Skp and SurA are both periplasmic chaperones involved in the biogenesis of Escherichia coli β-barrel outer membrane proteins (OMPs). It is commonly assumed that SurA plays a major role whereas Skp is a minor factor. However, there is no molecular evidence for whether their roles are redundant. Here, by using different dilution methods, we obtained monodisperse and aggregated forms of OmpC and studied their interactions with Skp and SurA by single-molecule fluorescence resonance energy transfer and fluorescence correlation spectroscopy. We found that Skp can dissolve aggregated OmpC while SurA cannot convert aggregated OmpC into the monodisperse form and the conformations of OmpC recognized by the two chaperones as well as their stoichiometries of binding are different. Our study demonstrates the functional distinctions between Skp and SurA. In particular, the role of Skp is not redundant and is probably more significant under stress conditions.

The cell envelope of gram-negative bacteria consists of the outer membrane (OM), inner membrane (IM), and periplasm. The β-barrel outer membrane proteins (OMPs) embedded in the OM perform diverse and significant functions such as signaling, transporting, and proteolysis. The OMPs of gram-negative bacteria share similar folding pathways with that of mitochondria and chloroplasts. Therefore, the study of the OMP folding mechanism not only provides insights into antimicrobial drug design but also helps elucidate mitochondrial and chloroplast biogenesis. Most knowledge about OMP folding was obtained from ensemble experiments where OMPs were usually at micromolar concentrations and prone to aggregate, which is different from the physiological environment in the cells. Unlike ensemble techniques, single-molecule detection (SMD) can measure OMPs from nano-to picomolar concentrations and prevent aggregation. In this work, we investigated the folding of OmpT, one of the OMPs, in Tween-20 and n-dodecyl β-d-maltopyranoside (DDM) micelles by SMD. We prepared monodisperse OmpT and observed both unfolded and folded OmpT in Tween-20 and DDM micelles under different urea concentrations by single-molecule fluorescence resonance energy transfer (FRET). The folded OmpT in Tween-20 is structurally similar to the native OmpT folded in DDM but exhibits weaker resistance to urea. In contrast, OmpA barely folds and OmpC hardly folds in Tween-20 micelles. We confirmed that folded OmpT forms complexes with detergent micelles and estimated the number of bound Tween-20 and DDM molecules per OmpT by fluorescence correlation spectroscopy. We compared the enzymatic activity of OmpT folded in two detergents with a fluorescent peptide as substrate, and found that the folded form of OmpT in Tween-20 possesses weaker enzymatic activity than that in DDM. We also investigated the folding properties of OmpT, OmpA, and OmpC in the presence of the β-barrel assembly machine (BAM) complex. OmpT folds efficiently in liposome even without the BAM complex; OmpA only folds with the help of the BAM complex; and OmpC does not fold with or without the BAM complex. Based on the comparison of the folding of OmpT, OmpA, and OmpC in detergent micelles and in the presence of the BAM complex, we propose that OmpT has stronger folding tendency than OmpA and OmpC, which supports the idea that the exact role of the BAM complex is dependent on the distinct folding properties of individual OMPs. Since Tween-20 is a widely used reagent to block nonspecific adsorption in SMD experiments, our results also remind people to exercise caution to prevent possible wrong interpretations caused by the interaction between proteins and Tween-20.