Fig. 4: PlpD is a homodimer.
a AlphaFold Multimer model of the PlpD dimer showing dimerization interfaces at the PL-domain and the β-barrel. Domains are color-coded as in Fig. 1a. Curving of the OM induced by dimerization of PlpD was calculated using the OPM server50. Residues selected to mutate to cysteine are highlighted: M249 (purple) to confirm dimerization of the PL-domains, I532 (green) and F427 (blue) to test interactions between the β-barrels, and L359 (gray) to serve as a negative control based on its location far from any predicted dimerization interface. b Magnified views of each selected residue illustrating its predicted proximity to the corresponding residue in the neighboring PlpD molecule. c SEH88 transformed with pSEH170, pSEH197, pSEH210, or pSEH213 were grown and the expression of the indicated plpD cysteine mutants was induced as described in Fig. 1b. Aliquots were then either mock-treated or treated with 4-DPS and analyzed by immunoblot using the anti-PlpDC antiserum. Both PlpD ( ~ 82 kDa) and an inter-molecularly crosslinked form ( ~ 300 kDa) were detected. d Quantitation of inter-molecular disulfide bond formation between selected cysteine residues. The levels of disulfide bond formation in mock-treated cells (‘spontaneous’, Ox-) and 4-DPS-treated cells (‘catalyzed’, Ox + ) are shown. Bars = median, N = 3 biologically independent samples obtained as in Fig. 3c. One-sided ANOVA and multiple comparison tests are shown in Table S5. e SEH88 transformed with pSEH213 were grown and the expression of the plpD I532C mutant was induced as described in Fig. 1b. 4-DPS was added and aliquots were removed from 0-120 mins to track the kinetics of inter-molecular disulfide bond formation. The samples were then analyzed by immunoblot using the anti-PlpDC antiserum. Curve of best fit is shown. Error bars=standard error from the mean, N = 3 biologically independent samples obtained as in Fig. 3c. f The experiment described in (e) was repeated except cells were transformed with pSEH197 to express the plpD M249C mutant and aliquots were removed from 0-90 min. Error bars and N are as in (e). Curve of best fit is shown. Representative experiments for (e) and (f) are shown in Fig. S8 and statistical analyzes are shown in Table S7.
