Abstract
<jats:p>The question what features of protein structure make it prone to reversibly alter shape and function under native conditions is a matter of considerable interest. To address this issue, we have in the past investigated whether fold-switching of lymphotactin, E. coli RfaH and Mad2 is brought about by a specific pattern of backbone density distribution. Metamorphic fold switching was found to occur when a backbone segment of marginal secondary-structure propensity toggles between folding basins of different permittivity. In backbone-polarization model, those conditions are defined by (i) folding potential, a function FPi of shielding tensors of Cα i atoms, which identifies secondary-structure propensity as marginal or robust; (ii) folding basin, a region of continuous dielectric of permittivity εr; and (iii) folding template, which accounts for charge separation in the medium e.g. Gosh-Debye-Hückel ionic matrix. To test the explanatory and predictive power of this model, we now review recent studies of thermal stability and structural dynamics of KOW domains in NusG/Spt5 and RfaH transcription factors, CD and NMR studies of a range of NusG/Spt5/LoaP/RfaH KOW variants, exchange-based NMR studies of refolding of EcRfaH KOW, and thermodynamic and kinetic studies of circadian clock protein KaiB, human lymphotactin, and a designed protein Sa1, in which permittivity of the folding basin is increased by low temperature and decreased by crowding agents. Lastly, we test the scope of this model by looking at putative metamorphism of Aβ42 paranuclei, and loop-to-helix transition in influenza hemagglutinin.</jats:p>