DsbA is a redox-switchable mechanical chaperone

DsbA is a ubiquitous bacterial oxidoreductase that associates with substrates during and after translocation, yet its involvement in protein folding and translocation remains an open question. Here we demonstrate a redox-controlled chaperone activity of DsbA, on both cysteine-containing and cysteine-free substrates, using magnetic tweezers-based single molecule force spectroscopy that enables independent measurements of oxidoreductase activity and chaperone behavior. Interestingly we found that this chaperone activity is tuned by the oxidation state of DsbA; oxidized DsbA is a strong promoter of folding, but the effect is weakened by the reduction of the catalytic CXXC motif. We further localize the chaperone binding site of DsbA using a seven-residue peptide which effectively blocks the chaperone activity. We found that the DsbA assisted folding of proteins in the periplasm generates enough mechanical work to decrease the ATP consumption needed for periplasmic translocation by up to 33%.


Supplementary Figure 1. Folding probability under force measured by magnetic tweezers (A) Representative magnetic tweezers demonstrating unfolding and refolding transitions of an eightrepeat (N=8) construct of the protein L domain:
The protein is first unfolded at a constant force of 45 pN (Pulse I) resulting in eight consecutive unfolding (upwards-step) transitions of 15 nm each (see inset for magnified). The force is then quenched to 8.1 pN (Pulse II) resulting in entropic recoil of the protein followed by relaxation to an equilibrium between folding (downwards-step) and unfolding transitions. The state i of the protein at any point during the recording is equal to the number of folded domains. The total residence time (t i ) of each folded state during the equilibrium phase is shown by the gray histogram on the right and the sum of all residence times (t T ) is equal to the duration of the equilibrium phase. After calculating the states i and the residence times t i the folding probability at a force of 8.1 pN is calculated according the presented equation (pink inset). (B) Schematic of the magnetic tweezers experiment: One end of octamer of protein L is attached to the glass coverslip via HaloTag covalent chemistry and the other end is tethered to a paramagnetic bead via biotin-streptavidin binding. A precise pulling force is applied by positioning a pair of permanent magnets above the tethered paramagnetic bead with sub-micron resolution. DsbA (red curls) can be washed into or out of the flow cell during the course of an experiment.

MFPT Calculations
Mean first passage times (MFPT) were calculated by averaging the trajectory durations for n > 5 molecules. To calculate the folding MFPT, the protein construct was first fully unfolded, the force was relaxed and the time required to achieve folding of all eight domains simultaneously was recorded. This was performed for several different folding forces in the range of 4-8 pN and was then fit with a single exponential equation of the form , and the fit parameters are recorded below:

Determination of oxidation state by DTNB assay
We checked the percentage of the oxidized and reduced DsbA using Ellman's test protocol (https://www.goldbio.com/documents/2359/Ellmans+Test+Protocol.pdf). After purification, we oxidized the DsbA by incubating with 0.3% H2O2 and reduced it by incubating with 100 µM TCEP for overnight at 4ºC.
The oxidized and reduced fractions are then purified by the size exclusion chromatography. Using these samples, we performed the magnetic tweezers experiments and DTNB assay. DTNB assays are used to measure the percentage of oxidized and reduced DsbA. Our results shows for oxidized DsbA, more than 99% DsbA molecule is oxidized; whereas for the reduced one, 98.75% DsbA molecule are reduced. For oxidized DsbA, the absorbance at 412 nm is almost 0, indicating no free thiols or the presence of any reduced DsbA which concludes that more than 99% DsbA is oxidized by H 2 O 2 . In case of reduced DsbA, 5.58 µM reduced DsbA was used in DTNB assay. We measured the absorbance 0.15 at 412 nm, which shows 5.51 µM DsbA is reduced. This result shows our reduced sample is 98.75% reduced.
Supporting Figure 9: Effect of peptide (PWATCDS) on protein L refolding: Folding probabilities of protein L, only in the presence of DsbA (red) and peptide (blue) are plotted against the force. The peptide does not interact with the protein L and thus, does not shift the folding probability and it overlaps with the control (black), whereas the DsbA has been observed to increase the folding probability. Each data point is calculated using > 2500s and over three molecules per force. Error bars represent standard error of the mean.