Cell surface display of bacterial proteins

Lead Research Organisation: University of Sheffield
Department Name: Molecular Biology and Biotechnology

Abstract

The proteins exposed at the cell surface of bacteria play a pivotal role in a multitude of fundamental processes such as cell growth and division. In pathogens, surface proteins promote host invasion and later during infection, evasion from the host immune response. Most surface proteins are anchored to a scaffold molecule called peptidoglycan, an essential component of the bacterial envelope. Peptidoglycan is a bag-shaped giant molecule surrounding the cell that made plays a protective role. Its synthesis is unique to bacteria and represents the target of the most important antibiotics ever discovered such as penicillin. To date, a plethora of surface proteins has been described but how they interact with the bacterial envelope remain poorly understood. We have recently described the molecular basis for peptidoglycan recognition by a ubiquitous protein domain called LysM in vitro using synthetic ligands. We are now ready to take the next step to actually understand how LysM proteins bind to the complex peptidoglycan molecule in live bacteria and are targeted to specific subcellular localizations. To achieve this aim, we will use a multidisciplinary approach combining bacterial genetics, state-of-the-art microscopy and biophysics.
Binding of LysM domains to the PG molecule represents a key bacterial mechanism for controlling the location, synthesis, degradation and decoration of cell wall features. The understanding generated by this project will give us the tools to dissect how bacteria grow, divide and interact with their environment, to develop new therapeutic strategies, and to learn how to apply synthetic biology to bacterial cell walls.

Technical Summary

The proteins displayed at the bacterial surface play key roles in fundamental processes such as cell growth and division. In pathogens, surface proteins promote host invasion and evasion from the host immune response. Core to all these functions is the ability of the bacterial proteins to bind to their own cell wall and thus be appropriately displayed for interaction with the environment. This project will focus on a ubiquitous domain (LysM) that mediates non-covalent binding of proteins to peptidoglycan (PG), the essential component of the bacterial cell wall which is the target of most important antibiotics ever discovered, penicillin. I recently elucidated the molecular basis underpinning LysM-PG interaction in vitro using the multimodular LysM domain from a PG hydrolase in the opportunistic pathogen Enterococcus faecalis.
Our unpublished data indicate that LysM proteins are targeted to distinct sites at the cell surface. The aim of this project is to understand how this is achieved. We will undertake a multidisciplinary approach combining microbial genetics and super-resolution microscopy to study the subcellular targeting of LysM-GFP fusions in live bacteria. We will explore the impact of LysM domains properties (e.g., modularity, charge), secretion signals and cellular factors (membrane potential and PG-associated polymers) to the subcellular localization of proteins. We will also investigate the functional diversity of naturally occurring LysM domains to understand how PG structure modulates binding. Finally, we will characterize the role of newly identified components in LysM-mediated surface display.
This project will show how different bacteria use their LysM domains to recognise and target different features of peptidoglycan. Enzymes linked to LysM are used by bacteria to remodel the cell wall, build and destroy septa, and in pathogenesis. The project will produce a toolkit to map bacterial surface features, and target novel features for drug design.

Planned Impact

The aim of this project is to understand how ubiquitous LysM domains mediate non-covalent cell surface display of proteins in bacteria. This project will show how different bacteria use their LysM domains to recognise and target different features of peptidoglycan. Enzymes linked to LysM are used by bacteria to remodel the cell wall, build and destroy septa, and in pathogenesis. The project will produce a toolkit to map bacterial surface features, and target novel features for drug design, an area of research of immense public concern. There will be a variety of impacts over a range of timescales and in different arenas.
(i) Impact on academic researchers (short to medium term). The work will directly benefit UK and international microbiologists. Due to the multidisciplinary nature of the project, we also expect this work to benefit a wide range of researchers including structural biologists, biophysicists, cell biologists and scientists working on the development of novel antimicrobials. We will communicate our findings through open access publications in peer-reviewed journals, presentations at symposia, seminars, and national/international conferences. All the protocols and biological materials generated during this work will be made available through the PI's website that will be regularly updated in a timely manner.
(ii) Impact on RA and technician employed (short term). Both RA and technician will acquire a wide range of skills to improve their employability, thereby having a direct impact on the knowledge-led economy of UK Plc. The skills will be transferred to students working on related projects, as well as to visiting scientists. The RA will also benefit from the support offered through the action plan for continued implementation of the Concordat to support the career development of researchers. The RA will be encouraged to undertake training offered for personal professional development by the Researcher Personal Development Team at the University of Sheffield.
(iii) Biotechnology and pharmaceutical companies developing and commercialising antimicrobials (short to medium term) -translational fusions between acidic/basic LysM motifs and the catalytic domain of colicin M wich cleaves peptidoglycan lipid precursors at the outer face of the cytoplasmic membrane will provide the proof of concept that LysM domains can be used as a means to target therapeutic molecules to pathogens. We will have a strong commitment to secure any intellectual property (IP) associated with this work and maximise the opportunities to exploit translation of research findings into commercial applications. Once IP has been secured, we will consider the opportunity to create a spin-out company with the support of the Commercialisation team at the University of Sheffield or licensing out. This project will provide the opportunity to initiate collaborations with established companies to seek funding in partnership with the support of Sheffield Science Gateway which offers several schemes. A contact has been established with Lisando GmbH to explore potential partnership.
(iv) The public, as potential users of the research (short to long term). The expected impact of this project (long term) is to increase the effectiveness of the health service offered by the NHS and enhance the quality of life and health in the UK and worldwide. This will be achieved through commercialisation of novel therapeutic molecules generated, with the PI and RA also being involved in various outreach activities (school visits and promotion of our research during work and family focussed scientific events).
 
Description We have made progress in the following research areas:

1) LysM -mediated cell surface display in live bacteria underpins specific subcellular localisation of proteins.
We have built a series of recombinant Enterococcus faecalis strains combining various targeting domains (LysM and signal peptides) fused to the GFP to identify their respective contribution to protein surface display. We have established that the combination of both LysM domains and the signal peptide of surface proteins are required to target proteins to specific subcellular compartments. In the case of E. faecalis AtlA glucosaminidase, both LysM and the secretion signal are required to target this enzyme to the cell division site where it can hydrolyse peptidoglycan and release the daughter cells.

2) LysM subcellular localisation of E. faecalis AtlA is required to maintain cell chain size and is essential for virulence
We have shown that in E. faecalis, the activity of the enzyme playing a major role in peptidoglycan hydrolysis for cell separation at the end of binary fission requires a complete LysM domains. In collaboration with Pr Stephen Renshaw and Pr David Dockrell, we showed that the formation of bacterial short chains is essential for virulence.

3) Characterisation of a novel locus required for LysM surface display.
We previously identified a gene (called admA) required for the surface display of the LysM protein AtlA. We have complemented this mutation and formally establish that this gene is recruiting the LysM domain to the bacterial septum using a split luciferase assay in Enterococcus faecalis. We are currently investigating the function of AdmA by X-ray crystallography and ITC, testing various potential binding partners.

4) Development of novel methods to study protein-peptidoglycan interactions
(i) We have set up a flow cytometry assay to quantify single cell binding of LysM to bacteria that will be potentially applicable to any bacterial surface protein. This assay will allow us to explore the binding specificity of LysM domains. This will be applied to a set of LysM domains found in bacteria with various peptidoglycan structures. 32 LysM domains have been chosen for expression and purification. A subset of 10 of them (including basic and acidic domains) will be further analysed.
(ii) In collaboration with proteinmetrics, a US company specialised in mass spectrometry softwares, we have developed a pipeline for the automated analysis of PG fragments. This has allowed us to identify and purify peptidoglycan fragments that we can use as ligands to study how LysM domains bind to peptidoglycan.
Exploitation Route The findings reported above are important (i) to develop peptidoglycan binding domains as a means to target therapeutic molecules to pathogens, (ii) to develop these domains as fluorescent probes to study peptidoglycan topology and (iii) to use LysM domains as a cell surface display system in bacteria to generate new live vaccines or bioremediation tools (iv) to design novel antimicrobial strategies against enterococcal infections.

During this work, we have provided the proof of concept that peptidoglycan fragments can be analysed by shotgun proteomics softwares following LC-MS/MS. This discovery, not initially identified as an objective of the project has led to 2 publications and is expected to have a major impact on structural analysis of peptidoglycan.

Two important concept have emerged from our work:
1) Peptidoglycan binding domains such as LysM are used as a targeting signal in the cytoplasm to target proteins to specific subcellular localisations; this mechanism is likely to be conserved in other bacteria
2) Despite their highly conserved fold, LysM domains can display distinct affinities for peptidoglycan and recognise different peptidoglycan structures with distinct affinities.
Sectors Healthcare,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology

 
Description BBSRC iCase - Development of an automated, high-resolution analysis pipeline for bacterial peptidoglycan structural analyses
Amount £110,000 (GBP)
Funding ID 2058718 
Organisation Biotechnology and Biological Sciences Research Council (BBSRC) 
Sector Public
Country United Kingdom
Start 09/2017 
End 09/2021
 
Description UK-France International exchange scheme
Amount £7,800 (GBP)
Funding ID IEC\R2\170260 
Organisation The Royal Society 
Sector Charity/Non Profit
Country United Kingdom
Start 02/2018 
End 01/2020
 
Title Automated analysis of LC-MS/MS PG fragments 
Description We have reported a novel strategy using shotgun proteomics techniques for a systematic and unbiased structural analysis of peptidoglycan fragments using high-resolution mass spectrometry and automated analysis of HCD and ETD fragmentation spectra with the Byonic software. 
Type Of Material Technology assay or reagent 
Year Produced 2017 
Provided To Others? Yes  
Impact The strategy described allows a non-biased, ultra sensitive and rapid analysis of peptidoglycan LC-MS/MS data. We have published the proof of concept of this approach and demonstrated that it has the potential to detect an unexpected diversity of peptidoglycan structures. This will allow to elucidate the role of a large number of enzymes involved in peptidoglycan remodelling previously unknown. Thanks to the completeness of the method developed, we can now envisage to study in great details how peptidoglycan structure changes following treatment by cell wall targeting antibiotics or during pathogenesis. Thanks to this breakthrough, we can now envisage studying very dilute peptidoglycan samples (for example, from individual abscesses or in individual organs from infected animals such as mice). This is expected to reduce the number of animals required to carry out such studies. 
 
Title Flow cytometry assay to quantify single-cell binding of proteins to bacterial cell surfaces 
Description Using the high throughput flow cytometer purchased with the funding, we have developed an assay which allows quantification of protein binding to bacterial cells. We have now set up the proof of concept with using a single LysM peptidoglycan binding domain (LysMA1) as a model. This assay invloves: - LysM labelling; covalent FITC or a translational fusion with a fluorescent protein can be used - incubation in the presence of cells - measurement of fluorescence associated with individual cells by flow cytometry We have defined the conditions (protein concentrations, cell densities, buffer conditions, flow cytometry parameters) which give a binding dose-response to saturation. 
Type Of Material Technology assay or reagent 
Year Produced 2018 
Provided To Others? No  
Impact This assay is instrumental to quantify protein-peptidoglycan interactions. It will allow us to measure the impact of peptidoglycan structure and cell wall components on binding (using isogenic mutants as substrates). We will also be able to compare the binding affinity of distinct LysM domains to the same cells. Beyond this project, this represents a technique widely applicable to any surface protein. 
 
Description Analysis of E. faecalis pathogenesis using the zebrafish experimental model of infection 
Organisation University of Sheffield
Department Department of Infection, Immunity and Cardiovascular Disease
Country United Kingdom 
Sector Academic/University 
PI Contribution We have built several Enterococcus faecalis mutants with mutations impairing the cell surface display of the major autolysin involved in septum cleavage. THe impact of these mutations, resulting in an increaded cell chain length has been tested during zebrafish infection.
Collaborator Contribution The group led by Prof Stephen Renshaw has developed zebrafish as an experimental model of infection. This gave us access to the methodology required to follow phagocytosis of Enterococccus faecalis cells during the infection process and to measure lethality. This led to a key finding published in Salamaga et al. (2017)
Impact Salamaga et al., PLoS pathogens, 2017
Start Year 2016
 
Description Biophysical analysis of LysM modularity 
Organisation Research Complex at Harwell
Country United Kingdom 
Sector Public 
PI Contribution We have purified LysM domains made of variable numbers of binding modules to explore their flexibility by Small Angle X-ray Scattering (SAXS) and analytical ultracentrifugation
Collaborator Contribution Dr David Scott (Research Complex at Harwell/University of Nottingham) has carried out Small Angle X-ray Scattering (SAXS) and analytical ultracentrifugation experiments and analysed them.
Impact Combined to our NMR results, these SAXS, AUC and molecular dynamics simulations indicate that LysM domain presents an intrinsic flexibility.
Start Year 2016
 
Description LysM-peptidoglycan interactions analysis by NMR 
Organisation University of Sheffield
Department Department of Physics and Astronomy
Country United Kingdom 
Sector Academic/University 
PI Contribution This collaboration is aimed at characterizing the molecular basis underpinning LysM-peptidoglycan interaction.
Collaborator Contribution Mike Williamson is studying the 3D structure of LysM domains by NMR and how they interact with peptidoglycan fragments.
Impact We have analysed the relaxation constants (T1/T2) of our model LysM domain (from E. faealis AtlA autolysin) and shown that this domain is flexible, with no detectable quaternary structure formed by this multimodular domain.
Start Year 2016
 
Description Structural analysis of LysM-peptidoglycan interactions 
Organisation University of Birmingham
Country United Kingdom 
Sector Academic/University 
PI Contribution We have purified model LysM that were provided alongside with synthetic ligands (chitooligosaccharides) that we provided to our collaborator to determine co-crystal structures of these interaction partners.
Collaborator Contribution The laboratory of Dr Andrew Lovering has expertise in structural biology (X-ray crystallography). His team has
Impact The laboratory of Dr Andrew Lovering has performed crystallogenesis trials and solved the high resolution structure (1.05 Angstroms) of one of our model LysM domains in its apo form. This information will be critical to analyse our NMR titrations carried out with this LysM domain.
Start Year 2017
 
Description Structural characterization of LysM-peptidoglycan interactions 
Organisation National Center for Scientific Research (Centre National de la Recherche Scientifique CNRS)
Department Architecture and Function of Biological Macromolecules (AFMB)
Country France 
Sector Academic/University 
PI Contribution My lab has expressed and purified recombinant LysM proteins to determine their structure in complex with chitooligosaccharides and peptidoglycan fragments.
Collaborator Contribution Our partner has solved the structure of a co-crystal between our model LysM domain and a tetrachitooligosaccharide (GlcNAc)4
Impact Our collaborators are structural biologists.
Start Year 2016
 
Title BYOS (suite of mass spectrometry software customized for automated peptidoglycan analysis) 
Description In collaboration with ProteinMetrics (https://www.proteinmetrics.com/), we are trying to customize available software solutions for peptidoglycan analysis. This has been facilitated by a BBSRC iCASE studentship (2017-2021). The source code has been modified to improve both the annotation of spectra and scoring method of peptidoglycan fragments (2017-2018). We are currently working towards a complete automated process whereby several software solutions are combined to provide a comprehensive analysis of MS1 and MS2 data as well as a quantification of peptidoglycan fragments. Since October 2018, we have generated a dedicated software which can predict fragmentation products from any peptidoglycan fragment using a graphic representation of molecules. This is being tested with Clostridium difficile peptidoglycan as a model system. 
Type Of Technology Software 
Year Produced 2019 
Impact The use of shotgun proteomics is a novel approach allowing an unbiased analysis of peptidoglycan structure. The automated approach we are developping has opened the possibility to carry out relatively high-throughput peptidoglycan analyses, currently impossible due to the labour-intensive methods available. 
 
Description KrebsFest: Exploring hidden worlds at Shambala festival (August 2016) & Cheltenham Science Festival (June 2017) 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact Presenter on University of Sheffield outreach event KrebsFest on tour. Intended purpose of the event is to raise awareness of the research happening at the University of Sheffield. This was through demonstrations (crystalalography & microscopy) and speaking about our specific research projects. A wide audience was reached through KrebsFest on tour at two different locations. The first, at Shambala festival in Northamptonshire (August 2016), drew a nationwide audience. Mainly children under the age of 15 participated in our demonstrations whilst parents were more interested in our specific projects. In June 2017, KrebsFest on tour was part of the Cheltenham Science festival. During the day this was mainly attended by local school children. The most impact specific to this project came from the evening adult only event, where local members of the public asked lots of questions about what we research, how and why. Many commented on their changed views and opinions of scientific research following conversations with them.
Year(s) Of Engagement Activity 2016,2017
 
Description Press release + interview BBC Sheffield to comment on published article (Salamaga et al., PLoS Pathogens, 2017) 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Media (as a channel to the public)
Results and Impact Following the publication of the research article (Salamaga et al.,2017), a press release has been published by the university of Sheffield media team. This led to an interview with BBC Sheffield (31/07/2017), an article in the national media support Metro ("killer bug too small for antibiotics"), an article in AOL online and international media (India and Taiwan).
Year(s) Of Engagement Activity 2016,2017