Sequential assembly of the bacterial flagellum outside the living cell

Lead Research Organisation: University of Cambridge
Department Name: Pathology

Abstract

Bacteria are the smallest of all cells and live lives of feast and famine. To survive they must sense and adapt quickly to their ever-changing external environment. To find food and escape danger, many bacteria build on their cell surface long rotating propellers called flagella, which allow them to move rapidly over surfaces and through liquids. Motility driven by surface flagella is important to colonization of rich niches in human, animal and plant hosts by disease-causing (pathogenic) bacteria. Both biologists and physicists have long found flagella fascinating as they illustrate beautifully how complex biological structures, comprising a number of discrete substructures, are assembled meticulously from thousands of distinct protein subunit building blocks to operate as tiny rotary 'nanomachines' outside the living cell. They have even been claimed by creationists, who cite their exquisite structural-functional complexity as evidence against slow evolution!

Remarkably, the flagellum is fundamentally self-assembling and this feat of nanoscale construction is rendered even more astonishing by the fact that each new subunit building block, made inside the cell, must travel up to 20 times the length of the cell through a narrow central channel in the growing flagellum to incorporate at its tip far outside the cell. Until recently, the mystery was how the subunit building blocks travel through the long channeI where there is no discernable energy source to propel them. We have discovered that this is powered by the subunits themselves, as they link in a chain that is pulled to the flagellum tip.

We now propose to use molecular, biochemical and structural biology approaches to build on this completely new and unanticipated chain mechanism for flagellum growth to address a major open question concerning the assembly of the complex flagellum structure: how are the many different subunit building blocks incorporated into the flagellum in the correct sequence? The proposed work has the potential to reveal new aspects of flagellum assembly and provide a deeper understanding of how multi-component biological machines can be assembled beyond the surface of living cells. Findings from this work could, ultimately, be exploited to discover new molecules to block the function of such machines, leading to the development of new drug therapies to treat bacterial infections.

Technical Summary

Flagella, the helical propellers that extend from the bacterial cell surface, are a paradigm for how complex molecular machines are built outside the living cell. Their sequential assembly requires export of thousands of structural subunits across the cell membrane and this is achieved by a type III export machine located at the flagellum base. Subunits then transit up to 20 cell lengths through a narrow channel in the flagellum to reach the assembly site at the flagellum tip. We have recently solved the mystery of how subunit transit is achieved without a conventional energy source in the external channel, showing that the energy for transit is intrinsic to the subunits themselves as they link in a chain pulled to the flagellum tip (Nature 504:287).

Our proposal builds on this new chain mechanism and preparatory experiments arising from it, and is primed to reveal serial key mechanisms for subunit recognition at the export machinery, sequential subunit assembly and length-control in the flagellum. We will use our biochemical approaches to determine whether subunit binding affinities determine the order in which subunits link into the chain and incorporate into the nascent flagellum. We will build on initial NMR and crystallography to gain a structural view of subunit interactions with the export gate, revealing binding interfaces and conformational changes triggered by subunit binding. Finally, we will assess interactions at the export gate and with subunit chains by the molecular ruler FliK, which monitors hook length and governs the export switch from rod/hook to filament subunits, an approach that could uncover the mechanism by which FliK reports hook length back to the export gate, a critical and unexplained event in the sequential assembly of flagella. In addition to resolving central features of flagella assembly, our work will be pertinent to understanding how bacterial virulence needles are built, and indeed to the wider subject of export and assembly.

Planned Impact

Private sector beneficiaries: While the immediate impact of our research will be in the academic community, in the longer term our work has potential for impact in the pharmaceutical, biotechnology and nanotechnology industries. Firstly, our work will potentially uncover new targets for novel antimicrobials that block bacterial motility or effector export. Secondly, protein export via the flagellar pathway has already become a focus for biotechnologists working to develop systems for rapid and simple purification of large amounts of recombinant proteins from the extracellular growth media (e.g. Nature Biotechnology 23:475-81). Our proposed work has the potential to reveal new aspects of flagellar export that could be exploited to further improve the secretion of heterologous proteins via the flagellar pathway. Lastly, as detailed in the academic beneficiaries section, our work on flagella export has already attracted attention in the field of nanotechnology and our work could have implications for the self-assembly of multicomponent nanotubes. Should we reach the point where aspects of our research show strong potential for translation to industry, we will take appropriate steps (see Pathways to Impact) and work with Cambridge Enterprise to engage with industry, patent discoveries, and translate our research into commercially exploitable resources.

We take very seriously the need to train and develop new cohorts of researchers skilled in current molecular microbiology, structural biology and biochemical techniques that are vital to the continued success of UK science and to maintain UK competitive advantage in the increasingly knowledge-led economy, especially in the face of increasing microbiological challenges of the future. Such researchers are in high demand in the industrial sector, and previous members of our labs have taken leading positions in UK industry. In addition to acquiring valuable technical expertise, the RA employed on this project will work with the PI, Co-I, Cambridge University Personal and Professional Development and the University Careers Service to develop transferable professional skills (e.g. Leadership and Management Development, Presentation and Communication) that will equip them for transition to many employment sectors.

Beneficiaries in the wider public : We anticipate a number of impacts. In the short-to-medium term, there will be benefits from our planned engagement activities (see Pathways to Impact). Part of the fight to combat infectious disease lies in generating public awareness: by learning about research like ours the public become aware of the science behind the global health threat that will be posed in the future by re-emergent resistant bacteria. Such an understanding is important in seeking to augment research support. A major aspect of our public engagement effort will be schools outreach work. We view this as essential to foster interest in STEM subjects. The PI, Co-I and RA employed on the project will regularly engage students (e.g. through Nuffield scholarships), giving us further opportunity to inspire the next generation of researchers.

Longer-term benefits to the public will emerge should our work translate to industry. Flagella and related needle structures are important virulence factors in a wide range of human, animal and plant infections. Treatment by existing antibiotics is threatened by rising resistance so fundamental research into virulence mechanisms is vital to our understanding of infections, and for our ability to develop new interventions. Ultimately, the target beneficiaries are the millions suffering from bacterial infections and our hope is that such work will facilitate design of new drugs that could attenuate bacterial motility and effector export and be used as adjuncts to antibiotic-based therapies. In this case, the work will also impact upon those developing new antimicrobial therapies both in academia and in pharmaceutical research.
 
Description The project has produced several significant findings that reveal new mechanistic aspects of bacterial
flagellar export. These are outlined below in relation to the original project objectives (i-iv). In addition,
our work generated a number of unanticipated findings (v).

(i) Do rod/hook subunit affinities for the export machinery reflect the order in which subunits link
in the chain and/or assemble into the flagellum? What is the structural basis for subunit
interactions with the export machinery?
Using isothermal titration calorimetry (ITC) and 2D/3D NMR, we determined the binding affinities of all
rod and hook structural subunits for the cytoplasmic domain of the FlhB flagellar export gate component
(FlhBc) and characterised structural changes in FlhBc associated with subunit binding. The NMR
studies confirmed that the subunit Gate Recognition Motif (GRM) contacts a conserved hydrophobic
patch on the surface of FlhBc. We found that there was no correlation between subunit-FlhB binding
affinities in vitro and the order in which subunits assemble into the flagellum. Using RNA-seq and Riboseq,
we extended this work to investigate the relationship between cellular levels of newly synthesised
subunits and assembly order/stoichiometry in the flagellum.

(ii) Are rod/hook subunits captured into the transiting chain in a specific order? Does this order
reflect the sequence of rod/hook subunit assembly?
Using in vitro and in vivo assays, we found no evidence to support the notion that subunits are captured
into the transiting chain in a specific order. Indeed, we have found that subunit chaining is not dependent
on specific residues at subunit N-/C-termini, but is driven by confinement of unfolded subunit
polypeptides within the narrow 2nm flagellar channel. We are now performing additional work on
subunit chaining in vivo.

(iii) How and where do filament subunit chains form? Are filament subunits captured into the
transiting chain in an order that reflects the sequence of filament subunit assembly?
As for rod/hook subunits, our findings indicate that filament subunits form chains once confined in the
flagellar channel1 and that chaining is not dependent on specific residues at subunit N-/C-termini. Our
RNA-seq and Ribo-seq data indicate that cellular levels of newly synthesised filament subunits correlate
with stoichiometry in the flagellum.

(iv) How does the FliK molecular ruler interact with the FlhB export gate? How does FliK transit
through the growing flagellum: does FliK make or break subunit chains?
Using isothermal titration calorimetry (ITC) and 2D NMR, we determined the binding affinity of the FliK
molecular ruler for the cytoplasmic domain of the FlhB flagellar export gate component (FlhBc) and
characterised structural changes in FlhBc associated with FliK GRM binding. We found that FliK drives a
major structural change in FlhBc that causes loss of the cleaved but non-covalently associated Cterminal
FlhBcc domain. This destroys the conserved hydrophobic patch on the surface of FlhBc that
binds rod/hook subunits, meaning that they are no longer recruited to the export machinery. This FliKdependent
destruction of the early export gate triggers a switch in the specificity of the export
machinery from rod/hook to filament protein export.

(v) Other significant achievements that have emerged from the project
1) Subunit N-terminal secretion signals trigger opening of the fT3SS FlhAB-FliPQR export gate
By analysing a suite of natural cargo subunits containing deletions in the N-terminal secretion signal, we
identified dominant-negative cargo variants with 'short' signals that stalled at the cytoplasmic domain of
the FlhB fT3SS export gate component (FlhBc), blocking secretion. We isolated intragenic revertants
that relieved the secretion block and found that the suppressor mutations restored the length of the
secretion signal and, in some cases, reinstated hydrophobic residues at the cargo's extreme N-terminus.
We reasoned that the 'short' dominant negative cargo variants might stall at FlhBc because their
secretion signals could not reach a binding site on the FlhAB-FliPQR export gate to trigger opening and
cargo translocation. To test this, we used the published atomic resolution structure of FliPQR (Prof S.
Lea and colleagues) to identify residues potentially involved in export gate opening. We found that
alanine replacement of FliP M210 or FliR I110 or I113 relieved the secretion block caused by 'short'
dominant negative cargo variants, possibly by destabilising the export gate closed state, promoting
secretion even in the absence of a cargo secretion signal. An extension of this work has revealed new mechanistic aspects of PMF-driven opening of the Type III
secretion periplasmic export gate.
Exploitation Route Data from the project have fed into new work that will engineer the E. coli K12 fT3SS for efficient 'one step'
secretion of high-value recombinant proteins directly into the extracellular environment, simplifying
downstream processing and reducing costs2. With Dr Graham Stafford (University of Sheffield) and his
collaborators at Fujifilm Diosynth Biotechnologies, we have developed this project, funded in part by a
BBSRC Business Innovation Voucher (Exploiting mechanistic understanding of the flagellar export machinery to inform engineering of
flagellar Type III secretion system (fT3SS) 'pipes' as a platform technology for industrial biotechnology). We now aim to secure further funding for to support this new
project.
Sectors Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology

 
Description BBSRC CBMNet Business Interaction Voucher
Amount £9,930 (GBP)
Organisation CBMNet 
Sector Academic/University
Country United Kingdom
Start 02/2017 
End 04/2017
 
Title Recognition of discrete export signals in flagellar subunits during bacterial Type III secretion 
Description Type III Secretion Systems (T3SS) deliver subunits from the bacterial cytosol to nascent cell surface flagella. Early flagellar subunits that form the rod and hook substructures are unchaperoned and contain their own export signals. A gate recognition motif (GRM) docks them at the FlhBc component of the FlhAB-FliPQR export gate, but the gate must then be opened and subunits must be unfolded to pass through the flagellar channel. This induced us to seek further signals on the subunits. Here, we identify a second signal at the extreme N-terminus of flagellar rod and hook subunits and determine that key to the signal is its hydrophobicity. We show that the two export signal elements are recognised separately and sequentially, as the N-terminal signal is recognised by the flagellar export machinery only after subunits have docked at FlhBC via the GRM. The position of the N-terminal hydrophobic signal in the subunit sequence relative to the GRM appeared to be important, as a FlgD deletion variant (FlgDshort), in which the distance between the N-terminal signal and the GRM was shortened, 'stalled' at the export machinery and was not exported. The attenuation of motility caused by FlgDshort was suppressed by mutations that destabilised the closed conformation of the FlhAB-FliPQR export gate, suggesting that the hydrophobic N-terminal signal might trigger opening of the flagellar export gate. 
Type Of Material Database/Collection of data 
Year Produced 2022 
Provided To Others? Yes  
URL http://datadryad.org/stash/dataset/doi:10.5061/dryad.66t1g1k3x
 
Description Engineering flagellar export pipes 
Organisation Fujifilm
Department Fujifilm Diosynth Biotechnologies
Country United States 
Sector Private 
PI Contribution My research team has developed novel expression vectors and performed molecular biochemical and imaging studies as part of a collaborative project to test a novel one-step flagella-based system for secretion of biotechnologically important proteins into the extracellular milieu.
Collaborator Contribution Our collaborator, Dr Graham Stafford, at the University of Sheffield has developed prototype strains and vectors, and a high-throughput assay to assess secretion efficiency. Fujifilm Diosynth Biotechnologies (FujiDB) will run a number of high-density industrial fermentations using the Sheffield/ Cambridge prototype strains. This would require 1xFTE for 3 weeks (£4000) plus materials (£1000) and consumables (£2000, including for preparation of recombinant gene of interest). FujiDB would also collect high quality fermentation data on growth (OD, dry weight, £1000). These would be supplied to the academic partners for analysis as detailed above. Ian Hodgsons time vs internal project supervision (3 days, £2000) plus attendance for project review meetings and travel costs will be provided by FujiDB.
Impact No impact yet.
Start Year 2016
 
Description Engineering flagellar export pipes 
Organisation University of Sheffield
Country United Kingdom 
Sector Academic/University 
PI Contribution My research team has developed novel expression vectors and performed molecular biochemical and imaging studies as part of a collaborative project to test a novel one-step flagella-based system for secretion of biotechnologically important proteins into the extracellular milieu.
Collaborator Contribution Our collaborator, Dr Graham Stafford, at the University of Sheffield has developed prototype strains and vectors, and a high-throughput assay to assess secretion efficiency. Fujifilm Diosynth Biotechnologies (FujiDB) will run a number of high-density industrial fermentations using the Sheffield/ Cambridge prototype strains. This would require 1xFTE for 3 weeks (£4000) plus materials (£1000) and consumables (£2000, including for preparation of recombinant gene of interest). FujiDB would also collect high quality fermentation data on growth (OD, dry weight, £1000). These would be supplied to the academic partners for analysis as detailed above. Ian Hodgsons time vs internal project supervision (3 days, £2000) plus attendance for project review meetings and travel costs will be provided by FujiDB.
Impact No impact yet.
Start Year 2016
 
Description NMR collaboration 
Organisation University of Cambridge
Department Department of Chemistry
Country United Kingdom 
Sector Academic/University 
PI Contribution The collaboration involved NMR studies of bacterial flagellar export machinery components. Experimental work and peak assignment was carried out by members of my lab group in the NMR Facility (Department of Biochemistry).
Collaborator Contribution Dr Daniel Nietlispach, head of the NMR Facility in the Department of Biochemistry, provided training for members of my lab group.
Impact No outputs, as yet.
Start Year 2015