. They will learn the latest molecular microbiology techniques, how to analyse high-throughput sequencing data, mine available genome sequences, and exploit newly-available sequencing technologies. Collaborations with the broader community of researchers, as outlined in our proposal, will provide for interactions with clinicians and public health experts. Other potential beneficiaries include the wider audience of students, academics and public addressed by our planned outreach activities.

Discovering novel regulatory mechanisms, which may have the potential to be valid for many species holds promise to identify potential drug targets and inhibitors. This will not be addressed in depth within the project, but throughout the research we expect to get critical scientific information that may need protection and consideration for commercialisation.