Sequence-Encoded Heterogeneity of Interactions Decouples Different Measures of Protein Sizes and Reconciles the Discrepant Inferences from Saxs versus Fret Experiments

Abstract

Quantifying the conformational properties of folded proteins as a function of denaturant concentration is important for understanding the mechanism of protein folding. Additionally, quantifying the conformational properties of intrinsically disordered proteins (IDPs) can yield insights regarding the functions of IDPs. Radius of gyration, RG, which quantifies the average size of a protein, and end-to-end distance, REE, are two commonly used conformational descriptors as these values can be determined using small-angle x-ray scattering (SAXS) and single molecule Förster resonance energy transfer (smFRET), respectively. However, for several proteins, the RG profile determined directly from SAXS as a function of denaturant concentration differs from the RG profile inferred using REE values determined from smFRET and a given homopolymer model. Here, we show that this discrepancy is not due to any inherent flaws of either measurement, but instead arises as a result of sequence-specific interactions that exist within proteins under low denaturant conditions. Sequence-specific interactions induce heterogeneity within the conformational ensemble of a protein, which in turn decouples globally averaged (e.g., RG) and length-scale-specific (e.g., REE) conformational descriptors. The extent of decoupling of conformational descriptors, and thus the appropriateness of using a homopolymer model to convert between RG and REE, is sequence specific and is often unknown a priori. Here, we investigate the relationship between sequence-complexity and deviations from homopolymer models. Additionally, we show that by combining results from experimental measurements that yield distinct conformational descriptors with molecular simulations we can provide complete conformational descriptions of heterogeneous ensembles.