Form and function in the ESX-1 secretion system; elucidation of mechanism and structural determinants of bacterial virulence.

Almost all bacteria cause disease by producing toxins that they secrete into the host. Many of these toxins are protein molecules, and the bacterium has specialised machines in the cell membrane that allows the controlled passage of these toxins to the outside. We are working on one of these machines; the recently discovered ESX-1 system seeking to understand a central aspect of microbiology, namely how protein secretion machines work. This ESX-1 system has been shown to be essential for the virulence of important pathogens (M. tuberculosis, B. anthracis and S. aureus) and so there is a biomedical and agricultural aspect to this basic research. We have identified that the Staphylococcus aureus ESX-1 appears to be the most tractable for structure-function studies. S. aureus is a commensal Gram positive bacterium responsible for a number of illnesses in humans and animals that range from minor skin infections, such as pimples and abscesses, Toxic shock syndrome (TSS) through to septicaemia and mastitis in dairy herds and pigs. However, it is most widely known as a major cause of hospital acquired infections, and is a frequent cause of post-surgical wound infections. This situation is exacerbated by the fact that some strains of S. aureus are resistant to many antibiotics (e.g. Methicillin Resistant Staphylococcus aureus), making it a severe and difficult to treat problem. The number of deaths attributed to S. aureus infection is comparable to that attributed to acquired immune deficiency syndrome. An understanding, at the molecular level, of S. aureus biology and pathogenesis is essential if we are to design new treatments to prevent or cure infection. The ESX-1 transport system drives the secretion of at least three different proteins from S. aureus, and it is known that when this system is inactivated S. aureus shows a dramatic reduction in its ability to cause infection. We seek knowledge of the architecture and function of this distinctive bacterial secretion machine and identification of its cargo. The ESX-1 system consists of 5-8 different protein components. We will investigate how these proteins interact to assemble the secretion machine, and the molecular basis for how the machine works. To attain this goal we must derive the structure and function of each component, elucidate how the components interact and quantify the association, determine the molecular basis for cargo recognition, and the generation of the motive force. We have already made significant progress with six of these proteins cloned, expressed and purified and two crystallised. We will exploit single crystal X-ray diffraction methods to derive accurate molecular structures and a battery of biophysical techniques to characterise the protein-protein interactions. Knowing the structures of the protein components and how they interact with each other is important because this might, in the longer term, lay the foundation for studies directed to the design or discovery of compounds that will prevent these proteins from working with each other or prevent the motive force from being used to secrete out the proteins that establish and prolong infection. Information on proteins that are secreted and of the structures found on the surface of the bacterium may also provide opportunities for vaccine design.

The ESX-1 protein secretion system is a general protein export pathway found in many Gram positive bacteria including the serious pathogens Mycobacterium tuberculosis and Staphylococcus aureus. Previous studies have shown that this secretion pathway is essential for the virulence of both of these organisms. Indeed the loss of the ESX-1 system from a strain of M. bovis is the basis for the attenuated BCG vaccine. Although the ESX-1 export pathway shares a number of common components between bacterial species there are clear differences in the composition of the machinery between Actinobacteria (including M. tuberculosis) and firmicutes (S. aureus). For the past two years we have been working on the characterisation of the ESX-1 secretion system from S. aureus. Of the eight components that are encoded at the esx-1 locus, we have overproduced and isolated six of these and have diffracting crystals for two of them. We have also raised antisera to these proteins and have preliminary data about pair-wise interactions between several of these components. We now seek to build on our excellent preliminary data to obtain high-resolution crystal structures for the proteins we have crystallised. We will optimise our purification strategies en route to obtaining diffracting crystals of the other components and of relevant protein-protein complexes. In addition we will probe the organisation of the complex in S. aureus using cross-linking and native level protein purification coupled with yeast two-hybrid studies and co-purification approaches. To this we will add quantification of the protein-protein interactions important in the assembly of the secretion machine and in recognition of its cargo. Our multi-disciplinary approach will advance knowledge of important, fundamental aspects of protein structure and function, protein-protein interactions, protein transport across a membrane, bacterial virulence and pathogenesis.

Aacademic researchers with interests in protein secretion, structure and function of multiprotein complexes and bacterial cell biology will benefit from the data, and understanding that our research will produce. Microbiologists with particular interests in Gram positive bacteria, in particularly the S. aureus, Bacillus, Listeria, Mycobacterium and Corynebacterium communities will be informed by our work. This study also has the potential to benefit researchers in wider disciplines such as bioengineering, drug discovery and vaccine development. A final academic benefit to arise from this work will be in the generation of new strains of S. aureus and antisera to components of the secretion machinery. Strains and antibodies will be archived by the University of Dundee and we will make these reagents available to the research community in a timely manner following publication of our findings. Our research findings will primarily be disseminated through publication in peer-review journals and at national and international scientific meetings. In the longer term the knowledge generated from our research has the potential to impact on major human health problems - infection by S. aureus, which today is responsible for a comparable number of deaths as HIV-AIDS in North America,1 infection by Mycobacteria (tuberculosis and leprosy). Staphylococcal infections are also of veterinary importance affecting milk and animal production in cattle and swine. Therefore, academics and/or industrial groups working on new drugs, vaccine development or improved diagnostic methods to promptly identify the type of infection would gain useful information for their work. The structures and reagents that we generate would ultimately also provide opportunities for design of inhibitors that prevent the action of this important virulence determinant, and any surface exposed proteins of the machinery would make excellent vaccine candidates. We will, through the Research and Innovations Office at the University of Dundee seek to make links with academics interested in bioengineering and vaccine development as we progress with our study. We will also seek to exploit any opportunities to interact with companies, large or small who wish to exploit our experience, data and reagents. The Research and Innovations office in Dundee have an excellent record of forging links with life sciences and pharmaceutical companies.