Dissecting the Effects of Periplasmic Chaperones on the In Vitro Folding of the Outer Membrane Protein PagP

Although many periplasmic folding factors have been identified, the mechanisms by which they interact with unfolded outer membrane proteins (OMPs) to promote correct folding and membrane insertion remain poorly understood. Here, we have investigated the effect of two chaperones, Skp and SurA, on the folding kinetics of the OMP, PagP. Folding kinetics of PagP into both zwitterionic diC12:0PC (1,2-dilauroyl-sn-glycero-3-phosphocholine) liposomes and negatively charged 80:20 diC12:0PC:diC12:0PG [1,2-dilauroyl-sn-glycero-3-phospho-(1′-rac-glycerol)] liposomes were investigated using a combination of spectroscopic and SDS-PAGE assays. The results indicate that Skp modulates the observed rate of PagP folding in a manner that is dependent on the composition of the membrane and the ionic strength of the buffer used. These data suggest that electrostatic interactions play an important role in Skp-assisted substrate delivery to the membrane. In contrast, SurA showed no effect on the observed folding rates of PagP, consistent with the view that these chaperones act by distinct mechanisms in partially redundant parallel chaperone pathways that facilitate OMP assembly. In addition to delivery of the substrate protein to the membrane, the ability of Skp to prevent OMP aggregation was investigated. The results show that folding and membrane insertion of PagP can be restored, in part, by Skp in conditions that strongly favour PagP aggregation. These results illustrate the utility of in vitro systems for dissecting the complex folding environment encountered by OMPs in the periplasm and demonstrate the key role of Skp in holding aggregation-prone OMPs prior to their direct or indirect delivery to the membrane.


Supplementary information SI Figure 1
Equilibrium unfolding curves for His-tagged PagP (HT PagP, circles) and untagged PagP (PagP, diamonds). The dark grey dashed lines represent the global fit of the HT PagP curves to a two-state model (described below). The fit parameters obtained (∆G°U N = -60.26 ± 6.24 kJ mol -1 , M UN = 6.92 ± 0.74 kJ mol -1 M -1 ) are in agreement with those previously measured for HT PagP 1 . The untagged construct did not fold reversibly under these conditions and the data therefore are not fitted. The black symbols indicate the <λ 320-370nm > for each PagP construct denatured in 10 M urea in the absence of liposomes. All experiments were carried out at a final protein concentration of 0.4 µM, in diC 12:0 PC liposomes, LPR 3200:1, 25 ºC, pH 8.0.
The average wavelength, <λ>, was calculated according to the following equation: in which <λ> is the average wavelength, λ i the wavelength and I i the fluorescence intensity at λ i . The <λ> is calculated over the range i = 320-370 nm.
Since <λ> has been shown to vary non-linearly with protein population, the fitting equation used must take into account the quantum yield ratio of the folded and unfolded states of PagP (Q R ) 2; 3 . Q R was calculated according to the following equation: where is the sum of the intensities of the folded state spectrum over the range of wavelengths used to calculate < λ >, is the sum of the intensities of the unfolded state spectrum over the range of wavelengths used to calculate < λ >.
This term is then incorporated into the equation for a two-state folding model as previously described 2; 3 , and the data fitted using this equation: where S obs is the observed signal, a and c are the signals of the native and denatured states, respectively, in the absence of denaturant, b and d are the denaturant dependence of the signal of the native and denatured states, respectively, [D] is the denaturant concentration, is the free energy of unfolding in the absence of denaturant, is the m-value (which reflects the denaturant dependence of ) and Q R is the quantum yield ratio as previously defined. . Four replicate samples were globally fitted to a single exponential function to obtain the reported rate constants.

SI Figure 2
Urea-dependence of PagP folding monitored using cold SDS-PAGE. All samples contained 4 µM PagP (diC 12:0 PC liposomes, LPR 3200:1) in 10 mM glycine buffer, pH 9.5, 2 mM EDTA and were refolded at 37 ºC for 15 hours prior to 1:1 dilution with 2 × SDS-PAGE loading buffer. Lanes on the SDS-PAGE gel are numbered to indicate the final concentration of urea (M) in each sample. The unfolded and folded forms of PagP are denoted by U and F, respectively. SI Table 2 PagP  PagP populates a single conformation under the conditions used for chaperone binding. 10 µM PagP in 50 mM glycine, pH 9.5 containing 0.24 M urea was incubated for 5 min at room temperature before injection on to a Superdex 75 10/300 GL column (see Methods).

SI Figure 5
Equilibrium unfolding of SurA monitored using tryptophan fluorescence emission. SurA dissolved in 50 mM glycine buffer, pH 9.5 was diluted into separate aliquots of 50 mM glycine buffer containing different concentrations of urea (0-8 M in 0.2 M increments). The final concentration of protein was 2.5 µM. The aliquots were equilibrated overnight (16 h) at 37 °C before measurement. The fluorescence emission intensity was measured at 335 nm (after excitation at 280 nm) due to the large intensity difference between the folded and unfolded states at this wavelength, for 60 s and the average signal calculated.

SI Figure 6
PagP and SurA do not interact under the conditions used in the kinetic folding assays. (a) 10 µM PagP was added to 60 µM SurA in 50 mM glycine, pH 9.5 containing 0.24 M urea and incubated for 5 min at room temperature before injection on to a Superdex 75 10/300 GL column (see Methods).  SI Table 3 Lipid Experiment Measured rates of PagP folding into liposomes in vitro. Rate constants were obtained by measuring Trp fluorescence emission at 335 nm over time, and fitting the observed transients to a single exponential function (see Methods). Global fits were obtained over four replicates from a single batch of liposomes, and the average of the global fits from folding reactions into three batches of liposomes calculated. The standard error of the mean was calculated by taking the number of liposome replicates to be 3.