Fig. 6. MreB twist angles predict filament length and orientation in vivo.

a Structured illumination microscopy of wildtype and the four EcMreB mutants in Fig. 4a constructed in E. coli cells with a sandwich fusion of msfGFP to MreB. Images are maximum projections of a z-stack, with red (membrane dye FM 4-64FX) and green (MreB-msfGFP) channels merged. White arrows highlight typical MreB filaments in each strain. Scale bar is 1 µm. b The cumulative distributions of MreB-msfGFP fluorescence filament length across mutants trend with their predicted intrinsic twist angles in Fig. 4b. The V55A and I141V strains had shorter filaments than wildtype, and the E276D and R124C strains contained longer filaments. MreB filaments were defined as continuous regions with high msfGFP signal on the cylindrical cell body with area larger than the diffraction limit. n > 1000 filaments were measured for each strain. Inset: the 99th percentile of filament length in each strain was highly correlated with the mean twist angle from Fig. 4b (Pearson’s r = −0.98, p = 0.004, Student’s t test, n = 5 strains), providing experimental validation of the coarse-grained model. c Cumulative distributions of MreB filament pitch angle. The V55A and I141V strains had larger pitch angles than wildtype, while E276D and R124C had smaller pitch angles that were closer to 90°. The pitch angle was defined as the angle between the main axis of each fluorescent patch and the long axis of the cell. Inset: the experimentally measured pitch angle was highly correlated with the mean twist angle from Fig. 4b (Pearson’s r = 0.94, p = 0.02, Student’s t test, n = 5 strains). Data points are mean ± S.E.M. for n > 1000 patches in each strain.