Molecular basis of protein translocation through outer membrane porins

Lead Research Organisation: University of Oxford
Department Name: Biochemistry


Gram-negative bacteria are surrounded by two membranes, the outer most of which (usually referred to as the 'outer membrane') is a highly effective barrier against toxic molecules; for example, bile salts which exist within the gut of a mammal. The effectiveness of the outer membrane is a double-edged sword for the organism since it is also very effective at keeping out essential nutrients such as sugars, which the organism needs to grow and divide. Hence all organisms that have an outer membrane also have specialised proteins within this membrane whose job it is to allow the exchange of nutrients and metabolites with the environment. These proteins are known as porins, which are barrel-like membrane proteins that have a hole or pore running through them which traverses the membrane. There are many barrel proteins in the outer membrane of Gram-negative bacteria (it is estimated that 2-3% of the E. coli genome encode such proteins) which serve a variety of functions. Those that allow the exchange of nutrients are usually referred to as general or classical porins, the best understood of which are the proteins OmpF and OmpC. These porins are also the major routes by which commonly used antibiotics (e.g. ampicillin) diffuse into the cell. Indeed, the channels of these porins are frequently found to be mutated in multidrug resistant bacteria.

The focus of this research proposal is a novel function we have recently discovered for the general porin OmpF which shows that porins can do more than just let small molecules diffuse through their pores. We have discovered that OmpF can also be used to allow proteins to pass (translocate) into the cell, as long as the protein is unfolded (i.e. a random coil) and can snake its way through the narrow pores of the porin. Even more remarkably, this translocating protein snakes through two of the three pores of OmpF, which is composed of three barrel subunits; i.e. it goes into the cell then comes back out again. Given the importance of porins to the physiology of bacteria and eukaryotes (mammals and plants have porins in the outer membranes of some of their organelles) our discovery has important ramifications for our understanding of the biology of organisms that have outer membranes since it shows that proteins can exploit their porins. Proteins have functions which are usually imparted by their having a three dimensional structure or fold. Although the proteins that pass through OmpF are unfolded polypeptides and so by implication have no function, these can become functional by virtue of their being able to bind to other molecules such as proteins, and hence in this way alter cellular behaviour. The twin aims of this proposal are to understand the molecular basis for protein translocation through the porin OmpF in the gut bacterium Escherichia coli (determining for example how much polypeptide can pass through the pore) and to discover how common this phenomenon is by looking at other porins and porins from other organisms. Another important question to be investigated is what drives protein translocation through porins? This is an important question since the outer membrane has no energy source to call upon in order to drag a protein into a cell (unlike the inner most membrane which does) and so answering this question will not only provide us with fundamental new insights into bacterial porins themselves but also the outer membrane in which they reside.

Technical Summary

We have discovered that an intrinsically disordered protein (IDP) domain of a colicin can thread its way through two subunits of the trimeric porin OmpF in order to deliver an epitope signal to the bacterial periplasm. This proposal aims to uncover the molecular mechanism of this new form of protein translocation and determine if it occurs in other systems.

(1) How does an IDP associate with a porin and what is the driving force for translocation?
This will be investigated by crystallographic, mutational, thermodynamic (calorimetry), and kinetic studies (stopped-flow FRET and planar lipid bilayer (PLB) single channel recordings), all ultimately feeding into molecular dynamics simulations of IDP binding and translocation through porins.

(2) How does an IDP thread through OmpF and how much sequence can be translocated?
The threading mechanism will be investigated using constructs of colicin E9 (1-62) that contain both its OmpF binding sites and its intervening signalling epitope in PLB experiments. We will also determine what constraints there are in the amount of polypeptide that can be translocated.

(3) What is the relationship between IDP sequence and porin binding/translocation?
We will identify OmpF binding sites within the many colicins that are known to utilise OmpF for their import to determine what governs their binding to porins, exploiting a newly devised native state electrospray ionisation mass spectrometry method we have developed.

(4) Do porins other than OmpF bind and translocate IDPs?
We will investigate IDP binding/translocation in other porins, principally OmpC and PhoE to determine differences/similarities with OmpF, and identify/study porins from the PDB with the potential translocate IDPs.

(5) Can a folded domain be delivered through OmpF if tethered to a colicin's IDP?
We will determine if a small, folded domain present in a colicin known to translocate its preceding IDP through OmpF can also be unfolded/translocated.

Planned Impact

Who might benefit from this research (excluding other academics) and how might they benefit?

The main beneficiaries of this research will be:

(i) the biotechnology industry, especially those with interests in the use of protein pores in analytical applications (biosensors, protein sequencing). As part of our Impact plan we will engage with scientists at Oxford Nanopore Technologies to discuss potential ways to exploit our discoveries in a way that dovetails with their priorities. Hagan Bayley was one of the co-founders of ONT, which has raised over £100M in venture capital. As part of our research programme we will discover how IDPs interact with/translocate through different porins, which could be valuable in the development of novel biosensors. Finding novel ways of delivering proteins and peptides into cells could have far reaching consequences for biological research and for biotechnology. We will uncover the constraints to the translocation mechanism through porins. We will uncover the ground rules for porin translocation and so be able to design peptide modules, for example, that could bind periplasmic proteins while carrying defined fluorophores. Porin translocation might also be a route by which epitopes could be displayed from the surfaces of bacteria, for example in response to an extracellular signal. Through the University's Technology Transfer Office (Isis Innovation) we will explore the possibility of patenting a novel bacterial surface display technology we are in the process of developing, which is based on IDP translocation through porins and the presentation of epitopes on the cell surface. We will also make links with companies that develop antimicrobial peptides, some of which likely use porins for entry into Gram-negative bacteria. Our research on IDP translocation through porins may provide avenues for the production of more efficient antimicrobial peptides. Protein secretion through bacterial porins has recently been described in the literature, which must almost certainly pass through porins as IDPs. Our work on IDP translocation will provide possible mechanisms for such transfer. The secretion of proteins to the extracellular medium is an important goal in biotechnology for the facile production of proteins (for use in analytical applications or as biologics).

(ii) the general public. A range of outreach and research dissemination activities are planned. The main objectives are to, first, explain the importance of this research for understanding how antimicrobial proteins enter bacteria as this may be a means of developing novel antibacterial strategies (although this is not an objective of this specific proposal), and second, using this as a springboard for explaining concepts as 'what an Intrinsically Disordered Protein (IDP) is?' and 'how protein antibiotics differ from traditional antibiotics'. These educational goals are important as part of a wider objective here in Oxford to prepare the public for new forms of antibiotics that include proteins. A variety of avenues will be used to conduct this outreach. Including Science Open day events organised through the Departments of Biochemistry and Chemistry, public talks (CK to give a secondary school lecture organised by the Biochemical Society), production of non-specialist websites (e.g. Proteopedia) that describe IDPs and in particular how they can pass through narrow porin pores and disseminating web material suitable for a lay audience to BBSRC (e.g. as was recently done for our Science paper which underpins this application; health/2013/130703-n-antibiotics-sneak-across-membranes.aspx)


10 25 50
Description We had discovered that unfolded parts of protein toxins known as colicins are able to bind to outer membrane proteins called porins in bacteria. In colicin E9 for example we had previously found that two regions of its disordered N-terminus were able to thread through the outer membrane porin OmpF and seemingly bind in opposite orientations to this trimeric protein. One of our objectives in this grant was to establish if there were preferences for the directionality of binding by these disorder sequences to the pores of OmpF and, if so, the mechanism of binding of each sequence. We have discovered that these sequences do indeed bind in opposite orientations to OmpF and that this preference appears linked to the length of the sequence and its overall charge. We have determined these preferences using both channel experiments and in vivo fluorescence microscopy experiments.
Exploitation Route We defined the ground rules by which disorder sequences bind and translocate through outer membrane porins in Gram-negative bacteria. This could have repercussions as to how antimicrobial peptides, many of which are disordered sequences, pass through such pores.
Sectors Agriculture, Food and Drink,Pharmaceuticals and Medical Biotechnology

Description Response mode grant
Amount £633,423 (GBP)
Funding ID BB/P009948/1 
Organisation Biotechnology and Biological Sciences Research Council (BBSRC) 
Sector Public
Country United Kingdom
Start 07/2017 
End 07/2020
Title Directionality of porin in planar lipid bilayers 
Description The bacterial outer membrane porin OmpF has been the subject of 1000s of studies. All Gram-negative bacteria have similar trimeric porins in their outer membrane. these allow passive diffusion of molecules into and out of the cell. OmpF has been studied for decades by planar lipid bilayer approaches in which single OmpF trimers are inserted into a bilayer separating two buffer compartments and conductance through the OmpF channel used to indicate the passage of ions and small molecules. However, the orientation of OmpF in these experiments was always inferred and never demonstrated. This is particularly important since OmpF can insert into lipid bilayers in either orientation (inside-in or inside-out). Using chemical modification using engineered OmpF trimers we established an unambiguous method by which to determine the orientation of OmpF in planar lipid bilayers that could related back to the asymmetry of current/voltage curves. This work has been published. 
Type Of Material Technology assay or reagent 
Year Produced 2017 
Provided To Others? Yes  
Impact many studies have investigated the effect of antibiotics on OmpF channels in planar lipid bilayers. Our work now allows the unambiguous interpretation of the orientation of OmpF in these experiments. 
Description Collaboration with Genentech 
Organisation Genetech, Inc
Department Department of Infectious Diseases
Country United States 
Sector Private 
PI Contribution Following from our published work in Nature, where we describe for the first time the spatial and temporal nature of outer membrane protein islands, scientists at Genentech invited me to give a seminar and then suggested a collaboration. Genentech are developing antimicrobial agents centred on the bacterial outer membrane. Our new collaboration will address fundamental questions concerning how the outer membrane is built. My laboratory will use some of the microscopy approaches we have developed to analyse reagents provided by Genentech.
Collaborator Contribution Genentech will be providing a variety of reagents, including bacterial strains and fluorescently-labelled molecules.
Impact None yet
Start Year 2016
Description Mode of action of protein antibiotics 
Organisation The Wellcome Trust Sanger Institute
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution My laboratory provide materials, manpower and expertise for the investigation of protein antibiotic action against pathogenic E. coli (UPEC) strains held at the Sanger. We addressed a long-standing problem in the field whereby protein antibiotics (colicins) made naturally by E. coli populations seem not to be active against many pathogenic species. We discovered that growth conditions influence the integrity of the outer membrane lipids which in turn affects susceptibility to colicins.
Collaborator Contribution This is a collaboration with Prof Julian Parkhill's laboratory at the Sanger. Two postdocs in Prof Parkhills laboratory conducted high-throughput Biolog and TraDIS screening in order to determine growth conditions optimal for killing of a UPEC strain and to identify susceptibility determinants. Prof Parkhill was eventually a co-applicant on a Wellcome Trust Collaborative Award which I led and is still running. Prof Parkhill recently moved to the University of Cambridge (Department of Veterinary Medicine).
Impact No outputs yet but Wellcome Trust Collaborative Award (£2M) was funded to develop protein antibiotics. This is in collaboration with groups in Oxford, Glasgow, Sanger and UCL
Start Year 2015
Description Production of a glossy brochure targeted at a lay audience outlining our work on the development of protein-based antibiotics 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact I wrote an article for Science Impact Ltd titled 'Developing next-generation antibiotics' which summarised the work (much of it funded by BBSRC) on the intoxication mechanisms of antibacterial proteins (bacteriocins). The article was produced as a glossy pamphlet which was distributed as hardcopy and electronically to various stakeholders (research councils etc).
Year(s) Of Engagement Activity 2017