The mechanism of BAM-assisted OMP folding

Outer membrane proteins (OMPs) mediate the survival and pathogenicity of Gram negative bacteria. The biogenesis of these proteins, however, presents problems as they must be transported to, inserted and folded correctly in the outer membrane in the absence of ATP. This problem is resolved by the β-barrel assembly machinery (BAM) complex: a ~203 kDa complex of five proteins (BamA-E) that enables the membrane insertion and folding of substrate OMPs on a physiological timescale. Despite available crystal structures, the mechanism of this vital protein complex remains poorly understood. In this thesis I use a variety of structural and biochemical tools to probe the nature of BAM-assisted OMP folding, particularly the role of BamA dynamics. Successful purification of the intact BAM complex in the detergent DDM allowed the first cryo-electron microscopy structure of the complex to be obtained, at a resolution of 4.9 Å. This reveals the intact BAM complex with BamA in a laterally-open conformation in which the first (β1) and last (β16) strands of the barrel are no longer hydrogen bonded. In addition, biochemical assays provide the first in vitro evidence of the functional importance of BamA lateral gating in OMP folding. These assays demonstrate that in a reconstituted system utilising the BAM complex, inhibiting the lateral gating of BamA by incorporating new disulphide bonds diminishes the ability of BAM to assist substrate folding. This is shown with two different OMP substrates, tOmpA (the β-barrel domain of OmpA) and OmpT. In synthetic lipids, however, the presence of prefolded BamA is sufficient to aid substrate folding and inhibition of lateral gating by disulphide bonding in this case does not diminish the catalytic effect. The results indicate that BamA likely adopts different roles depending on substrate and lipid. Furthermore, this thesis discusses preliminary experiments towards determining the significance of the β-signal: a conserved sequence found towards the C-terminus of OMPs hypothesised to be important for recognition by BamA. The results show that while some mutations may slow the protein’s intrinsic folding into the membrane they do not affect the apparent BamA-catalysed folding rate. However, other single amino-acid substitutions appear to incur a large energetic penalty, rendering the protein incapable of adopting its stable β-barrel structure. Combined, the data allow us to begin dissecting the mechanism of BAM-assisted folding of OMPs, particularly the role of BamA in passive membrane destabilization or active lateral opening.