Sequence charge decoration dictates coil-globule transition in intrinsically disordered proteins

Abstract

We present an analytical theory to compute conformations of heteropolymers-applicable to describe disordered proteins-as a function of temperature and charge sequence. The theory describes coil-globule transition for a given protein sequence when temperature is varied and has been benchmarked against the all-atom Monte Carlo simulation (using CAMPARI) of intrinsically disordered proteins (IDPs). In addition, the model quantitatively shows how subtle alterations of charge placement in the primary sequence-while maintaining the same charge composition-can lead to significant changes in conformation, even as drastic as a coil (swelled above a purely random coil) to globule (collapsed below a random coil) and vice versa. The theory provides insights on how to control (enhance or suppress) these changes by tuning the temperature (or solution condition) and charge decoration. As an application, we predict the distribution of conformations (at room temperature) of all naturally occurring IDPs in the DisProt database and notice significant size variation even among IDPs with a similar composition of positive and negative charges. Based on this, we provide a new diagram-of-states delineating the sequence-conformation relation for proteins in the DisProt database. Next, we study the effect of post-translational modification, e.g., phosphorylation, on IDP conformations. Modifications as little as two-site phosphorylation can significantly alter the size of an IDP with everything else being constant (temperature, salt concentration, etc.). However, not all possible modification sites have the same effect on protein conformations; there are certain "hot spots" that can cause maximal change in conformation. The location of these "hot spots" in the parent sequence can readily be identified by using a sequence charge decoration metric originally introduced by Sawle and Ghosh. The ability of our model to predict conformations (both expanded and collapsed states) of IDPs at a high-throughput level can provide valuable insights into the different mechanisms by which phosphorylation/charge mutation controls IDP function.

Thermodynamically driven assemblies and liquid–liquid phase separations in biology

“…Annotation using curated disordered sequences from the DisProt database 203 (Box 1) initially suggests that a vast majority (∼95%) of IDPs have amino acid compositions that predispose them to be globule formers ( Figure 10A). 204 However, most of these predicted globule formers are actually polyampholytes in that they are enriched in charged residues but have roughly equal numbers of positive and negative charges. 204 Although such sequences are classified as globule formers on the basis of their low net charge per residue, in reality the conformational properties of polyampholytes are governed by the linear sequence distribution of oppositely charged residues.…”

“…204 However, most of these predicted globule formers are actually polyampholytes in that they are enriched in charged residues but have roughly equal numbers of positive and negative charges. 204 Although such sequences are classified as globule formers on the basis of their low net charge per residue, in reality the conformational properties of polyampholytes are governed by the linear sequence distribution of oppositely charged residues. If the oppositely charged residues are segregated in the linear sequence, then electrostatic attractions between oppositely charged blocks cause chain collapse and result in hairpin or globular conformations.…”

“…In sequences with well-mixed oppositely charged r static repulsions and attractions sequences adopt random-coil depending on the total charge charged residues) ( Figure 10 polyampholytes with well-mixed charged residues. 204 Thus, IDPs a classes of random coils that f conformations ( Figure 10B). Su likely to help improve the sol structured domains (see sectio flexibility that is required for func which promote high local concen parts, or entropic bristles, which exploration of conformations.…”

“…97% of sequences used in this annotat less than 0.26 and are thus predicted to be globule formers. 204 instance, the original idea of Uversky et al [185] was explored by Ashbaugh and Hatch [200] using a simple bead-spring model of peptides in which each residue can only be of three types: hydrophobic, polar, and positively charged polar. They simulated sequences with a wide range of lengths (N ), net charges (q) and hydrophobicities (H), with 15-20 random peptides generated for each N , q and H. Even though they observed that coillike and collapsed globules occupy distinct regions of the (q, H) space, the boundary between them only becomes N -independent (as predicted by Uversky) when explicit counterions are included for charged residues (vs. utilizing screened charges).…”

Charge fluctuation effects on the shape of flexible polyampholytes with applications to Intrinsically disordered proteins

“…In order to obtain R g , we adopt the Edwards-Singh (ES) type variational calculation [26], which has been extensively used in the polymer literature [2,25,[27][28][29]. More recently, the method was used to study sequence dependence of collapse of polypeptide chains [30] and polyelectrolytes [20] with application to a special class of synthetic IDPs. In developing the theory, we assume that the interactions between charges exist only between specific monomers, described by the second and third term in Eq.(6).…”