Charge Patterned Sequences form Helical Structures through Charge Neutralization

Abstract

Many proteins with repeating blocks of oppositely charged residues form long single alpha helices (SAHs). The consensus sequence is a block of four Glu residues followed by a block of four Lys or Arg residues, (Glu4(Lys/Arg)4)n. The current working hypothesis is that SAHs are stabilized by i:i+4 salt bridges between opposite charges in consecutive helical turns. We test the merits of this hypothesis to understand the sequence-encoded preference for SAHs and the logic behind the failure of certain atomistic simulations in anticipating the preference for stable SAHs.In simulations with fixed charges the favorable free energy of solvation of charged residues and the associated loss of sidechain entropy hinders the formation of SAHs. We proposed that alterations to charge states induced by sequence context might play an important role in stabilizing SAHs. We tested this hypothesis using a (Glu4Lys4)n repeat protein and a simulation strategy that permits the substitution of charged residues with neutralized protonated or deprotonated variants of Glu / Lys. Our results predict that stable SAH structures derive from the neutralization of approximately half the Glu residues. These findings explain experimental observations and also provide a coherent rationale for the failure of simulations based on fixed charge models. Large-scale sequence analysis reveals that naturally occurring sequences often include “defects” in charge patterns such as Gln or Ala substitutions. This sequence-encoded incorporation of uncharged residues combined with neutralization of charged residues might tilt the balance toward alpha helical conformations.Our results highlight the need for developing robust methodologies for constant pH simulations that can be applied to sequences with high charge contents. They also highlight the need for generalizing bioinformatics predictors to account for sequence-encoded charge regulation that might influence disorder predictions for sequences with high fractions of charged residues.