Informing Nmr Experiments With Molecular Dynamics Simulations To Characterize The Dominant Open State Of Kcsa

The bacterial ion channel KcsA is an ideal model for eukaryotic potassium channels: it is indeed structurally simple, while featuring the hallmarks of eukaryotic potassium channels, primarily a conserved K+ selectivity filter. After a pH drop triggers its opening, KcsA, like most ion channels, becomes transiently conductive until it inactivates and the flow of ions ceases. X-Ray diffraction techniques cannot directly capture these conductive states and have to resort to mutated constructs. On the other hand, solid state NMR allows the study of ion channels in a bilayer environment and characterization of their room-temperature thermodynamic and dynamical properties. However, the interpretation of NMR data generally requires additional computational modelling, to link spectroscopic data to molecular information. In this work, we use molecular dynamics simulations in combination with solid-state NMR in order to investigate which of the proposed conductive states is dominant. To do so, we simulated the two most stable conductive states known: open (3FB5) and fully-open(5VK6). The chemical shift distribution of the configurations sampled in the simulations were then calculated and compared to the ones extracted from the NMR experiment. Bayesian inference was used to distinguish which state-dependent changes in chemical shifts were statistically significant. After this statistical filtering of chemical shifts, the comparison of the chemical shifts with experiment allowed us to identify the dominant conductive state to be the "open state". This finding is consistent with other computational investigations showing that the fully open state is less stable than the open state.