Stochastic assembly of pore forming toxin complexes

Mirko Bischofberger

Pore-forming toxins (PFTs) constitute the biggest class of bacterial protein toxins and confer potent virulence to many pathogens. Unlike other membrane proteins, which are inserted into the membrane co-translationally, PFTs are secreted by the bacteria in a water-soluble form. The monomers then diffuse towards the target cells, bind to them via specific receptors and oligomerize on their own into membrane-spanning, ring-like complexes. Some PFTs, such as aerolysin from A. hydrophila, form pores directly in the plasma membrane of target cells. In the case of mammalian cells this leads to ionic imbalances in the cell and ultimately cell death, in the case of erythrocytes to cell lysis. Other PFTs transport harmful secondary toxins into the cells via pores formed inside the cells. Pores built by PFTs vary greatly in stochiometry and pore diameter. Staphylococcal alpha-toxin and Aeromonas aerolysin form heptameric channels with a ~2-3 nm diameter, while members of the cholesterol-dependent cytolysins pores of variable sizes (30-50nm) that can be formed by of up to 50 monomers.

The assembly of pores by PFTs is a stochastic process. Using mathematical models we ask how the single toxin monomers dynamically come together to form multimeric pores on living cell membranes. For example, do the monomers assemble in a sequential or a random manner? Modeling the stochastic assembly of PFTs can be done on a theoretically correct basis through the formulation of chemical master equations. This approach models exact trajectories of individual reaction events and with it the stochastic nature of the process.

Figure 1 : The staphylococcal alpha-haemolysin pore (pink) is made up of seven subunits, which span the lipid bilayer as a beta-barrel, whereas ClyA (green) forms an iris-like barrel of alpha-helices from its twelve subunits. The assembly pathway of such PFTs is unknown (yellow). Modified from: “Membrane-protein structure: Piercing insights”, Bayley H., Nature 459, 651-652(4 June 2009)

At the moment we study the stochastic assembly process of the PFT aerolysin (PA), a well studied member of bacterial PFTs. Unlike other PFTs that form pores inside the cell PA forms pores on the cell surface of target cells. In the case of mammalian cells pore formation can indirectly be observed using a calcium-sensitive fluorescent dye, where the influx of calcium due to pore formation is monitored. In the case of erythrocytes pore formation can indirectly be observed by cell lysis. Quantification of the lag times of pore formation gives insights into the stochastic process of pore formation on the surface of individual cells in real-time. We compare the experimental data with different mathematical models of pore assembly.

Figure 2 : Two possible scenarios of how the heptameric pore of aerolysin could assemble. Top: Sequential model in which pores can only be made through the addition of monomers. Bottom: Random model in which all possible assemblies between oligomers of different stochiometries are allowed.