Sizes, conformational fluctuations, and SAXS profiles for intrinsically disordered proteins

The preponderance of intrinsically disordered proteins (IDPs) in the eukaryotic proteome, and their ability to interact with each other, and with folded proteins, RNA, and DNA for functional purposes, have made it important to quantitatively characterize their biophysical properties. Toward this end, we developed the transferable self-organized polymer (SOP-IDP) model to calculate the properties of several IDPs. The values of the radius of gyration ((Formula presented.)) obtained from SOP-IDP simulations are in excellent agreement (correlation coefficient of 0.96) with those estimated from SAXS experiments. For AP180 and Epsin, the predicted values of the hydrodynamic radii ((Formula presented.)) are in nearly quantitative agreement with those from fluorescence correlation spectroscopy (FCS) experiments. Strikingly, the calculated SAXS profiles for 36 IDPs are also nearly superimposable on the experimental profiles. The dependence of (Formula presented.) and the mean end-to-end distance ((Formula presented.)) on chain length, (Formula presented.), follows Flory's scaling law, (Formula presented.) ((Formula presented.) and (Formula presented.)), suggesting that globally IDPs behave as synthetic polymers in a good solvent. This finding depends on the solvent quality, which can be altered by changing variables such as pH and salt concentration. The values of (Formula presented.) and (Formula presented.) are 0.20 and 0.48 nm, respectively. Surprisingly, finite size corrections to scaling, expected on theoretical grounds, are negligible for (Formula presented.) and (Formula presented.). In contrast, only by accounting for the finite sizes of the IDPs, the dependence of experimentally measurable (Formula presented.) on (Formula presented.) can be quantitatively explained using (Formula presented.). Although Flory scaling law captures the estimates for (Formula presented.), (Formula presented.), and (Formula presented.) accurately, the spread of the simulated data around the theoretical curve is suggestive of of sequence-specific features that emerge through a fine-grained analysis of the conformational ensembles using hierarchical clustering. Typically, the ensemble of conformations partitions into three distinct clusters, having different equilibrium populations and structural properties. Without any further readjustments to the parameters of the SOP-IDP model, we also obtained nearly quantitative agreement with paramagnetic relaxation enhancement (PRE) measurements for α-synuclein. The transferable SOP-IDP model sets the stage for several applications, including the study of phase separation in IDPs and interactions with nucleic acids.