Supplementary Figure 7: Molecular models based on standard cIAP1 SAXS data.
(a) The best of ten molecular models of cIAP1-B3R-ΔC7 is shown superposed with the averaged, filtered ab initio model. A normalized spatial discrepancy (NSD) for the aligment is indicated. The fit (dark blue line) of the molecular model to the experimental data (red circles) is shown below the model, and the Chi value describing the fit is indicated. I stands for scattering intensity and q is proportional to the scattering angle (q = 4 sin(θ)/λ, where 2θ = the angle between the incident X-ray beam and the detector, and λ = the X-ray wavelength in Ångstroms). (b) The same data as in (a) is shown for cIAP1-B3R-ΔC7 in the presence of 1 mM AVPW. (c, d) Constraints used in the generation of closed monomeric molecular models are shown. All constraints are based on mutational data1. (e) The best of ten molecular models of cIAP1-B3R is shown superposed with the averaged, filtered ab initio model, in two views. The NSD between the two models is indicated. (f) The fit of the best cIAP1-B3R molecular model to the experimental data is shown. All colors and variables are as in (a). (g) A subset of calculated molecular models for cIAP1-B3R is shown aligned by their BIR3 domains to demonstrate the heterogeneity of the position of the CARD. The protein likely exists as an ensemble of states not fully reflected by any single model. (h) The distributions of Rg values in the ensembles of molecular models generated by EOM for cIAP1-B3R (black), cIAP1-B3RL617E (orange), cIAP1-B3R-ΔC7 (violet) and cIAP1-B3RL617E + AVPW (green) are shown. The initial, unoptimized pools are displayed as dotted lines, and the optimized ensembles as solid lines. Note that the closed states (cIAP1-B3R and cIAP1-B3RL617E) adopt tighter and smaller distributions than the open states (cIAP1-B3R-ΔC7 and cIAP1-B3RL617E + AVPW), reflecting the more rigid conformation of the closed states.
