Effects of Periplasmic Chaperones and Membrane Thickness on BamA-Catalyzed Outer-Membrane Protein Folding.

The biogenesis of outer-membrane proteins (OMPs) in gram-negative bacteria involves delivery by periplasmic chaperones to the ß-barrel assembly machinery (BAM), which catalyzes OMP insertion into the outer membrane. Here, we examine the effects of membrane thickness, the Escherichia coli periplasmic chaperones Skp and SurA, and BamA, the central subunit of the BAM complex, on the folding kinetics of a model OMP (tOmpA) using fluorescence spectroscopy, native mass spectrometry, and molecular dynamics simulations. We show that prefolded BamA promotes the release of tOmpA from Skp despite the nM affinity of the Skp:tOmpA complex. This activity is located in the BamA ß-barrel domain, but is greater when full-length BamA is present, indicating that both the ß-barrel and polypeptide transport-associated (POTRA) domains are required for maximal activity. By contrast, SurA is unable to release tOmpA from Skp, providing direct evidence against a sequential chaperone model. By varying lipid acyl chain length in synthetic liposomes we show that BamA has a greater catalytic effect on tOmpA folding in thicker bilayers, suggesting that BAM catalysis involves lowering of the kinetic barrier imposed by the hydrophobic thickness of the membrane. Consistent with this, molecular dynamics simulations reveal that increases in membrane thinning/disorder by the transmembrane domain of BamA is greatest in thicker bilayers. Finally, we demonstrate that cross-linking of the BamA barrel does not affect tOmpA folding kinetics in 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) liposomes, suggesting that lateral gating of the BamA barrel and/or hybrid barrel formation is not required, at least for the assembly of a small 8-stranded OMP in vitro.

Rapid and Robust Polyprotein Production Facilitates Single-Molecule Mechanical Characterization of ß-Barrel Assembly Machinery Polypeptide Transport Associated Domains.

Single-molecule force spectroscopy by atomic force microscopy exploits the use of multimeric protein constructs, namely, polyproteins, to decrease the impact of nonspecific interactions, to improve data accumulation, and to allow the accommodation of benchmarking reference domains within the construct. However, methods to generate such constructs are either time- and labor-intensive or lack control over the length or the domain sequence of the obtained construct. Here, we describe an approach that addresses both of these shortcomings that uses Gibson assembly (GA) to generate a defined recombinant polyprotein rapidly using linker sequences. To demonstrate the feasibility of this approach, we used GA to make a polyprotein composed of alternating domains of I27 and TmCsp, (I27-TmCsp)3-I27)(GA), and showed the mechanical fingerprint, mechanical strength, and pulling speed dependence are the same as an analogous polyprotein constructed using the classical approach. After this benchmarking, we exploited this approach to facilitiate the mechanical characterization of POTRA domain 2 of BamA from E. coli (EcPOTRA2) by assembling the polyprotein (I27-EcPOTRA2)3-I27(GA). We show that, as predicted from the α + ß topology, EcPOTRA2 domains are mechanically robust over a wide range of pulling speeds. Furthermore, we identify a clear correlation between mechanical robustness and brittleness for a range of other α + ß proteins that contain the structural feature of proximal terminal ß-strands in parallel geometry. We thus demonstrate that the GA approach is a powerful tool, as it circumvents the usual time- and labor-intensive polyprotein production process and allows for rapid production of new constructs for single-molecule studies. As shown for EcPOTRA2, this approach allows the exploration of the mechanical properties of a greater number of proteins and their variants. This improves our understanding of the relationship between structure and mechanical strength, increasing our ability to design proteins with tailored mechanical properties.